Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Phosphine complexes of zirconium, hafnium and the lanthanoid metals Haddad, Timothy Samir 1990

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

Item Metadata

Download

Media
831-UBC_1991_A1 H32.pdf [ 10.31MB ]
Metadata
JSON: 831-1.0059690.json
JSON-LD: 831-1.0059690-ld.json
RDF/XML (Pretty): 831-1.0059690-rdf.xml
RDF/JSON: 831-1.0059690-rdf.json
Turtle: 831-1.0059690-turtle.txt
N-Triples: 831-1.0059690-rdf-ntriples.txt
Original Record: 831-1.0059690-source.json
Full Text
831-1.0059690-fulltext.txt
Citation
831-1.0059690.ris

Full Text

PHOSPHINE COMPLEXES OF ZIRCONIUM, HAFNIUM AND THE LANTHANOID METALS By TIMOTHY SAMIR HADDAD B. Sc., The University of Regina, 1984 M. Sc., The University of British Columbia, 1987 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE STUDIES Department of Chemistry We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA DECEMBER 1990 © Timothy Samir Haddad 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 CLMem The University of British Columbia Vancouver, Canada Date Dec £ Q l ^ l O DE-6 (2/88) ii A B S T R A C T The synthesis of a variety of new lanthanoid phosphine complexes has been achieved by complexing either one or two amido-diphosphine ligands to yttrium, lutetium or lanthanum. At room temperature, the seven-coordinate bis(amido-diphosphine) complexes, MCl[N(SiMe2CH2PR2)2l2- are fluxional and display NMR spectra indicative of complexes where the phosphorus donors are rapidly exchanging, probably via a dissociation-reassociation pathway. It is possible to generate thermally unstable hydrocarbyl complexes of the type, M(R)[N(SiMe2CH2PMe2)2]2» which undergo a clean first order elimination of R-H to generate cyclometallated i 1 complexes of the type, M[N(SiMe2CHPMe2)(SiMe2CH2PMe2)][N(SiMe2CH2PMe2)2]; the yttrium derivative was crystallographically characterized. These thermally robust compounds will undergo a-bond metathesis with H2 and D2 at high temperature, but appear to be too sterically congested to react with larger molecules. The synthesis of a series of mono(amido-diphosphine) lanthanoid complexes, MCl2 [N(SiMe2CH2PR2)2]. (R = Me, Ph, Pr', Bul) has also been achieved. Complexes of the type, MCl2[N(SiMe2CH2PMe2)2L are insoluble in hydrocarbon solvents, presumably because they are oligomeric in nature. They will however, dissolve in THF probably forming seven-coordinate bis(THF) monomers. Attempts to alkylate these compounds generally led to decomposition; the cyclometallated 1 1 bis(ligand) complex, M[N(SiMe2CHPMe2)(SiMe2CH2PMe2)] [N(SiMe2CH2PMe2)2], was identified as the major product. A route to a dimeric mono(amido-diphosphine) allyl complex, {YCl(allyl)[N(SiMe2CH2PMe2)2]}2 (characterized by crystallography) was found via the reaction of allyl-MgCl or Mg(allyl)2(dioxane) with YCl[N(SiMe2CH2PMe2)2l2- The mono(ligand) complexes containing bulky i i i phosphine donors (R = Ph , B u l , Pr-) are soluble in hydrocarbon solvents; YCl2[N(SiMe2CH.2PPr-2)2] can be isolated as either a T H F adduct or as the base-free dimer. A new reaction, mediated by a zirconium or hafnium amido-diphosphine complex, where al ly l and butadiene moieties are coupled together to generate a coordinated [ r | 4 : r i - - C H 2 = C H C H = C H C H 2 C H 2 C H 2 ] 1 - fragment has been investigated. The process is very sensitive to the nature of the ancillary ligands at the metal. For MC l ( r i 4 -C4H 6 ) [N ( S iMe2CH2PR2 )2 ] complexes, after the addition of a l l y lMgC l , the transformation takes about one hour when M = H f & R = Pr-, two hours when M = Zr & R = Pr-, a week when M = H f & R = Me , and results only in decomposition when M = Zr & R = Me. Similarly, for the zirconium mediated coupling of 1-methylallyl with butadiene, when R = Me , decomposition occurs and when R = Pr-, after two hours the coupling is complete. Two of the four possible coupled products are formed in unequal amounts, and the coupl ing occurs exclusively at the substituted end of the 1-methylal ly l unit as determined by X-ray crystallography. Wh ich diastereomer is formed in excess was not determined. The reduction of ZrCl3[N(SiMe2CH2PR2)2] (R = Pr 1 or Bu-) with Na/Hg amalgam under nitrogen results in the formation of a binuclear zirconium dinitrogen complex, {ZrCl[N(SiMe2CH2PR2)2]}2(M--'n2:'n2-N2). X-ray crystallography (for R = Pr-) reveals that the N2 l igand is symmetrically bound in a side-on fashion to both metals. In addition, the N — N bond length of 1.548 (7) A, the longest bond length ever reported for a dinitrogen complex, indicates that the dinitrogen has been reduced to a N 2 4 ' hydrazido l igand. Protonation of the complex with HCl results in a quantitative formation of hydrazine. iv TABLE OF CONTENTS ABSTRACT ii TABLE OF CONTENTS iv LIST OF TABLES vii LIST OF FIGURES viii LIST OF ABBREVIATIONS xiii ACKNOWLEDGEMENTS xvii CHAPTER 1 GENERAL INTRODUCTION 1.1 Transition Metal Phosphine Complexes 1 1.2 Phosphine Complexes of the Lanthanid.es and Group 3 Metals 2 1.3 Phosphine Complexes of Hafnium and Zirconium 7 1.4 Summary of Previous Work from Our Laboratory 10 1.5 The Scope of this Thesis 16 1.6 References 18 CHAPTER 2 BIS(AMIDO-DIPHOSPHINE) COMPLEXES OF LANTHANUM, YTTRIUM AND LUTETIUM 2.1 Introduction 23 2.2 Synthesis of Bis(amido-diphosphine) Complexes 24 2.3 NMR Characterization and Fluxional Behaviour 26 2.4 X-ray Crystal Structure Determination of Y^N(SiMe2CHPMe2)(SiMe2CH2PMe2)][N(SiMe2CH2PMe2)2]44 2.5 Kinetics of the Thermolysis of MR[N(SiMe2CH2PMe2)2h 48 2.6 Reactivity of M[N(SiMe 2CHPMe 2)(SiMe2CH 2PMe 2)][N(SiMe 2CH2PMe 2) 2]53 2.7 Summary 55 2.8 Experimental Procedures 55 2.9 References 64 C H A P T E R 3 M O N O ( A M I D O - D I P H O S P H I N E ) C O M P L E X E S OF Y T T R I U M A N D L U T E T I U M 3.1 Introduction 68 3.2 Syntheses of M C l 2 [ N ( S i M e 2 C H 2 P M e 2)2] 69 3.3 Synthesis and Structure of {Y(a l ly l ) [N (S iMe 2 CH 2 PMe2)2] }2(M -^C1)2 74 3.4 Syntheses of YCl2 [N (S iMe 2 CH 2 PR2)2 ] R = Ph, Bu-, Pr- 82 3.5 Synthesis and Thermolysis of YCl(Ph)[N (SiMe 2CH2PPr- 2 )2] 88 3.6 Summary 91 3.7 Experimental Procedures 92 3.8 References 101 C H A P T E R 4 A L L Y L - D I E N E C O U P L I N G A T H A F N I U M A N D Z I R C O N I U M 4.1 Introduction 102 4.2 Synthesis of Allyl-Diene Coupled Complexes 103 4.3 Molecular Structures of HfCn4:T| --C7H11 ) [N (S iMe 2 CH 2 PPr- 2 ) 2 ] , 16c, and Zr(Ti 4:r|--C8Hi3)[N(SiMe 2CH2PPr-2)2], 18c 105 4.4 N M R Spectroscopic Characterization 110 4.5 Summary 117 4.6 Experimental Procedures 118 vi 4.7 References 122 C H A P T E R 5 S Y N T H E S I S A N D C H A R A C T E R I Z A T I O N O F A S I D E - O N D I N I T R O G E N C O M P L E X O F Z I R C O N I U M 5.1 Introduction 123 5.2 Synthesis of {ZrCl[N(SiMe2CH 2PPr- 2)2]}2(N2) 127 5.3 Molecular Structure of {ZrCl[N(SiMe 2CH 2PPr- 2)2]} 2(li-Tl2:ri2-N2) 128 5.4 NMR Spectra of {ZrCl[N(SiMe 2CH 2PR2)2]}2(N2) 133 5.5 Reactivity of {ZrCl[N(SiMe 2CH 2PPr i2)2])2(N2) 136 5.6 Experimental Procedures 139 5.7 References 141 T H E S I S S U M M A R Y A N D F U T U R E P R O S P E C T S 144 A P P E N D I X 148 A - l X-ray Crystallographic Analysis of Y[N(SiMe2CHPMe 2 )(SiMe 2 CH 2 PMe 2 )][N(SiMe 2CH2PMe 2 ) 2 ]148 A-2 X-ray Crystallographic Analysis of {Y(Ti3-allyl)[N(SiMe 2CH2PMe2)2]}2(li-Cl)2 155 A-3 X-ray Crystallographic Analysis of Hf(ri4:ri - - C 7 H 1 i)[N(SiMe2CH2PPr-2)2] 160 A-4 X-ray Crystallographic Analysis of Zr(n.4:r| l -C 8 Hi 3 ) [N(SiMe 2 CH 2 PPr- 2 ) 2 ] 165 A-5 X-ray Crystallographic Analysis of {ZrCl[N(SiMe 2 CH 2PPri 2)2] }2(M--'n2:Tl2-N2) 173 vii Table 1.1 Table 1.2 Table 1.3 Table 1.4 Table 2.1 Table 2.2 LIST OF TABLES Phosphine Complexes of the Group 3 Metals. Phosphine Complexes of the Lanthanides. Group 3 and Lanthanide Metal Phosphorus Bond Lengths. 3 4 6 Phosphine Complexes of Hafnium and Zirconium Prior to 1980. 8 1 Jp.y and 2Jp.p Coupling Constants for MR[N(SiMe2CH2PMe2)2l2- 33 -Jp.Y and 2Jp.p Coupling Constants for M[N(SiMe 2 CHPMe 2 ) (S iMe 2 CH 2 PMe 2 ) ] [N(SiMe 2 CH 2 PMe 2 )2] . 41 Table 2.3 Selected Bond Lengths for Table 2.4 Table 2.5 Table 2.6 Table 3.1 Table 3.2 Table 4.1 Table 4.2 Table 4.3 Table 4.4 Table 5.1 Table 5.2 Table 5.3 Y[N(SiMe 2 CHPMe 2 ) (S iMe 2 CH 2 PMe 2 ) ] [N(SiMe 2 CH 2 PMe 2 ) 2 ] . 44 Selected Bond Angles for I 1 46 50 50 74 76 Y[N(SiMe 2 CHPMe 2 ) (S iMe 2 CH 2 PMe 2 ) ] [N(SiMe 2 CH 2 PMe 2 ) 2 ] . Rate Data for the Thermolysis of YR[N(SiMe2CH2PMe2)2]2-Eyring Data for the Thermolysis of YR[N (SiMe2CH2PMe2)2h-Selected Bond Lengths for {Y(allyl)[N(SiMe2CH 2PMe 2)2]}2(u.-Cl)2. Selected Bond Angles for {Y(allyl)[N(SiMe2CH 2PMe2)2]}2(l i-Cl) 2 . Selected Bond Angles for Zr(ri4 : r jl-C8Hi3)[N(SiMe2CH 2 PPr i2)2]- 105 Selected Bond Lengths for Zr(r|4:n--C8Hi3)[N(SiMe2CH2PPr-2)2] 106 Selected Bond Lengths for Hf(r| 4:ri--C7Hii)[N(SiMe2CH2PPr-2)2] 106 Selected Bond Angles for Hf(rj4: T 1l-C7Hi 1)[N(SiMe2CH2PPr- 2)2]. 105 Selected Bond Angles for {ZrCl [N(SiMe 2CH 2PPr-2)2]} 2(H-il 2:ri2-N2) • 131 Selected Bond Lengths for {ZrCl[N(SiMe2CH2PPri2)2]}2(lA-Tl 2:ri 2-N2). 131 Theoretical and Calculated Hydrazine Concentrations. 138 vi i i LIST OF FIGURES Figure 1 . 1 A representation of Yb[N (SiMe3) '2 ] 2 (dmpe), t n e f- r s t crystallographically characterized lanthanide phosphine complex. 5 Figure 1 . 2 Some recently characterized group 4 phosphine complexes (left to right references 44, 45 and 46). 9 Figure 1 . 3 Two target lanthanoid phosphine complexes. 17 Figure 2 . 1 300 MHz -H NMR spectrum (top, in C 6 D 6 ) and 75.4 MHz 1 3 C { l U ) NMR spectrum (bottom, in C 6 D 5 C D 3 ) of Y (Ph ) [N (S iMe 2 CH 2 PMe2)2 l2 recorded at room temperature. Solvent peaks are marked with an asterisk. 27 Figure 2.2 Figure 2.2: 300 MHz -H NMR spectrum (top) and 75.4 MHz 1 3 C { -H ) N M R spect rum (bottom) of Y(Ph) [N(S iMe 2 CH 2 PMe 2 )2 ]2 in C 6 D 5 C D 3 recorded at -59 °C. Solvent peaks are marked with an asterisk. 28 Figure 2.3 121 M H z 3 - P { - H } N M R spectrum of Y(Ph) [N(S iMe 2 CH 2 PMe 2 )2 ]2 in C 6 D 5 C D 3 recorded at -49 °C (top) and a simulation (bottom) using Bruker PANIC software. See Figure 2.5 and Table 2.1 for details.30 Figure 2.4 121 M H z 3 1 P { 1 H} N M R spectrum of Lu(Ph)[N(SiMe 2CH 2 PMe2)2]2 in C 6 D 5 C D 3 recorded at -59 °C (top) and a simulation (bottom) using Bruker PANIC software. See Figure 2.5 and Table 2.1 for details. 31 Figure 2.5 Three possible structures based on a pentagonal bipyramidal geometry for MR[N (S iMe2CH2PMe2)2]2 complexes in the low temperature limit. The 3 1 P { -H } NMR spin system in the pentagonal plane is also shown. 32 ix Figure 2.6 121 MHz 3-P{-H} NMR spectrum of YCl[N(SiMe2CH2PPh2)2]2 in C6D5CD3 recorded at -59 e C (top) and a spectral simulation using Bruker PANIC software (below). 35 Figure 2.7 121 MHz 31P{ lH} NMR spectra of LaCl[N(SiMe2CH2PPh2)2]2 in CD3C6D5 as a function of temperature. The actual concentration is unknown. An impurity, HN(SiMe2CH2PPh 2) 2, is indicated with an asterisk. 37 Figure 2.8 121 MHz 3 1P{1H} NMR spectrum (top) and simulation (bottom) of Y^N(SiMe 2CHPMe 2 ) (SiMe 2CH 2PMe 2 ) ] [N(SiMe 2CH 2PMe 2 ) 2 ] ,7a, in CD3C6D5 recorded at -29 °C. See Table 2.2 for details. 39 Figure 2.9 121 MHz 3 1P{ 1H) NMR spectrum (top) and simulation (bottom) of Lu[N(SiMe 2CHPMe 2 ) (SiMe 2CH 2PMe 2 ) ] [N(SiMe 2CH 2PMe 2 ) 2 ] ,8a, in CD 3 C6D 5 recorded at -29 °C. See Table 2.2 for details. 40 Figure 2.10 8 9 Y NMR spectra of l a (top, 19.6 MHz) in CD3C6D5/THF and 7a (middle, 24.6 MHz) in CD3C6D5. The theoretical spectrum of 7a, based on the couplings constants in Table 2.2, is also shown (bottom). 43 Figure 2.11 Chem 3D® core view (top) and ORTEP stereoview (bottom) of Y[N(SiMe 2CHPMe 2 ) (SiMe 2CH 2PMe 2 ) ] [N(SiMe 2CH 2PMe 2) 2]. 45 Figure 2.12 A possible transition state for the elimination of R—H. 48 Figure 2.13 The first-order rate plots for the thermolysis of Y(CH 2 C6H 5 ) [N(SiMe 2 CH 2 PMe 2 ) 2 ] 2 , 5a, and the Eyring plots for the thermolysis of Y(R)[N(SiMe 2CH 2PMe 2 ) 2 ] 2 , R = C H 2 C 6 H 5 , C D 2 C 6 D 5 , C 6 H 5 . 49 Figure 3.1 NMR spectra of LuCl2 [N(SiMe 2CH 2 PMe2)2]- H a : 300 MHz -H spectrum (top) and 75.4 MHz 13C{-H} spectrum (bottom), both in C6D5CD3/THF. The C6D5CD3 peak is marked with an asterisk. 73 Figure 3.2 Three views of {Y(allyl)[N(SiMe 2CH 2PMe2)2]}2(H-C1) 2,12a. At the top are two Chem 3D® views; the figure at the top right shows the approximate pentagonal bipyramidal geometry at yttrium (based on the assignment of two coordination sites for the allyl moiety): the N and CI' atoms are axial while PI, P2, CI and the allyl are equatorial. The bottom view is the ORTEP stereoview showing the complete atom labelling scheme. 75 Figure 3.3 300 MHz *H NMR spectra of {Y(allyl)[N(SiMe 2-CH 2 PMe 2 ) 2 ] } 2 (u-C l ) 2 , 12a, at + 80 °C (top) and at -50 °C (bottom) in C 6 D 5 C D 3 (marked with asterisks). 78 Figure 3.4 Tempera ture invar iant N M R spectra of {YCl2[N(SiMe2CH2PPr- 2 )2]}2- 10c in C 6 D 6 : 121 MHz 3-P{lH} NMR spectrum (top) and 300 MHz -H NMR spectrum (bottom). 85 Figure 3.5 121 MHz 3 1P{-H} variable temperature NMR spectra of YCl2[N(SiMe2CH 2PPr-2)2](THF) in C D 3 C 6 D 5 . 87 Figure 3.6 Temperature invariant N M R spectra of YCl(Ph)-[N( S i M e 2 C H 2 PPr - 2 ) 2 ] , 14c, in C D 3 C 6 D 5 . 121 MHz 31P{-H} NMR spectrum (top) and 300 MHz -H NMR spectrum (bottom). The CD3C6D5 resonances are marked with an asterisk. 89 Figure 3.7 121 MHz 3-P{-H} NMR spectra (in C D 3 C 6 D 5 ) of 15c and a possible structure for this complex. 90 Figure 4.1 Four possible isomers for the coupling of 1-methylallyl with butadiene. 104 xi Figure 4.2 Molecular structure of Zr(rj 4 :Ti--CgHi3)[N(SiMe2CH2PPr-2)2]-1 8 c . Two Chem 3D® views showing the two different diastereomers and an ORTEP stereoview showing the complete atom numbering scheme. 108 Figure 4.3 Molecular structure of Hf ( r | 4 : r i . l - C 7 H i i ) [N(S iMe2CH 2 PPr- 2 )2 ]-16c. A Chem 3D® view for comparison with 16c and an ORTEP stereoview showing the complete atom numbering scheme. 109 Figure 4.4 121 MHz 31p{lH} NMR spectra in C 6 D 6 of Zr(Tl 4:Tl 1-C8Hi3)-[N(SiMe2CH2PPr-2)2]» 18c. The upper spectrum is of the crude reaction mixture and shows the different proportions of the two diastereomers, the lower spectrum is of a 1:1 mixture of crystallized products. ZrCl(T | 4 -C4H 6 ) [N(S iMe2CH2PPr -2 )2] is marked with an asterisk. I l l Figure 4.5 300 MHz lH NMR spectra of Zr(ri 4 :n. 1 - C 8 H i 3 ) -[N(SiMe2CH2PPr-2)2], 18c, in QDe. 112 Figure 4.6 A portion of the 75.4 MHz ^ C f - H } NMR spectra in C6D 6 of Zr(rt 4:Tii-C8Hi3)[N(SiMe 2CH2PPri2)2]. 18c. 113 Figure 4.7 121.4 MHz 3-P{ -H} NMR spectra in THF/C 6 D 6 of C 3 H 5 MgCl and ZrCl(ri i 4 -C4H6 ) [N (SiMe2CH2PPr i 2)2] after 50 mins (upper) and 120 mins (lower). Note that the two scales are not identical. 116 Figure 4.8 The numbering scheme used for the NMR spectroscopic assignments. 118 Figure 5.1 Representations of the first crystallographically characterized end-on mononuclear and bridging dinuclear N2 complexes. 123 Figure 5.2 Three views of the molecular structure of 19c. At the top are two Chem 3D® views showing the symmetric environment about the zirconium centres. The top left view is looking down the Zr—Zr xii axis, while at the top right is a view showing the close contact of the isopropyl groups across the N 2 bridge. At the bottom is a stereoview of this complex showing the disorder in the ligand backbone. 130 Figure 5.3 300 MHz -H NMR spectra of {ZrCl[N(SiMe2CH2PBu-2)2]}2(N2) in C 6 D 5 C D 3 . The upper trace at + 60 °C and the lower spectrum was recorded at 20 °C. 134 Figure 5.4 121 MHz 3 1P{-H} NMR spectra of {ZrCl[N(SiMe 2-CH2PBu-2)2]}2(N2) in C6D5CD3. The upper trace at + 25 °C and the lower spectrum was recorded at 0 °C. 135 Figure 5.5 121 MHz 31P{-H} (upper trace) and 300 MHz -H NMR (lower trace) spectra of the crude reaction mixture from {ZrCl[N(SiMe2CH2PPr-2)2] }2(N2) + CDCI3. 137 xiii L I S T O F A B B R E V I A T I O N S The following list of abbreviations, most of which are commonly used in the chemical literature, will be employed in this thesis: A angstrom Anal. analysis Ax axial br broad Bz benzyl Bu butyl Bu- tertiary butyl °C degree centigrade Calcd. calculated Ci ipso-carbon cf. compare Chem 3D® molecular modelling program for the Macintosch !3C{-H} observe carbon while decoupling proton cm centimetre Cm mera-carbon CN coordination number Co ortho-caibon c p para-carbon Cp cyclopentadienyl anion, C5H5-Cp* per(methyl)cyclopentadienyl anion, C s M e s -Cp' mono(methyl)cyclopentadienyl anion, MeCshLf A heat xiv AG* free energy of activation A H * enthalpy of activation AS* entropy of activation • Au c peak separation at coalescence 8 chemical shift d doublet d n n atoms of deuterium per molecule deg degree dmpe l,2-bis(dimethylphosphino)ethane dmpm l,l-bis(dimethylphosphino)methane dppe l,2-bis(diphenylphosphino)ethane e molar extinction coefficient e.g. for example Eq equatorial eu entropy units fac facial E gram h Planck's constant H a anri-hydrogen H m meso-hydrogen Hm mera-hydrogen H 0 orr/w-hydrogen H p para-hydrogen H s syn-hydrogen Hz Hertz i.e. that is IR infrared XV n T A-B n-bond coupling constant between A and B atoms K degree Kelvin k rate constant k B Boltzmann constant kc rate constant at the temperature of coalescence Kcal kilocalorie Kgq equilibrium constant k H observed rate constant for a protio-isomer k D observed rate constant for a deuterio-isomer kobs observed rate constant •^max wavelength of maximum absorbance L litre M molarity Me methyl mer meridional mg milligram MHz megaHertz mL millilitre mmol rnillimole mol mole M.Sc. master of science NMR nuclear magnetic resonance nm nanometre Np neopentyl ORTEP Oak Ridge Thermal Ellipsoid Plotting Program P pentet xvi PANIC Parameter Adjustment in NMR by Iteration Calculations Ph phenyl % percent % T percent transmittance 31P{!H} observe phosphorus while decoupling proton Pr- isopropyl q quartet s singlet sec second sept septet t triplet t-butyl tertiary butyl T c temperature of coalescence THF terahydrofuran trmpe MeC(CH2PMe2>3 T M E D A tetramethylethylenediamine UV-VIS ultraviolet-visible xvii ACKNOWLEDGEMENTS First and foremost, I would like to thank my supervisor Mike Fryzuk for all his support throughout my time here at UBC. Being a part of his ever-changing group has given me the opportunity to meet every kind of post-doc and grad-student imaginable. Thank-you, Pat, Myl, Guy, Cam, Neil, Lisa, Dave, Dave, Chas, Alan, Randy, Kiran, Jesse, Craig, Cindy, Brian, Bobbi, Warren, Graham, Pauline and Patrick for your own individual outlook on it all. I would also like to acknowledge the help I had from all the support staff here at UBC: the guys in the electrical, mechanical and glass-blowing shops, the NMR staff, P. Borda the elemental analysis expert, and Steve Rettig, the crystallographer who solved all the crystal structures in this thesis. Many thanks to all the grad-students whose interests (apart from chemistry) such as soccer, football, pub-crawls, skiing, hiking and auto-repair have allowed me a great deal of variation in my leisure time. And thank-you Michele for making these last two years so much better than the first four that I spent here in Vancouver. 1 CHAPTER 1: GENERAL INTRODUCTION 1.1 Transition Metal Phosphine Complexest The phosphine ligand has played a pivotal role in the development of transition metal coordination1 and organometallic chemistry.2 The synthetic flexibility and versatility of this ligand type have permitted the isolation of numerous unique and interesting complexes. The phosphorus donor has many beneficial features which can be used to design ligands to impart a desired stereo-electronic environment. It is possible to use a phosphine ligand which is a strong a-donor and either of small steric bulk (e.g., PMe3) or one which is enormous in relative size (e.g., PBul3). In addition, phosphorus donors capable of 7C-backbonding with the metal can be utilized and these are also available in a variety of sizes (e.g., the small PF3 or the much larger P(OBul)3 ligand). Phosphine donors have also been combined into chelate-type ligands to increase the stability of the ligand-metal interaction3, and this has aided the isolation and study of coordinatively unsaturated complexes. This donor type has been used in the synthesis of chiral ligands; complexes containing such ligands are used to bring about enantioselective processes in both the drug industry4 and the synthesis5 of natural products. In addition to this synthetic flexibility, this donor type allows an excellent insight into reactivity as phosphorus-31 is a spin 1/2 100% natural abundance nucleus. It is easily observable by NMR spectroscopy and both the chemical shift and the observed coupling constants yield valuable stereochemical and electronic information.6 t Throughout this thesis, the term "phosphine donor" is used to refer to a phosphorus-based ligand which forms a a-bond to a metal via donation of its lone pair of electrons. Included in this definition are phosphites, P(OR)3, while phosphides, PR2", are excluded. 2 While there are countless phosphine complexes of the transition metals, there are still areas of the periodic table for which such complexes are rare or even unknown.7 In particular, phosphine complexes of the early transition elements and lanthanides, where the metal is usually in a high oxidation state, are rare8 and these compounds are often unstable with respect to phosphine dissociation. One reason advanced for this instability is that these metals are hard acids9 while the phosphine ligand is considered to be a soft base.1 0 Therefore, such a mismatch in donor and acceptor should result in only weakly bonded phosphines. However, if the metal is exposed to no other ligand, such phosphine complexes should be isolable. It is with these early metal phosphine complexes (and the lack thereof) that this thesis is concerned. In particular, phosphine donor-containing complexes of the group 3 and 4 metals and the lanthanides will be discussed. 1.2 Phosphine Complexes of the Lanthanides and Group 3 Metals Table 1.1, listing the syntheses of all the known group 3 phosphine complexes, shows how rare are these species (new complexes synthesized for this thesis have been omitted from Tables 1.1-1.4). In fact, genuine phosphine complexes have only recently been reported. There is some doubt about the authenticity of the dppe adducts of SCX3 as the reported elemental analyses are very poor 1 1 , but the other three scandium phosphine complexes have been crystallographically characterized. These derivatives show how the small, basic PMe3 ligand can be used to stabilize and prevent oligomerization or decomposition of the the highly reactive scandium-hydride and -alkyl units. The chemistry associated with these PMe3 stabilized compounds has been reported1 2 to involve phosphine dissociation to yield a coordinatively unsaturated complex which then undergoes a variety of interesting organometallic reactions. The yttrium complex, Y(OCBu-2CH2PMe2)3, illustrates the use of hybrid 3 chelating ligands to generate phosphine complexes. 1 3 It is noteworthy that no phosphine complexes of lanthanum or actinium have been reported. Table 1.1: Phosphine Complexes of the Group 3 Metals. Complex Synthesis Year R e f-[ScCl3(dppe)]x ScCl 3 + dppe 196911 [ScBr 3(dppe)i. 5] x ScBr3 + dppe 196911 [Me 2Si(C 5Me 4 ) 2 ]ScH(PMe 3 ) [Me 2Si(C 5Me 4) 2]ScR + PMe 3 + H 2 19881-4 [(Cp*NBu-)Sc(n-H)(PMe3)]2 (Cp*NBu-)ScR + PMe 3 + H 2 19881 5 [(Cp*NBut)Sc(u-C 2H 4)(PMe 3)] 2 [(Cp*NBu-)Sc(ii-H)(PMe3)]2 + C 2 H 4 Y (OCBu t 2 CH 2 PMe 2 ) 3 YC1 3 + 3/2 [Li(OCBu-2CH 2PMe 2)]2 198813 Table 1.2 lists the syntheses of all the known lanthanide phosphine complexes. For only six of the fourteen lanthanide metals have such complexes been isolated. There is no reason to suspect that phosphine complexes of Pr, Pm, Sm, Gd, Tb, Dy, Er or Tm are not isolable. In fact Dr. David J. Berg, a former postdoctoral fellow in our laboratories, synthesized and isolated the first such complexes of Sm and Er, as well as other examples with Ce, Eu and Yb . 1 6 The first evidence for a lanthanide phosphine complex was obtained17 in 1965 and was based on the changes observed in the UV-VIS spectrum of Cp 3 Yb upon addition of PPh3. This adduct was isolated and characterized18 by elemental analysis. Thirteen years later another report was published, claiming the observation (via UV-VIS spectroscopy) of the formation of a variety of phosphine adducts of Cp 3 Yb. 4 However, only a poor elemental analysis of Cp3Yb(PH2Cy) was reported and the attempted isolation of other adducts led to mixtures contaminated with Yb(III) phosphides.19 Table 1.2: Phosphine Complexes of the Lanthanides. Complex Synthesis YearRef. Cp 3Yb(L): L = PPh 3 Cp 3Yb(L): L = PPh3, PHPh 2, PH2Ph, PCy 3, PHCy 2, PH 2Cy, PMe2Ph Yb[N(SiMe 3) 2] 2L 2: L 2 = 2PBu3, dmpe Eu[N(SiMe 3) 2] 2(dmpe)i. 5 Eu[N(SiMe 3) 2] 2(PBu 3) 2 Cp 3Yb(L): L = PEt 3 Cp*2Eu(L): L = dmpe, dmpm Cp*2Yb(L): L = dmpe, dmpm Cp*2YbCl(dmpm) Ho(BH 3Me) 3(dmpe)i. 5 Yb(BH 3Me) 3(dmpe)i. 5 Cp 2 Lu(CH 2 PMe 3 ) Cp 3 Ce(PMe 3 ) Cp' 3Ce[P(OCH 2 ) 3CEt] Nd(OCBu- 2CH 2PMe 2 ) 3 Cp 3 Yb + PPh3 19651 7 Cp 3 Yb + L (in situ) 197819 Yb[N(SiMe 3) 2] 2(OEt 2) + L 2 198220 2NaEu[N(SiMe 3) 2] 3 + 3 dmpe NaEu[N(SiMe 3) 2] 3 + 2 PBu 3 Cp 3 Yb + PEt3 198321 Cp* 2Eu(OEt 2) + L 19832 2 Cp* 2Yb(OEt 2) + L Cp*2Yb(dmpm) + YbCl 3 HoCl 3 + 3 LiBH 3 Me + dmpe 19852 3 YbCl 3 + 3 LiBH 3 Me + dmpe Cp 2LuCl + L iCH 2 PMe 3 19852 4 Cp' 3Ce(THF) + excess PMe 3 19872 5 Cp' 3Ce(THF) + P(OCH 2 ) 3CEt 198826 NdCl 3 + 3/2 [Li(OCBu t 2CH 2PMe 2 ) ] 2 19881 3 5 It was not until 1982 that the first lanthanide phosphine complex was fully characterized.20 Dmpe and PBU3 adducts of Yb[N(SiMe3)2l2 and Eu[N(SiMe3)2]2 were synthesized via displacement of coordinated diethyl ether, and the ytterbium derivative, Yb[N(SiMe3)2]2(dmpe), was characterized by X-ray crystallography. The X-ray structure revealed that this complex contained two unusual agostic silyl methyl groups, making the complex formally six-coordinate (Figure 1.1). Figure 1.1: A representation of Yb[N (S iMe3)2]2 (dmpe) , the first crystallographically characterized lanthanide phosphine complex. Since 1982 a few more phosphine complexes have been reported but such compounds are still rare. Three of these reports are of primary interest because they demonstrate that phosphine complexes of the lanthanides are not necessarily inherently unstable. Equilibrium studies showed21 that for Cp3Yb(L), PEt3 is as good a ligand as pyrrolidine and better than THF. The synthesis25-26 of Cp'3Ce(L) (L = PMe3, P(OCH2)3CEt) via displacement of THF by a phosphorus ligand demonstrated that this holds true for Ce(III) as well. These three examples prove that the phosphine donor can indeed coordinate strongly to these "hard" metal centres and in some cases will displace THF from the metal centre. Me3 Si SiMe3 6 Table 1.3: Group 3 and Lanthanide Metal Phosphorus Bond Lengths. Complex Bond Length C N a Crystal Radiusb Difference0 Cp' 3 Ce [P (OCH 2 )3CEt] 3.086 (3) 10 1.39 1.70 Cp ' 3 Ce ( PMe 3 ) 3.072 (4) 10 1.39 1.68 Cp*2YbCl(dmpm) 2.941 (3) 8 1.125 1.816 Yb [N ( S iMe 3 ) 2 ]2 (dmpe) 3.012 (4) 6 1.16 1.85 Nd (OCBu t 2 CH 2 P M e 2 ) 3 3.154 (2) 6 1.123 2.031 Y ( O C B u t 2 C H 2 P M e 2 ) 3 3.045 (2) 6 1.040 2.005 M e 2 S i ( C 5 M e 4 ) 2 S c H ( P M e 3 ) 2.752 (1) 8 1.01 1.74 [(Cp*NBu-)Sc(u-H)(PMe 3)] 2 2.996 (1) 7 0.95 2.05 [ (Cp*NBut)Sc(Lt-C 2 H 4 ) (PMe 3)]2 2.825 (3) 7 0.95 1.88 a) The formal coordination number at the metal centre. b) The effective ionic radius of the metal for that oxidation state and CN.; c) The difference between the M-P bond length and the crystal radius. The metal-phosphorus bond lengths of all of the crystallographically characterized complexes have been tabulated in Table 1.3; included in the table is the crystal radius for the particular metal in the appropriate oxidation state and coordination number. In the column labelled "Difference", the cation radius has been subtracted from the M—P bond length. The wide range of values thus obtained demonstrates that the variation in Ln—P bond length is not due to just the cation radius. Clearly the steric influence of the various other ligands about the metal has a large effect on the M—P interaction. 