UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Dinuclear palladium complexes with bridging hydrides Clentsmith, Guy Kenneth Bruce 1991

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


831-UBC_1991_A6_7 C63.pdf [ 3.33MB ]
JSON: 831-1.0059508.json
JSON-LD: 831-1.0059508-ld.json
RDF/XML (Pretty): 831-1.0059508-rdf.xml
RDF/JSON: 831-1.0059508-rdf.json
Turtle: 831-1.0059508-turtle.txt
N-Triples: 831-1.0059508-rdf-ntriples.txt
Original Record: 831-1.0059508-source.json
Full Text

Full Text

DINUCLEAR PALLADIUM COMPLEXES WITH BRIDGING HYDRIDES by Guy Kenneth Bruce Clentsmith, B. Sc. (Hons), University of New South Wales, 1988. A thesis submitted in partial fulfillment of the requirements for the degree of Master of Science in THE FACULTY OF GRADUATE STUDIES, Department of Chemistry. We accept this thesis as conforming to the required standard, The University of British Columbia, September 1991. © Guy Kenneth Bruce Clentsmith, 1991. In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of C H E M i g r O /  The University of British Columbia Vancouver, Canada Date l o - ' O - ^ V DE-6 (2/88) i i Abstract The palladium hydride dimer, [(dippp)Pd]2(M.-H)2, (2) (dippp = 1,3-bis(diisopropylphosphino)propane), was synthesized, and its interaction with lithium tetraethylborate examined. When the two compounds were mixed in stoichiometric amounts high yields of the adduct, [(dippp)Pd]2(|J.-H)2*LiBEt4 (1) — previously isolated adventitiously by a co-worker — were obtained. Variable temperature 3 1P{ 1H} NMR spectroscopy showed a condition of chemical equilibrium between 2 and 1, but the value of the equilibrium constant, Keq, could not be determined owing to its large magnitude. Extension of this chemistry to other sources of L i + led to the isolation of the aluminate adduct, [(dippp)Pd]2(|i-H)2,*LiAlEt4 (3). Likewise when NaBEt4 was substituted an adduct formulated as [(dippp)PdJ2(|i-H)2'NaBEt4 (4) was obtained. The interaction of tetraethylborate salts with other transition-metals was also examined, with evidence of adduct formation for both [(dippp)Rh]2(|J.-H)2 and [(dippp)Ni]2(|J.-H)2 (6) with LiBEt4. It was not possible to isolate these compounds as analytically pure samples. This general interaction is discussed in terms of metal basicity. The reactivity of 2 towards organic donor species was also studied. Addition of excess donor results in the formation of tricoordinate Pd(0) species: (dippp)Pd(rj2-H 2 C = C H 2 ) (5), (dippp)Pd(PPh3) (7), and (dippp)Pd(DMAD) (DMAD = dimethylacetylenedicarboxylate) (8) were thus isolated. When a stoichiometric equivalent of donor was added to 2, no dimeric intermediates could be observed. ii Abstract The palladium hydride dimer, [(dippp)Pd]2(M--H)2, (2) (dippp = 1,3-bis(diisopropylphosphino)propane), was synthesized, and its interaction with lithium tetraethylborate examined. When the two compounds were mixed in stoichiometric amounts high yields of the adduct, [(dippp)Pd]2(p.-H)2*LiBEt4 (1) — previously isolated adventitiously by a co-worker — were obtained. Variable temperature 3 1 P{ 1 H) NMR spectroscopy showed a condition of chemical equilibrium between 2 and 1, but the value of the equilibrium constant, Keq, could not be determined owing to its large magnitude. Extension of this chemistry to other sources of L i + led to the isolation of the aluminate adduct, [(dippp)Pd]2((i-H)2,'LiAlEt4 (3). Likewise when NaBEt4 was substituted an adduct formulated as [(dippp)Pd]2(p>H)2,NaBEt4 (4) was obtained. The interaction of tetraethylborate salts with other transition-metals was also examined, with evidence of adduct formation for both [(dippp)Rh]2(p>H)2 and [(dippp)Ni]2(|i-H)2 (6) with LiBEt4. It was not possible to isolate these compounds as analytically pure samples. This general interaction is discussed in terms of metal basicity. The reactivity of 2 towards organic donor species was also studied. Addition of excess donor results in the formation of tricoordinate Pd(0) species: (dippp)Pd(rj2-H 2 C = C H 2 ) (5), (dippp)Pd(PPh3) (7), and (dippp)Pd(DMAD) (8) (DMAD = dimethylacetylenedicarboxylate) were thus isolated. When a stoichiometric equivalent of donor was added to 2, no dimeric intermediates could be observed. i i i Table of Contents Page Abstract ••—ii Table of Contents iii List of figures vi List of Tables. viii Abbreviations Acknowledgements xi Chapter 1: INTRODUCTION 1 1.1 Preamble 1 1.2 Metal hydrides: 2 1.2.1 General 2 Historical background 2 Structures.... .• 3 Characterization 4 1.2.2 Theoretical considerations 6 Bonding description of (12 hydrides 6 Donor-Acceptor formalism as applied to bridging hydride complexes .9 1.2.3 Summary of previous work 10 1.3 Aim and scope of this work 11 1.4 References 12 Chapter 2: HYDRIDES OF PALLADIUM 15 2.1 Introduction 15 2.2 Survey of palladium hydrides • 15 2.2.1 Range 15 2.2.2 Palladium hydrides in catalysis 17 2.3 Isolation and structure of [(dippp)Pd]2(M--H)2,LiBEt4 19 2.4 Routes to [(dippp)Pd]2(u-H)2 . 24 2.5 Interaction of palladium hydride dimers with LiBEt4 and NaBEt4 30 2.6 Bonding considerations in [(dippp)Pd]2(li-H)2,LiBEt4 33 2.6.1 Lithium in association with transition-metals 34 2.6.2 Structures of LiBMe4 and related compounds 35 2.6.3 7 L i NMR Spectrum of [(dippp)Pd]2(M-H)2*LiBEt4 (1) 38 2.6.4 Metal basicity 38 2.7 Further chemistry of [(dippp)Pd]2(M.-H)2 41 2.8 Interaction of L1BE14 with other transition-metals 44 2.9 Summary and future prospects 45 2.10 References 46 Chapter 3: EXPERIMENTAL... 47 3.1 General. 47 3.1.1 Procedures 47 3.1.2 Reagents and starting materials 48 Chapter 3: EXPERIMENTAL. 50 3.1 General 50 3.1.1 Procedures 50 3.1.2 Reagents and starting materials 50 3.2 Preparations 51 3.2.1 (dippp)PdX2, (X = CI, I).. 51 3.2.2 [(dippp)Pd]2(Ll-H)2 (2) 52 3.2.3 [(dippp)Pd]2(p:-D)2 (2a) 53 3.2.4 [(dippp)Pd] 2(^-H) 2»LiBEt 4 (1) 53 3.2.5 [(dippp)Pd]2(p:-H)2*LiAlEu (3) 54 3.2.6 [(dippp)Pd]2(Li-H)2»NaBEu (4) 54 3.2.7 (dippp)Pd(ri2-H2C=CH2) (5) ...54 3.2.8 (dippp)Pd(PPh3) (7) 55 3.2.9 (dippp)Pd(DMAD) (8) 55 3.2.10 [(dippp)Ni]2(p>H)2 (6) 56 3.2.11 [(dippp)Ni]2(p>H)2»LiBEt4 (9) 56 3.2.12 [(dippp)Rh]2(pi-H)2»LiBEt4 (10) 56 3.3 References 57 List of figures Fig. 1.1: Structural configurations of transition-metal complexes with edge-bridging (H2) and face-bridging ({13) hydride ligands 5 Fig. 1.2: Metal hydride orbital's for M-H-M bridges and structural consequences 8 Fig. 1.3: Metal-hydride orbitals calculated for four-centre, four-electron bonds 8 Fig. 1.4: Donor-acceptor scheme in relation to \i2 (bridging) hydrides 9 Fig. 1.5: Reactivity of the rhodium hydride dimer series 11 Fig. 2.1: Catalytic cycle for nucleophilic attack at Pd(II) olefin complexes 18 Fig. 2.2: Chem 3D® core view (top) and ORTEP stereoview (bottom) of [(dipppPd)]20i-H)2-LiBEt4 (1) 20 Fig. 2.3: 300 MHz ! H NMR spectrum of [(dippp)Pd]2(H-H)2»LiBEt4 (1) 23 Fig. 2.4: Formation of [(dippp)Pd]2(rt-H)2*LiBEt4 (1) and [(dippp)Pd]2(H-H)2(2) 25 Fig. 2.5: Presumed mechanism of formation of [(dippp)Pd]2((i-H)2 (2) 26 Fig. 2.6: 300 MHz !H NMR spectrum of [(dippp)Pd]20i-H)2 (2) 27 Fig. 2.7: Newman projection along Pd-Pd axis; possible coordination geometries for [(dippp)Pd]2(|i-H>2 (2)... 28 Fig. 2.8: Proposed mechanism of hydride interchange for [(dippp)Pd]2(|i-H)2 (2) in solution 30 Fig. 2.9: Partner exchange observed in alkali tetraethylborate/aluminate adducts of [(dippp)Pd]2(|i-H)2 32 Fig. 2.10: Structure of Li4[(Me3P)3WH5]4 35 Fig. 2.11: Structural configurations in LiBMe4 35 Fig. 2.12: [(Et20)Na] 2[Et2Be 2]2(|l-H)2 37 v i i Fig. 2.13: 116.6 MHz ?Li NMR spectrum of [(dippp)Pd]2(u-H)2 (1), with selective proton irradiation 39 Fig. 2.14: Examples of transition-metal/Lewis-acid addticts 40 Fig. 2.15: Reactivity of [(dippp)Pd]2((J.-H)2 (2) 42 Fig. 2.16: 300 MHz ! H NMR spectrum of (dippp)Pd(Ti2-H2C=CH2) (5) 43 List of Tables Table 2.1: Palladium hydride specie s 16 Table 2.2: Selected intramolecular distances for [(dippp)Pd]2(M--H)2#LiBEt4 (1) 21 Table 2.3: Selected intramolecular angles for [(dippp)Pd]2(M--H)2*LiBEt4 (1) 22 ix Abbreviations A Angstrom. anal. analysis Bu" butyl, - C 4 H 9 . Bu' tertiary butyl, -C(CH 3) 3. °C degree Celsius calcd calculated. Cp ' cyclopentadienyl substituent, (C5H5)-. Cp* pentamethylcyclopentadienyl substituent, (CsMes)". 8 chemical shift (NMR). d doublet (NMR). dd doublet of doublets (NMR). dcypp l,3-bis(dicyclohexylphosphino)propane. dippp 1,3-bis(diisopropylphosphino)propane. DMAD dimethylacetylenedicarboxylate. dppe 1,2-bis(diphenylphosphino)ethane. dtbpm bis-(ditertiarybutylphosphino)methane. Et ethyl, - C 2 H 5 . Hz Hertz. IR infrared. / scalar nuclear spin-spin coupling constant (NMR). m multiplet (NMR). m meta. Me methyl, C H 3 . NMR Nuclear Magnetic Resonance. o ortho. p para. Ph phenyl. ppm parts per million. prZ isopropyl, -CH(CH 3) 2. quint quintet (NMR). R (alkyl) substituent. s singlet (NMR). sept septet (NMR). t triplet (NMR). Ti spin lattice relaxation time (NMR). THF tetrahydrofuran. Acknowledgements I would like to express my profound thanks to my supervisor, Prof. Michael Fryzuk, for the guidance, patience, and humour he has shown me throughout the course of this work. I would also like to thank past and present members of this research group, including Dr. Tim Haddad, Dr. Jesse Ng, Dr. Patrick Paglia, Bugger Moo, Mr. David McConville, and Ms. Lisa Rosenberg, for their friendship and advice. The staff of the glass workshop, Mr. Steve Rak, and Mr. Sean Adams, are to be thanked for their cheerful and speedy technical assistance. Also to be thanked is Mr. Peter Borda for performing the microanalyses, and Dr. Steve Rettig for determining the crystal structure. I would also like to say thanks very much to Clarko, Geoffrey Prenter, Dork, Rob, Doug, Crasher, Roy and H. G., and the State Bank and Channel Ten. G. K. B. C. Chapter 1: INTRODUCTION. 