3.. .. , ware \ ‘ Q} 1/! .zlzhfflvr I. I’fi¢l.li‘ a. > ‘ .. ‘ m: . 1. ; 3 n... 2.33.53. 2.. in? .3wa ; r5»... av ,I u! . ‘n‘lif 2,1: .3. 1 z!!! 5:; o ,.. o 31.3..) fiat , , (‘3. w\0ulvti ‘ . (5‘ \ zégwéésfii 1. Them ([373 all) Illlllllllllllllllll‘ll‘lllllllllll 3 1293 01389 This is to certify that the dissertation entitled The Synthesis of Coordination and Organometallic Compounds for Two-Electron Thermal and Photochemical Redox Reactions presented by Douglas H. Motry has been accepted towards fulfillment of the requirements for flit 0 degree in TQWg’B m‘o (:Aenm'a‘ify W @rC/W Major professor Date .2 '7ng MS U is an Affirmatiw Action/Equal Opportunity Institution 0-12771 LIBRARY Michigan State University PLACE II RETURN BOX to romovothb ohookout from your rooord. TO AVOID FINES return on or baton dot. duo. DATE DUE DATE DUE DATE DUE l I | MSU Is An Atfirmotivo Action/Equal Opponunlty Inothlon WWI THE SYNTHESIS OF COORDINATION AND ORGANOMETALLIC COMPOUNDS FOR TWO- ELECTRON THERMAL AND PHOTOCHEMICAL REDOX REACTIONS By Douglas H. Motry A Dissertation Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Chemistry 1995 ABSTRACT THE SYNTHESIS OF COORDINATION AND ORGANOMETALLIC COMPOUNDS FOR TWO-ELECTRON THERMAL AND PHOTOCHEMICAL REDOX REACTIONS By Douglas H. Motry Most important reactions minimally involve two—electron transformations. The synthesis of organometallic and coordination complexes targeted at effecting novel two-electron reactions from the ground and excited state have been investigated. The photochemist attempts to use the energy of a harnessed photon to promote the two-electron reaction. We have found that two-electron mixed valence compounds of the type M'H—Mn+1 provide a unique foundation for the design of multielectron photoreaction schemes. A two-electron mixed valence complex RhoRhII undergoes oxidation by halogens to give novel compounds which exhibit long-lived excited state. The synthesis of the two-electron mixed valency complexes has been expanded to iridium systems and serves to further generalize this class of compounds. Survey reactions of other transition metal complexes will also be presented. From the ground state, the organometallic chemist designs new molecules to overcome kinetic barriers to reveal new reactivity. This is the case for early transition complexes with metal-boron bonds. We have synthesized Group 5 boryls by the reaction of anionic tantalum complexes with chloroboranes and titanocene boronate esters by taking advantage TiILIV redox chemistry. The reactions of these compounds with olefin and hydride ligands offers mechanistic information on the formation of early metal boryls. For the titanium case, catecholborane shows unique reactivity with respect to hydroboration in the formation on vinyl boronate esters (hydroboration without bond reduction) when olefin is bonded to titanium. A proposed mechanism for this “hydroboration” reaction will be presented. To my mother, Barbara Motry iv ACKNOWLEDGMENTS I would like to express my appreciation to the many people who have helped me in achieving this goal and I am deeply indebted. First, I'd like to thank Dr. Mitch Smith III for his guidance and excellence in helping me to better understand chemical synthesis. I would also like to thank Dr. Daniel N ocera for the opportunity to work in his laboratory and to listen to his many entertaining stories. Second, I would like to thank many present and past group members. Janice Kadis and I shared many wonderful and happy times together. I would also like to thank Dean Lantero who has been very helpful to work with in the Smith lab. Other people who have been instrumental to my success in Graduate School are Mark Torgerson, Zoe Pikramenou, Jeff Zaleski, Claudia Tmro, Carolyn Hsu, and Carl Iverson, and Baixin Qian and the remaining members of the Smith and Nocera Groups. Lastly, I would like to thank my parents Frank and Barbara Motry. TABLE OF CONTENTS PAGE LIST OF FIGURES ............................................................................ x CHAPTER ONE Introduction to Multielectron Mixed Valence Systems .......................................................................................... l A. Energy Conversion Chemistry by Multielectron Processes .............. l B. Two-Electon Mixed Valence Compounds ....................................... 13 C. Choice of Ligand System for Two-Electron Mixed Valence ' Compounds .................................................................................. 15 CHAPTER TWO Late Metal Two-Electron Mixed Valence Compounds A. Strategies for the Synthesis of New Two-Electron Mixed Valence Systems of Cobalt ............................................................................. 22 B. Strategies for the Synthesis of New Two-Electron Mixed Valence Systems of Iridium ........................................................................... 35 C. Strategies for the Synthesis of New Two-Electron Mixed Valence Systems of Platinum ......................................................................... 40 CHAPTER THREE Introduction to Early Metal Interactions with Boryls ............................................................................................... 42 A. Reasons For Studying Metal-Boron Interactions ............................ 42 B. Metal-Boron vs. Metal-Carbon Bonds ...................... 43 C. Metal Boryl Complexes ................................................................... 44 D. Metal Mediated Hydroboration Chemistry ...................................... 48 vi ABCDEEmmAB. filiiii CHAPTER FOUR Regioisomers of the First Tantalum Boryl PCP”? Complexes Early Metal Boron Bonds Synthesis of the First Tantalum Boryl Complexes... Results and Discussion Conclusion CHAPTER FIVE Hydroboration of Decamethyl Titanocene Ethylene CHAPTER SIX Experimental Part1 A. B. Complex: A Dehydrogenative Boronation Reaction ....................... A Simple Early Metal Olefin Complex Results and Discussion Mechanistic Rationalization Elucidation of Mechanism Reactivity of Cp*2'l‘i(n2-CH2=C(I-I)(B02C10H5)) ............................ Conclusion and Future Directions Solvent Purification Syntheses 1. Synthesis of Dicobalt Complexes a. Starting Materials -- -- b. Attempted synthesis of C02(CH3N(PF7),)3(CO)C12, (1) c. Preparation of C02(CH3N(PF7)2)3Br4 in attempted synthesis of C02(CH3N(PFQ7)3(CO)Br2, (2) ................ d. Preparation of C02(CH3N(PF2)2)314 in attempted synthesis of C02(CH3N(PF7)2)3(C0)12, (3) ................... e. Preparation of (PPh3)2CoC142' in attempted synthesis of vii PAGE 62 62 62 63 79 80 8O 81 87 90 92 94 97 97 98 98 98 98 98 99 PAGE C02(CH3N(PF7)2)3(PPh3)C12 by disproportionation, (4) 9 9 f. Preparation of (PPh3)2CoBr42' in attempted synthesis of . 99 C02(CH3N(PF7)2)3(PPh3)Br2 by disproportionation, (5) . 99 g. Preparation of (PPh3)2CoI42’ in attempted synthesis of 99 . C02(CH3N(PF2)2)3(PPh3)12 by disproportionation, (6) ...... 99 2. Synthesis of Diiridium Complexes 100 3. Preparation of CH3N(PF?)21r2(}r-CH3N(PF7)7)3C12. (7) ..... 100 b. Preparation of CH3N(PF7)21r2(u-CH3N(PF7)2)3(SPh)2, (8) 100 0. Preparation of PhN(PF921r2(u-CH3N(PF2)2)3C12, (9) ......... 100 3. Synthesis of Platinum Complexes _- _ 100 a. Preparation of Pd4(CH3N(PF2)7)10, (10) ............................... 101 C . Spectroscopic Instrumentation and Methods ...................................... 101 1. Electronic Absorption Spectroscopy ........................................... 101 2 Nuclear Magnetic Resonance Spectroscopy .............................. 101 3. Mass Spectroscopy ................................................... 101 4 Infrared Spectroscopy .................................................................... 102 CHAPTER SIX Experimental Part II ......................................................... 103 A . General Procedtnes _- - -- - - - -- .................. 103 B . Syntheses -- _ - ......... ............ 103 l. Syntheses of Borane Reagents ...................................................... 103 a. HBOzCloI-IG, (11) - - -- - .......................... 104 b. DB02C6H4, (12) _ - -_ . - _ ...... 104 c. DBOZClng, (13) 104 2. Syntheses of Tantalum Boryls - i ..... 105 a. Endo-szTaH2(BOZC6H4), (14) and ’ Exo—szTaH2(BOZC6H4), (15) -- 105 viii PAGE 3. Syntheses of Titanium Starting Materials and the Vinylboronate EsterComplex - -- ...... - 106 a. Cp*2T1Cl,(16) _ _ _- -- 106 b. Cp*2T1(n2-CH2=CH7),(17) 106 c. cp*2n(n2-CH2=c1mozcar4), (13) .................................. 107 d. Cp*2Ti(1]2-CH2=CHBOZCH10H6), (19) _- - - 108 4. NMR Tube Reactions: The Reactivity of the Titanium Vinylboronate Ester Complex with Ethylene and Carbon Monoxide - 109 ~ a. Cp*2u(nZCH2=CrIBozcnloH.) + C0(g) ....................... 109 b. Cp*2u(n2CH2=CHB02CHmHs) + C2H4(g) ................... 109 ix 10. ll. 12. l3. 14. 15. LIST OF FIGURES PAGE Schematic diagram of the photosynthetic assembly 4 Two electron mixed valence excited states may be prepared by MMCT from weakly-coupled electrons in a dimetallic core or by light absorption from a bimetallic core with a mixed valency ground state - _ - -_ 7 An ORTEP view of Rh2[CI-I3N(PF2)2]3(PF3)C12 ll Redox chemistry of C02[CH3N(PF2)2]3(CO)2 19 Isotope peak intensity pattern for ions containing the indicated number of bromine atoms 25 FAB mass spectrum of C02[C113N(PF2)2]3Br4 - 27 Product distribution from reaction of cobalt dimer with molecular bromine _ _ - - 30 An ORTEP View of (PPh3)2Col4'2 - 34 FAB mass spectrum of lrz[tt-CH3N(PF2)2]3[‘1}CH3N(PF2)2](CI)2 38 Dinuclear iridium product, Ir2[tt-CH3N(PF2)2]3[‘I}CH3N(PF2)2](C1)2 39 Boron analogy to carbenes and carbynes -_ 43 Frontier orbitals of a bent metallocene interacting with a bent metallocene to create n—backbonding . 46 Boryl transfer which could complement known hydroboration .......... 48 Proposed catalytic cycle for rhodium mediated hydroboration of olefins - 55 Proposed catalytic cycle for lanthanum based hydroboration ............. 60 X 16. 17. 18. 19. 21. i3 27. 29. PAGE 1H NMR of the szTaH2(BO¢C6H4) endo isomer 65 1H NMR of the CpgTaH2(802C61-l4) eta isomer ...... -- 57 ORTEP drawing of endo-TaH2(802QH4) isomer 70 ORTEP drawing of exo-TaI-I2(802QH4) isomer _- -- 72 Tam species in resonance with a la" species ........... 73 me species with bridging hydrides in resonance with a me Metal d-orbitals capable of p-bacltbonding to the empty p-orbital of boron -- - - -- - - _ - ..... - - ......... 76 Kinetic and Thermodynamic Products for CO attack on a zirconium exo acyl . 78 , Tantalum analog to zirconium for formation of an and endo isomers 79 1H NMR spectrum of Cp‘z'l‘imz-CHFCO-IXBOngHq» showing vinyl proton coupling - -- ......... _ 84 Borane resonance illustrating why vinylboranes can serve as electron electrbn deficient Diels-Alder reagents ....... 86 Proposed Reaction Mechanism of Cp‘zTimZ-CHFCHz) with HBOszHa -- _ 89 Deuterium incorporation into Vinylboronate ester - - ......... 91 Reactivity of Vinylboronate ester complex with CO and C2114 ,,,,,,,,,,,,, 93 Catalytic Scheme for Generation of Vinylboronate Ester from Olefin ‘ and Borane - - - - _ - -- - -- - 96 xi Chapter One Multielectron Mixed Valence Systems A. Energy Conversion Chemistry by Multielectron Processes Nature relies upon many multielectron reactions to produce compounds essential for sustaining life. An example is the oxidation of H20 to 02 in photosynthesis, which represents a four-electron process and converts light energy into chemical energy.“4 A number of carefully controlled consecutive events diagrammed in Figure 1 form the basis for this significant energy storage process. A light harvesting center collects and transfers energy to Photosystem 11 (PS 11) generating holes and electrons. The holes and electrons are separated in distance and energy in a transmembrane assembly consisting of two types of quinones, cytochrome f, iron-sulfur proteins, and a plastocyanine. This separation is important in order to prevent recombination of the initially generated hole and electron. Upon reaching Photosystem 1 (PS I), the electron is used to fill a hole created in the photoexcitation of PS I. This electron is subsequently transferred through another series of acceptors, iron-sulfur proteins, ferridoxin, and ferridoxin NADP+ reductase where it is used to convert NADP+ to NADPH. The photogenerated holes remaining in PS II are transferred to a tetrameric manganese cluster, which will undergo four one-electron oxidations before it converts two water molecules to diatomic oxygen. 2 Attempts to model this elegant system of nature have focused on creating photocatalytic systems capable of achieving efficient energy storage by mimicking photosynthesis. One strategy is the synthesis of charge separating networks that propagate efficient electron /hole separation on different time scales.5'6 Most schemes involve the transport of charge separated electrons and holes away from a photochemical reaction center by covalently attaching donor and acceptors to a light harvesting center. This photogenerated charge separated state may persist into the microsecond range; the quantum yield however is usually low. Even with the design of charge separating networks, several obstacles must be overcome within this photochemical energy storage strategy. The chromophore must be highly absorbing to efficiently collect incident radiation. Once the light energy has been absorbed, the lifetime of the excited state must be sufficiently long as to allow for electron transfer events to occur. Once the charges (holes and electrons) have been separated, the equivalents must be trapped to prevent thermal back reaction (recombination of the hole and electron) which would lower efficiency. Finally, several single electron reactions must be coupled to lead to the overall multielectron chemistry of a catalytic center. The intricacies needed to molecularly engineer elegant charge separated networks containing the capacity for energy storage suggests the need to consider alternative strategies to achieve multielectron photocatalytic schemes. If more than one electron can be moved from a discrete excited state, then the structural complexity demanded for efficient charge separation should be relaxed. Moreover, multielectron reaction from an electronically excited core obviates the necessity for charge storage coupled to catalytic redox centers. The development of discrete multielectron processes represents a novel basic reaction type for electronic excited states; it could expand the framework in which to design Figure 1. Schematic diagram of the photosynthetic assembly. n 0.53..