7 1.3 Phosphine Complexes of Hafnium and Zirconium In recent years, the phosphine donor has been recognized as a suitable ligand for hafnium and zirconium as evidenced by numerous publications in this field.8 However, by the end of 1979 (when work from this laboratory on early metal phosphine complexes was started) there were only 13 literature reports of such complexes, half of which were published in 1979. Table 1.4 lists the synthesis of these compounds in chronological order. The simple dppe adducts of ZrCL^ and HTCI4, reported28 in 1965, were the first phosphine derivatives of these metals. Seven years later it was shown29 (via N M R spectroscopy) that tetrabenzyl-zirconium and -hafnium formed an adduct with PMe3. In 1974, PPh3 and dppe were reported30 to be capable of displacing THF from PhZrCl 3(THF) 3 to give PhZrCl3(THF)(PPh3) or PhZrCl3(dppe). In 1977, the small basic PMe3 ligand was shown31 to bind to Zr(II) by displacing one of the Tt-acceptor CO ligands from Cp2Zr(CO)2 to form Cp2Zr(CO)(PMe3), while in 1978 it was demonstrated32 that PF3 would displace N2 from [Cp*2Zr(N2)]2(M--N2) to give [Cp*2Zr(PF3)]2(|i-N2). There were six publications in 1979 on a wide variety of compounds, demonstrating that the phosphine ligand could stabilize complexes in the +4, +3, +2 and 0 oxidation states. Clearly, the phosphine donor can be an excellent ligand for these metals no matter what the metal oxidation state. In fact, the phosphine donor is no longer an uncommon ligand for zirconium, and it is often used to stabilize low-valent complexes. 8 Table 1.4: Phosphine Complexes of Hafnium and Zirconium Prior to 1980. Complex Synthesis YearRef-MX4(dppe): M = Hf, Zr; X = CI, Br M(CH 2C 6H5) 4(PMe3): M = Zr, Hf PhZrCl 3(THF)(PPh 3) PhZrCl3(dppe) Cp* 2Zr(H) 2(PF 3) Cp 2Zr(CO)(PMe 3) [Cp* 2Zr(PF 3)] 2(u-N 2) ZrCl4(dmpe)2 ZrH(n,5-C6H7)(dmpe)2 [Zr(Ti4-C 4H 6) 2(dmpe)] 2(dmpe) Zr(ri4-C4H6)2(dmpe)(PMe3) (C 7H 8 ) 2M(PMe 3 ) : M = Hf, Zr Cp 2 ZrL 2 : L 2 = dmpe, dppe, 2 PPh2Me, 2 PPhMe 2 Cp 2Zr(PPh 2Me)(CO) [CpZr(L)(u.-Til:Ti5-C5H4)]2: L = PPh2Me, PPhMe 2 [Zr(Ti 4-C 4H 6) 2(dmpe)] 2(CO) [Zr(tl4.c 4H 6) 2(dmpe)] 2 [Zr(Ti4-C4H6)2(dmpe)]2(L): L = PMe 3, PMe 2Ph, P(OMe)3 Cp 2Hf(CO)L: L = PPh3, PMe 3, PF 3 HfCl3[(Cl)CNBut](dppe) 19652 8 197229 197430 ! 9 7 633 197731 19783 2 19793 4 M X 4 + dppe M ( C H 2 C 6 H 5 ) 4 + PMe 3 (in situ) PhZrCl 3(THF) 3 + PPH 3 PhZrCl 3 (CH 3 CN) 2 + dppe Cp* 2 Zr(H) 2 + PF 3 (in situ) Cp 2 Zr(CO) 2 + PMe 3 [Cp* 2Zr(N 2)] 2(u-N 2) + 2 PF 3 ZrCl4 + 2 dmpe ZrCl4(dmpe)2 + C6Hg + 2 Na/Hg ZrCU(dmpe)2 + C 4H6 + 2 Na/Hg [Zr(ri4-C4H6)2(dmpe)]2(dmpe) + 2 L Toluene + PMe 3 + Metal (g) at -196 °C Cp2Zr(H)R + L 2 Cp 2Zr(PPh 2Me) 2 + CO Cp 2 ZrL 2 + A [Zr(Tt4-C4H6)2(dmpe)]2(dmpe) + CO 19793 7 [Zr(Ti4-C4H6)2(dmpe)]2(CO) + vacuum [Zr(ri4-C4H6)2(dmpe)]2 + L Cp 2Hf(CO) 2 + L 197938 {HfCl3[(Cl)CNBut](CNBut)}2 + 2 dppe 19793 9 1 9 7 9 3 5 197936 9 The first crystallographically characterized phosphine complexes of zirconium and hafnium were reported in 1980 and 1981. The synthesis of ZrH(rj 5-CgHn)(dmpe)2 was achieved via the reduction of ZrCl4(dmpe)2 with Na/Hg in the presence of 1,3-cyclooctadiene.40 The hafnium(O) complex, Hf(C4H6)2(dmpe), was synthesized by the reaction of HfCU with two equivalents of [Mg(C4H.6)(THF)2]n ti-the presence of dmpe. 4 1 Since then, at least 35 zirconium and 7 hafnium complexes containing phosphorus donors have been characterized by X-ray crystallography. From this data, the average M—P bond-length is 2.75 A in a range that spans from 2.972 (2) A for Zr(CH 2SiMe3) 4dmpe 4 2 to 2.619 (1) A for Cp 2Zr(CO)[P(OMe) 3] 43. There does not appear to be any correlation between bond length and metal oxidation state. Figure 1.2: Some recently characterized group 4 phosphine complexes (left to right references 44, 45 and 46). More recently, the phosphine donor has been used to stabilize some interesting and unusual complexes (Figure 1.2). The zirconocene fragment has been stabilized by the coordination of two PMe3 ligands; crystallographic characterization of this complex reveals a very short Zr—P bond length of 2.64 A . 4 4 Benzyne has also been trapped on the zirconocene fragment45; it was synthesized by the thermolysis of M = Hf, Zr 10 Cp2ZrPb.2 in the presence of excess PMe3. The reduction of MCl4 (THF)2 , M = Hf, Zr, with potassium napthalenide in the presence of trmpe, followed by the addition of CO, led to the unique zero-valent comp lexes 4 6 shown in Figure 1.2. Numerous other examples are given in reference 8. 1.4 Summary of Previous Work from Our Laboratory This summary of early metal amido-diphosphine complexes is based on four full journal pape r s , 4 7 , 4 8 , 4 9 * 5 0 two journal communications, 5 1 , 5 2 and two M.Sc. theses. 5 3 , 5 4 In 1979, the synthesis of a new six-electron, hybrid ancillary ligand was developed which contained both an amido donor and two phosphine donors in a chelate array (Scheme 1.1).55 Scheme 1.1 Me2 Me2 Si Si CI H CI Me2 Me2 3 LiPR2; R = Me, PH, Bu* .Si^ Si \l I JR, LiBu Me? Me2 2 LiPPh2 Si Si f x O -(2 LiCl) p h p ^ rW -(Butane) The choice of ligand donors was deliberate and based on a number of simple observations: i) the amide donor was an excellent ligand for early transition metals in 11 high oxidation states; ii) the phosphine donor was considered to be the ligand of choice for low oxidation state late transition metals; iii) amide complexes of the late transition elements were rare; iv) phosphine complexes of the early transition metals were also rare; and v) the choice of ancillary ligand will often dictate the type of chemistry that occurs at a particular metal centre. What these five points indicate is that a ligand that incorporates both of the above donor types should be able to stabilize most metals in a variety of oxidation states and this could result in the synthesis of some unusual complexes (i.e., early metal phosphine complexes or late metal amides). Such complexes might display unique reactivity patterns as the ancillary ligand arrangement at the metal will be different from the more common six-electron donor, (C5H5)- or (CsMes)-. As it is quite simple to vary the substituents at the phosphine donor (Scheme 1.1), this ligand also displays a synthetic flexibility that makes available a variety of ligands with different stereo-electronic properties. The reaction of two equivalents of LiN(SiMe2CH2PR-2)2 (R = Me or Ph) with either ZrCl4 or HfCU gave the six-coordinate derivatives shown in Scheme 1.2. Each ligand is only bound in a bidentate fashion and the metals are so sterically crowded that the chlorides would not react with Grignard or organolithium reagents.47 As the bis(ligand) complexes were so unreactive, mono(amido-diphosphine) complexes were investigated (Scheme 1.2). Their syntheses were achieved by the addition of one equivalent of the lithium-amide to hafnium or zirconium tetrachloride; for the complexes with methyl groups at phosphorus long reaction times and high dilution were found to be necessary.48 12 Scheme 1.2 MCI4 + 2 LiN(SiMe2CH2PR2)2 EtpO -(2 LiCl) MCI4+ LiN(SiMe2CH2PR2)2 toluene 1—5 days -(LiCl) MCl2[N(SiMe2CH2PR2)2]2 M = Hf, Zr R = Me, Ph MCI3[N(SiMe2CH2PR2)2] M = Hf, Zr R = Me, PH, Bu l ^ \ — Si J \ N / MeoSi \ £1 Me 2 P 'C\ .....>SiMe 2CH 2PMe 2 \ R | / \ \ / H i ^ : Z r C M e * P C / ^ N S i M e 2 C H 2 P M e 2 Mfi„s/ni/y •Sr CI Me 2 N Z r—C I C l ^ CI N CI Me 2Si' CI fac-X-ray Structures PH2 mer-These mono(amido-diphosphine) compounds were found to be quite reactive and a series of derivatives were prepared. H f C l 3 [ N ( S i M e 2 C H 2 P M e 2 ) 2 ] was converted to Hf(BH4 )3 [N (SiMe2CH2PMe2)2L which in turn was transformed into either the binuclear bridging trihydride Hf2(BH4)3[N(SiMe2CH2PMe2)2]2(M--H)3 or the dimeric bridging tetrahydride H f 2 ( B H 4 ) 2 [ N ( S i M e 2 C H 2 P M e 2 ) 2 J 2 ( ^ - H ) 4 , depending on the reaction conditions used (Scheme 1.3).49-51 13 Scheme 1.3 HfCl3lN(SiMe2CH2PMe 2)2l LiBH4 • Hf(BH4)3[N(SiMe2CH2PMe2)2] Toluene NMe 3 5 days -(H3BNMe3) Hf2(BH4)3[N(SiMe2CH2PMe2)2]2(u-H)3 H 3 BSMe 2 NMe 3 Hf2(BH4)2[N(SiMe2CH2PMe2)2]2(u-H)4 -(H3BNMe3) A series of fluxional trimethyl complexes has also been synthesized. M C l 3 [N ( S iMe 2 C H 2 P R 2 ) 2 ] (M = Hf, Zr; R = Me, Pr', But) r e a c t s w i t h three equivalents of MeMgCl to yield MMe3[N(SiMe2CH2PR2)2l- All of these molecules are fluxional and their NMR spectra display only one type of methyl resonance even when recorded at -90 °C. Phosphine dissociation of one arm of the amido-diphosphine ligand to generate a five-coordinate intermediate is a possible rationale for the observed spectra.53 As the amido-diphosphine ligand should be capable of stabilizing a low-valent zirconium or hafnium complex, a series of butadiene complexes were synthesized. Thus, reaction of colourless MCl3[N(SiMe2CH2PR2)2l with "magnesium butadiene" yields the coloured diene complexes MCl(r)4-C4H6)[N(SiMe2CH2PR2)2] (orange, M = Hf; red, M = Zr). The zirconium complexes can also be obtained by Na/Hg reduction of the trichloride in the presence of excess butadiene. All of these compounds can also be further derivatized by metathesis of the remaining chloride with organolithium reagents (Scheme 1.4). As seen with bis(cyclopentadienyl) group 4 diene complexes, 14 the MCl(r i 4 -C4H6 ) [N (S iMe2CH2PR2)2] compounds are also fluxional. However, they undergo a fundamentally different type of process. The fluxional behaviour for Cp2M(diene) compounds has been described as an "envelope flip" mechanism56 with inversion of the diene unit occurring via the a2-7i resonance structure shown below. The diene compounds incorporating the amido-diphosphine ligand display a variable temperature NMR behaviour that is best rationalized by a diene rotation process. This diene rotation, as well as analysis of the X-ray crystal structure data, indicate that the r|4-7t resonance form is an acceptable description of the diene-metal bonding interaction.50 Scheme 1.4 15 Scheme 1.5 pVelimination T The hafnium complexes, HfCl(rj4-C4H6)[N(SiMe2CH2PR2)2] R = Me, Pr1, were also found to undergo a unique carbon-carbon coupling reaction when an allyl moiety was substituted for the remaining chloride (Scheme 1.5). For the derivative with methyl groups at phosphorus, the allyl-diene complex displayed NMR spectra in 16 accord with a complex that contained a rotating diene and an allyl unit undergoing syn-anti exchange. Over a period of days this complex slowly converted to the coupled product. The derivative with the bulkier isopropyl groups at phosphorus underwent the coupling reaction in about one hour. Scheme 1.5 shows two possible mechanisms for arriving at the observed allyl-diene coupled product.52 It is interesting to note that the analogous Cp*2M(diene)(allyl)5 7 and Cp2M(diene)(allyl)5 8 complexes do not undergo this reaction and demonstrate again how a change in the ancillary ligand can affect the observed chemistry. 1.5 The Scope of this Thesis Chapters 2 and 3 deal with mono(amido-diphosphine) complexes and bis(amido-diphosphine) complexes of lanthanum, yttrium and lutetium. The synthetic strategy used is a continuation of the methodology described above. In order to obtain phosphine complexes of the lanfhanoidst, substitution of a chloride for an amido ligand containing pendant phosphine arms was expected to result in isolable lanthanoid-phosphorus complexes. The two basic target molecules are mono(amido-diphosphine) complexes and bis(amido-diphosphine) complexes (Figure 1.3). The systematic synthesis and study of a variety of these complexes with different substituents at phosphorus were undertaken and the results are described in chapters 2 and 3. The choice of La + 3 , Y + 3 and L u + 3 as the cations to work with was deliberate. All three metals are diamagnetic in their common +3 oxidation state, thus facilitating NMR spectroscopic studies. In addition, their collective radii span that of the lanthanide series. Both of these features make complexes of these metals excellent model compounds for the whole lanthanide series. t Due to their similar physicochemical properties, Y, La and the lanthanide metals shall be referred to as lanthanoids. 17 mono(ligand) bis(ligand) Figure 1.3: Two target lanthanoid phosphine complexes. The initial goal of this work was first to prove that lanthanoid complexes containing M—P bonds could be isolated, then examine the stereochemistry of the resulting complexes and determine the nuclearity of the complexes (i.e., monomers or oligomers). After these points had been addressed, the organometallic chemistry was to be investigated to see if the incorporation of the amido-diphosphine ligand induced new kinds of reactivity at lanthanoid-carbon bonds. Chapters 4 and 5 deal with some chemistry of amido-diphosphine complexes of hafnium and zirconium. Chapter 4 is a continuation of the study of the unique allyl-diene coupling reaction described earlier. The proposed mechanism for this coupling reaction involves two intermediate complexes; therefore, an investigation to look for evidence indicating such species was initiated. This coupling has also been investigated using the zirconium diene complexes with allyl-MgCl and 1-methylallyl-MgCl. The chemistry in Chapter 5 is a result of the continuing investigations into the possible stabilization of lower oxidation state zirconium(amido-diphosphine) 18 complexes. This chapter describes the discovery and characterization of an unusual side-on bound dinitrogen zirconium complex, synthesized via the reduction of ZrCl3[N(SiMe2CH-2PPri2)2] in the presence of dinitrogen. 1.6 References McAuliffe, C. A. In Comprehensive Coordination Chemistry, Wilkinson, G.; Gillard, R. D.; McCleverty, J. A. (eds.), Pergamon Press: London, 1987, Vol 2, 989. Collman, J. P.; Hegedus, L. S.; Norton, J. R.; Finke, R. G. Principles and Applications of Organotransition Metal Chemistry, University Science Books: California, 1987, 66-72. Huheey, J. E. Inorganic Chemistry, 3 ra" ed.; Harper and Roe: New York, 1978; 527-534. a) Knowles, W. S. Acc. Chem. Res. 1983,16, 206. b) Haggin, J. Chem. Eng. News, 1990, May 7, 58. Davies, S. G. Organotransition Metal Chemistry: Applications to Organic Synthesis, Pergamon Press: Toronto, 1982, pp. 326-330. a) Nixon, J. F.; Pidcock, A. Ann. Rev. NMR Spec. 1969,2, 345. b) Garrou, P. E. Chem. Rev. 1981, 57, 229. c) Meek, D. W.; Mazanec, T. J. Acc. Chem. Res. 1981,14, 266. McAuliffe, C. A.; Levason, W. S. Phosphine, Arsine and Stibine Complexes of the Transition Elements, Elsevier: Amsterdam, 1979, 72. Fryzuk, M. D.; Haddad, T. S.; Berg, D. J. Coord. Chem. Rev., 1990, 99, 137. Ahrland, S.; Chatt, J.; Davies, N. R. Quart. Rev., Chem. Soc. 1958,12, 265. Pearson, R. G. /. Am. Chem. Soc. 1963, 85, 3533. 19 ' 1 0 For a general discussion of the hard-soft acid-base theory see: Huheey, J. E. Inorganic Chemistry, 3rc* ed.; Harper and Roe: New York, 1978; p. 312. 1 1 Greenwood, N. N.; Tranter, R. L. /. Chem. Soc, A. 1969, 2678. 1 2 Piers, W. P.; Shapiro, P. J.; Bunel, E. O.; Bercaw, J.E. Synlett. 1990,1, 74. 13 Hitchcock, P. B.; Lappert, M. F.; MacKinnon, I. A. /. Chem. Soc, Chem. Commun. 1988, 1557. 1 4 Bunel, E. E.; Schomaker, V.; Bercaw, J. E. unpublished results, 1988. 1 5 Shapiro, P. J.; Schaefer, W. P.; Bercaw, J. E. unpublished results, 1988. 1 6 Fryzuk, M. D.; Berg, D. J. unpublished results. 1 7 Fischer, R. D.; Fischer, H. J. Organomet. Chem. 1965, 4, 412. 1 8 Fischer, E. O..; Fischer, H. J. Organomet. Chem. 1966, 6, 141. 1 9 Bielang, G.; Fischer, R. D. /. Organomet. Chem. 1978,161, 335. 2 0 Tilley, T. D.; Andersen, R. A.; Zalkin, A. J. Am. Chem. Soc. 1982,104, 3725. 2 1 Schlesener, C. J.; Ellis, A. B. Organometallics, 1983, 2, 529. 2 2 Tilley, T. D.; Andersen, R. A.; Zalkin, A. Inorg. Chem. 1983, 22, 856. 2 3 Shinomoto, R. S. Energy Res. Abstr., 1985,10, Abstr. No. 10165; Chem. Abstr., 1985,103, 80844m. 2 4 Schumann, H.; Reier, F-W.; Palamidis, E. J. Organomet. Chem. 1985, 297, C30. 20 2 5 Stults, S.; Zalkin, A. Acta Cryst., C43,1987, 430. 2 6 Brennan, J. G.; Stults, S.; Andersen, R. A.; Zalkin, A. Organometallics, 1988, 7, 1329. 2 7 Shannon, R. D. Acta Cryst., Sect. A, 1976,32, 1706. 2 8 Ray, T. C , Westland, A. D. Inorg. Chem. 1965,4, 1501. 2 9 Felton, J. J., Anderson, W. P. /. Organomet. Chem. 1972,36, 87. 3 0 Clarke, J. F.; Fowles, G. W. A.; Rice, D. A. J. Organomet. Chem. 1974, 76, 349. 3 1 Demersemen, B.; Bouquet, G.; Bigorgne, M. /. Organomet. Chem. 1977,132, 223. 3 2 Manriquez, J. M.; McAlister, D. R.; Rosenberg, E.; Shiller, A. M.; Williamson, K. L.; Chan, S. I.; Bercaw, J. E. J. Am. Chem. Soc. 1978,100, 3078. 3 3 Manriquez, J. M.; McAlister, D. R.; Sanner, R. D.; Bercaw, J. E. /. Am. Chem. Soc. 1976, 98, 6733. 3 4 Datta, S.; Wreford, S. S.; Beatty, R. P.; McNeese, T. J. J. Am. Chem. Soc. 1979,101, 1053. 3 5 Geoffrey, F.; Cloke, N.; Green, M. L. H. /. Chem. Soc, Chem. Commun. 1979, 127. 3 6 Gell, K. I.; Schwartz, J. /. Chem. Soc, Chem. Commun. 1979, 244. 3 7 Beatty, R. P.; Datta, S.; Wreford, S. S. Inorg. Chem. 1979,18, 3139. 3 8 Sikora, D. J.; Rausch, M. D.; Rogers, R. D.; Atwood, J. L. /. Am. Chem. Soc. 1979,101, 5079. 21 3 9 Behnam-Dehkordy, M.; Crociani, B.; Nicolini, M.; Richards, R. L. /. Organomet. Chem. 1979,181, 69. 4 0 Fischer, M. B.; James, E. J.; McNeese, T. J.; Nyburg, S. C ; Posin, B.; Wong-Ng, W.; Wreford, S. S. /. Am. Chem. Soc. 1980,102, 4941. 4 1 Wreford, S. S.; Whitney, J. F. Inorg. Chem. 1981,20, 3918. 4 2 Cayias, J. Z.; Babaian, E. A.; Hrncir, D. C ; Bott, S. G.; Atwood, J. L. J. Chem. Soc, Dalton Trans. 1986, 2743. 4 3 Erker, G.; Dorf, U.; Kruger, C ; Angermund, K. J. Organomet. Chem. 1986,301, 299. 4 4 Kool, L. B.; Rausch, M. D.; Alt, H. G.; Herberhold, M.; Honold, B.; Thewalt, U. /. Organomet. Chem. 1987,320, 37. 4 5 Buchwald, S. L.; Watson, B. T.; Huffman, H. C. /. Am. Chem. Soc. 1986,108, 7411. 4 6 Blackburn, D. W.; Chi, K. M.; Frerichs, S. C ; Tinkham, M. L.; Ellis, J. E. Angew. Chem., Int. Ed. Engl. 1988, 27, 437. 4 7 Fryzuk, M. D.; Williams, H. D.; Rettig, S. J. Inorg. Chem. 1983, 22, 863. 4 8 Fryzuk, M. D.; Carter, A.; Westerhaus, A. Inorg. Chem. 1985,24, 642. 4 9 Fryzuk, M. D.; Rettig, S. J.; Westerhaus, A.; Williams, H. Inorg. Chem. 1985, 24,4316. 5 0 Fryzuk, M. D.; Haddad, T. S.; Rettig, S. J. Organometallics, 1989, 8, 1723. 5 1 Fryzuk, M. D.; Williams, H. Organometallics, 1983,2, 162. 5 2 Fryzuk, M. D.; Haddad, T. S.; Rettig, S. J. Organometallics, 1988, 7, 1224. 22 Carter, A. M.Sc. Thesis, UBC, 1985. Haddad, T. S. M.Sc. Thesis, UBC, 1987. a) Fryzuk, M. D.; MacNeil, P. A. J. Am. Chem. Soc. 1981,103, 3592. b) Fryzuk, M. D.; MacNeil, P. A ; Rettig, S. J.; Secco, A. S.; Trotter, J. Organometallics, 1982,1, 918. c) Fryzuk, M. D.; Carter, A.; Westerhaus, A. Inorg. Chem. 1985, 24, 642. a) Erker, G.; Kruger, C ; Muller, G. Adv. Organometal. Chem. 1985,24, 1. b) Yasuda, H.; Nakamura, A. Acc. Chem. Res. 1985,18, 120. Blenkers, J.; de Liefde Meijer, H. J.; Teuben, J. H. /. Organomet. Chem. 1981, 218, 383. a) Erker, G.; Berg, K.; Kruger, C ; Muller, G.; Angermund, K.; Benn, R.; Schroth, G. Angew. Chem., Int. Ed. Engl. 1984,23, 455. b) Erker, G.; Berg, K.; Benn, R.; Schroth, G. Angew. Chem., Int. Ed. Engl. 1984, 23, 625. 23 CHAPTER 2: BIS(AMIDO-DIPHOSPHINE) COMPLEXES OF LANTHANUM, YTTRIUM AND LUTETIUM 2.1 Introduction This chapter is devoted to the chemistry of bis(amido-diphosphine) complexes of lanthanum, yttrium and lutetium. The synthesis and characterization of a series of such phosphine complexes with variation of both the metal and the type of phosphorus donor are described: MCl[N(SiMe2CH2PR2)2k R = Me, M = Y, La, Lu (a series); R = Ph, M = Y, La (b series) and R = Pr', M = Y (c series).''' In addition, the synthesis of some hydrocarbyl derivatives for the a series, their characterization, and the thermolysis of these organometallic species are presented. Lanthanoid hydrocarbyl complexes are rare and only recently have general procedures for their isolation been developed.1 In fact, a 1956 publication by Wilkinson and Birmingham states2: "it seems to be fairly certain that alkyl and aryl derivatives of scandium, yttrium, lanthanum and the rare earth elements either do not exist or have an existence so transitory that they cannot be isolated." Generally speaking, lanthanoid hydrocarbyl complexes are highly reactive, often unstable species, but given the right ancillary ligand arrangement about the metal such compounds can be isolated. A number of interesting hydrocarbyl lanthanoid complexes have been recently synthesized; a description of three such complexes will demonstrate their reactivity. t Throughout this thesis, the subscripts a-d shall be used to denote the type of amido-diphosphine ligand [N(SiMe2CH2PR2)2]" in the complex: a, R = Me; b, R = Ph; c, R = Pr'; d, R = Bul. 24 The lutetium complex Cp*2LuCH3(OEt2) was the first compound to directly model olefin insertion into a metal-alkyl bond, the lanthanide model for Ziegler-Natta polymerization. 3 This lutetium methyl complex is extremely reactive and will polymerize ethylene even at -90 °C. It was the first organometallic complex for which a well characterized reaction with methane was demonstrated and it also provided the first example of (3-methyl elimination at a metal centre.4 The [(CsH4R)2LnH(THF)]2 (Ln = Lu, Er, Y; R = H , Me) complexes, obtained by hydrogenolysis of (C5H4R)2Ln(Bu l)(THF), were the first fully characterized organometallic lanthanide hydrides 5, and provided the opportunity to study the reactivity of the lanthanide-hydride moiety in discrete well defined complexes.6 The Cp*2ScR complex7 has been used to generate the highly reactive deuteride Cp*2ScD which, under a D2 atmosphere, catalyses multiple H/D exchange converting Cs ( C H 3 ) 5H into C 5 ( C D 3 ) 5 H . 8 The isotopically labelled Cs ( CD3 )5H can easily be converted to perdeuterio LiCp* and used as a synthon for deuterium labelled transition metal Cp* complexes. 2.2 Synthesis of Bis(amido-diphosphine) Complexes The synthesis of bis(amido-diphosphine) complexes is relatively straight-forward. Addition of two equivalents of the amide salts [N (S iMe2CH2PR2)2]" to YCI3, L.UCI3 or LaCl3 in THF generates the desired compounds (Equation 2.1). THF MCI 3 + 2M'[N(SiMe 2CH 2PR 2) 2] • MCI[N(SiMe2CH 2PR 2) 2] 2 [2.1] M' = K or Li I LiCI^ M = Y : R = M e 1 a R = Ph 1b R = Prj 1c M = Lu: R = Me 2a M = La: R = Me 3a R = Ph 3b 25 With the smaller cations L u + 3 and Y + 3 (seven-coordinate ionic radii of 1.06 A and 1.10 A)9 either amide salt will generate good yields of the bis(ligand) complexes. However, the larger L a + 3 (ionic radius of 1.24 A)9 does not react with the lithium amides; long reaction times with the potassium salts are required to obtain the phosphine complexes. Scheme 2.1 Lu 8a La 9a The yttrium and lutetium bis(ligand) complexes MCl[N(SiMe2CH2PMe2)2]2, l a and 2a, can be further derivatized to generate hydrocarbyl complexes MR[N(SiMe2CH2PMe2)2]2» 4a-6a, which can be isolated and stored under nitrogen indefinitely as crystalline solids. These hydrocarbyl compounds are thermally unstable in solution and undergo a clean cyclometallation reaction by abstracting a 26 PCH2Si-methylene hydrogen to generate the cyclometallated product 7a or 8a v ia loss of R — H (Scheme 2.1). For the corresponding lanthanum complex LaCl [N(S iMe2CH2PMe2)2l2» 3a, reaction with phenyl l i thium yielded only the cyclometallated complex 9a; the lanthanum-phenyl derivative could not be observed at room temperature. The reaction of L iCH2S iMe3 with la or 3a leads directly to the cyclometallated products 7a and 9a. It is noteworthy that using T H F as a solvent in any of the aforementioned reactions has not resulted in formation of any T H F adducts. In fact, the N M R spectra of the MC l [N ( S iMe2CH 2 PMe2 )2 ]2 compounds obtained in THF/C6D6 or C6D6 are identical. This perhaps indicates that these compounds are coordinatively saturated. 2.3 NMR Characterization and Fluxional Behaviour A l l of the seven-coordinate chloro and hydrocarbyl complexes are highly f lux iona l as e x p e c t e d . 1 0 For those compounds with methyl substituents at phosphorus (a-series) the amido-diphosphine l igand displays three resonances in each of the room temperature ^CpH} and *H N M R spectra (i.e., the phosphorus methyl hydrogens and carbons, the methylene protons and carbons, and the s i ly l methyl hydrogens and carbons respectively) as shown in Figure 2.1. A t low temperatures, the *H N M R spectra of the chloro derivatives la, 2a and 3a, are uninformative as broad signals are observed right down to -95 °C. However, the low temperature *H N M R spectra of the phenyl complexes MPh[N(S iMe2CH2PMe2)2 l2 . 4a and 6a, display proton resonances for four different si ly l methyl, four phosphorus methyl and four methylene groups, while the low temperature ^CpH} N M R spectra show four s i ly l methyl, four phosphorus methyl and two methylene carbon resonances (Figure 2.2). 27 PCH 3 S i C H 3 H m SiCH2P Hc 8 6 8.2 7.8 7.4 PPM 1.2 0. 8 0.4 0 . 0 P P M SiCH2P Cm C r L L PCH 3 SiCH, j I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I [ I I I I I I I I 145 140 135 130 125 PP«5 20 15 10 5 PPM Figure 2.1: 300 MHz *H NMR spectrum (top, in C6D 6 ) and 75.4 MHz ^C{lH) NMR spectrum (bottom, in C6D5CD3) of Y(Ph)[N(SiMe2CH2PMe2)2]2 recorded at room temperature. Solvent peaks are marked with an asterisk. 