1.1 Preamble. The interaction of transition metals with dihydrogen has been the subject of much chemical research. The nineteenth century saw investigation into binary hydride phases of copper, iridium, palladium and platinum, and this and subsequent research has led to wide applications in surface chemistry and heterogeneous catalysis.1 In contrast it is only relatively recently that molecular complexes incorporating hydride ligands have been prepared and characterized. Undoubtedly research into this area was stymied until the development of spectroscopic techniques (lH NMR and IR spectroscopy, X-ray and neutron diffraction) which could adduce the nature of the transition-metal hydrogen interaction. The last few decades have witnessed a tremendous growth in the number and variety of transition-metal complexes containing hydrides. The hydride ligand can adopt a number of coordination geometries,. ranging from a terminal interaction, a metal hydrogen a bond, to edge-bridging (pL2) and face-bridging (JI3) hydrides which bridge across a number of metal nuclei. There are also complexes in which molecular hydrogen appears as a side bonded ligand.2"4 As a class, soluble metal-hydride derivatives have found wide applications in organic synthesis, homogeneous catalysis, and as models for heterogeneous, catalysis. The work described here deals with complexes of the lighter Group 10 metals of general formula [{R 2 P(CH 2 ), ,PR 2 }MH] 2 (M = Ni, Pd). While these complexes are of interest by themselves — they exhibit | i 2 bridging hydrides — it is their interaction with suitable derivatives of the alkali metals (M'ER4, M' = Li, Na; E = B, Al; R = Et) that form the basis of this work. To put this work in context the general chemistry of selected transition-metal hydride complexes will be examined. 1.2 Metal hydrides: 1.2.1 General. Historical background. The ability of transition-metals to occlude large volumes of dihydrogen was recognized in the nineteenth century. Several metallic and intermetallic phases involving chemisorbed hydrogen were investigated, of which the binary hydride, (3-PdH is one famous example.1 However, the appearance of well-characterized, molecular transition-metal hydride species was much more haphazard. In the 1930's the carbonylhydrido complexes, H2Fe(CO)4 and HCo(CO)4, were reported. In the late 1950's the isolation of a series of new hydride species, namely HRe(CsH5)2,5 H M ( C 5 H 5 ) ( C O ) 3 (M = Cr, Mo),6 and trans-HPt(PEt3)2Cl,7 signalled the start of active research in this area. Given the widespread application of transition metal complexes as homogeneous catalysts in the following years, and the suspected role of hydride intermediates, many such complexes were isolated and characterized. Early single crystal X-ray studies established that the hydride ligand was stereochemically non-innocent; but this work could only detect the hydride ligand by the presence of a vacant coordination site on the metal complex.8 It had been suggested originally that metal-hydride linkages were short — i.e. the hydride was separated from the metal by a distance not greater than the metal's covalent radius. On this basis H2Fe(CO) 4 and HCo(CO) 4 were analogues of Ni(CO)4, with "H2Fe" and "HCo" as virtual nickel nuclei.1 That the hydrogen atom was in the electronic cloud of the metal was also used to explain the up field position of hydride resonances observed in *H NMR spectroscopy. X-ray structural studies of K2ReH9 and RhH(CO)(PPh3)3, in which the hydrides were located, established that the metal-hydride linkage ranged from 1.6 to 1.7 A in length and were thus consistent with the sum of the covalent radii. Subsequent neutron diffraction experiments confirmed these results. The first example of a transition-metal complex containing a bridging hydride ligand was [CpW(CO)3]2(H)+.9 The bridging nature of the hydride was evidenced by its equivalent coupling to each 183\y nucleus. The first X-ray diffraction study of a transition-metal complex with bridging hydrides, performed on [HCr2(CO)io]", found an unusually long metal-metal separation of 3.41 A, but failed to locate the hydride.'8 This result was interpreted by proposing a linear M-H-M linkage. A subsequent neutron diffraction experiment revealed that the hydride was offset from the centre of the Cr-Cr axis by 0.3 A. 8 This and allied research established the bent nature of |i2 hydride ligands. Many other complexes bearing JJ.2, H3, and even interstitial hydrides at the centre of metal polyhedra, were isolated and characterized in the proceeding years. Bonding in these systems is akin to the electron-deficient bonding observed in the boranes. Quite recently, a series of complexes have been isolated in which molecular hydrogen acts as an rj2-bound ligand. 2 4 A side-bound dihydrogen ligand is intermediate between a cis-dihydride species, L n M H 2 , and a coordinatively unsaturated metal fragment, L„M. The interest in transition-metal hydrides continues, both as discrete molecular complexes and non-molecular, ternary phases. Aside from theoretical interest and their widespread implication in homogeneous catalysis, they have found some industrial application and have also been touted as advanced materials for hydrogen storage.1 Structures. Given the stereochemical activity of the hydride ligand, the coordination geometries of mononuclear hydride complexes are fairly predictable. The homoleptic complexes, [FeHg]4- and [PtrLi]2-, have octahedral and square planar geometries respectively;10 [ReHo]2- exists as a tri-capped trigonal prism.1 Where bulky ancillary ligands are present the exact position of the hydride ligand may become uncertain. This is the case for (Pl^P^RhH,11 in which the phosphine donors are tetrahedrally arrayed about the metal centre, and the coordination site of the hydride is unclear. In many instances the ancillary ligand is a bulky tertiary phosphine. The ability of such ligands to stabilize metal hydride complexes was recognized quite early.1,7 Bulky phosphine donors lend kinetic stability to the metal-core and prevent the formation of metal-metal bonds with the elimination of dihydrogen. The propensity of the hydride ligand to bridge between two or more metal centres results in a more varied coordination chemistry. The structural types that bridging hydrides can give rise to are summarized in Figure 1.1. Dimers of type'(/), with a single, unsupported H2 hydride bridge, are relatively rare. Up to four hydride ligands have been observed as bridges between two metal-nuclei. Literature examples of compounds bearing \i2 hydride ligands include [(dppe)PtH]2(p.-H)+, [(Cp*)2lrCl]2(p>H)2, {[(Ph3P)2lrH]2(Li-H)3}+, and [(Et2PhP)4Re]2(H-H)4.12 The hydride may also cap the face of a metal cluster (p.3); [CpNi]4(|i.3-H)3, structurally characterized by both X-ray and neutron diffraction,8 provides one example of this type of coordination (type v). In some instances the hydride may occupy an interstitial site; the structure of [HRu6(CO)i8]- exhibits an octahedral arrangement of metal atoms with the hydride located in the centre of the octahedron.1 Characterization The small X-ray scattering cross-sectional area of the hydrogen atom is well recognized, and scattering due to hydride ligands tends to be overwhelmed by that due to the transition metal to which it is bonded.8 Where hydrogen is bonded to two or 5 more metals, in a JJ.2 or jj.3 fashion, its contribution becomes an even smaller proportion of the overall diffraction pattern. H L „ M ™ ML„ L n M — — M L n H H H L»M4"^ ML" L"MIH|» HT H /// IV Fig. 1.1: Structural configurations of transition-metal complexes with edge-bridging (H2) and face-bridging (^3) hydride ligands. Early studies inferred the position of the hydride on the basis of the vacant coordination site, and it is only by using special techniques that hydrides may be located using single crystal X-ray diffraction.1 The application of neutron diffraction, in which scattering by the hydrogen atoms compares in magnitude to that by transition-metals, established unequivocally the range of metal hydride bond distances as 1.6 to 1.7 A for a terminal M-H bond. Neutron diffraction was also used to probe systems which included bridging hydride ligands.8 Structure determination by neutron diffraction is, however, non-routine, and in any case it is quite standard to undertake structure determination by X-ray crystallography prior to resorting to neutron diffraction methods. By contrast observation of hydride ligands by lH NMR spectroscopy is quite convenient and very informative.1 As the ancillary ligand is generally a phosphine, proton-phosphorus coupling can give rise to characteristic multiplets whose chemical shifts readily identify a hydride ligand. Hydride resonances usually appear upfield in the NMR spectrum due to paramagnetic shielding which results from the mixing between filled and unfilled metal orbitals.13 Coupling to NMR active transition-metal nuclei, such as 1 0 3 R h or 1 9 5 Pt, can be used to provide useful information on fluxional processes of transition-metal hydride complexes. Vibrational spectroscopy is also a characterization technique that is amenable to the synthetic chemist, especially for hydride complexes whose paramagnetism or insolubility precludes NMR spectroscopy. Terminal M-H bonds have stretching vibrations that occur between 2300 and 1600 cm"1. These may usually be identified by examining the IR spectrum of the corresponding metal-deuteride. Bridging hydrides in general give bands of less intensity and at much lower energy (1000-600 cm -1) where they may be masked by ligand vibrations.1 For these reasons IR spectroscopy applied to |i2 hydride complexes is not very diagnostic. 1.2.2 Theoretical considerations. Bonding description of \ii hydrides. Terminally bonded to a transition-metal the hydride ligand is unremarkable — it is a two-electron donor with a single negative charge. As such a M-H a bond is comparable to a metal-halide or a metal-alkyl bond, though its bond enthalpy is estimated to be much higher than that of the latter.12 The ability of hydride ligands to undergo multi-centre bonding requires a wider theoretical basis. The concept of electron-deficient, two-centre, three-electron bonding is familiar from the chemistry of the borane series. A bonding description of bridging M-H-M groups is briefly presented here. Unlike the F-H-F interaction in (HF)2,14 a M-H-M bond is inherently bent.8 Metal and hydrogen orbitals invoked for a M-H-M bond are depicted in Figure 1.2. Drawing from borane chemistry, such orbitals are designated as open (I), or closed (II). For this reason assignment of bond order between the metal nuclei becomes very much more of a formalism.1 The range of M-H-M bond angles for such complexes that have been unequivocally characterized is from 123 to 160°.1 Some degree of metal-metal interaction (II) may be inferred from these values: an open type of interaction (II) would give more obtuse angles.8 Another structural consequence of the metal-metal interaction is that a (J.2 hydride ligand lies in a position above the intersection of the vectors formed by the ancillary ligands (II). Typically a hydride bridge expands the metal-metal separation by 0.40 to 0.60 A . In general a M-H-M linkage is symmetrical, despite the off-axis location of the hydride ligand, although asymmetry can be induced if each metal has a