摘要:

DALTON PERSPECTIVE J. Chem. Soc., Dalton Trans., 1998, Pages 2771–2776 2771 Molecular modeling of d- and f-block metal complexes Thomas R. Cundari * Department of Chemistry, University of Memphis, Memphis, TN 38152-6060, USA. E-mail: tcundari@memphis.edu This Perspective gives an overview of challenges in the application of molecular mechanics to d- and f-block complexes. Molecular mechanics (MM) entails a classical mechanics description of chemical systems. Thus, computational challenges faced by the modeler are often diVerent from those encountered for quantum methods.However, as with main group compounds, there are considerable motivations to develop MM force fields for d- and fblock metals. Foremost among these is that MM calculations are very computationally inexpensive. This permits one to (a) study very large (hence generally more experimentally relevant) models, (b) utilize readily available hardware, and (c) carry out more complete conformational analyses.Challenges in the development of MM force fields for d- and f-block metals include (a) the scarcity of metric and vibrational data for parameterization, (b) extending ‘organic’ force fields to inorganic species, and (c) parameterizing force fields with static structures. However, these problems pale in comparison to the major roadblocks to metal force field development: treating angular distortions about metal atoms (which are more variable than typically seen in lighter main group elements) and transferability (which arises from the tremendous chemical diversity of these metals). 1 Introduction Perhaps no technological development has made a more signifi- cant impact on chemistry in the past two decades than the advent of aVordable, reliable, and powerful computers. With these advances, and the development of easy-to-use software, computational chemistry has become a valuable tool in the chemist’s arsenal for design and analysis of materials and processes. Although one always runs into the danger of overgeneralizing in a diverse field, one can loosely divide computational chemistry research themes into two general groups: development of new techniques, and extension of existing techniques to new chemical families.This Perspective will focus on the latter, because it is often the case in modeling inorganic chemistry that a good deal of time is spent extending techniques originally developed for organic chemistry. Computational techniques for modeling inorganic compounds run the gamut of sophistication. In general, the more comprehensive a model (classical or quantal), the greater are the computational resources required.Hence, there is great interest in developing less expensive, although reliable, methods.1 One approach is the use of semiempirical approximations for a full quantum description of chemical systems. Semiempirical quantum mechanical (SEQM) techniques often involve the approximation or neglect of computationally expensive integrals that describe the interactions among nuclei and electrons.2 Alternatively these integrals can be viewed as adjustable parameters whose selection is based on the ability to reproduce and ideally predict experimental observables.SEQM methods include the well known extended Hückel method,3 as well as the techniques included in programs such as MOPAC and recently extended to d-block metals.2 An alternative approach, the subject of this Perspective, entails a classical mechanics picture of metal complexes.Tom Cundari obtained his B.S. at Pace University in 1986, and a Ph.D. in 1990 at the University of Florida (Russ Drago, Advisor). After a year of postdoctoral study at North Dakota State University with Mark Gordon, he started as Assistant Professor at The University of Memphis. He was promoted to Associate Professor in 1994. His research interests include modeling of lanthanide and transition metal chemistry, artificial intelligence in inorganic chemistry, and cricket.Pictured in the recent group photo: (back row, left to right) Tie Zhou, Wentao Fu, Tom Klinckman, Tom Cundari and Mike Benson; (front row, left to right) Akihiko Yoshikawa, Leah Saunders, Mary Cocke and Jun Deng.2772 J. Chem. Soc., Dalton Trans., 1998, Pages 2771–2776 2 Goals and Themes This Perspective will focus on molecular mechanics (MM) techniques for d- and f-block complexes, which signifies a classical (‘ball-and-spring’) description of chemical bonding.Recent reviews are available describing the development and application of quantum methods for metal chemistry.4 Excellent overviews of the theory behind MM can be found in reviews by Hay5 and Landis et al.6 and the monographs by Comba and Hambley 7 and Allinger.8 The latter deals with organic chemistry but it is requisite reading for anyone wishing to learn MM. Another work that is unfortunately often overlooked is the seminal work by Kepert 9 on metal stereochemistry utilizing a points-on-a-sphere method.It is hoped that this Perspective will serve as an introduction to the challenges and opportunities in MM modeling for inorganic chemists (experimental or computational) who may be more familiar with quantum mechanical techniques. 3 Bigger is Better Computational chemists have one property in common with gases, they can expand to fill any volume. Hence, a good deal of research involves development of new methods to make feasible the study of larger systems, either with more eYcient techniques or the marriage of existing techniques (e.g.density functional theory 10) with more powerful architectures (e.g. parallel computing11). It seems as if once it becomes feasible to model silyl (SiH3) substituents, one’s experimental colleagues want to model SiMe3 or SiBut 3, Scheme 1. Larger substituents clearly engender a bigger computational problem, but they are also more experimentally relevant as indicated by a cursory glance at a crystallographic database.12 There are 3 and 1781 d-block complexes with silyl and SiMe3 substituents, respectively; even the very bulky SiBut 3 group, developed by Wolczanski group, is found in 17 d-block complexes.13 A fruitful approach to eYciently modeling larger systems is molecular mechanics,5–8 which entails a classical description of a chemical system.In its simplest implementation a complex is viewed as an assemblage of balls and springs, with the former modeling atoms and the latter the chemical bonds that join them.In molecular mechanics the total energy is the sum of individual contributions calculated by means of relatively simple algebraic equations. This is in contrast to the integrodi Verential equations whose solution in the Schrödinger equation makes quantum mechanical (QM) calculations computationally expensive. This diVerence immediately suggests that for similarly sized systems MM will be faster by order of magnitudes, which has several important implications. (1) Molecular mechanics can usually be readily applied to very large systems more eYciently than QM methods.This assertion is supported by the great body of work on the application of MM to large biomolecules. (2) The reduced computational demands of MM permit the use of less sophisticated (hence less expensive and more readily available) computers to attack chemical problems.This brings with it the potential for nearly all inorganic chemists to employ MM modeling in design and analysis. (3) Another exciting by-product of studying large systems is Scheme 1 that it allows for greater correspondence between experimental systems and computational models. Apart from the obvious fact that more realistic models are preferable, it is useful from a computational point of view. As the model closely approaches experiment, one may assume that substantial deviations (assuming the experimental results are correct!) are due to a deficit in the theoretical model.Although this is unwelcome news for the computational chemist, it is as important to identify systems that are not amenable to description by a model as it is to know those which are. (4) A major advantage to the routine and eYcient study of larger, more experimentally relevant chemical systems is that it allows the theory–experiment interface to be more dynamic. Clearly, if calculations take longer to do than the experiments, there is little advantage to utilizing theory to aid in the design of a new chemical. However, if the modeler can quickly obtain results that yield new insight and provide useful suggestions for further experiments then the synergism between theory and experiment is more fully realized. 4 Why Bother? Before embarking on some of the challenges inherent in MM modeling of d- and f-block metal complexes, it is prudent to discuss motivations for this pursuit lest the reader be unnecessarily disheartened.14 The d-block or transition metals have fascinated inorganic chemists since the time of the great debate between Jørgensen and Werner regarding the structure of coordination complexes.Perhaps the next major revolution in transition metal (TM) chemistry came about with the advent of organometallic chemistry as a distinct discipline. This can be traced to the research of Wilkinson and others with metallocenes and the work of Ziegler, Natta and co-workers on catalytic olefin polymerization.Interest in f-block metals has largely resulted from the utility of the actinides in nuclear chemistry. Recently, there has been growing interest in the lanthanides as experimentalists have sought to exploit their unique chemical, magnetic and photophysical properties for a variety of technological applications. d-Block complexes are typically found with a wide variety of formal oxidation and spin states, co-ordination geometries, bond types (single, double, triple, quadruple, and dative), and ligand types (hard or soft bases).For the academic chemist unraveling the how and why of this diversity is intrinsically Scheme 2 Catalytic olefin hydrogenation mechanics; sol = solvent. Adapted from ref. 15J. Chem. Soc., Dalton Trans., 1998, Pages 2771–2776 2773 interesting, but it also has a practical side. Consider a typical catalytic cycle (e.g.hydrogenation by Wilkinson’s catalyst,15 Scheme 2) with its changing ligand types, co-ordination geometries and oxidation states. The ability of transition metals to stabilize diVerent chemical environments is the reason for their utility in catalysis, advanced materials, biology, and medicine. The diversity provides experimentalists with many options for new materials and processes. However, this chemical diversity can be a thorn in the side of the modeler. To be truly successful a computational model must ideally be able to adequately describe not only a narrow subset of complexes, but rather a wide assortment with comparable accuracy. This challenge brings with it special complications in MM modeling of metal complexes that do not arise in quantum modeling.In the following section I attempt to outline some of these challenges and innovative approaches taken to treat them. 5 The Challenge of Molecular Mechanics for Metal Complexes The following is a brief overview of molecular mechanics, with emphasis on issues important for metal complexes.5–7 The interested reader is directed elsewhere for a more in-depth discussion of the theory.A minimal MM force field is given in equation (1).8 The steric energy (Usteric) of a compound is described as the Usteric = oUr 1 oUq 1 oUt 1 oUvdw (1) sum of individual bond-stretching (Ur), angle-bending (Uq), bond torsion (Ut), and van der Waals (Uvdw) interactions. The optimum geometry is the combination of internal coordinates with the lowest Usteric.Additional terms to describe other interactions (e.g. hydrogen-bonding or electrostatic interactions) can be added to equation (1) as needed. A simple MM picture of a molecule as a collection of balls and springs is instructive. A spring is described by Hooke’s law, equation (2), which introduces several parameter types. First, Uq = 0.5k(q 2 q0)2 (2) one must know the equilibrium geometry (q0) of the spring (bond).The second important quantity is the force constant (k) which describes the restoring force needed to bring the spring back to equilibrium. I refer to the former as metric parameters and the latter as vibrational parameters. Essentially one wishes to know the equilibrium value of an internal coordinate as well as the energy required to displace it from this equilibrium value. Obtaining MM parameters is dealt with in the following two subsections. Subsequent subsections address MM parameter transferability, and other thorny problems in MM modeling of metal complexes.a Metric parameters Metric parameters can be obtained from a variety of experimental sources including solid-state neutron and X-ray diVraction, or gas-phase electron diVraction. Alternatively, they can be obtained from high-level ab initio quantum calculations on suitable model compounds. If the past decade of quantum chemistry research has proven anything it is this: with the right computational method (the description of which occupies a large literature), one can accurately predict geometries.I qualify ‘accurately’ since its definition is subjective, but for most purposes this involves theory–experiment agreement of the order of ±0.01 Å for bond lengths and ±18 for bond angles. Torsional (or dihedral) angles are the softest internal modes and hence one’s vision of an accurately determined torsional angle depends on the problem at hand. Orpen’s research 16 on the systematic analysis of X-ray diVraction structures suggests that the aforementioned guidelines may actually be optimistic given the variability of metal complex geometries with their environment.b Vibrational parameters Generally, it is more diYcult to obtain vibrational than metric parameters from experiment as force constants must be extracted from spectroscopic, isotopic-labeling studies. Vibrational parameters can be determined from a QM-derived energy Hessian (second derivative of the energy with respect to atomic coordinates).17 The main sticking point of the QM approach is that for the sake of tractability most vibrational frequencies (and force constants derived from them) are calculated within the harmonic approximation.The reality is that anharmonic eVects occur in vibrational spectra, and that experimentally quantifying them is often diYcult. This dilemma was largely resolved by a crucial contribution from the Pople group.17 They showed that QM-calculated, harmonic frequencies generally diVer from experimental (and hence anharmonic) frequencies by a constant or scale factor.It is often observed that calculated harmonic frequencies are ª10% too high. The magnitude of the scale factor changes as one goes to diVerent levels of theory, but the relationship holds up remarkably well. Recent work by Cundari and Raby18 has sought to evaluate scaling approximations for estimating MM vibrational parameters for TM complexes.The big stumbling block, as it always seems to be in computational metal chemistry, is the lack of a large and diverse database of reliable experimental results against which one can ‘calibrate’ the theory. However, our work, as well as that of others, suggests that for a variety of metal systems the approximations first forwarded by Pople and co-workers seem to hold for metal complexes, particularly over a series of related complexes.17 c What is a typical gadolinium–nitrogen bond? The transferability issue Assuming one can obtain metric and vibrational parameters by experiment or calculation, the would-be modeler of metal complexes must deal with transferability.The relatively narrow range of atom and bond types in main group compounds, particularly those involving lighter elements, means that transferability is much less problematic in molecular modeling of organic molecules. The question in this subsection title marks the moment at which this thorny issue impressed itself upon my research group.If we return to the simple ball-and-spring analogy, MM requires the chemist to specify the nature of the balls as well as the springs that connect them. This is a problem as metals often come in a bewildering array of oxidation states [is the Gd]N bond in a gadolinium(III) co-ordination complex a reasonable facsimile of that in a gadolinium(0) organometallic?], bond types (e.g. the Gd]N bond involving an amine versus an imine ligand), co-ordination numbers, spin states, and coordination types.If we further consider the Gd]N bond one must distinguish a Gd]N linkage in a nine-co-ordinate tricapped trigonal prismatic TRPS-9) complex from that in a capped square antiprismatic complex, and also among Gd]N bonds involving capping and prismatic co-ordination sites.19 Two major approaches have been developed to deal with transferability: ignore it or develop atom/bond types in greater number and with more specificity.The former is more popular. The latter approach involves developing parameters for specific chemical environments, e.g. distinguish between the apical (Tc]La) bond in a square pyramidal TcL5 complex from the four basal (Tc]Lb) bonds by giving each set of bonds diVerent q0 and k, see equation (2) and Scheme 3. Advantages and disadvantages of this approach are obvious. One gains the potential for greater accuracy, but there is a great increase in the number of parameters.Less obvious is the loss of generality2774 J. Chem. Soc., Dalton Trans., 1998, Pages 2771–2776 resulting from a specific MM model. Using the square pyramidal TcL5 example, one would have to energetically distinguish two isomers (A and B in Scheme 3) and correctly predict A as the more stable candidate. Clearly, as the number (and variety) of ligands increases the situation becomes considerably more problematic. An interesting approach to transferability in MM was employed by Hay20 in his work on lanthanide (Ln) aqua and nitrate complexes.It is found, particularly for nine-co-ordinate TRPS-9 complexes, that Ln]L bond lengths can diVer quite substantially depending on whether the ligand L occupies a capping or prismatic co-ordination site. Prismatic ligands occupy more hindered positions than capping ligands and as a result they typically have longer q0 for comparable metal–ligand bond types. Given the weakness (i.e.low force constant) of gadolinium–ligand bonds, the equilibrium bond lengths can cover a range of 0.1 Å within a single complex. Hay’s approach involves the use of very small Gd]L force constants (k ª 0.1 mdyn Å21; dyn = 1025 N), allowing in essence the bonds to expand or contract in response to steric pressure resulting from non-bonded terms [Uvdw in equation (1)]. Although this approach must be employed judiciously, and the results evaluated critically, Hay has shown this to be an eYcient and eVective technique.Our group followed the lead of Hay for a variety of other co-ordinating atom types for high co-ordination number gadolinium(III) complexes.19 It is unclear how this simple technique could be applied to transition metal complexes, which have a larger covalent contribution to their bonding than the lanthanides, although research to address this question would be of interest. d Using organic force fields for inorganic complexes The majority of eVorts at developing MM force fields for metals amount to extending popular ‘organic’ force fields such as MM2 and AMBER by inclusion of new atom types and parameters.The MM parameters can be divided up into two groups, metal-dependent and metal-independent. It is generally assumed that the metal-independent MM parameters needed to describe an organic ligand are the same whether it is coordinated to a metal or not. In other words the Namine]Calkane bond in an edta complex of GdIII has the same force constant and equilibrium bond length whether or not it is co-ordinated to a metal.Chemical intuition suggests that such an approximation is more plausible for co-ordination complexes with their dative/ionic metal–ligand interactions as compared to organometallics which generally have more covalent bonding. A recent contribution from the Comba group21 has looked at this issue for co-ordination complexes and found that in some cases the use of ‘organic’ MM parameters for co-ordinated ligands can lead to significant errors.However, the majority of studies on MM modeling of metal complexes have successfully utilized ‘organic’ force fields for metal-independent parameters.5–7 e Parameterizing using static structures Another sticky issue involves the use of static structures (obtained principally by solid-state crystallography) to assess the predictive ability of a newly developed force field. This issue is not unique to MM modeling of metal complexes.Everyone who has given seminars on MM modeling has almost surely Scheme 3 heard the question how do you know the solid-state structure is the same as the solution structure? Answers to this question can range from unimaginative (how do you know it’s not?) to metaphysical (what do we mean by molecular structure anyway?). As the field of MM modeling of metal complexes matures it is increasingly desirable to carry out comparisons not only between single complexes, but ensembles of molecules to develop a statistical profile of calculated and experimental metric properties for chemical moieties (e.g.GdCN = 9]Namine bonds or N]C]C]N torsional angles in transition metal ethylenediamine complexes, Fig. 1). The wider availability of powerful, easy-to-use graphical packages for mining structural databases will assist in this task. As Orpen 16 points out in his review, by studying the structural variability of a chemical moiety in a variety of solid-state environments one is in some sense modeling the diVerent chemical environments seen in solution.A very narrow range of values for a particular metric parameter in diVerent crystal environments leads one to expect this parameter to be relatively unchanged upon going from the solid to solution phase. f Large co-ordination numbers A major challenge in MM studies of metal complexes concerns modeling the angular arrangement of ligands about high coordination number (CN) metals.Organic molecules tend to have co-ordination geometries that cover a narrow range of angular orientations: it takes a lot of energy and steric pressure to significantly displace an sp3 C from tetrahedral or to induce non-planarity about the Csp2]] Csp2 double bond of an olefin. Metals, particularly those of the d and f block, generally have a wider range of co-ordination geometries. The situation is particularly troublesome for co-ordination numbers of seven and higher (the norm in f-block chemistry) where there are often minuscule thermodynamic diVerences and small kinetic barriers among structural polytopes.Also, high CN complexes often have symmetry inequivalent co-ordination sites. Consider a simple homoleptic ML9, Scheme 4, with TRPS-9 geometry. There are two distinct ligand co-ordination sites (capped, Lc, and prismatic, Lp) and six unique L]M]L bond angle types with equilibrium values ranging from ª70 to ª1408 for a system Fig. 1 Histogram showing the range of N]C]C]N dihedral angles in d-block complexes 10 Scheme 4J. Chem. Soc., Dalton Trans., 1998, Pages 2771–2776 2775 with an idealized TRPS-9 (D3h) structure. Clearly, as one goes to larger co-ordination numbers and diVerent ligating atom types the treatment of the angular interactions about high CN metals becomes increasingly diYcult. One approach to dealing with the challenge of high coordination number complexes has its genesis in the work of Kepert.9 Kepert developed and extensively applied a points-ona- sphere (POS) model to investigate the stereochemistry of coordination complexes.This simple and intuitive (and therefore powerful) model predicts the stereochemistry of metal complexes on the basis of ligand–ligand repulsions (1,3-nonbonding interactions). Conceptually, the method can be thought of as an extension to metal co-ordination complexes of the well known VSEPR model. The lowest energy coordination geometry is determined from minimization of a simple functional of the type in equation (3) where r is the V µ rij 2n (3) distance between ligating atoms i and j, and n is an integer that can range from 1 (Coulombic) to infinity (hard sphere approximation). A value of n in the range of 6 seems to give the closest correspondence with experiment.In the simplest POS implementation ligating atoms are constrained to move on the surface of a sphere centered at the metal atom thereby fixing the metal-ligand distance.The POS approximation is expected to be most valid in situations in which the metal–ligand bond is highly ionic. In such cases there is minimal directionality arising from covalent, orbitally directed interactions and the preferred co-ordination geometry results from minimization of ligand–ligand repulsions within the constraints of chelation, the stereochemical requirements of the organic ligands, etc. Hence, lanthanide complexes are an ideal family of metal complexes for investigation with MM techniques that employ POS approximation.An MM implementation of the POS model to high coordination number complexes is well demonstrated by Hay’s work on lanthanide aqua and nitrate complexes.20 An extension of Hay’s approach to lanthanide SchiV base and related complexes was reported by Cundari et al.19 Based on the descriptive chemistry of the metals one would expect the POS approximation to also be particularly useful for co-ordination complexes of the alkali metals and alkaline earth metals. 6 Summary, Conclusion and Prospectus The MM techniques have become routine for many families or organic complexes and the exploitation of this computationally eYcient, chemically intuitive model for metal chemistry has attracted increased interest. This contribution has sought to outline some of the current challenges and opportunities in molecular mechanics modeling of metal complexes.Much of the preceding discussion is colored by experience in the author’s own lab.19,22–29 Alternative and complementary views can be found in the growing literature dedicated to molecular mechanics calculations on metal complexes.5–7 A combination of quantum calculations and structural databases seems an eVective solution to obtaining needed metric and vibrational parameters. Likewise, taking parameters originally derived for organic molecules and using them for the metal-independent parameters of ligands seems to generally be successful although the caveat of Comba and co-workers 21 must always be kept in mind.From the author’s perspective the two immediate, major challenges in MM modeling of metal complexes involve treating L]M]L angular interactions and parameter transferability. The points-on-a-sphere approach is successful for a wide range of metal complexes. In the author’s lab MM force fields have been developed using the POS approximation to describe complexes of gadolinium,19 platinum,22–23 vanadium,24,25 chromium28 and technetium.29 The greater degree of ionic/ dative metal–ligand bonding of metals in the first transition series as compared to second and third row congeners suggests the former will be more amenable to the POS description as will lanthanide complexes versus actinide analogues and co-ordination complexes versus organometallics. Systematic research on these issues will be of interest from the viewpoint of the important applications of the metals involved, and may yield important new insight into the bonding and structure of metal complexes.Landis et al.30 have developed an alternative method for treating L]M]L bond angles using a valence bondtype approach that designates a hybridization at the metal; preliminary applications to metal alkyl and metal hydride complexes are encouraging and further application of this approach will be of great interest. In the final scheme of things, the chemist must assess what level of accuracy is required for a particular application in order to select a suitable computational model. I have found in my research on metal complexes that one can carry out reliable (with respect to the tertiary structure of both the metals and ligands) MM conformational searches.This is in large part due to the fact that the preference for a particular conformer is often not inordinately influenced by reasonable uncertainty in the estimation of a few metal-dependent MM parameters.One advantage to a simple MM force field like that in equation (1) is that often these are quite robust with respect to small modifi- cations in the vibrational and metric parameters. Furthermore, the structural similarity of low energy conformations suggests that errors due to neglect of transferability will cancel out to a fair degree. Two final comments are germane. First, the highly variable nature of bonding in metal complexes, particular TM complexes, that leads to the transferability problem in MM also makes QM modeling of metal complexes challenging.Secondly, of great interest are combined MM/QM methodologies such as outlined by Maseras and Morokuma31 in which QM methods are used to model metal–ligand interactions (where transferability is probably most significant) while MM is used to describe the main group ligands. Alternatively, it is possible to use MM to perform a quick conformational analysis and obtain low-energy conformations which could then be submitted to further refinement in a separate quantum mechanical step.24,25,29 Another technique for the inclusion of electronic (and hence quantum) eVects such as Jahn–Teller distortions into MM calculations has been addressed by Deeth and Paget.32 These researchers have added another term to equation (1), the so-called cellular ligand field stabilization energy (CLFSE), to model stereochemical eVects arising from the variable occupancy of the d orbital manifold. 7 Acknowledgements T. R. C. acknowledges the graduate (Wentao Fu, Tom Klinckman, Phil Raby) and undergraduate (Melissa Beaugrand, Mary Cocke, Leah Saunders, Laura Sisterhen, Leigh-Anne Snyder, Chryssanthi Stylianopoulos) students doing MM research for their hard work and dedication. A special debt of gratitude is due to Dr. Eddie Moody (now at Los Alamos National Laboratory) for his perseverance in initiating this research in our lab.Likewise, Professors Lori Slavin (Austin Peay State University, Clarksville, TN, USA) and Shaun Sommerer (Barry University, Miami, FL, USA) were instrumental in this research. Initial support for this research was provided by the Petroleum Research Fund administered by the American Chemical Society. DiVerent portions of this research were supported by the United States National Science Foundation (NSF), grant CHE-9614346, and Department of Energy, grant DE-FG02-97ER14811. T.R. C. also acknowledges the NSF for support of computational chemistry at The University2776 J. Chem. Soc., Dalton Trans., 1998, Pages 2771–2776 of Memphis through the Academic Research Infrastructure (grant STI-9602656) and Chemical Research Instrumentation and Facilities (grant CHE-9708517) programs. 8 References 1 The Reviews in Computational Chemistry series, eds. D. B. Boyd and K. B. Lipkowitz, VCH, New York, provides an excellent overview of modern computational chemistry. 2 MOPAC, A Semiempirical Molecular Orbital Program, J. J. P. Stewart, J. Comput. Aided Mol. Des., 1990, 4, 1. 3 M. H. Whangbo, J. K. Burdett and T. A. Albright, Orbital Interactions in Chemistry, Wiley, New York, 1985. 4 M. T. Benson, T. R. Cundari, M. L. Lutz and S. O. Sommerer, in Reviews in Computational Chemistry, eds. D. B. Boyd and K. B. Lipkowitz, VCH, New York, 1996, vol. 8, pp. 145–202; G. Frenking, I. Antes, M. Böhme, S. Dapprich, A. W. Ehlers, V.Jonas, A. Neuhaus, M. Otto, R. Stegmann, A. Veldkamp and S. F. Vyboishchikov, in Reviews in Computational Chemistry, eds. D. B. Boyd and K. B. Lipkowitz, VCH, New York, 1996, vol. 8, pp. 63–143. 5 B. P. Hay, Coord. Chem. Rev., 1993, 126, 177. 6 C. L. Landis, D. M. Root and T. Cleveland, in Reviews in Computational Chemistry, eds. D. B. Boyd and K. B. Lipkowitz, VCH, New York, 1995, vol. 6, pp. 73–136. 7 P. Comba and T. W. Hambley, Molecular Modeling of Inorganic Compounds, VCH, New York, 1995. 8 U. Burkert and N. L. Allinger, ACS Symp. Ser., 1982, 177. 9 D. L. Kepert, Inorganic Stereochemistry, Springer, Berlin, 1982. 10 L. J. Bartolotti and K. Flurchick, in Reviews in Computational Chemistry, eds. D. B. Boyd and K. B. Lipkowitz, VCH, New York, 1996, vol. 7, pp. 187–216. 11 T. G. Mattson, ACS Symp. Ser., 1995, 592. 12 Cambridge Structural Database System, Version 5.14, Cambridge Crystallographic Data Centre, October 1997. 13 See for example, C. P. Schaller and P. T. Wolczanski, Inorg. Chem., 1993, 32, 131; J. B. Bonanno, P. T. Wolczanski and E. B. Lobkovsky, J. Am. Chem. Soc., 1994, 116, 11 159; C. P. Schaller, C. C. Cummins and P. T. Wolczanski, J. Am. Chem. Soc., 1996, 118, 591. 14 The diversity of interesting and important applications of d- and f-block metal chemistry can be inferred from inspection of a standard text such as F. A. Cotton and G. Wilkinson, Advanced Inorganic Chemistry, 5th edn., Wiley, New York, 1988. 15 A. Yamamoto, Organotransition Metal Chemistry, Wiley, New York, 1986. 16 A. G. Orpen, Chem. Soc. Rev., 1993, 191. 17 W. J. Hehre, L. Radom, P. v. R. Schleyer and J. A. Pople, Ab initio Molecular Orbital Theory, Wiley, New York, 1986. 18 T. R. Cundari and P. D. Raby, J. Phys. Chem. A, 1997, 101, 5783. 19 T. R. Cundari, E. W. Moody and S. O. Sommerer, Inorg. Chem., 1995, 34, 5989. 20 B. P. Hay, Inorg. Chem., 1991, 30, 2876. 21 J. E. Bol, C. Buning, P. Comba, J. Reedijk and M. Ströhle, J. Comput. Chem., 1998, 19, 512. 22 T. R. Cundari, W. Fu, T. R. Klinckman, E. W. Moody, L. L. Slavin, L. A. Snyder and S. O. Sommerer, J. Phys. Chem., 1996, 100, 18 157. 23 T. R. Cundari and W. Fu, J. Mol. Struct. (THEOCHEM), 1998, 425, 51. 24 T. R. Cundari, L. L. Sisterhen and C. L. Stylianopoulos, Inorg. Chem., 1997, 36, 4029. 25 T. R. Cundari, L. C. Saunders and L. L. Sisterhen, J. Phys. Chem., 1998, 102, 997. 26 T. R. Cundari, N. Matsunaga and E. W. Moody, J. Phys. Chem., 1996, 100, 6475. 27 ACA Transaction Symposium, T. R. Cundari, E. S. Ignarra, E. W. Moody, P. D. Raby and S. O. Sommerer, 1995, 31, 23. 28 M. V. Cocke and T. R. Cundari, unpublished work. 29 M. Beaugrand, T. R. Cundari and W. Fu, unpublished work. 30 C. R. Landis, T. Cleveland and T. K. Firman, J. Am. Chem. Soc., 1998, 120, 2641. 31 F. Maseras and K. Morokuma, J. Comput. Chem., 1995, 16, 170. 32 R. J. Deeth and V. J. Paget, J. Chem. Soc., Dalton Trans., 1997, 527 and refs. therein. Received 18th March 1998; Paper 8/02144I

ISSN:1477-9226    

DOI:10.1039/a802144i

出版商:RSC    

年代:1998

数据来源: RSC

摘要:

DALTON COMMUNICATION J. Chem. Soc., Dalton Trans., 1998, Pages 2777–2778 2777 Macropolyhedral boron-containing cluster chemistry. Mixed and multiple cluster fusion in platinaborane chemistry and the structure of [(PMe2Ph)2PtB16H17PtB10H11(PMe2Ph)] as determined by synchrotron X-ray diVraction analysis Jonathan Bould,a William Clegg,b John D. Kennedy a and Simon J. Teat b a The School of Chemistry of the University of Leeds, Leeds, UK LS2 9JT b CLRC Daresbury Laboratory, Daresbury, Warrington, Cheshire, UK WA4 4AD and The Department of Chemistry of the University of Newcastle, Newcastle upon Tyne, UK NE1 7RU Multiple cluster fusion between [(PMe2Ph)2PtB8H12] and molten B10H14 gave the 26-boron species [(PMe2Ph)2PtB16- H17PtB10H11(PMe2Ph)] which exhibits quite diVerent structural and assembly features to the 26-boron species [(PMe3)2IrB26H24Ir(CO)(PMe3)2] previously prepared from [(PMe3)2(CO)HIrB8H12] and B10H14.Extension of contiguous boron-containing cluster chemistry beyond icosahedral horizons requires intercluster fusion.1 The structural parameters associated with the intimate subcluster conjunction modes and the reaction parameters associated with the assembly processes, that are both required in order to develop this new area, need to be discovered and explored.Initial experiments show that heteroboranes, particularly, thus far, thiaboranes 2 and metallaboranes,3,4 have flexible architectures that can expedite this exploration.In this regard, they are much more malleable than the binary boranes themselves. Two-cluster fusion is well-recognised, 1 but more complex subcluster assembly is rare.5,6 We have recently demonstrated the principle of intimate three-cluster assembly in an iridaborane system by the thermolytic formation of the 26-boron species [(PMe3)2IrB26H24Ir(CO)(PMe3)2] (structure I) from the reaction of [(PMe3)2(CO)HIrB8H12] with molten B10H14 as solvent. 6 There is interest in extending this triple-cluster fusion principle to other systems.Here we report its extension to a platinaborane system to give a 26-boron species [(PMe2Ph)2PtB16- H17PtB10H11(PMe2Ph)] (Fig. 1 and structure II) that has quite diVerent structural features to that of the iridaborane. Thermolysis of [(PMe2Ph)2PtB8H12] (290 mmol) in molten B10H14 (1400 mmol) at 134 8C for 10 min, followed by repeated TLC and HPLC separation (silica, CH2Cl2–hexane mixtures as liquid phases), gave [(PMe2Ph)2PtB16H17PtB10H11(PMe2Ph)] as an orange crystalline solid (560 mg, 0.50 mmol). The yield was small, but this is to be expected and tolerated in exploratory work delineating new chemistry in an area where both the structural and the assembly principles are unknown.4 Single crystals (from diVusion of hexane through a benzene layer into a dichloromethane solution) also were small (fine needles of cross-section 10 × 10 mm), and required high-intensity synchrotron radiation to attain suYcient diVraction intensity for structural analysis.* The NMR spectroscopic † and FAB mass * Crystal data for [(PMe2Ph)2PtB16H17PtB10H11(PMe2Ph)]?C6H6? 0.5C6H14, M = 1235.1, monoclinic, space group P21/n, a = 14.7903(4), b = 13.0595(3), c = 28.3431(7) Å, b = 90.452(2)8, U = 5474.4(2) Å3, Z = 4, l = 0.6874 Å, m = 5.22 mm21, T = 160 K, R1 = 0.0375 for 9491 reflections and wR2 = 0.0861 for all 11 734 unique reflections.Methods and programs were as described elsewhere.7 CCDC reference number 186/1089.See http://www.rsc.org/suppdata/dt/1998/2777/ for crystallographic files in .cif format. spectrometric data were consistent with the crystallographically determined molecular constitution of the platinaborane. The crystal structure showed benzene and hexane of solvation. Additional products from the reaction include known [(PMe2- Ph)2Pt-anti-B18H20] (900 mg, 1.35 mmol), resulting from twocluster mixed fusion, and known [(PMe2Ph)2PtB10H12 ] (ca. 800 mg, 1.41 mmol), presumably resulting from transmetallation, with small quantities of other large platinaboranes which we hope to be able to report upon in due course. None appears so M M Pt P H Pt P P M M M I II nido 11-vertex III nido 8-vertex nido 11-vertex † NMR data, with chemical shifts d in ppm (CD2Cl2 at 300 K), are as follows. The 11B NMR spectrum (9.5 T, 128 MHz) features a region between ca. 22 and 115 ppm containing fairly broad, overlapping, and hence poorly resolved resonances arising from 18 of the 26 boron atoms, such that the individual peaks cannot yet be unequivocally distinguished.The data are presented in the order: relative 11B intensity [d(11B) {d(1H) of directly attached hydrogen atoms}]: ca. 4B [ca. 115, 112.7, ca. 111 {13.00, 12.89, 12.71, 12.62}], ca. 14B [ 17.1 (probable conjuncto site with no terminal 1H or 31P coupling), ca.17, 16.0, 12.4, 0.0, 21.5, 22.6 {14.04, 13.90, 13.82, 13.60, 13.75, 13.78, 13.00, 12.80, 12.49, 12.35}], 1B [-6.9 {12.08}], 1B [-9.7 {11.99}], 1B [-15.6 {11.24}], 1B [-19.1 {11.54}], 1B [-20.1 {11.44}], 1B [-25.6 {11.56}], 1B [-26.7 {11.64}], 1B [-36.0 {10.85}]; additionally d(1H)- (m-H) 21.14, 21.16 (2H), 21.70(2), 23.72, 28.69 {d, 2J(31P]1H) 71 Hz}, d(1H)(PMe3) 12.28 and 12.23 {overlapping d, each with 2J(31P]1H) 11 Hz}, 11.92 and 11.82 {doublets, each with 2J(31P]1H) 12 Hz}, 11.43 {d, 2J(31P]1H) 11 Hz}, d(31P)[P(8)] 25.57 (br), d(31P)[P(71) and P(72)] 210.3 {1J(195Pt]31P) 3746 Hz} and 211.3 {1J(195Pt]31P) 2660 Hz}, any 2J(31P]31P) coupling not resolved (<ca. 3 Hz).2778 J. Chem. Soc., Dalton Trans., 1998, Pages 2777.2778 far to identify with known products 3 resulting from the thermolytic autofusion of [(PMe2Ph)2PtB8H12] in benzene or toluene solvents. Structurally, [(PMe2Ph)2PtB16H17PtB10H11(PMe2Ph)] at first sight appears as a three-cluster assembly, based on two classical8,9 nido eleven-vertex {MB10} clusters and a nidoshaped ten-vertex {PtB9} cluster.In this work it is formally numbered as such, viz. as 7,7,80-tris(dimethylphenylphosphine)- nido-70-platinaundecaborano-(70:99)-nido-99-platinaundecaborano-( 89,79,39 : 2,5,6)-nido-7-platinadecaborane. The {PtB9} unit is, however, not of true nido character, as discussed below. The two nido eleven-vertex {PtB10} units are conjoined with a common platinum atom. A basic model for this double unit is the [Pt(B10H12)2]22 dianion.8,10 With a formal PMe2Ph hydride-ion replacement, this first subcluster (double-primed in the numbering system) would therefore have no net negative charge associated with it.The neutral two-cluster analogue, [Pt{B10H11(PMe2Ph)2}2] has not been reported. The second {PtB10} unit, i.e. the central (single-primed) subcluster in the molecule, is also of relatively straightforward constitution, with one of the two bridging hydrogen positions of a [Pt(B10H12)2]22 subcluster model being occupied by one bridging hydrogen atom and the other by the elements of what at first sight resembles a fused B(2)B(5)B(6) three-borons-in-common triangular link to the final nido-shaped ten-vertex {5-PtB9} subcluster. Although this latter {PtB9} subcluster approximates to the ten-vertex nido shape, the platinum.boron distance Pt(7)]B(2) is essentially non-bonding at 2.602(6) A (hatched line in structure II).This subcluster is thus better regarded as based on a nido eight-vertex {B8} subcluster which is fused to the central {PtB10} unit with a common two-boron edge B(2)]B(5) (structure III).The platinum centre Pt(7) is then regarded as bridging the B(3)]B(8) position, but it is also linked in turn to the B(6) atom in the central {PtB10} subcluster via an unusual type of Fig. 1 Drawing of the crystallographically determined molecular structure of [(PMe2Ph)2PtB16H17PtB10H11(PMe2Ph)], with P]organyl groups omitted for clarity. Selected interatomic distances (A) are as follows: Pt(7)]B(8) 2.299(7), Pt(7)]B(3) 2.321(6), Pt(7)]B(6) 2.357(6), Pt(7)]H(6) 1.84(6), Pt(7) ? ? ? B(2) 2.602(6), Pt(7)]P(71) 2.2652(15), Pt(6)]P(72) 2.3327(14), Pt(99)]B(49) 2.198(6), Pt(99)]B(20) 2.228(6), Pt(99)]B(39) 2.231(6), Pt(99)]B(30) 2.233(6), Pt(99)]B(2) 2.275(6), Pt(99)]B(110) 2.277(6), Pt(99)]B(109) 2.278(7), Pt(99)]B(80) 2.298(6), P(80)]B(80) 1.939(7) Pt]H]B bridging hydrogen atom.11 This latter link is unsupported by additional platinum bonding to boron atoms adjacent to the B]H (exo) site in question.This B]H (exo) ©¡�¡ Pt interaction may throw useful light on contemporary considerations of ¡®agostic¡� interaction between square-planar transition-element centres and B]H or C]H bonds that happen to be close by in solid-state structures.12 Inter-subcluster bridging links related to this are an increasingly recognised feature in macropolyhedral borane compounds, and the incidence of sulfur-bridged subclusters in [S2B16H14(PPh3)] and [(PPh3)NiS2B16H12(PPh3)],13 and of iridium-bridged subclusters in [(PMe3)2(CO)IrB16H14Ir(CO)(PMe3)2],4 are to be noted in this context.Acknowledgements We thank the EPSRC (UK) and CCLRC (UK) for financial support. References 1 J. D. Kennedy, in Advances in Boron Chemistry, ed. W. Siebert, Royal Society of Chemistry, Cambridge, 1997, pp. 451.462; R. N. Grimes, in Comprehensive Organometallic Chemistry I, eds. G. Wilkinson, F. G. A. Stone and E. Abel, Pergamon, Oxford, 1982, vol. 1, ch. 5.5, pp. 459.542; in Comprehensive Organometallic Chemistry II, eds. G. Wilkinson, F. G. A. Stone and E. Abel, Pergamon, Oxford, 1995, vol. 1, ch. 9, pp. 217.256. 2 P. Kaur, J. Holub, N. P. Rath, J. Bould, L. Barton, B. S¢§ tibr and J. D. Kennedy, Chem. Commun., 1996, 273; T. Jelinek, B. S¢§ tibr, J. D. Kennedy and M. Thornton-Pett, in Advances in Boron Chemistry, ed. W. Siebert, Royal Society of Chemistry, Cambridge, 1997, pp. 426.429; Inorg. Chem. Commun., 1998, 1, 79; T. Jelinek, I.Cisar¢§ova, B. S¢§ tibr, J. D. Kennedy and M. Thornton-Pett, J. Chem. Soc., Dalton Trans., 1998, in the press. 3 M. A. Beckett, J. E. Crook, N. N. Greenwood, J. D. Kennedy, J. Chem. Soc., Dalton Trans., 1986, 1879; M. A. Beckett, N. N. Greenwood, J. D. Kennedy, P. A. Salter and M. Thornton-Pett, J. Chem. Soc., Chem. Commun., 1986, 556. 4 L. Barton, J. Bould, J. D. Kennedy and N. P. Rath, J. Chem. Soc., Dalton Trans., 1996, 3145; J. Bould, W. Clegg, J. D. Kennedy, S. J.Teat and M. Thornton-Pett, J. Chem. Soc., Dalton Trans., 1997, 2005. 5 M. R. Churchill, A. H. Reis, J. N. Francis and M. F. Hawthorne, J Am. Chem. Soc., 1970, 92, 4993. 6 J. Bould, J. D. Kennedy, L. Barton and N. P. Rath, Chem. Commun., 1997, 2405. 7 W. Clegg, M. R. J. Elsegood, S. J. Teat, C. Redshaw and V. C. Gibson, J. Chem. Soc., Dalton Trans., 1998, in the press. 8 L. J. Guggenberger, J. Am. Chem. Soc., 1972, 94, 114. 9 R. E. Williams, Adv. Inorg. Chem. Radiochem., 1976, 18, 67; K. Wade, Adv. Inorg. Chem. Radiochem., 1976, 18, 1. 10 S. A. MacGregor, L. J. Yellowlees and A. J. Welch, Acta Crystallogr., Sect. C, 1990, 46, 1399; see also S. A. MacGregor, J. A. Scanlan, L. J. Yellowlees and A. J. Welch, Acta Crystallogr., Sect. C, 1991, 47, 513. 11 I. Blandford, J. C. JeVery, P. A. Jellis and F. G. A. Stone, Organometallics, 1998, 17, 1402. 12 R. Macias, J. Holub, W. Clegg, M. Thornton-Pett, B. S¢§ tibr and J. D. Kennedy, J. Chem. Soc., Dalton Trans., 1997, 149; K. J. Adams, T. D. McGrath, R. L. Thomas, A. S. Weller and A. J. Welch, J. Organomet. Chem., 1997, 527, 283; J. Bould, P. A. Cooke, U. Dorfler, J. D. Kennedy, L. Barton, N. P. Rath and M. Thornton- Pett, Inorg. Chim. Acta, 1998, in the press. 13 P. Kaur, M. Thornton-Pett, W. Clegg and J. D. Kennedy, J. Chem. Soc., Dalton Trans., 1996, 4155. Received 20th May 1998; Communication 8/03