— 589: 5.0.) 8.9.3:... 2. 8:. a :6 9 ~\ uzaeoaua Q Oasdbxh . «1.1 , . a u .02. Cl ea ex dir pre pro tl'er. sub: -d8 Hans selec and mulu' were 5 excited state reaction schemes and contribute significantly to our understanding of the mechanisms of multielectron processes. The excited states of binuclear metal complexes may be designed to undergo direct multielectron reactivity owing to several attractive features. These include the presence of a coordinatively unsaturated bimetallic core comprised of metal centers that can exhibit complementary redox function.7 A hint as to the design of multielectron excited states comes from the chemistry of d3----d8 complexes. Numerous spectroscopic studies have shown that the lowest energy d3----d8 excited state is metal-based with two electrons in a triplet configuration,” each localized on a metal center of the binuclear core“).ll Accordingly, the excited state of a binuclear (18 compound may be described chemically as a diradical tethered by a metal-metal bond ([d8°"°d3]* = [-M M-]"‘).12 The presence of a single electron at a coordinatively unsaturated metal center provides a site for substrate activation by photoinduced single electron or atom transfer to yield a mixed valence d"---d8 intermediate and the corresponding substrate radical intermediateuv13 Subsequent trapping of this radical by the d7-- -d8 photoproduct in an ensuing thermal reaction effects overall two-electron transformations such as isopropanol to acetone12 and the dehydrogenation of selected hydrocarbons.l4v15 With the analogy between one-electron chemistry and a biradical excited state suggested, we wondered if excited state multielectron reactions might be emphasized when two metal-localized electrons were singlet coupled within a bimetallic core. Two approaches to study two-electron mixed-valence excited states are described in Figure 2. For excited states derived from weakly-coupled electrons, the multielectron excited state may be prepared by exciting a metal-to—metal charge transfer (MMCI‘) transition. Electrons, originally localized on the individual metal centers of a bimetallic core in the ground state, are paired upon Figure 2. Two electron mixed valence excited states may be prepared by MMCT from weakly-coupled electrons in a dimetallic core or by light absorption from a bimetallic core with a mixed valency ground state. u 2:»...— su III me «a mill at NEE :5— A c5. :2 av $— 2 mi... #025 >5 >2 4 < .— NE.)— cS. _ A 7:5— 7.52 «V c— .u 2i§+ seam ween. 85828 mesa teem Beacon the absorption of a photon to produce an excited state that is zwitterionic, : 8 M+, in nature. The formation of such zwitterionic excited states is consistent with the electronic structure of quadruple metal-metal bond complexes. Metal- localized excited states associated with promotion of electrons to and from the 5 and 8" orbitals exhibit MMCT character (i.e. M2(II,II)* a M2(I,III)).16 With regard to two-electron mixed-valenc , a M"“1—M“’1 excited state is hoto enerated Y P 8 from a M“—M" ground state. Two-electron reductions of a substrate may be promoted at the :M" site, whereas substrates susceptible to two-electron oxidation may react at the M+ site. The visible irradiation of the quadru le P bonded W2(dppm)2Cl4 (dppm = (bis)diphenylphosphinomethane) induces a metal-metal charge transfer (MMCT) giving rise to a two-electron mixed valence excited state, Mn'l—M“+1, where the electron is contained in a localized or weakly coupled orbital. This intermediate mixed valence excited state is trapped by an intramolecular rearrangement to an edge-sharing bioctahedral complex, which reacts in the presence of CH3I by oxidative addition to one side of the molecule to generate a Wme binuclear complex. CH3I 1 on,“ (31’ m p ..‘.$C LII,“ I o.“\‘| w I ‘cn’ l ‘CHS P P (1) E—‘O 9 In the context of two-electron mixed-valence excited states resulting from electrons paired within a binuclear core, quadruplely bonded metal species are unique. Most excited states are derived from the population of molecular orbitals that are delocalized over the entire bimetallic core and, consequently, it is not appropriate to think about electron pair localization. For this reason, the generalization of two-electron excited state chemistry must involve new approaches. One promising line of research is the synthesis of complexes with the two-electron mixed valency already present in the ground state. Here the fomtal oxidation states of the metals within the bimetallic core differ by two, i.e. Mn—M'H'Z. In this approach, the absorption of the photon produces a more energetic excited state that is predisposed to react in two-electron steps at the individual metal centers of the bimetallic core. Two-electron ground state mixed valence complexes, M“—M““2 although rare, offer a unique and novel system from which a multielectron photochemical scheme can be developed. A singly-bonded RhORhII core was stabilized by a bis(difluorophosphino)methylamine ligand several years ago.17 The mixed- ‘valence dimer Rh2[CH3N(PF2)2]3C12(PF3) (designated RhoRhnClz) was oxidized to the symmetrical ClthnRhnClz complex Rh2[CH3N(PF2)2]3Cl4. The RhoRhn two-electron mixed valency is established by the formal electron count and more importantly coordination geometry. This unsymmetrical dinuclear complex displays a trigonal bipyramidal coordination geometry about the Rh0 center and an octahedral coordination geometry about the RhII center as seen in Figure 3. Reactivity studies on the molecule show that two-electron oxidation reactions can be performed at the Rb0 center to afford the symmetrical X2RhnRhnX2 (X = Cl, Br, 1) complex Rh2[CH3N(PF2)2]3X4. Alternatively, reduction of the mixed- valence complex yields the RhoRh0 dimer to complete a homologous d7—d7, d7—d9, d9—d9 series. 10 Figure 3. An ORTEP view of Rh2[CH3N(PF2)2]3(PF3)C12. Thermal parameters are shown at the 50% level. Hydrogen atoms are omitted for the sake of clarity. ll Figure 3 12 The near-ultraviolet absorption profile and red luminescence of this series is consistent with a simple picture of the electronic structure of singly bonded metal-metal complexes. In brief, the electronic structure of the X2RhHRhnX2, RhoRhnxz and the RhoRh0 series is best described as (d6)d1——dl(d6), (d3)d1— d1(d5), and (d3)d1—dl(d3), respectively. The metal-metal 0 interaction results from the pairing of an odd electron in dzz whether it is on a Rho or RhII metal center. Thus the electronic structures of the complexes are for the most part unchanged across the series, with lowest energy transitions resulting from promotion of electrons to the do* orbital. Not surprisingly, the absorption and luminescence spectra across the RhORho, RhoRhnxz and X2RhnRhnX2 series are derived from a dO'* excited state and therefore are similar. Because the do“ excited state presents the possibility for interconverting among the RhoRho, RhORhnXZ, and XthnRhnxz cores via two-electron steps, the system offers a foundation for the design of four-electron photocatalytic schemes. Dissociative photochemistry for RhORhnX2 is suggested by the disappearance of the do“ luminescence in fluid solutions. In THF solutions containing the Br2 trap, 2,3-dimethyl-2-butene, and RhORhHBrz cleanly converts to RhORho. L, TME THF occupies the vacant axial coordination site of the Rh center from Which bromine elimination took place. This two-electron photoreductive-elimination is 13 noteworthy because metal-halide bonds are kinetic and/or thermodynamic sinks in a photochemical cycle. It probably is the case here too, but the reactive do“ excited state permits us to overcome the barrier for metal-halide cleavage. This is a significant result because few systems have been developed in which a photoreagent can be thermally regenerated from a photoproduct and marks the first step toward designing a photocatalytically driven energy storage process. Due to the success of the mixed valence dirhodium system, and the potential applications for energy storage processes, we became interested in the synthesis of new ground state two-electron mixed valence systems and their uses as multielectron photochemical reagents. B. Two-Electron Mixed Valence Compounds The dirhodium system is clearly a member of a rare class of two-electron mixed valence complexes. Nevertheless two-electron mixed valence species, especially of the later transition metals, have long been recognized in the formalism of a two-electron dative metal bond. For instance, in 1985 Kubiak and coworkers18 reported an unprecedented structure of a dinuclear nickel cation with an overall plus two charge. The unusual feature about this structure comes from the differences in the coordination geometries about each nickel atom. One nickel atom possesses an idealized square-based pyramidal coordination geometry while the second possesses a geometry best described as a distorted trigonal bipyramid. Because of the differing metal geometries and presence of bridging atoms, simple assignment of the individual metal oxidation states and electron counts becomes unclear. In describing this complex, Kubiak states that this species is best viewed as an 18 electron NiI center having a dative coordinate bond with a 16 electron NiI center. However, complexes of this type generally do 14 not reflect two-electron mixed valency in their coordination chemistry and reactivity, and rather result from the convenience of electron counting. The two-electron dative bond motif is useful in the interpretation of other binuclear metal compounds. Puddephatt.19 has employed the formalism for the chemistry of binuclear Pt compounds. Reversible addition of carbon monoxide to [HPt(u-PP)2PtCO]+, PP = EtzPCHzPEtz or thPCHzPth generates a PtOPtII product. It should be noted, however, that these compounds could not be isolated due to the reversible nature of CO binding. The only evidence in claiming the formation of mixed valence species comes from multinuclear N MR data. Cowie20 has described [MM’(CO)3(dppm)2] (M. M’ = Rh, Ir) complexes as two electron mixed valence species instead of zero valent complexes. The argument behind considering these mixed valence species is made from coordination geometry. For these complexes, one metal has a coordination number of four while they other metal is five coordinate. The author suggests that instead of viewing these species as formally zero valent metal atoms, as would be implied by a symmetrical formulation, a more appropriate view is that of a mixed-valence MllM‘I complex in which the M'I atom functions as a pseudohalide bonded to the other metal via a dative bond. There are many other examples of “two- electron mixed valence” compounds along these lines. More generally, the electron counts of almost all binuclear metal-metal complexes possessing a metal- metal bond are two-electron mixed valence if a dative bond is postulated. The establishment of two-electron mixed valency by coordination geometry is less prevalent. The structure of H4Rh2{P[N(CH3)2]3]4, reported by Muetterties and coworkers21 in 1982, shows a formal d8 RhI center with square- planar coordination geometry singly-bonded to a d6 RhIII center with octahedral geometry. Similarly Lewis22 and coworkers have reported the crystal structure of an IrL—IrIII complex with bridging hydride and chloride ligands. The ORTEP 15 showing the molecular structure of [(114-C3H12)Ir(tr-H)(u-Cl)-IrH2(PPh3)2] establishes an IrI metal center with square-planar coordination geometry bonded to an 11“11 center with a pseudo-octahedral environment. In each of these examples, the oxidation states of the metals is confirmed through the coordination geometry of the molecules. The unequivocal establishment of two-electron mixed valency by coordination geomeu'y and oxidation state is less prevalent. In addition to the dirhodium fluorophosphine compounds, Walton has reported an (EtO)2C12Re —4 — ReC12(PPh3)2, which is an unsymmetric dirhenium complex with chelating dppe ligands, two bridging hydrides, and two terminal hydrides.23 These compounds are unique compared to other two-electron mixed valence dinuclear complexes because they do not possess bridging atoms, which may complicate the electron counting schemes to assign the formal oxidation states for the metals involved. These other systems assume that the single ligand bonded with both metal atoms affect each metal in the same electron counting manner. Because of the asymmetric structure, this may or may not be a valid assumption. The Rh2[CH3N(PF2)2]3C12(PF3) complex, on the other hand, has no bridging atoms to both rhodiums thereby clearly establishing it as a mixed valence species by all standards discussed here. C. Choice of Ligand System for Two-Electron Mixed Valence Compounds In order to generalize the chemistry of two-electron mixed valence species, and explore their reactivity, a ligand capable of stabilizing metal cores of diverse redox chemistry is critical. In preparation for the synthesis of other mixed valence 16 metal systems, the role played by the fluorophosphine ligand in stabilizing the two-electron mixed-valence dirhodium core must be understood. The fluorOphosphine ligand acts as a good n—acceptor and a moderate a— donor”29 which enables it to stabilize metals in both low and high oxidation states.30 As illustrated in the schematic represented by l for PF3, the lone pair electrons on phosphorus can be donated to a metal 0' orbital. This o—donor ability of the fluorophosphine acts to stabilize metals in high oxidation states. Conversely, the n—accepting properties of the fluorophosphines acts to stabilize metals in low oxidation states. The n—accepting ability of the ligand can be divided and understood as the combination of two bonding frameworks. First, the low-lying 3d orbitals of phosphorus can accept electron density form the metal (11: orbitals through a conventional n—backbonding argument as illustrated in by 2. 