different set of ancillary ligands.8 If the metal nuclei are bridged by two hydride ligands the situation changes entirely. In all known examples both hydrides approach the M-M vector more closely than does a single bridging hydride — i.e. for a [L„M]2(ji-H)2 complex each individual M-H-M angle is less acute than the corresponding angle in a [L„M]2(|i-H) system.8 A four-centre, four-electron bond is proposed to account for this phenomenon,15,16 as shown in Figure 1.3. The b2 u orbital results from the overlap of two metal d-orbitals, (d x y or d x z), with the two hydrogen atomic orbitals. The a g bonding orbital comprises a metal-metal interaction, overlapping both hydrogen orbitals. Thus both hydrides are closely constrained to the metal-metal axis. 8 Fig. 1.2: Metal hydride orbitals for M-H-M bridges and structural consequences. Fig. 1.3: Metal-hydride orbitals calculated for four-centre, four-electron bonds. A molecular orbital scheme has also been devised for hydride ligands coordinated in a face bridging mode.1 In such a metal cluster, a 113 hydride ligand is typically located 0.8 to 1.2 A away from a face; a four-centre, two-electron interaction is invoked but here the bonding description is less amenable to qualitative analysis. Donor-Acceptor formalism as applied to bridging hydride complexes. The synthesis and reactivity of many bridging hydride complexes may be rationalized by reference to a donor-acceptor formalism in which a M-H bond acts as a ligand towards an unsaturated metal complex.3,12 This is shown schematically in Figure 1.4. Fig. 1.4: Donor-acceptor scheme in relation to |I2 (bridging) hydrides. This formalism is familiar in terms of synergic bonding of rc-acid ligands, and can also be extended to other complexes which possess a-bonded groups as ligands. The build-up of electron density on the acceptor can be stabilized by back donation from the d-orbitals to the unfilled anti-bonding orbital of the M-H (or H-H, or C-H etc.) bond of the donor. Certain mixed metal dimers have been rationally synthesized by this route by adding a mononuclear cis-dihydride to a coordinatively unsaturated metal complex — notably those dimers in which CP2WH2 acts as the donor.12 It is also sometimes conceptually useful to consider a compound such as [L„M]2(M.-H)2 as deriving from a M(II) dihydride, L„MH2, and a M(0) fragment, L„M. 1.2.3 Summary of previous work. Previous work in this laboratory has established that certain hydride-bridged dimers of general formula [{R2P(CH2)nPR2}M]2( ri-H)2 (M = Rh) offer a wide spectrum of reactivity. The use of a bidentate, chelating phosphine is crucial in stabilizing these systems which are coordinatively unsaturated metal dimers of high reactivity. The steric and electronic properties of the ligand may be systematically varied by changing the alkyl group attached to the phosphorus and also by varying the chain length of the methylene backbone. Research by this group has found the use of the isopropyl group and chain lengths of two to four to be exceptionally versatile. This substitution guarantees a very electron-rich environment for the metal atom to which the ligand binds, and thus enhanced reactivity towards small molecules. Because of this reactivity, synthetic and reactivity studies are limited to reactions in inert solvents such as toluene and hexane. The rhodium complexes, [Pr'2P(CH2)„PPr'2Rh]2(|i-H)2 (n = 2, 3, 4), as 28 electron dimers, show remarkable reactivity, as shown in Figure 1 . 5 . 1 7 - 2 2 11 H H Rh Rh HV srRh ,Rh H Ph Rh Rh N / H .CH, V Rh Rh-V allene n = 2 1-alkene n = 2 r\ A /S H2C)„ Rh Rh (C (CH2)„ Zn(CH2Ph)2 n = 3 ' Zn Zn Rh Rh H H (H)R-/ \ Rh NRh REH (-H2) -butadiene = 2,3 (-Ha) n=2 E = 0, S,NR H Rh' NRh V Fig. 1.5: Reactivity of the rhodium hydride dimer series. Similar complexes have also been isolated for cobalt and these have been found relevant to the hydrogenation of arenes.23 An unusual mononuclear cobalt hydride complex, (dippp)CoH3, has also been isolated. The actual coordination of the hydrides in this complex has yet to be established, but low values of Ti suggest the possibility of an ri2-bound dihydrogen ligand.23 1.3 Aim and scope of this work. Given the interesting chemistry of the above rhodium and cobalt complexes it seemed natural to examine the chemistry of the analogous platinum group metal-complexes. Nickel and platinum hydride complexes are well represented in organometallic chemistry. For both metals there are well-characterized examples of phosphine stabilized, metal hydrides having mononuclear and dinuclear structures: for nickel, [(dcypp)Ni]2(|i-H)2;24 for platinum, (dcypp)PtH2 and [(dcypp)Pt]2(|i-H)2,25 [(dppe)Pt]2(|i-H)2,26 and cis-(dtbpm)PtH2.27 The corresponding palladium complexes are, however, conspicuously absent. A few hydrides of general formula trans-(R3P)2PdH2 exist, and these will be reviewed later, but no examples of higher complexity — for instance complexes bearing |i2-hydrides — have been isolated. This is somewhat surprising given the fact that homogeneous catalysis by low-valent palladium complexes and simple palladium salts is widespread in organic synthesis, and in many cases a palladium hydride intermediate is inferred. This work describes synthetic routes to [(dippp)Pd]2(|i-H)2, and the reactivity of this dinuclear complex towards organic substrates. A reaction with lithium tetraethylborate to form the unusual adduct, [(dippp)Pd]((l-H)2*LiBEt4, was discovered serendipitously by a co-worker. This will be also described and an attempt will be made to explain the bonding interaction in this neutral complex and to extend this type of interaction. 1.4 References. (1) Moore, D. S.; Robinson, S. D. Chem. Soc. Rev. 1983,12, 415-452. (2) Kubas, G. J. Acc. Chem. Res. 1988, 21, 120-128. (3) Crabtree, R. H.; Hamilton, D. G. Adv. Organomet. Chem. 1988, 28, 299-338. (4) Crabtree, R. H. Acc. Chem. Res. 1990, 23, 95-101. (5) Wilkinson, G.; Birmingham, J. M. / . Am. Chem. Soc. 1955, 77, 3421-3425. (6) Fischer, E. O. v.; Hafner, W.; Stahl, H. O. Z. Anorg. Allg. Chem. 1955,252, 47-62. (7) Chatt, J.; Duncanson, L. A.; Shaw, B. L. Proc. Chem. Soc. 1957, 343. 13 (8) Bau, R.; Teller, R. G.; Kirtley, S. W.; Koeztle, T. F. Acc. Chem. Res. 1979,12, 176-183. (9) Cotton, F. A.; Wilkinson, G. Advanced Inorganic Chemistry; Fifth ed.; John Wiley & Sons: New York, 1988, pp 1099. (10 (11 (12 (13 (14 (15 (16 (17 (18 (19 (20 (21 (22 (23 (24 (25 Cotton, F. A.; Wilkinson, G. op. cit., pp 1108-1109. Baker, R. W.; Pauling, P. J. Chem. Soc, Chem. Commun. 1969, 1495. Venanzi, L. M. Coord. Chem. Rev. 1982,43, 251-21A. Buckingham, A. D.; Stephens, P. J. / . Chem. Soc. 1964, 2747-2759. DeKock, R. L.; Gray, H. B. Chemical Structure and Bonding; Benjamin/Cummings: Menlo Park, California, 1980, pp 436. Broach, R. W.; Williams, J. M. Inorg. Chem. 1979,18, 314-319. Teller, R. G.; Williams, J. M.; Koetzle, T. F.; Burch, R. R.; Gavin, R. M. Inorg. Chem. 1981,20, 1806-1811. Fryzuk, M. D.; Jang, M.; Jones, T.; Einstein, F. W. B. Can. J. Chem. 1986, 64, 174-179. Fryzuk, M. D.; Piers, W. E.; Rettig, S. J.; Jones, T.; Einstein, F. W. B.; Albright, T. S. / . Am. Chem. Soc. 1989, 111, 5709-5721. Fryzuk, M. D.; Piers, W. E.; Einstein, F. W. B.; Jones, T. Can. J. Chem. 1989, 67, 883-896. Fryzuk, M. D.; McConville, D. H.; Rettig, S. J. Organometallics 1990, 9, 1359-1360. Fryzuk, M. D.; Piers, W. E. Organometallics 1990, 9, 986-998. Fryzuk, M. D.; Rosenberg, L.; Rettig, S. J. Organometallics 1991,10, 2537-2539. Ng, J. B. Ph. D. Thesis, University of British Columbia, 1990. Jonas, K.; Wilke; G.Angew. Chem., Int. Ed. Engl. 1970, 9, 312-313. Clark, H. C ; Smith, M. J. H. / . Am. Chem. Soc. 1986,108, 3829-3830. 14 (26) Carrnichael, D.; Hitchcock, P. B.; Nixon, J. F.; Pidcock, A. / . Chem. Soc, Chem. Commun. 1988, 1554-1556. (27) Hofmann, P.; Heiss, H.; Neiteler, P.; Muller, G.; Lachmann, J. Angew. Chem., Int. Ed. Engl. 1990, 29, 880-882. Chapter 2: HYDRIDES OF PALLADIUM. 2.1 Introduction. Soluble palladium complexes containing hydride ligands are surprisingly rare. In addition the few such species that have been reported in the literature are not well characterized. This literature will be reviewed in order to give some background to the work later described. The remainder of this chapter deals with the chemistry of the palladium hydride dimer, [(dippp)Pd]2(}J.-H)2. The synthesis and characterization of this complex is described, as well as its interaction with alkali metal salts to give the interesting adducts, [(dippp)Pd]2(H-H)2*M(ER4) (M = Li; E = B, Al; R = Et. M = Na; E = B; R = Et). The coordination chemistry of lithium and its interaction with transition-metals is relevant in this respect and will also be presented. 2.2 Survey of palladium hydrides. 2.2.1 Range. Compared to the number of hydride complexes characterized for nickel and platinum the range of palladium-hydride complexes is rather sparse. This is somewhat unexpected given the widespread application of palladium complexes in homogeneous catalysis and the implication of palladium hydride intermediates in such catalytic systems. The first palladium-hydride species to be isolated was trans-(Et3P) 2PdHCl, prepared by chlorine/hydrogen exchange between (Et3P)2PdCl2 and Me3GeH. 1 This and other palladium-hydride species have been listed in Table 2.1. 