ISSN:1477-9226    

DOI:10.1039/a803817a

出版商:RSC    

年代:1998

数据来源: RSC

摘要:

DALTON COMMUNICATION J. Chem. Soc., Dalton Trans., 1998, Pages 2779–2781 2779 Synthesis and structure of [BpBut,Pri]2Co: a bis(pyrazolyl)hydroborato cobalt(II) complex with trans [Co ? ? ?H]B] interactions Prasenjit Ghosh, JeVrey B. Bonanno and Gerard Parkin * Department of Chemistry, Columbia University, New York, NY 10027, USA The bis{bis(3-tert-butyl-5-isopropylpyrazolyl)hydroborato}- cobalt(II) complex [BpBut,Pri]2Co has been synthesized by reaction of Tl[BpBut,Pri] with Co(ClO4)2?6H2O; the molecular structure of [BpBut,Pri]2Co is based on a square planar array of pyrazolyl groups with two axial [Co ? ? ?H]B] interactions, in marked contrast to those of other [BpR,R9]2Co derivatives which have tetrahedral structures and are devoid of such interactions.It is well known that transition metal complexes with electronically unsaturated metal centers may supplement their bonding by participating in three-center–two-electron interactions, e.g. [M? ? ?H]B] and [M? ? ?H]C].1 The factors which influence the occurrence of these interactions, however, may be quite complex.2 In this paper, we describe how modifying the substituents in bis{bis(pyrazolyl)hydroborato}cobalt(II) complexes [BpR,R9]2Co3 has a dramatic eVect on the existence of three-center–two-electron [Co ? ? ?H]B] interactions.The bis{bis(3-tert-butyl-5-isopropylpyrazolyl)hydroborato}- cobalt(II) complex [BpBut,Pri]2Co is readily synthesized by treatment of the thallium derivative Tl[BpBut,Pri] with Co(ClO4)2? 6H2O (Scheme 1).Despite the fact that other [BpR,R9]2Co complexes have been synthesized (Table 1),4 the successful isolation of [BpBut,Pri]2Co is noteworthy because the related complex, [BpBut]2Co, which also incorporates tert-butyl substituents in the 3-positions of the pyrazolyl groups, has been reported to be unstable.5 The most interesting feature of [BpBut,Pri]2Co is concerned with its molecular structure (Fig. 1),6 which is strikingly diVerent from those of all other [BpR,R9]2Co complexes.Thus, rather than adopting the tetrahedral array of nitrogen donors typical for other [BpR,R9]2Co derivatives, the cobalt center of Scheme 1 [BpBut,Pri]2Co is co-ordinated in a square planar manner to the four pyrazolyl groups.7 Furthermore, the bonding to cobalt is augmented by three-center–two-electron [Co ? ? ?H]B] interactions, such that the overall co-ordination environment about cobalt is pseudo-octahedral.8 Although [Co ? ? ? H-B] interactions in poly(pyrazolyl)borate complexes have been observed previously, the Co ? ? ? H distance of 1.95 Å in [BpBut,Pri]2Co is significantly shorter than those in other derivatives (Table 1).9,10 Excellent support for the presence of a [Co ? ? ?H]B] interaction is provided by IR spectroscopy. Thus, [BpBut,Pri]2Co exhibits two distinct sets of n(B]H) absorptions in the IR spectrum at 2486 cm21 and 2099/2071 cm21,11,12 of which the lower energy set is assigned to that of the [Co ? ? ?H]B] interaction.Significantly, the n(B]H) absorptions attributed to the [Co ? ? ?H]B] interaction in [BpBut,Pri]2Co are lower in energy than those of other poly(pyrazolyl)borato cobalt complexes (Table 1).9 Whereas [BpBut,Pri]2Co is unique in being the only bis(pyrazolyl) hydroborato cobalt derivative to exhibit a square planar rather than tetrahedral array of nitrogen donors,7 the former co-ordination is common in other transition metal complexes for which the ligand field stabilization energies of a dn configuration (0 < n < 10) may favor a square planar over tetrahedral geometry.13 For example, the chromium, nickel, and copper complexes [BpRR9]2M (M = Cr, Ni or Cu; Table 1) all adopt a square planar array of nitrogen donors.In contrast, the zinc and cadmium complexes [BpRR9]2M (M = Zn or Cd; Table 1) adopt tetrahedral co-ordination since d10 metal centers show no ligand field preference for square planar geometries.The X-ray diVraction and IR spectroscopic studies described above indicate that the Co ? ? ? H bonds in [BpBut,Pri]2Co represent a significant interaction. An important issue, however, is concerned with the extent to which the [Co ? ? ?H]B] Fig. 1 Molecular structure of [BpBut,Pri]2Co. Selected bond lengths (Å) and angles (8): Co]N12 2.131(5), Co]N22 2.138(5); N12]Co]N129 180.0, N12]Co]N22 86.4(2), N129]Co]N22 93.6(2).2780 J. Chem. Soc., Dalton Trans., 1998, Pages 2779–2781 Table 1 Comparison of metrical and IR data for [BpR,R9]2M transition metal complexes [BpBut,Pri]2Co [Bp]2Co [BpMe2]2Co [BpPh]2Co [Bp]2Cr [Bp]2Ni [H2B(pz)(pzMe2)]2Ni [BpMe2]2Ni [BpPh]2Ni [Bp(CF3)2]2Cu [BpPh]2Zn [H2B(pzMe2)(pzPh2)]2Zn [BpMe2]2Zn [Bp(CF3)2]2Zn [Bp]2Cd M[N4] Co-ordination Square planar Tetrahedral Tetrahedral Tetrahedral Square planar Square planar Square planar Square planar Square planar Square planar Tetrahedral Tetrahedral Tetrahedral Tetrahedral Tetrahedral d(M ? ? ? H)/Å 1.95 3.06 3.15 — 3.12 3.05 2.90 2.81 — 2.58 — 3.28 3.09 2.98 2.91 d(M ? ? ? B)/Å 2.59 3.01 3.14 — 3.22 3.14 3.05 2.99 — 2.87 — 3.22 3.17 3.17 3.13 d(M]N)/Å range 2.13–2.14 1.93–1.98 1.991–1.992 — 2.06–2.07 1.89–1.90 1.88–1.89 1.888–1.893 — 1.997–2.004 — 1.98–2.02 2.001–2.007 2.014–2.039 2.181–2.243 n(B]H)/cm21 range 2486–2071 —— 2450–2295 ———— 2424 2572, 2970 2470–2315 2497, 2400 — 2937–2487 — Ref.This work T1 T2 T3 T4 T5 T2 T2 T3 T6 T3 T7 T6 T6 T8 T1 L.J. Guggenberger, C. T. Prewitt, P. Meakin, S. Trofimenko and J. P. Jesson, Inorg. Chem., 1973, 12, 508; T2 ref. 14. T3 ref. 5. T4 P. Dapporto, F. Mani and C. Mealli, Inorg. Chem., 1987, 17, 1323. T5 H. M. Echols and D. Dennis, Acta Crystallogr., Sect. B, 1976, 32, 1627. T6 H. V. R. Dias and J. D. Gordon, Inorg. Chem., 1996, 35, 318. T7 M. V. Capparelli and G. J. Agrifoglio, Crystallogr. Spectrosc. Res., 1992, 6, 651. T8 D. L. Reger, S. S. Mason and A. L. Rheingold, Inorg.Chim. Acta, 1995, 240, 669. interactions are responsible for promoting the structural change from tetrahedral, since it is also possible that the observed [Co ? ? ?H]B] interactions are a result of conformational changes due to interligand steric interactions. Thus, one possible reason for [BpBut,Pri]2Co adopting a square planar rather than tetrahedral array of nitrogen donors is to minimize interligand steric interactions between tert-butyl substituents:14 a square planar geometry allows the 3-tert-butyl groups of each ligand to be located on opposite sides of the [CoN4] plane, whereas a tetrahedral geometry would require the 3-tertbutyl groups on one ligand to mesh with those of the other.It is, therefore, possible that the [Co ? ? ?H]B] interactions in [BpBut,Pri]2Co may be sterically promoted as a result of a conformational change.15 However, since the Co ? ? ? H distance of 1.95 Å is substantially shorter than the M ? ? ? H separations in other square planar [BpRR9]2M derivatives, e.g. 3.05 Å for [Bp]2Ni (Table 1), it is evident that the [Co ? ? ?H]B] interactions are important in influencing the structure of [BpBut,Pri]2Co.16 A further observation which supports this notion is that square planar Co(II) complexes are uncommon, with tetrahedral and octahedral complexes being preferentially favored; as such, there is a clear electronic preference for square planar Co(II) to bind two additional ligands.1 In contrast, square planar coordination for Ni(II) is common,1 such that the nickel centers in square planar [BpRR9]2Ni derivatives do not partake in threecenter –two-electron [Ni ? ? ?H]B] interactions (Table 1).In summary, the molecular structure of [BpBut,Pri]2Co, based on a square planar array of pyrazolyl groups with two axial [Co ? ? ?H]B] interactions, provides a striking contrast with those of other [BpR,R9]2Co derivatives which adopt tetrahedral co-ordination geometries and are devoid of [Co ? ? ?H]B] interactions.Acknowledgements We thank the National Institutes of Health (Grant GM46502) for support of this research. G. P. is the recipient of a Presidential Faculty Fellowship Award (1992–1997). Notes and references 1 F. A. Cotton and G. Wilkinson, Advanced Inorganic Chemistry, John Wiley & Sons, New York, 1998. 2 See, for example: A. Haaland, W. Scherer, K. Ruund, G. S. McGrady, A. J. Downs and O. Swang, J. Am. Chem. Soc., 1998, 120, 3762. 3 Bis- and Tris-(pyrazolyl)hydroborato ligands are represented by the abbreviations [BpR,R9] and [TpR,R9], with the 3- and 5-alkyl substituents listed respectively as superscripts.Additional substituents on boron (other than hydrogen) are represented by a prefix, e.g. [R2Bp]. See: S. Trofimenko, Chem. Rev. 1993, 93, 943; G. Parkin, Adv. Inorg. Chem., 1995, 42, 291. 4 The compounds [BpMe3]2Co4a and [BpEt2]2Co4b have also been synthesized, but IR and metrical data have not been reported.(a) S. Trofimenko, J. Am. Chem. Soc., 1967, 89, 6288; (b) S. Trofimenko, Inorg. Chem., 1970, 9, 2493. 5 S. Trofimenko, J. C. Calabrese and J. S. Thompson, Inorg. Chem., 1987, 26, 1507. 6 [BpBut,Pri]2Co?2CHCl3: C42H74B2Cl6CoN8, M = 984.34, monoclinic, P21/n (no. 14), a = 10.398(2), b = 20.473(3), c = 12.839(2) Å, b = 101.342(8)8, U = 2679.7(6) Å3, Z = 2, m = 0.655 mm21, T = 293 K. Independent reflections 3493, R1 = 0.0722, wR2 = 0.1407 for I > 2s(I). CCDC reference number 186/1088.The hydrogen atoms attached to boron were located and refined isotropically, giving the following bond lengths (Å) and angles (8): B]Hterm 1.13, B]Hbridge 1.15; H]B]H 111 (Found: C, 51.4; H, 7.7; N, 11.5. Calc. for [BpBut,Pri]2Co?2CHCl3: C, 51.3; H, 7.6; N, 11.4%). 7 The compound [RBp]2Co (R = cyclooctane-1,5-diyl) does, however, have a similar co-ordination geometry with [Co ? ? ?H]C] agostic interactions [d(Co ? ? ? H) = 2.16 Å and n(C]H) of 2690 cm21].See: S. Trofimenko, J. C. Calabrese and J. S. Thompson, Angew. Chem., Int. Ed. Engl., 1989, 28, 205; S. Trofimenko, J. C. Calabrese and J. S. Thompson, Inorg. Chem., 1992, 31, 974. 8 The UV/VIS electronic spectrum of [BpBut,Pri]2Co in CHCl3 indicates that the pseudo-octahedral geometry is also maintained in solution. Specifically, [BpBut,Pri]2Co is pale pink and is characterized by absorptions (l/nm and e/M21 cm21) at 310 (2600), 540 (28) and 594 (33). The intensities of these absorptions are much more similar to that of yellow octahedral [Tp]2Co [459 (13), 515 (1.3), 641 (0.1)] than that of violet tetrahedral [Bp]2Co [525 (301), 552 (406), 585 (340)] in the visible range.8a For further comparison, low-spin (S = ��� ) [RBp]2Co (R = cyclooctane-1,5-diyl) is also pink [472 (sh), 500, 562 (sh), 769 nm].8b (a) J. P.Jesson, S. Trofimenko and D. R. Eaton, J. Am. Chem. Soc., 1967, 89, 3148; (b) S. Trofimenko, F. B. Hulsbergen and J. Reedijk, Inorg. Chim.Acta, 1991, 183, 203. 9 For example, [Tp3-Pri,4-Br]Co[BpPh] exhibits Co ? ? ? H and Co ? ? ?B separations of 2.37 and 2.72 Å, respectively, and n(B]H) absorptions in the range 2480–2160 cm21. Likewise, [Tp3-Pri,4-Br]Co[TpPh] exhibits Co ? ? ? H and Co ? ? ? B separations of 2.26 and 2.77 Å, respectively (for the [TpPh] ligand), and n(B]H) absorptions in the range 2490–2175 cm21. See: J. C. Calabrese, P. J. Domaille, J. S. Thompson and S. Trofimenko, Inorg. Chem., 1990, 29, 4429. 10 For further comparison, the mean terminal Co]H bond length for complexes listed in the Cambridge Structural Database is 1.45 Å, with a range of 1.12–1.71 Å.CSD Version 5.14. 3D Search and Research Using the Cambridge Structural Database, F. H. Allen and O. Kennard, Chem. Des. Automat. News, 1993, 8, pp. 1 and 31–37. 11 Assignments for n(B]H) absorptions have been confirmed by studies on the deuterium labelled isotopomer, [D2BpBut,Pri]2Co: thus n(B]D) absorptions are observed at 1859 and 1575 cm21, corresponding to n(B]H)/n(B]D) ratios of 1.34 and 1.33/1.32. 12 The IR spectra of {[BpR,R9]M} derivatives are often surprisingly complicated and frequently exhibit more than two n(B]H) absorptions. For example, Trofimenko has reported six n(B]H) absorptions for Tl[BpBut], at 2410, 2355, 2290, 2275, 2218 and 2178J. Chem. Soc., Dalton Trans., 1998, Pages 2779–2781 2781 cm21 (ref. 5). Trofimenko has also noted that pyrazabole exhibits a complex set of absorptions in both the solid state and solution. See: S.Trofimenko, J. Am. Chem. Soc., 1967, 89, 3165. 13 In fact, the preference for Co(II) to adopt tetrahedral co-ordination is so high that it is cited as forming more tetrahedral complexes than that of any other transition metal. See ref. 1. 14 In this regard, Kokusen has suggested that the subtle conformation changes between [Bp]2Co and [BpMe2]2Co are a result of interligand steric interactions between the 3-methyl substituents.See: H. Kokusen, Y. Sohrin, M. Matsui, Y. Hata and H. Hasegawa, J. Chem. Soc., Dalton Trans., 1996, 195. 15 Trofimenko has also noted that interligand steric interactions have the eVect of forcing six-membered [M(N2)2B] fragments into a “deep boat” conformation which may have the eVect of forcing the B]H groups closer to a metal. See ref. 9. 16 Furthermore, the observation that [RBp]2Co (R = cyclooctane-1,5- diyl), which is devoid of bulky pyrazolyl substituents, adopts a similar octahedral geometry with agostic [Co ? ? ?H]C] interactions (ref. 7), also suggests that the strengths of these three-center–twoelectron [Co ? ? ?H]X] interactions are important in influencing the geometrical preferences of [R02BpR,R9]2Co derivatives. Received 22nd May 1998; Communication 8/03862GJ. Chem. Soc., Dalton Trans., 1998, Pages 2779–2781 2781 cm21 (ref. 5). Trofimenko has also noted that pyrazabole exhibits a complex set of absorptions in both the solid state and solution. See: S. Trofimenko, J. Am. Chem. Soc., 1967, 89, 3165. 13 In fact, the preference for Co(II) to adopt tetrahedral co-ordination is so high that it is cited as forming more tetrahedral complexes than that of any other transition metal. See ref. 1. 14 In this regard, Kokusen has suggested that the subtle conformation changes between [Bp]2Co and [BpMe2]2Co are a result of interligand steric interactions between the 3-methyl substituents. See: H. Kokusen, Y. Sohrin, M. Matsui, Y. Hata and H. Hasegawa, J. Chem. Soc., Dalton Trans., 1996, 195. 15 Trofimenko has also noted that interligand steric interactions have the eVect of forcing six-membered [M(N2)2B] fragments into a “deep boat” conformation which may have the eVect of forcing the B]H groups closer to a metal. See ref. 9. 16 Furthermore, the observation that [RBp]2Co (R = cyclooctane-1,5- diyl), which is devoid of bulky pyrazolyl substituents, adopts a similar octahedral geometry with agostic [Co ? ? ?H]C] interactions (ref. 7), also suggests that the strengths of these three-center–twoelectron [Co ? ? ?H]X] interactions are important in influencing the geometrical preferences of [R02BpR,R9]2Co derivatives. Received 22nd May 1998; Communication 8/

ISSN:1477-9226    

DOI:10.1039/a803862g

出版商:RSC    

年代:1998

数据来源: RSC

摘要:

DALTON COMMUNICATION J. Chem. Soc., Dalton Trans., 1998, Pages 2783–2785 2783 A neutral uranyl dimeric complex and remarkable extraction properties of a 1-acid 3-diethyl amide substituted calix[4]arene ligand Paul D. Beer,*,a Michael G. B. Drew,b Dusan Hesek,a Mark Kan,c Graeme Nicholson,c Philippe Schmitt,a Paul D. Sheen b and Gareth Williams c a Inorganic Chemistry Laboratory, University of Oxford, South Parks Road, Oxford, UK OX1 3QR b Department of Chemistry, University of Reading, Whiteknights, Reading, UK RG6 2AD c AWE, Aldermaston, Reading, UK RG7 4PR The crystal structure of [(UO2)2L2] (L = 1-acid 3-diethyl amide substituted calix[4]arene) has been determined; the uranyl, lanthanide and heavy metal extraction properties of L and of a recyclable polymeric calix[4]arene based resin material are also described.Motivated by major economic and environmental considerations the selective extraction of uranium from sea water and radioactive nuclear waste is a topic of intense current interest.1,2 Many tailor-made ligands (uranophiles) have been designed to perform this operation at the molecular level, some of which have utilised the calixarene framework.2 In order to achieve the desired commercial viability, the ideal receptor must display a remarkable specificity and this may be achieved by taking into account many of the co-ordinating particularities of the uranyl cation (UO2 21).One of these key features is the ability of the UO2 21 ion to accommodate from four to six oxygen donor ligands in an equatorial pseudo-planar arrangement.3–5 We report here the novel crystal structure of a neutral uranyl dimeric complex and the remarkable and eYcient extraction properties of a 1-acid 3-diethyl amide substituted calix[4]arene ligand L.6 The addition of UO2(OAc)2?2H2O to L in acetonitrile† gave crystals of the uranium complex (Scheme 1) suitable for X-ray crystallographic determination.‡ The structure is a centrosymmetric dimer and is illustrated in Fig. 1. Each uranium atom is bonded to the two oxygen atoms in axial positions at 1.78(1) Å to make up the linear uranyl group. In the equatorial plane the uranium is bonded to only four oxygen atoms in an approximate plane (maximum deviation from the UO4 least squares plane is 0.13 Å). The uranium is bonded to one phenolic oxygen at the bottom rim of the calixarene, one amide oxygen atom and to two acid oxygen atoms from diVerent acid groups.These acid groups from the two adjacent calixarenes form together an eight-membered centrosymmetric ring of the type (U]O]C]O] U]O]C]O]). It is interesting that this UO4 equatorial plane is approximately perpendicular to the plane of the methylene groups of † A solution of UO2(OAc)2?2H2O (0.025 g, 5.9 × 1025 mol) in acetonitrile was added to a stirred solution of L (0.05 g, 5.9 × 1025 mol) in acetonitrile. Slow evaporation of the resulting solution gave red crystals (45 mg, 70% yield).Reproducible elemental analytical results could not be obtained because of the presence of variable amounts of included solvent, which is a well documented feature of calixarene chemistry. ‡ Crystal data: [(UO2)2L2]?10MeCN?2MeOH, C126H170N12O20U2, M 2648.80, triclinic, space group P1� , a = 12.350(9), b = 12.765(12), c = 23.67(3) Å a = 94.643(10), b = 97.523(10), g = 110.143(10)8, T = 293 K, U = 3441(6) Å3, Z = 1, m = 2.41 mm21, Dc = 1.278 g cm23, reflections collected 7941, R1 = 0.0711, wR2 = 0.1934 for 5998 data with I > 2s(I).CCDC reference number 186/1086. See http://www.rsc.org/suppdata/ dt/1998/2783/ for crystallographic files in .cif format. the calixarene (angle of intersection 87.28). The four U]O bond distances are all diVerent with the shortest bond to the phenolic oxygen at the bottom rim [U(1)]O(250) 2.162(7) Å]. The amide oxygen distance [U(1)]O(353)] is 2.379(9) Å while distances to the two acid oxygen atoms are [U(1)]O(154)] 2.321(8) and [U(1)]O(153I)] 2.403(9) Å.This irregularity of distance is concomitant with considerable variation in the cis O]U]O angles in the equatorial plane being 108.3(3) for O(154)]U]O(250), 92.5(3) for O(250)]U]O(353), 76.1(3) for O(353)]U]O(153I) and 83.3(3)8 for O(153I)]U]O(154). It seems likely that these variations are due to the steric constraints of the ligand. Computer modelling calculations had demonstrated that the uranyl ion could not fit into the calixarene so that the uranyl axis was coincident with the calixarene axis and the uranium bonded to the four oxygens at the bottom rim in an equatorial plane.It is interesting that the recent crystal structure 2b of a uranyl complex with calix[6]arene also shows dimer formation with two uranyls sandwiched between two calix[6]arenes and the uranyl axes perpendicular to the calixarene axes. In that structure the equatorial planes are made up of two oxygen atoms from each calixarene. Fig. 1 Structure of the [(UO2)2L2] dimer. Selected bond lengths (Å): U(1)]O(250) 2.162(7), U(1)]O(154) 2.321(8), U(1)]O(353) 2.379(9), U(1)]O(153I) 2.403(9). Symmetry element I 2x, 2y, 2z. Scheme 12784 J. Chem. Soc., Dalton Trans., 1998, Pages 2783–2785 Scheme 2 Uranyl extraction experiments and those of a variety of other selected metals were carried out using an aqueous phase of metal nitrate (0.4 × 1023 mol dm23) at diVering pH values with citrate buVer and the extractant dichloromethane solvent phase containing L at a concentration of 9.6 × 1023 mol dm23.After 1 h of rapid mixing of solutions, inductively coupled plasma atomic emission spectral (ICP-AES) analysis was used to determine the concentrations of metal in the respective phases. The extraction results as a function of pH displayed in Fig. 2 clearly show that quantitative extraction of uranyl, lanthanum, lutetium and mercury occurs at pH values > 6. This may be attributed to the calix[4]arene’s unique, lower rim, coordination environment forming neutral metal cation complexes which in combination with the ligand’s lipophilic exterior results in a remarkably eYcient extracting reagent.† Crystal structure determinations have previously shown L to form neutral dimeric and monomeric complexes with lanthanum and lutetium respectively.6 It is noteworthy that washing the organic phase with 1 mol dm23 nitric acid releases the extracted metal and allows calixarene re-use.Indeed recycling the calixarene ligand ten times did not result in any decline in extraction performance. Fig. 3 shows the preliminary extraction results at physiological pH with an extended range of metals which reveal L selectively extracts UO2 21, La31, Lu31, Hg21, Sr21, Pb21, Bi31 quantitatively, and to a lesser extent Y31, Ag1, and is a poor extractant for Group 1, 2 and transition metals. In order to further test the potential commercial viability of Fig. 2 Percentage of UO2 21, La31, Lu31 and Hg21 (0.4 × 1023 mol dm23) extracted by L (9.6 × 1023 mol dm23) in dichloromethane in the presence of citrate (1.2 × 1023 mol dm23) at 293 K as a function of pH. § Preliminary 1H NMR solution binding studies reveal L forms an empirical 1 : 1 stoichiometric complex with UO2 21 in CD2Cl2 solution. Obviously this result does not prove or disprove whether the dimeric 2 : 2 structure is retained in solution. this calixarene based extractant this ligand design was successfully attached to a Tentagel S NH2 resin polymer using the synthetic route shown in Scheme 2.Elemental analysis of the resulting material suggested a loading of 0.24 × 1023 mol of calix[4]arene ligand per gram of resin. Aqueous solutions of uranyl nitrate were passed through the resin material and quantitative extraction was observed over the range pH 4–9. The preliminary results of passing actual low level nuclear waste are also very exciting. The resin selectively extracts uranyl at concentrations as low as ppb levels and on addition of nitric acid uranyl can be removed and the resin recycled without any diminution in extraction eYciency.Modelling calculations show the uranyl dimeric structure is still possible with the polymeric attachment. Studies arin progress to elucidate whether the solution formation of the dimeric uranyl L complex is crucial to the eYcient extraction of the actinide either with L or the polymeric calix[4]arene based resin material.Fig. 3 Percentage extraction of metal ions (0.4 × 1023 mol dm23) in the presence of citrate (1.2 × 1023 mol dm23) in dichloromethane at 293 K. The pH value of the aqueous phase was 7.0 ± 0.1.J. Chem. Soc., Dalton Trans., 1998, Pages 2783–2785 2785 Acknowledgements We thank AWE (plc) for postdoctoral funding (D. H., P. S., P. D. S.) and EPSRC and the University of Reading for funds for the Image Plate System. References 1 P. Thuéry, N.Keller, M. Lange, J.-D. Vignier and M. Nierlich, New J. Chem., 1995, 19, 619 and refs. therein. 2 (a) P. Thuéry and M. Nierlich, J. Inclusion Phenom. Mol. Recognit. Chem., 1997, 27, 13; (b) P. Thuéry, M. Lance and M. Nierlich, Supramol. Chem., 1996, 7, 183; Y. Kubo, S. Maeda, M. Nakamura and S. Tokita, J. Chem. Soc., Chem. Commun., 1994, 1725; T. Nagasaki and S. Shinkai, J. Chem. Soc., Perkin Trans. 2, 1991, 1063; S. Shinkai, H. Koreishi, K. Ueda, T. Arimura and O. Manabe, J.Am. Chem. Soc., 1987, 109, 6371. 3 W. G. Van der Sluys and A. P. Sattleberger, Chem. Rev., 1990, 90, 1027. 4 P. Guilbaud and G. Wipf, J. Phys. Chem., 1993, 97, 5685. 5 P. H. Walton and K. N. Raymond, Inorg. Chim. Acta, 1995, 240, 593; T. S. Franczyk, K. R. Czerwinski and K. N. Raymond, J. Am. Chem. Soc., 1992, 114, 8138. 6 P. D. Beer, M. G. B. Drew, A. Grieve, M. Kan, P. B. Leeson, G. Nicholson, M. I. Ogden and G. Williams, Chem. Commun., 1996, 1117. Received 24th June 1998; Communication 8/04816IJ. Chem.Soc., Dalton Trans., 1998, Pages 2783–2785 2785 Acknowledgements We thank AWE (plc) for postdoctoral funding (D. H., P. S., P. D. S.) and EPSRC and the University of Reading for funds for the Image Plate System. References 1 P. Thuéry, N. Keller, M. Lange, J.-D. Vignier and M. Nierlich, New J. Chem., 1995, 19, 619 and refs. therein. 2 (a) P. Thuéry and M. Nierlich, J. Inclusion Phenom. Mol. Recognit. Chem., 1997, 27, 13; (b) P. Thuéry, M. Lance and M. Nierlich, Supramol. Chem., 1996, 7, 183; Y. Kubo, S. Maeda, M. Nakamura and S. Tokita, J. Chem. Soc., Chem. Commun., 1994, 1725; T. Nagasaki and S. Shinkai, J. Chem. Soc., Perkin Trans. 2, 1991, 1063; S. Shinkai, H. Koreishi, K. Ueda, T. Arimura and O. Manabe, J. Am. Chem. Soc., 1987, 109, 6371. 3 W. G. Van der Sluys and A. P. Sattleberger, Chem. Rev., 1990, 90, 1027. 4 P. Guilbaud and G. Wipf, J. Phys. Chem., 1993, 97, 5685. 5 P. H. Walton and K. N. Raymond, Inorg. Chim. Acta, 1995, 240, 593; T. S. Franczyk, K. R. Czerwinski and K. N. Raymond, J. Am. Chem. Soc., 1992, 114, 8138. 6 P. D. Beer, M. G. B. Drew, A. Grieve, M. Kan, P. B. Leeson, G. Nicholson, M. I. Ogden and G. Williams, Chem. Commun., 1996, 1117. Received 24th June 1998; Communication 8/04816I

ISSN:1477-9226    

DOI:10.1039/a804816i

出版商:RSC    

年代:1998

数据来源: RSC

摘要:

DALTON FULL PAPER J. Chem. Soc., Dalton Trans., 1998, Pages 2787–2791 2787 Molecular structure of monomeric scandium trichloride by gas electron diVraction and density functional theory calculations on ScCl3 and Sc2Cl6† Arne Haaland,*,a Kjell-Gunnar Martinsen,a Dmitry J. Shorokhov,a Georgiy V. Girichev *,b and Vasili I. Sokolov b a Department of Chemistry, University of Oslo, Box 1033 Blindern, N-0315 Oslo, Norway b Department of Physics, State Academy of Chemistry and Technology, Engels 7, 153460 Ivanovo, Russia The molecular structures of monomeric and dimeric scandium trichloride were optimised by DFT calculations with basis sets of valence shell TZ 1 P quality, and the molecular force fields and normal vibrational modes calculated.Optimisation of ScCl3 yielded an equilibrium geometry of D3h symmetry and bond distance Sc]Cl 228.5 pm. Optimisation of a model of the dimer with double Cl bridges indicated an equilibrium geometry of D2h symmetry, the terminal and bridging bond distances Sc]Clt 226.0 and Sc]Clb 247.5 pm, and the valence angles Clt]Sc]Clt 114.9 and Clb]Sc]Clb 86.68.Synchronous gas electron diVraction (GED) and mass spectrometric (MS) data were recorded with the eVusion cell kept at 900 ± 10 K. The gas was found to consist of 92 ± 2% monomer and 8 ± 2% dimer. Least-squares refinement of a trigonal pyramidal (C3v) model of the monomer yielded the bond distance rg(Sc]Cl) = 229.1(3) pm and a valence angle a Cl]Sc]Cl 119.8(5)8.The concentration of the dimer was too low for the GED data to give accurate structure parameters for this species. Bond energies for both monomer and dimer were calculated from thermochemical data in the literature and compared to corresponding energies in MCl3 and M2Cl6, M = Al, Ga or In. The monomeric scandium trihalides, ScX3, have been the subject of several studies aiming towards the establishment of the molecular structure, i.e. the determination of the Sc]X bond distance and the molecular shape; is the equilibrium geometry trigonal pyramidal, symmetry C3v, or planar, symmetry D3h? The first investigation of ScF3 by gas electron diVraction (GED) dates back to 1961.1 DiVraction data recorded at an unspecified temperature were found to be consistent with a monomer concentration of 100% and D3h symmetry.This, of course, is the symmetry indicated by a spherical ion model, by the VSEPR model. It is also consistent with a hybridisation model since sd2 hybrid orbitals formed from the 4s, 3dxy and 3dx2 2 y2 atomic orbitals on Sc have major lobes pointing in the appropriate directions.(Hybridisation of the 4s, 3dxz and 3dyz orbitals would, however, yield hybrids favorable for a trigonal pyramidal co-ordination geometry.) Three reports on the IR absorption spectra of ScF3 in rare gas matrices were published in the 60s or early 70s.2–4 Since the symmetric Sc]F stretching frequency (n1) could not be found, it was concluded that the molecule must be planar or near-planar.The electric deflection of molecular beams, on the other hand, indicated a polar, i.e. non-planar structure,5 and a second investigation by GED yielded a F]Sc]F valence angle of 110(2.5)8.6 Finally a third, careful analysis of GED data recorded with a nozzle temperature of 1750 K yielded a Sc]F bond distance of rg = 184.7(2) pm and a non-bonded F ? ? ? F distance which, after correction for thermal vibration, diVered from that calculated for a planar model by 0.0(15) pm;7 the molecule is clearly planar or very nearly so.Equilibrium geometries of D3h † Supplementary data available: experimental conditions for the synchronous GED/MS. For direct electronic access see http://www.rsc.org/ suppdata/dt/1998/2787/, otherwise available from BLDSC (No. SUP 57406, 2 pp.) or the RSC Library. See Instructions for Authors, 1998, Issue 1 (http://www.rsc.org/dalton). symmetry are also indicated by ab initio calculations at the CISD(Q) level 8 and by DFT calculations at the same level as those described below for ScCl3.9 It would seem that the question about the equilibrium structure of ScF3 has been settled for the time being! The gas-phase IR spectra of monomeric ScCl3, ScBr3 and ScI3 have been recorded by Selivanov.10 No symmetric Sc]X stretching frequencies (n1) could be assigned.The IR spectrum of matrix-isolated ScBr3 has also been reported; n2, n3 and n4 could be assigned, but n1 was not found.11 Neither the trichloride nor the tribromide appears to have been studied by GED up to the present, but Ezhov et al.12 have recently published the results of a GED study of gaseous ScI3 at 1050 K.The molecular beam was found to contain both monomeric and dimeric species with mole fractions of 21(3) and 79(3)% respectively. Least-squares refinement of the molecular structures of both monomer and dimer yielded a monomer bond distance of 262(1) pm and a monomer valence angle of 117(2)8: the concentration of the monomer is obviously too small to allow a distinction to be made between planar and pyramidal models.In this article we report the results of density functional theory (DFT) calculations on both monomeric and dimeric scandium trichloride and a GED investigation which shows that monomeric ScCl3 is trigonal planar or very nearly so. Density Functional Theory Calculations The original plan was to optimise the molecular structures of both ScCl3 and Sc2Cl6 by DFT calculations using the program system GAUSSIAN 9413 with the Becke exchange14 and the Perdew–Wang correlation functional 15 (BPW 91).Optimisation of a trigonal planar (D3h) model of ScCl3 with the standard eVective core potential (ECP) basis LanL2DZ 13 converged to a bond distance of 229.1 pm. The dimer was assumed to have a diborane-like structure with two bridging chlorine atoms (see2788 J. Chem. Soc., Dalton Trans., 1998, Pages 2787–2791 Table 1 Structure parameters of ScCl3 and Sc2Cl6 obtained by density functional theory calculations or gas electron diVraction.Interatomic distances (r), root-mean-square vibrational amplitudes (l), perpendicular amplitude correction coeYcients (K) and shrinkages (d) in pm, angles in 8 a DFTb GED re l K rg l ScCl3 Mole fraction 93(3)% Interatomic distances Sc]Cl Cl ? ? ? Cl 228.5 395.7 7.8 23.6 3.6 0.6 229.1(3) 390.8(11) 7.6(2) d 23.3(10) d Shrinkage d(Cl ? ? ? Cl) c Valence angle Cl]Sc]Cl 6.1 e 120.0 6.0(16) a 119.8(5) Sc2Cl6 Mole fraction 7(3)% Interatomic distances Sc]Clt Sc]Clb Clb ? ? ? Clb Sc ? ? ? Sc Clb ? ? ? Clt Clt ? ? ? Clt Sc ? ? ? Clt Clt ? ? ? Clt Clt ? ? ? Clt Valence angles Clt]Sc]Clt Clb]Sc]Clb Sc]Clb]Sc R factor e 226.0 247.5 339.5 360.1 394.9 380.4 517.8 603.0 713.1 e 114.9 86.6 93.4 7.7 11.7 15.5 14.0 37.2 25.5 64.4 165.2 39.0 27.6 8.8 3.5 2.1 26.0 42.4 8.3 4.3 1.1 227.5(10) d 246(2) 325(6) 349(4) 404(2) 412(2) 501(3) 543(3) 691(3) a [114.9] 86(2) 94(2) 0.053 7.6(2) d [11] [15.5] [14.0] 37.6(10) d [25.5] [45.0] d [51.0] d [39] a Estimated standard deviations in parentheses in units of the last digit.Non-refined parameters in square brackets. b The calculations on ScCl3 have been carried out with the ADF program and the TZ 1 P basis set IV, that on Sc2Cl6 with GAUSSIAN 94 and a 6-311G* basis set. See comment in Density Functional Theory Calculations. c The shrinkage is defined as d(Cl ? ? ? Cl) = ÷3 rg(Sc]Cl) 2 rg(Cl ? ? ? Cl).d See comment in Structure refinements. e ÷[Sw(Iobs 2 Icalc)2/Sw(Iobs)2]. sketch in Contents). Optimisation of a model of D2h symmetry with the LanL2DZ basis yielded the terminal and bridging bond distances Sc]Ct 227.5 and Sc]Clb 251.2 pm. Optimisation of a C2v model of the dimer (i.e. a model in which the central Sc2Cl2 ring is non-planar) with the standard all-electron (AE) basis set 6-311G*13 converged to D2h symmetry (planar Sc2Cl2 ring).Interatomic distances and valence angles are listed in Table 1. The normal vibrational modes are listed in Table 2. The molecular force field was transferred to the program ASYM 40 for calculation of root-mean-square vibrational amplitudes, l, and perpendicular amplitude correction coeYcients K16 (see Table 1). After several attempts to optimise the structure of the monomer at the BPW91/6-311G* level had failed to converge, we turned to the Amsterdam Density Functional (ADF) program.17 Calculations were carried out with the Vosko–Wilk– Nusair parametrisation,18 the gradient correction of Becke 14 for exchange and of Perdew19 for correlation.A standard basis set of TZ 1 P quality (IV) was used,17 with the atomic cores of Sc and Cl up to and including the 2p AOs frozen in their atomic shape. Structure optimisation of a C3v model of ScCl3 now converged nicely to yield a structure of D3h symmetry. The vibrational modes are listed in Table 2, the bond distance, root mean square (r.m.s.) vibrational amplitudes and perpendicular amplitude correction coeYcients are listed in Table 1.Experimental A sample of ScCl3?xH2O with a stated purity of 99.99% was purchased from Aldrich Chemical Company. The anhydrous trichloride was obtained by heating the sample under reflux with thionyl chloride as described in ref. 20. Gas electron diVraction and mass spectrometry Synchronous MS and GED experiments were carried out on the modified EMR-100/ApdM-1 unit in Ivanovo.The nickel oven containing the sample was kept at the lowest possible temperature at which suYcient vaporisation took place, about 900 K, corresponding to a vapour pressure of about 0.025 Torr (Torr ª 133 Pa).21 The ratio of evaporation surface to the nozzle orifice was approximately 400. The length to diameter ratio of the diVusion nozzle was optimised to keep equilibrium concentrations of the monomer and dimer in the vapor and a negligibly small scattering volume.22 Other experimental Table 2 Normal mode frequencies (cm21) of ScCl3 and Sc2Cl6 obtained by DFT calculations Symmetry Mode w Symmetry Mode w ScCl3 (D3h) A1 E 13 341 86 E A2 24 468 79 Sc2Cl6 (D2h) Ag Ag B1g B2g B3g B1u B1u B2u B3u 13579 11 13 15 17 438 151 237 465 73 474 12 56 273 Ag Ag B1g B2g Au B1u B2u B3u B3u 2468 10 12 14 16 18 287 71 81 58 36 109 319 412 94J.Chem. Soc., Dalton Trans., 1998, Pages 2787–2791 2789 conditions are summarised in SUP 57406.A portion of the mass spectrum is given in Fig. 1. For analysis of the gas composition we assumed that the ions ScCln 1, n = 1 to 3, are formed from the monomer, that the ions Sc2Cln 1, n = 1 to 5, are formed from the dimer, and that the ratio of the ionisation cross-sections of dimer to monomer is equal to 2. Atomic electron scattering factors were taken from ref. 23 and backgrounds were drawn as smooth least-squares adjusted polynomials to the diVerence between experimental and calculated molecular intensities.Structure refinements Structure refinement of the monomer was based on a geometrically consistent ra model of C3v symmetry. The mole fraction of monomer in the molecular beam, the Sc]Cl bond distance, the non-bonded Cl ? ? ? Cl distance and the Sc]Cl and Cl ? ? ? Cl r.m.s. vibrational amplitudes were refined as independent parameters. The asymmetry constant of the Sc]Cl bond distance (and of the terminal Sc]Cl distance in the dimer) was estimated from k = (1/6)l4÷[8p2cwexem/h].Molecular constants taken from the scandium monochloride molecule24 yielded k = 7.24 × 1026 pm3. Structure refinement of the dimer was based on a geometrically consistent ra model of D2h symmetry. Such a model is characterised by four independent structure parameters, e.g. the terminal and bridging Sc]Clt and Sc]Clb bond distances and the valence angles a Clt]Sc]Clt and Clb]Sc]Clb. Since the amount of dimer in the molecular beam was less than 10% we were unable to refine these four parameters without divergence, and the valence angle a Clt]Sc]Clt was fixed at the value obtained by the DFT calculations.The diVerence between the Fig. 1 A portion of the mass spectrum of scandium trichloride under the conditions of the gas electron diVraction experiment. The ionising potential is 50 V Fig. 2 Calculated (full lines) and experimental (squares) modified molecular intensity curves of ScCl3 with diVerence curves shown below Sc]Cl bond distance in the monomer and the terminal distance in the dimer, rg(Sc]Cl) 2 rg(Sc]Clt), was fixed at the value obtained by DFT calculations with LanL2DZ basis (1.6 pm).(The estimated standard deviation obtained for the Sc]Ct bond distance was expanded from 0.3 to 1.0 pm to include the uncertainty due to this constraint.) The vibrational amplitudes of the two bond distances were refined with a constant diVerence, as were the amplitudes of the Cl ? ? ? Cl distance in the monomer and of the Clb ? ? ? Clt distance in the dimer which turned out to be very similar. The calculated vibrational amplitudes of the non-bonded Sc ? ? ? Clt distance at about 500 pm and the non-bonded Clt ? ? ? Clt distance at about 540 pm were 64 and 165 pm respectively.These amplitudes were varied stepwise to minimise the square-error sum. The best fit was obtained for the values 45 and 51 pm respectively. Other amplitudes were fixed at their calculated values.The structures were refined by a modified version of the program KCED 25 originally written by H. M. Seip. The refinements converged to yield the best values listed in Table 1. Since the refinements were carried out with diagonal weight matrices the listed estimated standard deviations have been multiplied by a factor of 2.0 to include the uncertainty due to data correlation and expanded to include an estimated scale uncertainty of 0.1%. Experimental and calculated molecular intensity curves are compared in Fig. 2, radial distribution curves in Fig. 3. Results and Discussion The composition of the molecular beam The mass spectra recorded simultaneously with the GED diagrams indicated that the mole fractions of monomers and dimers in the molecular beam were 92 ± 2 and 8 ± 2% respectively, while the amount of trimer or higher species was negligible. These mole fractions are in good agreement with the less accurate values obtained by analysis of the electron diVraction data, 93(3) and 7(3)% respectively.The high concentration of the monomer allows an accurate determination of its molecular structure, while the concentration of the dimer was too low for the GED diagrams to contain much information about the molecular structure of Sc2Cl6. The molecular structure of ScCl3 Least-squares structure refinement of a molecular model of C3v symmetry to the GED data yielded a Cl]Sc]Cl valence angle Fig. 3 Calculated (full line) and experimental (squares) radial distribution curves of a mixture of ScCl3 (92%) and Sc2Cl6 (8%).Artificial damping constant k = 25 pm2. The two peaks at about 230 and 390 pm represent the Sc]Cl bond distance and the non-bonded Cl ? ? ? Cl distance in the monomer. Below: diVerence curve2790 J. Chem. Soc., Dalton Trans., 1998, Pages 2787–2791 of 119.8(5)8 while structure optimisation by DFT calculations with an all-electron basis of TZ 1 P quality yielded an equilibrium structure of D3h symmetry; calculations and experiment agree that the molecule is planar or very nearly so.A planar equilibrium structure is also indicated by the gas phase IR spectra since the symmetric Sc]Cl stretching mode (n1) could not be detected.10 The calculated Sc]Cl bond distance is 228.5 pm in good agreement with experimental (rg) distance of 229.1(3) pm. Before going on to discuss the molecular structures of the monomeric trichlorides of the heavier Group 3 metals, yttrium and lanthanum, we pause to note that while the bond distances in the Group 13 trichlorides MCl3, M = Al, Ga or In, are 6 to 11 pm shorter than the bond distance in the monochlorides MCl,25 the bond distance in ScCl3 is 6 pm longer than in ScCl, 222.9 pm.26 An early GED investigation of YCl3 indicated that the equilibrium structure is planar.27 More recently, Konings and Booij 28 have recorded the infrared spectrum of gaseous YCl3 and assigned the four normal modes under the assumption that the structure is pyramidal.This assignment has been challenged by Marsden and Smart29 who optimised the structure at the MP2 level with a ECP basis of DZ quality and obtained an equilibrium structure of D3h symmetry. Finally, a reinvestigation by a combination of GED and DFT calculations has shown that the structure is indeed planar or very nearly so.30 The equilibrium structure of monomeric LaCl3 is still not definitely established.Two relatively recent investigations of LaCl3 by GED led to the conclusion that the equilibrium geometry is pyramidal,31,32 while quantum chemical calculations at various levels indicate that it is planar.33–36 The molecular structure of Sc2Cl6 Density functional theory calculations on the dimer with a 6-311G* basis converged to a model of D2h symmetry. Bond distances and valence angles are listed in Table 1. Attempts to refine the four independent structure parameters characterising a D2h model, viz.the terminal and bridging Sc]Clt and Sc]Clb bond distances and the valence angles a Clt]Sc]Clt and Clb]Sc]Clb, to the GED data failed to converge. The diVerence between the Sc]Cl bond distance in the monomer and the terminal distance in the dimer, rg(Sc]Cl) 2 rg- (Sc]Clt), was therefore fixed at the value obtained by DFT calculations with LanL2DZ basis (1.6 pm), and Clt]Sc]Clt at the value obtained by the all-electron calculations on the dimer. The best values obtained for the two structure parameters that could be refined without constraints rb(Sc]Clb) = 246(2) pm and a Clb]Sc]Clb 86(2)8 are not significantly diVerent from their calculated values.In the following we base our discussion of the dimer on the calculated structure parameters. The compound Sc2Cl6 appears to be similar to the Group 13 analogues M2Cl6, M = Al,37 Ga37,38 or In,38 insofar as the bridging M]Cl distance is about 20 pm longer than the terminal and the Clb]Sc]Clb angle is close to 908, but to diVer from the Group 13 analogues by having a Clt]Sc]Clt less than 1208: Clt]Al]Clt 123.6(16),37 Clt]Ga]Clt 124.7(18) 37 and Clt]In]Clt ª 1308.38 The crystal structure of ScCl3 is constructed from ScCl6 octahedra, each Cl atom bridges two Sc atoms at a distance, 252 pm, about 5 pm longer than the Sc]Clb distance in the gaseous dimer.39 Bond energies The mean bond energy of monomeric ScCl3 at 298 K may be calculated from the standard enthalpy of formation:24 MBE(ScCl3) = {DH8f[Sc(g)]13DH8f[Cl(g)]2DH8f[ScCl3(g)]}/3 = 478(3) kJ mol21.Similarly the mean bond energy of gaseous LaCl3 calculated from the standard enthalpy of formation 40 is found to be 509 kJ mol21. Both the Sc]Cl and La]Cl MBEs are larger than those of the Group 13 analogues, MCl3, M = B, Al, Ga or In, which range from 456 to 327 kJ mol21.25 While the MBEs of the Group 13 trichlorides decrease as the group is descended, those of the Group 3 trichlorides appear to increase. Since the terminal Sc]Cl bond distance in Sc2Cl6 is very close to the bond distance of the monomer, we assume the bond energies to be equal; BE(Sc]Clt) = MBE(ScCl3).The mean energy of the bridge bonds may then be estimated from the dimerisation enthalpy,41 DH8d = 2199 kJ mol21 where DH8d = 2 BE(Sc]Clt) 2 4 BE(Sc]Clb) or BE(Sc]Clb) = 289 kJ mol21. The M]Clb bond is thus stronger in Sc2Cl6 than in Al2Cl6, Ga2Cl6 or In2Cl6.38 The ratio between terminal and bridging bond energies is however 1.7 ± 0.1 for both Sc and the Group 13 metals.38 Acknowledgements We are grateful to the Russian Basic Research Foundation for financial support (Grant 95-03-09852a) and to the Research Council of Norway (Programme for Supercomputing) for a grant of computing time.References 1 P. A. Akishin and V. A. Naumov, J. Struct. Chem., 1961, 2, 1. 2 D. McDonald, jun. and W. Weltner, jun., J. Phys. Chem., 1966, 70, 3293. 3 R. H. Hauge, J.W. Hastie and J. L. Margrave, J. Less Common Met., 1971, 23, 359. 4 J. W. Hastie, R. H. Hauge and J. L. Margrave, J. Less Common Met., 1975, 39, 309. 5 E. W. Kaiser, W. E. Falconer and W. Klemperer, J. Chem. Phys., 1972, 56, 5392. 6 N. I. Giricheva, E. Z. Zasorin, G. V. Girichev, K. S. Krasnov and V. P. Spiridonov, J. Struct. Chem., 1976, 17, 686. 7 E. Z. Zasorin, A. A. Ivanov, L. I. Ermolaeva and V. P. Spiridonov, Zh. Fiz. Khim., 1989, 63, 669. 8 V. G. Solomonik, V. V. Sliznev and N.B. Balabanov, Zh. Neorg. Khim., 1995, 40, 2024. 9 A. Haaland and K.-G. Martinsen, unpublished work. 10 G. K. Selivanov, Ph. D. Thesis, Moscow State University, 1972. 11 P. A. Perov, S. V. Nedyak and A. A. Maltsev, Vestn. Mosk. Univ., Ser. 2: Khim., 1975, 16, 281. 12 Yu. Ezhov, S. A. Komarov and V. G. Sevastyanov, Zh. Strukt. Khim., 1997, 38, 489. 13 GAUSSIAN 94, revision D.2, M. J. Frisch, G. W. Trucks, H. B. Schlegel, P. M. W. Gill, B. G. Johnson, M. A. Robb, J.R. Cheeseman, T. Keith, G. A. Peterson, J. A. Montgomery, K. Raghavachari, M. A. Al-Laham, V. G. Zakrzewski, J. V. Ortiz, J. B. Foresman, J. Cioslowski, B. B. Stefanov, A. Nanayakkara, M. Challacombe, C. Y. Peng, P. Y. Ayala, W. Chen, M. W. Wong, J. L. Andres, E. S. Repogle, R. Gomperts, R. L. Martin, D. J. Fox, J. S. Binkley, D. J. Defrees, J. Baker, J. P. Stewart, M. Head-Gordon, C. Gonzales and J. A. Pople, Gaussian, Inc., Pittsburgh, PA, 1995. 14 A. D. Becke, Phys. Rev.A, 1988, 38, 3098. 15 J. P. Perdew and Y. Wang, Phys. Rev. B, 1992, 45, 13 244. 16 L. Hedberg and I. M. Mills, J. Mol. Spectrosc., 1993, 160, 117. 17 Amsterdam Density Functional (ADF), Release 2.1.1, Vrije Universiteit, Amsterdam, 1995; E. J. Baerends, D. E. Ellis and P. Ros, Chem. Phys., 1973, 2, 41; E. J. Baerends, Ph. D. Thesis, Vrije Universiteit, Amsterdam, 1975; P. M. Boerigter, G. te Velde and E. J. Baerends, Int. J. Quantum Chem., 1988, 33, 87. 18 S. H. Vosko, L. Wilk and M.Nusair, Can. J. Phys., 1980, 58, 1200. 19 J. P. Perdew, Phys. Rev. B, 1986, 33, 8822. 20 J. H. Freeman and M. L. Smith, J. Inorg. Nucl. Chem., 1958, 7, 224. 21 Yu. B. Patrikeev, V. A. Morozova, G. P. Dudchik, O. G. Polyachonok and G. I. Novikov, Zh. Fiz. Khim., 1973, 47, 266. 22 G. V. Girichev, S. A. Shlykov and S. B. Lapshina, Zh. Fiz. Khim., 1990, 64, 899. 23 R. A. Bonham and L. Schäfer, in International Tables for X-Ray Crystallography, eds. J. A. Ibers and W. C. Hamilton, Kynoch Press, Birmingham, 1974, vol. 4.J. Chem. Soc., Dalton Trans., 1998, Pages 2787–2791 2791 24 Molecular Constants of Inorganic Compounds, ed. K. S. Krasnov, Chemistry, Leningrad, 1977. 25 A. Haaland, A. Hammel, K.-G. Martinsen, J. Tremmel and H. V. Volden, J. Chem. Soc., Dalton Trans., 1992, 2209. 26 K. P. Huber and G. Herzberg, Molecular Spectra and Molecular Structure, Van Nostrand, New York, 1979, vol. IV. 27 P. A. Akishin, V. A. Naumov and V. M. Tatevskii, Kristallografiya, 1959, 4, 194. 28 R. J. M. Konings and A. S. Booij, J. Mol. Struct., 1992, 271, 183. 29 C. J. Marsden and B. A. Smart, Aust. J. Chem., 1993, 46, 749. 30 J. Molnár, M. Kolonits, C. J. Marsden and M. Hargittai, manuscript in preparation. 31 T. G. Danilova, G. V. Girichev, N. I. Giricheva, K. S. Krasnov and E. Z. Zasorin, Izv. Vyssh. Uchen. Zaved. Khim. Khim. Tekhnol., 1979, 22, 101. 32 V. P. Spiridonov, A. G. Gershikov and V. S. Lyutsarev, J. Mol. Struct., 1990, 221, 79. 33 L. L. Lohr and Y. Q. Jia, Inorg. Chim. Acta, 1985, 119, 95. 34 M. Dolg, H. Stoll and H. Preuss, J. Mol. Struct. (THEOCHEM), 1991, 235, 67. 35 S. Di Bella, G. Lanza and I. L. Fragalà, Chem. Phys. Lett., 1993, 214, 598. 36 G. Lanza and I. L. Fragalà, Chem. Phys. Lett., 1996, 255, 341. 37 Q. Shen, Ph. D. Thesis, Oregon State University, Corvallis, 1974. 38 G. V. Girichev, N. I. Giricheva, V. A. Titov and T. P. Chusova, J. Struct. Chem., 1992, 33, 362. 39 H. Fjellvåg and P. Karen, Acta Chem.Scand., 1994, 48, 294. 40 I. Barin, Thermochemical Data of Pure Substances, VCH, Weinheim, 1993, Parts I and II. 41 K. Wagner and H. Schäfer, Z. Anorg. Allg. Chem., 1977, 430, 43. Received 5th May 1998; Paper 8/03339KJ. Chem. Soc., Dalton Trans., 1998, Pages 2787–2791 2791 24 Molecular Constants of Inorganic Compounds, ed. K. S. Krasnov, Chemistry, Leningrad, 1977. 25 A. Haaland, A. Hammel, K.-G. Martinsen, J. Tremmel and H. V. Volden, J. Chem. Soc., Dalton Trans., 1992, 2209. 26 K. P. Huber and G. Herzberg, Molecular Spectra and Molecular Structure, Van Nostrand, New York, 1979, vol. IV. 27 P. A. Akishin, V. A. Naumov and V. M. Tatevskii, Kristallografiya, 1959, 4, 194. 28 R. J. M. Konings and A. S. Booij, J. Mol. Struct., 1992, 271, 183. 29 C. J. Marsden and B. A. Smart, Aust. J. Chem., 1993, 46, 749. 30 J. Molnár, M. Kolonits, C. J. Marsden and M. Hargittai, manuscript in preparation. 31 T. G. Danilova, G. V. Girichev, N. I. Giricheva, K. S. Krasnov and E. Z. Zasorin, Izv. Vyssh. Uchen. Zaved. Khim. Khim. Tekhnol., 1979, 22, 101. 32 V. P. Spiridonov, A. G. Gershikov and V. S. Lyutsarev, J. Mol. Struct., 1990, 221, 79. 33 L. L. Lohr and Y. Q. Jia, Inorg. Chim. Acta, 1985, 119, 95. 34 M. Dolg, H. Stoll and H. Preuss, J. Mol. Struct. (THEOCHEM), 1991, 235, 67. 35 S. Di Bella, G. Lanza and I. L. Fragalà, Chem. Phys. Lett., 1993, 214, 598. 36 G. Lanza and I. L. Fragalà, Chem. Phys. Lett., 1996, 255, 341. 37 Q. Shen, Ph. D. Thesis, Oregon State University, Corvallis, 1974. 38 G. V. Girichev, N. I. Giricheva, V. A. Titov and T. P. Chusova, J. Struct. Chem., 1992, 33, 362. 39 H. Fjellvåg and P. Karen, Acta Chem. Scand., 1994, 48, 294. 40 I. Barin, Thermochemical Data of Pure Substances, VCH, Weinheim, 1993, Parts I and II. 41 K. Wagner and H. Schäfer, Z. Anorg. Allg. Chem., 1977, 430, 43. Received 5th May 1998; Paper 8/03339K

ISSN:1477-9226    

DOI:10.1039/a803339k

出版商:RSC    

年代:1998

数据来源: RSC

摘要:

DALTON FULL PAPER J. Chem. Soc., Dalton Trans., 1998, Pages 2793–2798 2793 NMR Studies of metal complexes and DNA binding of the quinone-containing antibiotic streptonigrin Xiangdong Wei and Li-June Ming *,† Department of Chemistry and Institute for Biomolecular Science, University of South Florida, Tampa, Florida 33620-5250, USA Optical and 1H NMR techniques have been applied to the study of a few metal complexes (Co21, Fe21, and Yb31) of the antitumor antibiotic streptonigrin (SN) produced by Streptomyces flocculus to elucidate the structure of the complexes.The hyperfine-shifted 1H NMR signals of these paramagnetic complexes were fully assigned by means of relaxation and two-dimensional NMR techniques. These studies revealed that SN binds transition metal and lanthanide ions and forms stable metal–drug complexes, with the metal located at the quinolinequinone– picolinate site. This configuration requires a ª1808 twist of the C2]C29 bond in the crystal structure of the drug.The hyperfine-shifted 1H NMR signals of the Co21–SN complex are significantly changed upon addition of calf thymus DNA or poly[dA-dT], indicating direct binding of Co21–SN complex with DNA. Streptonigrin (SN, also known as rufochromomycin and bruneomycin) is a metal-dependent quinone-containing antibiotic produced by Streptomyces flocculus 1 (Fig. 1 2). This antibiotic has been shown to exhibit active inhibition toward several tumors and cancers (e.g.lymphoma, melanoma, and breast and cervix cancers) as well as viruses in some early in vitro and clinical observations.3,4 However, the high toxicity and serious side eVects of this drug reduce its clinical value, and limit its use only as an experimental antitumor agent.3,4 Nevertheless, because of its antitumor potency and unique structure, SN has served as a lead drug molecule for chemical modifi- cation and synthesis in order to correlate specific structure features with the biological activity of the molecule.5 Since SN contains a quinone moiety, it may share some common mechanistic characteristics with other quinone-containing antibiotics such as the anthracyclines in inhibition of cancer growth.Two mechanisms for this action have been proposed:6 (1) by way of interference with cell respiration and (2) through disruption of cell replication and transcription. A key step in this action is reflected by the induction of severe irreversible damage to DNA and RNA in vitro and in vivo in the presence of reducing agents.6,7 Streptonigrin is able to bind several diVerent metal ions, and requires metal binding for full antibiotic and antitumor activity. 6,8 The transition metal ions Cu21 and Fe21 have been known to accelerate SN-mediated DNA scission in the presence of NADH (reduced nicotinamide adenine dinucleotide), thus enhancing the antitumor activity of this antibiotic.9,10 This antibiotic also exhibits a strong EPR signal upon reduction in the presence of a bound metal ion, indicating the formation of a metal–semiquinone form of this drug.11 These results indicate that metal ions are directly involved in the action of SN.Metal– SN complexes can be reduced to their semiquinone forms by NADH to induce cleavage of DNA. This reduction process is inhibited by superoxide dismutase and catalase, indicating the involvement of superoxide and peroxide.6,9d Moreover, the interaction of metal–SN complexes with DNA has also been proposed on the basis of some optical studies.12 However, the role of metal ion in the action of SN has not yet been fully defined, and the metal binding mode and structure of these metal complexes could not be definitely determined in previous studies.Particularly, two diVerent configurations of the metal– SN complexes have been proposed (Fig. 1):6 with the metal † E-Mail: ming@chuma.cas.usf.edu bound through the quinolinequinone–amine functionalities based on the crystal structure;2 and via the quinolinequinone– picolinate functionalities which requires a significant twist of the crystal structure.We report here a study of the binding of SN with paramagnetic metal ions, including the transition metal ions Co21 and Fe21, and the lanthanide Yb31. Since the chemical shift and the relaxation times of paramagnetic molecules are very sensitive to structural changes,13 they can be utilized as very sensitive ‘probes’ for the studies of molecular structures and interactions.The paramagnetically shifted 1H NMR signals of the metal–SN complexes have been fully assigned and their relaxation times measured, which aVord an accurate determination of their structures in solution. The interaction of the Fig. 1 (A) The molecular structure of streptonigrin based on the crystallographic study.2 (B) The molecular structure of a metal complex of streptonigrin based on the NMR studies discussed in this report.The metal is put in the quinolinequinone–picolinate site according to the results from the NMR studies. This structure requires ª1808 rotation of the bipyridine C2]C29 bond in the crystal structure of the drug (A). The numbering of SN follows the nomenclature: 3-amino- 2-(79-amino-6-methoxy-59,89-dioxoquinolin-29-yl)-6-carboxy-4-(20- hydroxy-30,40-dimethoxyphenyl)-5-methylpyridine2794 J. Chem. Soc., Dalton Trans., 1998, Pages 2793–2798 Co21–SN complex with DNA has also been monitored by the use of optical and NMR spectroscopies.A direct interaction was observed, where a significant change of the hyperfineshifted 1H NMR signals of the complex was detected in the presence of DNA. These paramagnetic metal–SN complexes can serve as prototypical model systems for future investigation of other paramagnetic metal–drug complexes and their binding with DNA. Results and Discussion Titration of streptonigrin with metal ions A freshly prepared methanol solution of SN gives a deep brown solution with lmax = 392 nm.An optical titration shows Co21 ion can bind SN tightly in methanol to form a very stable 1 : 1 Co21–SN complex (lmax = 404 nm, Fig. 2). A fitting of the change of the absorption at 404 nm of SN with respect to the amount of Co21 gives an aYnity constant of 3.30 × 106 M21 for the simple equilibrium Co21 1 SN Co21–SN (inset, Fig. 2). Similarly, the addition of Fe21 to SN in methanol under argon shifts the electronic absorption of the drug to 400 nm upon the formation of a 1 : 1 Fe21–SN complex with an aYnity constant of 5.43 × 106 M21.Upon the addition of Yb31 to SN in CH3CN the lmax shifts to 410 nm (greenish yellow) with an aYnity constant 1.58 × 106 M21 for the formation of a 1 : 1 Yb31–SN complex. The change of the electronic transition in SN upon the binding of these three metal ions (cf. Fig. 2) is similar to that observed previously for Cu21 and Zn21 binding to the drug.6 1H NMR of ytterbium(III)–streptonigrin complex The 1H NMR spectrum of a freshly prepared 1 : 1 Yb31–SN complex in methanol is shown in Fig. 3 (spectrum B), in which the signals due to the drug are paramagnetically shifted to the region of d 5 to 210. Since the Yb31-bound SN is undergoing chemical exchange with the free drug, signal assignment of the Yb31–SN complex can be achieved by the use of saturation transfer two-dimensional EXchange SpectroscopY (EXSY) on a sample with both the free drug and the complex present (Fig. 4).14 The paramagnetically shifted signals in an EXSY spectrum can thus show cross-peaks with their diamagnetic counterparts of the free drug, which can easily be assigned on the basis of chemical shift and COSY (Fig. 3A). For example, the signal at d 28.9 (which integrates to 3 protons with T1 = 114.5 ms) is assigned to 5-CH3 on the picolinate ring (Fig. 1), and the signals at d 25.5 (73.7) and 21.8 (167.9 ms) are assigned to quinolinequinone 39-H and 49-H protons, respectively (Fig. 4). Since the relaxation time T1 of a proton in paramagnetic molecules is proportional to the sixth power of the proton– Fig. 2 Electronic spectra of SN and its binding with Co21 in methanol. The formation of the 1 : 1 complex is clearly shown in a titration of Co21 into a 0.033 mM drug solution (inset). A fitting of the change of the absorption at 404 nm against [Co21] using the simple equilibrium Co21 1 SN Co21–SN gives an aYnity constant 3.30 × 106 M21.Similarly, the binding of the Fe21 and Yb31 to SN shifts lmax to 400 and 420 nm with aYnity constants 5.43 × 106 (CH3OH) and 1.58 × 106 M21 (CH3CN), respectively metal distance (i.e. T1 µ rM–H 6),15 it is therefore extremely sensitive to structural changes. Thus, it can be taken as a ‘ruler’ for the measurement of the proton–metal distances in paramagnetic molecules. The three most upfield shifted signals in the spectrum of the complex with the shortest relaxation times are attributable to the protons closest to the paramagnetic Yb31 center.The large paramagnetic shift and short T1 value of the 5-CH3 protons suggest that they are close to the bound Yb31. This T1 value is shorter than that of the 60-H protons (Table 1), indicating that the 5-CH3 protons are closer to the metal. This is Fig. 3 Proton NMR spectra (360.13 MHz, 298 K, 908 pulse ª7 ms) of (A) free drug and the 1 : 1 complexes (ª4 mM) Yb31–SN (B), Fe21–SN (C), and Co21–SN (E) in CD3OD, and (D) Co21–SN (ª2 mM) in borate– D2O buVer at pD 8.0.The signals are assigned based on their T1 values and EXSY studies (Figs. 4–6) Fig. 4 The 1H EXSY spectrum (360.13 MHz, 298 K, mixing time 20 ms) of the complex Yb31–SN in the presence of residual free drug (asterisked) in CD3OD. The numbers show the assignment of the signals to the structure in Fig. 1J. Chem. Soc., Dalton Trans., 1998, Pages 2793–2798 2795 Table 1 Proton NMR chemical shifts and T1 values and metal–proton distances of metal–SN complexes in CD3OD Signal Yb31–SN Fe21–SN Co21–SN M–H/Å assignment 39-H 49-H 69-OCH3 5-CH3 30-OCH3 40-OCH3 50-H 60-H d 25.5 21.8 20.4 28.9 d ª5.0 d 2.93 T1/ms 73.7 167.9 521.0 114.5 ddd 361 Yb31]Ha/Å 5.16 5.97 c 7.47 5.58 ——— 6.90 d 65.6 6.37 0.70 1.44 3.11 3.24 5.78 5.22 T1/ms 7.5 18.1 90.2 15.7 328.7 349.7 158.6 35.6 Fe21]Ha/Å 5.14 5.97 c 7.85 5.82 9.95 10.08 8.68 6.69 d 87.0 29.6 3.75 16.2 8.72 7.02 10.2 13.4 T1/ms 13.8 40.8 d 38.8 291.5 d 291.4 87.0 Co21]Ha/Å 4.97 5.97 c — 5.92 8.48 — 8.48 6.80 Model I b 5.06 5.97 6.77 e 6.08 e 9.95 e,f 11.24 e,f 9.05 f 6.73 f Model II b 5.06 5.97 6.77 e 6.19 e 7.66 e,f 8.44 e,f 6.41 f 4.57 f a A 0.5 s21 diamagnetic contribution has been added to the relaxation in the calculation of the distance, i.e.T1921 1 0.5 = T1 21. b Model I is shown in Fig. 1B with the metal bound to the drug through the bipyridyl moiety (quinolinequinone–picolinate) and the fourth ring perpendicular to the bipyridine moiety.The alternative configuration, Model II, is based on the crystal structure of the free drug (Fig. 1A) in which the metal is bound through the quinolinequinone–amine functionalities. The M]N distances are set to be 2.1 Å in these models. c This distance is used as the reference distance. The other metal–proton distances are calculated as (T19/T19(M–49H))1/6 × 5.97 Å.15,16 d Not resolved or measured. e Average with the assumption of free rotation of the methyl group.f Average of two distances with the ring rotated by 1808. consistent with binding of the Yb31 at the quinolinequinone– picolinate site as shown in Fig. 1B. The T1 values of other signals are also consistent with this binding mode for this complex (Table 1). Since Yb31, like alkaline earth metal ions,17 prefers an oxygenrich ligand binding environment with little covalency, the ethylenediamine diacetate-like binding mode shown in Fig. 1B is presumably the preferred binding mode for the biologically relevant Ca21 and Mg21 ions as well.As transition metal ions have been proposed to be involved in the binding of SN to DNA and cleavage of DNA by the drug, the study of metal–SN complexes is important to provide further mechanistic information about SN action.6,8 However, because the ligand binding preferences between transition metal ions and the lanthanides (and the alkaline earth metals) are very diVerent, whether or not this Yb31 binding mode is applicable to transition metal– SN complexes cannot be answered at this stage. 1H NMR of iron(II)–streptonigrin complex The redox-active Fe21 ion has been shown to enhance the activity of SN.9 Hence, it is important to reveal the exact binding mode of Fe21 with this antibiotic and solve the structure of the Fe21–SN complex in order to gain further insight into the mechanism of SN action and the role of metal ion in the action. Since Fe21 can aVord relatively sharp hyperfine-shifted 1H NMR signals,13 the Fe21 complex of SN can be thoroughly analysed by means of NMR techniques.The 1H NMR spectrum of a 1 : 1 Fe21–SN complex shows several well defined hyperfine-shifted signals (Fig. 3C). The ‘clean’ spectrum indicates that there is only one Fe21–SN complex formed under the experimental conditions. The binding mode of Fe21 ion can be determined when signal assignment is achieved, as discussed below. All the 1H NMR signals of the Fe21–SN complex can be assigned by means of two-dimensional NMR techniques (COSY and EXSY, Fig. 5) and T1 measurement (Table 1).The most downfield-shifted signal at d 65.6 (7.5 ms) can be assigned to the 39-H proton, which is four bonds away from the bound Fe21 and is the closest to the metal. Therefore, it should gain the largest through-bond contact shift and shortest relaxation time compared to all other protons. The rest of the hyperfine-shifted signals can be assigned based on their correlations with their diamagnetic counterparts of the free drug in the EXSY spectrum.The only COSY cross-peaks of the complex (inset, Fig. 5) are associated with the phenyl ring protons 50- and 60-H at d 5.78 and 5.22, respectively. A complete signal assignment is shown in Table 1. The 15.7 ms T1 value of the 5-CH3 signal at d 1.44 is shorter than that of the 60-H signal at d 5.22 (35.6 ms). This indicates that Fe21 is bound to SN at the quinolinequinone–picolinate site (Fig. 1B, Model I in Table 1), similar to that in Yb31–SN. This binding mode requires a ª1808 rotation of the bipyridine C2]C29 bond in the crystal structure (Fig. 1A). Another configuration with the metal bound through the 3-NH2 nitrogen of SN has been proposed in previous studies 6d based on the crystal structure of the free drug 2 (Model II, Table 1). This alternative would aVord a Fe]H (5-CH3) distance much longer than the Fe-H60 distance, thus a longer T1 value for the 5-CH3 protons than the 60-H proton.This configuration can be discarded based solely on the T1 values reported in our study (Table 1). The Fe21–SN complex is presumably the ironbound form of the drug under the reduction conditions in the cells. The unambiguous assignment of the 1H NMR signals and the determination of the structure of this complex described here provide an important step for further study of the interaction of this complex with biomolecules and cell components. The NMR results also indicate that the formation of a Fe31– SN (semiquinone) complex via electron transfer from Fe21 to SN is not likely to occur.This is because: (1) there is no indication of a high unpaired electron density on the quinone ring (as a result of the free radical on a semiquinone moiety), Fig. 5 The 1H EXSY spectrum (360.13 MHz, 298 K, mixing time 20 ms) of the complex Fe21–SN in the presence of residual free drug (asterisked) in CD3OD. The inset is a COSY spectrum showing the H50–H60 through-bond coupling.The numbers indicate the assignment of the signals to the structure shown in Fig. 12796 J. Chem. Soc., Dalton Trans., 1998, Pages 2793–2798 which would aVord a large contact shift and an even shorter relaxation time on the 69-OCH3 protons; and (2) there is no sign of a larger magnetic moment and longer electronic relaxation time due to the S = 5 2 – Fe31 (relative to the S = 2 Fe21) in a Fe31–SN (semiquinone) complex, which would aVord faster relaxing and broader hyperfine-shifted 1H NMR signals. 1H NMR of cobalt(II)–streptonigrin complex A 1: 1 Co21–SN complex is formed upon the addition of 1 equivalent Co21 to SN in methanol as shown by its electronic and 1H NMR spectra (Figs. 2 and 3E). The hyperfine-shifted 1H NMR signal at d 87 (T1 = 13.8 ms) can be assigned to the 39-H proton which is four bonds away from the metal and is the proton closest to the metal. The signal at d 29.6 (40.8 ms) can be assigned to 49-H five bonds away from the metal that gains significant contact shift via the aromatic pyridine ring.The assignment of most hyperfine-shifted signals can be achieved by the use of the EXSY technique to reveal saturation transfer between the complex and free drug (Fig. 6, Table 1). For example, the 5-CH3, 50-H, and 60-H are found at d 16.2 (38.8), 10.24 (291.4), and 13.41 (87.0 ms), respectively. The shortest T1 value of the 5-CH3 protons among all protons reflects that these protons have the shortest distance to the Fe21. The signal assignment and the T1 values of the hyperfine-shifted signals of Co21–SN (Table 1) are consistent with the structure shown in Fig. 1B, with the Co21 bound to the quinolinequinone– picolinate function groups (Model I, Table 1) rather than to the quinolinequinone and the 3-NH2 groups (Model II, Table 1). This binding mode is similar to that found in the Fe21–SN complex. Again, this indicates that the bipyridine C2]C29 bond of the free drug in the crystal structure 2 has to rotate by ª1808 upon metal binding.The electronic (lmax = 370 nm) and NMR (Fig. 3D) spectra of the Co21–SN complex observed in borate buVer solution at pH 8 are similar to those acquired in methanol solution. The acquisition of the NMR spectrum of a metal–SN complex in aqueous solution is important for further study of its interaction with DNA (see below). This complex shows broader isotropically shifted 1H NMR signals in water than in methanol, possibly attributable to a coagulation of this hydrophobic drug in aqueous solution.The broadness of 1H NMR signals in aqueous solution has also been observed for metal–anthracycline complexes which also contain an extended hydrophobic ring system.18 The virtually identical spectral features of the Co21–SN complex in water and methanol, however, indicate the formation of the same complex in these two solutions. This suggests that the structural information acquired in methanol can assist the assignment of the structure and better understanding of the action of metal–SN complexes in aqueous solutions under physiological conditions.Fig. 6 The 1H EXSY spectrum (360.13 MHz, 298 K, mixing time 20 ms) of the complex Co21–SN in the presence of residual free drug (asterisked) in CD3OD. The numbers indicate the assignment of the signals according to the structure shown in Fig. 1 Interaction of Co21–SN complex with poly[dA-dT] The air-sensitive Fe21–SN complex is diYcult to handle when sample transfer is necessary during experiments.The binding mode of Co21 with this drug is similar to that of Fe21, suggesting that the more air-resistant Co21–SN complex can serve as a substitute for Fe21–SN, and as a good model system to provide molecular information and DNA-binding property of metal– SN complexes. Moreover, the sensitivity of hyperfine-shifted signals toward subtle structural changes 13 also suggests that the paramagnetic Co21–SN complex can serve as a good probe for monitoring the binding of metal–SN complexes with DNA.Previous studies showed that SN exhibited a preferred cleavage site at cytosine bases adjacent to purine bases in DNA.10b Moreover, addition of poly[dA-dT] to the complex Cu1–SN was previously observed to cause small perturbation of the drug signals (0.22 to 0.31 ppm), which was suggested to be due to the binding of this complex to poly[dA-dT].10b Upon addition of 10 units of poly[dA-dT] to Co21–SN in borate buVer D2O solution at pD 8.0 three new 1H NMR signals appear, one sharp peak at d 16.8 and two broad peaks at d ª15 (overlapped) and 127 (Fig. 7D), with concomitant disappearance of the downfield hyperfine-shifted signals of the Co21–SN complex (Fig. 7A). This significant change of the paramagnetically shifted signals suggests that Co21–SN complex is bound to poly[dA-dT], forming a ternary Co21–SN– poly[dA-dT] complex. Binding of Co21–SN complex with calf thymus DNA The addition of a soluble form of calf thymus DNA to Co21– SN complex in 10 mM Tris buVer [tris(hydroxymethyl)methylamine] at pH 7.5 causes a shift of the electronic transition of the complex from 370 to 385 nm with a slight decrease in intensity and an isosbestic point at ª415 nm.This result indicates that the Co21–SN complex can also bind to naturally occurring DNA. This red-shift of the optical absorption is similar to that of the Zn21–SN complex upon the addition of calf thymus DNA.12 Upon the addition of calf thymus DNA to ª3 mM Co21–SN in borate buVer at pD 8.0 the 1H NMR signals of the complex at d 84 and 30.5 decrease in intensity, and two new signals appear at d 73 and 40 that are presumably due to the 39-H and 49-H protons, respectively (Fig. 7, A through C). These two new signals are not observed when ª5 mM Co21 is present in the DNA solution under the same conditions, suggesting that the complex is bound to the DNA (or that the drug assists the Fig. 7 Proton NMR spectra (360.13 MHz and 298 K) of (A) Co21–SN (200 ml at ª2 mM) and this sample with the addition of 220 (B) and 380 ml (C) calf thymus DNA (1 mg mL21), and the spectrum of the complex in the presence of 10 units poly[dA-dT] (D). All the samples were in borate–D2O buVer at pD 8.0J. Chem. Soc., Dalton Trans., 1998, Pages 2793–2798 2797 binding of metal ion to DNA). This observation clearly indicates that Co21 remains bound to SN upon binding of the Co21–SN complex to large pieces of native calf thymus DNA.Since irradiation of these signals does not reveal any noticeable saturation transfer, the two new signals may not be in fast exchange with the two hyperfine-shifted 39- and 49-H signals of the Co21–SN complex under the experimental conditions. This indicates that this complex binds calf thymus DNA to form a kinetically inert Co21–SN–DNA ternary complex. The results presented here clearly reveal the binding of the complex Co21–SN with poly[dA-dT]2 and native calf thymus DNA.Conclusion Optical titration and one- and two-dimensional NMR techniques have been applied to the study of the few metal complexes (Co21, Fe21, and Yb31) of the antitumor antibiotic streptonigrin. These studies reveal that SN binds transition metal and lanthanide ions with the metal located in the quinolinequinone –picolinate site, which aVords a configuration that requires a ª1808 rotation of the bipyridine C2]C29 bond in the crystal structure.The Co21–SN complex shows diVerent isotropically shifted 1H NMR signals upon addition of calf thymus DNA and poly[dA-dT], indicating direct binding of the complex with DNA. These studies provide the foundation for future investigation of the interactions between metal–SN complexes and diVerent oligonucleotide sequences to reveal detailed information about the mechanism of SN action and the structures of metal–SN–DNA ternary complexes. This report also demonstrates that NMR can be a versatile tool for the study of paramagnetic metal–DNA complexes.Experimental Chemicals and sample preparations Streptonigrin was purchased from Sigma Co., and was also supplied as a gift by Rhône-Poulenc Rorer, Recherche- Développement Laboratories (Pairs) and by the National Cancer Institute (Drug Synthesis & Chemistry Branch, Development Therapeutics Program, Division of Cancer Treatment). The drug is soluble in some organic solvents such as 1,4-dioxane, pyridine, dmf, dmso, and slightly soluble in alcohol and CHCl3.It is barely soluble in aqueous solution at pH < 7, but is slightly soluble at higher pHs to low mM levels. However, it is unstable and photosensitive at pH > 8.6d This low solubility in water causes diYculty for NMR studies. To overcome this problem, SN was first dissolved in aqueous solution at high pH and then adjusted to the desired pH value. The drug solution was prepared just before the experiments, and the concentration of the drug was determined by using e365 = 14 200 M21 cm21 at pH 7.2.6d The metal complexes of SN were prepared by direct addition of stoichiometric amount of metal ions to the SN solutions.All metal salts were obtained as the highest grade. Metal ion concentrations were determined by edta titration with xylenol orange as indicator. All the organic solvents used in the experiments were HPLC grade. The DNA solutions were prepared by dissolving, respectively, 1 mg soluble calf thymus DNA (Sigma Chemical Co.) and 10 units polydeoxy(adenylic acid–thymidylic acid) (poly[dA-dT], Sigma) in borate buVer D2O solution at pD 8.0 and stored at 4 8C.One unit of poly[dA-dT] yields A280 = 1.0 in 1.0 mL water (at 1 cm path length). The Co21 and Fe21 samples were prepared under anaerobic conditions, and transferred to an optical cell or an NMR tube under argon using a gas-tight syringe. The electronic spectra were acquired on a Hewlett Packard 8452A diode array spectrophotometer using a quartz cell of 1 cm path length.Metal titrations were performed by continuous addition of metal ions to SN solutions (e.g. 0.033 mM in the case of Co21 titration shown in Fig. 2). The spectra were recorded and calibrated against dilution factors. The aYnity constant can be obtained by fitting the change in the absorptions (i.e. DA = AM–SN 2 ASN) with respect to the metal concentration according to the equilibrium M 1 SN M–SN.Nuclear magnetic resonance experiments The metal complex concentrations in organic solvents for NMR studies were about 4 mM, whereas those in aqueous solutions were about 2 mM. All 1H NMR spectra were acquired on a Bruker AMX360 spectrometer at 360.13 MHz. The 1H chemical shift was referenced to external tetramethylsilane to avoid the eVect on the chemical shift of an internal reference by the paramagnetism of the metal complexes. A 908 pulse with presaturation for solvent suppression was used for the acquisition of one-dimensional 1H NMR spectra (8K data points).A linebroadening factor of 10–30 Hz was introduced to the spectra via exponential multiplication prior to Fourier transformation to enhance the signal-to-noise ratio. In the presence of chemical exchange (such as an equilibrium M 1 L M]L), saturation transfer can occur between counterparts, such as between the paramagnetically shifted signals in M]L and their diamagnetic counterparts in L in NMR experiments.This can be conveniently studied by the saturation transfer techniques used for detection of the nuclear Overhauser eVect (NOE), such as one-dimensional diVerence spectroscopy with the decoupler set on and oV the signal of interest and the two-dimensional EXSY pulse sequence (D1–908–t1–908– tmixing-free induction decay). Owing to the fast nuclear relaxation rates and the fast molecular rotational correlation time, NOE cannot be detected in small paramagnetic complexes.The cross-peaks observed in the EXSY spectra of the M–SN complexes are thus due to chemical exchange of the drug between its free and complexed forms. The EXSY spectra were acquired with presaturation for solvent suppression and 1024 × 512 data points. A 45–608 shifted sine-squared-bell window function was applied in both dimensions prior to Fourier transformation in phase sensitive EXSY spectra. Magnitude-COSY spectra of the complexes were acquired for the elucidation of through-bond proton couplings as shown in Fig. 5. The spectra were acquired with 1024 × 256 data points, and then a 08-shifted sine-squaredbell window function was applied to both dimensions and processed in magnitude mode. Proton spin–lattice relaxation times (T1) for all the metal complexes were determined by the use of the inversion-recovery method (D1–1808–t–908-free induction decay) with 16 diVerent t values and a recycle time of ª5T1.The peak intensities were fitted against the t values by a three-parameter fitting program on the spectrometer to give the T1 values. Since nuclear relaxation in paramagnetic molecules is dependent upon the metal– nucleus distance, relative distances can be obtained with respect to a reference nucleus [i.e. rM–H = (T1M/T1Mref ) 1/6 rM–Href]. The proton 49-H (rM–49H = 5.97 Å) was chosen as the reference proton. Since dipolar relaxation in paramagnetic metal–pyridine complexes has been demonstrated to be the predominant contribution to nuclear relaxation,16 the contact contribution to the nuclear relaxation was not taken into consideration in this study.In most cases, paramagnetic relaxation is the predominant contribution to nuclear relaxation. To demonstrate this, a 0.5 s21 diamagnetic contribution was considered in the calculation of the distance rM–H. There is no significant diVerence in the calculated rM–H with or without considering the diamagnetic contribution.Acknowledgements This work has been partially supported by a University of South Florida (USF) Research and Creative Scholarship Award, and by the Florida Division American Cancer Society2798 J. Chem. Soc., Dalton Trans., 1998, Pages 2793–2798 Edward L. Cole Research Grant (F94USF-3) on antitumor antibiotics. X. W. acknowledges a summer research fellowship (1995) awarded by the Institute for Biomolecular Science at USF. The gift of streptonigrin by Rhône Poulenc Laboratories and by the National Cancer Institute (Drug Synthesis and Chemistry Branch, Developmental Therapeutics Program, Division of Cancer Treatment) is gratefully acknowledged.References 1 W. S. Marsh, A. L. Garretson and E. M. Wesel, Proc. Am. Assoc. Cancer Res., 1960, 3, 131; K. V. Rao and W. P. Cullen, in Antibiotics Annual 1959–1960, eds. H. Welch and F. Marti-Ibanez, Medical Encyclopedia, New York, 1960. 2 Y. Chiu and W. N. Lipscomb, J. Am.Chem. Soc., 1975, 97, 2525. 3 Antibiot. Chemother., 1961, 11, 147. 4 T. J. McBride, J. J. Oleson and D. Woolf, Cancer Res., 1966, 26A, 727; H. L. White and J. R. White, Mol. Pharmacol., 1968, 4, 549; R. B. Livingston and S. K. Carter, Single Agents in Cancer Chemotherapy, Plenum, New York, 1970, pp. 389–392; M. A. Chirigos, J. W. Pearson, T. S. Papas, W. A. Woods, H. B. Wood, jun., and G. Spahn, Cancer Chemother. Rep., 1973, 57, 305; M. G. Brazhnikova and Y. V. Dudnik, Methods of Development of New Anticancer Drugs, National Cancer Institute Monograph: USAUSSR, 1975, pp. 207–212. 5 J. W. Lown and S.-K. Sim, Can. J. Chem., 1976, 54, 2563; K. V. Rao and J. W. Beach, J. Med. Chem., 1991, 34, 1871; D. L. Boger, K. C. Cassidy and S. Nakahara, J. Am. Chem. Soc., 1993, 115, 10 733. 6 (a) J. W. Lown and S.-K. Sim, Can. J. Biochem., 1976, 54, 446; (b) R. Cone, S. K. Hasan, J. W. Lown and A. R. Morgan, Can. J. Biochem., 1976, 54, 219; (c) N. R. Bachur, S. L. Gordon and M.V. Gee, Cancer Res., 1978, 38, 1745; (d ) J. Hajdu, in Metal Ions in Biological Systems, ed. H. Siegel, Marcel Dekker, New York, 1985, vol. 19; (e) M. S. Cohen, Y. Chai, B. Britigan, W. McKenna, J. Adams, T. Svendsen, K. Bean, D. Hassett and F. Sparling, Antimicrob. Agents Chemother., 1987, 31, 1507. 7 H. L. White and J. R. White, Biochim. Biophys. Acta, 1966, 123, 648. 8 J. Hajdu and E. C. Armstrong, J. Am. Chem. Soc., 1981, 103, 232; A. Moustatih and A. Garnier-Suillerot, J. Med. Chem., 1989, 32, 1426; M. M. L. Fiallo and A. Garnier-Suillerot, Inorg. Chem., 1990, 29, 893; G. Long and M. M. Harding, J. Chem. Soc., Dalton Trans., 1996, 549. 9 (a) J. R. White and H. N. Yeowell, Biochem. Biophys. Res. Commun., 1982, 106, 407; (b) M. L. Merryfield and H. A. Lardy, Biochem. Pharmacol., 1982, 31, 1123; (c) H. N. Yeowell and J. R. White, Antimicrob. Agents Chemother., 1982, 22, 961; (d ) J. Gutteridge, Biochem. Pharmacol., 1984, 33, 3059; (e) H. N. Yeowell and J. R. White, Biochim. Biophys. Acta, 1984, 797, 302; ( f ) P. H. Williams and N. H. Carbonetti, Infect. Immun., 1986, 51, 942. 10 (a) B. K. Sinha, Chem. Biol. Inter., 1981, 36, 179; (b) Y. Sugiura, J. Kuwahara and T. Suzuki, Biochim. Biophys. Acta, 1984, 782, 254. 11 H. S. Soedjak, B. L. Bales and J. Hajdu, in Oxygen Radicals in Biology and Medicine, eds. M. G. Simic, K. A. Taylor and C. V. Sonntag, Plenum, New York, 1987. 12 J. R. White, Biochem. Biophys. Res. Commun., 1977, 77, 387; K. V. Rao, J. Pharm. Sci., 1979, 68, 853. 13 I. Bertini and C. Luchinat, NMR of Paramagnetic Molecules in Biological System, Benjamin/Cummings, Menlo Park, CA, 1986. 14 L.-J. Ming and X. Wei, Inorg. Chem., 1994, 33, 4617. 15 I. Solomon, Phys. Rev., 1955, 99, 559. 16 L.-J. Ming, H. G. Jang and L. Que, jun., Inorg. Chem., 1992, 31, 359. 17 J.-C. G. Bunzli and G. R. Choppin, Lanthanide Probes in Life, Chemical and Earth Science, Elsevier, Amsterdam, 1989; C. H. Evans, Biochemistry of the Lanthanides, Plenum, New York, 1990; L.-J. Ming, in Nuclear Magnetic Resonance of Paramagnetic Macromolecules, ed. G.-N. La Mar, NATO-ASI, Kluwer, Dordrecht, 1995; Magn. Reson. Chem., 1993, 33, S104. 18 X. Wei, Ph.D. dissertation, University of South Florida, 1996; X. Wei and L.-J. Ming, Inorg. Chem., 1998, 37, 2255. Received 5th March 1998; Paper 8/01841C