17 Second, Marynick has proposed that the orbitals of 0'" symmetry with respect to the P-F bond axis can also accept electron density from the metal.25 This accepting orbital is formed from the P-F o“ bonding interaction shown in 3. The enhanced n-accepting ability of PF3 over other phosphines such as PH3 or PR3 (R: alkyl) can also be rationalized due to an inductive effect. First, the high electronegativity of fluorine vs. hydrogen (or alkyl) causes highly polar bonds that characteristically have low-lying 6‘ accepting orbitals. These low lying orbitals improve the electron withdrawing ability of the ligand. Second, because the o P—F bond is highly polar toward fluorine, the 0‘ orbital will necessarily be highly polar toward phosphorus and provide increased overlap with the metal drt orbitals. The demonstrated versatility of the bis(difluorophosphino)alkyl/arylamine ligand, RN(PF2)2, (R = methyl, phenyl) in stabilizing a variety of metal oxidation states was demonstrated by King and coworkers.31 One example of the ligand’s broad redox ability is illustrated by the dicobalt complex, C02[CH3N(PF2)2]3(CO)2. As shown in Figure 4, the Co0Co0 dimer reacts with an excess of bromine to give the C02[CH3N(PF2)2]3(Br)432 derivative, Br2ConConBr2, and upon electrochemical reduction yields the CO‘ICO‘I anion 18 Figure 4. RCdOX 011611118in 0f C02[CH3N(PF2)2]3(CO)2. 19 oo c|:c§co oc—co—co'i—co C a 0 8 8 T”’ N sz/ \PF2 ' h i i OC-Cp—Cq—CO 29. ZBIZ P P m 31' -l Ll L’s". If OC-jh—fq-CO ‘ i Bt—Co-g-Co—Br P T3 P 9P Bl’lp PEI: 18 Figure 4. Redox chemistry of C02[CH3N(PF2)2]3(CO)2. 19 oo clsc§co oc—Co—co'L-co / C e 0 8 8 T"’ N F2P/ \PF2 V . T i OC-Cp—Cq—CO 29. 281'2 P P P p P 81‘ - I-. Is: Ls 00—00702—00 < > Br—Co-g-Co—Br : a, r P To P P Bl‘lp Pl!» Figure 4 20 C02[CH3N(PF2)2]3(CO)22‘.33 Thus a six electron change in the dicobalt core is accommodated by the ligand. The special properties of the ligand appear to derived from its electronic properties. Even though the fluorophosphine ligand itself is symmetrical, upon coordination to the Rh; core, the ligand has the versatility to become asymmetric due to the lone pair of electrons on the bridging nitrogen heteroatom. We have speculated that the reason behind this ligand’s ability to stabilize mixed valence complexes is due to its sp2 hybridized bridging nitrogen heteroatom and very 1t- acidic fluorophosphines. The nitrogen heteroatom has a lone pair of electrons that are available to donate electron density towards the very electronegative fluorophosphine moieties ligated to the metal core. After careful examination of bond lengths from crystal structure data, the stabilizing of a mixed valence species appears to arise from the asymmetrical donation of electron density from the lone electron pair nitrogen to one PF2 moiety preferentially. This is observed in alternating long and short phosphorus-nitrogen bond distances. The average difference in P—N bond lengths is 0.04 A for a single bidentate fluorophosphine ligand in the crystal structure of Rh2[PF2N(CH3)PF2]C12. The PF2 moiety that 1: backbonds to the lone pair of nitrogen will be less electron deficient, have shorter P—N bonds, and therefore act as weaker n-acceptor on the RhII metal center to which it’s coordinated. This weaker rt-acceptor ability will serve to stabilize a metal’s higher oxidation state. The PF2 moiety that does not interact with the N lone pair will be more electron deficient and it consequently has longer P—N bonds. This mechanism engenders more electron withdrawing ability for PF2 coordinated Rho, which is stabilized in the low oxidation state. More practically, the fluorophosphine ligand is of interest to the development of multielectron photochemistry in addition to stabilizing a range of 21 oxidation states. The bidentate ligand can stabilize the metal-metal core toward photochemical degradation. Most singly bonded bimetallic cores either lose a ligand or break a metal-metal bond upon photolysis. An example of this photodegradation is the photocleavage of Mn2(CO)10 which has a single bond between metal atoms.34‘37 Irradiation of the lowest-energy transition in this dimer promotes an electron from a a bonding orbital to an antibonding metal-metal 6* orbital, thereby resulting in the cleavage of the metal-metal bond to yield a ~Mn(CO)5 radical fragment. Because this dissociative pathway is very efficient, the excited state lifetimes of metal-metal bonded dimers are typically very short. The bridging fluorophosphine suppresses bimetallic dissociation and gives rise to long-lived excited states whose chemistry may therefore be exploited. The subsequent Chapters of Part One explore the reaction chemistry of the bis(difluorophosphine)aminophosphine ligand with Groups 9 and 10 metals with emphasis on defining the guidelines for the design of two-electron mixed valence bimetallic compounds. Chapter Two presents the results of the attempted synthesis of two-electron mixed valence systems for cobalt, iridium and palladium. The identification of the products is accomplished by use of NMR, mass spectrometry, UV/Vrs and X-ray crystal structure analysis. An experimental section describing reaction conditions and procedures is presented in Chapter Six Part I. Chapter Two Late Metal Two-Electron Mixed Valence Compounds A. Strategies for the Synthesis of New Two-Electron Mixed Valence Systems of Cobalt Our synthetic program directed towards exploring the chemistry of M“— M“+1 compounds centered on homologous series of compounds using the fluorophosphine RN(PF2)2 (R = Me, Ph) ligation sphere. We began our studies by seeking to stabilize the two-electron mixed-valence bimetallic cores of other mixed valence systems of Group 9 metals. In designing a two-electron mixed valence binuclear cobalt complex, the C02[CH3N(PF2)2]3(CO)2 system seemed to be an ideal starting material. Our first strategy to enter into mixed valency in this dicobalt system is summarized below, where the controlled oxidative addition to the symmetrical metal-metal dimer, (3020.0 yields a mixed valence system, Cozn-O. 22 23 Accordingly, the C020,0 core complex, C02(CH3N(PF2)3(CO)2, was reacted with one equivalent of bromine. To our surprise, two compounds were isolated from the reaction which were analyzed using FT—IR, 1H NMR, UVNis and FAB/MS. One of the isolated compounds shows an intense absorption at 332 nm, which is energetically coincident with the intense o —) 0* absorption of C02(CH3N(PF2)2)3(CO)2. A 1H NMR of the isolated compound shows a peak at 2.92 ppm, also matching the 1H NMR shift of the starting material. Finally, infrared analysis for the unknown compound shows the two carbonyl stretches at 2027 cm‘1 and 2014 cm’1 found in C02(CH3N(PF2)2)3(CO)2. On the basis of these spectroscopic results we conclude that this recovered product was the C02(CH3N(PF2)2)3(CO)2 starting material. The second product is assigned by using mass spectral analysis. Figure 5 shows the isotope intensity pattern arising from compounds containing varying numbers of bromine atoms. Comparison of this figure to the mass spectra of the isolated compound reveals the correct number of bromine atoms present in the fragment. Figure 6 is the mass spectrum of the second unknown compound in the mass / charge [M/Z] range 600 - 1000 amu. The peak at 858 amu corresponds to the C02(CH3N(PF2)2)3(Br)3+ fragment; peaks at 779 amu and 700 amu, respectively indicate loss of one and two bromines. This mass spectrum was compared to the known compound, C02(CH3N(PF2)2)3(Br)4, which was prepared by reacting C02(CH3N(PF2)2)3(CO)2 with an excess of bromine according to literature methods.32 A mass spectrum of C02(CH3N(PF2)2)3(Br)4 revealed identical peaks 858, 779 and 700 amu. A molecular ion peak is not observed for this product which is not unexpected because a positive ion is generated by the loss of a bromine anion to create the observed peak at 858 amu. Thus, the other product from the oxidative addition of 1 eq of Br2 to the C020,0 dimer is assigned to C02(CH3N(PF2)2)3(Br)4, 2. In summary, reaction of C020,0 with one equivalent 24 Figure 5. Isotope peak intensity pattern for ions containing the indicated number of bromine atoms. 25 Number of Bromine Atoms Figure 5 26 Figure 6. FAB mass spectrum of Coz(CH3N(PF2)2)3(Br)4, 2. I 27 N\: coon o3”, e 2:»...— mhh 28 of Br2 does not produce the corresponding two-electron mixed valence complex, C0211"), or even a C021,I dimer but instead results in half the starting material being converted to a C0211.II dimer. The C02(CH3N(PF2)2)3(CO)2 reactivity with one equivalent of bromine is summarized in Figure 7. The oxidation reactions of C020,0 with one equivalent of other halogens were also investigated to synthesize complexes 1 and 3. The reaction of C02(CH3N(PF2)2)3(CO)2 with iodobenzenedichloride (a chlorine source) did not show the characteristic 1H NMR signal in the 0 - 10 ppm range for the methyl . group of the fluorophosphine ligand. The absence of this signal is indicative of the loss of bridging phosphine to yield a paramagnetic cobalt monomer, consistent with work described below showing that the methyl group signal is lost in the presence of paramagnetic cobalt monomers. Since, chlorine proved to be too strong an oxidizing agent for the oxidation of C02(CH3N(PF2)2)3(CO)2 to a two-electron mixed valence species, the reactions of C02(CH3N(PF2)2)3(CO)2 were performed in the presence of the more mildly oxidizing iodine. The results however, were the same as the oxidations with bromine. Two compounds were recovered as the result of disproportionation into C020,0 and (30211.11 dimers. No detectable mixed valence products were found. The second strategy to generate mixed valence cobalt complexes was to disproportionate CoI monomers with the bidentate fluorophosphine ligand to generate MIN-Mn+1 systems. The monomers, (PPh3)3CoIX (X = Cl, Br, I)38 are useful starting materials because the bulky triphenylphosphines may be easily displaced by excess of fluorophosphine ligand. By bridging the monomer units and relying on the properties of the fluorophosphine ligand, the disproportionation reaction of M“—Mn systems would result in the formation of mixed valence binuclear complexes, Mn'l—M‘h‘l. 29 Figure 7. Product distribution from reaction of cobalt dimer with molecular bromine. 30 Bra, 1 eq P P P P I psi?“ .s‘Br 1/2 OC—Co Co—CO + 1/2 Br—Co 2‘ Clo—Br P P P P 3' p P ‘p Figure 7 31 i I I I \\X FZPNPFz .P‘ /CO., > PhaP—Cq Co—X (4) Phap l ’9PPh3 X=C|,Br,l I '9 Pl l PhaP w These reactions however resulted in products which showed no 1H NMR signals in the range of 1-10 ppm. These products 4, 5, 6 are consistent with cobalt monomers as described below. A third approach to two electron mixed valency cobalt complexes was attempted through the oxidation of unsymmetrical bimetallic cobalt complexes where one carbonyl has been replaced, with a phosphine. By synthesizing asymmetrical cobalt dinuclear species the reactivity of the molecule may be altered predisposing one side of the molecule towards oxidation over the other. Accordingly, the C020-0 dimer was photolyzed in the presence of excess phosphine to yield asymmetrical phosphine substituted C02(CH3N(PF2)2)3(CO)(PR3) dinuclear complex.39 P P P P I I hv l i OC—Co 09—00 T OC—Co Co—PRa (5) 11““ PPPP P 32 Oxidation reactions of asymmetric C020!0 complexes were investigated with mild oxidants such as CC14 or HCBr3 with the goal of effecting the following. m m P T P P x l och, CBr3H I .~“ Co—PF'ta > oc—Cq Co—Y (6) l | p pPP ‘ \\\ These oxidation reactions however did not result in the formation of two-electron mixed valence species or even cobalt dimers. The oxidation reaction products of asymmetric C020,0 complexes mixed with C04 or HCBr3 were characterized with 1H NMR and X-ray crystallography. The 1H NMR spectra of products showed no signals in the range of 0-10 ppm which indicated the loss of bridging fluorophosphine ligand and formation of cobalt monomers. The formation of monomers was confirmed using X-ray crystallography; a representative structure of (PPh)2CoI4‘2, 6 is shown in Figure 8. From these studies, the following reactivity trend was generalized. 33 Figure 8. An ORTEP view of (PPh3)2CoI4“2, 6. Thermal ellipsoids are at the 25% probability level. 34 a 95»...— 35 P P I I CCI4, CBraH " 2* oc——lco [Op—PR3 > (PR3)200X4 (7) P P P P B. Strategies for the Synthesis of New Two-Electron Mixed Valence Systems of Iridium The strategy employed to synthesize a two-electron mixed valence species was to disproportionate the symmetrical In” system with bridging fluorophosphine barring the problem of oligomerization. This synthetic strategy requires that the iridium dimers have ligands that can easily be displaced by fluorophosphine ligand, r. m CH3 P P l lax Ill/1,". “(\X [1,, “(Dip P2PN PF2 / \ . C "l "Ir” = F P —Ic--'r' X (3) f 2 '9’ "/ [\X N —COD 2 w X = SPh, CI R = Ph, Me 36 Suitable starting materials for this chemistry are the In” complexes, [(114' (38H12)1r(ll-Cl)21r(1l4-CaH12)]40 and [(114-C8H12)1r(lJc-SP11)21r(114-C3H12)].41 These IrglvI dimers are ideal candidates for the synthesis of two-electron mixed valence species for two reasons. First, the weakly bound olefin ligands should be readily displaced during reaction from the bimetallic core yielding a diiridium fluorophosphine complex. Second, the disproportionation of metal oxidation states would yield complexes according to the established precedent for the dirhodium system, Rh2[(CH3N(PF2)]3C12. Ir2(u-Cl)2(cis-cyclooctene)4 was reacted with an excess of CH3N(PF2)2. The reaction product was characterized using Fast Atom Bombardment Mass Spectroscopy as shown in Figure 9. A molecular ion peak is observed at 1124 amu. The peak at 1089 amu corresponds to loss of one chlorine ligand firm the molecular parent. The peak at 957 amu corresponds to loss of one fluorophosphine ligand. The peak at 922 amu corresponds to loss of one fluorophosphine and one chlorine ligand. The peak at 790 amu corresponds to loss of two fluorophosphine ligands from the parent molecular ion. These mass spectra results indicate the formation of Ir2[u-CH3N(PF2)2]3[ln-CH3N(PF2)2](CI)2, 7 with a dangling fluorophosphine ligand as shown below in Figure 10. 37 Figure 9. FAB mass spectrum of Ir2[p.-CH3N(PF2)2]3[ln-CH3N(PF2)2](C1)2, 7. 