28 8 . 6 8 P C H , H m H S i C H 2 P [f \J S i C H ? V S i C H 2 P 2 7 . 8 7 . 4 7 . 0 PPM6 1 . 2 0 . 8 0-. 4 " 0 . 0 * c J m Ul P C H 3 S i C H 2 P SiCHs I I I | I I I I | I I II | I I II | I I I I I I I [ I I I I | I II I | I I I I | I I I 145 140 135 130 125 PP«5 20 15 10 5 PPM Figure 2.2: 300 MHz *H NMR spectrum (top) and 75.4 MHz 13cpH} NMR spectrum (bottom) of Y(Ph)[N(SiMe 2CH 2PMe 2)2]2 in Q D 5 C D 3 recorded at -49 °C. Solvent peaks are marked with an asterisk. 29 Similar results are obtained from the 31p{ lH } N M R spectra. The spectra obtained at room temperature show only one phosphorus resonance (a doublet for the yttrium derivatives) indicating four equivalent phosphines in fast exchange, while at low temperature two phosphorus environments can be discerned. For the chloro derivatives la-3a, only the yttrium complex la shows an analyzable pattern (Table 2.1) while the others show two poorly resolved signals. However, the yttrium- and lutetium-phenyl complexes 4a and 6a display wel l resolved spectra (AA ' BB ' for L u and A A ' B B ' X for spin 1/2 Y ) . The 3 1 P { 1 H} N M R spectra obtained at low temperature have been simulated; the experimental and simulated patterns for 4a and 6a are shown in Figures 2.3 and 2.4, while the coupling constants are tabulated in Table 2.1. Of the three standard seven-coordinate geometries (pentagonal bipyramid, capped octahedron or capped trigonal prism), assigning these complexes a pentagonal bipyramidal geometry generates the most reasonable structures. The three possible geometrical isomers which satisfy the observed N M R data are shown in Figure 2.5. The amido-diphosphine ligands are fac-fac (C2 symmetry where the phosphine donors of each ligand are cisoid), mer-mer (C2 symmetry where the phosphine donors of each ligand are transoid), or fac-mer (Cs symmetry where the phosphine donors of one ligand are transoid and the other are cisoid). A t this point there is no definitive way to distinguish among these possibilities, but the mer-mer arrangement of the ligands is the most reasonable as it would have the least strain induced by the planar Y — N R 2 moiety. In an ideal pentagonal bipyramid the P — M — P bond angle would be 144° for a mer ligand and 72° for a fac ligand. Further evidence indicating the preference for this ligand to bind in the mer fashion comes from some previous X-ray crystallographic results. From twelve structures the average P — M — P angle is 141°, the smallest P— M — P angle ever observed 1 1 for this ligand is 102° (in HfCl3 [N(SiMe2CH2PMe2)2]) and the two yttrium structures which have been solved (vide supra) have this angle between 148 and 160 °. 30 Figure 2.3: 121 MHz 31p{lH} NMR spectrum of Y(Ph)[N(SiMe2CH2PMe2)2]2 in C6D5CD3 recorded at -49 °C (top) and a simulation (bottom) using Bruker PANIC software. See Figure 2.5 and Table 2.1 for details. 31 Figure 2.4: 121 MHz 31p{lH} NMR spectrum of Lu(Ph)[N(SiMe2CH 2PMe2)2]2 in C6D5CD3 recorded at -59 °C (top) and a simulation (bottom) using Bruker PANIC software. See Figure 2.5 and Table 2.1 for details. 32 fac, mer Figure 2.5: Three possible structures based on a pentagonal bipyramidal geometry for MR[N(SiMe2CH2PMe2)2]2 complexes in the low temperature limit. The 3 1P{ 1H} NMR spin system in the pentagonal plane is also shown. 33 It is noteworthy that the magnitudes of the two bond phosphorus couplings (2Jp-p) correlate with the assumed positions in the equatorial pentagonal plane in that transoid are greater than cisoid couplings. These coupling constants also agree with previously reported12 2Jp.p values of 53 to 80 Hz for trans disposed phosphines in the Group 4 complexes, MR(ti 4-C 4H 6)[N(SiMe2CH2PMe2)2] (M = Hf, Zr; R = CI, Ph, Np). Table 2.1: Up.y and 2JP.p Coupling Constants for MR[N(SiMe2CH2PMe2)2]2-Coupling M=Y, R=Ph,» 4a M=Lu, R=Ph,b 6a M=Y, R=C1,C l a 2 JPA-P A ' +21.9 ±19.6 ±18.0 2 JPA-PB - 6 . 3 - 7 . 8 5 - 2 . 0 2JPA-PB- +49.1 +70.75 +49.0 2 JP A-PB +49.1 +70.75 +49.0 2JPA.-PB. - 6 . 3 - 7 . 8 5 - 2 . 0 2JP B .p B . ±44.1 ±62.1 ±80.0 1JPA-Y 36.5 — 52.3 1 JPA-Y 36.5 — 52.3 ^PB-Y 51.3 — 50.8 1JPB..Y 51.3 — 50.8 a) At - 4 9 °C: 8 P A = 8 P A « = -43.6 ppm; 5 P B = 8 P B - = -48.8 ppm. b) At - 5 9 °C: 5PA = 5PA- = -39.7 ppm; 8 P B = 5 P B - = -46.4 ppm. c) At - 9 0 °C: 8 P A = 5P A ' = -43.8 ppm; 8 P B = 6 P B - = -45.9 ppm. 34 To account for the observed fluxional behaviour of these phosphine complexes it is tempting to invoke pseudorotation or stereochemical nonrigidity, a known process for many seven-coordinate complexes in which no metal-ligand bonds are broken.10 However, a simpler explanation is that these processes occur by a series of rapid equilibria involving dissociation of the phosphine arms of the tridentate ligands, rotation about the M—NR2 bond, followed by reassociation. The free energy of activation, AG*, for the process that renders all four phosphines equivalent changes only slightly for the La (10 Kcal/mol) to Y (10.1 Kcal/mol) to Lu (9 Kcal/mol) series of M C l [ N ( S i M e 2 C H 2 P M e 2 ) 2 l 2 complexes. For the phenyl derivatives, M P h [ N ( S iMe2CH2PMe2)2k M = Y and Lu, the AG* are also very similar, 11.5 Kcal/mol for yttrium and 11.0 Kcal/mol for lutetium. When a more bulky ancillary ligand is coordinated to yttrium, the variable temperature NMR behaviour changes. For the complex which has the bulky, but less basic, phenyl groups at phosphorus, YCl[N (SiMe2CH2PPh 2 )2]2 l b , the 3 1P{ !H} NMR spectrum recorded at +29 "C shows the expected doublet ^Jy-P = 39 Hz), indicating four equivalent phosphines. However, at -59 °C the spectrum clearly shows four different phosphorus environments. This spectrum has been simulated and both the observed and calculated spectrum along with the coupling constants are shown in Figure 2.6. Three of the phosphines are coordinated to the metal, as shown by the lJy-P coupling constants, while the fourth appears to be dangling freely in solution (a sharp singlet which displays no couplings). The complex with a six-coordinate distorted octahedral geometry shown in Figure 2.6 is a possible structure for this compound. An analogous six-coordinate species from the a-series compounds M R [ N ( S iMe2CH 2 PMe2 )2 l2 . la-6a, could be an undetectable intermediate in the fluxional process which exchanges all four phosphine sites. Figure 2.6: 121 MHz 3 1P{ ]H} NMR spectrum of YCl[N(SiMe2CH2PPh2)2]2 in C6D5CD3 recorded at -59 °C (top) and a spectral simulation using Bruker PANIC software (below). 36 For the lanthanum complex with phenyl groups at phosphorus LaCl[N(SiMe2CH2PPh2>2]2» 3b, the 3 1P{ 1H) NMR spectra are again different as an equilibrium between two species is apparent (Figure 2.7). Above 80 °C only one compound is present. It displays NMR spectra typical of the seven-coordinate fluxional species: YCl[N(SiMe2CH2PR2)2]2, R = Me, Ph. However, below -40 °C, a different complex is present, one that displays two uncoupled phosphorus environments which exchange with one another at higher temperature. Between +80 °C and -40 °C both complexes are observable. It is possible that a monomer/dimer equilibrium is occurring, the dimer being favored at low temperature. That this monomer/dimer behaviour is not observed for yttrium and lutetium could be due to the smaller size of these metals relative to lanthanum, as the larger lanthanum is capable of the eight coordination geometry that a chloride-bridged dimer would require. It is possible that the complex LaCl[N(SiMe2CH 2PMe2)2]2> 3a, also participates in such an equilibrium. However, at the crucial temperatures where differentiation between the two species might be observed, the 3 1P{ 1H) NMR spectra of this compound are complex and only show broad lines; thus, no conclusions can be drawn. Molecular weight determinations show that 3a, 3b, and l a are monomelic in C^D^ at 20 °C (at concentrations of 10 - 20 mg / mL). 37 | I M I I I M I | U I I I I I I I | I I I I I I I I I | I I I I I I I I I | I I I I | I I I I ) U ' I | I I I I | I I I I | I I I I | I I I I | I I I I | I I I I | I I I I | I I I I | I I I I | I I I I -10 -12 -14 -16 -18 -20 -22 -24 -26 -26 PPM Figure 2.7: 121 MHz 3lp{lH} NMR spectra of LaCl[N(SiMe 2CH 2PPh2)2]2 in CD3C6D5 as a function of temperature. The actual concentration is unknown. An impurity, HN(SiMe2CH2PPh2)2> is indicated with an asterisk. 38 It is interesting to note that for the complex with the bulky and very basic isopropyl phosphorus substituents YCl[N(SiMe2CH2PPr i2)2l2. lc. yet another type of temperature dependent behaviour is observed. At room temperature a doublet (Uy-p = 44 Hz) indicating four equivalent phosphines is observed in the 3 1 P{ 1 H) NMR spectra. This doublet appears to go through "coalescence" around -40 "C and at least three new resonances appear, two singlets and one doublet (*Jy-p = 89 Hz) at -80 °C. A low temperature limit cannot be achieved in C6D5CD3 (at lower temperatures further line broadening occurs) and therefore it is difficult to reconcile these results. No further work was done with this complex because its high solubility made purification difficult. The cyclometallated complexes 7a-9a, derived from the loss of R-H from the hydrocarbyl complexes MR[N(SiMe2CH2PMe 2 ) 2 ]2 (See Scheme 2.1) are also fluxional. Although the lanthanum complex 9a does not reach a low temperature limit even at -100 °C, at -29 °C for yttrium and +20 °C for lutetium, static 3 1P{ 1H} NMR spectra with four inequivalent phosphines are obtained (Figures 2.8 and 2.9, Table 2.2). At higher temperatures (T c = -79 °C for lanthanum, +65 °C for yttrium and +90 °C for lutetium) two of the phosphines exchange, while the metallated ligand remains rigid (AG* = 8.8 Kcal/mol for lanthanum (estimated), 15.3 Kcal/mol for yttrium and 16.3 Kcal/mol for lutetium) The simplest explanation for these observations is that in solution, the non-metallated ligand has phosphines which rapidly come on and off the metal. This enables these two phosphines to exchange coordination sites via rotation about the M—NR2 bond. One end of the metallated ligand may be undergoing dissociation-reassociation as well, but this is difficult to substantiate. 39 M&2 M&z '1 I I I I I I I I I I I I I I I I I I I I I 1 I I I I I I I I I I I I I -34.0 -34.5 -41.5 -42 0 -42.5 -45.0 -45.5 Figure 2.8: 121 MHz 31p{lH} NMR spectrum (top) and simulation (bottom) of i 1 Y[N(SiMe 2CHPMe 2 ) (SiMe 2CH 2PMe 2 ) ] [N(SiMe 2CH 2PMe 2 ) 2 ] , 7a, in C D 3 C 6 D 5 recorded at -29 °C. See Table 2.2 for details. 40 Figure 2.9: 121 MHz 3 1P{ JH} NMR spectrum (top) and simulation (bottom) of L , u[N(SiMe2 ,CHPMe 2)(SiMe 2CH 2PMe 2)][N(SiMe 2CH 2 PMe2)2], 8a, in C D 3 C 6 D 5 recorded at -29 °C. See Table 2.2 for details. 41 Table 2.2: Up.y and 2Jp.p Coupling Constants for M[N(SiMe 2CHPR 2 ) (SiMe 2CH 2PR 2 ) ] [N(SiMe2CH 2PR 2) 2]. Coupling M=Lu, R=Me a, 8a M=Y, R=Me b, 7a M=Y, R=Phc, 7b 2 j P l - P 2 7.9 13.45 12.2 2 j P l - P 3 29.3 18.46 2.2 2 jPl-P4 30.3 20.86 12.8 2 Jp 2 -P 3 32.5 11.52 8.5 2 JP 2 -P 4 9.4 8.20 0 2 JP 3 -P 4 21.2 18.54 23.8 — 70.00 32.7 V Y — 36.61 38.1 — 47.00 63.4 1 JP 4 -Y — 67.98 61.1 a) At +20 °C: 8P1 = -38.6.ppm; 8P2 = -34.8 ppm; 6P3 = -36.9 ppm; 8P4 = -40.1 ppm. b) At -29 °C: 5P1 = -42.0 ppm; 8P2 = -34.2 ppm; 8P3 = -41.9 ppm; 8P4 = -45.3 ppm. c) At -19° C: 8P1= -8.8 ppm; 8P2 = +0.4 ppm; 8P3 = -7.0 ppm; 8P4 = -14.8 ppm. Addition of L i C H 2 S i M e 3 to YC l [N (S iMe 2 CH 2 PPh 2 ) 2 ] 2 , l b , the yttrium complex with phenyl groups at phosphorus, generates the ligand metallated complex Y[N(SiMe 2CHPPh 2)(SiMe 2CH 2PPh 2)][N(SiMe 2CH 2PPh 2) 2], 7b. This compound shows similar NMR spectra to those exhibited by 7a-9a (i.e., at low temperature a seven-coordinate complex with four inequivalent phosphorus resonances is observed, while at high temperature the two phosphines of the non-metallated ligand undergo chemical exchange). The static spectra for 7a, 8a and 7b have been simulated; the 42 experimental and calculated spectrum for 7a and 8a are shown in Figures 2.8 and 2.9, while the coupling constants for all three complexes are listed in Table 2.2. The 8 9 Y NMR spectra of two yttrium complexes were also obtained. The spectrum of Y[N(SiMe2CHPMe 2)(SiMe2CH2PMe 2)][N(SiMe2CH 2PMe 2) 2], 7a, and YCl[N(SiMe 2CH 2 PMe2)2l2. la. are shown in Figure 2.10. Both complexes display ^p.y couplings. Complex 7a shows an overlapping "doublet of doublet of triplets" pattern due to the coupling of two inequivalent and two equivalent phosphorus donors to the metal. Complex l a shows a much simpler five line pattern due to the coupling of four equivalent phosphines to the yttrium centre. These couplings match those observed in the 3 ^^H} NMR spectra. Both l a and 7a resonate considerably downfield of aqueous 3 M YCI3 (the standard set at 0 ppm), l a at +449 ppm and 7a at +533 ppm. There are not many compounds for which 8 9 Y NMR data are available to compare. The first papers1 3 published using this technique were on simple inorganic salts and demonstrated that such yttrium containing species resonate over a wide chemical shift range ( +150 to -130 ppm), display long relaxation times (about three minutes) and their chemical shifts are temperature and concentration dependant. More recently, a publication14 on 8 9 Y NMR spectra of some organometallic species reported the chemical shift range for a variety of [ C p 2 Y R ] n (n = 1, 2; R= CI, CH3, H) compounds to be between -371 ppm and +40 ppm. Another paper reports15 the 8 9 Y chemical shifts of [H2B(pz)2J3Y and [H2B (3,5-Me2pz )2 l3Y to be +238.8 and +105.6 ppm. Taken together, these data demonstrate that the coordination of the amido-diphosphine ligand to yttrium results in a metal that is more deshielded than when a cyclopentadienyl or pyrazolylborate ligand is used. It is worth noting that the chemical shift of l a was found to be +449 ppm in both C6D5CD3 and in a THF/C6D5CD3 mixture. 43 Figure 2.10: 8 9 Y NMR spectra of l a (top, 19.6 MHz) in CD3C6D5/THF and 7a (middle, 24.6 MHz) in CD3C6D5. The theoretical spectrum of 7a, based on the couplings constants in Table 2.2, is also shown (bottom). 44 2.4 X-ray Crystal Structure Determination of Y [ N ( S i M e 2 C H P M e 2 ) ( S i M e 2 C H 2 P M e 2 ) ] [ N ( S i M e 2 C H 2 P M e 2 ) 2 ] , 7a The molecular structure of 7a (Figure 2.11, Tables 2.3 and 2.4) is best described as a capped octahedron. The two nitrogens and the four phosphines form a distorted octahedron with the metallated carbon as the capping ligand. The low temperature 31P{!H} NMR spectrum is consistent with this structure and the atom designations PI, P2, P3 and P4 in the crystal structure have been assigned to the 3lp{lH} NMR spectrum (Table 2.2). PI and P2, which belong to the metallated ligand, are the non-fluxional phosphines. As P2 is in a strained three-membered ring, and therefore in the most different chemical environment, its assignment was made on the basis of the most unusual chemical shift (-34.2 ppm). P2 also has the closest contact with the metal, therefore we would expect it to show the largest coordination contact shift. This then places PI at -42.0 ppm. The assignments of the fluxional P3 (at -41.9 ppm) and P4 (at -45.3 ppm) atoms are purely arbitrary as there is no definitive way of distinguishing between these atoms. Table 2.3: Selected Bond Lengths for i 1 Y[N(SiMe2CHPMe2)(SiMe2CH2PMe2)][N(SiMe2CH2PMe2)2], 7a. Bond Length (A) Bond Length (A) Bond Length (A) Y—C2 2.65 (1) Y—NI 2.256 (7) Si4—C4 1.86 (1) Y—P2 2.817 (3) Y—N2 2.396 (7) P2—C2 1.73 (1) Y—PI 3.005 (3) Si2—C2 1.84 (1) PI—CI 1.80 (1) Y—P3 2.896 (3) Sil—CI 1.90 (1) P3—C3 1.80 (1) Y—P4 2.903 (3) Si3—C3 1.91 (1) P4—C4 1.80 (1) 45 Figure 2.11: Chem 3D® core view (top) and ORTEP stereoview (bottom) of Y[N(S iMe 2 CHPMe 2 ) (S iMe 2 CH 2 PMe 2 ) ] [N(S iMe 2 CH 2 PMe 2 ) 2 ] , 7a. 46 Table 2.4: Selected Bond Angles for Y[N(SiMe 2 CHPMe 2 ) (SiMe 2 CH 2 PMe 2 ) ] [N(SiMe 2 CH 2 PMe 2 ) 2 ] , 7a. Bonds Angle (deg) Bonds Angle (deg) Pl-- Y - -P2 161.38 (9) N l--Y—N2 165.2 (3) Pl-- Y - -P3 85.24 (8) P l --Y—C2 134.3 (2) Pl-- Y - -P4 86.5 (1) P2--Y—C2 36.7 (2) P2-- Y - -P3 106.2 (1) P3--Y—C2 83.9 (2) P2-- Y - -P4 90.0 (1) P4--Y—C2 121.3 (2) P3-- Y - -P4 150.2 (1) N l--Y—C2 69.5 (3) Pl-- Y - -NI 71.5 (2) N2--Y—C2 125.2 (3) P2-- Y - -NI 90.9 (2) Y— -C2—P2 76.8 (4) P3-- Y - -NI 106.7 (2) Y— -P2—C2 66.5 (4) P4-- Y - -NI 97.6 (2) Y - -NI—Si2 101.0 (3) Pl-- Y - -N2 94.4 (2) Y - -C2—Si2 84.7 (4) P2-- Y - -N2 102.6 (2) Nl--Si2—C2 104.4 (4) P3-- Y - -N2 75.6 (2) Si2-—C2—P2 123.3 (6) P4-- Y - -N2 76.6 (2) The four phosphorus-yttrium bond lengths of 3.005 (3) A, 2.903 (3) A, 2.896 A (3) and 2.817 (3) A span a large bond length domain. There is only one other r e p o r t e d 1 6 Y — P bond length to compare with; the six coordinate Y [OCBu l 2CH2PMe2]3 complex (ionic radius 1.040 A) has a Y — P bond length of 3.045 (2) A. The fact that the seven-coordinate complex 7a (ionic radius 1.10 A) has 47 shorter Y—P bond lengths indicates less steric congestion in 7a. Presumably, the short Y—P2 bond length of 2.817 A is due to the three-membered ring formed by the metallated carbon C2, yttrium and P2. There are five X-ray crystal structures reported for lanthanide phosphine complexes. Their M—P bond lengths are: (CsH4Me)3Ce(PMe3), 3.072 (4) A 1 7 ; (C5H4Me)3Ce[P(OCH2)3CEt], 3.086 (3) A 1 8 ; Nd[OCB U t 2 CH 2 PMe 2 ] 3 , 3.154 (2) A 1 6 ; (C5Me5)2YbCl(dmpm), 2.941 (3) A™; and Yb[N(SiMe3)2]2(dmpe), 3.012 (4) A 2 0 . Based solely on ionic radii, one should expect seven-coordinate Y + 3 (1.10 A) to have shorter metal phosphorus bond lengths than ten-coordinate C e + 3 (1.39 A), six-coordinate N d + 3 (1.123 A), eight-coordinate Y b + 3 (1.125 A) or six-coordinate Y b + 2 (1.16 A). The average Y—P bond length in 7a of 2.93 A (for the three non-metallated phosphines) is indeed shorter, however, as was noted in the introduction (Section 1.2), the steric effects of the other ligands present play a large role in determining the Ln—P bond length. The yttrium-carbon bond length of 2.65 (1) A is long for a yttrium rj-carbon bond. A typical bond length21 is 2.47 A, for example, (CsMe5)2YCH(SiMe3)2 has a Y — C bond length22 of 2.468 (7) A. Complexes that contain bridging alkyls have much longer Y — C bonds, e.g., [l,3-Me 2C5H 3) 2Y(Lt-Me)] 2 has Y-Me bond lengths23 of 2.60 (1) and 2.62 (2) A, and [(C 5H 5) 2Y(u-Me)] 2 has a Y-Me bond length24 of 2.655 (18) A. The long Y—C interaction that is seen in 7a could be due to the strain of the fused 3-and 4-membered Y-C2-P2 and Y-C2-Si2-Nl rings. 48 2.5 Kinetics of the Thermolysis of M R [ N ( S i M e 2 C H 2 P M e 2)2]2 The yttrium and lutetium hydrocarbyl complexes 4a-6a undergo clean first-order kinetics, eliminating R—H by abstraction of a PCH2Si-methylene hydrogen; this generates the cyclometallated complex shown in Scheme 2.1. The kinetics of this reaction in C6D6 can be conveniently followed by 3 1P{ 1H} NMR spectroscopy by measuring the decrease in concentration of YR[N(SiMe 2 CH 2 PMe 2 ) 2 ] 2 , R = Ph, CH 2C6H5, CD2C6D5, over time. The first-order rate and Eyring plots are shown in Figure 2.13 and the rate data are tabulated in Tables 2.5 and 2.6. From the Eyring plots, the activation parameters are remarkably similar: AH* = 21 kcal/mol and AS* = -3 eu. The four-centered transition state25 shown below (Figure 2.12) is a reasonable one for this process. For such a highly ordered transition state one might expect a large negative AS*. However, only a small negative value was found. To account for the rather small, but negative AS*'s observed, a transition state with a dissociated phosphine is proposed as this would be expected to have additional degrees of freedom and be better able to align into the necessary four-centered configuration for hydrocarbon elimination. As phosphine dissociation is rapid for these complexes, such a dissociation should not be the rate-determining step. Me2 - i * Si Me2 Figure 2.12: A possible transition state for the elimination of R—H. 49 Time (seconds) Eyring data for Y(R){N(SiMe2CH2PMe2)2}2 -11.0 -j 1 1 1 , 1 ,— 0.0028 0.0029 0.0030 0.0031 (K 1) temp Figure 2.13: The first-order rate plots for the thermolysis of Y(CH2C6H5)[N(SiMe2CH2PMe2)2]2,5a, and the Eyring plots for the thermolysis of Y(R)[N(SiMe2CH2PMe2)2]2, R = C H 2 C 6 H 5 , C D 2 C 6 D 5 , Q H 5 . 50 Table 2.5: Rate data for the thermolysis of YR[N(SiMe2CH2PMe2)2] 2-R = C6H 5 R = C H 2 C 6 H 5 R = CD 2C6D 5 kobs(CH2C6H5) Temp kobs Temp kobs Temp kobs kobS(CD2C6D5) — — 43.0 8.03 x IO'3 — — — 47.0 8.25 x IO"2 47.0 1.08 x IO-2 47.0 1.23 x IO"2 0.88 — — 50.0 1.43 x IO"2 — — — 56.6 2.43 x IO-2 56.5 2.97 x IO"2 56.5 3.36 x IO-2 0.88 60.0 3.25 x IO"2 — — — — — 66.0 5.86 x IO'2 66.5 7.19 x IO-2 66.5 8.49 x IO"2 0.85 73.0 9.91 x IO"2 73.5 1.42 x 10-1 73.5 1.55 x IO"2 0.92 — — 75.5 1.83 x IO"1 75.5 2.02 x 10-1 0.91 Table 2.6: Eyring data for the thermolysis of YR[N(SiMe2CH2PMe2>2]2-Function R = Q3H5 R = CH2C6H5 R = CD2C6D5 correlation -0.998 -0.999 -0.999 AS* (cal/k/mol) -4.i ± 3 -3.3 ± 3 -2.7 ± 3 AH* (Kcal/mol) 2 0 . 5 ± 1 2 0 . 6 ± 1 2 0 . 7 ± 1 Another interesting feature of this reaction is an apparent inverse isotope effect exhibited by the decomposition of YR[N(SiMe2CH2PMe2)2]2, R = CH2C6H5 and CD 2 C6D5. Even though the only C-H bonds being cleaved and formed in the transition state involve a PCH2Si-methylene hydrogen being transferred to the benzyl 51 ligand to form toluene (CH3C6H5 and C H D 2 C 6 D 5 ) , the formation of C H D 2 C 6 D 5 is faster than the formation of CH3C6H5: kn/ko = 0.85 to 0.92 in the temperature range studied (Table 2.5). One way to rationalize this observation is to invoke26 a pre-equilibrium involving an agostic T|2-benzyl ligand going to Tj1 -benzyl prior to the loss of toluene (Equation 2.2). H L n Y ^ C — Ph ^ = ^ r L n Y-CH 2 -Ph k* > Products [2.2] H 1 r|2-Benzyl T|1-Benzyl -d[ri2-Benzyl] / j^kg Rate- _ _ L i _ ) [ l f.Benzyl] [2.3] Rate = h^A h2-Benzyl] = (Keqk2)[ii2-Benzyl] [2.4] For these molecules, the breaking of an agostic benzyl C—H—»Y (or C—D - » Y ) interaction could be required in order for the benzyl ligand to achieve the required geometry for the elimination sequence. As an agostic C—H—»M is a stronger interaction then C—D—»M, 2 7 the pre-equilibrium would generate a higher concentration of Ti1-CD2C6D5 than r^-CF^CsHs resulting in faster elimination of C H D 2 C 6 D 5 than C H 3 C 6 H 5 . The measured deuterium isotope effect is therefore determined by the thermodynamic equilibrium isotope effect of the pre-equilibrium (Keq(H)/Keq(D) < 1) as the kinetic deuterium isotope effect should be zero. If the steady-state approximation is assumed (i.e., that the concentration of rj ^benzyl 52 complex remains constant), then the rate is defined as in Equation 2.3. If a further assumption is made that ki » k 2 then the rate equation simplifies to that in Equation 2.4 Unfortunately there is no conclusive evidence for the assumed agostic benzyl ligand. The a-benzyl carbon does display a reduced l T c-H coupling constant of 115 Hz, a typical agostic spectroscopic feature.27 However, no evidence for an agostic C—H—>M was evident in the solid state IR spectra (KBr pellet). It was hoped that the activation parameters for this reaction sequence would provide further evidence for this pre-equilbrium of the agostic benzyl ligand. Unfortunately, because the difference between the rates k0bs(H) and kDbs(D) is so small, the calculated activation parameters are, within experimental error, indistinguishable (Table 2.6). It is interesting to compare the rates of cyclometallation for the various MR [N (SiMe2CH 2PMe2)2J2 complexes as a function of M and R. At 73 °C, the lutetium-phenyl complex (k0bS = 1.88 x 10~4 sec-1, half life = 61.3 min) is an order of magnitude more stable than the corresponding yttrium phenyl complex (k0bS = 1.65 x 10"3 sec*1, half life = 7.0 min), which in turn is more stable than the yttrium-benzyl complex (kobs = 2.37 x 10"3 sec-1, half life = 4.9 min). The difference between the two yttrium complexes is rather small and may reflect the ability of the benzyl ligand to more easily attain the strained four-centered transition state than the phenyl derivative. The difference in rates between Lu and Y is probably due to the smaller size of lutetium relative to yttrium, the smaller metal making it more difficult to achieve the required geometry for the transition state. Consistent with this is the fact that the lanthanum-phenyl complex was not isolable; as the L a + 3 ion is the largest, the activation barrier is the lowest. 53 2.6 Reactivity of M[N (S iMe 2 CHPMe 2 ) (S iMe 2 CH 2 PMe 2 ) ] [N (S iMe 2 CH 2 PMe 2 ) 2 ] The strained structure depicted by the X-ray crystal structure determination of 7a prompted an investigation to see i f this complex might undergo o-bond metathesis react ions . 2 5 Indeed, in a sealed N M R tube under approximately four atmospheres of D 2 , the M — C bond does react reversibly, presumably generating the undetectable deuteride MD [N ( S iMe2CHDPMe2 ) ( S iMe 2 CH2PMe2 ) ] [N ( S iMe2CH 2 PMe2 ) 2 ] . This deuteride can then eliminate H D by a further ligand metallation reaction (Scheme 2.2). However, this reaction only proceeds at high temperatures. No reaction was detected until the N M R tube had been heated to about 90° C for several hours. A t this point, H D (triplet at 4.42 ppm, ' J H - D = 42.6 Hz) and H2 (at 4.46 ppm) were detectable. When deuterium N M R spectra were run on these complexes, the spectra showed that deuterium had become incorporated into the si ly l methyl (SiCjHj) and the methylene (S iCH^P) positions in the amido-diphosphine ligand. Clearly, it must be possible to metallate at either of these posit ions f rom the l igand to the metal, but the thermodynamic product is with metallation at the methylene position. The uranium hydride complex, HU[N (S iMe3) 2 ]3 , has been reported to undergo similar reaction with D 2 . 