16 Table 2.1: Palladium hydride species. Complex Preparation | Ref. trans-(Et3P)2PdHX (X = CI, Br, I, GePh3) trans-(Et3P)2PdCl2 + Me 3GeH 4, 40*C, sealed tube trans-(Et3P)2PdHCl + Me 3GeCl 1,2* trans-(R3P)2PdHX (R = Cy, X = CI, Br, I, NCS) (R 3P) 2PdX 2 + (R 3P) 2NiH(BH4) i trans-(R3P)2PdHX + (R 3P) 2NiX 2 3 (R = Pr'3P, X = CI) [(dppe)PdHPR3]PF6 (R = Pr, Cy) trans-(Et3P)2PdHCl + dppe + N H 4 P F 6 I C 6 H 6 / M e O H [(dppe)PdHPR3]PF6 + NH4CI 4 trans-(Cy3P)2PdH2 Pd(acac)2 + 2 P C y 3 4, Et20, AlEt3 (Cy 3P) 2PdH 2 5 -GH2PR2 CH2PR2 R = Bu f .CH0PR0 ^CHoPRo / = / 2l NaBH 4 s — ( 2| 2 F>d-ci P d _ H CH2PR2 CH2PR2 6 trans-[(Cy3P)2PdHL]BPh4 (L = Py, pyrazole, imidazole) ( C y 3 P ) 2 P d H N 0 3 + HLBPh4 [(Cy3P)2PdHL]BPh4 + H N 0 3 7 trans-L2PdHCl (L = PCy3, PPr'3, Bu'2Bu"P, Bu'2MeP) [(C 8Hi 2OCH 3)PdCl] 2 + 4L 4. MeOH 2L2PdHCl + 2C 8 HnOCH 3 8 trans-[(Cy3P)2PdH(OAr)].ArOH (Ar = C 6 H 5 , C 6 F 5 ) (Cy3P)2Pd + ArOH 4, toluene [(Cy 3P)2PdH(OAr)] • ArOH 9*,10* trans-[(Cy 3P) 2PdH(H 20)]BF 4 (Cy3P)2Pd + H B F 4 l ¥ [(Cy 3P) 2PdH(H 20)]BF 4 11* * Denotes compound that has been structurally characterized by single crystal X-ray diffraction. Of those listed only four have been characterized by single crystal X-ray diffraction. Characterization of the remainder has depended upon lH NMR spectroscopy, microanalysis, and observation of the Pd-H stretching frequency between 1900 and 2100 cm"1. Most of the hydrides listed have been synthesized by quite unconventional methods. Synthetic routes to the hydrides of other transition metals commonly employ reduction by main-group hydrides (such as NaBELi or similar), or addition of H 2 . The ease of reduction of Pd(II), as compared to Ni(II) or Pt(II), may preclude these more conventional methods. Interestingly, there have been to date no isolated examples of palladium complexes bearing cis-dihydride ligands; nor has a dinuclear palladium complex bearing |i2-hydride ligands been unequivocally characterized prior to work in this laboratory. However, such species have been postulated on the basis of solution data.1 2'1 8 2.2.2 Palladium hydrides in catalysis. Simple palladium complexes have been used extensively to catalyze a wide variety of organic transformations. In many of the schemes devised a palladium hydride species is inferred near the end of the catalytic cycle. Typically, a step involving elimination of "HX" is required to regenerate the catalyst precursor. One important example which serves to illustrate the role of palladium hydrides in catalysis is the nucleophilic attack at Pd(II) olefin complexes.19 The catalytic cycle is shown in Figure 2.1. (by p-hydride elimination) Fig. 2.1: Catalytic cycle for nucleophilic attack at Pd(II) olefin complexes. The cycle consists of (a) coordination by the olefin, (b) attack of the nucleophile, (c) P-hydride elimination, (d) and finally elimination of "PdHCl". The nature of the PdHCl intermediate is unknown. Nevertheless this methodology is very versatile and much variation can be introduced into the cycle. For example if a Pd(0) complex is used an aryl (or a vinyl) halide may be added with the olefin. The aryl halide oxidatively adds to the metal centre and the aryl residue migrates to the olefin once it has become coordinated. Completion of this cycle thus results in an elaborated olefin. Similarly a palladium hydride species is active in other transformations which palladium salts catalyze,, for example: reduction; 2 0 - 2 2 transmetallation;19 and telomerization of conjugated dienes. 4 , 2 3 , 2 4 It is also notable that the industrially important Wacker process for the palladium-catalyzed oxidation of ethylene to acetaldehyde follows the basic scheme of Figure 2.1 — in this case the olefin is ethylene and the nucleophile is water. Precise mechanistic details of this process are as yet uncertain.25 2.3 Isolation and structure of [(dipppJPdhflt-Hh'LiBEt^ The long-standing interest of this group in the synthesis and reactivity of hydride-bridged transition-metal dimers was incentive enough to attempt the preparation of [(dippp)Pd]2(p>H)2. Synthesis of this compound would complete the triad of Group 10 metal dimers of general formula [{R2P(CH2)nPR 2}M]2((i-H)m. It was also hoped that the Group 10 metal chemistry could be developed to complement that of the Group 9, which was discussed in the first chapter. Early investigations conducted by Dr. Brian Lloyd, a former post-doctoral fellow of this group, involved the addition of two equivalents of a stock solution of LiBEt3H to a slurry of (dippp)PdCl2 in THF at low temperature. A deep red solution formed from which maroon crystals could be isolated. The crystals were structurally characterized by X-ray crystallography and found to be consistent with the formula [(dippp)Pd]2(H-H)2»LiBEt4 ( l) . 2 6 The inclusion of L i + into the coordination sphere was quite adventitious and results from the disproportionation of proprietary LiBEt3H to form LiBEt2H 2 and LiBEt4. 2 7 , 2 8 The structure of 1 is shown in Figure 2.2. The dinuclearity of 1 is clearly evident as are the bridging hydrides which were located and refined. Selected intramolecular distances appear in Table 2.2; intramolecular angles are in Table 2.3. Elementary electron counting schemes indicate the presence of a metal-metal bond. The separation between the palladium nuclei is 2.8233 A. Metal-metal distances in comparable dimers range from 2.531 to 2.699 A;29 a non-bonding separation is taken to be from 3.2 to 3.4A. In the free metal the separation between palladium nuclei is 2.76 A.30 Fig. 2.2: Chem 3D® core view (top) and Ortep stereoview (bottom) of [(dippp)Pd]2(n-H)2«LiBEt4 (1). 21 Table 2.2: Selected intramolecular distances for [(dippp)Pd]2(n-H)2*LiBEu (1). Atom Atom Distance/(A) Pd(l) Pd(2) 2.8233(6) Pd(l) H(l) 1.62(5) Pd(l) H(2) 1.62(4) Pd(l) P ( D 2.342(1) Pd(l) P(2) 2.338(1) Pd(l) Li 2.737(8) Pd(2) H(l) 1.79(5) Pd(2) H(2) 1.78(4) Pd(2) P(3) 2.334(1) Pd(2) P(4) 2.334(1) Pd(2) Li 2.639(8) Li B 2.26(1) Li H(l) 2.65(7) Li H(2) 2.33(5) Table 2.3: Selected intramolecular angles for [(dippp)Pd](u,-H)2»LiBEt4 (1). Atom Atom Atom Angle/C) H(l) Pd(l) H(2) 71(2) H(l) Pd(2) H(2) 64(2) P(l) Pd(l) P(2) 124.52(4) P(3) Pd(2) P(4) 100.51(5) Pd(l) Li Pd(2) 63.3(2) Pd(l) H(l) Pd(2) 112(3) Pd(l) H(2) Pd(2) 112(2) The long Pd-Pd distance in 1 is probably a reflection of the steric demand of the ligand at each end of the dimer. The coordination around each palladium nucleus is square-planar to a first approximation. The mean P-Pd-P bond angle is 100.2° and the mean H-Pd-H angle is 67.5°. That the latter is significantly more acute than the former points again to a metal-metal interaction (Fig. 1.3). The Pd2H2 core of the dimer is roughly planar with an angle of 7.7° between the planes formed by each of the palladium nuclei and the hydrides. The dihedral angles between each set of phosphorus atoms disposed transoidally across the Pd2Fi2Core are 17.5 and 18.5°. The lithium is perched atop the two palladium nuclei at a mean distance of 2.69 A from each metal centre; it is separated from the bridging hydrides by a mean distance of 2.45 A. The l H NMR spectrum of 1 in C6D6 is shown in Figure 2.3. Each resonance is considerably broad and some overlapping occurs. Peaks due to the tetraethylborate protons are observed at 1.29 (methyl) and 0.42 (methylene) ppm. The corresponding values for actual LiBEt* in the same solvent are 1.30 and 0.21 ppm. Surprisingly, variable temperature *H NMR spectroscopy could detect no inequivalence amongst the ethyl groups of the borate unit even at -90° C. The borate group must undergo a rapid tumbling motion interchanging all the methylene protons which are presumably interacting with the lithium atom. Resonances due to the ligand protons occur between 1.07 and 1.78 ppm. Also noteworthy is the broad resonance at -3.66 ppm due to the bridging hydrides. When the spectrum is acquired at low temperature this resonance resolves into a quintet (2/p-H = 35 Hz). The 3 1P{ JH} NMR spectrum exhibits a broad singlet at 22.4 ppm. 2.4 Routes to [(dippp)Pd] 2 (^-H) 2 . Clearly the questions of bonding of the LiBEt4 unit raised by 1 could not be approached without the parent hydride dimer, [(dippp)Pd]2(n-H)2. In fact examination of the reaction mixture leading to 1 by *H NMR spectroscopy showed that [(dippp)Pd]2(ri-H)2 was formed initially. Only after the THF was removed did coordination of the free LiBEty occur. The use of authentic L i B E t 3 H in THF could in fact give [(dippp)Pd]2(ri-H)2, however once the hydride is delivered from this reagent free BEt3 is present which forms a persistent adduct with THF. The [(dippp)Pd]2(|i-H) 2 complex could be fractionally crystallized from pentane and separated from the THF/BEt3 adduct but the use of crystalline N a B M e 3 H or K B E t 3 H was preferable. These hydride transfer reagents may be deployed in toluene, and the liberated BEt3 may be pumped off. However, when a slurry of (dippp)PdCl2 in toluene was treated with K B E t 3 H unidentified reduction products were obtained. In order to enhance the solubility of (dippp)PdCl2, the complex was metathesized with two equivalents of NaI.2H20 in acetone to give (dippp)Pdi2. Subsequent treatment of the diiodide with KBEt3H in toluene did in fact give the desired hydride as shown in Figure 2.4. BEt2H2, LiBEt4 > " \ THF, -50° C * y CI ("LiCI, -BEt2CI) (-1/2H2) Li 'Pr, P-\ Pd--- -Pd > >P 'Pr P-'Pr? LiBEt4, Toluene 2KBEt3H Toluene, -50° C I (-2KI), (-1/2H2) - ( ) -CP. r \ ^ / Pd Pd Fig. 2.4: Formation of [(dippp)Pd]2(H-H)2«LiBEt4 (1) and [(dippp)Pd]2ai-H)2(2). Within ten minutes of addition of the KBEt3H to the slurry of Pd(dippp)I2 at -50° C, a deep red colour develops and the reaction mixture was seen to effervesce. When the lH NMR spectrum of the crude reaction mixture was sampled, a characteristic quintet was observed upfield at -2.52 ppm; 3 1P{ 1H} NMR spectroscopy showed only a singlet at 27.0 ppm, indicative of a single product. The yields obtained in these reactions were not outstanding (45-50%); the Celite was lined with a black, unidentified residue (presumably reduction products) after the solution was filtered. A plausible mechanism for the reaction is shown in Figure 2.5. Fig. 2.5: Presumed mechanism of formation of [(dippp)Pd]2(|i-H)2 (2). The presumed Pd(II) dihydride intermediate, (dippp)PdH.2, could not be observed when the reaction was monitored by low temperature 3 1 P NMR spectroscopy. b b' i i i 1 i i — i i j i — i — r — i — | — r — i — i — i — | — i — i i i | — r — i — i i j — i — i — i — i — | — i — i — i — i — i — r — i — i — r — i — r — i — i — i — i — [ — i — i — i — i — r ~ l — i — i — i — | — r 2 1 . 0 - 1 , - 2 - 3 P P M Fig. 2.6: Room temperature, 300 MHz lR NMR spectrum of [(dippp)Pd]2(H-H)2 (2) in C(,D(,. to 28 <9, C0, H' —\^ —c ^y- ^ c r ^ Y D2h D2h (4. = 0°, e = 9 0 ° ) (<)) = 0 ° , e = o*) D 2 (<|> * 0 ° , 6 = 90 - <t>/2°) Fig. 2.7: Newman projection along Pd-Pd axis; possible coordination geometries for [(dippp)Pd]2(H-H)2 (2). The *H NMR of 2 is shown in Figure 2.6. The most important signal in the spectrum is the binomial quintet observed at -2.52 ppm ( 2/p-H = 35 Hz). The possible structures for [(dippp)Pd]2(|i-H)2 are given in Figure 2.7: coordination geometries are defined by (J), the angle between the two coordination planes (P-M-P'), and 8, the angle formed between the plane of the hydrides and each coordination plane. Now if the dimer consists of two edge-sharing tetrahedra (0 = 90°, <|> = 0°) then the hydrides are magnetically equivalent and therefore consistent with the *H NMR spectrum. Other values of (|> render the hydrides magnetically inequivalent and should hence give rise to a second-order spectrum. However, the formally related d 9 nickel dimer, [(dcypp)Ni]2(|i-H)2, has in fact a structure based on D2 symmetry (0 = 31.7°, <j) = 116.7"),31 and yet the hydrides appear as a simple quintet in the ! H NMR spectrum.32 A molecular orbital analysis of the model complex [(H3P)2Ni]2((i-H)2 concluded that the energy minimum occurred for the square planar structure (D2h; 0 = 0°, <|> = 0°) and the observed departure from planarity in [(dcypp)Ni]2((i-H)2 was due to steric repulsion between the cyclohexyl groups on alternate ligands.31 For the 28 electron dimer [{(Pr'0)3P}2Rh]2(|i-H)2, the preference for planarity is even more pronounced, and the metal nuclei, the donor atoms, and the bridging-hydrides all lie in the same plane.33 Moreover, the value of <}> in the structure of 1, the LiBEty adduct, is also non-zero (0 = 17.5°, <|) = 145°). On this basis it is reasonable to expect similar structural features in [(dippp)Pd]2(|J.