38 a 9...»:— Nxz OONH Dead oooH oom , coo omh emu” wmo~ Nam mood «NOH CH3 F P P 9. I I 0| III.s N—P—lr——II—Cl / I l 'a ll FZP F w ‘7‘”3 N Figure 10 This coordination geometry has precedence in the chemistry of the fluorophosphine ligand with iridium. A binuclear rhodium two-electron mixed valence species with two bromines has been characterized by x-ray crystallography.42 The compound exhibits a octahedral geometry about one rhodium and a trigonal bipyramidal geometry about the second rhodium. A bidentate fluorophosphine ligand has replaced on terminal PF3 ligand and leaves one difluorophosphine uncoordinated. Another entry to the above compound comes from the Irz(u-SPh)2(CO)4 starting material. Its reaction with excess CH3N(PF2)2 yields Ir2[|.t- CH3N(PF2)2]3[ln-CH3N(PF2)2](SPh)2, 8 as characterized by FAB/MS in the mass/charge [M/Z] range of 800-1400 amu. A molecular ion peak is observed at 1270 amu followed by a peak at 1161 amu corresponds to loss of one thiophenol ligand. The peak at 1103 corresponds to subsequent loss of one fluorophosphine 4O ligand. The peak at 994 amu corresponds to loss of one fluorophosphine and one thiophenol ligand. These results establish that In“,0 dimers can be obtained as described by reaction (8) irrespective of the terminal anionic ligand. On this basis we feel that the chemistry of dirhodium and diiridium cores with the fluorophosphine is parallel and general for two electron mixed-valence complexes. C. Strategies for the Synthesis of New Two-Electron Mixed Valence Systems of Platinum The preparation of M“'1-M“"1 of Group 10 metals was not as straightforward as Group 9 metals. We relied on the bidentate fluorophosphine ligand to assemble M0 and Mn centers while maintaining the two-electron mixed- valency owing to the ligands aforementioned electronic properties. The complexes szoodba343’44 (dba = PhCHCHC(O)CHCHPh), a source of Pdo, and the well known (PhCN)2PdC1245 complex which has substitutionally labile nitrile ligands were reacted with fluorophosphine. l P P 0' o " I=2PNPI=2 I Is sz -dba3 + 2(PhCN)2Pd CI2 > 2 qu<——-Pd—Cl (9) P ’P P p w 41 The fluorophosphine, CH3N(PF2)2, serves the purpose of a bridge to link the metals of differing oxidation state together and has the demonstrated electronic ability to accommodate and stabilize mixed valency complexes. sz-dba3 was reacted with (PhCN)PdC12 in a 2:1 mole ratio. The reaction products of sz-dba3, (PhCN)PdC12 and CH3N(PF2)2 were characterized using Fast Atom Bombardment Mass Spectroscopy. The mass spectra of the product clearly shows a collection of peaks centered at 1260 [Mill indicating the formation of a Pd4[CH3N(PF2)2]1o, 10 cluster. The formulation of these peaks at 1260 [Mil] representing a Pd4[CH3N(PF2)2]10 cluster can be independently confirmed by computer simulation of Pd4[CH3N(PF2)2]10. The isotopic distribution of palladium shown by the computer simulation exactly matches the product peaks 1260 [ME] which - further verifies the product as Pd4[CH3N(PF2)2]10. There is a precedent for the formation of tetrameric palladium clusters reported in the literature with other bridging phosphines. Pd4(PMePh2)4(CO)5 has been obtained when Pd(PMePh2)2(NO)2 is reacted with carbon monoxide.“ Moiseev and co-workers have also noted that palladium acetate reacts with carbon monoxide in acetic acid to produce the tetranuclear cluster, [Pd4(CO)(OAc)4]~2AcOH.“7 Thus the reaction of sz-dba3, (PhCN)PdC12 and CH3N(PF2)2 does not result in the predicted formation of mixed valence binuclear species but rather Pd tetramers. Chapter Three Early Metal Interactions with Boryls A. Reasons For Studying Metal-Boron Interactions The chemistry of molecules with simple bonds between transition metals and boron (e.g. M—BRz) is underdeveloped when compared to the known chemistry of metal-carbon, metal-nitrogen and metal-oxygen bonds. Examples of molecules with simple metal-boron bonds exist, but the bulk of the known chemistry concerns metallocarborane species which generally serve as spectator ligands.48 Very few early transition metal boryl species exist, and of the well characterized systems, most are late transition metal boryl complexes bearing phosphite co-ligands. Borane analogs to Grignard reagents have long been sought and complexes of this type which stabilize a BRg' fragment are of synthetic interest. Also, well documented examples of terminal borylidene complexes have not been reported, although (CO)4FeBN(CH2)2 has been claimed on the basis of IR and NMR data.49 The lack of a class of well characterized metal-borylidene complexes is curious because their carbene analogs have been known for some time. The development of metal-borylidene and boryl complexes is of interest to scientists for fundamental as well as practical reasons. Fundamental interests in transitional metal complexes with metal boron bonds arise simply because few compounds exist and little is known of their reactivity. One particular point of 42 43 interest in understanding metal-boron interactions is what affect will the empty p- orbital of boron have on the stability and reactivity of these compounds. One possibility is the empty orbital could engage in n-backbonding which could stabilize the complex while still allowing an active site for nucleophilic attack on the ligand. The practical impetus for studying metal-boron interactions stems from recent reports of metal-mediated hydroboration. B. Metal-Boron vs. Metal-Carbon Bonds In the field of organometallic chemistry, there is a tremendous knowledge of structure and reactivity of metal carbon bonds.50 An important question in developing and understanding metal-boron interactions concerns how reactivity of the metal-boron linkage will compare with the known chemistry observed for metal alkyl and carbene complexes. An analogy can be drawn, as shown in Figure 11, between metal-boron and metal-carbon bonds since boryl and borylidene ligands are isolobal with carbene and carbyne radical cations. Figure 11 44 There are two fundamental differences between boron and carbon atoms bonded to transition metals. The first difference is in electronegativity; boron has an electronegativity of 2.0 whereas carbon has an electronegativity of 2.5 on the Pauling Scale. A second fundamental difference between the two atoms is the presence of an empty p-orbital on boron. This empty p-orbital, in some cases, may overlap with the occupied frontier orbital of metal fragments creating backbonding to the boryl ligand. This n-backbonding between the empty boron-based p—orbital and metal based electrons may have observable structural manifestations in metal- boryl complexes. The use of bent metallocene fragments in the development of metal boryls can provide insight into the nature of metal-boron interactions. Since the frontier orbitals for an idealized C2 bent metallocene fragment have been well defined,51 this information can be used as a tool for rationalizing solid-state metal- boron interactions. An illustration of n-backbonding created by the interaction of frontier orbitals of a bent metallocene with the empty p-orbital of a boron is shown in Figure 12. C. Metal Boryl Complexes In spite of the interest in metal-boron complexes, very little information regarding their structure and chemistry has ever been published. The first example of a metal-boron compound having a coordinate bond was reported by Grim in 1961.52 This compound was prepared by reaction of a lithium triphenylgerrnanate with triphenyl boron. LiGe(C6H5)3 + B(C5H5)3 LI [(C6H5)3GC-B (C6H5)3] ( 10) 45 Figure 12. Frontier orbitals of a bent metallocene interacting with a bent metallocene to create n-backbondin g. 46 Figure 12 47 A few years later, Parshall reported the borane adduct of the rhenium pentacarbonyl anion, the first transition metal-boron compound.53 Shortly following this report, Noth and Schmid prepared several metal boryls by the reaction of haloboranes (R2BX) with neutral and anionic transition metal complexes.54'58 All of the transition metals with metal-boron interactions are Group Six or later with the exception of szTi(Cl)BPh2. This last complex was prepared via oxidative addition of thBCl to "szTi". Considering the confusion surrounding early reports of titanocene,59 the formulation of this last titanium boryl complex is likely incorrect. The development of metal boryls is of interest for more than just fundamental reasons. Recent interest in the chemistry of simple metal-boron linkages‘mv‘s‘)’61 has been fueled by the successful metal-mediated functionalization of organic molecules by boron reagents.”65 An important step in the development of metal-boryls occurred in 1975 when Wilkinson's catalyst was found to oxidatively add catecholborane to form a rhodium boryl.66 Within the last few years, Merola, Baker, Marder and Hartwig have all synthesized and published new structures of metal-boryl complexes for the middle to late transition elements.‘5°“'523"67'69 There is now a renewed interest in the synthesis, characterization and reactivity of transition metal boryl complexes with special emphasis on complexes which could play a role in hydroboration chemistry. Surprisingly, little attention has been paid to early transition metal boryls whose reactivity is potentially quite different from later metals. Because early transition metals are quite electropositive, the polarity of a metal-boron bond may be reversed with respect to late metal boryls. Early transition metal centers are therefore ideally suited for stabilizing boryl anions (e.g. ’BR2) fragments. The synthesis and generalization of this class of compounds would be a significant achievement, since the fundamental regioselectivity of boryl transfer could 48 complement that observed in late metal hydroboration chemistry as illustrated in Figure 13. 8* ' 6+ 5- R2? —0. R20 0 56‘ :5‘“ 55‘ 55+ R2 B—M M—B R2 Early Metal Boryl Late Metal Boryl Figure 13 D. Metal Mediated Hydroboration Chemistry Hydroboration of unsaturated organic molecules has been an active area of research for the last 38 years. In 1956, the first additions of diborane to alkenes and alkynes were performed where the boron-hydrogen bond adds to a carbon- carbon multiple bond to form an organoborane with formal carbon-carbon bond reduction.70 Thus, olefins and alkynes can be transformed into their corresponding alkyl borane or vinylborane. —B / \ (12) _A_ c=c + H—B< ——> H—ti[ —CE — + H—B (13) I —U B /\ 49 Typically, hydroboration of aliphatic alkenes follows anti-Markovnikov addition of borane with good regioselectivity. As shown by the work of Brown,71 organoboranes are attractive from a synthetic point of view. Hydroboration is a powerful tool for the organic chemist because the boron substituent can be transformed into alcohols, ketones, amines or alkane products. Based on this synthetic versatility, chemists have been interested in controlling regio, stereo, and chemoselectivity of borane addition to organic molecules. The use of organometallic compounds can greatly contribute to the advancement of hydroboration chemistry since transition metal complexes can dramatically accelerate reaction rates, alter reaction mechanisms, and potentially change the chemo, regio, stereoselective addition of boranes to substrate molecules. The information learned by studying reaction mechanisms and gaining a mechanistic understanding of the reactivity, is key to controlling chemo, regio, stereoselectivity. A In 1975, Kono and Ito had structurally characterized the boryl product from oxidative addition of 4,4,6-trimethyl-l,3,2-dioxaborinane (TMDB) to Wilkinson's catalyst. The discovery that B-H bonds from TMDB or 1,3,2-benzodioxaborole (HBcat or catecholborane) undergo oxidative addition to yield RhIII boryl products formed the basis for metal catalyzed hydroboration.66 Me Me 0 \B—H 0/ Me T PPh Rh Ph (:1 not 3 or (P 3’3 > (RO)ZB—Rh‘ (14) I PPh3 O/ 50 In spite of the wide variety of rhodium based catalysts, little was done with this knowledge until 1985 when Mannig and Noth reported the first examples of rhodium-catalyzed olefin hydroboration.72 Catecholborane usually reacts very slowly with alkenes under ambient conditions, but Mannig and Noth found that small quantities of transition metal complexes can greatly enhance the rate of hydroboration. Metals can play a more significant role than just accelerating the rates of reaction, they can also alter the diastereoselectivity73 and regioselectivity73a’74 of hydroborations. With regard to diastereoselectivity, Evans has shown that rhodium-catalyzed hydroborations of allylic alcohols yield syn diastereomeric products.63a Whereas, uncatalyzed hydroboration reactions of allylic alcohol derivatives generally afford the anti isomer with 9-BBN providing the highest selectivity.75 The difference in selectivity of these two examples is a significant development because it demonstrates that rhodium catalyzed hydroboration of prochiral alkenes affords complementary diastereoselectivity. OR" 933” ’ HO/\é/'\R' (15) Fat OR" anti diastereomer \ R' R \ on" Flh', Catecholborane> HO 3' (16) R syn diastereomer 51 In regard to regioselectivity, catalysts can also serve to complement the products which are observed using conventional hydroboration reagents such as 9-BBN. Evans has shown that in the hydroboration of cyclic 1,2-disubstituted allylic alcohol derivatives, the regioselectivity of the catalyzed reaction with I-lBCat is opposite the uncatalyzed reaction with 9-BBN.73a on 13R2 Uncatalyzed > “I (17) l RZBH OR OR I Rh“) > : (RO)ZBH In contrast to the 9-BBN hydroborations, which appear to be dominated by (18) ” 13(0Ft)2 electronic effects favoring the anti 1,2 diols, the catalyzed reaction with catecholborane favors the anti 1,3 diols. The regioselectivity for rhodium catalyzed hydroboration is typically anti-Markovnikov addition of the B—H bond. One notable exception to this general rule is the catalyzed addition of boranes to styrenes and styrene derivatives.73a For these molecules, the borane is typically placed at the tit-position to the aromatic ring. 52 Catalyst B cat Metals can also affect the chemoselectivity of hydroboration to multifunctional organic substrates. By employing transition metals as catalysts, the hydroboration of carbon-carbon double bonds can occur even in the presence of more reactive functionalities such as ketones. This influence on chemoselectivity was demonstrated by Mannig, and Noth, when they reacted 5-hexen-2-one with catecholborane in the presence of less than 1 mol % of RhCl(PPh3)3.72 As shown below, the presence of a small amount of RhI catalyst alters the site of reactivity of the substrate and changes the final products. To gain further insight into metal catalyzed hydroboration, several mechanistic studies have been undertaken to determine reaction pathway(s) for 53 late transition metal complexesf’lt'm’77 The proposed mechanism for most rhodium catalyzed hydroborations follows an analogous sequence of events used to describe rhodium catalyzed hydrogenation of olefins.78 This mechanism for metal catalyzed hydroborations is diagrammed in Figure 14. First a borane oxidatively adds to a coordinatively unsaturated metal center to give a RhIII boryl hydride. This boryl hydride coordinates an olefin which then undergoes subsequent migratory insertion into the metal hydride bond to give an alkyl intermediate. The last step is reductive elimination of the alkyl and boryl ligands to yield the alkylborane product with regeneration of the active catalyst. The pathway displayed above is not the only possible pathway for late metal catalyzed hydroboration. A competing mechanism involves alkene insertion into a metal-boron bond (instead of a metal-hydride bond) forming a metal-boryl intermediate which is followed by B—H reductive elimination. Olefin insertion into a metal-boron bond is a precedented reaction step observed in the reaction of 4-vinylanisole with the bis(boryl) complex, [(PPh3)2RhCl(Bcat)2].5°d Isotopic labeling studies have also been used to address the mechanism of metal-mediated hydroboration. Two independent labeling studiesm‘,79 of the hydroboration of 2-methyl-3-[(tert-butyldimethylsilyl)oxy]but-1-ene with catecholborane-d1 mediated by Wilkinson's catalyst have been reported but the results differ dramatically from each other. In the first study, Evans found significant deuterium label (17%) on the terminal carbon when using deuterocatecholborane. 