2 8 Unfortunately, these complexes w i l l only activate their own C-H bonds. When a ds-toluene solution of the yttrium complex was placed under H 2 and heated to 110° C for one week, no H D was detectable in the N M R . If the ds-toluene were activated by this complex, then eventually the deuterium label would appear in the H 2 gas. In addition, a deuterium N M R spectrum was obtained which showed no deuterium incorporation in the complex. The same results were found from sealed samples of both 7a and 8a in just ds-toluene (no H 2 ) . After extensive heating, 2 H N M R spectra 54 showed no uptake of deuterium into the complexes. Also, a dn-cyclopentane solution of the yttrium complex was placed under about four atmospheres of D 2 and heated to 95° C for 1 week. Mass spectral analysis of the cyclopentane showed no increase above natural abundance deuterium. Thus, although these cyclometallated compounds are thermally robust (no decomposition of these complexes has been detected even when heated to 110° C for a week) and will engage in a-bond metathesis with H 2 and D 2 they appear to be too sterically crowded to react with larger molecules. Scheme 2.2 M*=M[N(SiMe2CH2PMe2)2] 55 2.7 Summary The synthesis of a variety of new lanthanoid phosphine complexes has been achieved by complexing two amido-diphosphine ligands to yttrium, lutetium or lanthanum. At room temperature these compounds are fluxional and display NMR spectra indicative of complexes where the phosphorus donors are rapidly exchanging, probably via a dissociation-reassociation pathway. The small lutetium and yttrium complexes are monomelic and appear to be coordinatively saturated. However, the lanthanum complexes may be dimeric at lower temperatures indicating that the larger L a + 3 centre may be capable of attaining eight coordination. It is possible to generate thermally unstable hydrocarbyl complexes of the type M(R)[N(SiMe2CH2PMe2)2]2 which undergo a clean first order elimination of R-H to generate cyclometallated l 1 complexes of the type M[N(SiMe2CHPR2)(SiMe2CH2PR2)][N(SiMe2CH2PR2)2]-These thermally robust compounds will undergo a-bond metathesis with H2 and D2 at high temperature, but appear to be too sterically congested to react with larger molecules. Therefore, it would appear that these compounds are less reactive than other seven-coordinate lanthanoid complexes of the type Cp2Ln(R). 2.8 Experimental Procedures 2.8.1 General Information. All manipulations were performed under prepurified nitrogen in a Vacuum Atmospheres HE-553-2 glovebox equipped with a MO-40-2H purification system or in standard Schlenk-type glassware on a dual vacuum/nitrogen line. Molecular weight determinations were carried out in C6Dg using the isopiestic method29 in a Signer molecular weight apparatus.30 Anhydrous LaCl3 and YCI3 were prepared from the oxides31 (Aldrich), while anhydrous LUCI3 was used as received (Aldrich). HN(SiMe2CH2Cl)2 (Aldrich or Silar) was distilled prior to use and stored 56 under vacuum. Phenyllithium,32 benzylpotassium ( K C D 2 C 6 D 5 was prepared using C 6 D 5 C D 3 ) , 3 3 L i C H 2 S i M e 3 3 4 , H P M e 2 , 3 5 H P ( C H M e 2 ) 2 , 3 6 H P ( C M e 3 ) 2 , 3 7 L i N ( S i M e 2 C H 2 P M e 2 ) 2 , 1 1 L iN(S iMe 2 CH 2 PPr i 2 ) 2 , 1 1 L iN(S iMe 2 CH 2 PPh 2 ) 2 , 3 8 HN(SiMe 2 CH 2 PPh 2 ) 2 , 3 7 were prepared according to the literature procedures. The syntheses of KN(SiMe 2 CH 2 PMe 2 ) 2 and KN(SiMe 2CH 2PPh 2) 2 are given in section 3.6. Hexanes and THF were initially dried over CaH 2 followed by distillation from sodium-benzophenone ketyl. Ether and toluene were distilled from sodium-benzophenone ketyl. The deuterated solvents C6D6 and CgD5CD 3 were dried overnight with activated 4A molecular sieves, vacuum transferred to an appropriate container, "freeze-pump-thawed" three times and stored in the glovebox. Carbon, hydrogen and nitrogen analyses were performed by P. Borda of this department. NMR spectra were recorded in C6D6 unless otherwise stated. LH NMR spectra (referenced to CgD 5 H at 7.15 ppm or C6DsCD 2 H at 2.09 ppm) were performed on one of the following instruments depending on the complexity of the particular spectrum: Bruker WP-80, Varian XL-300, Bruker WH-400 or a Bruker AM-500. 1 3 C NMR spectra (referenced to C 6 D 6 at 128.0 ppm or C6Ds£D 3 at 20.4 ppm) were run at 75.429 MHz OJQ-U coupling constants are reported in square brackets) and 3 1 P NMR spectra (referenced to external P(OMe)3 in CgDg or C6D5CD 3 at 141.0 ppm) were run at 121.421 MHz, both on the XL-300. 8 9 Y NMR spectra (referenced to external 3M YC1 3 in D 2 0 at 0 ppm) were run at 19.6 MHz (Bruker WH-400) and 24.9 MHz (Bruker AM-500). All chemical shifts are reported in ppm and all coupling constants are reported in Hz. 2.8.2 YCl[N(SiMe 2 CH 2 PMe 2 )2]2, la. To a slurry of YC13 (1.381 g, 7.07 mmol) in THF (100 mL) was added a solution of LiN(SiMe 2 CH 2 PMe 2 ) 2 (4.08 g, 14.20 mmol) in THF (10 mL) dropwise with good stirring. After 12 hours the THF was removed under vacuum, hexanes (50 mL) were added and the solution was filtered 57 through Celite® to remove LiCl. Cooling a saturated hexanes solution to -30 °C gave analytically pure crystals of YCl[N(SiMe2CH2PMe2)2k (3.60 g, y ield 74%). *H NMR: 8 1.06 (24H, s, PMe) , 0.81 (8H, s, PCH2Si) and 0.35 (24H, s, SiMe). 1 3 C NMR: 8 20.03 (s, PCH2Si), 15.89 (s, PMe) and 7.27 (s, SiMe). 31p{lH} NMR: 8 -45.7 (d, J = 52) (See Table 2.1 & Figure 2.10). Anal. Calcd. for C 2 o H 5 6 ClN2P4S i4Y : C, 35.05; H, 8.24; N, 4.09. Found: C, 35.05; H, 8.30; N, 4.02. Signer molecular weight method, calculated monomeric molecular weight : 685 g/mol. Found: 635 g/mol. 2.8.3 YCl [N(S iMe 2 CH 2 PPh 2 )2 ]2 » lb . The same procedure as for YCl[N(SiMe2CH2PMe2)2l2 was used except that the product was extracted with toluene and crystal l ized from toluene/hexanes. YCI3 (0.16 g, 0.83 mmol) and L iN ( S i M e 2 C H 2 P P h 2 ) 2 (0.89 g, 1.66 mmol) gave 0.78 g (80% y ield) of YCl[N(SiMe2CH2PPh2)2]2. *H NMR: 8 7.67 (16H, m, HQ), 7.14 (24H, m, H m & Hp), 1.80 (8H, s, PCH 2 Si) and 0.33 (24H, s, SiMe). " C NMR: 8 138.68 (br, Q ) , 133.53 (d, J = 14.1, C0), 128.53 (d, J = 3.3, Cm), 129.10 (s, Cp), 17.43 (m, PCH2Si) and 5.91 (s, SiMe). 31p{lH} NMR in C 6 D 5 CD 3 : at 29° C, 8 -16.3 (d, J = 39); at -59° C, -9.2 (m), -18.8 (m), -19.8 (m) , -20.6 (s) (See Figure 2.6). Ana l . Calcd. for C 6oH 72ClN2P 4Si4Y : C, 60.98; H, 6.14; N, 2.37. Found: C, 61.02; H, 6.33; N, 2.40. 2.8.4 YCl[N(SiMe 2 CH 2 PPr i 2 )2]2 , lc . The same procedure as for Y C l [ N ( S i M e 2 C H 2 P M e 2 ) 2 ] 2 was used. YC13 (0.28 g, 1.41 mmol ) and LiN (SiMe 2 CH 2 PPr i 2 )2 "(1.13 g, 2.82 mmol) gave YCl[N(SiMe2CH2PPri2)2]2. Because of the high solubility of the compound (1 gram wi l l dissolve in less than 1 m L of pentane) analytically pure material has not yet been obtained. *H NMR: 8 1.87 (8H, sept, J = 7.0, PCH), 1.16 (48H, m, PCMe), 0.91 (8H, s, PCH2Si) and 0.50 (24H, s, SiMe). 13c NMR: 8 24.47 (s, PCHMe2), 19.68 (t-4.0 Hz, P C H M e o ) . 11.23 (t-5 Hz, PCH2Si) and 6.00 (s, SiMe). 31p{lH} NMR: 8 -1.09 (d, J = 44). 58 2.8.5 LuCl[N(SiMe2CH2PMe2>2]2> 2a. The same procedure as for Y C l [N (S iMe2CH 2 PMe2)2 l2 was used. LuCl3 (0.50 g, 1.79 mmol) was reacted with L i N ( S i M e 2 C H 2 P M e 2 ) 2 (1.03 g, 3.58 mmol) to give 0.86 g (63% yield) o f LuCl[N(SiMe 2CH 2PMe2)2]2- N M R spectra are recorded in C6D5CD3. lH NMR: 8 1.06 (24H, s, PMe), 0.80 (8H, s, PCH2Si) and 0.34 (24H, s, SiMe). 13c{lH} NMR: [ !Jc-H] 5 20.48 (s, PCH2Si) [t, 120.8], 16.13 (s, PMe) [q, 128.3], and 7.11 (s, SiMe) [q, 117.3]. 31p{lH} NMR: at 20 °C 8 -42.93 (s); at -80 °C 8 -40.8 (br, s) and -42.8 (br, s). Anal. Calcd. for C 2 o H 5 6 ClN2P4S i4Lu: C, 31.14; H, 7.32; N, 3.63. Found: C, 31.34; H,7.42; N, 3.50. 2.8.6 LaCl[N(SiMe2CH2PMe2)2l2» 3a. The same procedure as for Y C l [ N ( S iMe2CH2PMe2)2 ]2 was used except that one week of stirring at room temperature was required. LaCl3 (0.38 g, 1.57 mmol) and KN (S iMe2CH2PMe2)2 (1.00 g, 3.13 mmol) gave 0.85 g (74% yield) of LaCl[N(SiMe2CH2PMe2)2]2- N M R spectra are recorded in C 6 D 5 C D 3 . *H NMR: 8 1.09 (24H, s, PMe) , 0.79 (8H, s, PCH 2 Si) and 0.27 (24H, s, SiMe). 13c{lH} NMR: [1JC-H] 5 20.05 (s, PCH2Si) [t, 118.4], 16.05 (s, PMe) [q, 127.7], and 6.81 (s, SiMe) [q, 117.1]. 31p{ lH} NMR: at 20 °C 8 -43.51 (s); at -60 °C 8 -40.42 (br, s) and -42.47 (br, s). Anal. Calcd. for C 2oH56ClN 2P4Si 4La: C, 32.67; H, 7.68; N , 3.81. Found: C, 32.67; H, 7.82; N , 3.70. Signer molecular weight method, calculated monomelic mol . wt.: 735 g/mol. Found: 685 g/mol. 2.8.7 LaCI[N(SiMe2CH 2PPh2)2]2, 3b. The same procedure as for Y C l [ N ( S i M e 2 C H 2 P P h 2 ) 2 ] 2 was used except that one week of stirring at room temperature was required. LaCl3 (0.25 g, 1.01 mmol) and KN(SiMe2CH2PPh2)2 (1.15 g, 2.02 mmol) gave 0.80 g (64% yield) of LaCl[N(SiMe2CH2PPh2)2]2- N M R spectra 59 are reported in C6D5CD3. *H NMR: at 80 °C, 8 7.50 (16H, s, HQ), 6.99 (24H, s, H M , Hp), 1.69 (8H, s, PCH2Si), 0.05 (24H, s, SiMe). *3c{lH} NMR: at 60 °C, 8 139.05 (br, Q), 133.71 (m, C0), 128.58 (s, Cm), 129.06 (s, Cp), 17.59 (m, PCH2Si) and 5.62 (s, SiMe). 31p{lH} NMR: at 80 °C, 8 -12.3 (monomer), at -39 °C 8 -11.2 (s) and -19.6 (s) (dimer). In between this temperature range both species are detectable (See Figure 2.7). This compound crystallizes in a 2:1 ratio with associated toluene: Anal. Calcd. for C60H72ClN2P4Si4Y(0.5 C 7H 8): C, 59.68; H, 5.99; N, 2.19. Found: C, 60.01; H, 6.00; N, 2.15. Signer molecular weight method, calculated monomelic molecular weight: 1232 g/mol. Found: 1337 g/mol. 2.8.8 Y(C 6H5 ) [N(SiMe2CH 2 PMe2)2]2, 4a. To an ether solution (25 mL) of YCl[N(SiMe 2CH 2PMe 2) 2] 2 (0.630 g, 0.919 mmol) was added an ether solution (5 mL) of PhLi (0.077 g, 0.916 mmol). After 20 minutes the ether was removed under vacuum, the resulting oil extracted with toluene and the solution was filtered through Celite® to remove LiCl. Addition of hexanes (1 mL) to a saturated toluene solution and cooling to -30 °C gave 0.382 g (57% yield) of colorless crystalline YPh[N(SiMe2CH2PMe2)2]2. *H NMR: at 20 °C 8 8.37 (2H, d, J = 6.9, HQ), 7.37 (2H, t, J = 6.8, Hm), 7.22 (IH, t, J = 6.9, Hp) 0.90 (24H, s, PMe), 0.73 (8H, s, PCH2Si) and 0.41 (24H, s, SiMe); at -49 °C in C6D5CD3 8 8.45 (H0), 7.43 (Hm), 7.25 (Hp) 0.98, 0.95, 0.78 & 0.72 (PMe), 1.41, 0.81, 0.39 & -0.01 (PCH2Si) and 0.64, 0.61, 0.45 &0.18 (SiMe). 13C NMR: 8 191.0 (d, J = 41.5, Q), 142.3 (s, C0), 125.7 (s, C m ) , 124.9 (s, Cp), 20.36 (s, PCH2Si), 16.64 (s, PMe) and 7.35 (s, SiMe); at -49° C in C 6 D 5 C D 3 8 142.70 (s, C 0), 125.52 (s, C m ) , 124.72 (s, Cp), 19.64 & 19.40 (s, PCH2Si), 17.32, 16.11, 15.78 & 15.77 (s, PMe) and 8.20, 7.42, 7.21 & 6.33 (s, SiMe). 3lp{lH} NMR: at 20 °C 8 -46.7 (d, J = 43); at -49 °C 8 -43.6 (m) & -48.8 (m) (See Table 2.1 & Figures 2.1, 2.2 and 2.3). Anal. Calcd. for C 2 6H 6iN 2P 4Si4Y: C, 42.96; H, 8.46; N, 3.85. Found: C, 43.20; H, 8.55; N, 3.81. 60 2.8.9 Y(CH 2C6H5)[N(SiMe2CH 2PMe2)2]2» 5a. The same procedure was followed as for the above compound except that THF was the solvent used. YCl[N(SiMe2CH2PMe2)2]2 (0.500 g, 0.730 mmol) and C6H5CH2K(0.095 g, 0.730 mmol) gave 0.413 g (76% yield) of colorless crystalline Y(CH 2C 6H 5)[N(SiMe 2CH 2PMe 2) 2] 2. *H NMR: 5 7.22 (2H, t, J = 7.5, Hm), 7.12 (2H, d, J = 6.9, Ho), 6.81 (IH, t, J = 7.2, Hp), 2.06 (2H, br, Y-CH2) 0.97 (24H, s, PMe), 0.74 (8H, s, PCH2Si) and 0.35 (24H, s, SiMe). NMR: [UC-H] 8 155.3 (s, Q), 128.2 (s, C0) [d, 154.3], 125.1 (s, Cm) [d, 152.4], 117.4 (s, Cp) [d,158.3], 55.85 (d of q-30.4 & 4.1 Hz, Y-CH2) [t,114.9], 20.16 (s, PCH2Si) [t,118.5], 16.64 (s, PMe) [q,128.2] and 7.35 (s, SiMe) [q,117.2]. 31P{lH} NMR: 8 -46.7 (d, J = 43). Anal. Calcd. for C2 7H63N2P4Si4Y: C, 43.77; H, 8.57; N, 3.78. Found: C, 43.56; H, 8.70; N, 3.60. Y(CD2C6D5)[N(SiMe2CH2PMe2)2]2 was synthesized in the identical manner using KCD2C6D5 as the alkylating agent. This compound shows identical 31P{1H} and *H NMR spectra to 5a except that no *H resonances are observed for the benzyl ligand. 2.8.10 Lu(C 6 H 5 )[N(SiMe 2 CH2PMe2)2]2»6a. The same procedure as for YPh[N(SiMe2CH2PMe2)2]2was used. LuCl[N(SiMe2CH2PMe2)2]2 (0.205 g, 0.266 mmol) and C 6 H 5 L i (0.023 g, 0.27 mmol) gave 0.19 g (87% yield) of colorless crystalline LuPh[N(SiMe2CH2PMe2)2]2. NMR spectra are reported in C6D5CD3. !H NMR: at 20 °C 8 8.24 (2H, d, J = 6.6, HQ), 7.28 (2H, t, J = 7.2, Hm), 7.11 (IH, t, J = 7.2, Hp), 0.94 (24H, s, PMe), 0.68 (8H, s, PCH2Si) and 0.37 (24H, s, SiMe); at -70 °C 8 8.52 (H0), 7.50 (Hm), 7.28 (Hp), 0.98, 0.95, 0.76 & 0.76 (PMe), 1.39, 0.80, 0.39 & -0.73 (PCH2Si) and 0.71, 0.65, 0.49 & 0.20 (SiMe). 13c{lH} NMR: at 20 °C 8 195 (s, Q), 142.77 (s, C0), 125.96 (s, Cm), 124.98 (s, Cp), 20.10 (s, PCH2Si), 16.99 (s, PMe) and 7.35 (s, SiMe); at -49 °C 8 195.5 (Q), 143.83 (C0), 125.73 (Cm), 124.73 61 (Cp), 20.05 &19.56 (PCH2Si), 17.68, 16.61, 16.34 & 16.21 (PMe) and 8.20, 7.78, 7.16 & 6.68 (SiMe). 31P{lH} NMR: at 20 °C 8 -43.58 (s); at -59 °C 8 -39.69 (m) and -46.40 (m) (See Table 2.1 & Figure 2.4). Anal. Calcd. for C 2 o H 5 6 C l N 2 P 4 S i 4 Y : C, 35.05; H, 8.24; N, 4.09. Found: C, 35.05; H, 8.30; N, 4.02. 2.8.11 Y [ N ( S i M e 2 C H P M e 2 ) ( S i M e 2 C H 2 P M e 2 ) ] [ N ( S i M e 2 C H 2 P M e 2 ) 2 ] , 7a. Thermolysis of YBz[N(SiMe 2CH 2PMe2)2]2 or YPh[N(SiMe 2 CH 2 PMe 2 ) 2 ] 2 in toluene results in quantitative conversion to this cyclometallated product. This compound is exceedingly soluble in pentane and has only been isolated in very low yield. NMR spectra are recorded in C6D5CD3. *H NMR: at -38 °C sharp resonances between 8 1.35 and 0.92 (PMe and YCHP), multiplets between 8 0.85 and 0.40 (PCH2Si, some are obscured), and 0.49, 0.38, 0.37, 0.34, 0.33, 0.29, 0.23, & 0.22 (8-SiMe). 1 3 C{ 1 H}NMR: at -28 °C. 8 32.29 (d of d, J = 44.7 & 9.3, YCHP), 22.17, 21.41, & 20.3 (3-PCH2Si), 18.82, 18.50, 18.40, 17.43, 16.16, 15.77, 15.53, &14.80 (8-PMe), and 10.03, 8.29, 7.59, 7.35, 7.14, 6.99, 6.76, & 5.63 (8 SiMe). 31P{!H} NMR: at -29 °C 8 -34.2 (m), -41.9 (m), -42.0 (m), & -45.3 (m) (See Table 2.2 and Figure 2.8) Anal. Calcd. for C20H55N2P4SUY: C, 37.02; H, 8.54; N, 4.32. Found: C, 37.41; H, 8.54; N, 4.22. 2.8.12 Y [ N ( S i M e 2 C H P P h 2 ) ( S i M e 2 C H 2 P P h 2 ) ] [ N ( S i M e 2 C H 2 P P h 2 ) 2 ] , 7b. To a toluene solution (50 mL) of YCl[N(SiMe2CH2PPh2)2]2 (0.615 g, 0.836 mmol) was added a toluene solution (10 mL) of LiCH2SiMe3 (79 mg, 0.84 mmol) dropwise, with good stirring. The mixture immediately became cloudy due to the formation of LiCl. After stirring for two hours, the solution was filtered through Celite® and the filtrate was evaporated to dryness to remove SiMe4. The crude crystalline product was redissolved in a few mL of toluene and cooled to -30 °C to give 0.395 g of product (68% yield) *H NMR: 8 8.1 to 6.8 (broad signals for phenyl resonances), 2.0 to -1.0 (broad signals for SiMe and PCHSi resonances). 31P{1H.} 62 (broad signals for SiMe and PCHSi resonances). 31p{ lH} N M R : 8 - 0.25 (m, P2), -9.07 (m, PI) ~ -11( very broad P3 & P4). (See Table 2.2 for analysis of low temperature 31p{lH} NMR data). Anal. Calcd. for C6oH 7 iN 2 P4Si4Y: C, 62.92; H, 6.25; N, 2.45. Found: C, 63.34; H, 6.45; N, 2.35. 2.8.13 L u [ N ( S i M e 2 C H P M e 2 ) ( S i M e 2 C H 2 P M e 2 ) ] [ N ( S i M e 2 C H 2 P M e 2 ) 2 ] , 8 a . This compound can be quantitatively formed from the phenyl-lutetium compound as with yttrium. It too has only been isolated in small yield due to its extreme solubility in pentane. * H N M R of this compound consists of sharp resonances between 8 1.35 and 0.93 (PMe and LuCHP), multiplets between 8 0.82 and 0.40 (PCH2Si), and 0.41, 0.33, 0.31, 0.29, 0.27, 0.24, 0.21, & 0.20 (8-SiMe). 1 3 C { 1 H } N M R : in C 6 D 5 C D 3 at -28 C. 8 34.03 (d of t, J = 43.6 & 4, LuCHP), 30.19, 22.77, & 20.0 (3-PCH2Si), resonances from 21 to 15, (8-PMe), and 10.36, 7.83, 7.72, 7.43, 7.17, 6.78, 6.55, & 5.69 (8 SiMe). 3 l p { l H } N M R : 8 -34.8 (m), -36.1 (m), -38.6 (m), & -40.1 (m) (See Table 2.2 and Figure 2.9). Anal. Calcd. for C 2 o H 5 5 N 2 P 4 S i4Lu : C, 32.69; H, 7.54; N,3.81. Found: C, 33.20; H, 7.57; N, 3.42. 2.8.14 L a [ N ( S i M e 2 C H P M e 2 ) ( S i M e 2 C H 2 P M e 2 ) ] [ N ( S i M e 2 C H 2 P M e 2 ) 2 ] , 9 a . To a hexanes solution (50 mL) of LaCl[N(SiMe 2CH 2PMe 2) 2] 2 (0.615 g, 0.836 mmol) was added a hexanes solution (10 mL) of LiCH 2 SiMe3 (79 mg, 0.84 mmol) dropwise, with good stirring. The mixture immediately became both cloudy (due to the formation of LiCl) and bright yellow (due to metallation of the ligand) in appearance. After stirring for 2 hrs., the solution was filtered through Celite® and the filtrate was evaporated to dryness to remove S iMe4 . The crude crystalline product was redissolved in a few mL of hexanes and cooled to -30 °C to give 0.395 g of product (68% yield). NMR spectra are reported in CD3C6D5. *H NMR at 20 °C: 8 1.30 (d, J = 6, PMe2), 1.24 (m, PCH(Y)(Si)), 1.03 (s, 2 PMe2), 1.01 (d, J = 4, PMe2), 0.90 (m, 63 PCH 2Si), 0.68 (m, 2 PCH2Si), 0.28 (s, SiMe2), 0.17 (s, SiMe2), 0.14 (s, 2 SiMe2). 3lP{lH} N M R at -99 °C 5 -21.3 (d, J = 27), -35.5 (broad), -37.3 (very broad), & -38.3 (d, J = 27). Anal. Calcd. for C2oH55N2P4Si4La: C, 34.33; H, 7.92; N, 4.00. Found; C, 34.05; H, 7.94; N, 4.00. 2.8.15 K ine t i cs of the Thermolys i s React ions. The first order decomposition of M(R)[N(SiMe 2 CH 2 PMe 2 ) 2 ] 2 complexes was monitored using 3 1P{!H} NMR spectroscopy by following the disappearance of the hydrocarbyl complex over time. In order to insure accurate integrations a twelve second delay between 20° pulses was utilized. Typically, the spectra were collected at room temperature on sealed 0.09 mol/L C(,D^ solutions in sealed NMR tubes. These samples were immersed in an oil bath (maintained at a specific temperature) for a known length of time and the reaction was quenched by freezing the sample. For the do- and d7-benzyl compounds the samples were treated identically; each sample spent exactly the same length of time in the oil bath, frozen, and in the spectrometer. The observed rate constants were determined from the slope of the graph: ln{([A]o / [A]o-x)} versus time by plotting the ln(l / % starting material) versus time (least squares fit). The AH* and AS* were determined from the Eyring plots: ln(k0bs / temp) versus 1 / temp (least squares fit). AH* = -R(slope) and AS* = R[intercept - ln(kB / h)] where R = gas constant, h = Planck's constant and k B = Boltzmann constant. 2.8.16 Ca l cu la t ion of A G * f r o m F l u x i o n a l N M R Behav iou r . The AG* were calculated using the value for the rate constant39 1^ (1^  = 7 tAu c / 21/2) in the Eyring equation AG* = -RTcln[(kch) / (k BT c)]. R = gas constant, T c = temperature of coalescence, Au c = peak separation in the low temperature limit, h = Planck's constant and k B = Boltzmann constant. 64 2.9 References 1 a) Marks, T. J.; Ernst, R. D. In Comprehensive Organometallic Chemistry; Wilkinson, G.; Stone, F. G. A.; Abel, E. W., (eds.); Pergamon: Oxford, 1982, Chapter 21 and references therein, b) Schumann, H. Angew. Chem., Int. Ed. Engl. 1984,23, 474 and references therein. 2 Wilkinson, G.; Birmingham, J. M. J. Am. Chem. Soc. 1956, 78, 42. 3 Watson, P. L. J. Am. Chem. Soc. 1982,104, 337. 4 Watson, P. L.; Parshall, G. W. Acc. Chem. Res. 1985, 75, 51. 5 Evans, W. J.; Meadows, J. H.; Wayda, A. L.; Hunter, W. E.; Atwood, J. L. / . Am. Chem. Soc. 1982,104, 2009. 6 a) Evans, W. J.; Meadows, J. H.; Wayda, A. L.; Hunter, W. E.; Atwood, J. L. / . Am. Chem. Soc. 1982,104, 2015. b) Evans, W. J.; Bloom, I.; Hunter, W. E.; Atwood, J. L. J. Am. Chem. Soc. 1983,105, 1401. c) Evans, W. J.; Meadows, J. H.; Hunter, W. E.; Atwood, J. L. Organometallics, 1983,2, 1252. d) Evans, W. J.; Meadows, J. H.; Hunter, W. E.; Atwood, J. L. J. Am. Chem. Soc. 1984,106, 1291. e) Evans, W. J.; Meadows, J. H.; Hanusa, T.P. / . Am. Chem. Soc. 1984, 106, 4454. 7 Thompson, M. E.; Baxter, S. M.; Bulls, A. R.; Burger, B. J.; Nolan, M. C; Santarsiero, B. D.; Schaefer, W. P.; Bercaw, J. E. / . Am. Chem. Soc. 1987, 709, 203. 8 O'Hare, D.; Manriquez, J.; Miller, J. S. / . Chem. Soc, Chem. Comun. 1988, 491. 9 Shannon, R. D. Acta Cryst. 1976, A32, 751. Drew, M. G. B. Prog. Inorg. Chem. 1977,23, 67. 65 Fryzuk, M. D.; Carter, A.; Westerhaus, A. Inorg. Chem. 1 9 8 5 , 2 4 , 642. Fryzuk, M. D.; Haddad, T. S.; Rettig, S. J . Organometallics, 1 9 8 9 , 5, 1723. a) Kronenbitter, J.; Schwenk, A. Z. Phys. A, 1 9 7 7 , 2 5 0 , 117. b) Kronenbitter, J.; Schwenk, A. / . Mag. Res. 1 9 7 7 , 2 5 , 147. c) Adam, R. M.; Fazakerley, G. V.; Reid, D. G.J. Mag. Res. 1979, 33, 655. d) Levy, G. C ; Rinaldi, P. L.; Bailey, J. T. / . Mag. Res. 1 9 8 0 , 4 0 , 167. e) Holloway, C. E.; Mastracci, A.; Walker, I. M. Inorg. Chim. Acta, 1 9 8 6 , 1 1 3 , 187. Evans, W. J.; Meadows, J. H.; Kostka, A. G.; Closs, G. L. Organometallics, 1 9 8 5 , 4 , 324. Reger, D.L.; Lindeman, Lebioda, L. Inorg. Chem. 1 9 8 8 , 2 7 , 1890. Hitchcock, P. B.; Lappert, M. F.; MacKinnon, I. A. / . Chem. Soc, Chem. Commun. 1 9 8 8 , 1557. Stults, S.; Zalkin, A. Acta Cryst. 1 9 8 7 , C43, 430. Brennan, J. G.; Stults, S. D.; Andersen, R. A.; Zalkin, A. Organometallics, 1 9 8 8 , 7, 1329. Tilley, T. D.; Andersen, R. A.; Zalkin, R. A., Inorg. Chem. 1983, 22, 856. Tilley, T. D.; Andersen, R. A.; Zalkin, R. A., / . Am. Chem. Soc. 1 9 8 2 , 1 0 4 , 3725. Booij, M.; Kiers, N. H.; Meetsma, A.; Teuben, J. H.; Smeets, W. J. J.; Spek, A. L. Organometallics, 1 9 8 9 , 5, 2454. Haan, K. H. den; Boer, J. L. de; Teuben, J. H.; Spek, A. L.; Kojic-Prodic, B.; Hays, G. R.; Huis, R. Organometallics, 1 9 8 6 , 5 , 2389. 66 2 3 Evans, W. J.; Drummond, D. K.; Hanusa, T. P.; Doedens, R. J. Organometallics, 1987, 6, 2279. 2 4 Holton, J.; Lappert, M. F.; Ballard, D. G. H; Pearce, R.; Atwood, J. L.; Hunter, W. E. / . Chem. Soc, Dalton Trans. 1979, 54. 2 5 Thompson, M. E.; Baxter, S. M.; Bulls, A. R.; Burger, B. J.; Nolan, M. C.; Santarsiero, B. D.; Schaefer, W. P.; Bercaw, J. E. J. Am. Chem. Soc. 1989,109, 203. 2 6 Parkin, G.; Bercaw, J. E. Organometallics, 1989,5, 1172. 2 7 Brookhart, M.; Green, M. L. H. J. Organomet. Chem. 1983,250, 395. 2 8 Simpson, S. J.; Turner, H. W.; Andersen, R. A. / . Am. Chem. Soc. 1979,101, 7728. 2 9 Gordon, A. J.; Ford, R. A. The Chemist Companion, John Wiley & Sons: Toronto, 1972, 454. 3 0 a) Signer, R. Liebigs Ann. Chem. 1930,475, 246. b) An apparatus, nearly identical to the one used in our laboratory, has recently been described: Zoellner, R. W. / . Chem. Ed. 1990, 67, 714. 3 1 Taylor, M. D.; Carter, C. P. / . Inorg. Nucl. Chem. 1962,24, 387. 3 2 Schlosser, M.; Ladenberger, V. / . Organomet. Chem. 1967,5, 193. 3 3 Schlosser, M.; Hartmann, J. Angew. Chem., Int. Ed. Engl. 1973,12, 508. 3 4 Schrock, R. R.; Fellmann, J. D. / . Am. Chem. Soc. 1978,100, 3359. 3 5 Parshall, G. W. Inorg Synth. 1968,11, 157. 3 6 Issleib, K.; Krech, F. J. Organomet. Chem. 1968,13, 283. 67 3 7 Hoffmann, H.; Schellenbeck, P. Chem. Ber. 1966, 99, 1134. 3 8 Fryzuk, M. D.; MacNeil, P. A.; Rettig, S. J.; Secco, A. S.; Trotter, J. Organometallics, 1982, i , 918. 3 9 Thomas, W. A. Annu. Rev. NMR Spectrosc. 1968,1, 43. 68 CHAPTER 3: MONO(AMIDO-DIPHOSPHINE) COMPLEXES OF YTTRIUM AND LUTETIUM 3.1 Introduction This chapter describes the syntheses and some preliminary reactivity studies of mono(amido-diphosphine) complexes of yttrium and lutetium, the second class of lanthanoid phosphine target molecules. As the bis(amido-diphosphine) complexes of yttrium and lutetium achieved steric saturation by being seven-coordinate, it seemed reasonable that mono(amido-diphosphine) complexes would generate more reactive and perhaps more complex compounds. The mononuclear mono(ligand) complex, MCl2 [N (S iMe2CH2PR2)2l> is ° n t y five-coordinate; in order for the metal to achieve steric saturation, coordination of a base such as THF or else dimerization of the complex would have to occur. As bulky phosphine donors can make oligomerization difficult (due to inter-ligand steric repulsions) the variation in the size of the phosphorus donor could also be important and perhaps enable the isolation of coordinatively unsaturated complexes. The major problem encountered with the bis(amido-diphosphine) complexes, MR[N(SiMe2CH2PMe2)2l2» is the tendency to eliminate R — H and form a not very reactive cyclometallated compound. It was anticipated that hydrocarbyl-containing mono(amido-diphosphine) derivatives might be more stable than the bis(amido-diphosphine) counterparts, and as a result, would exhibit more varied reactivity patterns. As these mono(amido-diphosphine) complexes can be considered analogous to mono(cyclopentadienyl) lanthanoid complexes, a short summary of some of this work is presented here. Lanthanoid derivatives containing only one cyclopentadienyl-type ligand, first synthesized1 in 1963, have only recently been structurally 69 characterized; both CpErCl2(THF)3 and Cp*Cel2(THF)3 are eight-coordinate in the sol id state. 2- 3 Complexes containing L i C l , Cp*LnCl3Li (THF)2 (Ln = La , Ce; structures unknown), are also obtained for the early lanthanoid chlor ides. 4 A six-coordinate dimer, [Cp*Ce(OCMe3)2l2. containing two bridging and two terminal alkoxide ligands has also been structurally characterized, 5 whi le Cp*Ce(0-2,6-C6H3Bu t 2 )2 is monomelic in the solid state, achieving six-coordination via interaction of one of the t-butyl methyl groups with the metal centre. 6 Mono(cyclopentadienyl) hydrocarbyl-containing derivatives are quite rare. The first such complexes were obtained with coordinated l i th ium; [Li(TMEDA)2]Cp*LuMe3 and [Li(THF)3J-Cp*Lu(Bu l)2Cl have both been structurally characterized. 7 The first such salt-and-solvent free complexes, Cp*Y(o-C6H4CH2NMe2)2 and Cp*La[CH(SiMe3)2]2» were both reported in 1989 4 . 8 Both C p*La [ C H ( S iMe3) 2 ]2 and its T H F adduct, Cp*La [CH ( S iMe3 )2 ]2 (THF ) have been crystallographically characterized. Each complex displays agostic P-silyl-methyl bonds with the T H F adduct having the weaker interaction. 