-H)2, especially as the longer metal-metal separation would tend to relieve the steric crowding. How then to explain the lH NMR spectrum? In solution a mechanism must exist whereby the hydrides are interchanged and coupling to each phosphorus nucleus becomes equivalent. Such a mechanism would unlikely involve dissociation of the phosphines or hydrides at a rate faster than the NMR time-scale. A more plausible mechanism is rotation of the (u,-H)2 unit around the Pd-Pd axis. Such a mechanism is depicted in Figure 2.8. Hydride equivalence can thus be achieved by a degenerate pair of M-H-M orbitals symmetrically disposed around the Pd-Pd axis.34 In any case this whole question must remain unresolved until a solid state structure of [(dippp)Pd]2(|i.-H)2 is obtained. Attempts to grow X-ray quality crystals have so far been unsuccessful. YY YY YY YY vw YY YY Fig. 2.8: Proposed mechanism of hydride interchange for [(dippp)Pd]2(p>H)2 (2) in solution. The remainder of the NMR spectrum is not very informative. Coincidence of the oc-methylene protons of the ligand backbone with the diastereotopic methyl protons of the alkyl groups attached to each phosphorus atom does not allow measurement of coupling constants between adjacent protons or phosphorus nuclei. The (3-methylene protons of the backbone are also superimposed upon the methine protons of the isopropyl groups. 2.5 Interaction of Palladium hydride dimers with LiBEt4 and NaBEt4. Given pure [(dippp)Pd]2(!t-H)2 and LiBEt4, we were in a position to examine the interaction of the two. Addition of one equivalent of LiBEt4 to a solution of [(dippp)Pd]2((J.-H)2 in toluene results in a decrease in the intensity of the red colour of the solution, and the adduct, [(dippp)Pd]2(n.-H)2]»LiBEt4 (1), may be isolated in over 90% yield. Likewise LiAlEt4 may be substituted to give the corresponding aluminate adduct, [(dippp)Pd]2(iu-H)2]»LiAlEt4 (3). The affinity between the lithium nucleus and the palladium hydride complex is quite profound. When the reaction between [(dippp)Pd]2(|J.-H)2 (2), and LiBEt4 was monitored by variable temperature 3 1P{ 1H} NMR, only a broad signal at 22.2 ppm, corresponding to 1, [(dippp)Pd]2(P--H)2*LiBEt4, is observed (-80° to 80° C). Similarly when the variable temperature 3 1P{ ^^H} NMR spectrum is taken of the pure adduct only the 22.2 ppm resonance is seen. Mixed solutions of [(dippp)Pd]2(fi-H)2 and [(dippp)Pd] 2(fi-H) 2»LiBEt4 show both characteristic resonances, 27.0 and 22.2 ppm respectively, but the measured integrals give only the ratio in which the two complexes were mixed initially. Exchange of LiBEt4 between the adduct and bare [(dippp)Pd]2(}i-H)2 was evidenced by the broadening of each resonance and ultimate coalescence at elevated temperatures (40° C). The equilibrium constant for the reaction could therefore not be measured by these means and is estimated to be very high. Adduct formation between LiBEt4 and 1 is therefore thermodynamically favoured, but the adduct is kinetically labile. The affinity of the palladium hydride dimer towards the tetraethylborate salt of another alkali-metal was also examined. Addition of one equivalent of NaBEt4 to a solution of [(dippp)Pd]2(iU-H)2 in toluene imparts a bottle-green colour to the solution. Recrystallization from toluene/pentane gives bright-blue crystals formulated as [(dippp)Pd] 2(|i-H) 2»NaBEt4 (4). The gross details of the lR NMR spectrum of 4 are similar to those of 1, and the 3 1P{ 1H} NMR spectrum features a singlet at 25.9 ppm. In the *H NMR spectrum at room temperature the bridging hydrides may be clearly resolved into a quintet (8 = -3.43 ppm; /p_H = 35.4 Hz). While structural characterization of this complex has yet to be accomplished, the fact that the sodium analogue of 1 can be isolated belies arguments based upon the covalent character of the alkali metal; sodium is expected to have an overwhelming ionic character. J, K A K Fig. 2.9: Partner exchange observed in alkali tetraethylborate/aluminate adducts of [(dippp)Pd]2(u-H)2. The relative affinities of Na+ and Li+ for [(dippp)Pd]2(|i-H)2 (1) were examined by performing what are effectively partner-exchange reactions in benzene or toluene, as shown in Figure 2.9. Each adduct could be distinguished by the chemical shift of its 3 1P{ lR} NMR signal. Addition of one equivalent of LiBEt4 to 4 results in the displacement of Na to give 1; addition of excess NaBEtt (five equivalents) to 1, results in formation of 4, although some of 1 remains (roughly one third as determined by 3 1P{ 1H} NMR spectroscopy). As expected, the smaller, more polarizing Li nucleus (from either an ionic or a covalent standpoint) exerts greater attraction for the dimer. Estimates of the affinity of each alkali-metal for the Pd2H 2 core remain qualitative at this time owing to the difficulty of dissolving the alkali-metal salts in benzene or toluene. Addition of excess NaBEt4 (five equivalents) to 3, [(dippp)Pd]2(|J.-H)2,LiAlEt4, results in the formation of 4, and also 1, in roughly a 3:1 ratio. That an exchange could be effected between the tetraethylaluminate and tetraethylborate anions, suggests the existence of a {[(dippp)Pd(jj. -Ff)2Pd(dippp)]Li}+ species in solution. When these adducts are treated with a suitable donor (ethylene, PPh3, or DMAD) the dimer undergoes reductive elimination to give tricoordinate Pd(0) species. The presence of LiBEt4.does not detract from the reactivity of the metal-hydride core. 2.6 Bonding considerations in [(dippp)Pd]2(|i-H)2*LiBEt4. While the interaction of L i + with transition-metals does have some precedent the association of L i + with the metal core in 1 is nonetheless intriguing. As a digression the structures of some other complexes in which an alkali-metal is associated with a transition-metal will be reviewed. The coordination chemistry of lithium with main-group compounds is also clearly relevant here, and the structures of related main-group compounds will be examined. Finally, in order to complete this picture, the concept of metal-basicity will be introduced and applied to 1. 2.6.1 Lithium in association with transition-metals. There exist many examples of anionic transition-metal hydrides or anionic complexes containing electron-rich rc-donors as ligands in which the cation is an alkali metal. In some cases the ion-pair contact is so close that covalent interactions have been invoked between the transition-metal nuclei and the alkali-metal centre. For example, by the treatment of (ri5-C5H5)2MH2 (M = Mo, W) with Bu"Li, tetrameric Li4[(T| 5-C5H5)2MH]4 is obtained.35 The structure features an eight-membered metallacycle composed of M - L i units, with a separation of 2.77 A.36 The lithium is bicoordinate and a L i - M covalent bond is inferred. Some degree of interaction has also been inferred between the lithium and the transition-metal centre in certain mixed-metal organometallic complexes. Examples include [Li(THF)2][Co(COD)2], [Li 2 (TMEDA)] 2 [Ni(norbornene)]2 and [ L i 2 ( T M E D A ) ] 2 [ F e ( C 2 H 4 ) 4 ] . 3 7 Similar considerations apply to certain alkali metal salts (Na, K) of transition-metal polyhydrides. These include K [ f a c - ( M e 2 P h P ) 3 0 s H 3 ] , 3 8 [K(18-crown-6 ) ] [ ( M e 3 P ) 3 W H 5 ] , 3 9 and [K(18-crown-6)][fac-(Ph3P)2Ph2RuH3] and its ortho-metallated product, K[(Ph3P)2(Ph2PC6H4)RuH2].40 The bonding between anion and cation has been described as an intimate ion-pair. In all cases the interaction is mediated by the hydride ligands. Distances between the alkali metal and the hydrides range from 2.52 to 2.78 A. An example in which an interaction between lithium and bridging hydrides is quite explicit occurs in Li4[(Me3P)3WH5]4, which is shown in Figure 2.10.41 Here the lithium-hydrogen contacts are as low as 1.73 A, and the transition-metal atoms are connected by one W(|i3-H)Li2 bridge, and two W(|i2-H)Li bridges. Due to the comparative shortness of the L i - H separation a multi-centre covalent interaction may be argued more convincingly. L = PMe 3 Fig. 2.10: Structure of Li4[(Me3P)3WH5]4. 2.6.2 Structures of LiBMe4 and related compounds. The structures observed in the allied molecules, LiAlEt4 and LiBMe4, are also clearly relevant to the structure of 1. The LiBMe4 molecule provides an important comparison because it is has been structurally characterized by neutron diffraction and the positions of the hydrogen atoms are known to a relatively high degree of certainty.42 The structure of LiBMe4 features Li-H-C interactions as shown in Figure 2.11. Each lithium atom participates in two dibridged interactions with one tetramethylborate residue, and two tribridged interactions with two other, distinct residues. Li Li i. Dibridged ii. Tribridged Fig. 2.11: Structural configurations in LiBMe4. Similar interactions of the dibridged type may also be inferred in the X-ray structure of LiAlEt4. It was originally proposed that these interactions are the result of the attraction between the Lewis-acidic lithium centre and the weakly basic hydrogen atoms.42 Similar arguments have also been advanced to explain the coordinative saturation of transition-metal nuclei in complexes of the pyrazolylborates 4 3 A more recent appraisal of bonding in organolithium compounds tends to discount the lithium nucleus as a participant in covalent bonding (vide infra), but the argument is valid even at the limit of ion-dipole bonding. In [(dippp)Pd]2(ri-H)2 ,LiBEt4, Li-H-C, dibridged interactions with the ce-methylene protons of the tetraethylborate unit are undoubtedly present. (The non-hydride hydrogens in the structure, Fig. 2.2, are interpolated from standard C/H parameters and cannot be used as a basis for argument.) From the structure these carbon atoms are separated from the lithium by distances of 2.32 and 2.35 A. The equivalent distances in the structures of LiBMe4 and LiAlEt4 are 2.36 and 2.37 A respectively. A similar interaction between