54 Figure 14. Proposed catalytic cycle for rhodium mediated hydroboration of olefins. 55 RhLacl I-.. \ H C’s/V RhLZCI L = PPh3 H o Kl?“ \ .\\‘\ B-RH Cl +L Figure 14 J)» B—H 56 OTBS 0.2% RhCI(PPh3)3 Me 0 \3—0 Me @O/ H2O °/o D Distribution When Burgess investigated the same reaction, he claimed that nearly all the deuterium label was found exclusively at the second carbon (>99%) and the recovered starting material also showed some deuterium incorporation. Evans later showed that impurities in the catalyst were responsible for Burgess' observations. It is now known that if Wilkinson's catalyst is used, the regiochemistry depends on the quality of the complex.80 Old or commercial Cth(PPh3)3 may be contaminated with RhII impurities81 and contain peroxides82 which are not found in properly, freshly prepared catalyst. These impurities can dramatically alter the regiochemistry of reaction products and nullify any conclusions which can be drawn from mechanistic studies. Careful reinvestigation of this reaction under optimal conditions showed that several steps in this catalytic cycle are reversible and the actual mechanism of metal mediated hydroboration is more complicated than originally surmised. 57 Reaction studies of catecholborane mixed with RhCl(PPh3)3 indicate that over 50% of all rhodium atoms form HthCl(PPh3)3 not HRh(BCat)Cl(PPh3).83 PPh3 o\ Ph3P—RIh—CI + O/B—H > PPh3 H H B—Rh" + Rh’ + Other (23) / IV V I‘ PPh H PPh 0 CI 3 Cl 3 <50% >50% Extraneous hydridorhodium complexes formed during reaction provide a separate catalytic manifold for olefin hydrogenation and are capable of redistributing the deuterium label. Other difficulties in labeling studies include side products such as RhH(PPh3)3, catecholborane disproportionation compounds, and phosphine- borane adducts, formed when catecholborane reacts with free phosphine.84 The list of possible complications of rhodium catalyzed hydroboration do not stop here. The analysis of organic products also indicate complicated side reactions that result from B—elimination processes and isomerization ‘of the carbon-carbon double bond.“ Not all catalyzed hydroborations have used rhodium or iridium metals. There have also been reports of early metal catalytic hydroborations of alkenes mediated by BH4-lTiCl3 or BH4-le2TiC12.86‘88 But these early metal hydroboration systems have been criticized as likely being "metal promoted" rather than "metal catalyzed" reactions.“ It is known that titanocene dichloride and borohydride react according to the following equation:89 58 H gym 4? “(H/n,” / 211'.‘ + 4BH4' ——> 2n;H’B\ + Bsz + H2 (24) H In the reported "metal catalyzed" reactions for this titanium complex nearly 20 mol % of titanium is needed per one equivalent of borohydride. The catalyzed hydroborations of this system is likely to be incorrect and arise from adventitious diborane or from BH4- bonded to titanocene. While metal catalyzed hydroboration chemistry of early transition metal complexes is undeveloped, lanthanide complexes, Cp*2LnR (Ln = La, Sm; R = H, CH(SiMe3)2), (MezsiMe4C5)2SmR (R = CH(SiMe3)2) and Cp*2Sm(THF) have recently been reported as effective hydroboration catalysts.64 Deuterium labeling studies of the catalyzed reactions are fully consistent with a mechanism involving o-bond metathesis. The proposed catalytic mechanism is displayed in Figure 15. [Cp*2LnH]2 the active hydroboration catalyst, is formed by metathesis of Cp*2LnR with HBCat. Once formed, the active catalyst can insert an olefin into the metal hydride bond to generate a metal alkyl. This is followed by o—bond metathesis of H-BCat with a metal alkyl to regenerate the metal hydride with extrusion of the substituted borane. Deuterium labeling experiments in this system show that the reaction is well-behaved because there is no scrambling of the label. 59 Figure 15. Proposed catalytic cycle for lanthanum based hydroboration 6O 2Ln TMS + Q: ,B—H % ms 0 \ fl]; 0 hi 0 \\ P ' or “sf? GA 43?”. as. Figure 15 61 Our work has focused on both developing synthetic routes to boryl complexes of the early transition metals and exploring reactivity of metal olefin complexes with borane reagents. Specifically, Chapter Four describes our initial results involving the reactivity of the szTaHz‘ anion with B- Chlorocatecholborane. Chapter Five describes the reactivity of the titanocene ethylene complex, Cp"'2Ti(112—CH2=CH2)90 (Cp* 2 n5-C5Me5), with hydroboranes and the experimental conditions are given in Chapter Six Part II. Chapter Four Regioisomers of the First Tantalum Boryl Complexes A. Introduction Simple bonds between transition metals and boron (e.g. M—BRg) are underdeveloped when compared to the existing chemistry of metal carbon bonds. Whereas most efforts have been directed towards late transition metal systems, our interests have centered on the chemistry of the early transition metal compounds. Due to the electropositive nature of the metal, these systems would seem to be the best candidates for stabilizing a boryl anion. In late metal systems, transition metal boryls have been prepared by reacting haloboranes with metal anions“,60c Due to the success of this approach to late metal-boryl synthesis, we chose to investigate reactions of Group Five metal anions with haloboranes. B. Synthesis of the First Tantalum Boryl Complexes The complex szTaH3 was prepared according to literature methods and deprotonated with n-B uLi to form yellow-orange {CmTaHzLi } x metal anion that has been shown to react as a source of szTaHz".9l The anion was isolated and mixed with one equivalent of fi-chlorocatecholborane in toluene at -78 °C. Upon 62 63 warming the suspension, the color changed from yellow-orange to very pale yellow with formation of a white precipitate (presumably LiCl). The toluene solution was filtered, and upon solvent removal, an off-white microcrystalline solid was isolated. 1H and 11B NMR spectra (C6D5 solution) of the crude reaction mixture indicated formation of two products. The 1H NMR spectra showed two major cyclopentadienyl resonances, two distinct catecholate regions and multiple hydride resonances. Likewise, 11B NMR spectra exhibited two distinct boron resonances with chemical shifts at 70.0 ppm and 64.7 ppm. From this data it appeared that two isomers of szTaHzB02C6H4 l4 and 15 were formed during the reaction. Indeed, two products could be readily separated by fractional crystallization from toluene when solutions were slowly cooled to -80 0C. C. Results and Discussion 1H NMR data of the less soluble isomer, 14 revealed a single hydride resonance (8 = -4.23 ppm) that integrated as two protons. The chemical equivalence of the two protons suggested that this is the endo-isomer of Cp2TaH2(BCat) where the boryl ligand likely occupies a central position in the wedge of the bent metallocene framework. With the boryl ligand located in the center of the wedge, the two hydrides are chemically equivalent which accounts for the observed singlet in the proton NMR. The 1H NMR of the tantalum endo- isomer is shown in Figure 16. 11B NMR spectra for this endo-isomer shows a resonance at 70.0 ppm. The 1H NMR spectra of the second, more soluble isomer, 15 is more complicated. The hydride region exhibits a doublet (6 = -4.20 ppm, J an = 5.6 Hz) and a broad singlet (5 = -5.15 ppm) each integrating as one proton vs. the cyclopentadienyl resonances. To determine if these two hydride resonances were Figure 16. 1H NMR of the szTaH2(B02C6H4) endo isomer, l4. 65 3 9...»...— la a- 7 a- a- a- o- u u n o n o p .uiLlrthL sift; . .. . . ... s .....-rp.—I>Lerrt—l... . . _ . . . ... Pt. t..—t[-LL..—I?Llr o..— . prick»; TIPPL . .. _ g _ : -'.— 66 Figure 17. 1H NMR of the szTaH2(B02C6H4) exo isomer, 15. 67 2 2:3... II a- o- a- a- a- or u u n o a o p —. ._—I>.>pLs. _..;. . . . . ... ..,otrL....—L1rrrhi—Ll.r... _ . . . . _ ...-....._ . . . . _ . ... . _ . o .. —.. p. . . . . . . I . - a i lll'I-fi III". I." 68 coupled, the broad resonance at -5.15 ppm was selectively decoupled which led to the collapse of the hydride doublet at -4.20 ppm. Thus, coupling between the chemically inequivalent hydrides was confirmed. The only geometry which is consistent with the observed data is one where the boryl ligand likely occupies an exo position in the metallocene wedge. The 1H NMR of the tantalum exo-isomer is shown in Figure 17. The 11B NMR shows a resonance at 64.7 ppm for this isomer. The two readily separable isomers were each characterized crystallographically and the resulting molecular structures with relevant metrical parameters are shown in Figures 18 and 19. Despite their proximity to heavy tantalum nuclei, hydride positions could be located in difference Fourier maps and refined isotropically for both compounds. In the endo isomer the boryl fragment adopts a "locked" geometry where the catecholate oxygens and cyclopentadienyl ligands are effectively eclipsed when viewed down the Ta-B vector. Other intriguing features of the endo isomer structure which merit comment are the metal-boron and boron-hydride bond distances. Significantly, the Ta-B distance (2.263(6) A) is shorter than the Nb—B distance 2.411(5) A in CpngH2(BC3H14), which is almost certainly a true borohydride complex.92 Also, the B---H contact distances for the hydride ligands (1.76 and 2.12 A) are longer than those found in archetypal borohydride complexes (B—H = 1.10-1.20 A)93t94 and closely related niobium complexes szNbH2(BozC6H4).95 Furthermore, the H—Ta—H bond angle of 113 (3)° in the endo isomer is much closer to the Hexo-Ta-H'exo bond angle found in szTaH3, 126°,96 and larger than the corresponding H—Nb—H angles in szNbH2(B02C6H4) and szNbH2(BC3H14) which are 92(2)0 and 70(3)O respectively.92 This crystallographic data suggests that the endo isomer depicted in Figure 17 is best formulated as an authentic TaV-boryl species represented as A as opposed to a Tam-borohydride complex represented as B in Figure 20. 69 Figure 18. ORTEP drawing of endo—szTaH2(802C6H4), 14, showing 25% probability thermal ellipsoids. 70 2 as»... .0 Au 50 CI 3 u, . C m. :3. .8 a: D ._ O _I\c 71 Figure 19. ORTEP drawing of exo—szTaH2(BOzC6H4), 15, showing 25% probability thermal ellipsoids. 72 a— 2:»...— 73 s” /o yaw. /o a: \ D <—> Ta~ (B\ H O % H O A B Figure 20 In the exo isomer, the plane of the boryl fragment is again eclipsed with the cyclopentadienyl centroids. Despite the location of the hydride ligands in the difference map, the usual caveats concerning hydride positions in heavy atom structures and the large uncertainties in the Ta-H distances prevent any meaningful analysis of B---H contact distances in comparison with other compounds. The crystal structure data, however, can provide insight into the oxidation state of the tantalum metal. Significantly, the sum of the angles about boron, defined by the tantalum center and catecholate oxygens, is 360.0 (22)°, thereby establishing planar sp2 hybridization at boron. Thus, the exo isomer is therefore best described as a TaV-boryl complex since any contribution due to a limiting Tam-agostic borane complex should be reflected by a deviation from planarity at the boron center. Lastly, the metal boron distance of 2.296(11) A for the exo isomer is also significantly shorter than the Nb-B distance of szNbH2(B02C5H4) previously mentioned. 1H, 2H, 11B NMR have also been used to probe B—H interactions in the closely related endo complex, szNbH2(B02C6H4).92 Isotopic perturbation of equilibrium has previously been used to probe hydrogen complexes and agostic 74 interactions.97’99 If a rapid chemical equilibrium between two forms exists where hydride position is intimately involved, substitution by deuterium tends to shift the equilibrium. This is accompanied by a perturbation in chemical shift of the affected resonance(s). 1H and 2H NMR data indicate that a rapid equilibrium exists between terminal and bridging hydrogens in the complex, szNbH2(302C6H4). This indicates that szNbH2(BozC6H4) complex should not be viewed as a resonance hybrid of two possible structures, but as two distinct molecules. One molecule is a NbV boryl complex which is in rapid equilibrium with a second me complex which has bridging hydrides as shown in Figure 21. Figure 21 Electronic bases favoring the "locked" geometry of the formally d0 tantalum complexes are not immediately obvious. In both cases, however, there exist filled metal-hydride bonding orbitals of proper symmetry for n—backbonding to the empty p-orbital of boron. The appropriate metal-hydride orbitals capable of backbonding from the metal center are illustrated in Figure 22. The presence of n-backbonding from the metal center to the boryl ligand could account for the observed solid-state structures. X-ray data for the endo compound, in conjunction 75 Figure 22. Metal d-orbitals capable of rt-backbonding to the empty p-orbital of boron. 76 «a 2a»...— Q 77 with the NMR data argue against B---H interactions. Thus, the adopted geometry is not an accurate measure for boryl interactions. The formation of regioisomeric products in Group 5 metallocene chemistry is rare. To our knowledge, the endo and exo isomers, l4 and 15, are the first case of Group 5 regioisomers which have been cleanly separated and structurally characterized. There is a precedence, however, for the formation of endo and exo isomers in the addition of silanes to CpgTaH3.1°0 Precedence cases of regio isomer formation were first recognized in zirconocene systems. For example, the regiochemistry of CO addition to Cp'zerz complexes (Cp' = C5Me5, C5H4R; R = Me, H) is sensitive to variations in the ligand sphere. The complex (1] 5- C5Me5)22rH2 is known to add CO preferentially to the central lobe of it's la1 LUMO51 to form a product where the hydride ligands are chemically equivalent.101 Conversely in Figure 23, the exo acyl isomer forms as the kinetic product from the reaction of CO with Cp2Zr(CH3)2 and argues strongly for initial "side-on" attack by CO to the la orbital.102 78 f + CEO : 1" » Me Me \ C \\ 0 .