ISSN:1477-9226    

DOI:10.1039/a801841c

出版商:RSC    

年代:1998

数据来源: RSC

摘要:

DALTON FULL PAPER J. Chem. Soc., Dalton Trans., 1998, Pages 2799–2805 2799 Insertion of a platinum(0) fragment into the strained silicon–carbon bond of a silicon-bridged [1]ferrocenophane: synthesis, alkyne insertion chemistry, and catalytic reactivity of the [2]platinasilaferrocenophane Fe(Á5-C5H4)2Pt(PEt3)2SiMe2 Karen Temple,a Alan J. Lough,a John B. Sheridan *,b and Ian Manners *,a a Department of Chemistry, University of Toronto, 80 St. George St., Toronto, Ontario M5S 3H6, Canada b Department of Chemistry, Rutgers, The State University of New Jersey, University Heights, Newark, NJ 07102, USA The reaction of Pt(PEt3)3 with the silicon-bridged [1]ferrocenophane Fe(h5-C5H4)2SiMe2 1 at 60 8C resulted in the insertion of a platinum(0) Pt(PEt3)2 fragment into the strained Si]C bond to yield the first [2]platinasilaferrocenophane Fe(h5-C5H4)2Pt(PEt3)2SiMe2 4.Complex 4 serves as a model for the proposed intermediate during the transition metal-catalyzed ring-opening polymerization (ROP) of 1.The reactivity of 4 was illustrated by the insertion of diphenylacetylene into the Pt]Si bond at elevated temperatures to yield a [4]ferrocenophane Fe(h5-C5H4)2Pt(PEt3)2C(Ph)C(Ph)SiMe2 5 with a cis Pt]C]] C]Si bridge. Both 4 and 5 were fully characterized spectroscopically and by single crystal X-ray diVraction. Despite the reactivity of the Pt]Si bond, the [2]platinasilaferrocenophane 4 was inactive as a ROP catalyst for 1 even at 95 8C.However, addition of BH3?THF co-catalyst rendered 4 active towards the ROP of 1 at 25 8C, presumably via abstraction of one or more PEt3 ligands, aVording low molecular weight [Mn (number average molecular weight) = ca. 1720–4695; PDI (polydispersity index) = 1.51–1.73], cyclic poly(ferrocenylsilanes) 6. Transition metal-catalyzed polymerizations of organic monomers have been well studied and are of widespread and still growing importance.1,2 Recently, transition metal-mediated routes to inorganic and organometallic polymer systems have attracted increasing attention.In particular, early transition metal-catalyzed dehydrocoupling reactions have provided attractive routes to s-conjugated materials such as polysilanes 3 and polystannanes 4 and metal-catalyzed ring-opening polymerization (ROP) of silacyclobutanes has also been well established as a route to high molecular weight polycarbosilanes.5,6 We are particularly interested in the synthesis, properties and applications of new classes of polymers with transition elements in the main chain such as poly(ferrocenes).7–9 The ambient temperature transition metal-catalyzed ROP of silicon-bridged [1]ferrocenophanes 1 has been reported using a variety of RhI, Pd0, PdII, Pt0 and PtII catalysts to yield high molecular weight poly(ferrocenylsilanes) 2 together with, in some cases, the cyclic dimer 3 as a by-product.10–12 This particularly mild, convenient and versatile route, which has been shown to permit molecular weight and architectural control,13 complements our previously described thermal7 and anionic,14 and recently discovered cationic 15 ROP routes to these interesting materials.The elucidation of the mechanism of these novel transition metal-catalyzed ROP reactions is therefore of considerable interest. By analogy with the tentatively proposed mechanism for the transition metal-catalyzed ROP of silacyclobutanes, 16–19 the initial step might be expected to involve insertion of the transition metal into the strained Si]C bond of 1.As a follow up to our recent communication,20 we report in detail our studies of the first example of this type of reaction and our investigation of the reactivity of the resulting [2]platinasilaferrocenophane. Results and Discussion EVective catalysts for the ambient temperature ROP of [1]silaferrocenophane 1 range from PtII and PdII species such as PtCl2, PdCl2 and Pd(cod)Cl2, to compounds such as Pt(cod)2 involving zero-valent platinum and the RhI complex, [Rh(cot)2- (m-Cl)]2.In contrast, ROP was not observed with a significant number of complexes including [Rh(cod)2(m-Cl)]2, RhCl(PPh3)3, and phosphine complexes of platinum and palladium such as Pt(PPh3)3.10,11 Interestingly, in the presence of catalytic amounts of Pd(PR3)2Cl2, (R = Cy or Bu), 1 has been reported to undergo exclusive cyclodimerization at elevated temperatures to aVord 3.12 Transition metal-catalyzed ROP of mixtures of 1 with other [1]ferrocenophanes 21,22 or sila- and disila-cyclobutanes 23 has also been shown to permit the formation of random copolymers.In addition, we have recently shown that the regiocontrolled metal-catalyzed ROP of unsymmetrically substituted [1]silaferrocenophanes with diVerent cyclopentadienyl ligands such as Fe(h5-C5H4)(h5-C5Me4)SiMe2 is possible,13 and that the addition of Et3SiH to 1 allows eVective molecular weight control. Moreover, the use of polysiloxanes with Si]H groups allows access to novel graft copolymers.13 A likely first step in the ROP mechanism involves insertion of a transition metal into the strained Si]C bond.Significantly, the strained Si]C bond of [1]silaferrocenophanes appears to mirror the reactivity of strained carbosilacycles which readily polymerize in the presence of Group 9 and 10 metal catalysts and for which the insertion of transition elements to yield metallacarbosilacycles has precedent.17,24 In eVorts to provide evidence for an analogous but unprecedented reaction for [1]silaferrocenophanes we have examined the reactivity of 1 towards low valent late transition metal centres.Si Me Me Fe Fe Si Me Me Fe Fe Si Si Me Me Me Me 3 1 n 22800 J. Chem. Soc., Dalton Trans., 1998, Pages 2799–2805 Synthesis and characterization of the [2]platinaferrocenophane, fcPt(PEt3)2SiMe2 4 (fc 5 Fe(Á5-C5H4)2) With prior knowledge that phosphine complexes such as Pt(PPh3)3 do not function as ROP catalysts for 1, we explored the analogous stoichiometric reaction with the more electron rich Pt0 complex Pt(PEt3)3, which is known to exhibit a rich range of oxidative-addition chemistry. Addition of equimolar quantities of 1 to toluene solutions of Pt(PEt3)3 followed by heating to 60 8C for 4 h led to a gradual disappearance of 1 as monitored by 1H NMR spectroscopy and formation of the insertion product 4 [equation (1)].Complex 4 crystallizes from hexane solutions as an air-stable orange crystalline solid and was isolated in ca. 80% yield. The new species was characterized using 1H, 13C, 31P, 29Si and 195Pt NMR spectroscopy. Noteworthy is the 29Si NMR resonance at d 5.18 which shows a large 1312 Hz coupling to Pt as well as couplings of 181 and 14 Hz to phosphorus nuclei of the trans and cis phosphine ligands respectively. The 31P NMR data show Pt]P couplings of 2160 and 910 Hz, the latter being small due to the strong trans influence of the silyl substituent. This is in excellent agreement with the 195Pt NMR spectrum which shows a doublet of doublets, centred at d 24661.0 with coupling to two inequivalent phosphorus nuclei of 2156 and 915 Hz.The 1H and 13C NMR data are consistent with a [2]platinasilaferrocenophane structure in that four and six cyclopentadienyl (Cp) resonances are observed respectively. Of these, the 13C-{1H} NMR signals for C1 and C6 are of particular interest.Thus, C1 appears as a multiplet centered at d 73.3 with a large Pt]C coupling of 850 Hz, whereas that for C6 is a multiplet at d 84.9 with a much smaller JPtC of 60 Hz. The downfield chemical shifts of C1 and C6 are notably diVerent from that of the ipso carbons in the highly strained parent [1]silaferrocenophane 1 which appear at d 33.5. In order to more fully investigate the novel structure of 4 an X-ray diVraction study was undertaken. Suitable crystals were grown from hexane solution at ca. 210 8C over 2 d and a labelled thermal ellipsoid plot of 4 is shown in Fig. 1. An accompanying summary of cell constants and data collection parameters are listed in Table 1. SiMe2 Fe Pt PEt3 PEt3 Pt(PEt3)3 toluene, 60 °C –PEt3 1 4 (1) Table 1 Summary of crystal data, details of intensity collection and least-squares refinement parameters Compound Empirical formula Mr Crystal system Space group T/K a/Å b/Å c/Å b/8 U/Å3 Z m(Mo-Ka)/cm21 Reflections collected Rint Independent reflections No.observed data [I > 2s(I)] R1[I > 2s(I)]* wR2(F2)* 4 C24H44FeP2PtSi 673.56 Monoclinic P21/c 293(2) 8.2699(14) 19.008(3) 17.719(2) 100.565(11) 2738.1(7) 4 58.03 6575 0.0387 5959 4337 0.0293 0.0614 5 C38H54FeP2PtSi 851.78 Monoclinic P21/n 293(2) 10.830(1) 15.807(2) 21.683(2) 96.044(7) 3691.1(7) 4 43.23 11 513 0.0318 10 629 6994 0.0365 0.0843 * R1 = S(Fo 2 Fc)/S(Fo); wR2 = {S[w(Fo 2 2 Fc 2)2]/S[w(Fo 2)2]}� �� . The structure reveals the molecule is slightly strained with a tilt angle of 11.6(3)8 between the C5H4 planes, much less than in 1 [20.8(5)8].8 The platinum centre is in a distorted square planar environment with a larger than expected P]Pt]P angle of 102.74(5)8, and somewhat compressed P]Pt]C(1) and C(1)]Pt]Si angles of 82.48(13) and 83.28(13)8, respectively.Interestingly, the angle b between the plane of the cyclopentadienyl ligand and the C(1)]Pt bond is only 1.2(3)8, whereas the analogous angle for the ipso C]Si bond is 12.8(3)8.The Pt]Si bond is also twisted with respect to the C5H4]Fe]C5H4 vector as revealed by a stagger in the C5H4 rings of 7.6(5)8. The oxidative addition of complex 1 to Pt(PEt3)3 represents the first well characterized example of an insertion of transition metal into the strained Si]C bond of a [1]silaferrocenophane.20 In order to gain insight into the chemistry of this novel complex, the reactivity and catalytic activity of 4 were probed. Insertion of diphenylacetylene into the Pt]Si bond of 4: synthesis and characterization of a [4]platinasilaferrocenophane 5 Hydrosilylation 25 and bis(silylation) 26 of unsaturated hydrocarbons, such as acetylenes, in the presence of organometallic catalysts represent well studied reactions. For example, alkynes have been inserted into the Si]C and Si]Si bonds of a variety of species including [2]disilaferrocenophane.27–29 We therefore attempted the insertion of diphenylacetylene into the Pt]Si bond of 4.30 The direct thermal reaction of 4 and diphenylacetylene in a toluene solution led to the 1,2-insertion of the alkyne into the Pt]Si bond and formation of compound 5 in 38% isolated yield [equation (2)].Recrystallization of 5 from toluene–n-hexanes at 255 8C gave orange, air-stable crystals and characterization by 1H, 13C, 31P and 29Si NMR spectroscopy was consistent with the proposed structure. More specifically, the insertion of diphenylacetylene removes the trans influence of the silyl group in 4 increasing one of the Pt]P coupling constants from 910 (for 4) to 1675 (for 5) Hz, while the JPtP coupling for the phosphorus Fig. 1 Molecular structure of 4 shown with 30% thermal ellipsoids Fe Pt PEt3 PEt3 Si MeMe PhC CPh Ph Ph 4 60 °C, toluene 5 (2)J.Chem. Soc., Dalton Trans., 1998, Pages 2799–2805 2801 trans to the C5H4 ligand remains almost relatively constant [cf. 2018 Hz (for 5), 2160 Hz (for 4) (Fig. 2)]. The 13C NMR spectrum of 5 shows broad resonances at d 131.5 (C28,32) and 127.1 (C29,31) assigned to the phenyl group in close proximity to one of the PEt3 ligands.These broad resonances indicate restricted rotation of this phenyl ring about the C25]C27 bond, presumably caused by steric interactions with the PEt3 group. Both the methyl and methylene resonances of 5 associated with the two PEt3 groups (d 8.6 and 17.3, respectively) are consistent with the corresponding peaks assigned in complex 4 (d 9.0 and 15.9).Interestingly, the two methyl groups of the SiMe2 moiety are separated by 1.0 ppm in the 1H NMR spectrum. Consideration of the molecular structure reveals the close proximity of one methyl to a phenyl group and hence ring current could contribute to the shift downfield. Unlike 4, complex 5 shows eight unique signals for the C5H4 protons. Likewise, the 13C spectrum displays ten individual resonances for the C5H4 carbons with those of the Cp bonded to platinum showing Pt coupling (cf. 1JPtC1 = 989, 2JPtC2,5 = 54 and 3JPtC3,4 = 77 Hz). A signal at d 177.2, typical of metal bound vinyl carbons, with coupling to both 195Pt (1J = 863 Hz) and 31P (2J = 112, 11 Hz) is assigned to C25. Further confirmation of the structure was obtained by an X-ray diVraction study. Single crystals of 5 suitable for an X-ray diVraction study were grown by cooling a solution of the compound in toluene–n-hexanes to 215 8C over 48 h. The molecular structure of 5 is shown in Fig. 3 with an accompanying summary of cell constants and data collection parameters listed in Table 1. The presence of the sterically encumbering Pt(PEt3)2 moiety resulted in a number of interesting structural features. In comparison with 4, the Pt]C(1)]Fe angle [138.4(2)8] is quite obtuse with the C5H4 ligand tilting slightly away from the bridging elements. Thus, Pt resides above the plane of the ring, causing the angle b to take on a value of 29.8(1)8. This leads to a longer Fe]C(1) bond length 2.145(4) Å and a shorter Fe]C(4) distance of 2.017(5) Å.Steric crowding of the phenyl groups also contracts the Pt]C(25)]C(27) bond angle to 108.3(3)8 causing a rather large deviation from the ideal 1208. One phenyl ring is close to the ethyl groups of the phosphine trans to C5H4 leading to restricted rotation about C(25)]C(27) as evidenced by the 1H and 13C NMR spectra (see above). The C]] C bond length [1.352(5) Å] is somewhat longer than the typical value of 1.317 Å and is only slightly twisted [2.9(1)8] with respect to the C5]Fe]C5 centroid axis.Noteworthy is the cis orientation of the alkene unit which is consistent with most transition metal-catalyzed alkyne silylation reactions. Undoubtedly, steric crowding of the bulky phenyl groups also plays an important role in the resulting conformation. However, the cis stereo- Fig. 2 The 31P NMR (121.5 MHz, C6D6) spectrum of complex 5 chemistry of silylation products is consistent with coordination of the alkyne prior to migratory insertion via the Pt centre.Catalytic activity of the [2]platinaferrocenophane 4 for the ROP of 1 in the absence and presence of BH3?THF as a co-catalyst As mentioned above, it is likely that the initial step in the platinum-catalyzed ROP of 1 involves insertion of a PtII or Pt0 atom into one of the fc ipso C]Si bonds of the ferrocenophane. 16–19 The formation of complex 4 represents the first direct evidence that such a step can occur.However, on addition of ca. 1 mol % of 4 to a room temperature toluene solution of 1 no polymerization nor dimerization was detected by NMR spectroscopy. Moreover, polymerization attempts under similar experimental conditions but at an elevated temperature of 95 8C, also proved unsuccessful. The lack of catalytic activity for 4 is consistent with the fact that phosphine derivatives of the Group 9 and 10 metals do not polymerize [1]ferrocenophanes. This contrasts with the active catalysts that have weakly co-ordinated ligands which allow further addition of ferrocenophane substrate to species akin to 4 leading to the high molecular weight organometallic polymers.10,11,31 Recent experiments involving an analogous complex of 4 but with a more labile cod ligand generated an eVective catalyst for the ROP of 1.31 However, in the presence of neat cod, all catalytic activity was arrested. This indicates that the displacement of the co-ordinating ligands to generate a reactive unsaturated platinum centre is necessary prior to polymerization.31 Therefore, in the case of 4 where the strongly co-ordinating phosphine ligands prevent catalytic activity, an alternative strategy to the transition metal-catalyzed ROP of 1 involves the abstraction of the phosphines by a strong Lewis acid.Thus, 2 mol % of BH3?THF was injected into a dichloromethane or THF solution containing an excess of 1 along with 1 mol % of 4 as the catalyst. The polymerization was observed to proceed smoothly with time over a period of 1.5 d reaching 100% conversion as monitored by 1H NMR spectroscopy.The polymer solution was concentrated and precipitated into n-hexanes yielding an orange-red polymer. The ROP of 1 by 4 proceeds in the presence of both 1 and 2 mol % of BH3?THF; however, the latter re in a qualitatively faster rate of polymerization. The complete absence of Fig. 3 Molecular structure of 5 shown with 30% thermal ellipsoids2802 J.Chem. Soc., Dalton Trans., 1998, Pages 2799–2805 any polymerization in a control experiment involving BH3?THF and 1 confirmed the inability of BH3?THF alone to initiate the ROP of 1. Despite a 2 : 1 ratio of BH3?THF to 4, not all of 4 is consumed as significant amounts are detectable by 31P NMR spectroscopy even at maximum conversion of 1. However, the ability of borane to abstract the phosphine ligands of 4 was proven by the appearance of a four line 31P NMR resonance centred at d 21.4 (JBP = 59 Hz) attributed to the formation of the H3B?PEt3 adduct.In contrast to other transition metal-catalyzed polymerizations of 1 which yield high molecular weight polymers 2 (Mw = 105–106, Mn > 105), GPC (gel-permeation chromatography) analyses of the poly(ferrocenylsilanes) using the two component 4–BH3?THF catalyst indicated the formation of low molecular weight polymer with Mn (number average molecular weight) = 1720–4695, PDI (polydispersity index) = 1.51–1.73. However, partial oxidation of poly(ferrocenes) can cause artificially low molecular weights to be observed by GPC in THF due to the associated decrease in hydrodynamic radius.32 We therefore considered it important to verify that the poly(ferrocenylsilanes) formed using the 4–BH3?THF catalyst system were indeed in their neutral form.Upon reduction of the product with sodium dihydronaphthylide, no significant apparent increases in molecular weight were observed which indicated that the materials are indeed of low molecular weight.Silicon-29 NMR spectroscopy was utilized to probe the structure of these oligomers which are formed during the phosphineabstraction route. Interestingly, the 29Si NMR spectra revealed a single resonance at d 26.4 assigned to SiMe2 units in the interior of the polymer chain [Fig. 4(a)]. However, the spectrum showed no signals that could be assigned to end groups which normally accompany low molecular weight linear poly(ferrocenylsilanes) with Mn < 8000.33 This indicated the important result that the materials have a cyclic structure [6, equation (3)]. For comparison, Fig. 4(b) illustrates the corresponding spectrum for linear 2 (Mn = 8000, PDI = 1.02) generated by living anionic ROP using fcLi as an initiator followed by quenching Fig. 4 (a) The 29Si NMR (79.5 MHz, C6D6) spectrum of poly- (ferrocenylsilane) 6 (Mn = 4700, PDI = 1.72) synthesized using catalytic amounts of 4 and BH3?THF.No resonances were detected in the range d 2200 to 200. (b) The 29Si NMR (79.5 MHz, C6D6) of a trimethylsilylcapped poly(ferrocenylsilane) (Mn = 8000, PDI = 1.02) synthesized using living anionic ROP initiated by fcLi and subsequently quenched by Me3SiCl Fe Si Me Me 1 mol % 4 (3) 1 n 6 2 mol % BH3•THF with Me3SiCl. Detectable fcSiMe2 end groups appear slightly upfield of the interior silicon resonances at d 26.6. These results lend support to the tentative mechanism proposed by us 34 (Scheme 1) which is analogous to that postulated by Tanaka and co-workers for the Pt0-catalyzed ROP of silacyclobutanes.24,35 This involves sequential addition of monomer to the metal centre followed by reductive elimination to yield a macrocyclic poly(ferrocene). Reductive elimination from 7 would yield the cyclic dimer, 3, which is an observed byproduct in transition metal-catalyzed ROP reactions in dilute solution. Conclusion Insertion of a platinum fragment into the strained Si]C bond has yielded a well defined [2]platinasilaferrocenophane 4 which serves as a model for the proposed key intermediate in the transition metal-catalyzed polymerization of 1.The ability of this complex to insert diphenylacetylene into the Pt]Si bond provides further evidence for the chemical similarity between the Pt]Si bond of 4 and the Pt]Si bond of platinacarbosilanes.30 The lack of catalytic activity of 4 towards the ROP of 1 is consistent with the inability of numerous phosphine derivatives of platinum to initiate the transition metal-catalyzed polymerization of 1.However, in the presence of BH3?THF co-catalyst, 4 is activated via phosphine abstraction to initiate ROP of 1. In contrast to the high molecular weight polymers typically produced by transition metal-catalyzed ROP,34 the molecular weights produced by this novel route are invariably low. The absence of end groups in the 29Si NMR of these polymers indicates that the polymer formed is macrocyclic rather than linear.Therefore, the mechanism may involve the generation of a reactive unsaturated platinum center, which might then undergo successive oxidative additions of 1 followed by reductive eliminations of the growing polymer chain to aVord macrocyclic poly(ferrocenylsilanes). Further work is currently underway which aims to elucidate the fate of the platinum catalyst and to provide further insight into the catalytic cycles for these interesting ROP processes.Experimental Materials Solvents were dried by standard procedures and distilled immediately prior to use. Diphenylacetylene, PEt3 and BH3? THF (1.0 M BH3 in THF) were purchased from Aldrich and were used as received. [1]Silaferrocenophane 17 and Pt(PEt3)3 were synthesized as described in the literature. Scheme 1 A mechanism for the transition metal-catalyzed ringopening polymerization of [1]silaferrocenophanes SiMe2 Fe M fc SiMe2 M fc Me2Si fc SiMe2 M fc Me2Si 7 –M fc = (h-C5H4)Fe(h-C5H4) M = Pt, Pd, Rh, etc. 1 6 M—L 1 –M fast n n–1 1 3 –LJ. Chem. Soc., Dalton Trans., 1998, Pages 2799–2805 2803 Equipment All reactions and manipulations were performed under an inert atmosphere (prepurified N2) using either standard Schlenk techniques or an inert-atmosphere glovebox (Vacuum Atmospheres), except for the polymers for which manipulations were carried out in air. The reactions and polymerizations were monitored by 1H and 31P NMR spectroscopy.Solution NMR spectra were recorded on Varian XL 400 instruments. Proton NMR spectra (400 MHz) were referenced to residual protonated solvent and 13C NMR spectra (100.5 MHz) were referenced to the residual protons of the deuteriated solvent. Proton and 13C assignments for 5 were based on an HMQC (Heteronuclear Multiple Quantum Coherence-reverse detection) heteronuclear correlation experiment.36 Silicon-29 NMR (79.5 MHz) spectra were referenced externally to SiMe4 utilizing a normal (proton decoupled) pulse sequence.For the 31P-{1H} NMR (121.5 MHz) spectra, H3PO4 served as the external reference. Solid-state 29Si NMR was run on a Bruker DSX 400 MHz spectrometer. Molecular weight distributions were analyzed by gel permeation chromatography using a Waters Associates 2690 separations unit. Ultrastyragel columns with a pore between 500, 103 and 105 Å, and a Waters 410 diVerential refractometer were used.A flow rate of 1.0 mL min21 was used and samples were dissolved in a THF solution of 0.1% tetra-n-butylammonium bromide. Polystyrene standards purchased from Aldrich were used for calibration purposes. Elemental analyses were performed by Quantitative Technologies Inc., Whitehouse, NJ. X-Ray structural characterization A summary of selected crystallographic data are given in Table 1. Data were collected on a Siemens P4 diVractometer using graphite-monochromated Mo-Ka radiation (l = 0.710 73 Å).The intensities of three standard reflections measured every 97 reflections showed no decay. The data were corrected for Lorentz and polarization eVects and a semiempirical absorption correction (calculated from y scans) was applied. Minimum and maximum absorption corrections were 0.2548 and 0.3955 for complex 4 and 0.2548 and 0.3955 for complex 5. The structures 37 were solved and refined using the SHELXTL PC package. Refinement was by full-matrix least squares on F 2 using all data (negative intensities included).The weighting scheme was w = 1/s2(Fo 2) 1 (0.0186P)2 for 4 and w = 1/s2(Fo 2) 1 (0.0371P)2 for 5 and where P = (Fo 2 1 2Fc 2)/3. Hydrogen atoms were included in calculated positions. Molecular structures are presented with ellipsoids at a 30% probability level for both 4 and 5. CCDC reference number 186/1047. Preparation of fcPt(PEt3)2SiMe2 4 [1]Silaferrocenophane 1 (2.15 g, 8.88 mmol) was added to a toluene solution (10 mL) of Pt(PEt3)3 (4.83 g, 8.80 mmol), prepared by heating Pt(PEt3)4 to 60 8C in vacuo for 5 h.The deep red solution was heated (60 8C) in an oil bath for 4 h after which time monitoring by 1H NMR spectroscopy indicated all of 1 had been consumed. The solvent was removed in vacuo and the residue extracted with hexanes (20 mL). Cooling to 255 8C overnight gave 4 as orange microcrystals: yield 4.73 g (80%). 1H NMR (C6D6): d 0.78 (9 H, m, PEt3), 0.84 (6 H, d, JPtH = 2.2, SiMe2), 0.89 (9 H, m, PEt3), 1.26 (6 H, m, PEt3), 1.61 (6 H, m, PEt3), 4.14 (2 H, m, JPtH = 23.5 Hz, H2,5), 4.40 (2 H, t, H7,10), 4.52 (2 H, m, H3,4), 4.80 (2 H, t, H8,9). 13C-{1H} NMR (C6D6): d 8.4 (m, JPtC = 12.5, PCH2CH3, trans to silyl group), 9.0 (m, JPtC = 21.1, PCH2CH3, cis to silyl group), 9.1 (m, JPtC = 84.7, JPC = 27.0, SiCH3), 15.9 (m, JPC = 16.9, JPtC = 10.9, PCH2CH3 trans to silyl group), 18.7 (m, JPC = 27.0 and 5.1, JPtC = 28.4, PCH2CH3), 70.0 (s, C7,10), 70.5 (m, JPtC = 52, JPC = 15, C2,5), 73.3 (m, JPtC = 850, JPC = 104 and 14, C1), 74.3 (m, JPtC = 65, JPC = 4.5 and 2, C3,4), 74.9 (s, C8,9), 84.9 (m, JPtC = 60, JPC = 4.5 and 2 Hz, C6). 31P-{1H} NMR (C6D6): d 10.4 (m, JPtP = 910, JPP= 19, PEt3 trans to silyl group), 11.2 (m, JPtP = 2160, JPP = 19 Hz, PEt3 trans to C5H4). 29Si-{H} NMR (C6D6): d 5.18 (m, JPtSi = 1312, JPSi = 181 and 14.5 Hz). 195Pt-{1H} NMR (C6D6): d 24661.0 (d of d, JPtP = 2156, JPtP = 915 Hz) (Found: C, 42.77; H, 6.68. Calc.for C24H44FeP2PtSi: C, 42.80; H, 6.53%). Preparation of fcPt(PEt3)2(CPh)2SiMe2 5 A solution of 4 (149 mg, 0.22 mmol) and diphenylacetylene (39 mg, 0.22 mmol) in toluene (3.0 mL) was heated to 60 8C for 24 h after which time the reaction was quantitatively complete as indicated by 1H NMR spectroscopy. Removal of the solvent in vacuo and purification by precipitation from toluene–n-hexane solutions at 255 8C gave 5 as an orange crystalline solid: yield 71 mg (38% isolated yield). 1H NMR (CD2Cl2): d 0.09 (3 H, s, SiMe2), 0.83 (9 H, m, PEt3), 0.99 (9 H, m, PEt3), 1.12 (3 H, s, SiMe2), 1.53 (6 H, m, PEt3), 1.61 (6 H, m, PEt3), 3.94 (1 H, m, C5H4), 3.98 (1 H, m, C5H4), 4.01 (1 H, m, C5H4), 4.04 (1 H, m, C5H4), 4.07 (1 H, m, C5H4), 4.09 (1 H, m, C5H4), 4.15 (1 H, m, C5H4), 4.26 (1 H, m, JPtH = 25 Hz, C5H4), 6.72 (2 H, t, Ph), 6.82 (2 H, m, Ph), 6.96 (2 H, t, Ph), 7.06 (3 H, br, Ph), 7.20 (1 H, m, Ph). 13C-{1H} NMR (CD2Cl2): d 2.7 [m, JPtC = 15, Si(CH3)2], 3.5 [s, Si(CH3)2], 8.6 [m, JPtC = 15, P(CH2CH3)3], 8.7 [m, JPtC = 15, JPC = 3 Hz, P(CH2CH3)3], 17.2 [m, JPtC = 25, JPC = 20, P(CH2CH3)3 cis to C5H4], 17.5 [m, JPtC = 26, JPC = 24, P(CH2CH3)3 cis to C5H4], 66.0 (m, JPtC = 53, JPC = 7, JPC = 1, C2 or 5), 66.6 (s, C6), 67.9 (m, JPtC = 55, JPC = 6, C2 or 5), 69.5 (s, C10 or 7), 70.8 (s, C7 or 10), 73.8 (s, C9 or 8), 74.4 (s, C8 or 9), 77.1 (m, JPtC = 77, JPC = 4, C3and4, two resonances partially overlapping), 93.6 (m, JPtC = 989, JPC = 118, JPC = 11, C1), 123.4 (s, C30), 123.8 (s, C36), 126.3 (s, C35,37), 127.1 (br, C29.31), 131.5 (br, C28,32), 131.9 (m, JPtC = 33, JPC = 2, C34,38), 142.0 (m, JPtC = 43, JPC = 6, JPC = 3, C27), 149.3 (m, JPtC = 86, JPC = 12, JPC = 2, C26), 151.2 (m, C33), 177.2 (m, JPtC = 863, JPC = 112, JPC = 11 Hz, C25). 31P-{1H} NMR (C6D6): d 25.73 (JPtP = 2018, JPP = 16, PEt3 trans to C5H4), 20.65 (JPtP = 1675, JPP = 16 Hz, PEt3 trans to silyl group). 29Si-{1H} CP-MAS solid-state NMR: d 16.3 (Found: C, 53.67; H, 6.33.Calc. for C38H54FeP2PtSi: C, 53.58; H, 6.39%). Attempted transition metal-catalyzed ROP of 1 by fcPt(PEt3)2- SiMe2 4 Complexes 1 (100 mg, 0.41 mmol) and 4 (2.6 mg, 4.1 mmol) were dissolved in C6D6, sealed in an NMR tube and heated to 95 8C over a period of 20 h. No polymerization was detected by 1H NMR spectroscopy. Transition metal-catalyzed ROP of 1 by fcPt(PEt3)2SiMe2 4 in the presence of BH3?THF as co-catalyst Polymerization: in a typical experiment, complex 1 (532 mg, 2.20 mmol) was dissolved in CH2Cl2 (ca. 5 mL) after which 1 mol % of 4 (15 mg, 0.02 mmol) and 2 mol % of BH3?THF (50 mL, 0.05 mmol) were added. After ca. 1.5 d at room temperature, the polymerization was complete as monitored by 1H NMR spectroscopy. Precipitation of the polymer into n-hexanes and drying under vacuum gave 260–348 mg of 6 as an orange solid (50–70%). 1H NMR (C6D6): d 0.36 (s, SiMe2 of 1), 3.94 (m, C5H4 of 1), 4.41 (m, Cp of 1), 0.54 (s, SiMe2 of 6), 4.09 (m, C5H4 of 6), 4.26 (m, C5H4 of 6). 13C NMR (C6D6): d 20.52 (s, SiMe2 of 6), 71.97 (s, C5H4 of 6), 72.00 (ipso C of C5H4 of 6), 73.82 (C5H4 of 6). 29Si NMR (C6D6): d 26.4 (s, SiMe2 of the interior of 6). In the 1H, 13C and 29Si NMR spectra of 6, no end groups were detectable. In addition, 31P NMR spectroscopy of the polymer did not reveal any phosphorus nuclei after ca. 2000 transients. GPC: run 1 (2 mol % BH3? THF): Mw = 6330, Mn = 3960, PDI = 1.60; run 2 (2 mol % BH3?THF): Mw = 8120, Mn = 4695, PDI = 1.73.Monitoring the polymerization using 31P NMR spectroscopy revealed formation of a BH3?PEt3 adduct in addition to a number of2804 J. Chem. Soc., Dalton Trans., 1998, Pages 2799–2805 unidentifiable by-products. 31P NMR (C6D6): d 21.4 (q, JBP = 59 Hz, Et3P?BH3) as identified by comparison with an authentic sample. Control experiment: complex 1 (75 mg, 0.31 mmol) was dissolved in ca. 0.5 mL C6D6 and 2 mol % of BH3?THF (0.76 mg, 8.5 mmol) was injected via syringe.Monitoring by 1H NMR spectroscopy at room temperature showed no reaction nor polymerization over a period of 7 d. Reduction experiment: to a THF (ca. 3 mL) solution of polymer 6 (300 mg, 1.24 mmol) was added a solution of Na[C10H8] in THF (1.0 mL, 1.26 mmol), prepared from the reaction of Na metal (ca. 8 g, 0.33 mol) with naphthalene (ca. 4 g, 0.03 mol) in the same solvent (25 mL). The resulting solution was precipitated into n-hexanes to yield an orange polymer.GPC: run 3 (2 mol % BH3?THF): (before reduction) Mw = 3910, Mn = 2590, PDI = 1.51; (after reduction) Mw = 5000, Mn = 2870, PDI = 1.74; run 4 (4 mol % BH3?THF): (before reduction) Mw = 2600, Mn = 1720, PDI = 1.51; (after reduction) Mw = 3360, Mn = 2085, PDI = 1.61. MS (EI, 70 eV) (for run 4): m/z 1939 [M1 = (fcSiMe2)8 1], 1694 [M1 2 fcSiMe2], 1453 [M1 2 (fcSiMe2)2], 1210 [M1 2 (fcSiMe2)3], 968 [M1 2 (fcSiMe2)4], 726 [M1 2 (fcSiMe2)5], 484 [M1 2 (fcSiMe2)6], 243 [M1 2 (fcSiMe2)7].Acknowledgements We would like to thank the donors of the Petroleum Research Fund, administered by the American Chemical Society. I. M. is grateful to the Alfred P. Sloan Foundation for a Research Fellowship (1994–1998), the Natural Sciences and Engineering Research Council of Canada (NSERC) for an E. W. R. Steacie Fellowship (1997–1999), and the University of Toronto for a McLean Fellowship (1997–2002). K. T. would like to thank NSERC (1997–1999) for a Graduate Fellowship.We would also like to express our gratitude to Dr. Patricia Aroca-Ouelette for running the solid-state NMR of 5, and Drs. Timothy Burrow and Howard Hunter for their assistance with obtaining the 195Pt NMR spectra of 4. References 1 For late transition metal catalysts for the copolymerization of olefins and CO, see: C. M. Killian, D. J. Tempel, L. K. Johnson and M. Brookhart, J. Am. Chem. Soc., 1996, 118, 11 664; L. K. Johnson, C. M. Killian and M. Brookhart, J. Am. Chem. Soc., 1995, 117, 6414; A.Sen, J. T. Chen, W. M. Vetter and R. R. Whittle, J. Am. Chem. Soc., 1987, 109, 148; F. C. Rix, M. Brookhart and P. S. White, J. Am. Chem. Soc., 1996, 118, 4746; S. Bronco, G. Consiglio, R. Hutter, A. Batistini and U. W. Suter, Macromolecules, 1994, 27, 4436. 2 R. F. Jordan, J. Organomet. Chem., 1991, 32, 325 and refs. therein. 3 See for example: B. P. S. Chauhan, T. Simizu and M. Tanaka, Chem. Lett., 1997, 785; J. A. Reichl, C. M. PopoV, L. A.Gallagher, E. E. Remsen and D. H. Berry, J. Am. Chem. Soc., 1996, 118, 9450; V. K. Dioumaev and J. F. Harrod, J. Organomet. Chem., 1996, 521, 133; T. Imori, R. H. Heyn, T. D. Tilley and A. L. Rheingold, J. Organomet. Chem., 1995, 493, 83. 4 See for example: J. R. Babcock and L. R. Sita, J. Am. Chem. Soc., 1996, 118, 12 481; V. Lu and T. D. Tilley, Macromolecules, 1996, 29, 5763; T. Imori, V. Lu, H. Cai and T. D. Tilley, J. Am. Chem. Soc., 1995, 117, 9931; T. Imori and T. D. Tilley, J.Chem. Soc., Chem. Commun., 1993, 1607. 5 C. S. Cundy and M. F. Lappert, J. Chem. Soc., Dalton Trans., 1978, 655 and refs. therein; D. R. Weyenberg and L. E. Nelson, J. Org. Chem., 1965, 30, 2618; H. Yamashita, M. Tanaka and K. Honda, J. Am. Chem. Soc., 1995, 117, 8873. 6 L. V. Interrante, H. J. Wu, T. Apple, Q. Shen, B. Ziemann and D. M. Narsavage, J. Am. Chem. Soc., 1994, 116, 12 085; H. J. Wu and L. V. Interrante, Macromolecules, 1992, 25, 1840. 7 D. A. Foucher, B.-Z.Tang and I. Manners, J. Am. Chem. Soc., 1992, 114, 6246. 8 I. Manners, Adv. Organomet. Chem., 1995, 37, 131; I. Manners, Can. J. Chem., 1998, 76, 731. 9 R. Bayer, T. Pöhlmann and O. Nuyken, Makromol. Chem., Rapid Commun., 1993, 14, 359; M. Altman and U. H. F. Bunz, Angew. Chem., Int. Ed. Engl., 1995, 34, 569; P. F. Brandt and T. B. Rauchfuss, J. Am. Chem. Soc., 1992, 114, 1926; M. Rosenblum, H. M. Nugent, K.-S. Jang, M. M. Labes, W. Cahalane, P. Klemarczyk and W. M. ReiV, Macromolecules, 1995, 28, 6330; M.Morán, M. C. Pascual, I. Cuadrado and J. Losada, Organometallics, 1993, 12, 811; I. Cuadrado, M. Móran, C. M. Casado, B. Alonso, F. Lobete, B. Garcia, M. Ibisate and J. Losada, Organometallics, 1996, 15, 5278; G. E. Sourthard, M. D. Curtis and J. W. Kampf, Organometallics, 1996, 15, 4667; R. N. Kapoor, G. M. Crawford, J. Mahmoud, V. V. Dementiev, M. T. Nguyen, A. F. Diaz and K. H. Pannell, Organometallics, 1996, 15, 2848; T. M. Alias, S. Barlow, J.S. Tudor, D. O’Hare, R. T. Perry, J. M. Nelson and I. Manners, J. Organomet. Chem., 1997, 528, 42. 10 Y. Ni, R. Rulkens, J. K. Pudelski and I. Manners, Makromol. Chem., Rapid Commun., 1995, 16, 637. 11 N. P. Reddy, H. Yamashita and M. Tanaka, J. Chem. Soc., Chem. Commun., 1995, 2263. 12 N. P. Reddy, N. Choi, S. Shimada and M. Tanaka, Chem. Lett., 1996, 649. 13 P. Gómez-Elipe, P. M. Macdonald and I. Manners, Angew. Chem., Int. Ed. Engl., 1997, 36, 762. 14 R. Rulkens, Y. Ni and I.Manners, J. Am.Chem. Soc., 1994, 116, 12 121. 15 R. Resendes, P. Nguyen, A. J. Lough and I. Manners, Chem. Commun., 1998, 1001. 16 W. R. Bamford, J. C. Lovie and J. A. C. Watt, J. Chem. Soc. C, 1966, 1137. 17 C. S. Cundy, C. Eaborn and M. F. Lappert, J. Organomet. Chem., 1972, 44, 291. 18 For the insertion of Fe into a silacyclobutane see: C. S. Cundy and M. F. Lappert, J. Chem. Soc., Chem. Commun., 1972, 445; C. S. Cundy and M. F. Lappert, J. Chem. Soc., Dalton Trans., 1976, 910; C.S. Cundy and M. F. Lappert, J. Chem. Soc., Dalton Trans., 1978, 665. 19 For insertion of Pt and/or Pd into a silacyclobutane: H. Yamashita, M. Tanaka and K. Honda, J. Am. Chem. Soc., 1995, 117, 8873; Y. Tanaka, H. Yamashita, S. Shimada and M. Tanaka, Organometallics, 1997, 16, 3246; P. Braunstein and M. Knorr, J. Organomet. Chem., 1995, 500, 21. 20 J. B. Sheridan, A. J. Lough and I. Manners, Organometallics, 1996, 15, 2195. 21 N. P. Reddy, H. Yamashita and M.Tanaka, J. Chem. Soc., Chem. Commun., 1995, 2263. 22 T. J. Peckham, J. Massey, M. Edwards, I. Manners and D. Foucher, Macromolecules, 1996, 29, 2296. 23 J. B. Sheridan, P. Gómez-Elipe and I. Manners, Makromol. Chem., Rapid. Commun., 1996, 17, 319. 24 H. Yamashita, M. Tanaka and K. Honda, J. Am. Chem. Soc., 1995, 117, 8873. 25 J. L. Speier, Adv. Organomet. Chem., 1979, 407; I. Ojima, in The Chemistry of Organic Silicon Compounds, eds. S. Patai and Z. Rappoport, Wiley, Chichester, 1989, p. 1479; A. J. Chalk and J. F. Harrod. J. Am. Chem. Soc., 1965, 87, 16; J. F. Harrod and A. J. Chalk, J. Am. Chem. Soc., 1965, 87, 1133; M. A. Schroeder and M. S. Wrighton, J. Organomet. Chem., 1977, 128, 345; R. G. Austin, R. S. Paonessa, P. J. Giordano and M. S. Wrighton, ACS Symp. Ser., 1978, 169, 189; B. Marciniec and J. Gulinski, J. Organomet. Chem,, 1983, 252, 349; C. L. Randolph and M. S. Wrighton, J. Am. Chem. Soc., 1986, 108, 3366. 26 M. Murakami, T. Yoshida and Y.Ito, Organometallics, 1994, 13, 2900. 27 See for example: W. S. Palmer and K. A. Woerpel, Organometallics, 1997, 16, 1097; A. Naka, S. Okazaki, M. Hayashi and M. Ishikawa, J. Organomet. Chem, 1995, 499, 35; W. S. Palmer and K. A. Woerpel, Organometallics, 1997, 16, 4827. 28 For insertion of alkynes into Si]Si bonds: H. Okinoshima, K. Yamamoto and M. Kumada, J. Am. Chem. Soc., 1972, 94, 9263; H. Okinoshima, K. Yamamoto and M. Kumada, J. Organomet. Chem., 1975, 86, C27; H. Sakurai, K.Kamiyama and Y. Nakadaira, J. Am. Chem. Soc., 1975, 97, 931; H. Yamashita, M. Catellani and M. Tanaka, Chem. Lett., 1991, 241; H. Watanabe, M. Kobayashi, K. Higuchi and Y. Nagai, J. Organomet. Chem., 1980, 186, 51; C. Liu and C. Cheng, J. Am. Chem. Soc., 1975, 97, 6746; C. W. Carlson and R. West, Organometallics, 1983, 2, 1801; D. Seyferth, E. W. Goldman and J. Escudie, J. Organomet. Chem., 1984, 271, 337; M. Murakami, T. Yoshida and Y. Ito, Organometallics, 1994, 13, 2900. 29 W. Finckh, B. Z. Tang, A. Lough and I. Manners, Organometallics, 1992, 11, 2904. 30 For reactivity of alkynes towards a Pt]Si bond: J. Chatt, C. Eaborn and P. N. Kapoor, J. Organomet. Chem., 1970, 23, 109; H. Yamashita, M. Tanaka and M. Goto, Organometallics, 1993, 12,J. Chem. Soc., Dalton Trans., 1998, Pages 2799–2805 2805 988; H. Yamashita, M. Tanaka and M. Goto, Organometallics, 1992, 11, 3227; F. Glockling and K. A. Gooton, J. Chem. Soc. A, 1967, 1066; C. Eaborn, T.N. Metham and A. Pidcock, J. Organomet. Chem., 1973, 63, 107; C. Eaborn, D. J. Tune and A. Pidcock, J. Chem. Soc., Dalton Trans., 1973, 2255; C. Eaborn, B. RatcliV and A. Pidcock, J. Organomet. Chem., 1974, 65, 181; C. Eaborn, T. N. Metham and A. Pidcock, J. Organomet. Chem., 1977, 131, 377. 31 J. B. Sheridan, K. Temple, A. J. Lough and I. Manners, J. Chem. Soc., Dalton Trans., 1997, 711. 32 P. Gómez-Elipe and I. Manners, unpublished work. Gel-permeation chromatography gives molecular weight data based on the hydrodynamic size of polymer molecules.Oxidation of the ferrocene units leads to contraction of the polymer coils due to poorer polymer/solvent interactions and so an apparent decrease in molecular weight is detected. For an introduction to GPC, see R. J. Young and P. A. Lovell, Introduction to Polymers, Chapman and Hall, 2nd edn., 1991, pp. 211–221. 33 R. Rulkens, A. J. Lough and I. Manners, J. Am. Chem. Soc., 1994, 116, 797; R. Rulkens, A. J. Lough and I. Manners. J. Am. Chem. Soc., 1996, 118, 12 683. 34 I. Manners, Polyhedron, 1996, 15, 4311. 35 N. P. Reddy, T. Hayashi and M. Tanaka, Chem. Commun., 1996, 1865. 36 M. F. Summers, L. G. Marzilli and A. Bax, J. Am. Chem. Soc., 1986, 108, 4285. 37 G. M. Sheldrick, SHELXTL PC, Siemens Analytical X-Ray Instruments Inc., Madison, WI, 1994. Received 10th March 1998; Paper 8/01935EJ. Chem. Soc., Dalton Trans., 1998, Pages 2799–2805 2805 988; H. Yamashita, M. Tanaka and M. Goto, Organometallics, 1992, 11, 3227; F. Glockling and K. A. Gooton, J. Chem. Soc. A, 1967, 1066; C. Eaborn, T. N. Metham and A. Pidcock, J. Organomet. Chem., 1973, 63, 107; C. Eaborn, D. J. Tune and A. Pidcock, J. Chem. Soc., Dalton Trans., 1973, 2255; C. Eaborn, B. RatcliV and A. Pidcock, J. Organomet. Chem., 1974, 65, 181; C. Eaborn, T. N. Metham and A. Pidcock, J. Organomet. Chem., 1977, 131, 377. 31 J. B. Sheridan, K. Temple, A. J. Lough and I. Manners, J. Chem. Soc., Dalton Trans., 1997, 711. 32 P. Gómez-Elipe and I. Manners, unpublished work. Gel-permeation chromatography gives molecular weight data based on the hydrodynamic size of polymer molecules. Oxidation of the ferrocene units leads to contraction of the polymer coils due to poorer polymer/solvent interactions and so an apparent decrease in molecular weight is detected. For an introduction to GPC, see R. J. Young and P. A. Lovell, Introduction to Polymers, Chapman and Hall, 2nd edn., 1991, pp. 211–221. 33 R. Rulkens, A. J. Lough and I. Manners, J. Am. Chem. Soc., 1994, 116, 797; R. Rulkens, A. J. Lough and I. Manners. J. Am. Chem. Soc., 1996, 118, 12 683. 34 I. Manners, Polyhedron, 1996, 15, 4311. 35 N. P. Reddy, T. Hayashi and M. Tanaka, Chem. Commun., 1996, 1865. 36 M. F. Summers, L. G. Marzilli and A. Bax, J. Am. Chem. Soc., 1986, 108, 4285. 37 G. M. Sheldrick, SHELXTL PC, Siemens Analytical X-Ray Instruments Inc., Madison, WI, 1994. Received 10th March 1998; Paper 8/01935E