8 3.2 Syntheses of MC l2 [ N ( S i M e 2 C H 2 PMe2 )2 ] The first attempts to generate mono(amido-diphosphine) complexes were made by reacting YCI3 with one equivalent of L i [N (S iMe2CH2PMe2)2] in either toluene or T H F (Equations 3.1 and 3.2). toluene YCI3 + L i [ N (S iMe 2 C H 2 PMe 2 ) 2 ] • no reaction [3.1] 24 hr THF YCI3 + Li[N(SiMe2CH2PMe2)2] • YCl 2[N(SiMe 2CH 2PMe 2) 2] + LiCl[3.2] 24 hr 70 No reaction occurred in toluene, presumably due to the insolubility of YCI3. However, in THF both solids dissolved to give a clear colourless solution. The 3 1P{ 1H} NMR spectrum of this crude mixture (in CeDe/THF) clearly showed phosphine coordination to yttrium (a doublet at -42 ppm with ! J Y - P = 77 Hz). However, the product obtained from this reaction by removal of the THF was insoluble in either toluene or ether, so separation from LiCl was not possible. To overcome this problem it was decided to synthesize a new ligand salt, one that contained potassium instead of lithium, K[N(SiMe2CH2PMe2)2l- The byproduct of reaction 3.2 would then be KCI which is insoluble in THF, thus facilitating the isolation of YCl 2[N(SiMe2CH 2PMe2)2]- The synthesis of K[N(SiMe2CH2PMe2)2] should be possible by reacting HN(SiMe2CH2Cl)2 with KPMe2 (Equation 3.3), in analogy to the formation of Li[N(SiMe2CH2PMe2)2]9 from HN(SiMe2CH 2Cl)2 and LiPMe2-THF 3KPMe 2 + HN(SiMe 2CH 2Cl) 2 • K[N(SiMe 2CH 2PMe 2) 2] + HPMe2[3.3] -2 KCI In order to obtain the required KPMe 2 , HPMe2 was deprotonated with benzylpotassium, by vacuum transferring an excess of HPMe2 onto a cold (-78 °C) THF solution of KCH2C6H5 and allowing the mixture to warm slowly to room temperature (Equation 3.4). THF K C H 2 C 6 H 5 + HPMe2 • KPMe 2 + C H 3 C 6 H 5 -78 °C [3.4] 71 Removal of the volatiles (toluene, THF, HPMe2), washing the solid with hexanes and drying under vacuum left a freely-running, off-white powder for a 96% yield of KPMe2, based on KCH2C6H5 used. It is crucial that an excess of HPMe2 be used or the KPMe2 will be contaminated with unreacted KCH2C6H5. The KPMe2 thus obtained was reacted with HN(SiMe 2CH 2Cl)2 as shown in equation 3.3, to generate the necessary potassium ligand salt, K[N(SiMe2CH2PMe2)2l. in 66% isolated yield. The latter was reacted with YCI3 in THF (Equation 3.5), the KCI formed was filtered off, and the resulting clear, colourless solution was evaporated to dryness leaving a freely-running white powder. After washing with hexanes and drying in vacuo it was determined to be analytically pure YCl2[N(SiMe2CH2PMe2)2]- Presumably, this compound is oligomeric in nature as it is obtained as a hydrocarbon-insoluble, THF-free powder, probably due to the fact that the small size of the methyl groups at phosphorus are incapable of preventing oligomerization. THF MCI3 + K[N(SiMe2CH2PMe2)2] • MCl 2[N(SiMe 2CH 2PMe 2) 2] + KCI [3.5] 24 hr M = Y, 10a Lu, 11a The 3lp{lH} NMR spectrum of 10a (in QDcj/THF) obtained via eq. 3.5 was identical to that of the LiCl-contaminated material obtained in eq. 3.1. The conditions in reaction 3.5 have also been used successfully to prepare the lutetium analog, 11a. Both complexes display temperature independent NMR spectra in THF/C6D5CD3, that is, one sharp resonance (a doublet for yttrium) in the 3 1 P{ 1 H) NMR spectra both at room temperature and at -90 °C. In addition to resonances for the large excess of THF (about 10 equivalents), each complex shows three resonances in their 13C!{lH} 72 and *H N M R spectra (i.e., the phosphorus-methyl hydrogens and carbons, the methylene protons and carbons and the si ly l-methyl hydrogens and carbons respectively). In C6D5CD3/THF solutions, the coordination of two T H F molecules (rapidly exchanging with the bulk solvent) would generate a seven-coordinate monomer, which would have N M R spectra such as those observed (Figure 3.1). In order to obtain hydrocarbon-soluble complexes, an attempt was made to metathesize the chlorides with several reagents. The substitution of a hydrocarbyl ligand for a chloride should enhance solubility as wel l as generate a reactive ligand on the metal. However, the reactivity of Y C l 2 [ N ( S i M e 2 C H 2 P M e 2 ) 2 L 10a, was somewhat disappointing. Attempts to alkylate this complex with a variety of reagents ( L iMe , L i C H 2 C M e 3 , K C H 2 C 6 H 5 ) invariably led to intractable mixtures. In order to gain some insight into the processes occurring, the reaction with two equivalents of benzylpotassium was monitored at 20 °C by 3 1 P { !H } N M R spectroscopy. Initially, a doublet at -45.9 ppm ^Jy.p = 38.6) was observed (no signal for 10a remained) which disappeared after a few minutes to be replaced primari ly by resonances attributable to Y [ N ( S i M e 2 C H P M e 2 ) ( S i M e 2 C H 2 P M e 2 ) ] [ N ( S i M e 2 C H 2 P M e 2 ) 2 ] , 7a, the cyclometallated complex discussed in Chapter 2. A reasonable explanation is that the init ia l ly formed dia lky l complex, Y R 2 [N ( S i M e 2 C H 2 P M e 2 ) 2 l is unstable and decomposition occurs via ligand metallation and disproportionation. P C H 3 73 S i C H 3 Me 2 P. Me2Si' \THF N Lu" M e o S i THF P Me 2 S i C H 2 P \ J i i i i i i i ' i | i i i ' r • i i i i i i i i i i ' ' 1 1 r^ T TTTTTl^o'.'e ' 0/4 0.2 PPM S i C H 2 P 4.V P C H 3 r i i i i i i | i ' i r 20 18 16 ' S i C H 3 1 4 ' " 12 10 8 6 Figure 3.1: NMR spectra of LuCl 2 [N(SiMe 2 CH 2 PMe 2 ) 2 ] , H a : 300 MHz lH spectrum (top) and 75.4 MHz 13C{*H} spectrum (bottom), both in C 6 D 5 C D 3 / T H F . The C6D5CD3 peak is marked with an asterisk. 74 3.3 Synthesis and Structure of {Y(aIlyl)[N(SiMe2CH 2PMe 2)2]}2(u-Cl)2 The reaction of YCl 2[N(SiMe 2CH 2PMe 2) 2], 10a, with C3H5MgCl was more promising than the previously discussed alkylation attempts. With one equivalent of this Grignard reagent some decomposition did occur, but white crystalline product mixtures were obtainable. Unfortunately, attempts to separate the products via fractional crystallization proved futile. Reaction with two equivalents of CsHsMgCl led to a lemon yellow solution which contained some of the expected bis(allyl) complex, Y(allyl) 2[N(SiMe 2CH 2PMe 2) 2]. The^PpH) NMR spectrum showed a doublet at -38.3 ppm (Uy.p = 81 Hz) but this compound also could not be separated from the decomposition products present. However, a straightforward preparation of a mono(ligand) allyl complex was discovered. Reaction of CsHsMgCl with the bis(ligand) complex, YCl[N(SiMe 2 CH 2 PMe 2 ) 2 ] 2 , l a , produced the dimeric, mono(amido-diphosphine) compound, {Y(allyl)[N(SiMe 2CH 2PMe 2) 2]} 2(|I-C1) 2, 12a, which has been crystallography characterized (See Tables 3.1 & 3.2 and Figure 3.2). Table 3.1: Selected Bond Lengths for {Y(allyl)[N(SiMe2CH2PMe2)2]}2(u-Cl)2. Bond Length (A) Bond Length (A) Y-Cl 2.746 (1) Y - C l l 2.587 (5) Y-Cl' 2.795 (1) Y-C12 2.609 (5) Y-Pl 2.931 (1) Y-C13 2.621 (5) Y-P2 2.892 (1) C11-C12 1.387 (9) Y-N 2.292 (4) C12-C13 1.356 (8) 75 C12 Figure 3.2: Three views of {Y(allyl)[N(SiMe2CH2PMe2)2]h(u-Cl)2,12a. At the top are two Chem 3D® views; the figure at the top right shows the approximate pentagonal bipyramidal geometry at yttrium (based on the assignment of two coordination sites for the allyl moiety): the N and CI' atoms are axial while PI, P2, CI and the allyl are equatorial. The bottom view is the ORTEP stereoview showing the complete atom labelling scheme. 76 Table 3.2: Selected Bond Angles for {Y(allyl)[N(SiMe2CH2PMe2)2]}2(u-Cl)2. Bond3 Angle (deg) Bond Typeb Bond3 Angle (deg) BondTypeb N-Y-Cl ' 162.37 (10) Ax-Y-Ax Cl -Y-Pl 81.02 (4) Eq-Y-Eq N - Y - C l 97.88 (9) Ax-Y-Eq C1-Y-P2 81.26 (4) Eq-Y-Eq N - Y - P l 76.01 (10) Ax-Y-Eq P1-Y-P2 148.38 (4) Eq-Y-Eq N-Y-P2 80.75 (10) Ax-Y-Eq P l - Y - A 100.7 (1) Eq-Y-Eq c N - Y - A 100.0 (1) Ax-Y-Eq P2-Y-A 104.3 (1) Eq-Y-Eq c Cl' -Y-Cl 76.11 (4) Ax-Y-Eq C l - Y - A 161.9 (1) Eq-Y-Eqd c r - Y - p i 118.52 (4) Ax-Y-Eq C11-C12-C13 126.1 (6) Cl'-Y-P2 81.94 (4) Ax-Y-Eq Y - C l - Y ' 103.89 (4) C l ' - Y - A 87.5 (1) Ax-Y-Eq a) A = centroid of the allyl moiety. b) Ax = axial; Eq = equatorial for the pentagonal bipyramidal geometry. c) As the allyl centroid "A" centres 2 coordination sites, ideally this angle should be 108° (72 + 36). d) As the allyl centroid "A" centres 2 coordination sites, ideally this angle should be 180* (144 + 36). Assigning two coordination sites to the Tj3-allyl moiety makes the complex seven-coordinate at each metal. This arrangement is best described as two pentagonal bipyramids fused together via two bridging chlorides; each chloride is axial on one metal and equatorial on the other. The amide ligands take up the other axial sites. In an ideal pentagonal bipyramid the axial-metal-axial angle is 180°, the axial-metal-equatorial angles are 90° and the equatorial-metal-equatorial angles are 72° or 144°. In the allyl complex the axial-Y-axial bond angle is 162.37°, a deviation of 17.6° from the ideal. The axial-Y-equatorial bond angles range from 76.01° to 118.52° but average to 89.8°, very close to the expected 90°. See Tables 3.1 and 3.2 for the bond lengths and angles. 77 This allyl complex is fluxional and undergoes fast syn-anti allyl exchange.10 At high temperature, the *H and 1 3 C{ 1 H) NMR spectra each show five resonances. One signal for the central allyl proton (or its carbon), one signal for the equivalent syn and anti allyl hydrogens (or their carbons) and three signals for the amido-diphosphine ligand. However, syn-anti exchange does not completely explain the high symmetry of the NMR spectra (Figure 3.3). A fluxional process which equates both sides of the amido-diphosphine ligand must also be invoked. If the compound maintains its dimeric structure, then the simplest explanation is that phosphine donors exchange sites via dissociation, rotation about the metal amide bond and phosphine re-coordination. At low temperature, the NMR spectra are in agreement with the X-ray structure. The Hi, l3C{lH) and 3 1 P{ 1 H) NMR spectra are consistent with C 2 h symmetry (i.e., in solution and at low temperature one would expect the dimer to have a C 2 axis not present in the molecular structure) and reflect the inequivalent environments on either side of tridentate ligand plane. There are two signals for the phosphorus methyl protons and carbons, two for the silyl methyl hydrogens and carbons, two resonances for the methylene protons and one signal for the methylene carbons. The syn-anti exchange of the allyl moiety is frozen out such that two carbon and three proton resonances, showing resolved 3 J H - H couplings, are present (See Figure 3.3). A doublet ( ! J Y - P = 82 Hz) is observed in the 3 1P{ lH} NMR spectrum regardless of the temperature. Figure 3.3 300 MHz J H NMR spectra of {Y(allyl)[N(SiMe2CH2PMe2)2] h(H-Cl)2, 12a, at + 80 °C (top) and at -50 °C (bottom) in C6D5CD3 (marked with asterisks). 79 The formation of {Y(allyl)[N(SiMe2CH2PMe2)2]}2(H-Cl)2,12a, from allyl-MgCl and YCl[N(SiMe2CH2PMe2)2]2, la, was somewhat puzzling. The reaction was done in THF and worked up in toluene; extraction of the crude reaction mixture with toluene gave a clear colourless solution free of any precipitates. When this reaction was monitored by 31P{1H} NMR spectroscopy the spectrum showed the presence of two phosphorus containing compounds, a doublet for the allyl dimer and a singlet at -57 ppm due to an unknown compound. It seemed reasonable that, based on mass balance, this reaction gave two products, the allyl dimer and the new complex MgCl[N(SiMe2CH2PMe2)2] (Equation 3.6). However, the elemental analysis for the allyl complex did not agree with this formulation and on standing, solutions of this compound deposited some insoluble material. When crystals of "MgCl[N(SiMe2CH2PMe2)2l" were isolated, they gave an elemental analysis consistent with a formulation as Mg[N(SiMe2CH2PMe2)2l2-The analysis of the recrystallized allyl complex was now reinterpreted as indicating the presence of MgCl2- A formulation of {YCl(allyl)[N(SiMe2CH2PMe2)2]h-(MgCl2)o.7 fit the data and indicated that MgCh was the insoluble material that slowly precipitated from solution. This then implied that the products of the reaction are {Y(allyl)[N(SiMe2CH2PMe2)2] h(u-CT)2, Mg[N(SiMe2CH2PMe2)2]2 and MgCb (Equation 3.7). + C3H5MgCl MgCl[N(SiMe2CH2PMe2)2] + YCl[N(SiMe2CH 2PMe 2)2] 2 [3.6] 1/2 {YCl(allyl)[N(SiMe2CH2PMe2)2]}2 2YCl[N(SiMe2CH2PMe2)2]2 [3.7] 80 It is possible that the reaction proceeds initially via an exchange of ligands between the magnesium and yttrium metals (an allyl moiety for an amido-diphosphine ligand) followed by disproportionation of the MgCl[N(SiMe2CH2PMe2)2] to give Mg[N(SiMe2CH2PMe2)2l2 a n d MgCl2- The monomelic yttrium allyl complex, YCi(allyl)[N(SiMe2CH2PMe2)2L could then either dimerize to give 12a or coordinate the MgCl 2to giveYCl(allyl)[N(SiMe2CH2PMe2)2](MgCl2). To test the validity of equation 3.7, Mg(allyl)2(dioxane) was synthesized and reacted with 2 equivalents of YCl[N(SiMe2CH2PMe2)2l2» l a . As expected {Y(allyl)[N(SiMe2CH2PMe2)2]}2(H-Cl)2, 12a, and Mg[N(SiMe 2CH2PMe2)2]2 were formed (Equation 3.8). Both complexes gave satisfactory elemental analyses and identical NMR data to that recorded for the compounds isolated using equation 3.7. M g ( / \ ^ ) 2 Mg[N(SiMe 2CH 2PMe 2) 2]2 + 2YCl[N(SiMe 2CH 2PMe 2) 2] 2 • [3.8] THF {YCl(allyl)[N(SiMe2CH2PMe2)2] }2 The allyl dimer could now be prepared pure and in high yield; unfortunately it had disappointing reactivity. It did not react with H2 and it decomposed in the presence of ethylene. However, the ability of the allyl ligand to stabilize a mono(amido-diphosphine) complex had been demonstrated. This made the diallyl complex, Y(allyl)2[N(SiMe2CH2PMe2)2]» 13a, the next target molecule, as it should be a seven-coordinate monomer and also contain two-reactive hydrocarbyl ligands. 81 Y ( a l l y l ) 2 [ N ( S i M e 2 C H 2 P M e 2 ) 2 L 13a, was synthesized using the Mg(allyl)2(dioxane) reagent (Equation 3.9). As already mentioned, a previous attempt to make this molecule using (allyl)MgCl had resulted in formation of an inseparable mixtures of products. M g ( ^ ) 2 YCl 2[N(SiMe 2CH 2PMe 2)2] • Y(allyl) 2[N(SiMe2CH2PMe2)2] [3.9] THF -(MgCl 2) The lemon yellow diallyl complex is extremely soluble in hexanes and could not be induced to crystallize even though NMR spectra on the crude reaction material indicated that a very clean reaction with little or no decomposition had occurred. Consequently, this complex has only been characterized by NMR spectroscopy. Apart from syn-anti exchange for the equivalent allyl ligands this compound displays temperature independent NMR behaviour. The reactivity of H2 and ethylene with Y(allyl)2[N(SiMe2CH2PMe2)2] has been investigated. The thermolysis of Y(allyl)2[N(SiMe2CH2PMe2)2]» 13a, in a sealed NMR tube at 95° C for two days under approximately four atmospheres of H2 resulted in the complete disappearance of the diallyl complex. The resulting NMR spectra were complex and indicated that a number of species were present. The 1 1 cyclometallated Y[N(SiMe 2CHPMe 2)(SiMe 2CH 2PMe 2)][N(SiMe 2CH 2PMe 2) 2], 7a, bis(ligand) complex was the major product and a transitory hydride species (indicated by a doublet of pentets at +3.9 ppm in the *H NMR spectrum) could also be identified. Exposing just the diallyl complex to the same conditions (95 °C for 2 days) resulted in only a partial conversion (< 10% by NMR) to the cyclometallated bis(ligand) complex and no hydride formation, thus demonstrating the thermal stability of this bis(allyl) 82 complex. A sealed NMR tube of Y(al lyl )2[N (SiMe2CH2PMe2)2l. 13a, under approximately 1 atmosphere of ethylene (at room temperature) produced some crystalline polyethylene, but the NMR spectra only showed the presence of 13a, thus, the nature of the catalyst remains unknown. These are somewhat promising results as they show that given the right hydrocarbyl ligands Y(R)2[N(SiMe2CH2PMe2)2] complexes can be moderately thermally stable and can undergo hydrogenolysis to generate hydride species (albeit unstable hydrides). It is also possible that these complexes may be polymerization catalysts. However, the formation of i 1 Y[N(SiMe2CHPMe2)(SiMe2CH2PMe2)][N(SiMe2CH2PMe2)2], 7a, can still occur and seems to be problematic for any reactivity of complexes containing this particular amido-diphosphine ligand. 3.4 Syntheses of YCl 2 [N(SiMe 2 CH 2PR2)2] R = Ph, Bu*, Pr5 Although some success had been achieved with the mono(amido-diphosphine) complexes incorporating the [N(SiMe 2CH 2PMe 2)2l" ligand, there were two drawbacks: (i) the insolubility of the dichloride complex and (ii) the tendency of the alkylated derivatives to disproportionate and decompose. Therefore, the next phase of this project was to attempt to make more soluble mono(ligand) complexes by using ligands with bulkier phosphine donors. The basic idea was that large substituents at phosphorus would prevent higher order oligomers from forming and that greasy alkyl groups at phosphorus would aid in solubility. In addition, by bulking up the phosphine donor it might be possible to isolate complexes that would be unstable if the less bulky [N(SiMe2CH2PMe2)2l" ligand were used. The same strategy used for the formation of YCl2 [N (S iMe2CH2PMe2)2L 10a, was employed. Two new ligands, K [ N (S iMe2CH2PR2)2 l (R = Bul, Ph) were 83 synthesized (Equations 3.10 and 3.11); the KPBu l2 used was synthesized from benzylpotassium and HPBu^. THF 3KPBu l 2 + HN(SiMe 2CH 2Cl) 2 •K[N(SiMe 2 CH 2 PBu t 2 ) 2 ] + HPBu 2[3.10] -2 KCI THF K C H 2 C 6 H 5 + HN(SiMe2CH2PPh2)2 • K[N(SiMe 2CH 2PPh 2) 2] [3.11] - C 7 H 8 Both of these potassium salts reacted with YCI3 to give poor yields of toluene soluble complexes, 10b and lOd, that tended to crystallize as THF adducts (Equation 3.12). The complex with phenyl substituents at phosphorus was isolated as a THF adduct while for lOd, the first crop of crystals contained THF and the later batches of crystals contained little or no THF. Repeated recrystallization from toluene enabled removal of the THF, but the yields of pure complex were unsatisfactory. THF Y C I 3 + K[N(SiMe 2CH 2PR 2) 2] •YCl 2 [N(SiMe 2 CH 2 PR 2 ) 2 ](THF) n [3.12] -KCI R = Ph 10b R = Bu l lOd These complexes were not further investigated because a superior route to a soluble mono(ligand) compound was discovered. The reaction of YCI3 with the lithium salt of the ligand having isopropyl groups at phosphorus, Li[N(SiMe2CH2PPr J 2)2L gave the desired complex in high yield (Equation 3.13). Separation from the LiCl byproduct was possible as this mono(ligand) complex is hexanes soluble. 84 THF YC1 3 + Li[N(SiMe2CH2PPr1 2)2] •YCl 2 [N(SiMe 2 CH 2 PPr 2) 2](THF) n [3.13] -LiCl 10c The first two crops of crystals isolated were THF adducts, while the rest of the compound was isolated base free. A molecular weight determination of the THF-free complex showed it to be the dimer {YChtNCSih^C^PPr^hlh , 1 0 c > at 20 °C in C6D6. This dimer shows a doublet (Jy.p = 88 Hz) in its 3 1P{  lH) NMR spectrum at all temperatures (+20 to -80 °C). The *H NMR spectra are somewhat different. At room temperature a highly symmetric spectrum is observed; one silyl methyl, one methylene, one isopropyl methine and two isopropyl methyl resonances are present (Figure 3.4). At -78 °C the *H NMR spectrum is broad, but two silyl methyl environments can be discerned. Based on these data, there are two reasonable dimeric structures. A six-coordinate, dichloride bridged dimer, A, is consistent with the J H NMR spectrum obtained at low temperature, whereas a seven-coordinate, tetrachloride bridged dimer, B, explains the NMR data obtained at room temperature. A fluxional process whereby at high temperature A interconverts bridging and terminal chlorides via B also fits the data. Alternatively, compound A could be undergoing phosphine dissociation-reassociation equilibria. A rapid monomer-dimer equilibrium is unlikely as the molecular weight determination demonstrated that this complex was dimeric at 20 °C. Me 2 Si Me 2 Si A B Figure 3.4: Temperature invariant NMR spectra of {YCl2[N(SiMe2CH2PPri2)2]h. 10c in C 6 D 6 : 121 MHz 31p{lH) NMR spectrum (top) and 300 MHz *H NMR spectrum (bottom). 86 The isolated THF adduct, YCl 2[N(SiMe2CH 2PPr i2)2](THF), has a very complex NMR behaviour (different from that displayed by the base free complex) as shown by its variable temperature 31P{!H} NMR spectra (Figure 3.5). In C6D5CD3 at +60 °C there is only one doublet (Jy.p = 89 Hz) present. As the temperature is lowered to 20 °C coalescence occurs and at -34 °C the presence of three species is indicated by three doublets of 91, 87 and 73 Hz in a ratio of 74%, 13% and 13% respectively; because the nuclearity of these species is unknown the actual concentrations were not determined. As the temperature is lowered further, signal collapse occurs again and at -79 °C only two compounds are present, as indicated by two doublets of 92 and 73 Hz in ratios of 89% and 11%. If a large excess of THF is added to the C 6 D 5 C D 3 solution of YCl2[N(SiMe2CH2PPr i2)2](THF) the observed behaviour changes. The 31p{lH) NMR spectra now show just one doublet (Jy.p = 75 Hz) from high temperature down to -20 °C. Coalescence occurs around -50 °C, but even down to -100 °C, the low temperature limit is not achieved and broad complex signals are observed. While the effect of addition of THF was not quantified, it is reasonable to assume that equilibria involving the transfer of THF between yttrium and the bulk solvent is occurring. A number of reasonable structures may be postulated. At high temperature, addition of THF should favor a seven coordinate bis(THF) monomer that exchanges coordinated THF with free THF in the solvent mixture. At lower temperatures where THF exchange could be slow, mono(THF) adducts or even dimers with variable amounts of coordinated THF could be present. 87 +60 *C -79°C Figure 3.5: 121 MHz 3 1 P { 1 H ) variable temperature NMR spectra of YCl2[N(SiMe2CH2PPri2)2](THF) in C D 3 C 6 D 5 . 88 3.5 Synthesis and Thermolysis of YC l (Ph ) [N (S iMe 2 CH 2 PPr» 2)2] YCl 2[N(SiMe 2CH 2PPr i 2) 2](THF) was reacted with 2.1 equivalents of LiPh in an attempt to generate the diaryl complex YPh 2[N(SiMe 2CH 2PPr i 2) 2]. A low yield of some colourless crystals was isolated, tentatively assigned as the monophenyl derivative YCl(Ph)[N(SiMe 2 CH 2 PPri 2 ) 2 ] , 14c. This compound was only characterized by ^H, ^ CpH} and 3 1P{ 1H} NMR spectroscopy, and shows highly symmetric temperature-invariant spectra (Figure 3.6). A C6D5CD3 solution of 14c in a sealed NMR tube slowly changed from colourless to yellow over a period of weeks at room temperature. The 3 1P{ 1H} NMR spectrum showed that a new complex with resonances for four inequivalent phosphines was slowly growing in, while the *H NMR spectrum indicated that C^H6 was being formed. A ligand metallation reaction between two monomelic units is a possible reaction. Heating the tube to 47 °C for two weeks resulted in a complete conversion of 14c to this new complex. The 3 1P{ ]H} NMR spectrum (Figure 3.7) shows 4 inequivalent phosphines, and the core geometry shown in the figure fits the data. P 2 at 34.6 ppm is only coupled to one Y atom (39.1 Hz). Pi at 44.5 ppm is coupled to P3 (7.1 Hz), P4 (7.1 Hz) and two Y atoms (47.7 and 12.4 Hz). P3 and P4 at 12 and 13 ppm are not coupled to each other (therefore cisoid) but are coupled to Pi (7.1 and 7.1 Hz) and one Y atom (61.2 and 63.1 Hz). The structure shown in Figure 3.7 has each metal centre as being 6 coordinate. Another possibility is to have the phenyl group bridge both metals generating a dimer with one 7-coordinate and one 6-coordinate centre. 89 Figure 3 .6: Temperature invariant NMR spectra of YCl(Ph)[N(SiMe2CH2PPri2)2J, 14c, in C D 3 C 6 D 5 . 121 MHz 3lp{lHJ NMR spectrum (top) and 300 MHz *H NMR spectrum (bottom). The CD3C6D5 resonances are marked with an asterisk. 90 Figure 3.7: 121 MHz 31P{1H} NMR spectra (in CD3C6D5) of 15c and a possible structure for this complex. 91 This new complex is interesting in that only mono(ligand) metallation has occurred, and the remaining phenyl group shows no tendency to abstract another ligand proton. The reactivity of this complex could be of great interest and perhaps it will prove to be more versatile than the cyclometallated bis(amido-diphosphine) complex discussed in Chapter 2. 3.6 Summary The synthesis of a number of mono(amido-diphosphine) lanthanoid complexes has been achieved. As expected, these species were more complex than the bis(amido-diphosphine) compounds described in Chapter 2. Complexes of the type MCl2[N(SiMe2CH2PMe2)2l are insoluble in hydrocarbon solvents, presumably because they are oligomeric in nature. They will, however, dissolve in THF probably forming seven-coordinate bis(THF) monomers. Attempts to alkylate these compounds generally led to decomposition; the cyclometallated bis(ligand) complex i 1 Y[N(SiMe 2CHPMe 2)(SiMe 2CH 2PMe 2)][N(SiMe 2CH 2PMe 2) 2], 7a, was identified as the major product. A route to a dimeric mono(amido-diphosphine) allyl complex, {YCl(allyl)[N(SiMe2CH2PMe2)2]h was found via the reaction of allyl-MgCl or Mg(allyl)2(dioxane) with YCl[N(SiMe2CH2PMe 2 )2k- The diallyl compound Y(al ly l)2 [N(SiMe2CH2PMe2)2] was also synthesized, demonstrating that bis(hydrocarbyl) complexes can be stabilized on a Y[N(SiMe2CH2PMe2)2] fragment. Unfortunately the diallyl compound could not be purified, and when it was allowed to react with H2, decomposition occurred with formation of 7a as the major product. In an attempt to isolate hydrocarbon-soluble mono(amido-diphosphine) complexes, three yttrium compounds containing bulky phosphine donors were synthesized: YCl2[N(SiMe 2 CH 2 PR2)2]. R = Ph, Bu l, Pr1. The complex with 92 isopropyl groups at phosphorus was synthesized in good yield. This hexanes soluble isopropyl derivative can be isolated as either a THF adduct or as the base-free dimer. It can be derivatized, probably generating YCl(Ph)[N(SiMe2CH2PPri2)2l» which is thermally sensitive and undergoes a cyclometallation reaction (losing C6Hg) to form I 1 Y 2Cl 2(Ph)[N(SiMe 2CHPPr , 2)(SiMe 2CH 2PPr 1 2)][N(SiMe 2CH 2PPr 1 2) 2], 15c. 3.7 Experimental Procedures 3.7.1 General Information. The same general procedures as outlined in section 2.8.1 were followed. 3.7.2 K P M e 2 . H P M e 2 was vacuum transferred into a tared, evacuated heavy-walled Schlenk tube and its weight determined (9.02 g, 0.145 mol). This HPMe2 was then vacuum transferred onto a cold (-78 °C) THF solution (120 mL) of K C H 2 C 6 H 5 (12.5 g, 0.0960 mol) and slowly allowed to warm to room temperature. The bright orange colour of the THF solution immediately becomes dark orange brown. After warming to room temperature most of the volatiles were removed in vacuo. In order to prevent a solid mass from forming, toluene (50 mL) was added, the mixture stirred for a few minutes and then evaporated to dryness. The flask was then removed to a glovebox, the solid was scraped out of the flask, washed several times with hexanes and dried in vacuo. The yield of off-white K P M e 2 (9.19 g, 0.0918 mol) based on K C H 2 C 6 H 5 used was 96%. 3.7.3 K[N(SiMe 2CH 2PMe 2)2]. HN(SiMe 2 CH 2 Cl) 2 (3.63 mL, 16.6 mmol) was slowly added dropwise to a cooled (-10 °C) THF solution (100 mL) of KPMe 2 (5.00 g, 49.9 mmol) with good stirring. During the addition the dark red colour due to the phosphide faded to a pale yellow colour. The THF was removed in vacuo and the 93 resulting glass was extracted with hexanes (150 mL), and filtered through a thick pad of Celite® to remove the gelatinous KCI. Concentrating the hexanes solution and cooling to -30 °C yielded 3.50 g (66% yield) of colourless crystalline K[N(SiMe 2CH 2PMe 2)2]. *H N M R : 8 1.00 (12H, s, PMe), 0.67 (4H, d, J = 4.2, PCH 2Si) and 0.18 (12H, s, SiMe). 1 3C { 1H} N M R : 8 25.92 (d, J = 20.5, PCH 2Si), 17.54 (d, J = 9.4, PMe) and 7.86 (s, SiMe). 31p{lH} N M R : 8 -56.5 (s). Anal. Calcd. for C i 0 H 2 8 N P 2 S i 2 K : C, 37.59; H, 8.83; N, 4.38. Found: C, 37.80; H, 9.00; N, 4.51. 