&Me .‘\\Me 2? < VMe % Me % endo exo Thermodynamic Product Kinetic Product Figure 23 Thus, the electrophilic attack of the chloroborane can occur at two sites of the filled 1a1 orbital. Similar rationalizations for regioisomer formation can be made for the reaction of szTaHz" with ClBCat. The reaction is viewed as proceeding by initial attack of the filled 1a1 orbital of the lithium tantalum anion on the empty p—orbital of the chlorocatecholborane. On purely electronic grounds we might expect the exo isomer to be the favored one, while steric arguments suggest favored attack at the central site as shown in Figure 24. 79 751-05 35% Cp Cp 0 O \J Cng Me o .6 CPeO/H I \Me 9 3° CP ep’ \H I Ox 0 (CV \CI III 0 Figure 24 D. Conclusion This is the first case where Group Five regioisomers have been cleanly separated and structurally characterized. Furthermore, the endo and exo isomers represent the first unambiguous examples of Group Five boryl complexes. Significantly, the exo isomer appears to be the kinetic isomer and is gradually converted to the endo isomer on heating, which explains difficulties in preparing the endo isomer via thermal or photochemical reaction routes from szTaH3 and catecholborane. We are currently investigating the details of the isomerization of these two complexes and the reactivity of these species. The syntheses of other early transition metal boryl complexes is also currently being pursued. Chapter Five Hydroboration of Decamethyl Titanocene Ethylene Complex: A Dehydrogenative Boronation Reaction A. Introduction In the investigation of early transition metal mediated hydroboration, we decided to study the reactions of olefin complexes with boranes whose chemistry has been largely overlooked. The lime-green crystalline compound, Cp*2Ti(n2- CH2=CH2) was an obvious choice for these studies, since it is one of the simplest olefin complexes known, and is readily accessible from TiCl3(TI-IF)3 in two high- yield steps. TlCla-THFa + '8 Ti—Cl + Hzc=CH2 M» re. 2 Nan‘ ——> 80 i Ti— Cl (26) \ §\ Blue Crystals ’CHz TI— le} CH2 ‘2" , \ Green Crystals 81 Olefin complexes can serve as latent sources of electron-rich low-valent metal centers that can potentially serve as good precursors for synthesizing metal-boryl complexes.103 Cp*2Ti(n2-CH2=CH2) is a sixteen-electron metal center with a rich reaction chemistry. In accord with the Dewar-Chatt—Duncanson model for olefin binding, this complex exhibits reactivity consistent with a TiI V metallacyclopropane or Tial) olefin adductm‘itlo5 iii-“‘IHZ + 2HCI —-—> Ti” + ”2.0—CH3 (28) 2 . . 0 Based on these precedents, two modes of reactivity for Cp*2Ti(n2-CH2=CH2) with hydroboranes were envisioned. The olefin complex could react as a TiIV species to form the ring opened product, Cp*2Ti(H)(CH2CH2BR2), or oxidatively add the borane to form Cp*2Ti(H)(BR2) which would likely undergo insertion chemistry with the liberated olefin. B. Results and Discussion The complex Cp*2Ti(n2—CH2=CH2) reacts immediately with catecholborane in toluene at -78 °C to form a yellow solution. The reaction is then 82 allow“ to warm to room temperature and solvent is removed under reduced pteSSUre to leave a lemon-yellow solid. This yellow product was characterized by \h and 11B NMR spectroscopy. The 1H NMR spectra indicated clean formation of '3 diamagnetic product with inequivalent Cp* environments and three vinyl multiplets (characteristic of an unsymmetrical olefin coordinated to a metallocene center) and two aryl multiplets due to one catecholate group. 11B NMR shows a very broad resonance at 36 ppm, which in conjunction with 1H spectral data conclusively establishes the product as, Cp*2Ti(n2-CH2=C(H)(302C6H4)) 18, a Vinylboronate ester adduct of decamethyltitanocene. $2 CH «gm TI—r 2+ HBOZCGH4 W w“: Ii + *szTI (30) \ CH2 6 \ c; \ \ B\ l o A 1H NMR spectrum with of the Vinylboronate ester complex illustrating the coupling between the vinyl protons is shown in Figure 25. In order to investigate this reaction using NMR techniques, we required a solid borane to facilitate accurate weighing of small quantities in NMR tubes. When 2,3-dihydroxynapthalene is reacted with BH3-THF (Aldrich) a solid borane analogous to catecholborane 11, is formed with concomitant loss of dihydrogen. OH ‘ 0 BH -THF OH O 83 Figure 25. 1H NMR spectrum of Cp*2Ti(n2-CH2=C(H)(B02C6Ha)), 18. 84 3 ea...— Ian win or" 9.“ a.« Q." ..N ..N 0.0 «.0 v.» 0.0 0.0 Err—L 331., PIPELLLr—j _ i u L. 85 Th: Catecholborane analog, HB02C10H5, also reacts with Cp*2Ti(n2-CH2=CH2) ’0 yield a lemon-yellow solid, Cp*2Ti(n2-CH2=C(H)(802C10H6) 20 whose 1H M data were analogous to the catecholate derivative. When this reaction was mOrritored by NMR, ethane was also observed as a reaction side product. Cp*2Ti(n2-CH2=C(H)(B02C10H 5)) was further characterized by mass spectroscopy, but no molecular ion was observed. Instead, spectral data indicated fragment ion peaks for the Vinylboronate ester CH2C(H)(B02CloI-16)+ and Cp*2Ti+ at 196 and 318 m/z, respectively. This Vinylboronate ester is less soluble than the product derived from catecholborane and can be isolated more readily. The overall product yield is typically 40-45% and, significantly, never exceeds 50%. The formation of Cp*2Ti(n2—CH2_=C(H)(B02C10H5)) is significant for the following reasons. First, the stability of the complex is surprising because the coordination sphere of the Cp*2Ti fragment is crowded. In fact, inaccessibility of the propylene analog of Cp*2Ti(n2-CH2=CH2) has been attributed to unfavorable steric interactions.106 The second significant feature of this reaction is that boron substitution on the coordinated olefin has proceeded without carbon-carbon bond reduction! The overall transformation is consistent with B-H/C-H bond metathesis accompanied by formal loss of dihydrogen. Recent interest in vinyldialkylboranes has been stimulated by the fact that these molecules are excellent dienophiles in Diels-Alder reactions. The high reactivity of these potent dienophiles has been attributed the to the empty p—orbital on boron which presumably enhances the electron-deficiency of the dienophile. Interest in using boron substituted olefins comes from the understanding that a trivalent atom with an empty p-orbital should render the olefin electron deficient based on the resonance structures in Figure 26.107 Thus, these electron deficient and highly polarized dienophiles should exhibit enhanced reactivity and excellent 86 r . egloselectivity in Diels-Alder reactions. This has proven to be the case for alkyl su . bsutlned vinylboranes. Fi R Figure 26 One drawback to the utility of synthetic work on reactive dienophiles is the use of toxic tin reagents derived from vinyl Grignard reagents. Vinylboranes are synthesized by the reaction of haloboranes with a very toxic vinyltributyltin reagent shown below. /\ Br + M9 —’ /\MgBr ML /\SnBu3 (32) /\Sn3u3 + Br—B ——-> /—B (33) R 2= + “—3: D ; /\B\’R + Csz (34) R 12 Synthesis of vinylboranes from olefins and boranes is an attractive alternative to this transmetallation route. When the reaction in equation 30 is reduced to olefin and borane components, the significance of this transformation is clear: a transition metal can mediate dehydrogenative borylation of an olefin where the olefin siphons off the H2 as alkane. The discovered synthesis of Cp*2Ti(r|2- CH2=C(H)(B02C10H5) from Cp*2Ti(n2-CH2=CH2) is quite significant because it 87 0f - fem a new route to the synthesis of vinyldialkylboranes. Thrs has been observed in rhodium based systems; however, vinylboranes are only produced for styrene he, ~ “V auves (vide infra). C. Mechanistic Rationalization In an effort to understand the transformation of the ethylene ligand into a boronate ester, the reaction was monitored by NMR Spectroscopy. The reaction was run in a sealed NMR tube at low temperatures (-80 °C) to follow the course of the reaction and identify intermediates. Unfortunately the reaction is extremely rapid at low temperatures in toluene-d8 and no intermediates were observed. Since rapid reaction rates at —80 °C precluded classical kinetic analysis we chose to interrogate the mechanism with a series of isotopic labeling experiments and chemical tests. Inspite of the rapid reaction rate, useful information from NMR experiments were obtained and their spectra indicated formation of ethane along with the final Cp*2Ti(I-I)(n2—CH2=C(H)(B02C10H6)) product. A proposed mechanism that accounts for the formation of Cp*2Ti(n2- CH2=C(H)(B02C10H6)), CH3CH3, and mass balance observed for this chemical transformation is shown in Scheme 27. 88 Scheme 27. Proposed Reaction Mechanism of Cp*2Ti(n2-CH2=CH2) with HBOzCloHo- ' 89 um 95»...— P—rwoao t 7 2165480 foo i oIOpONOm—r. + w . O / / zlor Wyn a: we: 90 e believe that the borane initially attacks the ethylene ligand to induce a ring °Peneed TiIV intemcdiatc, Cp*2Ti(H)(CH2CH2B02C1()H6). This intermediate than tax“\‘uy undergoes ,B-H elimination to form an intermediate titanocene olefin Q“thride complex. CP*2Ti(112'CH2=C(H)(BOZC10H6)) results from H2 elimination from this intermediate, and the liberated H2 reacts with Cp*2Ti(n2- CH2=CH2) to generate ethane and "Cp*2Ti". D. Elucidation of Mechanism The simplest isotopic labeling experiment invites reaction of Cp*2Ti(n2- CH2=CH2) with D—BOzC1OI’I6. In the mechanism outlined in Scheme 27, it should be emphasized that if fi—H elimination is reversible, it could scramble deuterium label into the fi—position of the vinylborane ligand. Furthermore, the ability of Cp*2TiH2 to exchange hydrogen between the hydride positions and methyl sites of the pentamethylcyclopentadienyl ligand is well documented.90 This rapid hydrogen exchange could provide a pathway for deuterium scrambling into the ligands methyl groups. Thus, the results from an isotopic labeling study were potentially very complicated since the only site where deuterium incorporation is unlikely is the (It-position of the Vinylboronate ester fragment. To perform the labeling study, deuteroborane (DB02C10H6) 13 was first prepared by a modification of a procedure developed by H. C. Brown.88 Borontrifluoroetherate (Aldrich) was reacted with sodiumborodeuteride (Aldrich) generating diborane gas. The generated diborane was then trapped in a cooled solution of tetrahydrofuran containing dry 2,3-dihydroxynapthalene. Upon warming the solution, the desired deuteroborane was formed with liberation of HD gas. D/ D\ at 1,, /D 4BF3-OE12 + 3NaBDa : 3NaBF. + , g B\ (35) o 04 o D\B,s\D/o,B/D o’ r 4 \o D o ., THF > 2 ‘3—0 + 4HD (36) OH O/ 2‘ OH By following the deuterium label during the reaction using 2H NMR and 1H NMR spectroscopy, these experiments should provide insight into the reaction. When DB02C10H5 was reacted with Cp*2Ti(n2-CH2=CH2) under the same conditions as for the proton reaction 2H NMR shows that most of the label appears in the evolved ethane. When the pure Vinylboronate ester adduct is isolated from the isotopic labeling reaction, the product was analyzed using 1H vam, 2H NMR which indicates residual deuterium incorporation in the [i-position of the Vinylboronate ester. Integration of the vinylborane ligand in the 1H NMR spectra indicates a 15% loss of intensity at the site as illustrated in Figure 28. % D Incorporation Figure 28 92 Deuterium incorporation into the Vinylboronate ester is also confirmed by Mass spectral analysis. A mass spectrum of isolated product from the labeling study shows an increase in intensity of the peak at 197 m/z relative to 196 m/z. 2H NMR spectra rule against deuterium incorporation at the or-vinyl or Cp“ methyl positions, but instead show a broad hump which appears at 3.58 ppm corresponding to the position of the B-hydride in the Vinylboronate ester. The results from this labeling study are consistent with the mechanism proposed in Scheme 33 where B—H elimination is reversible, and H—D elimination from an intermediate olefin dihydride titanium complex and hydrogenation of Cp*2Ti(n2- CH2=CH2) are substantially faster than scrambling between hydride and Cp“ methyl sites. One feature of the proposed mechanism is the requirement that Cp*2Ti(n2- CH2=CH2) is preferentially hydrogenated in the presence of Cp*2Ti(n2- CH2=C(H)(B02C10H5)). This mechanistic requirement could be tested independently since the pure compounds can be subjected to competitive hydrogenation reaction. Cp*2T i(112-CH2=CH2) and Cp*2 T i (112- CH2=C(H)(302C10H5)) were dissolved in a 1:1 mixture in dg-toluene at -78 °C and exposed to 0.5 molar equivalents of Hz. Upon warming the solution, ethylene complex is exclusively hydrogenated in the presence ‘of the vinylborane complex as required by the proposed mechanism. E. Reactivity of Cp“2Ti(n2-CH2=C(H)(B02C10H6) Owing to the unfavorable steric interactions between the substituted olefin and the crowded coordination sphere of the Cp*2Ti fragment, the vinylborane is likely weakly ligated to the metal center and might be displaced by less sterically demanding ligands such as CO and C2H4. The reactivity of the vinylborane 93 complex with is shown in Figure 29. The complex Cp*2Ti(n2- CH2=C(H)(BOzC10H5) does indeed react with both carbon monoxide and ethylene to displace the Vinylboronate ester. Carbon monoxide reacts rapidly with Cp*2Ti(n2-CH2=C(H)(B02C10H6) to displace the substituted olefin ligand forming a TiII dicarbonyl complex and free CHz=C(H)(B02C10H6). An excess of ethylene also partially reacts with the Vinylboronate ester complex forming Cp*2Ti(n2-CH2=CH2) and free CI12=C(II)(B02C10H6). This latter substitution reaction with ethylene, however, occurs much more slowly with only 40% conversion 18 hours and is accompanied by some decomposition to unknown C 0 M00 > 1'] + CH2=C(H)302C10H6 (Fast) \VCO §\ products. CZH4 * KCHZ - (Slow, with T' (SI/H2 + CHZ-CIHIBOZCwHe decomposition) \\ Figure 29 94 F. Conclusion and Future Directions This latter reaction of ethylene displacing the coordinated Vinylboronate ester is significant because closes the catalytic loop and allows for the possible catalytic formation of vinylboranes from olefins and boranes as depicted in Figure 30. Attempts to catalytically convert ethylene to the corresponding vinylborane have not yet been successful using catecholborane or the napthalborane derivative. The slow displacement of the Vinylboronate ester by ethylene is a likely problem in designing an overall catalytic process because the Vinylboronate ester complex has not been proven to be inert to excess borane. This suggests that altering the rate of reactivity of the boranes or changing the steric properties of the coordinated Vinylboronate ester to facilitate displacement by ethylene might result in a successful catalytic cycle of olefin and borane conversion to Vinylboronate ester. Such a catalytic cycle capable of generating vinylboranes is relevant for two reasons. Vinylboranes are important to organic synthesis because they have been shown to react as excellent dienophiles in Diels-Alder reactions and vinylboranes can serve as benign vinyltin equivalents. 