the lithium and the hydrides attached to the palladium nuclei in 1 can be advanced as a further stabilizing factor. Certainly the concept of a metal hydride ligand acting as a Lewis base is well established;44 when the Lewis acid is a •coordinatively unsaturated transition-metal centre, the resultant complex is a hydride-bridged dimer (Fig. 1.4). The Lewis basicity of a-bonded groups is also a well established phenomenon and several transition-metal complexes have been isolated where a H-H bond, or a C-H bond acts as a ligand.45 The ability of an alkali metal to act as a corresponding acceptor is much more uncertain. The most recent, scholarly reviews of X-ray structures of organoalkali compounds hold that the observed bonding between lithium and other atoms is largely ion-dipole in origin and that coordination around the lithium nucleus is largely a matter of sterics 4 6 , 4 7 The reviewers cite several molecular orbital calculations that have been performed on organolithium species of known structure. The inclusion of lithium as a point-charge can approximate experimental parameters. If lithium is treated as a net acceptor of electrons, that is as a participant in covalent bonding, quite different geometries are generated. The dipole moment of the C-Li bond is also bought forward as evidence — 6.0 Debye for methyl lithium versus 1.85 for C H 3 F . This is to be compared with the dipole moment of 9.0 Debye estimated for a purely ionic interaction. In this case it is quite incorrect to speak of L i + as a participant in covalent bonding, inasmuch as it accepts only a small fraction of electron density. The marginal degree of covalency for lithium explains the occurrence of 1 3 C -^Li coupling in NMR spectroscopy. Traditional arguments to explain the solubility of Me Li versus MeNa relied on contrasting the covalent character of the former as opposed to the ionic character of the latter. The current argument draws attention to a periodic trend: larger ions such as N a + and K + are not as effectively encapsulated by the alkyl residues, and larger insoluble crystal lattices are deposited. Another parallel to the structure of 1 is found in a main group hydride, [Et2Be2]2(M.-H)2*Na2(Et20)2,48 which is represented in Figure 2.12. H 3 C C H 3 Fig. 2.12: [(Et20)Na]2[Et2Be2]2(li-H)2. The distance between each equivalent sodium atom and each hydride is 2.40 A. In 1 the distance between the lone lithium atom and the hydrides is 2.45 A. In [(Et20)Nah[Et2Be2]2(M--H)2 the interaction between the Na nuclei and the hydrides is surely ion-dipole (again transmitted through the hydrides), and the comparable distances between this dimer and 1 do not support the idea of covalent bonding between Li and the hydrides in 1. The lithium nucleus should be able to approach the hydrides more closely than sodium due to its smaller size. 2.6.3 7 L i NMR Spectrum of [(dippp)Pd] 2(|i-H)2«LiBEt 4 (1). In order to assess the degree of interaction between the Li and the hydrides in 1 the 7 L i NMR spectrum of the complex was examined. It features a broad singlet at -3.9 ppm. When the protons are successively decoupled the signal sharpens appreciably when the bridging hydrides are irradiated as shown in Figure 2.13. Because of the broadness of lines in the 7 L i NMR spectrum the interaction between Li and the hydrides could not be demonstrated unequivocally. 2.6.4 Metal basicity. It must be stressed that in almost all of the examples quoted above the interaction between the alkali-metal and the transition-metal complex is the result of ionic bonding between a cation and an anion of large electrostatic potential. Comparison to such complexes as [(dippp)Pd]2(|i-H)2»LiBEt4 is inappropriate: the hydrides in 1, 3, and 4 are bonded in a (12 fashion, and, more importantly, the metal core is not anionic. However, there are other classes of transition-metal compounds which can be regarded as the adducts of an electron-rich transition-metal complex coordinatively bonded to some Lewis-acid. Examples of this type of compound are depicted in Figure 2.14. i r i I i i i i i i i i i i 11 i i I i i i i i i i i i j i i 11 i i i i i I i i i i i i i i i 1 11 i i i ) i i i i 11 i i i i i i i i i i i i i i i i 30 20 10 0 -10 -20 -30 HZ Spectrum Position of irradiation *H NMR chemical shift/ ppm Peak width at half-height/ Hz A proton decoupled 10.3 B P{C//(CH 3) 2} 2 1.79-1.20 10.3 C BC//2CH3 0.38 8.2 D 0 # ) 2 -3.66 4.9 Fig. 2.13: Room temperature, 116.6 MHz 7 L i NMR spectrum of [(dippp)]2((J--H)2»LiBEt4 (1) in C^D^, with selective proton irradiation. Fig. 2.14: Examples of transition-metal/Lewis-acid adducts. The d 2 metal complexes [CP2MH2] (M = Mo, W) will form adducts with Al2Me6 and the boron halides; 4 9- 5 0 certain transition-metal carbonyls will form adducts with H g C h ; 5 1 Vaska type complexes, trans-[L2MX(CO)] (M = Rh, Ir; X = CI, Br, I, N3; L = R3P, R3AS) will form the corresponding adducts of B X 3 and SO2. 5 2 This type of interaction illustrates a phenomenon that Shriver has formalized as 'metal-basicity',51 in which a transition-metal complex serves as a donor towards a Lewis-acid species. The concept has been extended by Werner to include the reactions of electron-rich half-sandwich compounds.5 2 For the tungsten dihydride adduct, [Cp 2WH2] #AlMe3, more recent low-temperature X-ray diffraction studies show that the aluminum binds through (0-2 hydrides rather than by a dative bond from the metal, as shown in Figure 2.14,53 however there are other examples where a dative covalent bond is more explicit. Examples of these latter include the mercuric chloride adducts of [CpM(CO)2] (M = Co, Rh), 5 2 and the alkyl aluminum adducts of (R3P) 2RhCp. 5 4 In the light of these examples, another possible rationale for the bonding in 1 is by way of a dative bond between the transition-metal complex and the LiBEt4 entity. The Lewis-acid character of LiBEt4 has not been noted previously. Whether electron density is donated by orbitals centred on the palladium nuclei or from the metal hydride bonds is unclear. As mentioned, the Pd-Li contacts had an average value of 2.68 A. This is to be compared with an arbitrary value of 2.90 A from the sum of the metallic radii, 3 0 and also with an actual value of 2.64 A observed in the ternary hydride species, Li2PdH2.55 Both of these distances are greater than the interatomic distance found in ionic LiH, reported as 2.04 A.56 Interestingly, when one half of one equivalent of Al2Me6 was added to a solution of [(dippp)Pd]2(|l-H)2, a new resonance was observed at 20.8 ppm in the 31P{1H} NMR spectrum. In the *H NMR spectrum a new hydride signal appeared as a quintet at -3.78 ppm ( 2 / p - H = 28 Hz). This complex was not further characterized but it is presumed to be the corresponding AlMe3 adduct, [(dippp)Pd]2(|i.-H)2#AlMe3. Thus [(dippp)Pd]2(|i-H)2 will form a more conventional type of metal-base, Lewis acid adduct. Whether Al2Me6 will displace LiBEt4 from 1 is as yet unknown. 2.7 Further chemistry of [(dippp)Pd]2(|i-H)2. The palladium hydride dimer is also found to undergo very facile reductive elimination with the standard range of donor species. Addition of excess donor results in complete discharge of colour and colourless crystals are obtained after recrystallization. These reactions are represented in Figure 2.15. Reaction with ethylene, for instance, gives (dippp)Pd(T|2-H2C=CH2), (5). The lH NMR spectrum of 5 is shown in Figure 2.16. The ethylene protons appear as a doublet due to coupling with each trans phosphorus atom. Fig. 2.15: Reactivity of [(dippp)Pd]2(u.-H)2 (2). When a solution of 1 or 2 is placed under an atmosphere of deuterium the bridging hydrogen atoms are replaced by deuterium as shown. The dimer is, however, quite stable and will not undergo oxidative addition with H2 or D2 to give (dippp)PdH2 or a higher homologue of [(dippp)Pd]2(M--H)2. In fact when a solution of 1 was placed under 100 atmospheres of dihydrogen in a sapphire NMR tube the dimer remained intact and a Pd(II) species was not observed. 2.8 Interaction of LiBEt4 with other transition-metals. 44 An obvious extension of this research was to examine the behaviour of other transition-metal complexes with LiBEt4. The first examined was nickel, for which, using the same procedure as for 2, the analogous dimer, [(dippp)Ni]((i-H)2 (6), was isolated and characterized. This gives a singlet in the 3 1P{1H} NMR spectrum at 36.3 ppm and bridging hydrides appear as a quintet at -10.79 (2/p-H = 18.5 Hz) in the lH NMR spectrum. Addition of a stoichiometric amount of LiBEt4 to 6 resulted in a very large broadening of the 3 ^^H} NMR signal. This was attributed to the formation of an equilibrium mixture of [(dippp)Ni]2(|i-H)2»LiBEt4 (9) and the parent dimer, 6. The adduct was not isolated in analytically pure form; however, in solution when an excess of LiBEt4 was added to 6 (greater than 5 equivalents), a fairly sharp singlet in the 3 1 P{ 1 H) NMR spectrum was observed at 31.9 ppm. Low temperature 3 1P{ 1H} NMR experiments gave a sharp singlet at 36.3 ppm at -80° C, followed by progressive broadening of the signal as the sample was warmed up to room temperature. This indicates that adduct formation between [(dippp)Ni]2(|J.-H)2 and LiBEt4 is endothermic. Two singlets were never distinguished in the 3 1 P NMR spectrum. The upfield shift, AS, of 4.4 ppm observed in the 31P{!H} spectra of the adduct and free dimer is equivalent to that observed for the formation of [(dippp)Pd]2(ri-H)2,LiBEt4, AS = 5.0 ppm. Interestingly, the hydrides in [(dippp)Ni]2(ji-H)2»LiBEt4 do not shift upfield in the *H NMR spectrum, but appear as a broad singlet with no coupling to phosphorus discernible. The reactivity of [(dippp)Rh]2(|i-H)2 towards LiBEt4 was also examined. Adduct formation was again evidenced by the 3 1P{ 1H} NMR spectrum. The starting material has a 3 1 P NMR chemical shift of 50.5 ppm (/Rh-p = 160 Hz); after LiBEt4 was added a clearly distinct resonance was observed at 47.5 ppm (^ Rh-p = 155 Hz). In the lH NMR spectrum two sets of hydride resonances were apparent: (i) a multiplet at -6.65 ppm, corresponding to [(dippp)Rh]2((i-H)2 (10); (ii) a multiplet at -9.50 ppm, due to [(dippp)Rh] 2(u-H)2»LiBEt 4. That [(dippp)Rh]2(M.-H)2 should undergo adduct formation was not unexpected because the steric properties of this complex would be almost identical to to those of [(dippp)Pd]2(^-H)2. Attempts to grow crystals from the reaction mixture resulted in isolation of the starting material rather than the adduct, due to the former's greater insolubility. 