ISSN:1477-9226    

DOI:10.1039/a801935e

出版商:RSC    

年代:1998

数据来源: RSC

摘要:

DALTON FULL PAPER J. Chem. Soc., Dalton Trans., 1998, Pages 2807–2811 2807 Synthesis, characterization and structure of rhodium(I) carbonyl complexes with O,P-chelating 19-(diphenylphosphino)ferrocenecarboxylate or P-monodentate 19-(diphenylphosphino)ferrocenecarboxylic acid Petr S¡ te¡pnic¡ka * and Ivana Císar¡ová Department of Inorganic Chemistry, Charles University, Hlavova 2030, 12840 Prague, Czech Republic. E-mail: stepnic@natur.cuni.cz The reaction of 19-(diphenylphosphino)ferrocenecarboxylic acid (Hdpf) with [{Rh(m-X)(CO)2}2] (X = Cl or Br) aVorded rhodium(I) complexes trans-[Rh(Hdpf-P)2X(CO)] containing the ligand as the P-bonded phosphine.On the other hand, pentane-2,4-dionato rhodium(I) complexes reacted with Hdpf by an acid–base reaction yielding novel O,P-chelated rhodium(I) complexes and pentane-2,4-dione (Hacac). The compound [Rh(acac)(CO)2] reacted with 2 equivalents of Hdpf to give trans-[Rh(dpf-O,P)(Hdpf-P)(CO)] which exhibits proton exchange between the two forms of the ligand.Likewise, related complexes [Rh(acac)(PR3)(CO)], where PR3 = PCy3, PPh3 or PPh2Fc (Cy = cyclohexyl, Fc = ferrocenyl), aVorded the corresponding complexes trans-[Rh(dpf-O,P)- (PR3)(CO)]. The formation of the O,P chelates is regioselective and might be considered as a rather unusual displacement of pentane-2,4-dionate by the phosphinocarboxylate dpf2 with concurrent proton exchange. All the compounds were characterized by 1H, 13C, 31P and IR spectroscopies and by FAB mass spectrometry.The crystal structure determination of trans-[Rh(dpf-O,P)(PCy3)(CO)] confirmed the presence of an unprecedented heteroannular O,P-chelating ferrocene ligand. Hybrid ligands that possess simultaneously hard and soft donor groups according to the Pearson’s hard/soft acid/base (HSAB) concept are known to co-ordinate transition metals in a variety of manners. The bond between the hard donor atom and a soft metal may readily be cleaved to produce a free coordination site while the ligand remains bonded to the metal centre by its soft donor group.For this reason, the complexes of catalytically active metals such as RhI, PtII and PdII with hybrid phosphines have been widely used as catalysts.1 In addition, the catalytic properties of hybrid ligands may be finely tuned by changing stereoelectronic properties of substituents and/or the backbone of the ligand. Furthermore, hybrid ligands may bear substituents suitable for attachment of the ligand to a solid support.The use of redox active (ferrocene-based) ligands enables one to assemble redox active groups at the surface.2 Recently, we reported the synthesis of the organometallic carboxyphosphine 19-(diphenylphosphino)ferrocenecarboxylic acid [Fe(h5-C5H4PPh2)(h5-C5H4CO2H)] (Hdpf) 3 and of its complexes with PdII and PtII 4 in which the ligand behaves as the P-donor, the unco-ordinated carboxyl group being involved in hydrogen bonding of various types.In order to force the O,P-chelation of this ligand, we have studied its reactivity towards pentane-2,4-dionato complexes of RhI in analogy to the reaction of a simpler ligand, (diphenylphosphino)acetic acid, which forms O,P-chelated (diphenylphosphino)acetato complexes.5,6 Transition-metal complexes containing chelating ferrocene ligands appear to be limited almost exclusively to P,P-donors such as 1,19-bis(diphenylphosphino)ferrocene or N,P ligands of the 1-(dialkylamino)methyl-2-phosphinoferrocene family with much less attention being paid to ferrocenebased O,P-donors in general.In this paper we report syntheses and spectral characterization of rhodium(I) complexes containing the O,P-chelating dpf2, [Rh(dpf-O,P)L(CO)], where L = Hdpf-P, PCy3, PPh3 or PPh2Fc (Cy = cyclohexyl, Fc = ferrocenyl) and of the analogous complexes trans-[Rh(Hdpf-P)2- X(CO)] (X = Cl or Br) containing Hdpf as the P-bonded phosphine. The crystal structure of trans-[Rh(dpf-O,P)(PCy3)(CO)] as representative of heteroannular O,P chelation of the ferrocene ligand is also presented.Results and discussion Rhodium(I) complexes with Hdpf as the P-bonded phosphine The complexes trans-[Rh(Hdpf-P)2X(CO)], where X = Cl 1 or Br 2, were synthesized by cleavage of the halogeno bridges in the [{Rh(m-X)(CO)2}2] dimers with a stoichiometric amount of Hdpf in benzene (Scheme 1). They were characterized by elemental analyses, FAB mass spectrometry and 1H, 13C, 31P NMR and IR spectroscopies.The 31P NMR spectra of 1 and 2 exhibit one doublet shifted downfield in comparison to the free Hdpf. Coupling constants 1J(RhP) 126 Hz for both complexes Scheme 1 Hacac = pentane-2,4-dione. [Rh(acac)(CO)2] 2 Hdpf –Hacac, –CO trans-[Rh(dpf- O,P)(Hdpf- P)(CO)] [Rh(acac)(PR3)(CO)] PR3 2 Hdpf trans-[Rh(dpf- O,P)(PR3)(CO)] 3 PR3 = PCy3 PPh3 PPh2Fc 4 5 6 trans-[Rh(Hdpf- P)2X(CO)] 2 Hdpf X = Cl Br 12 –Hacac, –CO –CO –CO 1/2[{Rh(m–X)2(CO)2}2] PPh2 CO2H Fe Hdpf2808 J.Chem. Soc., Dalton Trans., 1998, Pages 2807–2811 evidence the trans configuration of the phosphine ligands.7 In the 13C NMR spectra, the signals of the phosphinylated cyclopentadienyl ring and those of the phenyl rings (o-, m-CH) appear as apparent triplets of AA9X spin systems (A = P, X = C) typical for trans-bis(phosphine) complexes with large J(PP) values.8 On the contrary, the signal of the terminal carbonyl group was observed as a regular doublet of triplets at dC 186.9 (186.3) with 1J(RhC) 74 (76) and 2J(PC) 16 (16) Hz for 1 and 2, respectively.These values are close to those reported for other mononuclear rhodium(I) carbonyl complexes.9 Fouriertransform IR spectra of 1 and 2 exhibit bands of terminal carbonyl groups at 1957 cm21 as well as nC]] O stretching frequencies of the protonated carboxyl groups at 1674 cm21. The latter values indicate that the carboxyl groups are involved in hydrogen bonding (cf. 1696–1703 cm21 for Hdpf). In the positive-ion FAB mass spectra measured in m-nitrobenzyl alcohol matrix the molecular ions, ions [M 2 CO]1 and ions due to elimination of the halogen atom [M 2 X]1, m/z 959, are observed. As the latter are isobaric with [3 1 H]1, further ions in the mass spectra of 1 and 2 are the same as those originating from 3 under the same conditions: m/z [(Hdpf)Rh(dpf)]1, 849 [959 2 C6H6O2]1, 821 [849 2 CO]1, 545 [(Hdpf)Rh(CO)]1 and 414 [Hdpf]1. Complexes with O,P-chelating 19-(diphenylphosphino)ferrocenecarboxylate The pentane-2,4-dionato complex [Rh(acac)(CO)2] reacts with diVerent phosphine and phosphite ligands by substitution of either one or two CO molecules to give complexes [Rh- (acac)Ln(CO)22n] (n = 1 or 2), the course of the reaction (n) depending on the stereoelectronic properties of the ligand applied.10 On the other hand, (diphenylphosphino)acetic acid replaces pentane-2,4-dione and one CO molecule yielding [Rh(Ph2PCH2CO2-O,P)(Ph2PCH2CO2H-P)(CO)].Similarly, the reaction of [Rh(acac)(CO)2] with 2 equivalents of 19- (diphenylphosphino)ferrocenecarboxylic acid in hot toluene aVorded cinnamon orange trans-[Rh(dpf-O,P)(Hdpf-P)(CO)] 3 (see Scheme 1). In the IR spectrum of 3 the carbonyl stretching frequency nC]] ] O appears at 1962 cm21. The bands at 1703 and 1551 cm21 were assigned to protonated and chelating carboxyl groups respectively. Unexpectedly, 1H, 13C and 31P NMR spectra of this mixed Hdpf–dpf complex display only one set of ligand resonances indicating the equivalence of both phosphine ligands due to the proton exchange which is fast on the NMR timescale at 294 K.Its 31P NMR spectrum exhibits one doublet at dP 20.2 (co-ordination shift, DP 37.8 ppm), i.e. roughly halfway between that observed for Hdpf-P (dP 22.2) and other dpf- O,P (dP ca. 18.6) complexes. The phosphine ligands occupy trans positions in the square-planar environment around RhI as deduced from the 1J(RhP) coupling constant.Similarly to 1 and 2, the 13C NMR spectra show apparent triplets due to AA9X spin systems, whereas the signal of the terminal carbonyl group is observed as a normal doublet of triplets with dC 189.8, 1J(RhC) 75 and 2J(PC) 17 Hz. In [2H6]dimethyl sulfoxide solution the signals in the NMR spectra of complex 3 are significantly broader, most likely as the result of a lowered rate of the chemical exchange (Scheme 2) on introduction of the good hydrogen-bond acceptor.The 31P and 1H NMR spectra of the mixtures with [Rh(acac)(CO)2] :Hdpf ratios of 1 : 1, 1 : 2 and 1 : 3 in CDCl3 demonstrated that the formation of 3 is fast and proceeds with the displacement of pentane-2,4-dione even at 1 : 1 molar ratio where 0.5 equivalent of the parent rhodium(I) complex remains. Addition of the second equivalent of Hdpf completes the reaction and the third equivalent remains unconsumed. No further 1H NMR signals were observed down to dH 230 in this system.In order to eliminate the factor of proton exchange, we synthesized analogous complexes containing other monodentate phosphines in the place of undissociated Hdpf. The complexes [Rh(dpf-O,P)L(CO)], where L = PCy3 4, PPh3 5 or PPh2Fc 6, were all prepared in a similar manner, i.e. by mixing solutions of Hdpf and the corresponding pentane-2,4-dionato complex [Rh(acac)L(CO)] in hot butan-2-one followed by cooling. The IR spectra of 4–6 exhibit a strong band for terminal carbonyl between 1961 and 1965 cm21 and the carboxylate band in the range 1567–1606 cm21.The 1H, 13C and 31P NMR data are consistent with the proposed structures. The degeneracy of the A9AX spin system (to give A2X; A = P, X = C) observed in the case of 3 due to the proton exchange is removed and the 31P NMR signals of 4–6 appear as double doublets of ABM systems (A, B = P; M = Rh). The coupling constants 2J(PP) ª 325– 353 Hz and 1J(RhP) ª 128–134 Hz imply that the phosphines are mutually trans. Similarly, the signals of the phosphinylated cyclopentadienyl and phenyl rings in the 13C NMR spectra appear as simple doublets (or dd) located in the usual range.In 1H NMR spectra the resonances associated with the cyclopentadienyl hydrogen atoms of 3–6 are observed as ill resolved apparent multiplets of AA9BB9X and AA9BB9 spin systems (A, B = H, X = P) for phosphinylated and carboxylated cyclopentadienyls, respectively. The FAB MS spectra of complexes 3–6 in a m-nitrobenzyl alcohol matrix display protonated molecular ions [M 1 H]1, ions due to loss of carbon monoxide, i.e.[M 2 CO]1 for 3 and [M 2 CO 1 H]1 for 4–6. Another common feature is the presence of peaks due to free (Hdpf for all compounds, FcPPh2) or protonated phosphines (Cy3PH1 and Ph3PH1) and species at m/z 545, [(Hdpf)Rh(CO)]1. Ions at m/z 849 due to loss of the carboxylated cyclopentadienyl ring were observed for 3, while the spectra of complexes 4–6 display ions [M 2 72]1 reflecting most likely the simultaneous loss of CO and CO2. Crystal structure of complex 4 Complex 4 crystallizes in space group P1� with one discrete molecule in the asymmetric unit.The structure is shown in Fig. 1. Selected bond lengths and angles are given in Table 1. The four donor atoms form a square-planar environment around RhI. The perpendicular distance of Rh from the ligand Fig. 1 Molecular structure of trans-[Rh(dpf-O,P)(CO)(PCy3)] 4.Thermal ellipsoids are shown at the 30% probability level. The hydrogen atoms were omitted for clarity. Scheme 2 Rh P O P CO OH Rh P O P CO HOJ. Chem. Soc., Dalton Trans., 1998, Pages 2807–2811 2809 plane is 0.111(1) Å (cf. mean deviation of the plane defining atoms of 0.07 Å) and the sum of the bond angles around Rh is 3608. A slight opening of the O]Rh]P angle to 95.99(6)8 reflects the steric requirements of the chelating ligand. The Rh]P distances 2.335(1) and 2.342(1) Å are in keeping with those found in 1,19-bis(diphenylphosphino)ferrocene complexes [Rh(dppf-P, P9)(h4-nbd)] 11 and [Rh(dppf-P,P9)(MeCN)2] 12 of 2.335(2), 2.317(2) and 2.247(1), 2.232(1) Å, respectively.The dpf2 anion is further bonded through its deprotonated hydroxy oxygen atom thus completing the chelation. The geometry around the oxygen donor of the ligand is typical for a covalently bonded carboxylate: Rh]O 2.071(2), C]] O 1.232(4) and C]O 1.273(4) Å, O]C]] O 123.6(3)8 with the dihedral angle subtended by the carboxyl group and the co-ordination polyhedron defined by atoms P(1), O(2), P(2) and C(24) of 60.9(3)8.The bond lengths resemble that reported for the cationic h1-acetato complex cis-[NBun 4][Rh(O2CMe)2(CO)2] 13 and analogous chelates trans-[Pd(Ph2PCH2CO2-O,P)2] 14 and trans-[Rh(Ph2PCH2CO2- O,P)(Ph2PCH2CO2H-P)(CO)].5 The arrangement of the Rh]C]] ] O moiety is nearly linear with no unexceptional features when compared to trans-[Rh(PR3)2Cl(CO)] complexes 15 [Rh]C 1.793(3), C]] ] O 1.153(4) Å, Rh]C]] ] O 179.2(3)8].With respect to the solid-state structure of unco-ordinated Hdpf, the ferrocene moiety exhibits no significant deformation of its bond lengths and angles on chelate formation. The iron– centroid distance is 1.614(2) Å for both cyclopentadienyls (Cp). The Cp rings are slightly tilted at the dihedral angle of 4.8(3)8 and adopt an eclipsed conformation with syn-arranged substituents: the torsion angle P(1)]Cp1]Cp2]C(11) is 260.0(1)8.In contrast to the free ferrocene ligand, the carboxyl group is rotated towards its parent Cp plane by 25.9(4)8 as required by the O,P chelation. A similar syn-eclipsed conformation was observed in another complex of a chelating ferrocene derivative, [Rh{(h5-C5H4PPh2)Fe[h5-C5H4(2-C5H4N)]-N,P}- (h4-cod)].16 For the cases of dppf chelates mentioned above, however, the Cp rings are syn-staggered, most likely due to the absence of a ‘spacer’ between one of the two donor atoms directly bonded to the ferrocene framework.The cyclohexyl rings of the PCy3 ligand are bonded to phosphorus in equatorial positions and adopt an almost exact chair conformation with ring puckering coordinates 17 Q = 0.577(4), 0.582(4) and 0.566(5) Å and q = 0.0(4), 178.1(4) and 1.9(5)8 for Table 1 Selected bond lengths (Å), angles (8) and dihedral angles of least-squares planes a (8) with estimated standard deviations in parentheses for complex 4 Rh]P(1) Rh]P(2) Rh]O(2) Rh]C(24) O(3)]C(24) O(1)]C(11) O(2)]C(11) C]C (Fc) b C]C (Cy) b C(24)]Rh]O(2) C(24)]Rh]P(1) O(2)]Rh]P(1) C(24)]Rh]P(2) O(2)]Rh]P(2) P(1)]Rh]P(2) O(3)]C(24)]Rh Cp1 vs.Cp2 Cp1 vs. Ph1 Cp1 vs. Ph2 2.335(1) 2.342(1) 2.071(2) 1.793(3) 1.153(4) 1.232(4) 1.273(4) 1.425(7) 1.53(1) 175.8(1) 87.4(1) 95.99(6) 89.8(1) 86.48(6) 170.60(3) 179.2(3) 4.8(3) 77.6(1) 73.8(1) C(06)]C(11) P(1)]C(01) P(1)]C(12) P(1)]C(18) P(2)]C(31) P(2)]C(25) P(2)]C(37) C]C (Ph) b O(1)]C(11)]O(1) C(01)]P(1)]C(12) C(01)]P(1)]C(18) C(12)]P(1)]C(18) C(31)]P(2)]C(25) C(31)]P(2)]C(37) C(25)]P(2)]C(37) RhL vs.CO2 Cp2 vs. CO2 Ph1 vs. Ph2 1.498(4) 1.804(3) 1.820(3) 1.835(3) 1.850(3) 1.850(3) 1.855(3) 1.382(9) 123.6(3) 103.48(1) 102.45(1) 103.47(1) 110.9(2) 104.5(2) 104.8(2) 60.9(3) 25.9(4) 87.3(1) a The planes are defined as follows: Cp1, C(1)–C(5), phosphinylated cyclopentadienyl ring; Cp2, C(6)–C(10), carboxylated cyclopentadienyl ring; Ph1, C(12)–C(17); Ph2, C(18)–C(23); CO2, C(11), O(1), O(2); RhL, P(1), P(2), O(2), C(24).b Mean value. the rings involving C(25), C(31) and C(37), respectively. The P]C bonds of both phosphine ligands are almost perfectly eclipsed when looking along the P(1) ? ? ? P(2) line. As the result of fixing the positions of the substituents on the Cp rings by co-ordination, the dpf ligand exhibits conformational chirality. The RFc enantiomer chosen arbitrarily for the refinement is related to its enantiomeric counterpart through the crystallographic symmetry centre to form the racemic crystal.There are no significant intermolecular contacts below the sum of van der Waals radii between the molecules in the crystal. Conclusion The results described here exemplify the ability of 19- (diphenylphosphino)ferrocenecarboxylic acid to displace pentane-2,4-dionato ligand in its rhodium(I) complexes [Rh(acac)(CO)L] (L = CO or PR3) with concomitant proton transfer, aVording O,P-chelated complexes in good yields. In accordance with the trans eVect, only one regioisomer is formed in which the P-donors are mutually trans.The mechanism of the O,P-chelate formation might involve oxidative addition (with RhIII]H intermediates) or substitution wi h1-pentane- 2,4-dione as an intermediate. The reaction is relatively fast and no direct evidence of an intermediate was obtained. However, such intermediates could hardly be expected to be stable towards subsequent chelation.According to recent calorimetric data,18 the formation of [Rh(acac)(PR3)(CO)] from [Rh(acac)- (CO)2] is rapid and quantitative and the reaction enthalpy depends upon stereoelectronic properties of the incoming phosphine. Therefore, substitution of one of the carbonyl ligands by Hdpf may represent the first step of the formation of complex 3. Experimental General comments All manipulations were carried out in an argon atmosphere. The solvents were purified and dried by refluxing and distillation from potassium (benzene, toluene, diethyl ether) or standing over K2CO3 followed by distillation (butan-2-one).Light petroleum (fraction with bp 40–60 8C) and methanol were used as received. Infrared spectra were recorded in Nujol mulls between KBr plates on an FT IR Mattson Genesis instrument, 1H, 13C-{1H} and 31P-{1H} NMR spectra on a Varian UNITY Inova 400 spectrometer. Chemical shifts (d) are in ppm. Standards: internal tetramethylsilane (1H, 13C) or external 85% aqueous H3PO4 (31P).Coupling constants ( J) are given as absolute values. The assignment of the signals was based on 13C APT (attached proton test), COSY-45 and 13C HMQC (heteronuclear multiple quantum correlation) experiments. The multiplets are labelled as usual with ‘a’ indicating an apparent multiplet of a secondorder spin system. Positive ion FAB mass spectra in a mnitrobenzyl alcohol matrix were measured on a VG-7070E spectrometer (xenon fast atoms; 8 kV, 2 mA; accelerating voltage 6 kV).The spectra were interpreted by comparison of measured and simulated isotopic patterns. The mass of selected fragment ions given here corresponds to the isotopomer containing 79Br, 35Cl, 56Fe and 103Rh. The compounds [Rh(acac)(CO)2],19 [Rh(acac)(PR3)(CO)] 6 (R = Cy or Ph), [{Rh(m-X)(CO)2}2] (X = Cl 20 or Br21) and Hdpf3 were prepared by literature procedures. Preparations trans-[Rh(Hdpf-P)2Cl(CO)] 1. Following the general procedure,22 a solution of Hdpf (83.0 mg, 0.20 mmol) in hot benzene (4 cm3) was added to a solution of [{Rh(m-Cl)(CO)2}2] (19.4 mg, 0.050 mmol) in the same solvent (2 cm3).The2810 J. Chem. Soc., Dalton Trans., 1998, Pages 2807–2811 resulting clear orange solution was left to stand at room temperature overnight. The precipitate formed was filtered oV, washed with benzene (5 cm3) and light petroleum (3 × 5 cm3), and dried under reduced pressure to yield complex 1 as an orange solid.Yield 96.5 mg, 96% (Found: C, 56.55; H, 4.03. C47H38ClFe2O5P2Rh requires C, 56.75; H, 3.85%). IR (Nujol): n & /cm21 1957s, 1674s, 1299m, 1164m, 1097m, 1033m, 836m, 750m, 696m, 682m, 575m, 695s, 510s, 506s and 471m. NMR [(CD3)2SO, 298 K]: 1H, d 4.47 (2 H, br s, C5H4C CH), 4.51 (2 H, br at, C5H4P CH), 4.64 (2 H, at, C5H4C CH), 4.82 (2 H, at, C5H4C CH), 7.46–7.52 [6 H, m, P(C6H5)2], 7.58–7.66 [4 H, m, P(C6H5)2] and 12.38 (1 H, s, CO2H); 13C-{1H}, d 71.2 (s, C5H4C CH), 73.1 (s, C5H4C Cipso), 73.4 (at, C5H4C CH), 73.7 (at, C5H4P CH), 74.9 (at, C5H4P CH), 75.5 (at, C5H4P Cipso), 128.1 [at, P(C6H5)2 CHm], 130.2 [at, P(C6H5)2 CHp], 133.3 [at, P(C6H5)2 CHo], 133.7 [at, P(C6H5)2 Cipso], 171.3 (s, C]] O) and 186.9 [dt, J(RhC) 74, J(PC) 16 Hz, C]] ] O]; 31P-{1H}, d 22.2 [d, J(RhP) 126 Hz, Hdpf].FAB1: m/z 994, M1; 959, [M 2 Cl]1; 930, [(Hdpf)Rh(dpf)]1; 849, [959 2 C6H6O2 (i.e. C5H4CO2- H 1 H)]1; 821, [849 2 CO]1; 545, [(Hdpf)Rh(CO)]1; and 414, [Hdpf]1.trans-[Rh(Hdpf-P)2Br(CO)] 2. The reaction of [{Rh(m-Br)- (CO)2}2] (23.5 mg, 0.050 mmol) and Hdpf (83.0 mg, 0.20 mmol) was carried out using the same procedure as for 1. A similar work-up gave 2 as an orange solid. Yield 100.7 mg, 97% (Found: C, 54.68; H, 3.92. C47H38BrFe2O5P2Rh requires C, 54.32; H, 3.69%). IR (Nujol): n& /cm21 1957s, 1674s, 1297m, 1161m, 1096m, 1032m, 837m, 732m, 694m, 567m, 512s, 504s and 470m. NMR [(CD3)2SO, 298 K]: 1H, d 4.50 (2 H, br s, C5H4C CH), 4.52 (2 H, br at, C5H4P CH), 4.60 (2 H, at, C5H4C CH), 4.82 (2 H, at, C5H4C CH), 7.46–7.51 [6 H, m, P(C6H5)2], 7.59–7.66 [4 H, m, P(C6H5)2] and 12.40 (1 H, s, CO2H); 13C-{1H}, d 71.3 (s, C5H4C CH), 73.1 (s, C5H4C Cipso), 73.3 (at, C5H4C CH), 73.6 (at, C5H4P CH), 75.1 (at, C5H4P CH), 75.8 (at, C5H4P Cipso), 128.0 [at, P(C6H5)2 CHm], 130.2 [at, P(C6H5)2 CHp], 133.3 [at, P(C6H5)2 CHo], 134.2 [at, P(C6H5)2 Cipso], 171.3 (s, C]] O) and 186.3 [dt, J(RhC) 76, J(PC) 16 Hz, C]] ] O]; 31P-{1H}, d 22.2 [d, J(RhP) 126 Hz, Hdpf].FAB1: m/z 1038, M1; 1010, [M 2 CO]1; 959, [M 2 Br]1; 930, [(Hdpf)- Rh(dpf)]1; 849, [959 2 C6H6O2 (i.e. C5H4CO2H 1 H)]1; 821, [849 2 CO]1; 545, [(Hdpf)Rh(CO)]1; and 414, [Hdpf]1. [Rh(acac)(PPh2Fc)(CO)]. A slurry of [Rh(acac)(CO)2] (258 mg, 1.00 mmol) and FcPPh2 (408 mg, 1.10 mmol) in diethyl ether (20 cm3) was heated to boiling until a clear orange solution resulted (CO evolution). Methanol (15 cm3) was added and the volume was reduced to ca. 10 cm3 in vacuo. The resulting precipitate was filtered oV, washed with a little methanol and dried in air to give the complex as an orange microcrystalline solid. Yield 524 mg, 87% (Found: C, 56.06; H, 4.36. C28H26- FeO3PRh requires C, 56.03; H, 4.37%). IR (Nujol): n& /cm21 1960s, 1574s, 1567s, 1524s, 1275m, 1164m, 1097m, 1040m, 821m, 748m, 745m, 696s, 524m, 497s and 469m. NMR (CDCl3, 294 K): 1H, d 1.62 (3 H, s, CH3), 2.11 (3 H, s, CH3), 4.23 (5 H, s, C5H5), 4.39 (2 H, aq, C5H4), 4.45 (2 H, m, C5H4), 7.31–7.43 [6 H, m, P(C6H5)2] and 7.61–7.69 [4 H, m, P(C6H5)2]; 13C-{1H}, d 26.6 (s, CH3), 27.6 [d, J(PC) 6, CH3], 70.0 (s, C5H5), 71.0 [d, J(PC) 8, C5H4 CH], 74.2 [d, J(PC) 11, C5H4 CH], 75.1 [d, J(PC) 60, C5H4P Cipso], 100.7 [d, J(PC) 2, ]] CH]], 127.6 [d, J(PC) 11, P(C6H5)2 CHm], 129.9 [d, J(PC) 2, P(C6H5)2 CHp], 133.9 [d, J(PC) 112, P(C6H5)2 CHo], 134.6 [d, J(PC) 52, P(C6H5)2 Cipso], 185.2 (s, C]] O), 187.5 (s, C]] O) and 189.8 [dd, J(RhC) 76, J(PC) 25 Hz, C]] ] O]; 31P-{1H}, d 42.5 [d, J(RhP) 176 Hz, FcPPh2].FAB1: m/z 600, M1; 572, [M 2 CO]1; 501, [M 2 acac]1; and 370, [FcPPh2]1. trans-[Rh(dpf-O,P)(Hdpf-P)(CO)] 3. Complex [Rh(acac)- (CO)2] (51.6 mg, 0.20 mmol) and Hdpf (166 mg, 0.40 mmol) were suspended in toluene (10 cm3). A vigorous gas evolution (CO) was observed instantly. The mixture was heated to boiling and the resulting clear orange solution was cooled to room temperature and left to stand at 0 8C overnight.Filtration, washing with diethyl ether (3 × 5 cm3) and light petroleum (3 × 5 cm3), and drying in vacuo aVorded 3 as a cinnamon orange powder. Yield 179 mg, 93% (Found: C, 58.80; H, 3.98. C47H37Fe2O5P2Rh requires C, 58.90; H, 3.89%). IR (Nujol): n& /cm21 1962s, 1703s, 1551s, 1348m, 1255m, 1164m, 1097m, 1030m, 834m, 695s and 503s. NMR (CDCl3, 294 K): 1H, d 4.42 (2 H, br s, C5H4C CH), 4.50 (4 H, m, C5H4P CH), 4.99 (2 H, at, C5H4C CH), 7.33–7.42 [6 H, m, P(C6H5)2] and 7.65–7.75 [4 H, m, P(C6H5)2]; 13C-{1H}, d 72.3 (s, C5H4C CH), 72.5 (s, C5H4C CH), 72.9 (at, C5H4P Cipso), 73.2 (at, C5H4P CH), 74.9 (at, C5H4P CH), 76.5 (s, C5H4C Cipso), 128.4 [at, P(C6H5)2 CHm], 130.4 [at, P(C6H5)2 CHp], 133.2 [at, P(C6H5)2 Cipso], 133.7 [at, P(C6H5)2 CHo], 175.2 (s, C]] O) and 189.8 [dt, J(RhC) 75, J(PC) 17 Hz, C]] ] O]; 31P-{1H}, d 20.2 [d, J(RhP) 132 Hz, Hdpf and dpf].FAB1: m/z 959, [M 1 H]1; 930, [M 2 CO]1; 849, [M 2 C6H5O2]1; 821, [M 2 C6H5O2 2 CO]1; 545, [(Hdpf)- Rh(CO)]1; and 414, [Hdpf]1.trans-[Rh(dpf-O,P)(PCy3)(CO)] 4. The complex [Rh(acac)- (PCy3)(CO)] (256 mg, 0.50 mmol) was dissolved in boiling butan-2-one (5 cm3) and a solution of Hdpf (217 mg, 0.52 mmol) in the same solvent (2 cm3) was added. The mixture was refluxed for 10 min, cooled to room temperature and left to stand at 0 8C overnight. Filtration, washing with cold butan-2- one (2 cm3) and drying in air aVorded 4 as a bright yellow crystalline solid. Yield 368 mg, 89% (Found: C, 60.91; H, 6.29.C42H51FeO3P2Rh requires C, 61.18; H, 6.23%). IR (Nujol): n& /cm21 1961s, 1602s, 1582m, 1567m, 1321s, 1177m, 1164m, 1095m, 1031m, 849m, 813m, 782m, 755m, 693m, 591m, 509s and 469m. NMR (CDCl3, 294 K): 1H, d 1.20–2.32 [33 H, m, P(C6H11)3], 4.06 (2 H, aq, C5H4P CH), 4.28 (2 H, at, C5H4C CH), 4.43 (2 H, at, C5H4P CH), 5.37 (2 H, at, C5H4C CH), 7.36–7.44 [6 H, m, P(C6H5)2] and 7.69–7.78 [4 H, m, P(C6H5)2]; 13C-{1H}, d 26.6 (s, C6H11P g-CH2), 27.6 [d, J(PC) 11 Hz, C6H11P a-CH2], 30.2 (s, C6H11P b-CH2), 33.6 [d, J(PC) 11, C6H11P CH], 70.9 (s, C5H4C CH), 71.8 [d, J(PC) 5 Hz, C5H4P CH), 72.7 [dd, J(PC) 10 and 2, C5H4P Cipso], 74.0 (s, C5H4C CH), 75.8 [d, J(PC) 10, C5H4P CH), 78.7 (s, C5H4C Cipso), 128.2 [d, J(PC) 10, P(C6H5)2 CHm], 130.2 [d, J(PC) 2, P(C6H5)2 CHp], 133.2 [d, J(PC) 12, P(C6H5)2 CHo], 133.7 [d, J(PC) 42, P(C6H5)2 Cipso], 174.7 (s, C]] O) and 190.5 [dt, J(RhC) 72, J(PC) 17 Hz, C]] ] O]; 31P-{1H}, d 18.4 [dd, J(RhP) 128, J(PP) 325, dpf] and 41.8 [dd, J(RhP) 127, J(PP) 325 Hz, PCy3].FAB1: m/z 825, [M 1 H]1; 797, [M 2 CO 1 H]1; 752, [M 2 CO 2 CO2]1; 545, [(Hdpf)- Rh(CO)]1; and 281, [PCy3H]1. trans-[Rh(dpf-O,P)(PPh3)(CO)] 5. The complex [Rh(acac)- (PPh3)(CO)] (246 mg, 0.50 mmol) and Hdpf (217 mg, 0.52 mmol) were treated in a similar fashion as for 4 to give 5 as an orange solid. Yield 251 mg, 62% (Found: C, 62.12; H, 4.27. C42H33FeO3P2Rh requires C, 62.56; H, 4.12%). IR (Nujol): n& /cm21 1965s, 1606m, 1583m, 1326m, 1164m, 1096m, 1027m, 743m, 694s and 510s.NMR (CDCl3, 294 K): 1H, d 4.15 (2 H, at, C5H4C CH), 4.35 (2 H, aq, C5H4P CH), 4.47 (2 H, at, C5H4P CH), 4.83 (2 H, at, C5H4C CH), 7.30–7.44 [15 H, m, P(C6H5)2 and P(C6H5)3], 7.63–7.78 [10 H, m, P(C6H5)2 and P(C6H5)3]; 13C-{1H}, d 70.5 (s, C5H4C CH), 71.0 [d, J(PC) 48, C5H4P Cipso], 71.9 [d, J(PC) 7, C5H4P CH], 72.3 (s, C5H4C CH), 74.6 [d, J(PC) 10, C5H4P CH], 80.7 (s, C5H4C Cipso), 128.3 [d, J(PC) 10, P(C6H5)2 CHm], 128.5 [d, J(PC) 10, P(C6H5)3 CHm], 130.3 [d, J(PC) 2, P(C6H5)2 CHp], 130.5 [d, J(PC) 2, P(C6H5)3 CHp], 131.4 [dd, J(PC) 42 and 2, P(C6H5)3 Cipso], 133.9 [d, J(PC) 44, P(C6H5)2 Cipso], 133.9 [d, J(PC) 13, P(C6H5)2 CHo], 134.4 [d, J(PC) 12, P(C6H5)3 CHo], 175.1 (s, C]] O) and 189.9 [dt, J(RhC) 71, J(PC) 18 Hz, C]] ] O]; 31P-{1H}, d 18.8 [dd, J(RhP) 134, J(PP) 344, dpf] and 28.1 [dd, J(RhP) 131, J(PP) 344 Hz, PPh3].FAB1: m/z 807, [M 1 H]1; 779, [M 2 CO 1 H]1; 734, [M 2 CO 2J.Chem. Soc., Dalton Trans., 1998, Pages 2807¡V2811 2811CO2]1; 545, [(Hdpf)Rh(CO)]1; 414, [Hdpf]1; and 263,[PPh3H]1.trans-[Rh(dpf-O,P)(PPh2Fc)(CO)] 6. Starting from [Rh-(acac)(PPh2Fc)(CO)] (301 mg, 0.50 mmol) and Hdpf (217 mg,0.52 mmol), the same procedure as for complex 4 aVorded 6 asan orange microcrystalline solid. Yield 193 mg, 42% (Found: C,60.51; H, 4.15. C46H37Fe2O3P2Rh requires C, 60.43; H, 4.08%).IR (Nujol): n& /cm21 1961s, 1603m, 1583m, 1324m, 1164m,1096m, 1028m, 744m, 694s, 586m and 497s.NMR (CDCl3, 294K): 1H, d 4.12 (2 H, at, C5H4C CH), 4.31 (2 H, aq, C5H4P CH,FcPPh2 or dpf), 4.39 (5 H, s, C5H5 FcPPh2), 4.41 (2 H, m,C5H4P CH, FcPPh2 or dpf), 4.46 (4 H, m, C5H4P CH, FcPPh2or dpf), 4.83 (2 H, at, C5H4C CH), 7.31¡V7.43 [12 H, m, P(C6H5)2,FcPPh2 or dpf] and 7.64¡V7.81 [8 H, m, P(C6H5)2, FcPPh2 ordpf]; 13C-{1H}, d 70.1 (s, C5H5 FcPPh2), 70.5 (s, C5H4C CH),71.0 [dd, J(PC) 47, unresolved J, C5H4P Cipso, FcPPh2 or dpf],71.5 [d, J(PC) 7, C5H4P CH, FcPPh2 or dpf], 71.9 [d, J(PC) 6,C5H4P CH, FcPPh2 or dpf], 72.6 (s, C5H4C CH), 73.1 [dd,J(PC) 47, J 5, C5H4P Cipso, FcPPh2 or dpf], 74.2 [d, J(PC) 11,C5H4P CH, FcPPh2 or dpf], 74.7 [d, J(PC) 10, C5H4P CH,FcPPh2 or dpf], 80.4 (s, C5H4C Cipso), 128.1 [d, J(PC) 10,P(C6H5)2 CHm, FcPPh2 or dpf], 128.3 [d, J(PC) 10, P(C6H5)2CHm, FcPPh2 or dpf], 130.1 [d, J(PC) 2, P(C6H5)2 CHp, FcPPh2or dpf], 130.3 [d, J(PC) 2, P(C6H5)2 CHp, FcPPh2 or dpf], 133.2[dd, J(PC) 44, J 2, P(C6H5)2 Cipso, FcPPh2 or dpf], 133.8 [d,J(PC) 8, P(C6H5)2 CHo, FcPPh2 or dpf], 133.9 [d, J(PC) 8,P(C6H5)2 CHo, FcPPh2 or dpf], 134.0 [dd, J(PC) ca. 40, J 2,P(C6H5)2 Cipso, FcPPh2 or dpf], 175.2 (s, C]]O) and 190.3 [dt,J(RhC) 72, J(PC) 18 Hz, C]] ]O]; 31P-{1H}, d 18.7 [dd, J(RhP)133, J(PP) 353, dpf] and 24.0 [dd, J(RhP) 132, J(PP) 353 Hz,PPh3]. FAB1: m/z 915, [M 1 H]1; 887, [M 2 CO 1 H]1; 894,[M 2 C5H5]1; 842, [M 2 CO 2 CO2]1; 545, [(Hdpf)Rh(CO)]1;and 370, [PPh2Fc]1.X-Ray crystallographyCrystal data and intensity collection parameters.C42H51FeO3-P2Rh, M = 824.5, triclinic, space group P1(no. 2), a = 9.901(1),b = 12.875(5), c = 15.427(2) , a = 96.69(1), b = 101.28(1),g = 94.46(1)8, U = 1905.2(8) 3 (by least squares from 25automatically centered diVractions with 24 < 2q < 288), T =150.0(1) K, graphite-monochromated Mo-Ka radiation,l = 0.710 73 , Z = 2, Dc = 1.437 g cm23, F(000) = 856,m(Mo-Ka) = 0.94 mm21, yellow prism grown by a slow coolingof a hot butan-2-one solution, dimensions 0.1 ¡Ñ 0.2 ¡Ñ 0.4 mm,Enraf-Nonius CAD4 four circle diVractometer, q¡V2q scan, datacollection range 211 < h < 11, 0 < k < 14, 217 < l < 17(2qmax = 488), variation of three periodically measured standarddiVractions 4.7%; 5963 unique diVractions were measured(Rs = 0.018) and used in all calculations.The data were correctedfor Lorentz-polarization eVects.Structure solution and refinement. The structure was solvedby direct methods (SIR 92, ref. 23) and refined by full-matrixleast squares on F2 (SHELXL 97, ref. 24). Weighting schemew = [s2(Fo2) 1 (w1P)2 1 w2P]21, where P = [max(Fo2) 1 2Fc2]/3,w1 = 0.0702 and w2 = 2.4986 was applied. All non-hydrogenatoms were refined anisotropically. The hydrogen atoms wereincluded in calculated positions and then freely isotropicallyrefined. Final R = 0.035 and R9 = 0.097 for 5377 observed diffractions[I > 2s(I)] and R = 0.039 and R9 = 0.101 (all data); 646parameters, largest D/s 0.001, goodness of fit 1.027, extremeson the residual electron density map 11.69 and 20.79 e 23.CCDC reference number 186/1077.See http://www.rsc.org/suppdata/dt/1998/2807/ for crystallographicfiles in .cif format.AcknowledgementsWe thank Mr. M.Pols£¾ek for FAB MS measurements andProfessor J. Podlaha for useful discussion and comments.Financial support from the Grant Agency of CharlesUniversity (grant No. 209/96/B) and Czech Republic (No. 203/97/0242) is gratefully acknowledged.References1 C.D. Frohning, C. W. Kohlpainter and H. Brunner, in AppliedHomogeneous Catalysis with Organometallic Compounds, eds.B. Cornils and W. A. Hermann, VCH, Weinheim, 1996, vol. 1,pp. 29¡V60, 201¡V219.2 R. D. Eagling, J. E. Bateman, N. J. Goodwin, W. Henderson, B. R.Horrocks and A. Houlton, J. Chem. Soc., Dalton Trans., 1998, 1273.3 J. Podlaha, P. S£¾ te£¾pnic£¾ka, I. Csar£¾ov and J. Ludvk, Organometallics,1996, 15, 543.4 P. S£¾ te£¾pnic£¾ka, J.Podlaha, R. Gyepes and M. Pols£¾ek, J. Organomet.Chem., 1998, 552, 293.5 A. Jegorov, B. Kratochvl, V. Langer and J. Podlahov, Inorg.Chem., 1984, 23, 4288.6 A. Jegorov, J. Podlaha, J. Podlahov and F. Turec£¾ek, J. Chem. Soc.,Dalton Trans., 1990, 3259.7 B. E. Mann, C. Masters and B. L. Shaw, J. Chem. Soc. A, 1971,1104.8 A. W. Verstuyft, J. H. Nelson and L. W. Carry, Inorg. Chem., 1976,15, 732 and refs. therein.9 L. S. Bresler, N. A. Buzina, Yu. S. Varshavky, N. V. Kiseleva andT.G. Cherkasova, J. Organomet. Chem., 1979, 171, 229.10 A. M. Trzeciak, T. Gowiak, R. Grzybek and J. J. Ziokowski,J. Chem. Soc., Dalton Trans., 1997, 1831 and refs. therein.11 W. R. Cullen, T.-J. Kim, F. W. B. Einstein and T. Jones,Organometallics, 1985, 4, 346.12 H. Wang, R. J. Barton and B. E. Robertson, Acta Crystallogr., Sect.C, 1991, 47, 504.13 A. Fulford, N. A. Bailey, H. Adams and P. M. Maitlis,J. Organomet. Chem., 1991, 417, 139.14 S. Civis£¾, J. Podlahov, J. Loub and J.Jec£¾n, Acta Crystallogr., Sect.B, 1980, 36, 1395.15 K. R. Dunbar and S. C. Haefner, Inorg. Chem., 1992, 31, 3676; A. L.Rheingold and S. J. Geib, Acta Crystallogr., Sect. C, 1987, 43, 784;F. Dahan and R. Choukroun, Acta Crystallogr., Sect. C, 1985, 41,704; S. E. Boyd, L. D. Field, T. W. Hambley and M. G. Partridge,Organometallics, 1993, 12, 1720.16 T. Yoshida, K. Tani, T. Yamagata, Y. Tatsuno and T. Saito, J. Chem.Soc., Chem. Commun., 1990, 292.17 D. Cremer and J. A. Pople, J.Am. Chem. Soc., 1975, 97, 1354.18 S. Serron, J. Huang and S. P. Nolan, Organometallics, 1998, 17, 534.19 Yu. S. Varshavskii and T. G. Cherkasova, Zh. Neorg. Khim., 1967,12, 1709.20 J. A. McCleverty and G. Wilkinson, Inorg. Synth., ed. H. F.Holtzclaw, jun., McGraw-Hill, New York, 1966, vol. 8, pp. 211¡V214.21 B. F. G. Johnson, J. Lewis, P. W. Robinson and J. R. Miller, J. Chem.Soc. A, 1969, 2693.22 L. Vallarino, J. Chem. Soc., 1957, 2287; J. A. McCleverty andG. Wilkinson, Inorganic Syntheses, ed.H. F. Holtzlaw, McGraw-Hill, New York, 1966, vol. 8, pp. 214¡V217.23 A. Altomare, M. C. Burla, M. Camalli, G. Cascarano,C. Giacovazzo, A. Guagliardi and G. Polidori, J. Appl. Crystallogr.,1994, 27, 435.24 G. M. Sheldrick, SHELXL 97, Program for Crystal StructureRefinement from DiVraction Data, University of Gttingen, 1997.Received 19th May 1998; Paper 8/03743DJ. Chem. Soc., Dalton Trans., 1998, Pages 2807¡V2811 2811CO2]1; 545, [(Hdpf)Rh(CO)]1; 414, [Hdpf]1; and 263,[PPh3H]1.trans-[Rh(dpf-O,P)(PPh2Fc)(CO)] 6.Starting from [Rh-(acac)(PPh2Fc)(CO)] (301 mg, 0.50 mmol) and Hdpf (217 mg,0.52 mmol), the same procedure as for complex 4 aVorded 6 asan orange microcrystalline solid. Yield 193 mg, 42% (Found: C,60.51; H, 4.15. C46H37Fe2O3P2Rh requires C, 60.43; H, 4.08%).IR (Nujol): n& /cm21 1961s, 1603m, 1583m, 1324m, 1164m,1096m, 1028m, 744m, 694s, 586m and 497s. NMR (CDCl3, 294K): 1H, d 4.12 (2 H, at, C5H4C CH), 4.31 (2 H, aq, C5H4P CH,FcPPh2 or dpf), 4.39 (5 H, s, C5H5 FcPPh2), 4.41 (2 H, m,C5H4P CH, FcPPh2 or dpf), 4.46 (4 H, m, C5H4P CH, FcPPh2or dpf), 4.83 (2 H, at, C5H4C CH), 7.31¡V7.43 [12 H, m, P(C6H5)2,FcPPh2 or dpf] and 7.64¡V7.81 [8 H, m, P(C6H5)2, FcPPh2 ordpf]; 13C-{1H}, d 70.1 (s, C5H5 FcPPh2), 70.5 (s, C5H4C CH),71.0 [dd, J(PC) 47, unresolved J, C5H4P Cipso, FcPPh2 or dpf],71.5 [d, J(PC) 7, C5H4P CH, FcPPh2 or dpf], 71.9 [d, J(PC) 6,C5H4P CH, FcPPh2 or dpf], 72.6 (s, C5H4C CH), 73.1 [dd,J(PC) 47, J 5, C5H4P Cipso, FcPPh2 or dpf], 74.2 [d, J(PC) 11,C5H4P CH, FcPPh2 or dpf], 74.7 [d, J(PC) 10, C5H4P CH,FcPPh2 or dpf], 80.4 (s, C5H4C Cipso), 128.1 [d, J(PC) 10,P(C6H5)2 CHm, FcPPh2 or dpf], 128.3 [d, J(PC) 10, P(C6H5)2CHm, FcPPh2 or dpf], 130.1 [d, J(PC) 2, P(C6H5)2 CHp, FcPPh2or dpf], 130.3 [d, J(PC) 2, P(C6H5)2 CHp, FcPPh2 or dpf], 133.2[dd, J(PC) 44, J 2, P(C6H5)2 Cipso, FcPPh2 or dpf], 133.8 [d,J(PC) 8, P(C6H5)2 CHo, FcPPh2 or dpf], 133.9 [d, J(PC) 8,P(C6H5)2 CHo, FcPPh2 or dpf], 134.0 [dd, J(PC) ca. 40, J 2,P(C6H5)2 Cipso, FcPPh2 or dpf], 175.2 (s, C]]O) and 190.3 [dt,J(RhC) 72, J(PC) 18 Hz, C]] ]O]; 31P-{1H}, d 18.7 [dd, J(RhP)133, J(PP) 353, dpf] and 24.0 [dd, J(RhP) 132, J(PP) 353 Hz,PPh3]. FAB1: m/z 915, [M 1 H]1; 887, [M 2 CO 1 H]1; 894,[M 2 C5H5]1; 842, [M 2 CO 2 CO2]1; 545, [(Hdpf)Rh(CO)]1;and 370, [PPh2Fc]1.X-Ray crystallographyCrystal data and intensity collection parameters. C42H51FeO3-P2Rh, M = 824.5, triclinic, space group P1(no. 2), a = 9.901(1),b = 12.875(5), c = 15.427(2) , a = 96.69(1), b = 101.28(1),g = 94.46(1)8, U = 1905.2(8) 3 (by least squares from 25automatically centered diVractions with 24 < 2q < 288), T =150.0(1) K, graphite-monochromated Mo-Ka radiation,l = 0.710 73 , Z = 2, Dc = 1.437 g cm23, F(000) = 856,m(Mo-Ka) = 0.94 mm21, yellow prism grown by a slow coolingof a hot butan-2-one solution, dimensions 0.1 ¡Ñ 0.2 ¡Ñ 0.4 mm,Enraf-Nonius CAD4 four circle diVractometer, q¡V2q scan, datacollection range 211 < h < 11, 0 < k < 14, 217 < l < 17(2qmax = 488), variation of three periodically measured standarddiVractions 4.7%; 5963 unique diVractions were measured(Rs = 0.018) and used in all calculations.The data were correctedfor Lorentz-polarization eVects.Structure solution and refinement. The structure was solvedby direct methods (SIR 92, ref. 23) and refined by full-matrixleast squares on F2 (SHELXL 97, ref. 24). Weighting schemew = [s2(Fo2) 1 (w1P)2 1 w2P]21, where P = [max(Fo2) 1 2Fc2]/3,w1 = 0.0702 and w2 = 2.4986 was applied.All non-hydrogenatoms were refined anisotropically. The hydrogen atoms wereincluded in calculated positions and then freely isotropicallyrefined. Final R = 0.035 and R9 = 0.097 for 5377 observed diffractions[I > 2s(I)] and R = 0.039 and R9 = 0.101 (all data); 646parameters, largest D/s 0.001, goodness of fit 1.027, extremeson the residual electron density map 11.69 and 20.79 e 23.CCDC reference number 186/1077.See http://www.rsc.org/suppdata/dt/1998/2807/ for crystallographicfiles in .cif format.AcknowledgementsWe thank Mr.M. Pols£¾ek for FAB MS measurements andProfessor J. Podlaha for useful discussion and comments.Financial support from the Grant Agency of CharlesUniversity (grant No. 209/96/B) and Czech Republic (No. 203/97/0242) is gratefully acknowledged.References1 C. D. Frohning, C. W. Kohlpainter and H. Brunner, in AppliedHomogeneous Catalysis with Organometallic Compounds, eds.B.Cornils and W. A. Hermann, VCH, Weinheim, 1996, vol. 1,pp. 29¡V60, 201¡V219.2 R. D. Eagling, J. E. Bateman, N. J. Goodwin, W. Henderson, B. R.Horrocks and A. Houlton, J. Chem. Soc., Dalton Trans., 1998, 1273.3 J. Podlaha, P. S£¾ te£¾pnic£¾ka, I. Csar£¾ov and J. Ludvk, Organometallics,1996, 15, 543.4 P. S£¾ te£¾pnic£¾ka, J. Podlaha, R. Gyepes and M. Pols£¾ek, J. Organomet.Chem., 1998, 552, 293.5 A. Jegorov, B. Kratochvl, V. Langer and J. Podlahov, Inorg.Chem., 1984, 23, 4288.6 A. Jegorov, J. Podlaha, J. Podlahov and F. Turec£¾ek, J. Chem. Soc.,Dalton Trans., 1990, 3259.7 B. E. Mann, C. Masters and B. L. Shaw, J. Chem. Soc. A, 1971,1104.8 A. W. Verstuyft, J. H. Nelson and L. W. Carry, Inorg. Chem., 1976,15, 732 and refs. therein.9 L. S. Bresler, N. A. Buzina, Yu. S. Varshavky, N. V. Kiseleva andT. G. Cherkasova, J. Organomet. Chem., 1979, 171, 229.10 A. M. Trzeciak, T. Gowiak, R. Grzybek and J. J. Ziokowski,J. Chem. Soc., Dalton Trans., 1997, 1831 and refs. therein.11 W. R. Cullen, T.-J. Kim, F. W. B. Einstein and T. Jones,Organometallics, 1985, 4, 346.12 H. Wang, R. J. Barton and B. E. Robertson, Acta Crystallogr., Sect.C, 1991, 47, 504.13 A. Fulford, N. A. Bailey, H. Adams and P. M. Maitlis,J. Organomet. Chem., 1991, 417, 139.14 S. Civis£¾, J. Podlahov, J. Loub and J. Jec£¾n, Acta Crystallogr., Sect.B, 1980, 36, 1395.15 K. R. Dunbar and S. C. Haefner, Inorg. Chem., 1992, 31, 3676; A. L.Rheingold and S. J. Geib, Acta Crystallogr., Sect. C, 1987, 43, 784;F. Dahan and R. Choukroun, Acta Crystallogr., Sect. C, 1985, 41,704; S. E. Boyd, L. D. Field, T. W. Hambley and M. G. Partridge,Organometallics, 1993, 12, 1720.16 T. Yoshida, K. Tani, T. Yamagata, Y. Tatsuno and T. Saito, J. Chem.Soc., Chem. Commun., 1990, 292.17 D. Cremer and J. A. Pople, J. Am. Chem. Soc., 1975, 97, 1354.18 S. Serron, J. Huang and S. P. Nolan, Organometallics, 1998, 17, 534.19 Yu. S. Varshavskii and T. G. Cherkasova, Zh. Neorg. Khim., 1967,12, 1709.20 J. A. McCleverty and G. Wilkinson, Inorg. Synth., ed. H. F.Holtzclaw, jun., McGraw-Hill, New York, 1966, vol. 8, pp. 211¡V214.21 B. F. G. Johnson, J. Lewis, P. W. Robinson and J. R. Miller, J. Chem.Soc. A, 1969, 2693.22 L. Vallarino, J. Chem. Soc., 1957, 2287; J. A. McCleverty andG. Wilkinson, Inorganic Syntheses, ed. H. F. Holtzlaw, McGraw-Hill, New York, 1966, vol. 8, pp. 214¡V217.23 A. Altomare, M. C. Burla, M. Camalli, G. Cascarano,C. Giacovazzo, A. Guagliardi and G. Polidori, J. Appl. Crystallogr.,1994, 27, 435.24 G. M. Sheldrick, SHELXL 97, Program for Crystal StructureRefinement from DiVraction Data, University of Gttingen, 1997.Received 19th May 1998; Paper 8/03743D

ISSN:1477-9226    

DOI:10.1039/a803743d

出版商:RSC    

年代:1998

数据来源: RSC

摘要:

DALTON FULL PAPER J. Chem. Soc., Dalton Trans., 1998, Pages 2813–2817 2813 Synthesis of dirhenium species with benzamidate ligands via hydrolysis of benzonitrile † Cary B. Bauer,a Thomas E. Concolino,b Judith L. Eglin,*,b Robin D. Rogers c and Richard J. Staples d a Institute of Enzyme Research, University of Wisconsin, 1710 University Ave., Madison, WI 53705, USA b Department of Chemistry, Mississippi State University, Mississippi State, MS 39762, USA c Department of Chemistry, The University of Alabama, Tuscaloosa, AL 35487, USA d Department of Chemistry, Harvard University, Cambridge, MA 02138, USA Benzonitrile co-ordinated to a metal–metal bonded dirhenium core has been shown to undergo hydrolysis in ethanol solvent systems to form benzamidate and remains co-ordinated to the dirhenium core as a bridging ligand.The compounds [NBu4][Re2Cl6{m-PhC(O)NH}]?0.5CH2Cl2 1 and [Re2Cl4(m-dppm)2{m-PhC(O)NH}] 2, where dppm is bis(diphenylphosphino)methane, were synthesized from the reaction of [NBu4]2[Re2Cl8] in an ethanol solvent system with benzonitrile containing 3% water.While compound 1 is the hydrolysis product of the reaction of [NBu4]2[Re2Cl8] with benzonitrile, 2 arises from the reduction of 1 upon co-ordination of the bidentate phosphine ligand dppm. The structures of 1 and 2 were determined by X-ray crystallography and the compounds characterized by a variety of spectroscopic methods. The hydrolysis or partial hydrolysis of nitriles to carboxylates or amides under milder reaction conditions is a process that has been studied for many years, demonstrating the importance of this reaction in organic chemistry.1,2 In most syntheses, the reaction mixture is first made very basic via the addition of KOH and subsequently neutralized by the addition of a concentrated acid.3–6 However, these harsh conditions are not always viable in the presence of other functional groups and researchers have been examining new methods to synthesize amides for their use in the fine chemical and pharmaceutical industry.7 With the potential to limit the hydrolysis of nitriles to the formation of amides by co-ordination of the nitrogen atom, complexes with a number of transition metal centers such as Pt,7–15 Pd,9,16 Rh,17 Ir,17 Cu,18 Ni,19 W20 and Co21 are currently under investigation.In addition to the relevance of this research to organic synthesis, the enzymatic hydrolysis mechanisms of nitriles are under investigation to understand the chemo-, regioand stereo-selectivity of the reactions which are performed under mild conditions.22–25 For example, cobalt complexes as artificial enzymes have been widely studied for their ability to form amides from coordinated nitriles under neutral pH conditions.3–6,26,27 For [Co(cyclen)(CO)3][ClO4], where cyclen is 1,4,7,11-tetraazacyclododecane, studies have shown a diaqua complex [Co(cyclen)- (OH2)2]31 is formed prior to the co-ordination of benzonitrile and the subsequent hydrolysis of the nitrile results in the formation of a benzamidate intermediate.21 The resultant benzamidate intermediate in the cobalt complex is chelated to the metal center, as in the case of the dirhenium core of [NBu4]- [Re2Cl6{m-PhC(O)NH}]?0.5CH2Cl2 1 and [Re2Cl4(m-dppm)2- {m-PhC(O)NH}] 2 where dppm is bis(diphenylphosphino)- methane.27 Bimetallic or polymetallic cores are often found in the active site of hydrolytic enzymes, such as urease, phosphatases and esterases.22–25 The commonly accepted rationale, termed ‘twometal mechanism’ by Steitz and Steitz,28 is that the transfer of positive charge from one metal center enhances the electrophilicity of the bound substrate, and the second metal ion facilitates the deprotonation of the co-ordinated water to yield a † Non-SI unit employed: mB ª 9.27 × 10224 J T21.bound hydroxide that serves as an intramolecular nucleophile. Co-operativity between two metal centers is not limited to enzymes and has also been proposed as the key of dirhodiumbased hydroformylation catalysts.29,30 Especially interesting in the synthesis of [NBu4][Re2Cl6{m-PhC(O)NH}]?0.5CH2Cl2 1 is the retention of a Re]Re quadruple bond in the product and the ability to study a potential two-metal mechanism for the hydrolysis of nitriles. Dirhenium-(III) and -(II) species have a diverse reactivity due to their high electron density, s2p4d2 and s2p4d2d*2 respectively, which allows them to co-ordinate and activate a variety of unsaturated organic substrates.31–33 For instance, reductive coupling of co-ordinated acetonitrile solvent molecules at the Re2 core leads to the formation of the HN2C2Me2 fragment.33–35 Our research combines the activation of unsaturated organic molecules with the study of the hydrolysis of benzonitrile and demonstrates that bridging benzamidate, m-PhC(O)NH2, results from the hydrolysis of a nitrile co-ordinated to Re2Cl8 22 without oxidation of the transition metal core.The structural and spectroscopic characterization of two diVerent species, a ReIII 2 and a ReIIIReII core, from the dinuclear starting material [NBu4]2[Re2Cl8] are presented. Results and Discussion Synthesis and reaction mechanism As demonstrated by related studies with mononuclear transition metal complexes, the synthesis of amides from the hydrolysis by water 27,36,37 or nucleophilic attack by base 3–6,26 on a nitrile co-ordinated to a single transition metal center is a well established synthetic strategy.However, a bimetallic pathway is proposed as the mechanism for the hydration of nitriles to amides in dipalladium(II) complexes with thiolate-hinged ligands.16,38 Co-ordination of the nitrile to a single palladium center with co-ordination of OH2 on the adjacent metal center leads to a concerted formation of the amide, as shown in Scheme 1. Unlike the mononuclear platinum systems,7,9,11,13–15 the reaction of the dipalladium(II) complex is performed in water–MeCN mixtures and is acid catalyzed.16 Compound 1, [NBu4][Re2Cl6{m-PhC(O)NH}]?0.5CH2Cl2, a ReIII 2 system, is synthesized in the reaction between [NBu4]2-2814 J.Chem. Soc., Dalton Trans., 1998, Pages 2813–2817 [Re2Cl8] and benzonitrile in ethanol solvent mixtures. Without the addition of acid or base the reaction medium is slightly acidic based on the character of ethanol as a weak acid, similar to water.39 In order to determine the oxygen source in our studies of benzonitrile, 1H NMR spectra were obtained for the PhCN and the ethanol solvent used in the reactions.The spectrum of the benzonitrile indicates 3% water is present and, since the reactions were performed under an argon atmosphere, O2 is eliminated as the source of the oxygen in the resultant benzamidate. Therefore, hydrolysis of the nitrile occurs due to the presence of water in the reaction mixture, similar to the water– MeCN solvent mixtures in the dipalladium systems.16 One potential reaction mechanism, analogous to the dipalladium system,16 is the co-ordination of either water or base to an open co-ordination site of the dirhenium core followed by hydrolysis of the nitrile via an intramolecular reaction.However, the intermolecular reaction resulting from direct attack of water on the carbon of the nitrile is also a viable reaction mechanism. Co-ordination of the nitrile in the first step of Scheme 2 results in a more electrophilic C]N bond of the nitrile from the positive charge of the Re31 which facilitates the nucleophilic attack of the OH2 co-ordinated to the adjacent Re31 center, illustrated in Scheme 2(a).Since OH2 is a stronger nucleophile than H2O, the reaction is shown as the attack of the lone pair of electrons of a co-ordinated hydroxide. After the HO]C(Me)N bond formation of the anionic ligand, proton rearrangement occurs to result in 1.Scheme 2(b) illustrates the other possible mechanism, the nucleophilic attack of an H2O at the carbon on the co-ordinated nitrile. In related studies mononuclear rhenium nitrile compounds such as [ReCl3(PPh3)2(MeCN)],40 [NEt4][Re(NO)Br4- (MeCN)] 41 and [ReCl4(MeCN)2] 42 have been synthesized and show some interesting reactivity.43,44 For example, the reaction of [ReCl4(MeCN)2] with primary aromatic amines yields substituted amidines of the general formula [ReCl4{MeC(NH)- NHR}2], as confirmed by IR analysis and the isolation of N-ptolylacetamide formed from the reaction of p-toluidine with [ReCl4(MeCN)2].42 Treatment of [ReCl4(MeCN)2] with ethanol results in the formation of [ReCl4{Me(NH)OEt}2] from the nucleophilic addition of alcohol at the nitrile carbon.42 Note that both of these reactions result in a change in the oxidation state on the central rhenium atom, unlike the reaction between [NBu4]2[Re2Cl8], water and benzonitrile.42 Direct attack of OH2 is unlikely since ethanol is slightly acidic and preliminary studies indicate that the direct attack of OH2 at the nitrile carbon is not probable because the reaction does not occur in a basic medium, instead a brown insoluble powder is formed.In contrast, the reaction proceeds under acidic conditions resulting from the addition of 0.5 mL of concentrated HCl(aq) to the reaction mixture. Compound 1 undergoes a one-electron reduction upon addition of the bidentate phosphine ligand dppm to form 2, [Re2Cl4(m-dppm)2{m-PhC(O)NH}], a ReIIIReII system with the bridging phosphines in a trans geometry. The reduction of a ReIII 2 core by a phosphine ligand is not unusual as demonstrated in the synthesis [Re2Cl4(m-dppm)2] from [Re2Cl6- (PBun 3)2], a two-electron reduction.45,46 Confirmation of structure Since one of the diYculties in the hydrolysis of nitriles is the tendency to form carboxylate species, a methodology to deter- Scheme 1 Proposed mechanism of the formation of co-ordinated benzamide at a dipalladium core Pd Pd O H N C CH3 Pd Pd O H Pd Pd C CH3 N N H C O CH3 CH3CN H2O mine if the dirhenium reaction terminates at the formation of the benzamidate ligand, m-PhC(O)NH2, or reacts further to form the benzoate ligand, m-PhCO2 2, has been established.The structure of compound 1 allowed the nitrogen and oxygen atoms to be distinguished in the benzamidate ligand, disorder in the oxygen and nitrogen atom positions in 2 required fitting the positions as 50% O and 50% N and the nature of the ligand could not be confirmed directly using NMR spectroscopy since the compound is paramagnetic, s2p4d2d*1. Magnetic susceptibility studies performed on 2 confirmed the paramagnetic ReIIIReII system.A meff value of 1.45 mB was obtained after performing a diamagnetic correction for 2 and the value compares well with the spin only meff value of 1.73 mB predicted for a system with one unpaired electron.47 Using the oxidizing agent AgBF4, a diamagnetic ReII 2 derivative of 2 was synthesized in order to perform NMR spectroscopy.Subsequent 31P-{1H} NMR studies confirm that the benzamidate ligand is synthesized and not the benzoate ligand. As expected for a co-ordinated benzamidate ligand, the 31P-{1H} NMR spectrum of the oxidized [Re2Cl4(m-dppm)2- {m-PhC(O)NH}] compound contains a complex pattern indicating inequivalent phosphorus magnetic environments with peaks centered at d 29.43 and 217.69.A singlet in the 31P-{1H} NMR spectrum is predicted if [Re2Cl4(m-dppm)2(m-O2CPh)] had been synthesized. Crystal structures The Re]Re bond length of 2.2209(5) Å for compound 1 (Fig. 1, Table 1), a ReIII 2 system, falls within the range observed for both the starting material [NBu4]2[Re2Cl8] [2.222(2) Å] 49 and other quadruply bonded dirhenium species.50 In contrast, the Re]Re bond length of 2.3129(7) Å for compound 2 (Fig. 2, Scheme 2 Proposed reaction mechanism of the formation of the co-ordinated benzamidate ligand Re Re Cl Cl Cl Cl Cl Cl Cl Cl Re Re Cl Cl N Cl Cl Cl Cl Cl C R 2– – –[N(C4H9)4]Cl + R C N benzonitrile Re Re OH2 Cl N Cl Cl Cl Cl Cl C R – Re Re Cl Cl N Cl Cl Cl Cl Cl C R – Re Re O Cl N Cl Cl Cl Cl Cl C R – Re Re O Cl N Cl Cl Cl Cl Cl C R – Re Re O Cl N– Cl Cl Cl Cl Cl – Re Re O Cl N Cl Cl Cl Cl Cl – C H R H H C R H H2 O H2 O (a) (b) –HCl –HCl ClJ.Chem. Soc., Dalton Trans., 1998, Pages 2813–2817 2815 Table 1) marks one of the longest distances observed for a ReIIIReII species where Re]Re bond distances of 2.20 to 2.30 Å are typical.50 However, two axially co-ordinated chlorides are present and the Re]Re bond lengths in systems with axial halides are usually longer than those in ReIIIReII systems with no axial halides.Weakening of the Re]Re s bond results from the axially co-ordinated ligand, as in the case of the ReIII 2 system [Re2(m-dfm)4(OMe)2] (Hdfm is di-p-tolylformamidine) with a Re]Re bond distance of 2.3047 Å.50,51 The Re]O and Re]N bond distances for compound 1 of 2.009(6) and 2.044(7) Å are only slightly shorter than those observed for the respective Re]O and Re]N bond distances in dirhenium(III) carboxylato {2.025(4) Å for [Re2Cl2(m-O2- CMe)4]} and amidinato complexes (2.055(9) Å for [Re2Cl2- {(MeN)2CPh}4]?CCl4) with axial chlorides.52,53 The Re]ONX bond distance of 2.078 Å (average) for 2 is similar to the 2.07 Å (average) Re]N bond distance for the amidinato complex without axial halides [Re2Cl4{(PhN)2CMe}2] but a direct comparison of this ‘average’ Re]ONX bond distance to those of other systems holds significant error.53 The C]O and C]N bond distances for unco-ordinated neutral benzamide are 1.24 and 1.31 Å, respectively.54 The co- Fig. 1 An ORTEP48 drawing of the crystal structure of [NBu4]- [Re2Cl6(PhCONH)]?0.5CH2Cl2 with thermal ellipsoids drawn at 50% probability. Hydrogen atoms were omitted for clarity Table 1 Selected bond lengths (Å) and angles (8) for compounds 1 and 2 when the numbering scheme diVers from 1 to 2 superscripts were attached to denote the respective compounds Re(1)]Re(2) Re(1)]N(1)1 and Re(1)]ONX(2)2 Re(2)]O(1)1 and Re(1)]ONX(1)2 Re(1)]Cl(1) Re(1)]Cl(2) Re(1)]Cl(3) Re(2)]Cl(4)1 and Re(2)]Cl(3)2 Re(2)]Cl(5)1 and Re(2)]Cl(4)2 Re(2)]Cl(6) N(1)]C(1)1 and ONX(1)]C(51)2 O(1)]C(1)1 and ONX(2)]C(51)2 Re(1)]Re(2)]O(1)1 and Re(1)]Re(2)] ONX(1)2 Re(2)]Re(1)]N(1)1 and Re(2)]Re(1)] ONX(2)2 Re(2)]Re(1)]Cl(1) Re(2)]Re(1)]Cl(2) Re(2)]Re(1)]Cl(3) Re(1)]Re(2)]Cl(4)1 and Re(1)]Re(2)] Cl(3)2 Re(1)]Re(2)]Cl(5)1 and Re(1)]Re(2)] Cl(4)2 Re(1)]Re(2)]Cl(6) 1 2.2209(5) 2.044(7) 2.009(6) 2.334(2) 2.375(2) 2.332(2) 2.322(3) 2.310(3) 2.306(3) 1.315(11) 1.302(11) 92.7(2) 88.1(2) 105.37(7) 100.07(6) 101.39(6) 103.86(8) 106.17(7) 106.61(7) 2 2.3129(7) 2.06(2) 2.096(14) 2.618(3) 2.375(3) 2.589(3) 2.389(3) 1.28(2) 1.34(2) 90.1(5) 88.5(5) 162.40(8) 104.41(9) 167.95(8) 99.89(8) ordinated benzamidate ligand in compound 1 has bond distances of 1.302(11) and 1.315(11) Å for C]O and C]N, while 2 has a distance of 1.31 Å (average) for the disordered ONX]C positions, distances that are within the range expected for the anionic ligand.In both compounds the benzamidate ligand retains an sp2 hybridization so little deviation in the N]C]O bond angle from 1208 is predicted. As expected, 1 has an N]C]O bond angle of 118.3(8)8 and 2 an ONX]C]ONX angle of 118.2(12)8.Conclusion This study marks the first use of a multiply bonded dirhenium species to synthesize co-ordinated benzamidate ligands from benzonitrile. It is currently being expanded to include other substituted nitriles to determine if these species will undergo an analogous selective hydrolysis process. Experiments are being carried out fully to characterize the diamagnetic product of the oxidation of [Re2Cl4(m-dppm)2{m-PhC(O)NH}], including elucidation of the coupling constants observed in the 31P-{1H} NMR spectra.Experimental Starting materials Reagent ethanol was dried over magnesium turnings, methylene chloride over P2O5, and hexanes over potassium/ sodium–benzophenone. All solvents were freshly distilled under an atmosphere of argon prior to use. The starting material, [NBu4]2[Re2Cl8], was synthesized as previously reported.55 Bis- (diphenylphosphino)methane (dppm) from Aldrich Chemical Co. was evacuated overnight under dynamic vacuum to remove any residual oxygen or moisture.Benzonitrile in a Sure-Seal bottle from Aldrich Chemical Co. was used without further purification. All reagent transfers were performed using standard Schlenk, vacuum line and dry-box techniques under an inert atmosphere of argon. Syntheses [NBu4][Re2Cl6{Ï-PhC(O)NH}]?0.5CH2Cl2 1. The complex [NBu4]2[Re2Cl8] (0.25 g, 2.19 mmol) was suspended in ethanol (30 cm3) and PhCN (1 cm3, 9.77 mmol) added. The solution was stirred for 3 d at room temperature and a deep blue-green solution formed.All solvent was removed under dynamic vacuum overnight to yield a blue product (0.168 g, 85%) (Found: C, Fig. 2 An ORTEP drawing of the crystal structure of [Re2Cl4- (m-dppm)2(PhCONH)]. Details as in Fig. 12816 J. Chem. Soc., Dalton Trans., 1998, Pages 2813–2817 28.70; H, 4.40; N, 2.91. Calc. for C23.50H42Cl7N2ORe2: C, 28.54; H, 4.28; N, 2.83%). lmax/nm (CH2Cl2): 614. n& max/cm21 2955, 2930 and 2873 (NH and CH) and 1530, 1469 and 1379 (CO and CN).dH(400 MHz; solvent CDCl3; standard SiMe4) 1.05 (t, J 6.6), 1.54 (s, broad), 1.77 (s, broad), 3.31 (s, broad), 7.46 (d, J 7.3), 7.54 (t, J 7), 7.94 (d, J 7.3 Hz) and 11.34 (s). [Re2Cl4(Ï-dppm)2{Ï-PhC(O)NH}] 2. Compound 1 (0.216 g, 2.19 mmol) dissolved in ethanol (30 cm3) was transferred via cannula to a Schlenk flask containing dppm (0.17 g, 4.38 mmol) and heated at reflux for 45 min. The brown product precipitated and was washed three times with 30 cm3 aliquots of ethanol and dried under dynamic vacuum (0.174 g, 59%).Compound 2 can also be synthesized from 1 prepared in situ by the addition of dppm (Found: C, 48.55; H, 3.99; N, 1.18. Calc. for C57H49- Cl4NOP4Re2: C, 48.82; H, 3.53; N, 1.00%). lmax/nm (CH2Cl2): 482 (sh), 722, 1066. n& max/cm21 2949, 2926 and 2872 (NH and CH) and 1466 and 1378 (CO and CN). Instrumentation The UV/VIS spectra were recorded on a Hewlett-Packard model 8453 diode array spectrophotometer from 400 to 1100 nm, 31P-{1H} (162 MHz) and 1H NMR spectra (400 MHz) of compounds 1 and 2 in CH2Cl2 on a General Electric Omega spectrometer with a 10 mm variable temperature broad band probe referenced to H3PO4 (d 0.00) and SiMe4 (d 0.00), respectively.The 1H NMR spectra (300 MHz) of neat benzonitrile and ethanol were recorded on a General Electric QE NMR spectrometer. Magnetic susceptibility measurements were performed on a Johnson Matthey magnetic susceptibility balance and diamagnetic corrections were applied.47 Infrared measurements were recorded on a MIDAC FT-IR spectrometer as liquid samples in CsI cells using CH2Cl2 as the solvent.Elemental analyses were performed on compound 1 by Exeter Analytical, Inc. and on 2 by Oneida Research Services. X-Ray crystallography Suitable crystals for X-ray crystallography were grown by layering CH2Cl2 solutions of the respective compounds with hexanes. Crystals of 1 and 2 were mounted on glass fibers with grease. A summary of data collection parameters is provided in Table 2.Data were collected using a Siemens SMART CCD (charge coupled device) based diVractometer equipped with a LT-2 low-temperature apparatus operating at 213 K for compound 1 and 173 K for 2. Omega scans of 0.38 per frame for 30 s were used, such that a hemisphere was collected. A total of 1271 frames were collected with a final resolution of 0.85 Å. The first 50 frames were recollected at the end of data collection to monitor for decay.Cell parameters were retrieved using SMART56 software and refined using SAINT 57 on all observed reflections. Data reduction was performed using the SAINT software, which corrects for Lorentz polarization and decay. Absorption corrections were applied using SADABS58 supplied by George Sheldrick. The structures were solved by the direct method using the SHELXS 9059 program and refined by least squares on F2 using SHELXL 97,60 incorporated in SHELXTLPC V 5.03.61 Neither of the crystals showed decomposition during data collection.Compound 1 crystallized in the triclinic crystal system. The space group P1� was assumed and confirmed by the successful solution refinement of the structure. All non-hydrogen atoms were refined anisotropically. Hydrogen atoms were found by Fourier-diVerence methods and refined isotropically. The structure of compound 2 was solved in the monoclinic crystal system. The space group P21/n was assumed and con- firmed by the successful solution and refinement of the structure.Although the majority of the structure was readily apparent, the co-ordinated O and N atoms of the benzamidate ligand could not be distinguished. The two positions, ONX1 and ONX2, were each refined as 50% oxygen and 50% nitrogen. The geometrically constrained hydrogen atoms were placed in calculated positions with isotropic vibrational factors equal to 120% of the atom to which they were bonded.Refinement of nonhydrogen atoms was carried out with anisotropic thermal parameters except for C20, a carbon on one of the phenyl rings of the dppm ligand. CCDC reference number 186/1053. Acknowledgements T. E. C. and J. L. E. would like to acknowledge the Donors of The Petroleum Research Fund, administered by the American Chemical Society, for the partial support of this research and National Science Foundation (NSF) Grant no. CHE 9630192 and the NSF EPSCoR program Grant no.EHR 9108767. The NMR measurements were made at the Mississippi Magnetic Research Facility at Mississippi State University, supported by the NSF Grant no. CHE-9214521 and Mississippi State University. J. L. E. would like to acknowledge Drs. T. Ren and R. Hicks for helpful discussions. References 1 R. O. C. Norman, Principles of Organic Synthesis, 2nd edn., Chapman and Hall, London, 1978. 2 J. McMurry, Organic Chemistry, Wadsworth, Inc., Belmont, CA, 1988, vol. 2. 3 K. B. Nolan and R. W. Hay, J. Chem. Soc., Dalton Trans., 1974, 914. 4 D. Pinnell, G. B. Wright and R. B. Jordan, J. Am. Chem. Soc., 1972, 94, 6104. 5 C. R. Clark and R. W. Hay, J. Chem. Soc., Dalton Trans., 1974, 2148. 6 N. E. Dixon, D. P. Fairlie, W. G. Jackson and A. M. Sargeson, Inorg. Chem., 1983, 22, 4038. 7 A. W. Parkins, Platinum Met. Rev., 1996, 40, 169. 8 F. D. Rochon, P. C. Kong and R. Melanson, Inorg. Chem., 1990, 29, 1352. 9 A. Erxleben, I. Mutikainen and B. Lippert, J.Chem. Soc., Dalton Trans., 1994, 3667. Table 2 Crystallographic and experimental details of the X-ray studies for compounds 1 and 2 Formula M l/Å Crystal system Space group a/Å b/Å c/Å a/8 b/8 g/8 U/Å3 Z Dc/g cm23 m/mm21 Crystal size/mm q Range for data collection/8 Index ranges Reflections collected Independent reflections (Rint) Data/restraints/ parameters R1 [I > 2s(I)] wR2 [I > 2s(I)] 1 C23.50H42Cl7N2ORe2 989.14 0.710 73 Triclinic P1� 11.9970(7) 12.2518(7) 12.9875(7) 79.379(1) 73.206(1) 79.478(1) 1779.2(2) 2 1.846 7.342 0.05 × 0.15 × 0.15 1.65 to 24.72 214 < h < 14, 211 < k < 14, 212 < l < 15 9289 5980 (0.0372) 5980/0/334 0.0419 0.0887 2 C57H49Cl4NOP4Re2 1402.05 0.710 73 Monoclinic P21/n 17.9845(2) 14.4872(1) 21.2250(2) 105.937(1) 5317.52(9) 4 1.751 4.911 0.05 × 0.10 × 0.10 1.32 to 23.20 219 < h < 19, 215 < k < 15, 223 < l < 13 19 872 7438 (0.1509) 7436/0/618 0.0656 0.1606J. Chem.Soc., Dalton Trans., 1998, Pages 2813–2817 2817 10 A.Erxleben and B. Lippert, J. Chem. Soc., Dalton Trans., 1996, 2329. 11 F. P. Fanizzi, F. P. Intini and G. Natile, J. Chem. Soc., Dalton Trans., 1989, 947. 12 R. Cini, F. P. Fanizzi, F. P. Intini, L. Maresca and G. Natile, J. Am. Chem. Soc., 1993, 115, 5123. 13 R. Cini, F. P. Fanizzi, F. P. Intini, G. Natile and C. Pacifico, Inorg. Chim. Acta, 1996, 251, 111. 14 R. Cini, F. P. Fanizzi, F. P. Intini, C. Pacifico and G. Natile, Inorg.Chim. Acta, 1997, 264, 279. 15 C. M. Jensen and W. C. Trogler, J. Am. Chem. Soc., 1986, 108, 723. 16 C. J. McKenzie and R. Robson, J. Chem. Soc., Dalton Trans., 1988, 112. 17 N. J. Curtis and A. M. Sargeson, J. Am. Chem. Soc., 1984, 106, 625. 18 R. Breslow, R. Fairweather and J. Keana, J. Am. Chem. Soc., 1967, 89, 2135. 19 M. P. Suh, K. Y. Oh, J. W. Lee and Y. Y. Bae, J. Am. Chem. Soc., 1996, 118, 777. 20 S. Thomas, P. J. Lim, R. W. Gable and C. G. Young, Inorg. Chem., 1998, 37, 590. 21 J. Chin, Acc. Chem. Res., 1991, 24, 145. 22 S. J. Lippard and J. M. Berg, Principles of Bioinorganic Chemistry, University Science Books, Mill Valley, CA, 1994. 23 N. Stra W. N. Lipscomb, T. Klabunde and B. Krebs, Angew. Chem., Int. Ed. Engl., 1996, 35, 2024. 24 O. Meth-Cohn and M.-Z. Wang, J. Chem. Soc., Perkin Trans. 1, 1997, 3197. 25 A. L. Nivorozhkin, A. I. Uraev, G. I. Bondarenko, A. S. Antsyshkina, V. P. Kurbatov, A. D. Garnovskii, C. I. Turta and N. D. Brashoveanu, Chem.Commun., 1997, 1711. 26 D. A. Buckingham, F. R. Keene and A. M. Sargeson, J. Am. Chem. Soc., 1973, 95, 5649. 27 J. H. Kim, J. Britten and J. Chin, J. Am. Chem. Soc., 1993, 115, 3618. 28 T. A. Steitz and J. A. Steitz, Proc. Natl. Acad. Sci. USA, 1993, 90, 6498. 29 W. Jones, J. Huggins and R. Bergman, J. Am. Chem. Soc., 1981, 103, 4415. 30 M. E. Broussard, B. Juma, S. G. Train, W. J. Peng, S. A. Laneman and G. G. Stanley, Science, 1993, 260, 1784. 31 T. E. Concolino and J. L.Eglin, J. Cluster Sci., 1997, 8, 461. 32 R. A. Walton, Polyhedron, 1989, 8, 1689. 33 K.-Y. Shih, P. E. Fanwick and R. A. Walton, J. Am. Chem. Soc., 1993, 115, 9319. 34 D. Esjornson, D. R. Derringer, P. E. Fanwick and R. A. Walton, Inorg. Chem., 1989, 28, 2821. 35 D. Esjornson, P. E. Fanwick and R. A. Walton, Inorg. Chem., 1988, 27, 3066. 36 K. Watanabe, S. Komiya and S. Suzuki, Bull. Chem. Soc. Jpn., 1973, 46, 2792. 37 D. P. Fairlie, W. G. Jackson and G. M. McLaughlin, Inorg.Chem., 1989, 28, 1983. 38 B. F. Hoskins, C. J. McKenzie, I. A. S. MacDonald and R. Robson, J. Chem. Soc., Dalton Trans., 1996, 2227. 39 A. Streitwieser and C. H. Heathcock, Introduction to Organic Chemistry, 2nd edn., MacMillan, New York, 1981. 40 M. G. B. Drew, D. G. Tisley and R. A. Walton, Chem. Commun., 1970, 600. 41 G. Ciani, D. Giusto, M. Manassero and M. Sansoni, J. Chem. Soc., Dalton Trans., 1975, 2156. 42 G. Rouschias and G. Wilkinson, J. Chem. Soc. A, 1968, 489. 43 J. J. R. F. d. Silva, M. F. C. G. d. Silva, R. A. Henderson, A. J. L. Pombeiro and R. L. Richards, J. Organomet. Chem., 1993, 461, 141. 44 B. StorhoV and H. Lewis, Coord. Chem. Rev., 1977, 23, 1. 45 T. J. Barder, F. A. Cotton, D. Lewis, W. Schwotzer, S. M. Tetrick and R. A. Walton, J. Am. Chem. Soc., 1984, 106, 2882. 46 T. J. Barder, F. A. Cotton, K. R. Dunbar, G. L. Powell, W. Schwotzer and R. A. Walton, Inorg. Chem., 1985, 24, 2550. 47 R. S. Drago, Physical Methods for Chemists, 2nd edn., Saunders College Publishing, Ft.Worth, 1992. 48 C. K. Johnson, ORTEP, Report ORNL-5138, Oak Ridge National Laboratory, Oak Ridge, TN, 1976. 49 F. A. Cotton, B. A. Frenz, B. R. Stults and T. R. Webb, J. Am. Chem. Soc., 1976, 98, 2768. 50 F. A. Cotton and R. A. Walton, Multiple Bonds between Metal Atoms, 2nd edn., University Press, Oxford, 1993. 51 F. A. Cotton and T. Ren, J. Am. Chem. Soc., 1992, 114, 2495. 52 D. M. Collins, F. A. Cotton and L. D. Gage, Inorg.Chem., 1979, 18, 1712. 53 F. A. Cotton, W. H. Ilsley and W. Kaim, Inorg. Chem., 1980, 19, 2360. 54 B. R. Penfold and J. C. B. White, Acta Crystallogr., Sect. C, 1959, 12, 130. 55 T. J. Barder and R. A. Walton, Inorg. Chem., 1982, 21, 2510. 56 SMART V 4.043, Software for the CCD Detector System, Siemens Analytical Instruments Division, Madison, WI, 1996. 57 SAINT V 4.035, Software for the CCD Detector System, Siemens Analytical Instruments Division, Madison, WI, 1995. 58 SADABS, Program for Absorption corrections using Siemens CCD based on the method of Bob Blessing, Acta Crystallogr., Sect.A, 1995, 51, 33. 59 G. M. Sheldrick, SHELXS 90, Program for the Solution of Crystal Structures, University of Göttingen, 1986. 60 G. M. Sheldrick, SHELXL 97, Program for the Refinement of Crystal Structure, University of Göttingen, 1997. 61 SHELXTL 5.03 [PC-Version], Program Library for Structure Solution and Molecular Graphics, Siemens Analytical Instruments Division, Madison, WI, 1995.Received 27th February 1998; Paper 8/01746HJ. Chem. Soc., Dalton Trans., 1998, Pages 2813–2817 2817 10 A. Erxleben and B. Lippert, J. Chem. Soc., Dalton Trans., 1996, 2329. 11 F. P. Fanizzi, F. P. Intini and G. Natile, J. Chem. Soc., Dalton Trans., 1989, 947. 12 R. Cini, F. P. Fanizzi, F. P. Intini, L. Maresca and G. Natile, J. Am. Chem. Soc., 1993, 115, 5123. 13 R. Cini, F. P. Fanizzi, F. P. Intini, G. Natile and C. Pacifico, Inorg. Chim.Acta, 1996, 251, 111. 14 R. Cini, F. P. Fanizzi, F. P. Intini, C. Pacifico and G. Natile, Inorg. Chim. Acta, 1997, 264, 279. 15 C. M. Jensen and W. C. Trogler, J. Am. Chem. Soc., 1986, 108, 723. 16 C. J. McKenzie and R. Robson, J. Chem. Soc., Dalton Trans., 1988, 112. 17 N. J. Curtis and A. M. Sargeson, J. Am. Chem. Soc., 1984, 106, 625. 18 R. Breslow, R. Fairweather and J. Keana, J. Am. Chem. Soc., 1967, 89, 2135. 19 M. P. Suh, K. Y. Oh, J. W. Lee and Y. Y. Bae, J. Am. Chem.Soc., 1996, 118, 777. 20 S. Thomas, P. J. Lim, R. W. Gable and C. G. Young, Inorg. Chem., 1998, 37, 590. 21 J. Chin, Acc. Chem. Res., 1991, 24, 145. 22 S. J. Lippard and J. M. Berg, Principles of Bioinorganic Chemistry, University Science Books, Mill Valley, CA, 1994. 23 N. Strater, W. N. Lipscomb, T. Klabunde and B. Krebs, Angew. Chem., Int. Ed. Engl., 1996, 35, 2024. 24 O. Meth-Cohn and M.-Z. Wang, J. Chem. Soc., Perkin Trans. 1, 1997, 3197. 25 A. L. Nivorozhkin, A. I. Uraev, G.I. Bondarenko, A. S. Antsyshkina, V. P. Kurbatov, A. D. Garnovskii, C. I. Turta and N. D. Brashoveanu, Chem. Commun., 1997, 1711. 26 D. A. Buckingham, F. R. Keene and A. M. Sargeson, J. Am. Chem. Soc., 1973, 95, 5649. 27 J. H. Kim, J. Britten and J. Chin, J. Am. Chem. Soc., 1993, 115, 3618. 28 T. A. Steitz and J. A. Steitz, Proc. Natl. Acad. Sci. USA, 1993, 90, 6498. 29 W. Jones, J. Huggins and R. Bergman, J. Am. Chem. Soc., 1981, 103, 4415. 30 M. E. Broussard, B. Juma, S.G. Train, W. J. Peng, S. A. Laneman and G. G. Stanley, Science, 1993, 260, 1784. 31 T. E. Concolino and J. L. Eglin, J. Cluster Sci., 1997, 8, 461. 32 R. A. Walton, Polyhedron, 1989, 8, 1689. 33 K.-Y. Shih, P. E. Fanwick and R. A. Walton, J. Am. Chem. Soc., 1993, 115, 9319. 34 D. Esjornson, D. R. Derringer, P. E. Fanwick and R. A. Walton, Inorg. Chem., 1989, 28, 2821. 35 D. Esjornson, P. E. Fanwick and R. A. Walton, Inorg. Chem., 1988, 27, 3066. 36 K. Watanabe, S. Komiya and S.Suzuki, Bull. Chem. Soc. Jpn., 1973, 46, 2792. 37 D. P. Fairlie, W. G. Jackson and G. M. McLaughlin, Inorg. Chem., 1989, 28, 1983. 38 B. F. Hoskins, C. J. McKenzie, I. A. S. MacDonald and R. Robson, J. Chem. Soc., Dalton Trans., 1996, 2227. 39 A. Streitwieser and C. H. Heathcock, Introduction to Organic Chemistry, 2nd edn., MacMillan, New York, 1981. 40 M. G. B. Drew, D. G. Tisley and R. A. Walton, Chem. Commun., 1970, 600. 41 G. Ciani, D. Giusto, M. Manassero and M. Sansoni, J.Chem. Soc., Dalton Trans., 1975, 2156. 42 G. Rouschias and G. Wilkinson, J. Chem. Soc. A, 1968, 489. 43 J. J. R. F. d. Silva, M. F. C. G. d. Silva, R. A. Henderson, A. J. L. Pombeiro and R. L. Richards, J. Organomet. Chem., 1993, 461, 141. 44 B. StorhoV and H. Lewis, Coord. Chem. Rev., 1977, 23, 1. 45 T. J. Barder, F. A. Cotton, D. Lewis, W. Schwotzer, S. M. Tetrick and R. A. Walton, J. Am. Chem. Soc., 1984, 106, 2882. 46 T. J. Barder, F. A. Cotton, K. R. Dunbar, G. L. Powell, W. Schwotzer and R. A. Walton, Inorg. Chem., 1985, 24, 2550. 47 R. S. Drago, Physical Methods for Chemists, 2nd edn., Saunders College Publishing, Ft. Worth, 1992. 48 C. K. Johnson, ORTEP, Report ORNL-5138, Oak Ridge National Laboratory, Oak Ridge, TN, 1976. 49 F. A. Cotton, B. A. Frenz, B. R. Stults and T. R. Webb, J. Am. Chem. Soc., 1976, 98, 2768. 50 F. A. Cotton and R. A. Walton, Multiple Bonds between Metal Atoms, 2nd edn., University Press, Oxford, 1993. 51 F. A. Cotton and T. Ren, J. Am. Chem. Soc., 1992, 114, 2495. 52 D. M. Collins, F. A. Cotton and L. D. Gage, Inorg. Chem., 1979, 18, 1712. 53 F. A. Cotton, W. H. Ilsley and W. Kaim, Inorg. Chem., 1980, 19, 2360. 54 B. R. Penfold and J. C. B. White, Acta Crystallogr., Sect. C, 1959, 12, 130. 55 T. J. Barder and R. A. Walton, Inorg. Chem., 1982, 21, 2510. 56 SMART V 4.043, Software for the CCD Detector System, Siemens Analytical Instruments Division, Madison, WI, 1996. 57 SAINT V 4.035, Software for the CCD Detector System, Siemens Analytical Instruments Division, Madison, WI, 1995. 58 SADABS, Program for Absorption corrections using Siemens CCD based on the method of Bob Blessing, Acta Crystallogr., Sect. A, 1995, 51, 33. 59 G. M. Sheldrick, SHELXS 90, Program for the Solution of Crystal Structures, University of Göttingen, 1986. 60 G. M. Sheldrick, SHELXL 97, Program for the Refinement of Crystal Structure, University of Göttingen, 1997. 61 SHELXTL 5.03 [PC-Version], Program Library for Structure Solution and Molecular Graphics, Siemens Analytical Instruments Division, Madison, WI, 1995. Received 27th February 1998; Paper 8/01746H

ISSN:1477-9226    

DOI:10.1039/a801746h

出版商:RSC    

年代:1998

数据来源: RSC

摘要:

DALTON FULL PAPER J. Chem. Soc., Dalton Trans., 1998, Pages 2819–2825 2819 Ruthenate(VI)-catalysed dehydrogenation of primary amines to nitriles, and crystal structures of cis-[Ru(bipy)2(NH2CH2Ph)2]- [PF6]2?0.5MeOH and cis-[Ru(bipy)2(NCPh)2][PF6]2?CH2Cl2† William P. GriYth,* Bharti Reddy, Abdel G. F. Shoair, Maria Suriaatmaja, Andrew J. P. White and David J. Williams * Inorganic Chemistry and Chemical Crystallographic Laboratories, Department of Chemistry, Imperial College of Science, Technology and Medicine, London, UK SW7 2AY Catalytic dehydrogenation of benzylic and other primary amines RCH2NH2 to the corresponding nitriles RCN by the system trans-[Ru(OH)2O3]22/S2O8 22 has been investigated. The complex cis-[Ru(bipy)2(NH2CH2Ph)2]21 and the new cis-[Ru(bipy)2(NH2CH2R)2]21 (R = o-, m- or p-ClC6H4, o-MeC6H4, o- or p-MeOC6H4); cis- [Ru(phen)2(NH2CH2R)2]21 (R = Ph or p-MeOC6H4) and cis-[Os(bipy)2(NH2CH2Ph)2]21 have been made, and dehydrogenation of the co-ordinated amine in the ruthenium complexes to the corresponding nitriles in cis-[Ru(L]L)2(NCR)2]21 (L]L = bipy or phen) by peroxodisulfate demonstrated.The crystal structures of cis-[Ru(bipy)2(NH2CH2Ph)2][PF6]?0.5MeOH and cis-[Ru(bipy)2(NCPh)2][PF6]2?CH2Cl2, the latter a product of co-ordinated amine dehydrogenation by peroxodisulfate to give cis-[Ru(bipy)2(NH2CH2Ph)2]21, were determined. Raman, infrared and 1H NMR data for the complexes have been measured; the latter suggest that the cis configurations of the amine complexes are retained in solution.As part of our continuing studies on the application of oxoruthenates and other ruthenium complexes as catalysts for the oxidation of alcohols,2–4 alkyl halides and nitro compounds, 3 sulfides,5 alkenes and alkanes 6 we have extended and developed our earlier brief observations 7 that benzylamine is dehydrogenated to benzonitrile by the catalytic trans-[Ru(OH)2- O3]22/S2O8 22 reagent. We find that the latter is eVective for the catalytic dehydrogenation (or oxidative dehydrogenation) of primary aromatic and some primary aliphatic amines RCH2NH2 to nitriles RCN, and in some cases it will further catalyse the hydration of nitriles to amides RCONH2.The use of other oxoruthenate systems to eVect these conversions has been studied, and a study made of the stoichiometric dehydrogenation of amine complexes cis-[Ru(L]L)2(NH2- CH2R)2]21 (L]L = bipy or phen) to [Ru(L]L)2(NCR)2]21 by peroxodisulfate. A number of new amine and nitrile complexes have been isolated. Relatively few reagents are known for the catalytic conversion of amines into nitriles; there are a number of stoichiometric procedures, e.g.the use of hypochlorite 8 or silver(II) compounds.9 The best catalytic system hitherto reported is a nickel(II) sulfate–peroxodisulfate reagent in 0.4 M aqueous base; this gives good yields and selectivities but is slow, taking a day for most reactions.10 The copper(I) chloride–pyridine–dioxygen system is also slow and requires higher temperatures (60 8C).11 The complex trans-[RuVIO2- (tmp)] (tmp = dianion of 5,10,15,20-tetramesitylporphyrin) in benzene aerobically catalyses dehydrogenation of benzylamine and of n-butylamine quantitatively to the corresponding nitriles at 50 8C over a 24 h period;12 RuCl3?nH2O in toluene dehydrogenates these substrates to a mixture of the nitriles and amides at 100 8C under 2 atm (atm = 101 325 Pa) of dioxygen,13 and [Ru(PPh3)2(NH2CH2Ph)2Cl2] has been shown to catalyse the aerobic conversion of benzylamine into benzonitrile at 80 8C.14 Apart then from the slow nickel system, none of these catalytic reagents operates eYciently at room temperatures.† Studies on transition-metal nitrido and oxo complexes. Part 18.1 Results and discussion (a) Dehydrogenation of amines to nitriles by trans- [Ru(OH)2O3]22/S2O8 22 This reagent comprised of 1 × 1024 M Ru (initially as RuCl3? nH2O or RuO2?nH2O), 0.1 M sodium peroxodisulfate and molar aqueous potassium hydroxide, is an eYcient catalyst for the conversion of a wide range of primary benzylic amines into the corresponding nitriles at room temperatures, and we have optimised the reaction conditions giving good yields and selectivities over relatively short periods of time (1–2 h; Table 1).With primary aliphatic amines however the system is more capricious: n-hexylamine and n-octylamine gave reasonable yields of the nitriles but took much longer (24 h) to react than did the aromatic amines; n-butylamine gave butyric acid, perhaps due to hydrolysis at the high pH of the reagent.None of these reactions occurs to any appreciable extent in the absence of ruthenium. As with most other oxidations involving ruthenate as a catalyst 2,7 these are self-indicating: the orange trans- [Ru(OH)2O3]22 turns dark green on addition of the amine, the orange ruthenate colour returning when the reactions are complete.The GC-MS studies show that in most cases only the nitriles are present after the reaction with traces of aldehyde and amide side-products; only in the case of benzylamine were significant quantities of the imine PhCH2N]] CHPh and benzaldehyde also formed (ca. 28 and 5% respectively as determined by GC-MS). The purity of all the nitrile products was checked by their melting or boiling points as appropriate, and their 1H NMR and GC-MS spectra measured.Two large-scale oxidations were carried out: thus 6.8 g (0.05 mol) of p-methoxybenzylamine gave 3.2 g (0.03 mol) of p-methoxybenzonitrile when treated with 0.1 g of RuCl3?3H2O (3.8 × 1024 mol) in 500 cm3 of 1 M aqueous KOH containing 10.4 g (0.04 mol) of K2S2O8 for 24 h. Under the same conditions 5.4 g (0.05 mol) of benzylamine gave 2.6 g (0.025 mol) of benzonitrile. When the reaction is conducted stoichiometrically the ruthenium-containing product is RuO2, so that 2 mol of ruthenate should dehydrogenate 1 mol of amine (i.e.an overall four-electron reaction). Stoichiometrically, solid barium ruthenate (0.34 g, 1.0 mmol) dehydrogenated 0.07 g (0.582820 J. Chem. Soc., Dalton Trans., 1998, Pages 2819–2825 Table 1 Catalytic oxidation * of amines to nitriles by trans-[Ru(OH)2O3]22/S2O8 22 Substrate Benzylamine o-Chlorobenzylamine m-Chlorobenzylamine p-Chlorobenzylamine o-Methylbenzylamine m-Methylbenzylamine p-Methylbenzylamine o-Methoxybenzylamine p-Methoxybenzylamine o-Bromobenzylamine hydrochloride m-Bromobenzylamine hydrochloride n-Butylamine n-Hexylamine n-Octylamine Product Benzonitrile o-Chlorobenzonitrile m-Chlorobenzonitrile p-Chlorobenzonitrile o-Methylbenzonitrile m-Methylbenzonitrile p-Methylbenzonitrile o-Methoxybenzonitrile p-Methoxybenzonitrile o-Bromobenzonitrile m-Bromobenzonitrile Butyric acid Hexanenitrile Octanenitrile Stirring time/h 1.5 1.0 1.0 1.0 1.0 1.0 1.5 0.5 0.75 1.5 1.5 24 24 24 Yield (%) 61 70 60 60 81 98 70 68 90 83 75 9 15 50 * Oxidations were carried out with 2 mmol substrate, 0.1 mmol RuCl3?3H2O and an excess of K2S2O8 (2.8 g) in 1 M KOH solution (25 cm3).mmol) of p-methoxybenzylamine to give 0.065 g (0.049 mmol) of p-methoxybenzonitrile, corresponding to a 3.8 electron change, in reasonable agreement with the expected four electron change. The use of a phase-transfer catalyst, (NBun 4)OH (following the eVective use of (NBun 4)HSO4 for the stoichiometric oxidations of amines to nitriles by hypochlorite 8), was attempted but did not improve yields or turnovers; and likewise sonication or heating the solution to 50 8C gave little improvement. Other ruthenium-containing systems were also used but were inferior to the trans-[Ru(OH)2O3]22/S2O8 22 reagent.Surprisingly, the use of perruthenate (in the catalytic [RuO4]2/BrO3 2 system), known to be eVective for oxidation of alcohols, halides and nitro compounds,2 was completely ineVectual, as was [RuO4]2/S2O8 22; however, this latter system is known not to function as a catalyst for alcohol oxidations.2 No amine dehydrogenation was observed with the [RuO4]2/BrO3 2 system, suggesting that peroxodisulfate is necessary for the reaction. However, [NPrn 4][RuO4], with N-methylmorpholine N-oxide as cooxidant,4 which has been shown recently to be an eVective and clean reagent for conversion of secondary amines R1CH2- NHR2 into the corresponding imine R1CH]] NR2,15 does convert benzylic amines into nitriles over 3 h periods at room temperatures, but substantial quantities of aldehyde and other side-products are also formed. Over a 24 h period, the amines are dehydrogenated and then hydrated in relatively small yields to the corresponding amides RCONH2 (ca. 10% for benzylamine and p-methoxybenzylamine) as we noted earlier;16 this does not occur in the absence of peroxodisulfate. The stoichiometric hydration of amines to nitriles in the presence of [RuII(NH3)5(H2O)]21 has been noted.17 It is clear that the platinum(II) phosphinito complexes recently reported are much better nitrile hydration catalysts.18 (b) Oxidations of co-ordinated primary amines to co-ordinated nitriles by peroxodisulfate We find that when aromatic amines are added to the trans-[Ru- (OH)2O3]22/S2O8 22 reagent or to a pure solution of trans-[Ru- (OH)2O3]22 in aqueous base a green species is formed, probably an amine complex such as [Ru(OH)2O3(NH2CH2R)]22.No such colour is formed with the corresponding nitrile. Attempts to isolate this green species have failed, and no 1H NMR spectrum could be measured owing to the paramagnetism of trans- [Ru(OH)2O3]22 and/or the complex. Direct reaction of amines with solid trans-Ba[Ru(OH)2O3] or solid trans-K2[Ru(OH)2O3] or of trans-[Ru(OH)2O3]22 in solution in the absence or presence of stabilising coligands such as pyridine or 2,29-bipyridyl did not give identifiable products.The intermediacy of such an amine complex seems likely, however: Bailey and James12 found, during their work on trans-[RuVIO2(tmp)] with benzylamine, that trans-[RuII(tmp)(NH2CH2Ph)2] is formed when the amine is present in excess. As a model for our postulated amine complex we treated cis-[Ru(bipy)2(NH2CH2Ph)2]21 with an excess of aqueous peroxodisulfate to establish whether it was dehydrogenated to a co-ordinated benzonitrile complex. No such conversions with peroxodisulfate have been reported, though Meyer and coworkers 19 have shown that cis-[Ru(bipy)2(NH2CH2Ph)2]21 is converted electrochemically into cis-[Ru(bipy)2(NH2CH2Ph)- (NCPh)]21 and cis-[Ru(bipy)2(NCPh)2]21, and Taube and co-workers 20 aerobically dehydrogenated [Ru(NH3)5(NH2- CH2Ph)]21 to [Ru(NH3)5(NCPh)]21.We find that cis-[Ru(bipy)2- (NH2CH2Ph)2]21 is indeed converted into cis-[Ru(bipy)2- (NCPh)2]21 by an excess of aqueous peroxodisulfate; i.e. coordinated benzylamine is oxidised to co-ordinated benzonitrile.The constitutions of these two complexes as their hexafluorophosphate salts 1 and 2 have been unambiguously established by X-ray crystallography (see below). We have made a number of new amine complexes of ruthenium cis-[Ru(L]L)2- (NH2CH2R)2]21 (L]L = bipy or phen) and find that they too are converted into salts of the corresponding new nitrile complexes [Ru(bipy)2(NCR)2]21 by an excess of aqueous peroxodisulfate at room temperatures (Table 2). Although peroxodisulfate is an eVective stoichiometric oxidant for such conversions, it is surprising that bromate, which we have previously shown3 to function as an eVective cooxidant with trans-[Ru(OH)2O3]22, does not oxidise these amine complexes.Neither peroxodisulfate nor bromate will oxidise the co-ordinated amine ligands in [Os(bipy)2(NH2CH2Ph)2]21, [Ru(CO)2Cl2(NH2CH2Ph)2] or [Ru(NH2CH2Ph)6]Cl2. (c) X-Ray crystallography (i) Crystal structure of cis-[Ru(bipy)2(NH2CH2Ph)2][PF6]2? 0.5MeOH 1.Orange-red crystals of the complex were prepared by reaction of cis-[Ru(bipy)2Cl2] and benzylamine under reflux with subsequent addition of NH4PF6, and recrystallised from methanol. The X-ray structural analysis of complex 1 (Fig. 1) confirms it to have the expected cis-configuration for the benzylamine ligands. The co-ordination geometry at ruthenium is slightly distorted octahedral, with angles in the ranges 78.7(2) to 98.9(3)8 and 169.3(2) to 175.6(2)8, the marked contractions in the cis angles being due to the bite of the 2,29-bipyridyl ligands.The six Ru]N bond lengths (Table 3) clearly emphasise their diVering chemical natures, those to the two benzylamine ligands being noticeably longer at 2.166(6) [N(2)] and 2.174(7) Å [N(1)] than those to the chelating 2,29-bipyridyl ligands [ranging between 2.041(6) and 2.075(6) Å]. These diVerences reflect the sp3 and sp2 nature of the respective co-ordinated nitrogen centres. The two N]C (benzyl) distances (average 1.43J. Chem.Soc., Dalton Trans., 1998, Pages 2819–2825 2821 Table 2 Analytical and spectroscopic data for cis-[Ru(L]L)2(NH2CH2R)2]21 andcis-[Ru(L]L)2(NCR)2]21 complexes Analytical data a (%) Vibrational data b/cm21 1H NMRc (d) Complex [Ru(bipy)2(NH2CH2Ph)2][PF6]2 [Ru(bipy)2(NCPh)2][PF6]2 [Ru(bipy)2(NH2CH2C6H4Cl-o)2][PF6]2?H2O [Ru(bipy)2(NCC6H4Cl-o)2][PF6]2 [Ru(bipy)2(NH2CH2C6H4Cl-m)2][PF6]2?H2O [Ru(bipy)2NCC6H4Cl-m)2][PF6]2?H2O [Ru(bipy)2(NH2CH2C6H4Me-o)2][PF6]2 [Ru(bipy)2(NCC6H4Me-o)2][PF6]2 [Ru(bipy)2(NH2CH2C6H4OMe-o)2][PF6]2?H2O [Ru(bipy)2(NCC6H4OMe-o)2][PF6]2?2H2O [Ru(bipy)2(NH2CH2C6H4OMe-p)2][PF6]2 [Ru(bipy)2(NCC6H4OMe-p)2][PF6]2 [Ru(phen)2(NH2CH2Ph)2][PF6]2?H2O [Ru(phen)2(NCPh)2][PF6]2 [Ru(phen)2(NH2CH2C6H4OMe-p)2][PF6]2?H2O [Ru(phen)2(NCC6H4OMe-p)2][PF6]2?H2O C 44.3 (44.5) 44.3 (44.9) 41.3 (40.7) 41.5 (41.8) 40.7 (40.7) 41.3 (41.0) 45.3 (45.7) 46.3 (46.1) 43.5 (43.4) 43.5 (43.0) 44.5 (44.2) 44.5 (44.6) 46.8 (46.4) 47.5 (47.6) 45.3 (46.0) 46.3 (46.4) H 3.8 (3.7) 3.1 (2.9) 3.3 (3.4) 3.4 (2.5) 3.4 (3.3) 2.3 (2.6) 3.7 (4.0) 3.5 (3.2) 4.2 (4.1) 3.5 (3.4) 4.2 (3.9) 3.5 (3.1) 3.4 (3.7) 2.8 (2.9) 3.6 (3.9) 3.2 (3.1) N 8.8 (9.2) 8.8 (9.2) 8.6 (8.4) 8.6 (8.6) 8.5 (8.2) 8.5 (8.4) 8.8 (8.9) 8.9 (9.0) 8.3 (8.4) 8.7 (8.4) 8.8 (8.6) 8.7 (8.7) 8.5 (8.5) 8.6 (8.8) 8.0 (8.0) 8.0 (8.1) nasym(NH2) 3316m 3320m 3314m 3311m 3260m 3320m 3330m 3287m 3280m nsym(NH2) 3188w 3244w 3134m 3266m 3160w 3217m 3233m 3244m 3212m n(CN) – 2240w 2245s, 2240w – 2240w 2245s – 2230w 2241s – 2235w – 2240w – 2241w – 2242w – 2236w 2244s d(NH2) 1600m 1617m 1615m 1603m 1620m 1618m 1622m 1623m CH2 3.2 (m), 3.8 (m) 3.4 (m), 3.7 (m) 3.2 (m), 3.8 (m) 3.3 (m), 3.6 (m) 3.4 (m), 3.7 (m) 3.5 (m), 3.8 (m) 3.2 (m), 3.8 (m) 3.3 (m), 3.7 (m) NH2 4.6 (t), 4.8 (t) 4.4 (t), 4.8 (t) 4.4 (t), 4.7 (t) 4.6 (t), 4.8 (t) 4.4 (t), 4.7 (t) 4.3 (t), 4.8 (t) 4.4 (t), 4.9 (t) 4.4 (t), 4.8 (t) a Calculated values in parentheses.b Raman data italicised.c In (CD3)2CO vs. SiMe4; resonances due to bipy/phen and phenyl omitted. Å) are unexceptional, reflecting their single-bond character; the angles at the benzylamine nitrogen atoms are sightly enlarged at 121.7(7) [N(1)] and 124.6(6)8 [N(2)]. Fig. 1 The molecular structure of the cation in complex 1, showing the overlap between one of the benzylamine ligands and one of the 2,29- bipyridyl units. An interesting feature of the conformation of the cation is the adoption of a gauche geometry about the N]CH2 bond in one benzylamine ligand [N(2)], whereas in the other [N(1)] the geometry is anti.The former conformation is stabilised by an intramolecular p–p stacking interaction between the benzyl ring and adjacent bipyridyl ligand (mean interplanar separation ca. 3.2 Å). The opposite face of the benzyl ring is involved in an intermolecular aromatic–aromatic edge-to-edge interaction with the phenyl ring of the other benzylamine ligand (centroid– centroid separation 4.78 Å).The combined eVect of these two interactions is to produce loosely linked chains of molecules that extend in the crystallographic a direction (Fig. 2). Centrosymmetrically related pairs of chains are cross-linked by additional T type aromatic–aromatic edge-to-edge interactions between the face of the N(1) benzylamine and the edge of the N(5) pyridine ring and vice versa (centroid–centroid separation 4.93 Å). (ii) Crystal structure of cis-[Ru(bipy)2(NCPh)2][PF6]2?CH2Cl2 2.An aqueous solution of cis-[Ru(bipy)2(NH2CH2Ph)2]21 was treated with an excess of aqueous peroxodisulfate and the yellow product, cis-[Ru(bipy)2(NCPh)2][PF6]2, was isolated by addition of NH4PF6. It was recrystallised from dichloromethane as yellow crystals. Fig. 2 Part of one of the aromatic–aromatic edge-to-face linked chains of cations present in the crystals of complex 1.2822 J. Chem. Soc., Dalton Trans., 1998, Pages 2819–2825 Table 3 Selected bond lengths (Å) and angles (8) for complex 1 Ru]N(1) Ru]N(4) N(1)]C(7) N(6)]Ru]N(4) N(6)]Ru]N(3) N(6)]Ru]N(2) N(3)]Ru]N(2) N(5)]Ru]N(1) C(7)]N(1)]Ru 2.174(7) 2.058(6) 1.401(13) 92.2(2) 94.9(2) 89.3(2) 94.6(2) 98.9(3) 121.7(7) Ru]N(2) Ru]N(5) N(2)]C(14) N(6)]Ru]N(5) N(4)]Ru]N(3) N(4)]Ru]N(2) N(6)]Ru]N(1) N(3)]Ru]N(1) C(14)]N(2)]Ru 2.166(6) 2.063(6) 1.454(11) 78.7(2) 78.8(2) 173.4(2) 175.6(2) 87.9(3) 124.6(6) Ru]N(3) Ru]N(6) N(4)]Ru]N(5) N(5)]Ru]N(3) N(5)]Ru]N(2) N(4)]Ru]N(1) N(2)]Ru]N(1) 2.075(6) 2.041(6) 92.7(2) 169.3(2) 93.9(2) 91.6(3) 87.2(3) Table 4 Selected bond lengths (Å) and angles (8) for complex 2 Ru]N(1) N(1)]C(1) N(1)]Ru]N(19) N(2)]Ru]N(29) N(1)]Ru]N(39) C(1)]N(1)]Ru 2.032(4) 1.140(6) 91.9(2) 90.2(2) 96.3(2) 177.6(4) Ru]N(2) C(1)]C(7) N(1)]Ru]N(2) N(1)]Ru]N(3) N(2)]Ru]N(39) N(1)]C(1)]C(7) 2.045(4) 1.415(6) 89.2(2) 87.7(2) 96.9(2) 178.0(6) Ru]N(3) N(1)]Ru]N(29) N(2)]Ru]N(3) N(3)]Ru]N(39) 2.064(4) 175.1(2) 78.9(2) 174.2(2) The X-ray structural analysis of complex 2 (Fig. 3) confirms that the expected stoichiometric oxidation from co-ordinated benzylamine to benzonitrile has occurred. The two benzylamine ligands present in 1 have retained their cis relationship in the oxidised product 2, the N(1)]Ru]N(19) angle being 91.9(2)8. The complex possesses crystallographic C2 symmetry about an axis passing through the ruthenium centre and bisecting the two benzonitrile ligands. The co-ordination geometry at ruthenium is slightly distorted octahedral, with angles in the ranges 78.9(2) to 96.9(2) and 174.2(2) to 175.1(2)8, the marked contractions observed in the cis angles being as expected due to the bite of the 2,29-bipyridyl ligands.The independent Ru]N bond lengths (Table 4) clearly reflect their diVering chemical natures, with those to the benzonitrile ligands [2.032(4) Å] being noticeably shorter than those to the 2,29-bipyridyl ligands (see above). The corresponding bonds to the benzylamine ligands in 1 are longer [at 2.166(6) and 2.174(7) Å], consistent with the change from an sp3 hybridisation in 1 to sp in 2.The bond distances to the sp2 hybridised 2,29-bipyridyl ligands are essentially the same in both structures [2.041(6) to 2.075(6) Å in 1 and 2.045(4) and 2.064(4) Å in 2] being, as expected, intermediate with respect to those to the sp3 and sp hybridised benzylamine and benzonitrile ligands in 1 and 2. The oxidation of the benzonitrile ligand in 1 is clearly demonstrated in 2 by the unambiguous triple-bond Fig. 3 The molecular structure of the C2-symmetric cation in complex 2. character for N(1)]C(1) [1.140(6) Å] and the linear geometries at C(1) and N(1) [178.0(6) and 177.6(4)8 respectively]. The orientation of the terminal phenyl rings of the benzonitrile ligands is such that they lie virtually coplanar with their associated trans 2,2-bipyridyl ligands. There is, surprisingly, a marked absence of any intramolecular p–p interactions.The only intermolecular interactions of any note are weak cation–anion C]H? ? ? F interactions between C(11) and C(14) of one of the PF6 anions (the H ? ? ? F distances are 2.49 and 2.36 Å with C]H]F angles of 173 and 1658 respectively). (d) Vibrational and 1H NMR spectra of amine and nitrile complexes There are very slight diVerences in the elemental analyses between cis-[Ru(bipy)2(NH2CH2R)2]21 and cis-[Ru(bipy)2- (NCR)2]21 salts, but vibrational and 1H NMR spectra clearly demonstrate the presence of either NH2CH2R or NCR ligands.In Table 2 we list infrared, Raman and 1H NMR data on the ruthenium amine and nitrile complexes isolated in this work. The NH2 stretches n(NH2) and deformations d(NH2) are present in the spectra of the amine complexes but absent in those of the nitrile species. Bands near 2240 cm21 of moderate intensity appear in the infrared spectra of the nitrile complexes and as strong bands in the Raman, clearly arising from the CN stretch n(CN).A cis geometry is indicated for the [Ru(bipy)2(NCR)2]21 and [Ru(phen)2(NCR)2]21 species by the fact that the infrared and Raman bands have significantly diVerent frequencies: in the former it is the asymmetric CN stretch nasym(CN) which is the strongest band while the symmetric stretch nsym(CN) is stronger in the Raman. The 1H NMR spectra demonstrate that a cis configuration for the amine complexes is retained in solution. Thus, for cis- [Ru(bipy)2(NH2CH2Ph)2] the peaks due to bipyridyl are very complex suggesting a cis rather than a trans structure; furthermore, the amine protons appear as two multiplets (at d 4.8 and 4.6 vs.SiMe4); on shaking a solution of the complex with 2H2O these peaks disappear due to exchange with deuterium. The methylene protons also appear as two multiplets (at d 3.2 and 3.4); for trans-[Ru(bipy)2(NH2CH2R)2]21 only one set of resonances for amine and methylene groups would be expected, but for the cis isomer there will be two sets since this isomer is diastereotopic.Conclusion We have shown that the trans-[Ru(OH)2O3]22/S2O8 22 reagent isJ. Chem. Soc., Dalton Trans., 1998, Pages 2819–2825 2823 eVective for the dehydrogenation of primary amines (particularly benzylic amines) to the corresponding nitriles under ambient conditions; over longer periods of time nitrile hydration to amides occurs. The reaction may proceed via initial formation of a co-ordinated amine complex; as models for reaction of co-ordinated amine species with peroxodisulfate we made a number of new complexes cis-[Ru(L]L)2(NH2CH2R)2]21 (L]L = bipy or phen) and have shown that these are oxidised by an excess of peroxodisulfate to the corresponding nitrile complexes cis-[Ru(L]L)2(NCR)2]21.The crystal structures of two such species, cis-[Ru(bipy)2(NH2CH2Ph)2][PF6]2?0.5MeOH and cis-[Ru(bipy)2(NCPh)2][PF6]2?CH2Cl2, have been determined. Experimental Chemicals were from Aldrich and used without further purifi- cation.The compounds RuCl3?nH2O and Na2[OsCl6]?nH2O were obtained from Johnson Matthey Ltd. Preparation of the trans-[RuO3(OH)2]22/S2O8 22 reagent The literature procedure 3 was used but with slightly diVerent concentrations: RuCl3?nH2O (0.024 g, 0.1 mmol) was predissolved in water (5 cm3) and an excess of K2S2O8 (2.8 g, 0.01 mol) in aqueous molar KOH (25 cm3) was added to give an orange solution. Catalytic dehydrogenation of amines to nitriles The reactions were performed at room temperature by dropwise addition of the amines (RCH2NH2; 2 mmol) over a period of 5 min to a vigorously stirred solution (100 cm3) of the trans- [Ru(OH)2O3]22/S2O8 22 reagent.The initial reaction mixture is dark green; when the reaction is complete the original orange colour of ruthenate reappears. The mixture was then extracted with diethyl ether (3 × 25 cm3), the ether extracts dried over anhydrous MgSO4 and the ether removed.Products were characterised by 1H NMR, IR spectra and melting points where appropriate. Hydration of nitriles to amides Reactions were carried out as above, but for 24 h periods; benzene was used rather than diethyl ether for extracting the products. Preparation and reactions of ruthenium amine and nitrile complexes The complex [Ru(bipy)2(NH2CH2Ph)2][PF6]2 was made by a method based on that of Meyer and co-workers 19 but the hexa- fluorophosphate salt was isolated in place of the perchlorate salt. The complex cis-[RuCl2(bipy)2], made by the literature method21 (0.2 g, 0.4 mmol), was suspended in 50% aqueous methanol (30 cm3).Benzylamine (2 g, 18.7 mmol) was added and the solution refluxed under nitrogen for 2 h. Methanol was evaporated oV, the solution cooled and extracted with diethyl ether (3 × 20 cm3) to remove the excess of benzylamine. The remaining aqueous solution was filtered and the complex precipitated by slow addition of a saturated solution of NH4PF6, and the red precipitate filtered oV, washed with water, diethyl ether and then dried in vacuo.Yield of red crystals 0.33 g, 0.36 mmol (90%). The methanol adduct 1 was made by recrystallisation of this material from MeOH. Other [Ru(bipy)2(NH2CH2R)2][PF6]2 salts. The complex cis- [RuCl2(bipy)2]?2H2O (0.2 g, 0.38 mmol) was suspended in 50% aqueous methanol (30 cm3). The amine (2 g) was added and the solution refluxed under nitrogen for 2 h. The methanol was evaporated oV, the solution cooled and extracted with diethyl ether (3 × 20 cm3) to remove the excess of amine.The remaining aqueous solution was filtered and the complex precipitated by slow addition of a saturated solution of NH4PF6. The precipitate was filtered oV, washed with water, diethyl ether and then dried in vacuo. The complexes cis-[Ru(phen)2(NH2CH2Ph)2][PF6]2?H2O and cis-[Ru(phen)2(NH2CH2C6H4OMe-p)2][PF6]2?H2O were similarly prepared, cis-[RuCl2(phen)2]?2H2O (made by the literature method21) (0.2 g, 0.35 mmol) replacing cis-[RuCl2(bipy)2]? 2H2O.Dehydrogenation of cis-[Ru(bipy)2(NH2CH2Ph)2]21 to cis- [Ru(bipy)2(NCPh)2][PF6]2. The complex cis-[RuCl2(bipy)2] (0.2 g, 0.4 mmol) was suspended in 50% aqueous methanol (30 cm3). Benzylamine (2 g, 18.7 mmol) was added and the solution refluxed under nitrogen for 2 h. Methanol was evaporated oV, the solution cooled and extracted with diethyl ether (3 × 20 cm3) to remove the excess of benzylamine.The remaining aqueous solution of [Ru(bipy)2(NH2CH2Ph)2]21 was degassed and aqueous K2S2O8 (3%, 10 cm3) was added with stirring under nitrogen at room temperature for 1.5 h; the mixture changed gradually from red to orange and finally to yellow. The yellow solution was filtered and a yellow precipitate was formed by adding a saturated solution (10 cm3, 10%) of NH4PF6. The precipitate of cis-[Ru(bipy)2(NCPh)2][PF6]2 was collected, washed with water, diethyl ether and dried in vacuo.Yield 0.33 g, 0.36 mmol (90%). The dichloromethane adduct 2 was made by recrystallisation of this material from CH2Cl2. General procedure for dehydrogenation of cis-[Ru(bipy)2- (NH2CH2R)2]21 to cis-[Ru(bipy)2(NCR)2]21 by peroxodisulfate. The complex cis-[RuCl2(bipy)2]?2H2O (0.2 g, 0.4 mmol) was suspended in 50% aqueous methanol (30 cm3), the amine (2 g) added and the solution refluxed under nitrogen for 2 h. Methanol was evaporated oV, the solution cooled and extracted with diethyl ether (3 × 20 cm3) to remove the excess of amine.The remaining aqueous solution was filtered and degassed by nitrogen, then aqueous K2S2O8 (3%, 10 cm3) solution added with stirring under nitrogen at room temperature for 2 h; the mixture changed gradually from red to orange and finally to yellow. The yellow solution was filtered and a yellow precipitate formed by adding a saturated solution (10 cm3, 10%) of NH4PF6. The precipitate was collected, washed with water and dried in vacuo.The complexes cis-[Ru(phen)2(NCPh)2][PF6]2 and cis-[Ru- (phen)2(NCC6H4OMe-p)2][PF6]2?H2O were similarly prepared, using cis-[RuCl2(phen)2]?2H2O (0.2 g, 0.35 mmol) in place of cis-[RuCl2(bipy)2]?2H2O. Ruthenium carbonyl complexes. A ruthenium carbonylcontaining solution using ethanol as the solvent was prepared by following the procedure of Chatt et al.22 The compound RuCl3?nH2O (4.2 g) was added to ethanol (75 cm3), heated at reflux and carbon monoxide passed into the solution.A blood red colour was formed after 5 h. This solution was used for the following reactions. [RuCl2(CO)2(NH2CH2Ph)2]. The procedure of Wilkinson and co-workers 23 was used with some modifications. Benzylamine (0.4 g, 3.7 mmol) was added slowly to the red solution (9 cm3). After 5 min a change to green occurred and a pale green precipitate was formed. This was filtered oV, washed with ethanol, diethyl ether and dried in vacuo. Yield 0.3 g, 0.6 mmol (67%) (Found: C, 43.5; H, 4.7; N, 6.3.Calc. for C16H18Cl2- N2O2Ru: C, 43.4; H, 4.1; N, 6.3%). [Ru(NH2CH2Ph)6]Cl2. Benzylamine (2 g, 18.7 mmol) was added slowly to the red solution (9 cm3); there was an immediate change to green and the reaction mixture was heated at reflux for 15 min, after which time a red crystalline precipitate was formed. The precipitate was filtered oV, washed with ethanol and diethyl ether (4 × 25 cm3) to remove the excess of2824 J. Chem.Soc., Dalton Trans., 1998, Pages 2819–2825 benzylamine and dried in vacuo. Yield 0.6 g, 0.73 mmol (73.7%) (Found: C, 61.5; H, 6.5; N, 10.1. Calc. for C42H54Cl2N6Ru: C, 61.9; H, 6.7; N, 10.3%). Neither of the above two complexes could be oxidised to the corresponding nitrile complexes with an excess of peroxodisulfate. Osmium complexes cis-[OsCl2(bipy)2]. This was prepared by a variation of the literature method.24 The salt Na2[OsCl6]?nH2O (1 g, 2.2 mmol) and 2,29-bipyridyl (0.72 g, 4.6 mmol) were added to ethylene glycol (50 cm3) and the mixture was heated at reflux for 45 min under nitrogen.Since the crude reaction mixture contained both cis-[OsCl2(bipy)2] and cis-[Os(bipy)2Cl2]1, an equal volume of saturated sodium dithionite was added to the cooled reaction mixture in order to reduce the excess of OsIII to OsII. The purple-black precipitate formed was isolated by filtration, washed with water to remove [Os(bipy)3]21 and other ionic products, and washed with a large volume of diethyl ether.Yield 0.5 g, 0.87 mmol (87%) (Found: C, 41.6; H, 2.3; N, 9.7. Calc. for C20H16Cl2N4Os: C, 41.9; H, 2.8; N, 9.8%). [Os(bipy)2(NH2CH2Ph)2][PF6]2. The complex cis-[OsCl2- (bipy)2] (0.1 g, 0.17 mmol) was suspended in 50% aqueous ethanol (30 cm3), benzylamine (2 g, 18.7 mmol) was added and the solution refluxed under nitrogen for 4 h. It changed from purple to dark yellow, then ethanol was evaporated oV, the solution cooled and extracted with diethyl ether (3 × 20 cm3) to remove the excess of benzylamine.The remaining aqueous solution was filtered and the complex precipitated by slow addition of a saturated solution of NH4PF6. The brown precipitate was filtered oV, washed with water, diethyl ether and then dried in vacuo. Yield 0.06 g, 0.06 mmol (35%) (Found: C, 39.8; H, 3.1; N, 8. Calc. for C34H34F12N6OsP2: C, 40.6; H, 3.4; N, 8.4%). This complex was not oxidised by peroxodisulfate under the conditions used for the ruthenium analogue.X-Ray crystallography Crystal data. [C34H34N6Ru][PF6]2?0.5CH3OH 1, M = 933.7, monoclinic, space group P21/c (no. 14), a = 11.987(1), b = 20.692(2), c = 16.544(1) Å, b = 106.73(1)8, U = 3929.7(4) Å3, Z = 4, Dc = 1.578 g cm23, m(Cu-Ka) = 48.4 cm21, F(000) = 1884, T = 293 K, orange-red block, 0.27 × 0.17 × 0.12 mm. [C34H26N6Ru][PF6]2?CH2Cl2 2, M = 994.5, monoclinic, space group C2/c (no. 15), a = 13.644(1), b = 26.405(2), c = 11.371(1) Å, b = 90.21(1)8, U = 4096.7(6) Å3, Z = 4 (the molecule has crystallographic C2 symmetry), Dc = 1.612 g cm23, m(Mo-Ka) = 6.81 cm21, F(000) = 1984, T = 293 K, orange prism, 0.67 × 0.67 × 0.23 mm.Data collection and processing. Data were measured on Siemens P4/PC diVractometers with graphite monochromated Cu-Ka (Mo-Ka) radiation for complex 1 (2) using w scans. 5839 (3596) Independent reflections were measured [2q < 120 (50)8] of which 4278 (2780) had |Fo| > 4s(|Fo|) and were considered to be observed.The data were corrected for Lorentz-polarisation factors, and semiempirical absorption corrections (based on y scans) applied; the maximum and minimum transmission factors were 0.51 and 0.39 for 1 and 0.83 and 0.71 for 2 respectively. Structure analysis and refinement. The structures were solved by direct methods and the non-hydrogen atoms of the cationic complexes refined anisotropically. In 1 both of the hexafluorophosphate anions were disordered; in each case this disorder was resolved into two, discrete, partial occupancy orientations, with the atoms of the major occupancy orientation being refined anisotropically.The half occupancy included solvent methanol molecule in 1 was found to be distributed over three discrete sites, all of which were refined isotropically. The included dichloromethane solvent molecule in 2 was disordered over a crystallographic C2 axis, and this was resolved into two symmetry related half occupancy orientations, both of which were refined anisotropically.The positions of the hydrogen atoms in both structures were idealised, assigned isotropic thermal parameters [U(H) = 1.2Ueq(C/N), U(H) = 1.5Ueq(O)], and allowed to ride on their parent atoms. Refinements were by full matrix least squares based on F2 to give R1 = 0.062 (0.050), wR2 = 0.153 (0.112) for the observed data and 537 (266) parameters for 1 (2) respectively. The maximum and minimum residual electron densities in the final DF map were 0.70 and 20.75 e Å23 for 1 and 0.35 and 20.23 e Å23 for 2 respectively.The mean and maximum shift/error ratios in the final refinement cycle were 0.001 and 20.031 for 1 and 0.000 and 0.000 for 2 respectively. All computations were carried out using the SHELXTL PC program system.25 CCDC reference number 186/1076. Instrumentation Infrared spectra were measured on a Perkin-Elmer series 1720 FTIR instrument, FT Raman spectra on a Perkin-Elmer series 1700 instrument with Nd-YAG laser excitation at 1064 nm and 1H NMR spectra on a JEOL EX-270 spectrometer. Microanalyses were carried out by the Imperial College Microanalytical Service.The GC-MS data were obtained by Mr. John Barton on a Micromass AutoSpec, fitted with a Hewlett-Packard 5890 gas chromatograph and an SGE BPX5 column. Acknowledgements We thank the Egyptian Ministry of Education for a grant to one of us (A. G. F. S.), Johnson Matthey Ltd. for a loan of ruthenium trichloride, Dr.A. J. Bailey for obtaining crystals of cis-[Ru(bipy)2(NCPh)2][PF6]2?CH2Cl2 and John Barton for GC-MS measurements. We also thank the University of London Intercollegiate Research Service (ULIRS) for the Raman spectrometer. References 1 Part 17, A. J. Bailey, M. G. Bhowon, W. P. GriYth, A. G. F. Shoair, A. J. P. White and D. J. Williams, J. Chem. Soc., Dalton Trans., 1997, 3245. 2 A. J. Bailey, L. D. Cother, W. P. GriYth and D. M. Hankin, Transition Met. Chem., 1995, 20, 590; W.P. GriYth, Chem. Soc. Rev., 1992, 21, 179. 3 A. J. Bailey, W. P. GriYth, S. I. Mostafa and P. A. Sherwood, Inorg. Chem., 1993, 32, 268. 4 S. V. Ley, J. Norman, W. P. GriYth and S. P. Marsden, Synthesis, 1994, 639; W. P. GriYth, S. V. Ley, G. P. Whitcombe and A. D. White, J. Chem. Soc., Chem. Commun., 1987, 1625. 5 J. M. JolliVe, S. V. Ley and D. J. Williams, J. Chem. Soc., Chem. Commun., 1990, 1219; A. C. Dengel, A. M. El-Hendawy, W. P. GriYth, S. I. Mostafa and D.J. Williams, J. Chem. Soc., Dalton Trans., 1992, 3489. 6 A. J. Bailey, W. P. GriYth and P. D. Savage, J. Chem. Soc., Dalton Trans., 1995, 3537; A. J. Bailey, W. P. GriYth, A. J. P. White and D. J. Williams, J. Chem. Soc., Chem. Commun., 1994, 1833. 7 G. Green, W. P. GriYth, D. M. Hollinshead, S. V. Ley and M. Schröder, J. Chem. Soc., Perkin Trans. 1, 1984, 681; M. Schröder and W. P. GriYth, J. Chem. Soc., Chem. Commun., 1979, 58. 8 G. A. Lee and H. H. Freedman, Tetrahedron Lett., 1976, 20, 1641. 9 J. B. Lee, C. Parkin, M. J. Shaw, N. A. Hampson and K. I. MacDonald, Tetrahedron, 1973, 29, 751; T. G. Clarke, N. A. Hampson, J. B. Lee, J. R. Morley and R. Scanlon, Tetrahedron Lett., 1968, 5685. 10 S. Yamazaki and Y. Yamazaki, Bull. Chem. Soc. Jpn., 1990, 63, 301. 11 P. Capdevielle, A. Lavine and M. Maumy, Synthesis, 1989, 453. 12 A. J. Bailey and B. R. James, Chem. Commun., 1996, 2343; 1997, 507. 13 R. Tang, S. E. Diamond, N. Neary and F. Mares, J. Chem.Soc., Chem. Commun., 1978, 562.J. Chem. Soc., Dalton Trans., 1998, Pages 2819–2825 2825 14 S. Cenini, F. Porta and M. Pizzotti, J. Mol. Catal., 1982, 15, 297. 15 A. Goti and M. Romani, Tetrahedron Lett., 1994, 35, 6567. 16 M. Schröder, Ph.D. Thesis, University of London, 1978. 17 S. E. Diamond, B. Grant, G. M. Tom and H. Taube, Tetrahedron Lett., 1974, 4025. 18 T. GhaYar and A. W. Parkins, Tetrahedron Lett., 1995, 36, 8657; A. W. Parkins, Platinum Met. Rev., 1996, 49, 169. 19 B. P. Sullivan, D. J. Salmon and T. J. Meyer, Inorg. Chem., 1978, 17, 3334. 20 S. E. Diamond, G. M. Tom and H. Taube, J. Am. Chem. Soc., 1975, 97, 2661. 21 G. Sprintschnik, H. W. Sprintschnik, P. P. Kirsch and D. G. Whitten, J. Am. Chem. Soc., 1977, 99, 4947. 22 J. Chatt, B. L. Shaw and A. E. Field, J. Chem. Soc., 1964, 3466. 23 J. V. Kingston, J. W. S. Jamieson and G. Wilkinson, J. Inorg. Nucl. Chem., 1967, 29, 133. 24 D. A. Buckingham, F. P. Dwyer, H. A. Goodwin and A. M. Sargeson, Aust. J. Chem., 1964, 17, 325. 25 SHELXTL PC, version 5.03, Siemens Analytical X-Ray Instruments, Inc., Madison, WI, 1994. Received 1st June 1998; Paper 8/04071KJ. Chem. Soc., Dalton Trans., 1998, Pages 2819–2825 2825 14 S. Cenini, F. Porta and M. Pizzotti, J. Mol. Catal., 1982, 15, 297. 15 A. Goti and M. Romani, Tetrahedron Lett., 1994, 35, 6567. 16 M. Schröder, Ph.D. Thesis, University of London, 1978. 17 S. E. Diamond, B. Grant, G. M. Tom and H. Taube, Tetrahedron Lett., 1974, 4025. 18 T. GhaYar and A. W. Parkins, Tetrahedron Lett., 1995, 36, 8657; A. W. Parkins, Platinum Met. Rev., 1996, 49, 169. 19 B. P. Sullivan, D. J. Salmon and T. J. Meyer, Inorg. Chem., 1978, 17, 3334. 20 S. E. Diamond, G. M. Tom and H. Taube, J. Am. Chem. Soc., 1975, 97, 2661. 21 G. Sprintschnik, H. W. Sprintschnik, P. P. Kirsch and D. G. Whitten, J. Am. Chem. Soc., 1977, 99, 4947. 22 J. Chatt, B. L. Shaw and A. E. Field, J. Chem. Soc., 1964, 3466. 23 J. V. Kingston, J. W. S. Jamieson and G. Wilkinson, J. Inorg. Nucl. Chem., 1967, 29, 133. 24 D. A. Buckingham, F. P. Dwyer, H. A. Goodwin and A. M. Sargeson, Aust. J. Chem., 1964, 17, 325. 25 SHELXTL PC, version 5.03, Siemens Analytical X-Ray Instruments, Inc., Madison, WI, 1994. Received 1st June 1998; Paper 8/04071K

ISSN:1477-9226    

DOI:10.1039/a804071k

出版商:RSC    

年代:1998

数据来源: RSC