3.7.4 K[N(SiMe 2 CH 2 PPh 2 ) 2 ]. K C H 2 C 6 H 5 (0.80 g, 6.14 mmol) in 25 mL of THF was slowly added dropwise to a cooled (-10 °C) THF/toluene solution (100 mL) of HN(SiMe 2CH 2PPh 2) 2 (3.25 g, 6.14 mmol) with good stirring. The deep orange colour due to the benzylpotassium is discharged as it deprotonates the amine leaving a yellow solution. The THF/toluene was removed in vacuo and the resulting glass was extracted with toluene (150 mL), and filtered through Celite® to remove any unreacted benzylpotassium. Concentrating the toluene solution, adding hexanes (50 mL) and cooling to -30 °C yielded 3.0 g (86% yield) of powdery K[N(SiMe 2CH 2PPh 2) 2]. * H N M R : 8 7.52 (8H, m, HQ), 7.07 (12H, m, H m & H p), I. 48 (4H, d, J = 3.2, PCH2Si) and 0.06 (12H, s, SiMe). 1 3 C N M R : 8 142.22 (d, J = II. 0, CO, 132.99 (d, J = 18.6, CQ), 128.50 (d, J = 12.2, C m ) , 128.49 (s, C p), 20.29 (d, J = 26.4, PCH2Si) and 7.82 (d-3.9 Hz, SiMe). 31p{lH} N M R : 8 -20.0. Anal. Calcd. for C3oH 3 6NP 2Si 2K: C, 63.46; H, 6.39; N, 2.47. Found: C, 63.57; H, 6.48; N, 2.30. 3.7.5 KP(Bu*)2. HPBut2 (10.15 g, 69.42 mmol) was syringed onto a cold -78 °C THF/hexanes solution (120 mL) of K C H 2 C 6 H 5 (8.55 g, 65.65 mmol) and slowly allowed to warm to room temperature. The bright orange colour of the THF solution immediately becomes dark greenish brown. After warming to room temperature the 94 volatiles were removed in vacuo and a white crystalline mass formed. The flask was left under vacuum overnight and the following day hexanes (60 mL) was added, the mixture stirred for a 30 minutes and then evaporated to dryness. The flask was then removed to a glovebox, the solid was scraped out of the flask, washed several times with hexanes and dried in vacuo. The yield of KPBu^ (11.14 g, 60.44 mmol) based on KCH 2 C6H 5 used was 92%. 3.7.6 K[N(SiMe2CH2PBu T2)2]. In a 250 mL flask, KPBu^ (7.61 g, 41.3 mmol) was suspended in hexanes and cooled to -10 °C. THF (100 mL) was slowly added to the phosphide suspension resulting in a greenish-brown solution. To this solution HN(SiMe 2 CH 2 Cl) 2 (3.00 mL, 13.8 mmol) was slowly added dropwise. After the addition, the solution was warmed to room temperature and allowed to stir for 2 hours prior to evaporation of the THF. The resulting glass was extracted with hexanes, filtered through Celite® and the filtrate cooled to -30 °C to yield 3.20 g of K[N(SiMe 2CH 2PBut 2) 2] (48%). *H NMR: 1.14 (36H, d, J = 11.1, PCMe), 0.66 (4H, d, J = 7.5, PCH2Si) and 0.42 (12H, s, SiMe). *3c{lH} NMR: 8 33.20 (s, PCMe 3), 29.93 (d, J = 12.9, PCMe3), 13.06 (d, J = 18, PCH2Si) and 7.52 (s, SiMe). 31P{lH} NMR: 8 19.0 (s). Anal. Calcd. for C 2 2 H 5 2 N P 2 S i 2 K : C, 54.16; H, 10.74; N, 2.87. Found: C, 53.80; H, 10.74; N, 2.70. 3.7.7 Mg(alIyl)2(dioxane). The synthesis of this compound is based on that reported for Mg(CH 2 SiMe 3 ) 2 . 1 1 To a THF solution (60 mL) of (allyl)MgCl (0.0614 mol) was slowly added 1,4-dioxane (10 mL), and the solution allowed to stir overnight. The following day the stirring was stopped and the suspension of MgCl2(dioxane) was allowed 6 hrs to settle out from the solution. The clear supernatent solution was cannula filtered to another flask and the solvent evaporated to dryness leaving a white powder. This powder was dried overnight under vacuum 95 and elemental analysis showed it to be a dioxane adduct. NMR spectrum reported in C 6D 6miF. The yield of Mg(allyl)2(dioxane) (2.24 g) was 19%. *H NMR: 6.51 (IH, p, J = 12.2, H m ) and 2.65 (4H, d, J = 12.2, H s , H m ) . Anal. Calcd. for CioH1 8Mg02: C, 61.74; H, 9.32. Found: C, 61.60; H, 9.30. 3.7.8 LuCI 2[N(SiMe 2CH 2PMe2)2l» Ha* In a glovebox, an apparatus designed for filtering under N 2 , was charged with LuCl3 (0.440 g, 1.56 mmol) and K[N(SiMe2CH2PMe2)2] (0.500 g, 1.56 mmol). Celite® was added to the filter pad and held in place with a wad of cotton wool. The apparatus was sealed under N 2 , removed from the glovebox, placed on a vacuum/nitrogen line and THF (100 mL) was added to the solids. This mixture was allowed to stir for 1 hour before filtering off the KCI generated. The filtrate was evaporated to dryness (leaving a freely-flowing white powder), the whole apparatus was sealed under vacuum and removed to the glovebox. The white powder was washed with hexanes and dried under vacuum to yield 0.423 g of LuCl 2[N(SiMe 2CH 2PMe 2) 2] (52% yield). NMR spectra are reported in C 6 D 5 CD 3 /THF IH NMR: 6 1.05 (12H, t, J = 2.1, PMe), 1.00 (4H, t, J = 3.1, PCH2Si) and 0.31 (12H, s, SiMe). ^C^H} NMR: 8 19.31 (s, PCH2Si), 13.93 (t, J = 3.1, PMe) and 6.15 (s, SiMe). 31P{lH} NMR: 8 -34.1 (s). Anal. Calcd. for CioH28NP2Si2LuCl2: C, 22.82; H, 5.36; N, 2.66. Found: C, 23.13; H, 5.20; N, 2.90. 3.7.9 YCI 2[N(SiMe 2CH 2PMe2)2L 10a. The same procedure as for the above compound was used. YCI3 (0.665 g, 3.41 mmol) was reacted with K[N(SiMe 2CH 2PMe 2) 2] (1-088 g, 3.40 mmol) to give 1.40 g (93% yield) of YCl 2[N(SiMe 2CH 2PMe 2) 2]. NMR spectra are reported in C6D5CD3/THF *H NMR: 8 1.04 (12H, t, J = 1.9, PMe), 1.02 (4H, t, J = 2.8, PCH2Si) and 0.30 (12H, s, SiMe). 13C{!H} NMR: 8 19.75 (s, PCH2Si), 14.06 (s, PMe) and 6.34 (s,SiMe). 96 3lp{lH} NMR: 5 -41.8 (d, J = 77). Anal. Calcd. for CioH 28NP 2Si2YCl2: C, 27.28; H, 6.41; N, 3.18; CI, 16.11. Found: C, 27.50; H, 6.50; N, 3.08; CI, 15.95. 3.7.10 Y C l 2 [ N ( S i M e 2 C H 2 P P h 2 ) 2 ] , 10b. The same procedure employed for the synthesis of YCl 2 [N(S iMe 2 CH 2 PMe 2 ) 2 ] was used except that the product was dissolved in minimum toluene and precipitated by the addition of hexanes. YCI3 (0.250 g, 1.28 mmol) was reacted with K[N(SiMe2CH2PPh2)2] (0.727 g, 1.28 mmol) to give 0.482 g of this compound isolated as a THF adduct (49% yield). * H NMR: 8 7.66 (8H, broad, H 0), 7.04, (12H, broad, H m , Hp), 3.44 (4H, broad, THF-a), 1.85 (4H, broad, PCH 2Si), 1.02 (4H, broad, THF-0), and 0.11 (12H, s, SiMe). 3 1 P { 1 H } NMR: 8 -15.7 (d, J = 68.7) and -11 (broad). This compound has not been isolated in analytically pure form. 3.7.11 Y C l 2 [ N ( S i M e 2 C H 2 P P r ! 2 ) 2 ] , 10c. A Schlenk tube was charged with Y C I 3 (0.793 g, 4.06 mmol), Li[N(SiMe2CH2PPri2)2] (1.623 g, 4.061 mmol) and THF (50 mL). The flask was sealed under N 2 and stirred 24 hrs. Then the THF was evaporated off and the resulting glass was extracted with toluene and filtered through a medium frit. The pale yellow filtrate was concentrated, cooled to -30 °C and several crops of crystals were collected. The first two crops of crystals contained one equivalent of THF, YCl 2[N(SiMe 2CH 2PPri 2) 2](THF) (1.030 g, 41% yield). * H NMR: in C 6 D 5 C D 3 8 3.90 (4H, broad, THF-a), 1.96 (4H, broad, PCHMe 2), 1.41 (4H, broad, THF-p), 1.24 (12H, d of d, J = 7.2 & 13.9, PCHMe), 1.14 (12H, d of d, J = 7.3 & 13.6, PCHMe), 1-02 (4H, m, PCH2Si) and 0.42 (12H, s, SiMe). 1 3 C { ! H } NMR: 8 70.44 (s, THF-a), 25.36 (s, THF-J3), 23.30 (s, PCHMe 2 ) , 19.89 (s, PCHMe). 19.46 (s, PCHMe). 9.87 (t, J = 4.9, PCH2Si) and 6.16 (s, SiMe). a i P ^ H } NMR: in C 6 D 5 C D 3 at 20 °C 8 0.0 (broad); at + 60 °C 8 1.8 (d, J = 89). After drying on a vacuum line overnight these crystals gave an elemental analysis consistent with 97 a formulation of Y C l 2 [ N ( S i M e 2 C H 2 P P r i 2 ) 2 ] ( T H F ) 0 . 4 . Anal. Calcd. for C3oH54NP2Si2Y(C4H80)o.4: C, 40.50; H, 8.18; N, 2.41. Found: C, 40.37; H, 7.88; N, 2.20. The 3 r d , 4 t h and 5 t h crops of crystals were the base free dimer {YCl 2[N(SiMe 2CH 2PPri 2) 2]} 2 (1.060 g, 47% yield). Molecular Weight Calcd.: 1105 g/mol; Found: 1107 g/mol. *H NMR: 8 2.05 (4H, m, PCHMe2), 1.28 (12H, d of d, J = 7.2 & 13.5, PCHMfi), 1.21 (12H, d of d, J = 7.2 & 13.2, PCHMeJ, 1-02 (4H, d, J = 4.8, PCH2Si) and 0.38 (12H, s, SiMe). 31P{*H} NMR: 8 -0.2 (d, J = 87.9). Anal. Calcd. for C i 8 H 4 4 N P 2 S i 2 Y C l 2 : C, 39.13; H, 8.03; N, 2.54. Found: C, 39.35; H, 8.19; N, 2.42. 3.7.12 Y C l 2 [ N ( S i M e 2 C H 2 P B u t 2 ) 2 ] , 10d. This compound was synthesized in the same manner as YCl 2 [N(SiMe2CH2PMe2)2l except that the product was recrystallized from toluene. YCI3 (167 mg, 0.855 mmol) was reacted with K[N(SiMe2CH2PBut2)2] (417 mg, 0.855 mol). The first crop of crystals contained one equivalent of THF, (0.030 g) for a 5% yield of YCl2[N(SiMe2CH2PBut2)2](THF). lH NMR: in C 6 D 5 C D 3 8 4.05 (4H, broad, THF-a), 1.40 (4H, broad, THF-p), 1.28 (36H, d, J = 12.3, PCMe), 1.04 (4H, d, J = 5.1, PCH2Si) and 0.49 (12H, s, SiMe). 1 3C{!H} NMR: 8 70.4 (s, THF-a), 32.68 (s, P £ M e 3 ) , 29.99 (m, PCMe3), 25.27 (s, THF-p), 13.06 (broad, PCH 2Si) and 6.62 (s, SiMe). 31p{lH} NMR: in C6D5CD3 8 21.9 (d, J = 87). After drying overnight under vacuum this compound gave an elemental analysis formulation as YCl 2[N(SiMe 2CH 2PBu t 2) 2](THF)o.7. Anal. Calcd. for C 2 2 H 5 2 N P 2 S i 2 Y C l 2 ( C 4 H 8 0 ) o . 7 : C, 45.20; H, 8.81; N, 2.13. Found: C, 45.16; H, 8.89; N, 2.20. The base free complex, YCl 2[N(SiMe 2CH 2PBu t 2) 2] (0.140 g, 27% yield) crystallizes last. *H NMR: 1.21 (36H, d, J = 12.9, PCMe), 0.76 (4H, d, J = 5, PCH2Si) and 0.36 (12H, s, SiMe). 31P{*H} NMR: 8 19.6 (d, J = 93.3). 98 3.7.13 Mg[N(SiMe 2 CH 2 PMe2)2]2 , { Y C l ( a I l y l ) [ N ( S i M e 2 C H 2 P M e 2 ) 2 ] } 2 , 12a. In a glovebox, a Schlenk tube was charged with YCl[N(SiMe 2 CH 2 PMe 2 ) 2 ] 2 (0.200 g, 0.292 mmol) and Mg(allyl)2(dioxane) (28 mg, 0.14 mmol). The Schlenk tube was removed from the glovebox and attached to a vacuum/nitrogen line. THF (50 mL) was added onto the two solids and the solution was stirred for 30 minutes. The THF was removed in vacuo (resulting in formation of a white crystalline mass), the flask was sealed under vacuum and taken into the glovebox. The crystalline mass was extracted with toluene to give a slightly murky solution which was filtered and concentrated to about 20 mL. The solution was cooled to -30 °C and the first crop of crystals (0.104 g, 76% yield) was {Y(u-Cl)(allyl)[N(SiMe 2CH 2PMe 2) 2]} 2,12a. NMR spectra are reported in C6D5CD3. *H N M R : 8 at +80 °C, 6.15 (IH, p, J = 13, H m ) , 3.15 (4H, broad, H s , H a), 1.18 (12H, s, PMe), 0.70 (4H, s, PCH2Si) and 0.14 (12H, s, SiMe); at +20 °C, 6.19 (IH, sept, J = 12, H m ) , 3.64 (2H, broad, H s), 2.83 (2H, broad, H a), 1.13 (12H, s, PMe), 0.65 (4H, s, PCH2Si) and 0.18 (12H, s, SiMe); at -50 °C, 6.30 (IH, t of t, J = 8.6 & 15.6, H m ) , 3.75 (2H, d, J = 8.6, H s), 2.95 (2H, d, J = 15.6, H a), 1.14 (6H, s, PMe), 1.04 (6H, s, PMe), 0.69 (2H, s, PCH2Si), 0.64 (2H, s, PCH 2Si) , 0.31 (6H, s, SiMe) and 0.23 (6H, s, SiMe). 1 3 C{!H} N M R : [ U C - H ] & 144.24 (s, Q) [d, 143.5], 72.68 (s, CG) [t, 150.1], 20.09 (s, PCH2Si) [t, 120.9], 15 (broad, PMe) and 6.34 (s, SiMe) [q, 117.1]. 31P{*H} N M R : 8 -39.7 (d, J = 81.8, major), -41.5 (d, J = 75.8, minor). Anal. Calcd. for C i 3 H 3 3 N P 2 S i 2 Y C l : C, 35.02; H, 7.46; N, 3.14; CI, 7.95. Found: C, 35.40; H, 7.52; N, 3.16; CI, 7.71. The second crop of crystals obtained were Mg[N(SiMe 2CH 2PMe 2) 2] 2 (0.065 g, 76% yield). *H N M R : 8 0.95 (12H, s, PMe), 0.61 (4H, t, J = , PCH2Si) and 0.35 (12H, s, SiMe). "Cf^H} N M R : 8 20.87 (s, PCH2Si), 15.28 (s, PMe) and 7.05 (s, SiMe). 3*P{lH} N M R : 8 -55.7 (s). Anal. Calcd. for C 2oH 56N 2P 4Si4Mg: C, 41.05; H, 9.64; N, 4.79. Found: C, 40.77; H, 9.44; N, 4.78. 99 3.7.14 {Y(^-Cl ) (a l ly I ) [N(S iMe 2 CH 2 PMe2)2]}2-(MgCl2 )x . To a THF solution (25 mL) of YCl[N(SiMe 2CH 2PMe 2) 2] 2 (0.200 g, 0.292 mmol) was added 0.70 M (allyl)MgCl (0.42 mL, 0.29 mmol). After 10 minutes the THF was removed under vacuum and the resulting oil was extracted with toluene. The clear toluene filtrate was concentrated and cooled to -30 °C. The crystalline product of this reaction, {Y(p:-Cl)(allyl)[N(SiMe2CH2PMe2)2]}2, is contaminated with cocrystallized MgCl 2; 0.130 g of such product was isolated. After recrystallization from toluene, this compound gave elemental analysis consistent with an empirical formula of {Y(^-Cl)(al lyl ) [N(SiMe 2 CH 2 PMe 2 )2]}2-(MgCl 2 )o .7 . Anal. Calcd. for C 26H66N 2P4Si 4Y 2Cl 2(MgCl 2) 0.7: C, 32.58; H, 6.94; N, 2.92. Found: C, 32.68; H, 7.07; N, 2.86. 3.7.15 Y(aHyl) 2 [N(SiMe 2 CH2PMe2)2]» 13a« In a glovebox, an apparatus designed for filtering under N 2 was charged with Mg(allyl)2(dioxane) (69 mg, 0.35 mmol) and YCl 2 [N(SiMe 2 CH 2 PMe 2 ) 2 ] (0.156 g, 0.354 mmol). The apparatus was sealed under N 2 , removed from the glovebox, placed on a vacuum/nitrogen line and THF (60 mL) was added to the solids. This mixture was allowed to stir for 1 hour before the THF was evaporated off. To the resulting oil, hexanes (30 mL) and 1,4-dioxane (10 mL) was added. The solution was filtered to remove MgCl2(dioxane) and the filtrate was evaporated to dryness. The apparatus was taken into a glovebox and the filtrate residues were reextracted with hexanes, concentrated to a few mL and cooled to -30 °C. No crystals of this compound could be isolated even though the 3 1 P{ 1 H) NMR spectrum showed only 1 compound to be present. *H NMR: 5 at 20 °C, 6.09 (2H, p, J = 11.5), 3.4 (4H, broad, H s), 2.6 (4H, broad, H a), 0.91 (12H, t, J = 1.8, PMe), 0.58 (4H, t, J = 3.9, PCH2Si) and 0.18 (12H, s, SiMe); in C 6 D 5 C D 3 at -40 °C, 6.11 (2H, t of t, J = 8.8 & 15.4), 3.44 (4H, d, J = 8.8, H s), 2.56 (4H, d, J = 15.4, H a), 0.88 (12H, broad, PMe), 0.49 (4H, broad, PCH 2Si) and 0.20 (12H, s, SiMe). 100 13C{1H} NMR: [HQ-H] 8 143.54 (s, Q) [d, 142.6], 67.71 (s, C0) [t, 150.0], 19.56 (t, J = 3, PCH2Si) [t, 120.1], 14.65 (t, J = 3, PMe) and 6.74 (s, SiMe) [q, 117.4]. 31p{lH} NMR: 5 -38.3 (d, J = 80.8). 3.7.16 YCl(Ph)[N(SiMe 2CH2PPri 2)2], 14c. To a rapidly stirred ethereal solution (25 mL) of YCl2[N(SiMe2CH2PPri2)2](TFTF) (0.175 g, 0.280 mmol) was added an excess of an ethereal LiCgHs (0.052 g, 0.62 mmol) solution. The mixture immediately became cloudy and after 15 minutes the solvent was removed in vacuo. The oily residue was extracted with toluene, filtered through Celite®, concentrated to about 10 mL and cooled to -30 "C. After a few weeks, a low yield of some colourless crystals were isolated. NMR data are reported in C6D5CD3. *H NMR: 8 8.23 (d, J = 6.3, Ho), 7.42, (t, J = 7.1, H M ) , 7.26 (t, J = 7.2, H P ) , 1.57 (4H, m, PCHMe2), 0.87 (4H, m, PCH2Si), 0.79 (12H, d of d, J = 7.2 & 14.4, PCHMe). 0.73 (12H, d of d, J = 7.2 & 14.4, PCHMe), and 0.43 (12H, s, SiMe). "C^H} NMR: 8 (Q was not located), 134.81 (s, CQ), 127.27, (s, Cp), 127.17, (s, Cm), 23.11 (s, PC_HMe2), 19.07 (broad, PCHMe), 8.63 (t, J = 6.2, PCH2Si) and 6.72 (s, SiMe). 31P{1H} NMR: 8 0.6 (d, J = 75.3). Anal. Calcd. for C3oH5 4NP2Si2Y: C, 56.67; H , 8.56; N, 2.20. Found: C, 53.36; H , 8.61; N, 2.20. 3.7.17 Y 2 Cl 2 (Ph)[N (SiMe2CHPPri2)(SiMe2CH2PPri 2 ) ] [N (SiMe2CH2PPri2)2]> 15c. Thermolysis of a C6D5CD3 solution of YCl(Ph)[N(SiMe2CH2PPri2)2] in a sealed NMR for two weeks at 47 °C resulted in a bright yellow solution. On cooling this solution to room temperature tiny yellow crystals precipitated from solution. J H NMR: 8 7.82(d, HQ), 7.24, (t, H M ) , 7.08 (t, H P ) , 2.4 to 1.5 (m, PCHMe2), 1.4 to 1.1 (m, PCHMe), 0.90 to 0.65 (m, PCH2Si), 0.60 to 0.20 (s, SiMe). 31p{lH} NMR: 8 44.5 (d of d of d of d , ijy.p = 47.7,2JY-p = 12.4, 2j Y. p = 7.1, 2j Y. P = 7.1), 34.6 (d, 2 J Y . 101 P = 39.1), 12.9 (d of d, lJY-p = 61.2, 2 J Y -p = 7.1), 12.2 (d of d, !jY-p = 63.1, 2 J Y -p = 7.1). This compound has not yet been isolated in an analytically pure form. 3.8 References 1 Manastyrskyj, S.; Maginn, R. E.; Dubeck, M. Inorg. Chem. 1963,2, 904. 2 Day, C. S.; Day, V. W.; Ernst, R. D.; Vollmer, S. H. Organometallics, 1982,1, 998. 3 Hazin, P. N.; Huffman, J. C ; Bruno, J. W. Organometallics, 1987,6, 23. 4 Booij, M.; Kiers, N. H.; Heeres, H. J.; Teuben, J. H. J. Organomet. Chem. 1989, 364,79. 5 Heeres, H. J.; Teuben, J. H; Rogers, R. D. / . Organomet. Chem. 1989, 364, 87. 6 Heeres, H. J.; Meetsma, A.; Teuben, J. H. J. Chem. Soc, Chem. Commun. 1988, 962. 7 a) Schumann, H.; Albrecht, I.; Pickardt, J.; Hahn, E. / . Organomet. Chem. 1984, 276, C5. b) Albrecht, I.; Schumann, H. J. Organomet. Chem. 1986,310, C29. 8 van der Heijden, H; Schaverien, C. J.; Orpen, A. G. Organometallics, 1989, 8, 255. 9 Fryzuk, M. D.; Carter, A.; Westerhaus, A. Inorg. Chem. 1985, 24, 642. 1 0 Huheey, J. E. Inorganic Chemistry, 3 r d ed., Harper & Roe, New York, 1978, 670-671. 1 1 Andersen, R. A.; Wilkinson, G. Inorg. Syn. Vol XIX, Shriver, D. F. (ed.), John Wiley & Sons: Toronto, 1979, 262. 102 CHAPTER 4: ALLYL-DDENE COUPL ING A T Z IRCONIUM AND HAFN IUM 4.1 Introduction Carbon-carbon coupling reactions are probably the single most important transformation in synthetic organic methodology.1 Of particular interest are those reactions which occur within the coordination sphere of a transition-metal complex as the resulting stereoselectivity of the process can often be fine-tuned by changes in the ancillary ligands of that metal centre.2 As described in Section 1.4, a unique allyl-diene coupling reaction at hafnium was discovered3 where the compounds HfCl(ri4-QH6)[N(SiMe2CH2PR2)2] (R = Me, Pr'), react with allyl-MgCl to generate the transient allyl-diene species, Hf(C3H5)(C4H6)[N(SiMe2CH2PR2)2l, which are transformed into the coupled products Hf(ri4:r|l-C7H11)[N(SiMe2CH2PR2)2] (Scheme 4.1). Scheme 4.1 103 This transformation is interesting for two reasons: i) this metal-mediated coupling between a C3 and a C4 fragment does not appear to have precedent and, ii) the reaction is quite sensitive to the nature of the ancillary ligand. The length of time it takes for these two complexes to complete the transformation (at room temperature) varies considerably; the compound with methyl groups at phosphorus takes about seven days, while the derivative with isopropyl groups at phosphorus takes about one hour. Consequently, it would appear that it is relief of steric strain in the diene-allyl complex that drives the reaction. However, it is interesting to note that neither of the analogous diene-allyl complexes CpM(diene)(allyl)4 or Cp*M(diene)(allyl)5, M = Hf, Zr, have been reported as undergoing any C—C coupling. This chapter describes some further investigations into the nature of this reaction. 4.2 Synthesis of AUyl-Diene Coupled Complexes The addition of allyl-MgCl to the diene complexes, MC1(T) 4 -C4H6)-[N(SiMe2CH2PR.2)2l> is summarized in equation 4.1. C3H5MgCI MCI(n4-C4H6)[N(SiMe2CH2PR2)2] • M(TiV-C7Hii)[N(SiMe 2CH 2PR 2)2] [4-1] THF M = Hf, FUPr' purple 16c M = Hf, R = Me purple 16a M = Zr, R = Pr' green 17c M = Zr, R = Me brown decomposition Interestingly, the synthesis of the zirconium derivative, 17c, takes about two hours at room temperature while the analogous zirconium derivative to 16a was not observed to form even after two weeks of stirring. The 3 1P{ 1H) NMR spectrum of this crude brown mixture from this reaction was indicative of decomposition. 104 As these allyl-diene coupled molecules are chiral, the reaction with a substituted allyl Grignard reagent was of interest because it was possible that the formation of diastereomers might occur in unequal amounts. Therefore, the reaction of 1-methylallyl-MgCl with the diene complexes was investigated (Equation 4.2). 1-Me-allyl-MgCI ZrCI(ri4-C4H6)[N(SiMe2CH2PR2)2] • ZrfoV-CeH^INfSiMegCHzPR^] I4"2! THF R = Me brown decomposition R = Pr' green 18c Once again, the less bulky zirconium derivative (methyl groups at phosphorus) reacted with the Grignard reagent to yield a product mixture that 31P{ *H} NMR spectroscopy indicated to be decomposition. The more bulky derivative (isopropyl groups at phosphorus) reacted with this Grignard reagent to yield two products in unequal amounts (approximately 2:1 by 31P{  lH} NMR spectroscopy). There are four possible isomers which can be formed due to the fact that the coupling can occur at either end of the allyl moiety and the methyl group can have two orientations (Figure 4.1). Of the four isomers, X-ray crystallography showed that C and D are the two which are formed. Figure 4.1 Four possible isomers for the coupling of 1-methylallyl with butadiene. It is interesting to note that although C and D are formed in unequal amounts, the 31P{ !H} NMR spectrum of the crystalline product shows equal amounts of C and D. 105 4.3 Molecular structures of Hf(Ti4:nl-C7Hii)[N(SiMe2CH2PPri2)2], 16c, and Zr(Ti4:nl-C8Hi3)[N(SiMe2CH2PPri2)2], 1 8 c Both Hf(Ti4:T1l-C7Hii)[N(SiMe2CH2PPr» 2)2], 16c, and Zr(Ti4:nl-C8Hi3)-[N(SiMe2CH2PPri2)2], 18c, have been crystallographically characterized. Tables 4.1-4.4 contain selected bond length and bond angle data, while Figures 4.2 and 4.3 show the structures of these compounds and the atom numbering scheme. Table 4.1: Selected Bond Angles for Zr(Ti4:rj1-C8Hi3)[N(SiMe2CH2PPri2)2], 18c. Bond Length Bond Length PI—Zr—P2 149.26 (4) Zr—C25—C24 72.7 (3) PI—Zr—N 76.1 (1) C25—C24—C23 118.9 (5) P2—Zr—N 74.6 (1) C24—C23—C22 122.5 (6) C25—Zr—N 110.5 (2) C23—C22—C21A 117.8 (7) C24—Zr—N 121.5 (2) C23—C22—C21B 147 (1) C23—Zr—N 148.7 (2) C22—C21A—C20A 104.0 (8) C22—Zr—N 179.0 (2) C22—C21B—C20B 100(2) C19—Zr—N 111.1 (2) C21A—C20A—C19 114.2 (9) C22—C21A—C26 116.3 (7) C21B—C20B—C19 114(2) C22—C21B—C26 107 (1) C20A—C19—Zr 120.8 (6) C20A—C21A—C26 114.7 (8) C20B—C19—Zr 115(1) C20B—C21B—C26 121 (2) C19—Zr—C25 132.2 (2) 106 Table 4.2: Selected Bond Lengths for Zr(r|4:T|l-C8Hi3)[N(SiMe2CH2PPri2)2], 18c. Bond Length Bond Length Zr—PI 2.782 (2) C25—C24 1.437 (8) Zr—P2 2.850 (2) C24—C23 1.405 (8) Zr—N 2.190 (4) C23—C22 1.376 (9) Zr—C25 2.345 (5) C22—C21A 1.52 (1) Zr—C24 2.358 (4) C22—C21B 1.57 (2) Zr—C23 2.500 (5) C21A—C20A 1.51 (4) Zr—C22 2.573 (5) C21B—C20B 1.59 (4) Zr—C19 2.305 (5) C20A—C19 1.54 (2) C26—C21A 1.44 (1) C20B—C19 1.28 (4) C26—C21B 1.57 (2) Table 4.3: Selected Bond Lengths for Hf(ri4:r|1-C7Hii)[N(SiMe2CH2PPr i2)2], 16c. Bond Length (A) Bond Length (A) Hf—PI 2.767 (6) Hf—N 2.129 (13) Hf—P2 2.812 (6) C19—C20 1.43 (3) Hf—C19 2.32 (2) C20—C21 1.45 (5) Hf—C20 2.36 (2) C21—C22 1.36 (5) Hf—C21 2.56 (2) C22—C23 1.44 (4) Hf—C22 2.58 (3) C23—C24 1.53 (4) Hf— C25 2.22 (3) C24—C25 1.52 (4) 107 Table 4.4: Selected Bond Angles for Hf(Ti4:ril-C7Hii)[N(SiMe2CH2PPri2)2], 16c. Bonds Angle (deg) Bonds Angle (deg) PI—Hf— -P2 150.3 (2) Hf—C19—C20 73.8 (13) PI—Hf— -N 77.8 (4) C19—C20—C21 120 (3) P2—Hf— -N 74.6 (4) C20—C21—C22 118(4) C19—Hf-- N 105.1 (7) C21—C22—C23 122 (4) C20—Hf-- N 118.4(9) C22—C23—C24 107 (2) C21—Hf-- N 148.1 (14) C23—C24—C25 110(3) C22—Hf-- N 174.9 (8) C24—C25—Hf 121 (2) C25—Hf-- N 115.9 (8) C25—Hf—C19 135.4 (9) The hafnium complex 16c crystallizes in the chiral space group PA\. The particular crystal which was crystallographically characterized is enantiomerically pure and the absolute configuration of the molecule is shown in Figure 4.3. The zirconium complex 18c however, crystallizes in the space group PI as a mixture of both diastereomers C and D and is not chiral (enantiomers C and D' are both present). The presence of two different molecules in one crystal complicated the solution and the structure was modeled as a 70:30 mixture (D to C) of both diastereomers. The net result is that only two atoms in the T| 5-C8H13 fragment (C20 and C21) differ in position between the two molecules, with the "allyl-methyl carbon" actually occupying the same spatial position in both molecules. Figure 4.2 Molecular structure of Zr(Tl4:r 1l-C8Hi3)[N(SiMe2CH2PPr i2)2], Two Chem 3D® views showing the two different diastereomers and an 01 stereoview showing the complete atom numbering scheme. Figure 4.3 Molecular structure of Hf(Ti 4:Til-C7Hii)[N(SiMe2CH 2PPr i2)2], 16c. A Chem 3D® view for comparison with 16c and an ORTEP stereoview showing the complete atom numbering scheme. 110 4.4 N M R Spectroscopic Characterization The allyl-diene coupled products 16c-18c display temperature invariant NMR spectra that are in accord with the crystal structure. All groups of atoms are inequivalent and the NMR spectra reflect this lack of symmetry. The 3 1P{ NMR spectra of the r|4:r|1-C7Hii-derivatives 16c and 17c show an AB quartet with transoid 2Jp.p coupling constants of 65 Hz and 66 Hz, respectively, while the spectrum of the zirconium complex 18c shows two AB quartets (2jp.p coupling constants of 67 and 59 Hz) for the two diastereomers (Figure 4.4). The *H NMR spectra are complex due to all the inequivalent groups of atoms but most of the resonances for 16c and 17c can be assigned. Assignment of all the proton resonances for 18c is not possible due to signal overlap: there are eight silyl-methyl (SiMe), four methylene (SiCHP) and sixteen isopropyl-methyl (PCHMe) resonances present over a two ppm range (Figure 4.5). The assignments of the resonances in the 1 3C{ lH} NMR spectrum of 16c and 17c can be made, while again, the diastereomeric mixture of 18c has too many overlapping resonances for accurate assignments to be made. The ^ Cf^H} NMR spectrum of just the CgHi3 fragments are shown in Figure 4.6. U l ! ! 1 i i 1 | '—' i — ' l » i—i—i—|—f—f—i—i—r~r—i—r i i — i — r — i — i — [ 1 ' i T f i i i i T j r i — n - r r r t T J I I I—r i i i T T J I — i — i — i i i i—r 22 ab lb . tb w ife lb PPH Figure 4.4: 121 MHz 31p{lH) NMR spectra in C6D 6 of Z r ( r i 4 : r i l - C 8 H i 3 ) -[N (SiMe2CH2PPr i2)2], 18c. The upper spectrum is of the crude reaction mixture and shows the different proportions of the two diastereomers, the lower spectrum is of a 1:1 mixture of crystallized products. ZrCl(T|4-C4H6)[N(SiMe2CH 2PPr i2)2] is marked with an asterisk. 112 Figure 4.5: 300 MHz J H NMR spectra of Zr(Ti4:-nl-C8Hi3)[N(SiMe2CH2PPri2)2], 18c, in C6D6. 113 Figure 4.6: A portion of the 75.4 M H z ^ C p H } N M R spectra in C 6 D 6 of Zr(Ti4:Til.C8Hi3)[N(SiMe2CH2PPri2)2]. 18c. 114 In order to explain the formation of the two diastereomers, 18c, the mechanism in Scheme 4.2 (which is different than the one presented in section 1.4, page 15) is suggested. Me2Si \ Zr<* Me2Si' j electrocyclic rearrangement Me7Sl; Me2Si Scheme 4.2 n3-allyl^=^ri1-allyl Me2Si Me2Si ^2 electrocyclic rearrangement Me2Si p-hydride elimination Me 2Si^ ^ Me 2Si^ j i Pr'p Me 2Si^ ^ H N Z r ^ A Me 2 s / j Me2Si Me9Si 115 All of the steps in the proposed mechanism of Scheme 4.2 have precedent. The first step where the allyl moiety is undergoing an T|3- to r|^ transformation is reasonable as the allyl-diene intermediate Hf(C3H5)(C4H6)[N(SiMe2CH2PMe2)2] can be identified by NMR spectroscopy as containing an allyl unit undergoing syn-anti exchange.3 The regioselectivity of this reaction can be explained by having the 1-methylallyl unit go rj 1 with the methyl group in the y position; this also makes intuitive sense because the reaction is sterically driven. As the methyl group can have two orientations, two different transition states are formed. The second step involves an electrocyclic six-membered ring transition state in a "chair type conformation". This type of transition state has been invoked to explain the addition of crotyltitanocenes to aldehydes.6 The final two steps, p-hydride elimination and olefin insertion, are well founded processes in organometallic chemistry.7 In an attempt to better understand this allyl-diene coupling transformation, the reaction of allylmagnesium chloride and ZrCl(T)4-C4H6)[N(SiMe2CH2PPr i2)2] was monitored by 3 1P{ 1H} NMR spectroscopy (Figure 4.7). The singlet due to the diene-chloride complex disappears after fifteen minutes, and doublet of doublet resonances (AX patterns) for three other compounds can be seen. The AX pattern at 26.1 and 17.6 ppm (2Jp.p = 50.9 Hz) can be assigned to the intermediate allyl-diene complex, and another AX pattern at 20.6 and 12.1 ppm (2Jp.p = 65.6 Hz) can be identified as the T| 4:r| 1-C7Hii coupled product. The other AX set of doublets at 27.4 and 11.1 ppm (2Jp.p = 34.2 Hz) is presumably due to an intermediate in the reaction sequence. Of the intermediates postulated in Scheme 4.2, the diene hydride species formed after P-hydride elimination step has the appropriate symmetry for a complex with inequivalent phosphines. Attempts to detect a hydride species in the *H NMR failed, although such a signal could still be present and be buried under another resonance. 