2.9 Summary and future prospects. The chemistry of [(dippp)Pd]2(M--H)2 was examined with a view to understanding its interaction with organic substrates. It would appear that given the correct ligand environment, in this case bulky, chelating phosphines, palladium hydride species may be accessed using fairly standard procedures. The absence of a cis-dihydride monomer is provocative. It is tempting to speculate that the absence of cis-dihydride complexes of palladium and nickel is due to the thermodynamic stability of Tj2-dihydrogen coordination, which is well represented for these first two members of Group 10,5 7*5 8 over a cis-dihydride mode. Certainly for the platinum dimers discussed in the first chapter there is evidence of an equilibrium between r|2-dihydrogen and cis-dihydride modes of coordination.59 In the present system, varying the bite-angle of the ligand (by varying the length of the methylene backbone) might give interesting results. The reactivity of 2 towards small organic molecules is disappointing. The dimer is reduced directly to a Pd(0) species without observation of any dinuclear intermediates. Nevertheless, [(dippp)Pd]2(|i-H)2 is a convenient source of Pd(0) fragments. ' What is clearly the most important feature of the chemistry of this complex — its ability to undergo adduct formation with tetraethylborate salts of lithium and sodium — was discovered by accident. This type of interaction was shown to have some generality among late transition-metal hydrides and obviously wider investigation is in order with bridging and terminal hydride complexes. The question of the actual interaction between the alkali metal ion and the transition-metal core remains unresolved and is an impetus to further study. Of all the bonding types in which a transition-metal is known to participate, a dative bond to the lithium nucleus is the category which can best describe the bonding in 1. Given that an interaction can be demonstrated between the lithium nucleus and the hydrides in the 7 L i NMR spectrum, it is reasonable to" assume that the hydrides are involved in binding the lithium nucleus and that a classical dative bond between lithium and palladium is not solely responsible for adduct formation. The interaction of [(dippp)Pd]2(|i-H)2 with AgBEt4 might prove informative, given the spin activity of 1 0 7 A g and 1 0 9 A g (51.8% and 48.2% respective abundances; I = 1/2). Also important is to learn whether the archetypal Lewis-acid, H + , in a form soluble in toluene, will bind to the dimer as well. Whether the bonding in 1 can be described as covalent, or as the result of ion-dipole attraction between L i + and the induced dipoles of the palladium-hydride bonds is again uncertain. The degree of charge separation within the complex is hard to assess but it nevertheless must be present if the lithium nucleus is predominately ionic. 2.10 References. (1) Brooks, E. H.; Glockling, F. / . Chem. Soc. (A) 1967, 1030-1034. (2) Brooks, E. H.; Glockling, F. / . Chem. Soc. (A) 1966, 1241-1243. (3) Green, M. L. H.; Munakata, H.; Saito, T. / . Chem. Soc. (A) 1971, 469-474. (4) Green, M. L. H.; Munakata, H. / . Chem. Soc, Chem. Commun. 1971, 549. (5) Kudo, K.; ffidai, M.; Uchida, Y. / . Organomet. Chem. 1973,56, 413-418. (6) Moulton, C. J.; Shaw, B. L. J. Chem. Soc, Dalton Trans. 1976, 1020-1024. (7) Moriyama, H.; Saito, T.; Sasaki, Y. Chem. Lett. 1976, 2, 175. 47 (8) Goel, A. B.; Goel, S. Inorg. Chim. Acta 1980, 45, L85-L86. (9) Braga, D.; P, S.; Di Bugno, C ; P, L.; M, P. / . Organomet. Chem. 1987,334, C46-C48. (10 (11 (12 (13 (14 (15 (16 (17 (18 (19 (20 (21. (22 (23 (24 Di Bugno, C.; Pasquali, M ; Leoni, P.; Sabatino, P.; Braga, D. Inorg. Chem. 1989, 28, 1390-1394. Leoni, P.; Sommovigo, M.; Pasquali, M.; Midollini, S.; Braga, D.; Sabatino, P. Organometallics 1991,10, 1038-1044. Schunn, R. A. Inorg. Chem. 1976,15, 208-212. Kellenberger, B.; Young, S. J.; Stille, J. K. / . Am. Chem. Soc. 1985,107, 6105-6107. Zudin, V. D.; Chinakov, V. D.; Nekipelov, V. M.; Likholobov, V. A.; Ermakov, Y. I. J. Organomet. Chem. 1985,289, 425-430. Young, S. J.; Kellenberger, B.; Reibenspies, J. H.; Hirnmel, S. E.; Manning, M.; Anderson, O. P.; Stille, J. K. / . Am. Chem. Soc. 1988,110, 5744-5753. Taqui Khan, M. M.; Taqui Khan, B.; Begum, S. / . Mol. Catal. 1988, 45, 305-307. Vijay Sen Reddy, V. J. Mol. Catal. 1988,45, 85-90. Kirss, R. U.; Eisenberg, R. Inorg. Chem. 1989,28, 3372-3378. Hegedus, L. S. Tetrahedron 1984,40, 2415-2434. Helquist, P. Tetrahedron Lett. 1978,22, 1913-1914. Zask, A.; Helquist, P. / . Org. Chem. 1978, 43, 1619-1620. Oshima, M.; Shimizu, I.; Yamamoto, A.; Ozawa, F. Organometallics 1991,10, 1221-1223. Tsuji, J. Organic Synthesis with Palladium Complexes; Springer-Verlag: Berlin, 1980. Benn, R.; Jolly, P. W.; Mynott, R.; Schenker, G. Organometallics 1985, 4, 1136-1138. (25) Collman, J. P.; Hegedus, L. S.; Norton, J. R.; Finke, R. J. Principles and Applications of Organotransition Metal Chemistry; University Science Books: Mill Valley, CA, 1987, pp 412-415. (26) Fryzuk, M. D.; Lloyd, B. R.; Clentsmith, G. K. B.; Rettig, S. J. J. Am. Chem. Soc. 1991,113, 4332-4334. (27) Honeycutt, J. B., Jr.; Riddle, J. M. / . Am. Chem. Soc. 1961, 83, 369-373. (28) Thaler, E.; Folting, K.; Huffman, J. C ; Caulton, K. G. Inorg. Chem. 1987,26, 314-311. (29) Maitlis, P. M.; Espinet, P.; Russell, M. J. H. in Comprehensive Organometallic Chemistry; G. Wilkinson, Ed.; Pergamon Press: Oxford, 1982; Vol. 6; pp 265-278. (30) Aylward, G. H.; Findlay, T. J. V. SI Chemical Data; Second ed.; Jacaranda Wiley Ltd.: Adelaide, 1974, pp 10. (31) Barnett, B. L.; Kruger, C ; Tsay, Y., Summerville, R. H.; Hoffmann, R. Chem. Ber. 1977,110, 3900-3907. (32) Jonas, K.; Wilke, G. Angew. Chem., Int. Ed. Engl. 1970, 9, 312-313. (33) Teller, R. G.; Williams, J. M.; Koetzle, T. F.; Burch, R. R.; Gavin, R. M. Inorg Chem 1981,20, 1806-1811. (34) Scioly, A. P.; Leutkens, M. L., Jr.; Wilson, R. B., Jr.; Huffman, J. C ; Sattelberger, A. P. Polyhedron 1987, 6, 741-757. (35) Francis, B. R.; Green, M. L. H.; Roberts, G. G. / . Chem. Soc, Chem. Commun. 1971, 1290. (36) Forder, R. A.; Prout, K. Acta Cryst. 1974, B30, 2318-2322. (37) Jonas, K.; Kruger, C. Angew. Chem., Int. Ed. Engl. 1980,19, 520-537. (38) Huffman, J. C ; Green, M. A.; Kaiser, S. L.; Caulton, K. G. / . Am. Chem. Soc. 1985,707,5111-5115. 49 (39) Bandy, J. A.; Berry, A.; Green, M. L. H.; Prout, K. / . Chem. Soc, Chem. Commun. 1985, 1462-1463. (40) Pez, G. P.; Grey, R. A.; Corsi, J. / . Am. Chem. Soc. 1981,103, 7528-7535. (41) Barron, A. R.; Wilkinson, G.; Motevalli, M.; Hursthouse, M. B. / . Chem. Soc, Dalton Trans. 1987, 837-846. (42) Rhine, W. E.; Stucky, G.; Peterson, S. W. / Am Chem Soc 1975, 97, 6401-6406. (43) Trofimenko, S. Prog. Inorg. Chem. 1986,34, 115-210. (44) Venanzi, L. M. Coord. Chem. Rev. 1982,43, 251-274. (45) Crabtree, R. FL; Hamilton, D. G. Adv. Organomet. Chem. 1988, 28, 299-338. (46) Setzer, W. N.; Schleyer, P. v. R. Adv. Organomet. Chem. 1985, 24, 353-451. (47) Schade, C ; Schleyer, P. v. R. Adv. Organomet. Chem. 1987, 27, 169-278. (48) Adamson, G. W.; Bell, N. A.; Shearer, H. M. M. Acta Cryst. 1981, B37, 68-71. (49) Johnson, M. P.; Shriver, D. F. J. Am. Chem. Soc. 1966, 88, 301-304. (50) Storr, A.; Thomas, B. S. Can. J. Chem. 1971,49, 2504-2507. (51) Shriver, D. F. Acc. Chem. Res. 1970, 3, 231-238. (52) Werner, H. Angew. Chem., Int. Ed. Engl. 1983, 22, 297-949. (53) Bruno, J. W.; Huffman, J. C ; Caulton, K. G. / . Am. Chem. Soc. 1984,106, 444-445. (54) Mayer, J. M.; Calabrese, J. C. Organometallics 1984, 3, 1292-1298. (55) Kadir, K.; Noreus, D. Z. Phys. Chemie Neue Folge 1989,163, 231-232. (56) Cotton, F. A.; Wilkinson, G. Advanced Inorganic Chemistry; Fifth ed.; Wiley-Interscience: New York, 1988, pp 118. (57) Kubas, G. J. Acc Chem. Res. 1988, 21, 120-128. (58) Crabtree, R. H. Acc. Chem. Res. 1990,23, 95-101. (59) Clark, H. C ; Smith, M. J. H. / . Am. Chem. Soc. 1986,108, 3829-3830. Chapter 3: EXPERIMENTAL. 3.1 General. 3.1.1 Procedures. Unless otherwise stated all manipulations were performed under an atmosphere of dry, oxygen-free1 dinitrogen using standard Schlenk or glove-box techniques. The glove-box used was a Vacuum Atmospheres HE-553-2 device equipped with a MO-40-2H purification system and a -30° C freezer. 1 H and 3 1 P NMR spectroscopy was performed on a Varian XL-300 instrument operating at 300 MHz and 121.4 MHz respectively; *H NMR spectra were referenced to internal C6D5H (7.15 ppm) or C6D5CD2H (2.09 ppm); 3 1 P NMR spectra were referenced to external P(OMe3)3 (141.00 ppm, with respect to H3PO4 at 0.00 ppm). 7 L i and 2 D NMR spectra were also acquired on the Varian instrument at 116.6 and 46.0 MHz respectively; 7 L i NMR spectra were referenced to external LiCl in D2O. Routine ! H and 3 1 P NMR spectroscopy was performed on a Bruker AC-200 instrument operating at 200.1 MHz and 81.1 MHz respectively. All chemical shifts are reported in ppm and all spectra were acquired at room temperature. The deuterated solvents used for NMR spectroscopy, C^D^ and CyDg, were dried by standing over activated 3 A molecular sieves; oxygen was removed by trap to trap distillation and three freeze-pump-thaw cycles. Microanalyses (C, H, halogen) were performed were performed by Mr. P. Borda of this department. 3.1.2 Reagents and starting materials. The organophosphorus ligand, dippp, was prepared by a literature method.2 Ethyl lithium was prepared by a published procedure3 and recrystallized from toluene. Lithium tetraethylborate was prepared by adding BEt3 (Aldrich) to a solution of ethyl lithium in Et 2 0 and recrystallized from toluene. Lithium tetraethylaluminate was prepared similarly from AlEt3 (Aldrich). Sodium tetraethylborate (Strem) was used as supplied or prepared by a literature method.4 Potassium deuteride was prepared by a literature method.5 Potassium triethylborohydride was prepared by addition of BEt3 to a slurry of KH (Aldrich) in toluene and was recrystallized from toluene; KBEt3D was prepared likewise from KD. Palladium chloride was obtained on loan from Johnson-Matthey and used to prepare (C6H5C=N)2PdCl2 by a literature method;6 (dippp)NiCl2 was prepared by adding one equivalent of dippp to an ethanolic solution of NiCl2.6H20; [(dippp)Rh]2(li-H)2 was prepared by a literature method.7 All solvents used in these and the following preparations were subjected to rigorous drying and removal of oxygen; as a preliminary each solvent was let stand over 4 A molecular sieves. Distillation from the appropriate drying agent was performed under an argon or dinitrogen atmosphere. Hexanes and toluene were heated to reflux over CaH2 prior to a final distillation from sodium benzophenone ketyl (hexanes) or sodium metal (toluene). Diethyl ether and pentane were distilled from sodium benzophenone ketyl. A small volume of glyme was added to the still-pot used for the aliphatic solvents. Acetone was sparged with dinitrogen. 