95 Figure 30. Catalytic Scheme for Generation of Vinylboronate Ester from an Olefin and Borane. ' 96 azoazo Poem: em 6...»...— fi/o or“. I/ / WK fawn: azoaxo Chapter Six Experimental Part I A. Solvent Purification All solvents for synthesis were refluxed under N2 for no less than 8 hours and freshly distilled prior to use. Hexane, cyclohexane, pentane, diethyl ether, benzene, toluene and tetrahydrofuran were refluxed over sodium. In the case of the latter three solvents, small amounts of benzophenone were added to form the blue or pln'ple ketyl radical anion or dianion, respectively, as an indicator of water content. Methylcyclohexane and acetoniuile were refluxed over calcium hydride and all halogenated solvents were refluxed over P205. Solvents used for NMR spectroscopy were deuterated dichloromethane (Cambridge Isotope Laboratories, 99.96%) and deuterated chloroform (Cambridge Isotope Laboratories, 99.8%). Dichloromethane and chloroform were dficxygenated by the freeze-pump-thaw method (10'6 torr) and vacuum distilled into a storage flask containing Linde 4 A molecular sieves, which were activated urltier dynamic vacuum (10'6 torr) at 250 °C for 12 hours. 97 9 8 B. Syntheses 1. Syntheses of Dicobalt Complexes 3. Starting Materials. The starting compounds C121C6H5108 CH3N(PF,_)2,109 C02(CH3N(PF3)2)3(CO)2,31 C02(CH3N(PF2)2)3(CO)(PPh3),39 and (PPh3)3CoX (X = Cl, Br, I) 33 and were prepared by literature methods. Phosphorus t1ifluoride(Ozark-Mahoning Company) was used as received. b. Attempted Synthesis of C02(CH3N(PF2)2)3(CO)Cl2, (1). To a 11 mL solution of dry degassed benzene, C02(CH3N(PF3)2)3(CO)2 (0.101g, 0.150 mmol) and C121C6H5 (0.041 g, 0.149 mmol) were added. A green precipitate formed almost immediately. The solution was evaporated to dryness to yield a green residue. The residue was washed once with hexane. Yield 0.012g. c. Preparation of C02(CH3N(PF2)2)3Br4 in attempted synthesis of C02(CH3N(PF2)2)3(CO)Br2, (2). In a flask, C02(CH3N(PF3)2)3(CO)2 (0.255g, 0.333 rnmol) was dissolved in 20 mL of dry degassed CH2C12. The reaction was initiated upon the addition of Br2 (0.053g, 0.333 mmol) in 3.4 mL of CH2C12. The solution is allowed to stir for three hours. The solution was evaporated to dryness under reduced pressure and the residue was washed with hexane. The remaining brown product was identified to be C02(CH3N(PF7)7)3Br4. Yield 0.125g (0.133mmol, 40%) The filtrate is evaporated to dryness to yield a purple solid, C02(CH3N(PF3)7)3(CO)2. Yield 0.095g (0.140 mmol, 42%). 9 9 (1. Preparation of C02(CH3N(PF2)2)3I4 in attempted synthesis of C02(CH3N(PF2)2)3(CO)I,_, (3). C02(CH3N(PF3)2)3(CO)2 (0.305g, 0.452 mrnol) was dissolved in 25 mLs of dry degassed CH2C12. To this solution is added 12 (0.115g, 0.452 mol) in 5.011115 of CH2C12. The solution is allowed to stir for three hours. The solution is then evaporated to dryness under reduced pressure and the residue is washed with hexane. The brown remaining product is C02(CH3N(PF2)2)3I4. Yield 0.183g (0.163 mmol, 36%) The filtrate is evaporated to dryness to yield a purple solid, C02(CH3N(PF3)2)3(CO)2. Yield 0.130g (0.194 mmol, 43%) e. Preparation of (PPh3)2CoCI42- in attempted synthesis of C02(CH3N(PF2)2)3(PPh3)Clz by disproportionation, (4). The fluorophosphine ligand, CH3N(PF2)2 (0.50 mL, 4.29 mol), was added to 10 mL of dry degassed THF containing CoCl(PPh3)3 (1.02g, 1.16 mmol). After 2 minutes of stirring, a blue precipitate formed. The solution was allowed to stir for an additional 30 minutes and it was then filtered. Yield 0.128g (0.177 mmol, 15.2%) f. Preparation of (PPh3)2C0Br42' in attempted synthesis of C02(CH3N(PF2)2)3(PPh3)Br2 by disproportionation, (S). Adry and degassed 10 mL THF solution of CoBr(PPh3)3 (0.600g, 0.649 mmol) was charged with CH3N(PF2)2 (0.50 1111., 4.29 mmol). After 15 minutes of stining, the Solution turned green. The solution was cooled to —15 °C for four hours during vWhich time a red-brown precipitate formed. Yield 0.120g (0.133 mmol, 20.5%) g. Preparation of (PPh3)zC 0142- in attempted synthesis 0 f C 02(CH3N(PF2)2)3(PPh3)Iz by disproportionation, (6). CoI(PPh3)3 (0.134g, 0.138 mmol) was dissolved in 5 mL of dry degassed THF. To this solution 1 0 0 is added CH3N(PF2)2 (0.10 ml. 0.86 mmol). After 15 minutes of stirring, the solution turned red-brown. A precipitate formed upon the addition of 17 mL of hexane. A red-brown solid is obtained upon filtering. Yield 0.020g (0.018 mnol, 13%) 2. Syntheses of Diiridium Complexes 3. Preparation of CH3N(PF2)2Ir2(tt-CH3N(PF2)2)3CI2, (7). To a flask of [Ir(Cl)(C8H14)2]2 (0.500g, 0.559 mmol) dissolved in 20 mL of dry degassed benzene, CH3N(PF2)2 (1.0 mL, 8.6 mmol) was added. The solution was allowed to stir for 30 minutes over which time it turned from yellow-orange to deep red. A 30 ml. addition of hexane prompted the formation of a deep red product, which was isolated by filtration. Yield 0.171 g (0.152 mmol, 27%) b. Preparation of CH3N(PF2)21r2(p.-MeN(PF2)2)3(SPh)2, (8). A flask of [Ir(u-SPh)(CO)2]2 (0.124g, 0.174 mmol) dissolved in 30 mL of dry degassed benzene was charged with CH3N(PF7)2 (0.30 mL 2.6 mmol) and the solution is allowed to stir for 10 minutes. The solution turns deep red. Upon removing the solvent under reduced pressure, the residue was washed with 20 mL of hexane to yield a orange solid. Yield 0.081g (0.064 mmol, 37%) c. Preparation of PhN(PF2)2Ir2(u-PhN(PF2)2)3Clz, (9). In a flask was placed [Ir(C1)(C3H14)2]2 (0.256g, 0.286 mmol), which is dissolved in 20 mL of dry degassed benzene. To this solution was added PhN(PF2)2 (0.98g, 4-28 mmol) and the solution was stirred for 20 minutes. The solution turned from yellow-orange to deep red. The resulting solution afforded a brown precipitate with the addition of 30 mL of hexane. Yield 0.076g (0.055 mmol, 19%) 101 3. Syntheses of Palladium Complexes a. Preparation of Pd4(CH3N(PF2)2)10, (10). A disproportionation reaction was attempted by mixing (PhCN)2PdC12 (0.100g, 0.261 mmol) and Pd2(dba)3 (0.119g, 0.130 mmol) in 10 mL of CH2C12 and in the presence of CH3N(PF2)2 (1.00 mL, 8.6 mmol). The green precipitate formed from the addition of hexane (15 mL) was isolated by filtration. Yield 0.085g (0.041mmol, 31%) C. Spectroscopic Instrumentation and Methods 1. Electronic Absorption Spectroscopy. Electronic absorption spectra were recorded on either a Cary 17 or Varian 2300 UV-Vis-NIR spectrometer. 2. Nuclear Magnetic Resonance Spectroscopy. The NMR spectra were recorded at the Max T. Rogers NMR facility at Michigan State University. The 31P{1H} NMR spectra were recorded on a Varian VXR-3OOS spectrometer. Phosphorous chemical shifts are reported in parts per million (5 scale), and measured relative to 85% I-I3PO4. Positive chemical shifts are downfield from the standard. The 1H NMR were recorded on either a Varian VXR-3OOS or Varian Gemini 300 spectrometer and chemical shifts were referenced to the chemical shifts of the deuterated solvent. 102 3. Mass Spectrometry. The fast atom bombardment mass spectra (FABMS) were recorded on a JEOL I-D( 110 double focusing mass spectrometer housed in the NII-I/MSU Mass Spectrometry Facility. Samples were dissolved in o- nitrobenzyl alcohol matrices. The acceleration potential for this work was 10 kV. Data were acquired, stored, and processed on a JEOL DA50000 data system. Fast Atom Bombardment (FAB) ionization was performed using a 6kV neutral beam of Xenon atoms. 4. Infrared Spectroscopy. Infrared spectra were obtained as KBr pellets on a Nicolet IR/42 spectrometer. 103 Experimental Part II A. General Procedures All manipulations were performed using glove box, Schlenk or vacuum- line techniques. All solvents for synthetic use were refluxed under N2 for no less than 8 hours and freshly distilled prior to use. Hexane, cyclohexane, pentane, diethyl ether, benzene, and tetrahydrofuran were refluxed over sodium/benzophenone ketyl before being distilled. Toluene was distilled over sodium metal. All solvents were rigorously deoxygenated before use. Solvents used for NMR spectroscopy were benzene-56 and toluene-68. These solvents were dried by refluxing one day over sodium then distilled. Argon was purified by passage through a column of MnO on silica. All chemicals were reagent grade and used as received unless otherwise noted. Elemental analyses were performed by Desert Analytics, Tucson, AZ. B. Syntheses l. Syntheses of Borane Reagents BF3-Et20 (Aldrich) was vacuum distilled (60 °C at 15 torr) from CaH2 before use. 2,3-Dihydroxynaphthalene (Aldrich), NaBD4 (Aldrich) and BH3-THF (l M solution) (Aldrich) were used as received. Catecholborane (Aldrich) was purified by vacuum transfer before use. Ethylene (99.9%) and Carbon Monoxide (99.9%) were purchased from Matheson and used as received. 104 a. HBOzClng, (11). A 20 mL tetrahydrofuran solution of 2,3- dihydroxynaphthalene (2.00 g, 12.5 mmol) was placed in a flask and cooled to - 78 °C. BH3-THF (1 M Sol'n in THF, 13.8 mls) was cooled to -78 °C and transferred via cannula to the 2,3-dihydroxynaphthalene with vigorous stirring. The solution was warmed to room temperature with evolution of hydrogen. The solvent was then removed under reduced pressure to leave a white residue. The residue was sublimed at 71 0c at 1x104 torr to give 1.75 g (73%) of a white solid (mp. 84 °C). lH NMR(C6D5); 67.51 (m, 2H. 02C10H6)r 6 7.30 (s, 2H, 02C10H6), 6 7.20 (m, 2H, 02C10H6), 6 5.90-2.55 (qt, 1H, HB), llB NMR(C6D6);629.8. b. DBOZC6H4, (12). In a flask catechol (1.00 g, 9.09 mmol) was dissolved in 15 mL tetrahydrofuran and precooled to -78 °C. In a second flask, 2 mL of BF3-Et20 was added dropwise over 20 minutes to a precooled solution (0 °C) of NaBD4 (0.468 g, 11.2 mmol) in 3 mL dirnethoxyethane. After the addition of BF3-Et20 was complete, the NaBD4 solution was heated with a water bath to 40 °C for 1/2 hr. Diborane, generated during the addition, was sent through a cannula by a slow nitrogen purge to the catechol. The vigorously stirred solution of catechol was then warmed to room temperature and hydrogen evolution was observed. The tetrahydrofuran was removed under reduced pressure to give 2-3 mls of crude liquid. The liquid was vacuum distilled (62-67 °C. 50 torr) to give 0.604 g (45%) of deuterated catecholborane. lH NMR(C6D6); 5 6.97 (m, 2H, 02C6H4). 6 6.74 (m, 2H, 02C6H4), 1113 NMR(C6D6);629.6. c. DBOZC10H6, (13). In a flask 2,3-dihydroxynaphthalene (1.38 g, 8.62 mmol) was dissolved in 15 mL tetrahydrofuran and precooled to -78 °C. In a second flask, 2 mL of BF3-Et20 was added dropwise over 20 minutes to a precooled solution (0 0C) of NaBD4 (0.487 g, 11.6 mmol) in 3 mL 10 5 dimethoxyethane. After the addition of BF3-Et20 was complete, the NaBD4 solution was heated with a water bath to 40 °C for 1/2 hr. Diborane, generated during the addition, was sent through a cannula by a slow nitrogen purge to the 2,3-dihydroxynaphthalene. The vigorously stirred solution of 2,3- dihydroxynaphthalene was then warmed to room temperature and hydrogen evolution was observed. The tetrahydrofuran was removed under reduced pressure to give a white residue. The residue was sublimed at 71 °C at 1x10‘4 torr to give 0.848 g (58%) of white solid. 1H NMR(C6D6); 6 7.51 (m, 2H, 02C10H6), 6 7.30 (s, 2H,02C10H6), 67.20 (m, 2H, ozclong), 1113 NMR(C6D6);6 29.8. 2. Syntheses of Tantalum Boryls a. Endo-szTaH2(B02C6H4) (14) and Exo- szTaH2(BOzC6H4) (15). A 10 mL toluene solution of B-chloro- catecholborane (418 mg, 1.31 mmol) was added dropwise to a vigorously stirred toluene suspension (25 mL) of {CpgTaHzLi}x (202 mg, 1.31 mmol) at -78 °C. The mixture was gradually warmed and stirred for 30 min. at room temperature during which time the orange solids were replaced with an off-white precipitate (LiCl) accompanied by the formation of a pale amber solution. The solvent from the filtrate was stripped and the crude product (321 mg, 71 mmol, 57% yield) was analyzed by 1H NMR where integration revealed a 32:68 (14:15) mixture of isomers. The isomers were separated by fractional crystallization fiom toluene. 14 was isolated as pale yellow microcrystals (80 mg, 19 %) as the less soluble isomer. Concentration of the mother liquor and further cooling afforded 15 as pale microcrystals (132 mg, 23%). 14le NMR (C6D6); 6 7.14 (m, 2H, 02C6H4), 6 6.82 (m. 2H. 02C6H4), 64.85 (s, 10H C5H5). 6424 (s, 2H, TaHz). “B NMR 6 70.0, Dnm = 250 Hz. IR (cm'l): 1768 "Tan Anal. calcd. for C14.2H20_3B02Ta: C, 49.78; 1 0 6 H, 4.27. Found: C, 49.55; H, 4.17. 15: 1H NMR (C6D6); 6 7.11 (m, 2H, 02C5H4). 6 6.84 (m, 2H, 02C6H4). 64.83 (s, 10H C5H5). 6 -4.20 (d, 1H, TaH2). 6 517 (hr, 1H, TaI-Iz). "B NMR 664.7, Dnm = 190 Hz. IR (cm-l): 1771, 1703 am. 3. Syntheses of Titanium Starting Materials and the Vinylboronate Ester Complex. 