116 30 25 ' 2 ' 0 15 ' I " 5 P P M ° Figure 4.7: 121.4 MHz 31p{lH} NMR spectra in THF/C6D6 of C 3H 5MgCl and ZrCl(ri4-C4H6)[N(SiMe2CH2PPri2)2] after 50 mins (upper) and 120 mins (lower). Note that the two scales are not identical. 117 4 . 5 Summary The coupling of allyl and butadiene fragments at hafnium or zirconium is a process which is very sensitive to the nature of the ancillary ligands at the metal. For the complexes, MCl(r|4-C4H6)[N(SiMe2CH2PR2)2L addition of an allyl-Grignard reagent results in formation of the C — C coupling products, M ( r | 4 : T | 1 -C7H1 i)[N(SiMe2CH2PR2)2]- The reaction takes about one hour when M = Hf & R = Pr1, two hours when M = Zr & R = Pr1, a week when M = Hf & R = Me, and results only in decomposition when M = Zr & R = Me. No coupling has been reported to occur for (C5R5)M(C3H5)(C4H6) (M = Hf, Zr; R = Me, H) compounds. Similar results are obtained when 1-methylallylmagnesium chloride is reacted with ZrCl(rj4-C4H6)[N(SiMe2CH2PR2)2]- When R = Me, decomposition occurs and when R = Pr*, after two hours the coupling is complete. Two of the four possible coupled products are formed in unequal amounts, and the coupling occurs exclusively at the substituted end of the 1-methylallyl unit as determined by X-ray crystallography. Which diastereomer is formed in higher yield was not determined. A mechanism for converting the MCl(ri4-C4H6)[N(SiMe2CH2PPri2)2] complexes to the coupled products via two intermediate species has been suggested. Evidence for one intermediate was found. The coupling reaction can be followed by 3 1P{ 1H} NMR spectroscopy and signals for the allyl-diene complex, an unknown intermediate as well as the final coupled product can be observed. 118 4.6 Experimental Procedures 4.6.1 General Procedures. The same general procedures as outlined in Section 2.7 were followed. The diene complexes MCl(T|4-C4H6)[N(SiMe2CH2PR2)2] were synthesized according to the literature procedures8, and the Grignard reagents were made from magnesium and allylchloride or 3-chloro-l-butene in THF. Figure 4.8 The numbering scheme used for the NMR spectroscopic assignments. 4.6.2 Hf(Ti4:Til.C7Hii)[N(SiMe2CH2PMe2)2], 16a. To a red solution (20 mL) of HfCl(ri4-C4H6)[N(SiMe2CH2PMe2)2] (0.270 g, 0.492 mmol) 0.73 M allylmagnesium chloride (0.67 mL, 0.49 mmol) in THF was added dropwise with rapid stirring. The red solution initially faded to orange and then over a period of days became a dark burgundy colour. After one week the solvent was removed under vacuum, the residue was extracted with hexanes and filtered through Celite® to remove MgCl2. 1 H NMR: 8 6.00 (m, H5), 4.48 (m, H6), 3.20 (m, H 2 a), 2.78 (m, H 3 b), 2.10 (m, H 3 a), 2.10 (m, H4), 1.96 (m, H2b), 1.57 (m, H 7 b), 1.41 (m, H 7 a), 0.41 (m, Hib), (Hi a is obscured), 1.25 (d, 2 J P . H = 6, PMe), 1.17 (d, 2Jp-H = 6, PMe), 0.93 (d, 2 J P . H = 6, PMe), 0.91 (d, 2 J P . H = 6, PMe), 0.89 (obscured SiMe2CHPMe2), 0.73 (dd, 2 J P . H = 9, 2 J H -H = 14, SiMe2CHPMe2), 0.50 119 (dd, 2Jp.H = 8 2 J H - H = 14, SiMe2CHPMe2), 0.34 (dd, 2 j P . H = 9.5 2 J H . H = 14, SiMe2CHPMe2), 0.24 (s, SiMe), 0.16 (s, SiMe), 0.14 (s, SiMe), 0.07 (s, SiMe). 13C{lH} NMR: 8 126.14 (s, C5) [d, l J c . H = 149], 93.42 (s, Qj) [d, 1J C-H = 144], 81.82 (s, C4) [d, iJc-H = 161], 63.41 (s, Ci) [t, lJC-H = 109], 49.21 (s, C7) [t, lJc-U = 141], 39.75 (s, C2) [t, iJc-H = 124], 35.71 (s, C3) [t, iJrj-H = 126], 18.35 (s, PCH2Si), 17.20 (s, PCH2Si), 15.08 (d, ijp-c = 15.2, PMe), 14.48 (d, iJp.c = 9.3, PMe), 14.12 (d, iJp.c = 15.4, PMe), 13.95 (d, l]P.C = 9.9, PMe), 6.50 (s, SiMe), 6.12 (s, SiMe), 6.03 (s, SiMe), 5.98 (s, SiMe). 31p{lH} NMR: 8 -8.8 (d, 2j P . P = 76), -16.6 (d, 2 J P . P = 76). Thus far this compound has not been obtained in an analytically pure form. 4.6.3 Hf(r|4:nl.C7Hii)[N(SiMe2CH2PPr i 2)2], 16c. The identical procedure described for 15a was used except that the reaction only requires one hour of stirring prior to workup. Red HfCl(Ti4-C4H6)[N(SiMe2CH2PPri2)2] (0.300 g, 0.454 mmol) reacts with 0.70 M allylmagnesium chloride (0.65 mL, 0.46 mmol) to yield 0.248 g (82%) of purple product. !H NMR: 8 6.36 (m, H5), 4.98 (m, HQO, 3.13 (m, H 2 a), 2.74 (m, H 3 b), 2.66 (m, H 3 a), 2.52 (m, H4), 2.44 (m, H 2 b), 1.54 (m, H 7 b), 1.30 (m, H 7 a), 1.05 (m, Hi a), 0.06 (Hib), 2.30 (dsept, 2 j P . H = 2, 3 J H - H = 7, PCHMe), 2.23 (dsept, 2 j P . H = 2, 3 J H - H = 7, PCHMe), 1.96 (m, PCHMe), 1.94 (m, PCHMe), 1.27, 1.26, 1.16, 1.15, 1.08, 1.06, 1.03 & 1.10 (for all 8 resonances: dd, 3 J P . H = 4 , 3 J H . H = 7, PCHMe)0.98 (dd, 2 J P . H = 8, 2JH-H = 14, SiMe2CHPMe2), 0.84 (dd, 2 J P . H = 8, 2 J H . H = 14, SiMe2CHPMe2), 0.57 (d, 2jp_H = 7.6, SiMe2CH2PMe2), 0.32 (s, SiMe), 0.27 (s, SiMe), 0.24 (s, SiMe), 0.21 (s, SiMe). 13C{1H} NMR: 8 123.48 (s, C5) [d, iJrj-H = 147], 90.22 (s, C6) [d, HQ-H = 160], 88.10 (s, C4) [d, lJc-H = 148], 69.42 (s, Ci) [t, lJC-H = HO], 51.52 (s, C7) [t, ^C-H = 139], 37.85 (s, C2) [t, lJC-H = 121], 34.72 (s, C3) [t, lJC-H = 127], 25.95 (d, 1J P . C = 7.7, PCHMe2), 25.48 (d, l J P . C = 5.2, PCHMe2), 25.18 (d, l J P . c = 6.4, P£HMe2), 25.07 (d, lJp.C = 7.8, P£HMe2), 20.25 (d, 2 J P . C = 4.5, PCHMe?). 20.19 (d, 2jp_c = 5.3, PCHMS2), 19.70 (s, PCHMe?). lv.?0 (s, PCHMe?). 19.41 (d, 2 J P . C = 3.7, 120 PCHMS2), 19.34 (s, PCHM£2), 19.25 (s, PCHMS2), 19.00 (s, PCHM&2), 11.03 (d, U P . C = 4.0, PCH2Si), 8.63 (d, 1JP.C = 3.8, PCH2Si), 6.40 (s, SiMe), 6.31 (s, SiMe), 6.11 (s, SiMe), 6.08 (s, SiMe). 3lp{lH} NMR: 6 +22.8 (d, 2 J P . P = 65), +17.3 (d, 2j P . P = 65). Anal. Calcd. for C2 5H55HfNP2Si2: C, 45.06; H, 8.32; N, 2.10; Found: C, 45.09; H, 8.50; N, 2.24. 4.6.4 Zr(ri 4:ri 1-C7Hii)[N(SiMe 2CH 2PPri 2) 2],17c. The identical procedure described for 15a was used except that the reaction only requires two hours of stirring prior to workup. Dark red ZrCl(Ti4-C4H6)[N(SiMe2CH2PPri2)2] (0.350 g, 0.610 mmol) reacts with 0.80 M allylmagnesium chloride (0.76 mL, 0.61 mmol) to yield 0.254 g (72%) of dark green product. *H NMR: 6 6.28 (m, H5), 4.52 (m, H6), 3.18 (m, H 3 a), 2.93 (m, H 3 b), 2.75 (m, H4), 2.3 (obscured, & H 2 b), 1.57 (m, H 7 b), 1.50 (m, H 7 a), 0.10 (obscured, Hia), -0.25 (Hib), 2.20 (m, PCHMe), 2.10 (m, PCHMe), 1.88 (m, PCHMe), I. 88 (m, PCHMe), 1.4-0.9 (8 PCHMs_), 0.55 (m, SiCH2P), other SiCH2P resonances are obscured, 0.29 (s, SiMe), 0.22 (s, SiMe), 0.20 (s, SiMe), 0.15 (s, SiMe). 13c{lH} NMR: 8 124.91 (s, C5), 101.97 (s, C6), 79.28 (s, C4), 69.40 (s, Ci) 51.52 (dd, 2jp_p = 3.1, 2 j P . P = 5.8, C7), 41.94 (s, C2), 34.93 (s, C3), 26.61 (d, 2 J P . C = 7.0, PCHMe), 25.46 (d, 2 j P . c = 2.7, PCHMe), 25.16 (d, 2 J P . C = 4.4, P£HMe), 24.89 (d, 2 J P . C = 5.8, P£HMe), 20.45 (s, PCHMfis), 20.40 (s, PCHM&2), 20.31 (s, PCHM&2), 20.22 (s, PCHM£2), 19.67 (s, PCHM£2), 19.62 (s, PCHMe^), 19.49 (s, PCHM&2), 18.31 (s, PCHMe?). 12.33 (d, lJ?.c = 6.6, PCH2Si), 9.95 (d, l J P . c = 5.8, PCH2Si), 6.49 (d, J = 3.0, SiMe), 6.14 (d, J = 2.8, SiMe), 5.94 (s, SiMe2). 31p{lH} NMR: 8 +21.2 (d, 2 J P . P = 66), +12.5 (d, 2J P.p = 66). Anal. Calcd. for C 2 5 H 5 5 ZrNP 2 Si 2 : C, 51.86; H, 9.57; N, 2.41; Found: C, 51.96; H, 9.59; N, 2.47. 4.6.5 Zr(ri4:ri1.C8Hi3)[N(SiMe2CH2PPr i2)2], 18c. The identical procedure described for 15a was used except that the reaction only requires two hours of stirring 121 prior to workup. Dark red ZrCl(Ti4-C4H6)[N(SiMe2CH2PPri2)2] (0.292 g, 0.509 mmol) reacts with 0.33 M 1-methylallylmagnesium chloride (1.54 mL, 0.51 mmol) to yield 0.181 g (60%) of dark green product. These green crystals are a 1:1 rnixture of diastereomers. IH NMR: 8 6.42 (m, H5), 6.20 (m, H5), 5.20 (m, He), 4.34 (m, H6), 3.22 (m, H4), 2.68 (m), 2.54 (m), 2.47 (m), 2.22 (m), 2.12 (m), 1.91 (m), 1.64 (m), 1.50 (m), 1.46 (d, J = 6, C8-Me), 1.40 (d, J = 6, C8-Me), 1.39 to 0.90 (overlapping resonances), 0.55 (d, J = 7), 0.35 (s, SiMe), 0.26 (s, SiMe), 0.23 (s, SiMe), 0.22 (s, SiMe), 0.21 (s, SiMe), 0.18 (s, SiMe), 0.14 (s, SiMe), -0.54 (m, Hi), 2.20 (m, PCHMe), 2.10 (m, PCHMe), 1.88 (m, PCHMe), 1.88 (m, PCHMe), 1.4-0.9 (8 PCHMa), 0.55 (m, SiCH2P), other SiCH2P resonances are obscured, 0.29 (s, SiMe), 0.22 (s, SiMe), 0.20 (s, SiMe), 0.15 (s, SiMe). 13C{1H} NMR: 8 124.42 (s, C5), 117.12 (s, C5), 108.74 (s, C6), 95.40 (s, C6), 94.65 (s, C4), 76.48 (s, C4), 68.00 (s, Ci) 65.18 (s, Ci) 50.04 (s, C2), 47.79 (s, C 2), 49.88 (m, C7), 49.84 (m, C7), 42.68 (s, C3), 37.87 (s, C3), 23.44 (s,C8), 22.12 (s,C8), 26.59 (d, 2 j P . c = 6.8, P£HMe), 26.49 (d, 2 J P . C = 7.8, P£HMe), 25.72 (d, 2 j P . c = 2.6, PCHMe), 25.30 (d, 2 J P . C = 5.4, PCHMe), 25.11 (d, 2 J P . C = 4.5, PCHMe), 24.93 (d, 2jp_c = 6.0, 2 PCHMe), 24.58 (d, 2 j p . c = 3.5, PCHMe), 20.67 to 19.49 (12 PCHMS2), 19.24 (s, PCHMe?). 19.10 (s, PCHMS2), 18.43 (s, PCHMe2), 18.27 (s, PCHMe?). 12.46 (d, Up.c = 6.2, PCH2Si), 10.82 (d, l J P . c = 6.5, PCH2Si), 10.02 (d, Up-c = 5.3, PCH2Si), 9.35 (d, iJ^c = 5.6, PCH2Si), 6.53 to 5.68 (SiMe). 31p{lH} NMR: 8 +18.5 (d, 2 J P . P = 59), +15.7 (d, 2 J P . P = 59) major diastereomer, +21.4 (d, 2 J P . p = 67), +11.5 (d, 2 J P . P = 67) minor diastereomer. Anal. Calcd. for C26H57ZrNP2Si2: C, 52.65; H, 9.69; N, 2.36; Found: C, 52.66; H, 9.82; N, 2.29. 122 4.7 References 1 Yamamoto, A. Organotransition Metal Chemistry; Wiley-Interscience: Toronto, 1986,374, and references therein. 2 Stille, J. K. In Modern Synthetic Methods; Scheffold, R.; Ed.; 1984, Vol. 3, 2. 3 a) Fryzuk, M. D.; Haddad, T. S.; Rettig, S. J. Organometallics, 1988, 7, 1224. b) Haddad, T. S. M.Sc. Thesis, UBC, 1987. 4 a) Erker, G.; Berg, K.; Kruger, C ; Muller, G.; Angermund, K.; Benn, R.; Schroth, G. Angew. Chem., Int. Ed. Engl. 1984,23, 455. b) Erker, G.; Berg, K.; Benn, R.; Schroth, G. Angew. Chem., Int. Ed. Engl. 1984,23, 625. 5 Blenkers, J.; de Liefde Meijer, H. J.; Teuben, J. H.; / . Organomet. Chem. 1981, 218, 383. 6 Collins, S.; Dean, W. P; Ward, D. G. Organometallics, 1988, 7, 2289. 7 Collman, J. P.; Hegedus, L. S.; Norton, J. R.; Finke, R. G. Principles and Applications of Organotransition Metal Chemistry, University Science Books, California, 1987, 2 n d ed., 380-387. 8 Fryzuk, M. D.; Haddad, T. S.; Rettig, S. J. Organometallics, 1989,5, 1723. 123 C H A P T E R 5: SYNTHESIS A N D C H A R A C T E R I Z A T I O N O F A SIDE-ON D I N I T R O G E N C O M P L E X O F Z I R C O N I U M 5.1 Introduction One of the milestones of modem inorganic chemistry was the discovery that dinitrogen (N=N) could act as a ligand. Since the original report1 by Allen and Senoff in 1965 on the preparation of [Ru(NH3)5N2]2+ (Equation 5.1), numerous N 2 complexes have been synthesized and structurally characterized.2 RuCl3(H20) x + excess H 2 NNH 2 (H 2 0) • [Ru(NH 3) 5(N 2)] + 2 [5.1] By far the most prevalent mode of coordination to a metal is end-on to either one metal or end-on bridging to two metals. The first such crystallographically characterized examples3' 4 are shown in Figure 5.1. NH, H3N-HQN 2+ > N H 3 •Ru- •N = N H3N' H,N NH 3 , * N H 3 Ru' N==N-H,N NH, 4+ Ru-* N H 8 •NH, NH 3 NH 3 NH 3 Figure 5.1: Representations of the first crystallographically characterized end-on mononuclear and bridging dinuclear N 2 complexes. Activation of the coordinated N 2 is indicated by a lengthening of the N-N bond distance (cf., 1.0975 A for that of free N 2 ) . 5 Typical N-N bond lengths are 1.03 to 1.16 A for mononuclear complexes and 1.12 to 1.33 A for binuclear compounds. The side-on mode 124 of N2 coordination is extremely rare and only a few polynuclear complexes which display this mode of coordination have been crystallographically characterized.6 These are summarized in Scheme 5.1. Side-on bonded N2 has also been observed in low temperature matrix isolated CoN 2 7 and ThN28 and theoretical predictions23 have been made which suggest that the side-on mode of binding might be the preferred coordination mode for the early transition elements. Core geometry for the complex cluster Molecular structure of [KN2Co(PMe3)3]6 [Ph5Ni2N2NaaLi6(OEt)4(OEt2)3J2. the related N—N: 1.16 to 1.18 A cluster [(PhLi)6Ni2N2(OEt2)2]2 is similar. N—N: 1.35 & 1.36 A Molecular structure of (C10H8)(C5H5),(C5H4)Ti4N2 N—N: 1.301 A Molecular structure of (Cp'2Sm)2N2 N—N: 1.088 A 125 Prior to the discovery of Allen and Senoff, the N 2 molecule was considered to be too inert to form transition metal complexes. The first edition of Cotton and Wilkinson's Advanced Inorganic Chemistry states9 that "the only reactions of N 2 at room temperature are with metallic lithium to give IJ3N and with nitrogen-fixing bacteria The fact that the enzyme nitrogenase will reduce N 2 to ammonia under ambient conditions demonstrates that the chemical inertness of dinitrogen can be overcome. Nitrogenase is thought to contain two molybdenum sites which, acting independently of one another, carry out the six-electron transformation of N 2 into two equivalents of N H 3 . 1 0 Consequently, much research is devoted to developing a transition metal based catalyst which will carry out this transformation efficiently.11 Interest in dinitrogen complexes was heightened when Chatt reported12 the first conversion of dinitrogen to ammonia in a protic media using a molybdenum compound; the news media was soon convinced that a major scientific breakthrough had occurred13 and some thought that inexpensive fertilizer would soon be available. The Farmers Weekly carried the story14 as "Cheaper nitrogen by 1990" and even our local newspaper The Province reported15 the story with the somewhat exaggerated title "Basic life process created in UK lab". These fanciful news stories reflect the ultimate and as yet unattained goal in this field of research: the synthesis of a useful catalyst that will convert dinitrogen to ammonia under mild conditions. Consequently, a fundamental test of any new dinitrogen complex is to see if in fact it can be induced to convert the N 2 ligand to ammonia or hydrazine. There have only been a few reports on zirconium dinitrogen complexes. A series of papers16 about the Zr(II) dimer {Cp*2Zr(N2)}2(^-N2) and two reports17 on the Zr(IJJ) monomer Cp2Zr[CH(SiMe3)2]N2 have been published. Both complexes display unique features. The Zr(II) dimer has been crystallographically characterized and is the first 126 complex to display both terminal and bridging end-on N2; this was also shown to be an important feature in the conversion of N2 to hydrazine. Protonation with HCl (g) yielded one equivalent of hydrazine and released two equivalents of N2 (Equation 5.2). (Cp* 2ZrN 2) 2N 2 + 4 H C l • 2Cp* 2ZrCl 2 + 2 N 2 + H 2 N N H 2 [5.2] Labelling studies revealed that the reaction with HCl proceeded via first protonation of one of the terminal N2 ligands, with release of the other terminal N2 ligand. One half of the hydrazine subsequently produced came from the remaining terminal N2 and the other half from the bridging N2. Formation of Cp*2ZrCl2 and a symmetric intermediate complex Cp* 2Zr(N2H)2 was inferred from the data. This latter complex reacts further with additional HCl, releasing N2 generating Cp*2ZrCl2 and hydrazine. The Zr(III) monomer Cp2Zr[CH(SiMe3)2]N2, is unique in that solution spectroscopy shows that the N2 ligand is probably bound side-on to the zirconium centre. The most compelling evidence is the temperature invariant ESR spectrum of the 1 4 N2 (or 1 5 N 2 ) complex, which shows the interaction of two equivalent 1 4 N (or 1 5N) centres with one Zr(III) nucleus. Protonation of this complex with HCl (aq) gave about 25% conversion of N2 to H2NNH2. There has been only one reported dinitrogen hafnium complex. {Cp*2Hf(N2)}2(lt-N2) has been prepared via the Na/K alloy reduction of Cp*2Hfl2 under one atmosphere of N2. It is thought to have the same structure as its zirconium analog.18 127 5.2 Synthesis of {ZrCl[N(SiMe2CH2PPri2)2]}2(N2) As part of an investigation into the stabilization of lower oxidation states of zirconium utilizing the amido-diphosphine ligand, it was found that phosphine complexes of zirconium(IV) can bind 1,3-butadiene under reducing conditions. Thus, reaction of ZrCl3[N(SiMe 2CH 2PPr i2) 2] with Na/Hg and 1,3-butadiene generates red ZrCl(T|4-C4H6)[N(SiMe 2CH 2PPr i 2) 2] which can also be prepared19 from magnesium butadiene, [MgC4H6(THF) 2 ]n, and ZrCl3[N(SiMe 2CH 2PPr i2)2]. However, if reduction of ZrCl3[N(SiMe2CH2PPri2)2] by Na/Hg is carried out in toluene under just N 2 , a deep blue solution is obtained from which dark blue crystals of the formula {ZrCl[N(SiMe 2 CH 2 PPr i 2 ) 2 ]} 2 (N 2 ) , 19c, can be isolated in moderate yield; the same reduction in the absence of N 2 results only in decomposition. Addition of 1,3-butadiene to the N 2 complex 19c does not generate the butadiene complex ZrCl(ri 4-C4H6)[N(SiMe 2 CH 2 PPr i 2 ) 2 ] . Attempts to add hydrazine to the butadiene complex resulted only in decomposition (Scheme 5.2). The reduction of the more bulky derivative ZrCl3[N(SiMe 2 CH 2 PBu t 2 ) 2 ] with Na/Hg under N 2 yields a green complex {ZrCl[N(SiMe 2CH 2PBu t 2) 2] }2(N2), 18d. Me 2 5.3 Molecular Structure of {ZrCl[N(SiMe2CH2PPr i2)2]}2(li-Tl2:T|2-N2) The X-ray crystal structure of {ZrCl[N(SiMe2CH2PPri2)2]}2(^-'n2:'ri2-N2), 19c, is shown in Figure 5.2 and Scheme 5.2; bond angles and bond lengths are given in Tables 5.1 and 5.2. The anionic ligands (NI, NI', N2, N2', CI & CY) are bound tetrahedrally to each zirconium while the two metals are fused together via a bridging side-on N 2 unit; the 129 two phosphine donors at each metal lie about three degrees out of the Zr2N2 plane. The most important feature of the structure is the bridging side-on bound N2 ligand that generates a completely planar, symmetric Zr2N 2 unit in which the zirconium-nitrogen bond lengths are essentially identical at 2.024 and 2.027 (4) A, significantly shorter than the typical zirconium-amide bond length (Zr-Nl) of 2.175 (3) A found in the ancillary ligand. A survey of 11 structures containing these amido-diphosphine ligands coordinated to hafnium or zirconium reveals a range of 2.096-2.264 A for the M-NR2 bond. By comparison, a very short Zr-N bond length of 1.826 (4) A, consistent with a double bond (Zr=N), has been reported20 for the imide derivative Cp2Zr(NBut)(THF). The N-N bond distance of the bridging N2 ligand in 19c is 1.548 (7) A, which is longer than that found in hydrazine (1.47 A), the prototypical N - N single bond. In fact, this is the longest N - N bond length ever measured for a transition metal dinitrogen complex. The only other planar side-on bridging dimer 6 3, (Cp*2Sm)2(^i-T|2:T|2-N2), has a very short N - N bond of 1.088 (12) A, and readily loses N2 in solution or under vacuum. There is one structure of a tetranuclear titanium N2 complex6b, (CifjH8)(C5H5)5(C5H4)Ti4(u.3-N2), where the N2 is terminally bound to two titaniums and side-on bound to one titanium. It has a long N - N bond distance of 1.301 A, somewhat longer than a N - N double bond (cf., azomethane at 1.24 A). The structure of the cobalt cluster [KN2Co(PMe3)3]6 has been reported60, and contains N2 ligands bridging end-on between potassium and cobalt centres. In addition, each dinitrogen ligand has two or three side-on interactions with other potassium atoms. The N-N bond lengths are between 1.16 and 1.18 A. There are two reported 6 d i 6 e nickel clusters, [(PhLi)6Ni2N2(OEt2)2l2 and [Ph5Ni2N2Na3Li6(OEt)4(OEt2)3]2. which display non-planar side-on N2 bonding. These clusters have tetrahedral M2N2 cores with additional bonding interactions to four lithium atoms; the N - N bond lengths are 1.35 and 1.36 (2) A, respectively. 130 Figure 5.2: Three views of the molecular structure of 19c. At the top are two Chem 3D® views showing the symmetric environment about the zirconium centres. The top left view is looking down the Zr—Zr axis, while at the top right is a view showing the close contact of the isopropyl groups across the N 2 bridge. At the bottom is a stereoview of this complex showing the disorder in the ligand backbone. 131 Table 5.1: Selected Bond Angles for {ZrCl[N(SiMe2CH2PPri2)2]}2(^-Tl2:T|2.N2). Bonds Angle (deg) Bonds Angle (deg) N2—Zr—N2' 44.9 (2) NI—Zr—N2 105.2 (1) Zr—N2—Zr' 135.1 (2) NI—Zr—N2' 105.5 (2) Zr—N2—N2' 67.6 (2) NI—Zr—PI 81.9 (1) Zr—N2'—N2 67.5 (2) NI—Zr—P2 80.9 (1) CI—Zr—NI 139.3 (1) PI—Zr—P2 141.00 (5) CI—Zr—N2 112.3 (1) PI—Zr—N2 84.7 (1) CI—Zr—N2* 111.9 (1) PI—Zr—N2* 129.6 (1) CI—Zr—PI 86.39 (5) P2—Zr— N2 133.7 (1) CI—Zr—P2 84.27 (5) P2—Zr—N2* 88.8 (1) Table 5.2: Selected Bond Lengths for {ZrCl[N(SiMe2CH2PPri2)2]}2(^-ri2:Ti2-N2). Bond Length (A) Bond Length (A) N2—N2' 1.548 (7) Zr—CI 2.493 (1) Zr—N2 2.024 (4) Zr—PI 2.764 (1) Zr—N2' 2.027 (4) Zr—P2 2.772 (1) Zr—NI 2.175 (3) The structure of the binuclear zirconium dinitrogen complex 19c is somewhat surprising. Crystal structure determinations of other dinitrogen complexes of group 4 metals16b.l6c.2i (cfM {Cp*2Zr(N2)h(H-N2) and (Cp*2Ti)2(N2)) show linear end-on N 2 coordination. However, Cp2Zr(R)N2 is believed17 to display side on bonding and there 132 are some theoretical predictions23 that a side-on structure should be more stable than a linear end-on structure, particularly for the early transition elements. Given the long N-N bond length in 19c, it is reasonable to suggest that this complex consists of two Zr4-*" centres and a bridging N24" unit. Even with this formalism, it is difficult to reconcile the planar, symmetric structure A, particularly since this same formalism should also hold for a linear diimide structure of the type B (Scheme 5.3), a common occurrence particularly for N2 complexes of groups 5 and 6. 2 2 Although an oversimplification, valence bond arguments suggest that unsymmetrical structures similar to C might be possible if the metal is sterically and electronically unsaturated as is the case in complex 19c. That the sum of these structures results in a symmetrical, planar structure A' (Scheme 5.3) having delocalized bonds is consistent with the relatively short Zr-N bond lengths found for the Zr2N2 core of 19c and with theoretical considerations of related bridging imide systems.23 Scheme 5.3 N Zr N N Zr N A B Zr Zr - - Zr Zr C C 133 5.4 N M R Spectra of {ZrCl[N(SiMe 2 CH 2 PR2)2]}2(N 2 ) {ZrCl[N(SiMe2CH2PPri2)2]}2(N2), 19c, shows temperature invariant lH and 3 1P{ *H} NMR spectra that are in accord with the C 2h symmetry of the structure. All four phosphorus atoms are equivalent and give rise to a singlet at -10.0 ppm. The lH NMR spectrum reflects the lack of symmetry on either side of the tridentate amido-diphosphine plane. Two silyl methyl, two methylene, two isopropyl methine and four isopropyl methyl resonances are observed at room temperature; the NMR spectra recorded at -80 °C are similar except that broadened signals are observed. The NMR spectra of {ZrCl[N(SiMe2CH2PBut2)2]}2(N2), 19d, (containing the very bulky t-butyl group at phosphorus) are different. For this complex variable temperature NMR spectra are observed. At +60 "C the lH NMR spectra show the same symmetry as the isopropyl derivative 19c. However, the two t-butyl and two silyl methyl resonances coalesce into four t-butyl and four silyl methyl resonances at 20 °C (Figure 5.3). This temperature-induced asymmetry manifests itself in the 3 1P{ ^H) NMR spectra as the collapse of a singlet at 25 °C into an AB quartet (showing a trans-2Jp.p of 75 Hz) at 0 °C (Figure 5.4). In the low-temperature limit, PI and P2 are no longer related by a mirror plane. The fluxional process can be explained by invoking rupture of the mirror plane which contains the Zr, CI and amide-N atoms (Figure 5.2). Inspection of the crystal structure of 19c reveals that the isopropyl groups have oriented themselves such that four of the eight isopropyl methines are directed toward each other (Figure 5.2, top right). Replacement of the methine proton with a methyl group (i.e., transforming an isopropyl group to a t-butyl group, 19c to 19d) would lead to severe inter-ligand repulsion. Such repulsion could force a rearrangement of the four phosphorus donors and lead to the observed lowering of the symmetry in 19d. 134 Figure 5.3: 300 MHz *H NMR spectra of (ZrCltNCSiMe^H^Bu^hlhCN?.) in C6D5CD3. The upper trace at + 60 °C and the lower spectrum was recorded at 20 °C. 135 Figure 5.4: 121 MHz31p{lH) NMR spectra of {ZrCl[N(SiMe2CH2PBut2)2]}2(N2) in QD5CD3. The upper trace at + 25 "C and the lower spectrum was recorded at 0 "C. 136 5.5 Reactivity of {ZrCl[N(SiMe2CH2PPri2)2]}2(N2) {ZrCl[N(SiMe2CH2PPri2)2]}2(N2), 19c, is only poorly soluble in hydrocarbon solvents. In an attempt to collect 13C{ :H} NMR data on 19c, it was dissolved in CDCI3, in which it is very soluble. However, after a few seconds a reaction took place as indicated by the blue solution turning green and then orange, with formation of a white precipitate. A similar result was obtained with CD2C12, except that the reaction took about 1 minute to complete. In both cases, the orange solution showed similar, but not identical NMR resonances (Figure 5.5). The NMR spectra were highly symmetric, similar to that obtained for ZrCl3[N(SiMe2CH2PPri2)2]. The 31P{1H} NMR spectra show only a singlet about two ppm upfield from where ZrCl3[N(SiMe2CH2PPri2)2] resonates. The *H NMR spectra show one silyl methyl, one methylene, one isopropyl methine and two isopropyl methyl resonances, again somewhat shifted from where ZrCl3[N(SiMe2CH2PPri2)2] resonates. Attempts to isolate these new complexes by crystallization from toluene resulted only in decomposition. A possible explanation for this reaction is that the halogen solvent is acting as an acid, cleaving the Zr—N bond with formation of cLj-hydrazine and either Z r C l ( C C l 3 ) 2 [ N ( S i M e 2 C H 2 P P r i 2 ) 2 ] (in the CDCI3 reaction) or ZrCl(CDCl2)2[N(SiMe2CH2PPri2)2] (for the CD2C12 reaction). Although it is not certain that this is the reaction taking place, it is known that ZrCl3[N(SiMe2CH2PPri2)2] is stable in the CDCI3 used; the 31P{1H} NMR spectrum of ZrCl3[N(SiMe2CH2PPri2)2] recorded after 1 week in CDCI3 did not show any decomposition. Therefore, it is assumed that the observed reaction is not due to a trace impurity that could be present in the CDCI3, but is in fact a genuine reaction with the halogenated solvent. 