3.2 Preparations. 3.2.1 (dippp)PdX2, (X = CI, I). To a solution of (PhC=N)2PdCl2 (4.37 g; 0,0114 mol) in acetone (100 mL) was added dippp (3.15 g; 0.0114 mol). With each drop of phosphine a yellow precipitate formed and the deep amber colour of the initial solution discharged completely to give a clear supernatant. The reaction vessel was exposed to the air and the fine yellow precipitate was collected upon a frit, washed with copious acetone (2 x 50 mL), and air-dried (4.34 g; 84% yield). Anal. Calcd for C15H34CI2P2PCI: C, 39.71; H, 7.55; CI, 15.63. Found: C, 40.00; H, 7.65; CI, 15.95. The product was too insoluble to allow further characterization. To a slurry of (dippp)PdCl2 (1.70 g; 3.75 x 10'3 mol) in acetone (50 mL) was added NaI.2H20 (1.53 g; 8.23 x 10-3 mol). The colour of the slurry deepened to ochre and stirring was continued for two hours. The product was exposed to the air, collected on a frit as a fine ochre powder, well-washed with acetone (2 x 50 mL), and then air dried (2.12 g, 89% yield). Anal. Calcd for C i s F ^ ^ P d : C, 28.30; H, 5.38; I, 39.87. Found: C, 27.98; H, 5.12; I, 40.05.. The product was too insoluble to allow further characterization. The diiodide could also be prepared directly by adding successively dippp and NaI.2H20 directly to a solution of (PhC=N)2PdCl2 (76% yield). 3.2.2 [(dippp)Pd]2(p.-H)2 (2). To a slurry of (dippp)PdI2 (1.15 g; 1.80 x 10-3 mol) in toluene (100 mL) at -60° C was slowly added a solution of KBEt3H (0.499 g; 3.81 x 10-3 mol) in toluene (10 mL). As the temperature was raised to -40° C a slight effervescence was noted and the red colour of the dimer developed as the (dippp)Pdl2 reacted. The temperature was maintained at -40° C for two hours after which time gas evolution had ceased and the solution was deep red in colour. The temperature was further raised to -20° C for thirty minutes, after which the reaction mixture was passed through a frit lined with Celite. The solvent was removed in vacuo and the residue dissolved in pentane (9 mL). The filtered solution was cooled to -30° C and deep-red crystals appeared after twenty-four hours (0.300 g, 44% yield): *H NMR ( C 6 D 6 ) 5 1.93 (m, 6H, P C H 2 C / / 2 C H 2 P , C7/Me2), 1.35 (m, 4H, PC// 2 CH 2 Ctf 2 P), 1.30 and 1.10 (m, 24H, CHMe2), -2.52 (quint, IH, ^-H, 2 / P . H = 34.9 Hz); 31p NMR (C 6 D 6 ) 8 27.0. Anal. Calcd for C3oH7oP4Pd2: C, 46.94; H, 9.19. Found: C, 47.22; H, 9.25. 3.2.3 [(dippp)Pd] 2(u-D) 2 (2a). The deuterium analogue was prepared by a procedure identical to that for [(dippp)Pd]2(n-H)2 using KBEt 3D (0.518 g; 3.72 x 10"3 mol) and (dippp)PdI2 (1.16 g; 1.82 x 10-3 mol).. Recrysallization from pentane (10 mL) gave deep-red crystals (0.356 g, 51% yield): lR NMR (C 6P 6) 5 1.93 (m, 6H, PCH 2 C/ / 2 CH 2 P, C//Me 2), 1.35 (m, 4H, PC/ / 2 CH 2 C/ / 2 P) , 1.30 and 1.10 (m, 24H, CHMe2)\ 3 1 P NMR (C 6D 6) 5 27.0 (quint, 2/P-D = 5.5 Hz); 2 D NMR 5 -2.49 (quint of t, U.-D). Anal. Calcd for C 3 oH 6 8 D 2 P 4 Pd 2 : C, 46.81; H, 9.17. Found: C, 46.95; H, 9.12. 3.2.4 [ ( d i p p p ) P d ] 2 ( p > H ) 2 « L i B E t 4 (1). Lithium tetraethylborate (0.023 g; 1.73 x 10"4 mol) was added to a solution of [(dippp)Pd]2(p>H)2 (0.133 g; 1.73 x lO^mol) in toluene (10 mL) and the mixture stirred for ten minutes The red colour of the solution became slightly less intense during this time. The solution was filtered through Celite and the solvent was stripped off under vacuum. The residue was taken up in toluene (0.5 mL) and the solution was layered with pentane (4.0 mL). The solution was cooled to -30° C and the product appeared as maroon needles after twelve hours (0.142 g; 91% yield): *H NMR (C 6D 6) 6 1.78 (m, 6H, PCH 2 C/ / 2 CH 2 P, C//Me 2), 1.29 (m, 10H, B C H 2 C / / 3 and P C / / 2 C H 2 C / / 2 P ), 1.20 and 1.07 (m, 24H, CHMe2), 0.42 (m, 4H, B C / / 2 C H 3 ) , -3.66 (quint, IH, ^l-H, 2/ p . H = 32.6 Hz); 31p NMR (C 6D 6) 8 22.4; 7 L i NMR (C 6D 6) 8 -3.9. Anal. Calcd for C38H9oBLiP4Pd2: C, 50.62; H, 10.06. Found: C, 50.57; H, 10.06. 3.2.5 [(dippp)Pd] 2(|I-H) 2*LiAlEt 4 (3). [(dippp)Pd]2(|!-H)2 (0.100 g; 1.30 x 10-4 mol) and LiAlEu (0.020 g; 1.33 x 10"4 mol) were treated as for the preparation of [(dippp)Pd] 2(M.-H)2»LiBE.t 4. Recrystallization from pentane/toluene (3:1; 10 mL) gave purple crystals after forty-eight hours (0.109 g; 89% yield): lH NMR (C 6D 6) 8 1.75 (m, 6H, PCH2C//2CH2P, C//Me 2), 1.21 (t, 6H, AlCH 2 C// 3 ) , 1.16 (m, 4H, PC//2CH2C//2P), 1-20 and 1.07 (m, 24H, CHMe2), 0.89 (m, 4H, A1C// 2CH 3), -3.87 (quint, IH, (l-H, 2 / P . H = 35.0 Hz); 31p NMR (C 6D 6) 8 23.3. Anal. Calcd for C38H9oAlLiP4Pd2: C, 49.73; H, 9.88. Found: C, 49.48; H, 9.85. 3.2.6 [(dippp)Pd] 2(n-H) 2 »NaBEt 4 (4). To a solution of [(dippp)Pd]2(|i-H)2 (0.100 g; 1.30 x 10"4 mol) in toluene (10 mL) was added NaBEty (0.020 g; 1.33 x 10"4 mol). Upon adding the NaBEty the colour darkened to a bottle-green. The solution was stirred for thirty minutes. Removal of the solvent gave a blue, crystalline residue. The residue was taken up in toluene (1.2 mL) and the green solution was layered with pentane (5.0 mL). Blue crystals appeared after six hours (0.050 g; 41% yield): lH NMR (C6D6) 8 1.76 (m, 6H, PCH 2C//2CH2P, CHMc2), 1-26 (t, 6H, BCH 2 C/ / 3 ) , 1.10 (m, 4H, P C / / 2 C H 2 C / / 2 P ) , 1.10 and 0.99 (m, 24H, CHMe2), 0.89 (m, 4H, BC/ / 2 CH 3 ) , -3.43 (quint, IH, ji-H, 2 / P . H = 35.4 Hz); 3ip NMR (C 6D 6) 8 25.4 . Anal. Calcd for C 3 8 H 9 oBNaP 4 Pd 2 : C, 49.74; H, 9.89. Found: C, 50.00; H, 10.07. 3.2.7 (dippp)Pd(r|2-H2C = CH2) (5). To a solution of [(dippp)Pd]2(n-H)2 (0.081 g; 1.06 x 10"4 mol) in toluene (10 mL), subjected to several freeze-pump-thaw cycles, was added ethylene (33 mm Hg; 5.0 x 10-4 mol) from a constant volume bomb. The initial red colour of the solution rapidly discharged to give a colourless solution. The solvent was stripped off, the residue dissolved in minimum pentane (2.5 mL), and then cooled to -30° C. Colourless crystals appeared after forty-eight hours (0.070 g; 80% yield): *H NMR (C 6 D 6 ) 8 2.86 (d, 2H, H 2 C=CH 2 , 2 / P H = 1.5 Hz), 1.69 (sept, 2H, C/ /Me 2 ) / H H ' = 7.0 Hz), 1.68 (m, IH, PCH2C//2CH2P), 1.25 (m, 2H, PC// 2 CH 2 C# 2 P), 1.08 and 0.96 (dd, 12H, CHMe 2, 2JPR = 15.0 Hz, / H H ' = 7.0 Hz); 31p NMR (C 6D 6) 8 29.5 . Anal. Calcd for Ci 7 H 3 8 P2Pd: C, 49.70; H, 9.32. Found: C, 49.90; H, 9.41. 3.2.8 (dippp)Pd(PPh3) (7). To a solution of [(dippp)Pd]2(^-H)2 (0.090 g; 1.17 x 10"4 mol) in toluene (15 mL), was added PPh3 (0.061 g; 2.33 x 10-4 mol). The red colour discharged rapidly and a vigorous evolution of gas was observed. The solvent was stripped off and the yellow residue taken up in pentane/toluene (1:1, 15 mL); a fine yellow powder deposited on standing ( 0.132 g; 87% yield): lH NMR (C6D6) 8 7.88 (m, 6H, o-Ph), 7.15 (m, 9H, m,p-Ph) 1.75 (m, 6H, P C H 2 C / / 2 C H 2 P , 0 / M e 2 ) , 1.23 (m, 4H, PC//2CH2C//2P), 1.08 and 0.96 (dd, 24H, CHMe 2, 2Jpn = 9.6 Hz, JHH< = 3.0 Hz); 3lp NMR (C 6D 6) 8 33.8 (t, IP, /pP> = 88.5 Hz), 23.7 (d, 2P). Correct micronanalysis for C 3 3 H 4 Q P 3 P d was not obtained. 3.2.9 (dippp)Pd(DMAD) (8). To a solution of [(dippp)Pd]2(li-H)2 (0.075 g; 9.77 x 10"5 mol) in toluene (10 mL), was added DMAD (0.014 g; 9.85 x 10"5 mol). The initial red colour of the solution slowly discharged to give a colourless solution within one hour. The solvent was stripped off, and the residue dissolved in minimum pentane (5 mL). The colourless solution was cooled to -30° C and colourless crystals appeared after forty-eight hours (0.087 g; 85% yield): *H NMR (C 6 D 6 ) 8 3.54 (s, 3H, C 0 2 C H 3 ) , 1.82 (sept, 2H, C7/Me 2, JHH' = 7.0 Hz), 1.51 (m, IH, P C H 2 C / / 2 C H 2 P ) , 1.22 (m, 2H, P C / / 2 C H 2 C / / 2 P ) , L H and 0.87 (dd, 12H, CHMe2, 2 / P H = 16.0 Hz, / H H ' = 7.0 Hz); 3lp NMR (C 6D 6) 8 34.5 . Anal. Calcd for C 2 iH4 0 P 2 O 4 Pd: C, 48.05; H, 7.68. Found: C, 48.27; H, 7.90. 3.2.10 ['(dippp)'Ni]2(n-H)2 (6). (dippp)NiCl2 (1.058 g; 2.60 x 10-3 mol) and KBEt 3H (0.737 g; 5.34 x 10-3 mol) was treated as for the preparation of [(dippp)Pd]2(}i-H)2. After removal of the solvent, the residue was recrystallized from toluene/pentane/ (3:1, 20 mL) to give red-black platelets (0.345 g, 42% yield): *H NMR (C 6D 6) 8 1.61 (m, 6H, PCH 2 C/ / 2 CH 2 P, CHMe2), 1-35 and 1.16 (dd, 24H, CHMe2, JH'-U = 6.4 Hz, / P . H = 7.2 Hz ), 1.23 (m, 4H, PC/ / 2 CH 2 C/ / 2 P) , -10.79 (quint, IH, n-H, 2 / p . H = 18.5 Hz); 31p NMR (C 6D 6) 8 36.3. Anal. Calcd for C 3oH 7oNi 2P 4: C, 53.61; H, 10.50. Found: C, 53.49; H, 10.44. 3.2.11 [ (dippp)Ni] 2(n-H) 2 .LiBEt 4 (9). [(dippp)Ni]2(n-H)2 (0.130 g; 1.93 x 10"4 mol) and LiBEt4 (0.026 g; 1.93 x 10-4 mol) were treated as for the preparation of [ (dippp)Pd] 2 (p . -H) 2 »LiBEt 4 . Recrystallization from toluene (1.8 mL) gave red-black crystals which were mostly [(dippp)Ni]2(|i-H)2.LiBEt4 (0.094 g; 60% yield): *H NMR (C7D8) 8 1.80 (m, 6H, P C H 2 C / / 2 C H 2 P , C//Me 2), 1.29 and 1.13 (m, 30H, CHMe2, BCH 2 C/ / 3 ) , 1.23 (m, 4H, PC/ / 2 CH 2 C/ / 2 P) , 0.40 (m, 4H, B O / 2 C H 3 ) , -10.9 (br s, IH, i^-H); 31p NMR (C 7D 8) 8 31.9 (extra LiBEty added). Anal. Calcd for C 3 8 H 9 0 B L i N i 2 P 4 : C, 56.62; H, 11.25. Found: C, 55.70; H, 11.11. 3.2.12 [(dippp)Rh] 2(n-H) 2 .LiBEt 4 (10). [(dippp)Rh]2(u-H)2 (0.100 g; 1.30 x 10'4 mol) andLiBEt4 (0.020 g; 1.33 x 10" 4 mol) were treated as for the preparation of [(dippp)Pd] 2(u.-H) 2»LiBEt4. The product was characterized in solution: *H NMR (C6D6) 6 1.78 (m, P C H 2 C / / 2 C H 2 P , CHMe2), 1.29 (m, P C / / 2 C H 2 C / / 2 P , B C H 2 C / / 3 ) , 1.22 (m, CUMe2), 0.41 (m, 4H, BC/ / 2 CH 3 ) , -6.65 (m, u-H); 31p NMR (C 6D 6) 5 47.5 (d, 7 R h . P = 155 Hz). 3.3 References. (1) Brown, T. L.; Dickerhoof, D. W.; Bafus, D. A.; Morgan, G. L. Rev. Sci. Instrum. 1962,33, 491-492. (2) Tani, K.; Tanigawa, E.; Yatsuno, Y.; Otsuka, S. / . Organomet. Chem. 1985, 279, 87-95. (3) Bryce-Smith, D.; Turner, E. E. / . Chem. Soc. 1953, 861-867. (4) Honeycutt, J. B., Jr.; Riddle, J. M. J. Am. Chem. Soc. 1961, 83, 369-373. (5) Klusener, P. A. A.; Brandsma, L.; Verkruijsse, H. D.; Schleyer, P. v. R.; Friedl, T.; Pi, R. Angew. Chem., Int. Ed. Engl. 1986, 25, 465-466. (6) Doyle, J. R.; Slade, P. E.; Jonassen, H. B. Inorg. Synth. 1960, 6, 216-219. (7) Fryzuk, M. D.; Piers, W. E.; Einstein, F. W. B.; Jones, T. Can. J. Chem. 1989, 67, 883-896. 


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items