3. I"szTiCl, (16). TiCl3-(THF)3 (5.00g. 13.5 mmol) and *CpNa (4.70 g. 29.7 mmol) were place in a Schlenk flask. To this flask was added 100 mls of dry, degassed THF and the resulting solution was then refluxed for 8 hrs under argon. The solution was then evaporated to dryness under reduced pressure leaving a dark blue residue. The dark blue’residue was dried by heating with a warm water bath under vacuum for 4 hrs. The residue was then extracted with pentane and filtered. The filtrate was then cooled to -80 °C leaving dark blue crystals. Yield 4.78g (9.32 mmol, 69%) b. Cp*,'ri(n2-CH,=CH,), (17). In a Schlenk flask was made 0.9% Na/Hg amalgam. To this flask was added *szTiCl (2.50 g, 7.07 mmol) dissolved in 150 mls of dry degassed toluene. The flask was then pressurized to 700 torr with ethylene and stirred vigorously. After six hrs more ethylene was added to maintain a pressure of 700 torr in the flask. The reaction was then stirred for 60 hrs and filtered. The gray-green filtrate was evaporated to dryness and the residue extracted with pentane. The pentane solution was then filtered and the filtrate was then cooled to -80 °C leaving green crystals. Yield 1.93 g (5.58 mmol, 79 %) 1 0 7 c. Cp'zTim2-CH2=CHB02C6H4), (18). Cp*2Ti(n2-C2H4) (0.658 g, 1.90 mmol) was placed in a flask, dissolved in 50 mL toluene and cooled to -78 °C. In a second flask, HB02C6H4 (0.200 mL, 0.188 mmol) was dissolved in 10 mL toluene, cooled to -78 °C, and transferred by cannula to the Cp*2Ti(112- C2Ha). The reaction immediately turned amber. The reaction was warmed to room temperature and stirred for 1 hr. The solvent was removed under reduced pressure and the residue was washed twice with cold 10 mL pentane (-78 °C) to afford of lemon-yellow solid. Yield 0.368 g (0.80 mmol, 42%) 1H NMR(C6D6); 6 7.04 (m, 2H, 02C6H4), 6 6.78 (m, 2H, 02C6H4), 6 3.56 (dd, lI-I, H2C=CHB), 6 2.85 (dd, 1H, H2C=CHB), 62.63 (dd, 1H, H2C=CHB), 6 1.72 (s, 15H, C5(CH3)5), 6 1.70 (s, 15H, C5(CH3)5), 118 NMR(C6D5); 636. 13c NMR(C6D5):6149.58 (BOzCz), 6130.99 (02C5H4), 6124.16 (02C6H4), 6 121.67 (C5(CH3)5), 6121.37 (C5(CH3)5), 6 108.86 (112C=CI-IB), 6 107.16 (OzClng), 6 88.12 (H2C=CHB), 6 11.95 (C5(CH3)5), 6 11.83 (C5(CH3)5). 1 0 8 d. Cp*sTi(n2-CH2=C(H>(BOZCrant», (19). CP*2Ti(112-C2H4) (0.280 g, 1.33 mmol) was placed in a flask, dissolved in 20 mL toluene and cooled to -78 °C. In a second flask, P1802C10116 (0.131 g, 0.768 mmol) was dissolved in 5 mL toluene, cooled to -78 °C, and transferred by cannula to the Cp*2Ti(rI2- C2114). The reaction immediately turned amber. The reaction was warmed to room temperature and stirred for 1 hr. The solvent was removed under reduced pressure and the residue was washed twice with 15 mL pentane to afford 0.183 g (44%) of lemon-yellow solid. When lH NMR(C6D6); 6 7.52 (m, 2H, 02C10H6), 6 7.33 (s, 2H, 02C10H6), 67.18 (m, 2H, OZCIOHG), 6 3.58 (dd, 1H, H2C=CI-IB), 6 2.89 (dd, 1H, H2C=CHB), 62.64 (dd, 1H, H2C=CHB), 61.73 (s, 15H, C5(CH3)5), 6 1.72 (s, 15H, C5(CH3)5), 118 NMR(C5D5); 636. 13C NMR(C6D6);6149.57 (s, B02C2), 6130.98 (s, 02C10H6), 6 124.16 ((1, IJCH = 159 Hz, OZCIOHG), 6 121.66 (s, C5(CH3)5), 6 121.36 (s, C5(CH3)5), 6108.86 (t, 1ch = 146 Hz, H2C=CHB), 6 107.18 (d, 1ch = 162 Hz, 02C10Hg), 6 88.12 ((1, br, H2C=CHB), 6 11.97 (qt, 1JCH = 126 Hz, C5(CH3)5), 611.86 (qt, IJCH = 126 Hz, C5(CH3)5). Mass. Spectrum. m/e 318 (47, Cp*2Ti), 196 (100, H2C=C(H)(B02C10H6)) Anal. Calcd for Cp*2Ti(h2- CH2=C(H)(B02C10H6): C, 74.73; H, 7.64. Found: C, 75.04 H, 7.86. 109 4. WTM Reactivity of the Titanium Vinylboronate Ester Complex with Ethylene and Carbon Monoxide a. Cp*zTi(n2-CH2=C(H)(BOZC10H6))+ (30(3). In an NMR tube Cp*2Ti(n2-CH2=C(H)(B02C10H5)) (0.018 g, 0.035 mmol) was dissolved in benzene-66 and degassed on a high-vacuum line by three freeze-pump-thaw cycles. The solution was warmed to room temperature and pressurized to 1 atrn with CO(g). 1H NMR (C6D6); CH2=C(H)(B02C10H6) 6 7.53 (m, 2H, 02C10H5), 67.34 (s, 2H, 02C10H6), 6 7.20 (m, 2H, 02C10H6), 6 6.47 (m, 1H, CH2=CHB), 6 6.07 (m, 1H, CH2=CHB), 6 5.92 (m, H. CngCHB), lH NMR(C5D6); Cp*2Ti(CO)2 61.65 (C5(CH3)5). b. Cp*2Ti(n2-CH2=C(H)(BOZCloHsD + C2H4(g). Cp*2Ti(n2- CH2=C(H)(B02C10H6)) (0.015 g, 0.029 mmol) was dissolved in benzene-66 in an NMR tube. The sample was degassed on a high-vacuum line by three freeze- pump-thaw cycles, warmed to room temperature, and pressurized to l atrn with C2H4-(g)- lH NMR(C6D6); CH2=C(H)(1302C 10116) 5 7 .53 (m. 2H. 02C10H6). 5 7.34 (s, 2H, 02C10H6), 67.20 (m, 2H, 02C10H6), 6 6.47 (m, 1H, CH2=CHB) 6 6.07 (m, 1H, CH2=CHB), 65.92 (m. H. CH2=CHB), 1H NMR(C6D6); Cp*2ri(h2- C2H4) 6 2.05 (C2H4) 6 1.68 (C5(CH3)5). List of References 10. References “Proton Transfer in Reaction Centers from Photosynthetic Bacteria”, M. Y. Okamura and G. Feher Annu. Rev. Biochemistry 1992, 61, 861. “Coupled Proton and Electron Transfer Pathways in the Acceptor Quinone Complex of the Reaction Centers”, E. Takahashi, P. Maroti and C. Wraight In Electron and Proton Transfer in Chemistry and Biology; A. Muller, H. Ratajczaks, W. Junge and E. Diemann, Ed.; Elsevier: Amsterdam, 1992; pg.219. “Water Oxidation in Photosystem II: From Radical Chemistry to Multielectron Chemistry”, G. T. Babcock, B. A. Barry, R. J. Debus, C. W. Hoganson, M. Atarnian, L. McIntosh, I. Sithole and C. F. Yocum Biochemistry 1989, 28, 9557. The Photosynthetic Bacterial Reaction Center-Structure and Dynamics; J. Breton and A. Vermeglio, Ed.; Plenum: New York, 1988. Electron Transfer in Biology and the Solid State; Johnson, M. K., King, R. B., Kurtz, D. M., Jr., Kutal, C., Norton, M. L., Scott, R. A., Eds.; Advances in Chemistry 226; American Chemical Society: Washington, DC, 1990. Electron Transfer in Inorganic, Organic, and Biological Systems; Bolton, J ., Mataga, N., McLendon, G., Eds.; Advances in Chemistry 228; American Chemical Society: Washington, DC, 1992. Chisholm, M. H. In Reactivity of Metal-Metal Bonds; Chisholm, M. H., Eds.; ACS Symposium Series 155; American Chemical Society: Washington, DC, 1981; Chapter 2. Rice, S. F.; Gray, H. B. J. Am. Chem. Soc. 1983, 105, 4571. Rice, S. F.; Miskowski, V. M.; Gray, H. B. Inorg. Chem. 1988, 27, 4704. Dallinger, R. F.; Miskowski, V. M.; Gray, H. B.; Woodruff, W. H. J. Am. Chem. Soc. 1981, 103, 1595. 110 ll. 12. 13. 14. 15. 16. l7. 18. 19. 21. 22. 23. 24. 25. 26. 27. 28. 111 Che, C.-M.; Butler, L. G.; Gray, H. B.; Crooks, R. M.; Woodruff, W. H. J. Am. Chem. Soc. 1983, 105, 5492. Roundhill, D. M.; Gray, H. B.; Che, C.-M. Acc. Chem. Res. 1989, 22, 55. Hill, M. G.; Mann, K. R. Inorg. Chem. 1991, 30, 1429. Smith, D. C.; Gray, H. B. In The Challenge of d and f Electrons; Salahub, D. R., Zemer, M. C., Eds.; ACS Symposium Series 394; American Chemical Society: Washington, DC, 1989; pp 356. Smith, D. C.; Gray, H. B. Coord. Chem. Rev. 1990, 100, 169. Engebretson, D. 8.; Zaleski, J. M.; Leroi, G. E.; Nocera, D. G. Science 1994, 265, 759. Duleboth. 1.; Ward, D. L.; Nocera, D. G. J. Am. Chem. Soc. 1990, 112, 2969. DeLaet, D. L.; Powell, D. R.; Kubiak, C. P. Organometallics, 1985, 4, 954. McLennan, A. J.; Puddephatt, R. J. Organometallics, 1986, 5, 811. McDonald, R.; Cowie, M. Inorg. Chem. 1990, 29, 1564. Meier, E. B.; Burch, R. R.; Muetterties, E. L.; Day, V. W. J . Am. Chem. Soc., 1982, 2661. Hlatky, G. G.; Johnson, B. F. G.; Lewis, J.; Raithby, P. R. J . Chem. Soc., Dalton Trans., 1985, 1277. Meyer, K. E.; Fanwick, P. E.; Walton, R. A. Inorg. Chem. 1992, 31, 4486. Orpen, A. G.; Connelly, N. G. J. Chem. Soc., Chem. Commun. 1985, 1310. Marynick, D. S. J. Am. Chem. Soc. 1984, 106, 4064. Xiao, S. X.; Trogler, W. C. ; Ellis, D. E.; Berkovitch-Yellin, Z. J. Am. Chem. Soc. 1983, 105 , 7033. Avanzino, S. C.; Chen, H. W.; Donahue, C. J .; Jolly, W. L. Inorg. Chem. 1980, 19, 2201. Corderrnan, R. R.; Beauchamp, J. L. Inorg. Chem. 1978, 17, 1585. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 41. 42. 43. 45. 46. 47. 112 Savariault, J. M.; Serafini, A.; Pelissier, M.; Cassoux, P. Theor. Chim. Acta. 1976, 42, 155. Balzanii, V.; Sabbatini, N .; Scandola, F. Chem. Rev. 1986, 86, 319. King, R. B. Acc. Chem. Res. 1980, I3, 243. Newton, M. G.; Pantaleo, N. S.; King, R. B.; Lotz, T. J. J. Chem. Soc., Chem. Commun. 1978, 514. Chaloyard, A.; El Murr, N.; King, R. B. J. Organometal. Chem. 1980, 188, C13. Gard, D. R.; Brown, T. L. J. Am. Chem. Soc. 1982, 104, 6340. Freedman, A.; Bersohn, R. J. Am. Chem. Soc. 1978, 100, 4116. Rothberg, L. J .; Cooper, N. J .; Peters, K. S.; Vaida, V. J. Am. Chem. Soc. 1982, 104, 3536. Morse, D. L.; Wn'ghton, M. S. J. Am. Chem. Soc. 1982, 104, 3536. Klein, H. F.; Karsch, H. H. Inorg. Chem. 1975, 14, 473. King, R. B.; Gimeno, J .; Lotz, T. J. Inorg. Chem. 1975, I 7, 2401. Herde, J. L.; Lambert, J. C.; Senott, C. V. Inorg. Synth. 1974, 15, 18. Bonnet, J. J .; Galy, J .; De Montauzon, D.; Poilblanc, R. J. Chem. Soc.Chem. Comm, 1977, 47. Kadis, J. Ph.D. Thesis, Michigan State University, 1993. Takahashi, Y.; Ito, T.; Sakai, S.; Ishii, Y. Chem. C ommun., 1970, 1065. Moseley, K.; Maitlis, P. M. J. Chem. Soc., Dalton Trans., 1974, 169. Wilke, G.; Bogdanovic, B.; Hardt, P., Heimbach, P.; Keim, W.; Kroner, M.; Oberkirch, W.; Tanaka, K.; Steinriicke, E.; Walter, D.; Zimmerman, H. Angew. Chem, Int. Ed. Engl., 1966, 5, 151. Dubrawski, J .; Kriege-Simondsen, J. C.; Feltham, R. D. J. Am. Chem. Soc. 1980, 102, 2089. Moiseev, I. I.; Stromnova, T. A.; Vargaftig, M.N.; Mazo, G. J .; Kuz'mina L. G.; Struchkov, Y. T. J. Chem. Soc.Chem. Comm, 1978, 27. 48. 49. 50. 51. 52. 53. 54. 55. 56. 57. 58. 59. 61. 62. 63. 113 For recent reviews of metallocarborane chemistry see: (a) Stone, F. G. A. Adv. Organomet. Chem. 1990, 3 1 , 53. (b) Grimes, R. N.; In "Comprehensive Organometallic Chemistry," Wilkinson, G.; Stone, F. G. A.; Abel, E.; eds,. Vol. 1, Chap. 5.5 Pergamon, Oxford, 1982. Schmid, G.; Petz, W.; N 6th, H. Inorg. Chim. Acta. 1970, 4, 423. G. Wilkinson, F. G. A. Stone and E. W. Abel Comprehensive Organometallic Chemistry; Pergamon Press: Oxford, 1982. Lauher, J. W.; Hoffman, R. J. Am. Chem. Soc 1976, 98, 1729. Seyferth, D.; Raab, G.; Grim, S. O. J. Org. Chem. 1961, 26, 3034. Parshall, G. W. J. Am. Chem. Soc. 1964, 86, 361. Néth, H.; Schmid, G. Angew. Chem. 1963, 75, 861. Noth, H.; Schmid, G. J. Organometal. Chem. 1966, 5, 109. Noth, H.; Schmid, G. Allgem. Prakt. Chem. 1966, 17, 610. Nam, H.; Schmid, G. Z. Anorg. Allg. Chem. 1966, 345, 69. Noth, H.; Schmid, G. J. Organometal. Chem. 1967, 7, 129. Knorr, J. R.; Merola, J. S. Organometallics, 1990, 9, 3008. (a) Gilbert, K. B.; Boocock, S. K.; Shore, 8. G. In Comprehensive Organometallic Chemistry; G. Wilkinson, F. G. A. Stone and E. W. Abel, Eds.; Pergamon: Oxford, 1982; Vol. 67, Chap. 41.1; pp 879. (b) Schmid, G. Angew. Chem, Int. Ed Engl. 1970, 9, 819. (c) Hartwig, J. F.; Buber, S. J. Am Chem. Soc. 1993, 115 , 4908. (d) Nguyen, P.; Blom. H. P.k Westcott, S. A.; Taylor, H. J .; Marder, T. B. J. Am. Chem. Soc. 1993, 115, 9329. Our description of "simple" bonds between metals and boron excludes the vast metallocarborane literature. (a) Baker, R. T .; Ovenal, D.; Calabrese, J. C.; Westcott, S. A.; Taylor, N. J.; Williams, I. D.; Marder, T. B. J. Am. Chem. Soc. 1990, 112, 9399. (b) Burgess, K.; van der Donk, W. A.; Westcott, S. A.; Marder, T. E.; Baker, R. T.; Calabrese, J. C. J. Am. Chem. Soc. 1992, 114, 9350. (a) Evans, D. A.; Fu, G. C.; Hoveyda, A. H. J. Am. Chem. Soc. 1992, 114, 6671. (b) Evans, D. A.; Fu, G. C. Anderson, B. A. J. Am. Chem. Soc. 1992, 114, 6679. 65. 66. 67. 68. 69. 70. 71. 72. 73. 74. 75. 76. 77. 78. 79. 80. 114 Harrison, K. M.; Marks, T. J. J. Am. Chem. Soc.1992, 114, 9220. Burgess, K.; Ohlmeyer, M. J. Chem. Rev. 1991, 91, 1179 and references within. Kono, H.; Ito, K. Chem. Lett. 1975, 1095. Westcott, S. A.; Taylor, N. J.; Marder, T. E.; Baker, R. T.; Jones, N. J.; Calabrese, J. C. J. Chem. Soc., Chem. Commun. 1991, 304. Nguyen, P.; Blom, H. P.; Westcott, S. A.; Taylor, N. J.; Marder, T. B. J. Am. Chem. Soc. 1993, 115, 9329. 56. Baker, R. T.; Calabrese, J. C. J. Am. Chem. Soc. 1993, 115, 4367. Brown, H. C.; Subba Rao, B. C. J. Am. Chem. Soc. 1956, 78, 2582. Pelter, A.; Smith, K.; Brown, H. C. In Borane Reagents; Academic Press: New York, 1988. Mannig, D.; N6th, H. Angew. Chem., Int. Ed. Engl. 1985, 24, 878. (a) Evans, D. A.; Fu, G. C.; Hoveyda, A. H. J. Am. Chem. Soc. 1988, 110, 6917. (b) Burgess, K.; Cassidy, J .; Ohlmeyer, M. J. J. Org. Chem. 1991, 56, 1020. (c) Burgess, K.; Ohlmeyer, M. J. J. Org. Chem. 1991, 56, 1027. d) Burgess, K.; Ohlmeyer, M. J. Tetrahedron Lett. 1989, 30, 395, 5857, 5861. (e) Burgess, K.; van der Donk, W. A.; Jarstfer, M. B.; Ohlmeyer, M. J. J. Am. Chem. Soc. 1991, 113, 6139. (a) Evans, D. A.; Fu, G. C. J. Org. Chem. 1990, 55, 2280. (b) Evans, D. A.; Fu, G. C. J. Am. Chem. Soc. 1991, 113, 4042. Still, W. C.; Barrish, J. C. J. Am. Chem. Soc. 1983, 105 , 2487. Brown, H. C. In Hydroboration; Wiley-Interscience: New York, 1962. Brown, H. C.; Kramer, G. W.; Levy, A. B.; Midland, M. M. In Organic Synthesis via Boranes; Wiley-Interscience: New York, 197 5. Halpem, J .; Okamoto, T.; Zakhariev, A. J. Mol Catal. 1976, 2, 65. Burgess, K.; van der Donk, W. A.; Kook, A. M. J. Org. Chem. 1991, 56, 2924. Brown, J. M.; Lloyd-Jones, G. C. J. Am. Chem. Soc. 1994, 116, 866. 81. 82. 83. 84. 85. 86. 87. 88. 89. 91. 92. 93. 94. 95. 96. 97. 98. 99. 100. 115 Ogle, C. A.; Hubbard, J. C.; Masterman, J. L. J. Chem. Soc., Chem. Commun. 1991, 1733. Carvalho, M.; Weiserman, L. F.; Hercules, D. M. Appl. Spectrosc. 1982, 36, 290. Westcott, S. A.; Taylor, N. J.; Marder, T. B.; Baker, R. T.; Jones, J. J. Chem. Soc., Chem. Commun. 1991, 304. Burgess, K.; Ohlmeyer, M. J. Chem. Rev. 1991, 91, 1179 and references within. Burgess, K.; Ohlmeyer, M. J. J. Org. Chem. 1988, 53, 5178. Isagawa, K.; Sano, H.; Hattori, M.; Otsuji, Y. Chem. Lett. 1979, 1069. Lee, H. S.; Isagawa, K.; Otsuji, Y.Chem. Lett. 1984, 363. Lee, H. S.; Isagawa, K.; Toyoda, H.; Otsuji, Y. Chem. Lett. 1984, 673. Lucas, C. R. Inorg. Synth. 1977, 17, 91. Bercaw, J. E. J. Am. Chem. Soc. 1974, 96, 5087. Bunker, J. M.; De Cian, A.; Green, M. L. H.; Moreau, J. J. E.,; Siganporia, N. J. Chem. Soc., Dalton Trans. 1980, 2155. Hartwig, J. F.; De Gala, S. R. J. Am. Chem. Soc. 1994, 116, 3661. Melmed, K. M.; Coucouvanis, D.; Lippard, S. J. Inorg. Chem. 1973, 12, 232. ‘ Jensen, J. A.; Girolami, G. S. J. Am. Chem. Soc. 1988, 110, 4450. Hartwig, J. F., personal communication. Wilson, R. D.; K6etxle, R. F.; Hart, D. W.; Kvick A; Tipton, D. L.; Bau, R. J. Am. Chem. Soc. 1977, 99, 1775. Brookhart, M.; Green, M. L. H. J. Organomet. Chem. 1983, 250, 395. Calvert, B. R.; Shapley, J. R. J. Am. Chem. Soc. 1978, 100, 7726. Crabtree, R. H.; Holt, E. M.; Lavin, M.; Morehouse, S. M. Inorg Chem. 1985, 24, 1986. (a) Berry, D. H.; Jiang, Q. J. Am. Chem. Soc. 1987, 109, 6210. (b) Jiang, Q.; Carroll, P. J .; Berry, D. H. Organometallics 1991, 10, 3648. 101. 102. 103. 104. 105. 106. 107. 108. 109. 116 Manriquez, J. M.; Mcalister, D. R.; Sanner, R. D.; Bercaw, J. E. J. Am. Chem. Soc. 1976, 98, 6733. Erker, G.; Rosenfeldt, F. Angew. Chem. Int. Ed. Engl. 1978, 17, 605. Lantero, D. R.; Motry, D. H.; Smith, M. R. J. Am. Chem. Soc. 1994, 116, 1081]. Cohen, S. A.; Auburn, P. R.; Bercaw, J. E. J. Am. Chem. Soc. 1983, 105, 1136. Cohen, S. A.; Bercaw, J. E. Organometallics 1985, 4, 1006. Smith, M. R., personal communication. Matteson, D. S. J. Am. Chem. Soc. 1960, 82, 4228. Rempel, G. A.; Legzdine, P.; Smith, H.; Wilkinson, G. Inorg. Synthesis, 1972, 13, 90. Nixon, J. F. J. Chem. Soc. A. 1968, 2689. "711111111111117115