137 Figure 5.5: 121 MHz 31p{lH) (upper trace) and 300 MHz *H NMR (lower trace) spectra of the crude reaction mixture from {ZrCl[N(SiMe2CH2PPri2)2] h(N2) + CDCI3. 138 In order to ascertain the degree of N2 activation in {ZrCl[N(SiMe2CH2PPr i2)2])2(N2), 18c, its reactivity with excess HCl has been investigated. Upon addition of excess HCl (g) the dinitrogen compound was completely decomposed with conversion of the N2 ligand to hydrazine (equation 5.3). excess HCl {ZrCl[N(SiMe 2CH 2PPr i 2) 2]}(N 2) •decomposition +H 2 NNH 2 [5.3] This process has been quantified (Table 5.3) and the conversion was measured to be 97% (an average of three trials). The possible interference on the hydrazine test from the zirconium metal or the ancillary ligands was probed by decomposing a sample of ZrCl3[N(SiMe2CH2PPri2)2]- No hydrazine formation was detected in this reaction. Table 5.3: Theoretical and Calculated Hydrazine Concentrations From Reaction 5.3. Trial 18c(mg) Theoretical %T Found [Found] [Hydrazine] [Hydrazine] [Theoretical] 1 10 3.0 x 10-6 M 61.5 2.94 x 10-6 M 0.98 2 12.5 3.75xl0"6M 54.5 3.70xlO-6M 0.99 3 21 6.3 x IO'6 M 40.0 5.94 x 10"6 M 0.94 That the conversion of the N2 ligand into hydrazine was essentially quantitative supports the claim this moiety is best described as a N24- ligand. If a method can be found whereby the HCl decomposition can be controlled, then along with the conversion of N 2 to hydrazine it might be possible to recycle 18c into ZrCl3[N(SiMe2CH2PPri2)2]- The reducing agent provides the 4 electrons and the HCl provides the four H + and four Cl" atoms necessary for the full cycle. 139 5.6 Experimental Procedures 5.6.1 General Experimental Procedures The same general procedures as outlined in section 2.8.1 were followed. 5.6.2 {ZrCUNCSiMeiClhPPr^hlhCNi), 18c. ZrCl3[N(SiMe2CH 2PPr i2)2] (0.680 g, 1.15 mmol) in toluene (60 mL) was added to a 0.1 % Na/Hg amalgam (0.106 g, 4.60 mmol). The flask was then cooled to -196 °C, filled with N2, sealed and allowed to warm slowly to room temperature with stirring. The colourless solution slowly takes on the deep blue colour of the product. After two days of stirring, the N 2 was removed and the flask taken into a glovebox for product isolation. The Hg slurry was extracted several times with toluene (150 mL in total, until no more blue colour appeared in the slurry) and filtered through a column of Celite®. The Celite® from the column was then placed in a 50 mL flask, extracted several times with toluene (200 mL in total, until no more blue colour appeared in the Celite® toluene mixture) and filtered through Celite®. The toluene filtrate was then concentrated and cooled to -30° C. Several lots of crystals were collected (until the filtrate was no longer blue) to yield 0.270 g of product (44% conversion). The filtrate was then evaporated to dryness and 0.084 g of unreacted ZrCl3[N(SiMe2CH2PPri2)2] was precipitated by addition of hexanes (12% unconverted starting material). *H NMR (8, 500 MHz, C6D 6 ): 2.41 (d of sept) 3jH_H = 7.3, 2 J P . H = 4.2, PCHMe 2; 2.15 (d of sept) 3jH-H = 7.2, 2 Jp. H = 4.1, PCHMe 2; 1.43 (d of d) 3 j H . H = 7.3, 3 J P . H = 14.0, PCHMe?: 1.33 (d of d) 3JH-H = 7.2, 3 j P . H = 13.3, PCHMS2; 1.29 (d of d) 3 j H . H = 7.3, 3 j P . H = 12.2, PCHMe?: 1.27 (d of d) 3 j H . H = 7.2,3jP.H = 13.2, PCHMe?: 1.07 (d of d) 2 J H - H = 13.9, 2 Jp. H = 7.0, PCH2S1; 1.00 (d of d) 2 J H - H = 13.9, 2 J P . H = 7.0, PCH^Si; 0.38 (s) SiMe; 0.33 (s) SiMe. 31p{lH} NMR (8, 121.4 MHz, C 6 D 6 ) : +10.0 (s). Visible spectrum: (Pentane, 1 cm quartz cell): "km&x = 580 nm, e = 3500 L mol - 1 cm - 1. Anal. 140 Calcd. for Zr2C36H88Cl2N4P4Si4: C, 40.54; H, 8.31; N, 5.25. Found: C, 40.65; H, 8.48; N, 5.20. 5.6.3 {ZrCl[N(SiMe2CH2PBut2)2]}2(N2), 18d. The same procedure as for the above compound resulted in only a trace yield of product The best results came from stirring two equivalents of sodium sand (12 mg, 0.52 mmol) with ZrCl3[N(SiMe2CH2PBut2)2] (0.161 mg, 0.249 mmol) under one atmosphere of N 2 in toluene (60 mL) for two weeks. The yield of product (66 mg, 0.053 mmol) was 42%. *H NMR (8, 300 MHz, C 7D 8): at 60 °C 8 1.52 (br-s, 2 PBu»), 1.29 (br-m, 2 PBu*), 1.09 (d of d, J = 12 & 2, PCH2Si), 0.43 (s, SiMe), 0.28 (s, SiMe). at 20 °C 8 1.59 (t, J P . H = 6.3, PBu1), 1.47 (t, Jp.H = 6.0, PBu1), 1.29 (m, 2 PBu1), 1.1-0.8 (m, 4 PCH2Si), 0.47 (s, SiMe), 0.44 (s, SiMe), 0.32 (s, SiMe), and 0.29 (s, SiMe). 31p{lH} NMR (8, 121.4 MHz C 7 D 8 ) at 20 CC: +35.7 at -30° C: +35.1, +34.8 (AB quartet, 2 J P . P = 75.2 ). Anal. Calcd. for Zr2C44Hio4Cl2N4P4Si4: C, 44.83; H, 8.89; N, 4.75; CI, 6.01. Found: C, 45.10; H, 9.10; N, 4.87; CI, 5.90. 5.6.4 Hydrazine Analysis Hydrazine was analyzed colourimetrically according to the method of Watt and Chrisp.24 In a glovebox three samples of 18c, {ZrCl[N(SiMe2CH2PPri2)2])2(N2), (10, 12.5 & 21 mg each in 10 mL of toluene), were loaded into Schlenk tubes, sealed under N 2 , removed from the glovebox and attached to a vacuum line. The nitrogen was removed and replaced with anhydrous HCl, resulting in an instantaneous loss of blue colour and formation of a white precipitate. The toluene solutions were extracted with distilled water and filtered into 250 mL volumetric flasks; 2 mL aliquots of the 250 mL solutions were added to 10 mL aliquots of a colour developer (p-dimethylaminobenzaldehyde, 4.0 g; ethanol, 200 mL; concentrated HCl, 20 mL) and diluted to 25 mL with IM HCl. After 10 minutes, percent transmittance readings were taken. The spectrometer was set to 100% transmittance using a blank solution consisting 141 of 10 mL of colour developer diluted to 25 mL with IM. HCl. Percent transmittance readings were converted to hydrazine concentrations using equation 5.4 (this equation was determined earlier from solutions of known hydrazine concentration).24 The results are tabulated in Table 5.3. log[hydrazine] = -6.0780 + (1.4193 x 10-2)(100 - % transmittance) [5.4] A 24 mg sample of ZrCl3[N(SiMe2CH2PPri2)2] was treated in a similar manner and it was found to give 100% transmittance. Therefore, there is no interference with this hydrazine test from the zirconium or the other ancillary ligands. 5.7 References 1 Allen, A. D.; Senoff, C. V. J. Chem. Soc, Chem Commun. 1965, 621. 2 a) Pelikan, P.; Boca, R. Coord. Chem. Rev. 1984,55, 55. b) Henderson, R. A.; Leigh, G. J.; Pickett, C. J. Adv. Inorg. Chem. Radiochem. 1983, 27, 197. c) Dilworth, J. R.; Richards, R. L. in Comprehensive Organometallic Chemistry; Wilkinson, G.; Stone, F. G. A.; Abel, E. W. (eds.); Pergamon: Oxford, England, 1982, 8, Chapter 60, 1073. d) Chatt, J.; Dilworth, J. R.; Richards, R. L. Chem. Rev. 1978, 78, 589. 3 Bottomley, F.; Nyburg, S. C. / . Chem. Soc, Chem Commun 1966, 897. 4 Treitel, I. M.; Flood, M. T.; Marsh, R. E.; Gray, H. B. / . Am.. Chem. Soc. 1969, 91, 6512. 5 "Tables of Interatomic Distances and Configurations in Molecules and Ions", Chemical Society Special Publications; Sutton, L. E., (ed.); The Chemical Society: London, 1958, Vol 11. 6 a) Evans, W. J.; Ulibarri, T. A.; Ziller, J. W. J. Am. Chem. Soc. 1988,110, 6877. b) Pez, G. P.; Apgar, P.; Crissey, R. K. J. Am. Chem. Soc. 1982,104, 482. c) Hammer, R.; Klein, H.; Friedrich, P.; Huttner, G. Angew. Chem., Int. 142 Ed. Engl. 1977,16, 485. d) Jonas, K.; Brauer, D. J.; Kruger, C ; Roberts, P. J.; Tsay, Y.-H. / . Am. Chem. Soc. 1976, 98, 74. e) Kruger, C ; Tsay, Y.-H. Angew. Chem., Int. Ed. Engl. 1973,12, 998. 7 Ozin, G. A.; Voet, A. V. Can. J. Chem. 1973, J i , 638. 8 Green, D. W.; Reedy, G. T. / . Mol. Spectrosc. 1979, 74, 423. 9 Cotton, F. A.; Wilkinson, G. Advanced Inorganic Chemistry, 1s t Ed., John Wiley & Sons: Great Britain, 1962, 245. 1 0 Eady, R. R.; Postgate, J.R. Nature, 1974,249, 805. 1 1 Chatt, J. / . Organomet. Chem. 1975,100, 17. 1 2 a) Chatt, J. Nature, 1975,253, 39. b) Chatt J. / . Organometal. Chem., 1975, 100, 17. 1 3 Huheey, J. E. Inorganic Chemistry, 3 r d ed., Harper and Roe: New York, 1983, 892. 14 Farmers Weekly, Jan 10,1975. 15 The Province, Vancouver, British Columbia, Jan. 15,1975. 1 6 a) Manriquez, J. M.; Bercaw, J. E. / . Am. Chem. Soc, 1974, 96, 6229. b) Manriquez, J. M.; Sanner, R. D.; Marsh, R. E.; Bercaw, J. E. / . Am. Chem. Soc, 1976, 98, 3042. c) Sanner, R. D.; Manriquez, J. M.; Marsh, R. E.; D. R.; Rosenberg, E.; Shiller, A. M.; Williamson, K. L.; Bercaw, J. E. / . Am. Chem. Soc, 1978, 98, 3078. 1 7 a) Gynane, M. J. S.; Jeffery, J.; Lappert, M. F. / . Chem. Soc, Chem. Commun. 1978, 34. b) Jeffery, J.; Lappert, M. F.; Riley, P. I. / . Organomet. Chem. 1979, 181, 25. 1 8 Roddick, D. M.; Fryzuk, M. D.; Seidler, P. F.; Hillhouse, G. L; Bercaw, J. E. Organometallics, 1985,1, 97. 19 Fryzuk, M. D.; Haddad, T. S.; Rettig, S. J. Organometallics, 1989,5, 1723. 143 20 Walsh, P. J.; Hollander, F. J.; Bergman, R. G. / . Am. Chem. Soc. 1988,110, 8729. 21 Sanner, R. D.; Duggan, M.; McKenzie, T. C; Marsh, R. E.; Bercaw, J. E. / . Am. Chem. Soc. 1976, 98, 8358. 22 (a) O'Regan, M. B.; Liu, A. H.; Finch, W. C; Schrock, R. R.; Davis, W. M. / . Am. Chem. Soc. 1990,112, 4331. (b) Schrock, R. R.; Kolodziej, R. M.; Liu, A. H.; Davis, A. M.; Vale, M. G. / . Am. Chem. Soc. 1990,112, 4338. 23 24 Thorn, D. L.; Nugent, W. A.; Harlow, R. L. / . Am. Chem. Soc. 1981,103, 357. Watt, G. W.; Chrisp, J. D. Anal. Chem., 1952,24, 2006. 144 THESIS SUMMARY AND FUTURE PROSPECTS The synthesis of a variety of new lanthanoid phosphine complexes has been achieved by complexing either one or two amido-diphosphine ligands to yttrium, lutetium and lanthanum. The bis(ligand) compounds, MCl[N(SiMe2CH2PR2)2l2> appear to be sterically saturated seven-coordinate monomers; none of these complexes show any tendency to coordinate THF, indicating that the metal coordination sphere is more saturated than a bis (cyclopentadienyl) lanthanoid complex, Cp2LnC l . However, the amido-diphosphine complex with the largest cation and the least-basic ligand, LaCl[N(SiMe2CH2PPh2)2]2> seem to display a monomer-dimer equilibrium, indicating that for lanthanum, eight coordination is at least a possibility. All of the bis(ligand) compounds are fluxional and display NMR spectra indicative of complexes where the phosphorus donors are rapidly exchanging, probably via a dissociation-reassociation pathway. As the ligands are anchored to the host metal via the amide donor, the phosphines cannot be lost to the bulk solvent. Although these phosphines are coming on and off the metal, it may be that the chelate effect is too strong, and that the complexation of two amido-diphosphine ligands shuts down the chemistry one might expect to be associated with these strong Lewis acids. Of all the thermally unstable hydrocarbyl complexes that were synthesized, the only one for which it may be possible to study the reactivity of the lanthanoid-carbon bond is the lutetium-phenyl complex, Lu(Ph)[N(SiMe2CH2PMe2)2]2- However, the inevitable formation of the cyclometallated bis(ligand) complex will be problematic. Of all the cyclometallated complexes, the least sterically crowded one (and therefore the i 1 most reactive) should be the lanthanum derivative, La[N(SiMe2CHPMe2)(SiMe2CH2-PMe2 ) ] [N (S iMe2CH2PMe2)2l- However, it should be noted, that this compound 145 does not react with propylene and is far less reactive than any bis(cyclopentadienyl)lanthanoid-alkyl complex. It would be fortunate if the stabilizing effect of two of these ligands on one metal could be used to generate new and unusual complexes (e.g., low valent Ln(II) compounds). Preliminary attempts to synthesize the unprecedented Y(II) complex, Y[N(SiMe2CH2PMe2)2l2> via the reduction of the Y(III) chloride, YCl[N (S iMe2CH2PMe2)2]2, with various sodium based reducing agents gave intensely-blue solutions from which the ubiquitous cyclometallated complex was isolated. While the bis(ligand) chemistry appears to always lead to the cyclometallated complexes, compounds with just one amido-diphosphine ligand show much more promise. A variety of these complexes, MCl2 [N (S iMe2CH2PR2)2l , (R = Me, Ph, PrJ, Bu l), were synthesized and just these simple compounds showed a varied chemistry. The derivatives with methyl substituents at phosphorus are too sterically unsaturated, and consequently, are obtained as insoluble polymeric powders. They will dissolve in THF, but most attempts at alkylation led to decomposition; once again the cyclometallated bis(ligand) complex was identified as the major product. A route to a mono(ligand) allyl complex, {Y(allyl)[N(SiMe2CH2PMe2)2l }2(M--C1)2, w a s found, and it may be that by derivatizing this compound with a bulky alkoxide, a reactive monomeric yttrium-allyl complex, Y(OR)(allyl)[N(SiMe2CH2PMe2)2] can be obtained. The other three mono(ligand) yttrium complexes YCl2 [N (S iMe2CH2PR2)2l> (R = Ph, Pr\ BU1) are hydrocarbon-soluble and can all be obtained as THF adducts. The derivative with isopropyl substituents at phosphorus was also obtained as a THF-free dimer, indicating that the metal is not sterically saturated. This dimer, { Y C l 2 [ N ( S iMe2CH2PPr i 2)2 ] }2 , h a s good solubility properties and it should be 146 possible to generate some stable hydrocarbyl derivatives. Related mono(ligand) derivatives of the trivalent metals Yb, Sm, Eu should be accessible using the strategies developed in this thesis. With the added flexibility of the divalent state available for these elements (e.g., {LnX[N(SiMe2CH2PPr i2)2])n) the potential exists for a rich and varied chemistry. The group 4 chemistry in this thesis was developed from the simple concept that the amido-diphosphine ligand should be capable of stabilizing the lower oxidation states of these metals. In fact, for the diene complexes, MR(r j 4 -C4H6)-[N(SiMe 2CH2PR2)2L the M(II) resonance structure is a better description than the M(IV) metallacyclopentene resonance form. The reaction where where allyl and butadiene fragments are coupled together to generate a coordinated [rj 4:r| 1-C H 2 = C H C H = C H C H 2 C H 2 C H 2 ] 1 " fragment can also be explained in terms of an electrocyclic rearrangement at a M(II) centre. This reaction is both regioselective and partially diastereoselective when 1-methylallyl and butadiene units are coupled, as two of the four possible coupled products are formed in unequal amounts; the coupling occurs exclusively at the substituted end of the 1-methylallyl unit (as determined by X-ray crystallography) and one diastereomer is formed in higher yield than the other. The selectivity demonstrated by this coupling warrants further investigation. It would be interesting to see how general this reaction is by varying both the diene and the allyl substituents. Another variation would be to synthesize a chiral amido-diphosphine ligand to see if this reaction could also be made enantioselective. As diene rotation interconverts enantiomers, it is possible that an enantioselective coupling could occur whereby only one enantiomer would be formed. Finally, it might be possible to do a complete kinetic study on the coupling reaction as the transformation from the allyl-diene complex to the final coupled product (via an unknown intermediate) can be followed by 3 1 P { 1 H} NMR spectroscopy. 147 As a continuation of the investigation into low oxidation state zirconium complexes, the trichlorides, ZrCl3[N(SiMe2CH2PR.2)2] (R = Pr5 or Bul), were reduced with sodium. Dimeric dinitrogen complexes, {ZrCl[N(SiMe2CH2PR2)2l}(M-- rl 2: rl 2-N2), were isolated, and are best formulated as containing Zr(IV) centres. It would appear that the side-on binding of the N 2 ligand favors electron transfer to form a N 2 4 _ hydrazido ligand. As these complexes have a chloride ligand available for substitution, it would be interesting to find out what effect the introduction of a hydride, alkyl or cyclopentadienyl ligand will have on the structure and chemistry of these compounds. The chemistry associated with this unique hydrazido (4~) ligand should be further investigated, in particular, the reaction with methylenechloride and chloroform (or other alkylhalides). It would also be interesting to find a way of recycling the dinitrogen complex back into ZrCl3[N(SiMe2CH2PR2)2L along with the release of hydrazine. Further investigations into the chemistry of low-valent zirconium compounds may also prove interesting; the synthesis of a phosphine complex such as ZrCl (PR3)2 [N (S iMe2CH2PR2)2l may be a way of stabilizing a very reactive Zr(II) species. The results from this thesis demonstrate that the strategy of using an amido-diphosphine ligand to generate phosphine complexes of the early metals is successful. In addition, these phosphine complexes display reactivity patterns different from those observed for the analogous complexes incorporating cyclopentadienyl ligands. From the observed chemistry, it appears that the amido-diphosphine ligands are better able to coordinatively saturate these metals than a Cp or Cp* ligand. 148 A P P E N D I X A -1 Xray Crystallographic Analysis of r Y [ N ( S i M e 2 C H P M e 2 ) ( S i M e 2 C H 2 P M e 2 ) ] [ N ( S i M e 2 C H 2 P M e 2 ) 2 ] 149 E m p i r i c a l Formula Formula Weight C r y s t a l System L a t t i c e P a r a m e t e r s : Space Group Z v a l u e D c a l c F000 mu(Cu K-alpha) D i f f T a c t o m e t e r R a d i a t i o n Temperature 2-theta(max) No. O b s e r v a t i o n s (I>3 . 0 0 ( s i g ( I ) ) No. V a r i a b l e s R e s i d u a l s : R; Rw Goodness of F i t I n d i c a t o r Maximum S h i f t i n F i n a l C y c l e L a r g e s t Peak i n F i n a l D i f f . Map C(20)H(55)N(2)P ( 4)Si ( 4)Y(l) 648.81 M o n o c l i n i c a - 18.115 (1) angstroms b - 15.875 (8) angstroms c - 12.706 (2) angstroms b e t a - 92.619 (7) degrees V - 3650 (2) angstroms**3 P21/c (#14) 4 1.18 g/cm**3 1376 53.58 cm**-l R i g a k u AFC6 Cu K-alpha (lambda= 1.54178) G r a p h i te-monochromated 21 degrees Cent. 155.3 degrees ) 3270 281 0.063; 0.073 2.16 0.10 1.42 e/angstrom**3 150 Positional parameters and B(eq) atom X y z B(eq) Y(l) 0.25215(4) 0 .51693(6) 0 .22160(7) 4 .27(4) P(D 0.3611(1) 0 .3742(2) 0 .2363(2) 4 .9(1) P(2) 0.1913(2) 0 .6800(2) 0 .2297(3) 6 .0(2) P(3) 0.2156(1) 0 .4535(2) 0 .0119(2) 4 .7(1) P(4) 0.2449(2) 0 .5025(2) 0 .4487(2) 6 .2 (2 ) Si(l') 0.4476(1) 0 .5377(2) 0 .2740(2) 4 .8(1) Si (2 ) 0.3510(1) 0 .6610(2) 0 .1454(2) 4 .8 (1 ) Si (3 ) 0.1256(1) 0 .3467(2) 0 .1621(2) 4 .6(1) Si (4 ) 0.0940(2) 0 .4540(2) 0 .3451(2) 5.3(2) N(l) 0.3672(4) 0 .5710(5) 0 .2184(6) 4 .2(4) N(2) 0.1467(4) 0 .4286(5) 0 .2431(6) 4 .3(4) C(l) 0.4319(5) 0 .4251(7) 0 .3178(8) 5.6(6) C(2) 0.2496(5) 0 .6659(7) 0 .1270(8) 5.8(6) C(3) 0.1858(5) 0 .3484(6) 0 .0430(8) 5.3(5) C(4) 0.1466(6) 0 .5180(8) 0 .4462(9) 6 .9(7) C(5) 0.4082(6) 0 .3532(8) 0 .116(1) 7.6(7) C(6) 0.3552(6) 0 .2684(8) 0 .292(1 ) 8 .8(8) C(7) 0.1013(6) 0 .702(1) 0 .167(1) 10(1) C(8) 0.2080(8) 0 .782(1) 0 .301(1) 11(1) C(9) 0.1323(7) 0 .5027(8) -0 .043(1) 8.2(8) C(10) 0.2696(7) 0 .4429(9) -0 .1048(9) 7.5(7) C ( l l ) 0.2854(8) 0 .576(1) 0 .546(1) 12(1) C(12) 0.2577(7) 0 .404(1) 0 .517(1) 10(1) C(13) 0.5267(6) 0 .5379(9) 0 .184(1) 8.5(8) 151 P o s i t i o n a l parameters and B(eq) atom X y z B(eq) C(14) 0 .4802(6) 0 .5987(8) 0 .393(1) 7.8(8) C(15) 0 .3983(6) 0 .7542(7) 0 .205(1) 7.5(7) C(16) 0 .3870(7) 0 .6545(9) 0 .009(1) 8.1(8) C(17) 0 .1426(6) 0 .2408(7) 0 .2242(9) 6 .4(6) C(18) 0 .0273(5) 0 .3452(8) 0 .1103(9) 6 .5(7) C(19) 0 .0096(6) 0 .517(1) 0 .306(1) 9.0(8) C(20) 0 .0583(6) 0 .3599(9) 0 .418(1) 8.1(8) 152 I n t r a m o l e c u l a r D i s t a n c e s atom atom distance Y(l) N(l) 2 .256(7) Y(l) N(2) 2 .396(7) Y d ) C(2) 2 .65(1) Y d ) P(2) 2 .817(3) Y d ) P(3) 2 .896(3) Y d ) P(4) 2 .903(3) Y(l) P d ) 3 .005(3) P d ) C d ) 1 .80(1) P d ) C(5) 1 .82(1) P d ) C(6) 1 .83(1) P(2) C(2) 1 .73(1) P(2) C(7) 1 .81(1) P(2) C(8) 1 .86(2) P(3) C(3) 1 .80(1) P(3) C(9) 1 .81(1) P(3) C(10) 1 .82(1) P(4) C(4) 1 .80(1) P(4) C(12) 1 .80(2) atom atom distance P(4) C ( l l ) 1 .83(1) S i d ) N d ) 1 .675(7) S i d ) C(14) 1 .87(1) S i d ) C(13) 1 .88(1) S i d ) C(l) 1 .90(1) Si(2) N(l) 1 .721(8) Si(2) C(2) 1 .84(1) Si(2) C(15) 1 .86(1) Si(2) C(16) 1 .88(1) Si(3) N(2) 1 .690(8) Si(3) C(18) 1 .87(1) Si(3) C(17) 1 .88(1) Si(3) C(3) 1 .91(1) Si(4) N(2) 1 .692(7) Si(4) C(4) 1 .86(1) Si(4) C(19) 1 .88(1) Si(4) C(20) 1 .89(1) 153 Intramolecular Bond Angles atom atom atom angle N(l) Y(l) N ( 2 ) 1 6 5 . 2 ( 3 ) N(l) Y d ) C ( 2 ) 6 9 . 5 ( 3 ) N(l) Y d ) P ( 2 ) 9 0 . 9 ( 2 ) N(l) Y d ) P ( 3 ) 1 0 6 . 7 ( 2 ) N(l) Y d ) P ( 4 ) 9 7 . 6 ( 2 ) N(l) Y d ) P d ) 7 1 . 5 ( 2 ) N ( 2 ) Y(l) C ( 2 ) 1 2 5 . 2 ( 3 ) N ( 2 ) Y d ) P ( 2 ) 1 0 2 . 6 ( 2 ) N ( 2 ) Y d ) P ( 3 ) 7 5 . 6 ( 2 ) N( 2 ) Y(l) P ( 4 ) 7 6 . 6 ( 2 ) N ( 2 ) Y d ) P d ) 9 4 . 4 ( 2 ) C ( 2 ) Y d ) P ( 2 ) 3 6 . 7 ( 2 ) C ( 2 ) Y d ) P ( 3 ) 8 3 . 9 ( 2 ) C ( 2 ) Y d ) P( 4 ) 1 2 1 . 3 ( 2 ) C ( 2 ) Y d ) P d ) 1 3 4 . 3 ( 2 ) P ( 2 ) Y d ) P ( 3 ) 1 0 6 . 2 ( 1 ) P ( 2 ) Y d ) P ( 4 ) 9 0 . 0 ( 1 ) P ( 2 ) Y(l) P d ) 1 6 1 . 3 8 ( 9 ) P ( 3 ) Y(l) P ( 4 ) 1 5 0 . 2 ( 1 ) P ( 3 ) Y d ) P d ) 6 5 . 2 4 ( 8 ) P ( 4 ) Y d ) P d ) 8 6 . 5 ( 1 ) C(l) P d ) C ( 5 ) 1 0 2 . 5 ( 5 ) C(l) P d ) C ( 6 ) 1 0 4 . 2 ( 6 ) C(l) P d ) Y d ) 9 8 . 4 ( 4 ) C( 5 ) P d ) C ( 6 ) 1 0 1 . 2 ( 6 ) C ( 5 ) P d ) Y d ) 1 1 4 . 7 ( 4 ) atom atom atom angle C(6) P d ) Y(l) 131.9(4) C ( 2 ) P ( 2 ) C(7) 105.2(6) C ( 2 ) P ( 2 ) C ( 8 ) 112.6(6) C ( 2 ) P(2) Y d ) 66.5(4) C(7) P ( 2 ) C ( 8 ) 99.6(7) C(7) P ( 2 ) Y d ) 120.5(5) C ( 8 ) P(2) Y(l) 139.2(5) C(3) P(3) C(9) 103.3(5) C(3) P(3) C(10) 105.6(6) C(3) P(3) Y d ) 100.3(3) C(9) P(3) C(10) 101.4(6) C(9) P(3) Y d ) 110.8(5) CdO) P(3) Y d ) 132.3(4) C(4) P( 4 ) C(12) 103.5(6) C(4) P( 4 ) C ( l l ) 106.9(6) C(4) P( 4 ) Y d ) 93.5(4) C(12) P( 4 ) C ( l l ) 101.0(8) C(12) P( 4 ) Y d ) 122.4(5) C d l ) P(4) Y(l) 126.0(5) N(l) S i ( l ) C(14) 114.5(5) N(l) S i d ) C(13) 114.5(5) N(l) S i ( l ) C(l) 106.4(4) C(14) S i ( l ) C(13) 105.8(6) C(14) S i d ) C(l) 107.2(5) C(13) S i ( l ) C(l) 108.1(5) N(l) Si(2) C(2) 104.4(4) 154 atom atom atom angle N(l) £i(2) C(15) 112 .2(4) N(l) Si(2) C(16) 113 .1(5) C(2) Si(2) C(15) 117 .2(5) C(2) Si(2) C(16) 105 .8(5) C(15) Si(2) C{16) 104 .3(6) N(2) Si(3) C(18) 114 .1(5) N(2) Si(3) C(17) 113 .9(4) N(2) Si(3) C(3) 110 .6(4) C(18) Si(3) C(17) 105 .8(5) C(18) Si(3) C(3) 106 .9(5) C(17) Si ( 3 ) C(3) 105 .0(5) N(2) Si ( 4 ) C(4) 111 .7(4) N(2) Si(4) Cd9) 113 • 9(5) N{2) Si(4) C(20) 113 .8(5) C(4) Si(4) Cd9) 106 .1(6) C(4) Si ( 4 ) C(20) 105 .6(6) C(19) Si ( 4 ) C(20) 105 .0(6) S i d ) Nd) Si(2) 127 .8(4) S i d ) N d ) Y d ) 131 .2(4) Si(2) N d ) Y d ) 101 .0(3) Si(3) N(2) Si(4) 122 .0(4) Si(3) N(2) Y d ) 122 .6(4) Si(4) N(2) Y d ) 115 .3(4) P d ) C d ) S i d ) 111 .5(5) P(2) C(2) Si(2) 123 .3(6) P(2) C(2) Y(l) 76. 8(4) atom atom atom angle Si(2) C(2) Y d ) 84.7(4) P(3) C(3) Si(3) 112.0(5) P(4) C(4) Si(4) 114.3(6) 155 A - 2 Xray Crystallographic Analysis of {Y(Ti3-allyl)[N(SiMe2CH2PMe2)2]h(u-Cl)2 156 formula C 26H66Cl2N2P4Si 4 Y2 crystal size , mm 0.13x0.25x0.37 crystal system monoclinic space group P2l/a a, A 12.0949(9) b, A 14.2647(11) c A 13.278(2) P, deg 101.555(8) < y , A 3 2244.3(4) z 2 Dc, g/cm 3 1.320 radiation M o wavelength (A) 0.71073 |1, cm" 1 29.8 2 0 max> d eS 55.0 total no. of reflections 5136 no. of unique reflections 2272 no. of variables 181 R 0.035 Rw 0.038 g.o.f. 1.15 residual density e/A3 0.46 157 Final positional (fractional x 10*; Y, CI, P, and Si x 10 5) and isotropic thermal parameters (U x 103 A2) with estimated standard deviations in parentheses Atom X y z Y 42306( 4) 45717( 3) 34398( 4) 41 C l 4 2 6 6 2 0 0 ) 60283( 8) 4 7 9 7 7 0 0 ) 50 P O ) 18189 (11 ) 4 6 4 7 0 0 0 ) 3 4 8 4 1 O 1 ) 53 P (2 ) 62229( 1 1 ) 55287( 9) 3 0 7 7 7 ( 1 1 ) 53 S i O ) 2 1 3 5 4 ( 1 3 ) 5 0 8 7 6 ( 1 2 ) 12831 (12 ) 59 S i ( 2 ) 4 1 7 3 5 ( 1 3 ) 62584( 9) 15244 (11 ) 49 N 3451 ( 3) 5364( 3) 1966( 3) 45 C O ) 1 338 ( 4 > 4400( 4) 21 28 ( 4) 64 C ( 2 ) 5729( 4) 5952( 4) 1781 ( 4) 57 C ( 3 ) 1282( 5) 581 0( 4) 3647( 5) 73 C ( 4 ) 894( 4) 3941 ( 4) 41 10( 4) 68 C ( 5 ) 6682( 5) 6586 ( 4) 3800( 4) 65 C ( 6 ) 7572( 5) 4945( 4) 3060( 5) 79 C ( 7 ) 1 233 ( 5) 6141 ( 5) 830( 5) 88 C ( 8 ) 2166( 5) 4335( 5) 1 23 ( 5) 86 C ( 9 ) 3947( 4) 7389 ( 3) 2151 ( 4) 59 C O O ) 3844( 5) 6477( 4) 97( 4) 78 C O D 3607( 5) 2837 ( 4) 3227 ( 6) 79 C ( 1 2 ) 4203( 6) 2988( 4) 2453( 5) 75 C ( 1 3 ) 5287( 6) 3287( 4) 2582( 5) 66 158 0 Bond lengths (A) with estimated standard deviations in parentheses Bond Length(A) Bond Length(A) Y -CI 2.746(1) P(2)-C(5) 1.813(6) y -Pd) 2.931(1) P(2)-C(6) 1.835(6) Y -P(2) 2.892(1) Si(1)-N 1.713(4) Y -N 2.292(4) S i d ) - C ( l ) 1.893(6) y -c( 1 1 ) 2.587(5) Si(1)-C(7) 1.884(6) Y -C(12) 2.609(5) S i ( D - C ( 8 ) 1.884(6) y -c(i3) 2.621(5) Si(2)-N 1.715(4) Y -CI' 2.795(1) Si(2)-C(2) 1.894(5) y - A 2.389(3) Si(2)-C(9) 1.860(5) P(1)-C(1) 1.812(6) SK2)-C(10) 1.883(6) P ( 1 )-C(3) 1.810(6) C(1 1 )-Cd2) 1.387(9) P(1 )-C(4) 1.825(5) C( 12)-C(13) 1.356(8) P ( 2 ) - C ( 2 ) 1.810(5) 159 Bond angles (deg) with estimated standard deviations in parentheses* Bonds Angle(deg) Bonds Angle(deg) CI - Y -p(D 81.02(4) Y -P(2)-C(5) 118.8(2) CI - Y -P(2) 81.26(4) Y -P(2)-C(6) 124.1(2) CI - Y -N 97.88(9) C(2)-P(2)-C(5) 103.5(2) CI - Y -CI' 76.11(4) C(2)-P(2)-C(6) 105.3(3) CI - Y -A 161.9(1) C(5)-P(2)-C(6) 101.8(3) p(D - Y -P(2) 148.38(4) N - S i ( D -Cd ) 109.6(2) P d ) -y -N 76.01(10) N -Si( l )-C(7) 113.8(2) p(D - Y -CI' 118.52(4) N -Si(D-C(8) 113.3(2) p(D -y -A 100.7(1 ) C(1)-Si(1)-C(7) 105.7(3) P(2) - Y - N 80.75(10) C (D-Si( l )-C(8) 106.6(3) P(2) -y -CI' 81.94(4) C(7)-S id)-C(8) 107.3(3) P(2) - Y -A 104.3(1 ) N -Si(2)-C(2) 108.8(2) N -y -CI' 162.37(10) N -Si(2)-C(9) 1 11.2(2) N -y -A 100.0(1 ) N -Si(2)-C(l0) 116.6(2) CI ' - Y -A 87.5(1) C(2)-Si(2)-C(9) 110.5(2) y -c i - y 103.89(4) C(2)-Si (2)-Cd0) 102.8(3) Y - P d ) - c d ) 95.5(2) C(9)-Si (2)-Cd0) 106.6(3) Y - P d ) -C(3) 114.7(2) Y -N - S i d ) 121.2(2) y - P d ) -CU) 133.8(2) Y - K -Si(2) 120.6(2) c d ) - p d ) -C(3) 104.2(3) Si(D-N -Si(2) 118.2(2) c d ) - P d ) -C(4) 104.0(3) P(1)-C(1)-Si(1) 112.8(3) C(3) - P d ) -C(4) 100.6(3) P(2)-C(2)-Si(2) 112.0(2) Y -P (2) -C(2) 100.9(2) C(1 1 )-Cd2)-C(13) 126.1(6) * Here and e l s e w h e r e , primed atoms a r e r e l a t e d t o th o s e i n Tab l e 1 by the c r y s t a l l o g r a p h i c i n v e r s i o n c e n t e r and A r e f e r s t o the c e n t r o i d of the a l l y l l i g a n d . 160 A-3 Xray Crystallographic Analysis of Hf (Ti4 :nl -C7Hi i ) [N(SiMe 2 CH2PPr i 2)2] 161 formula crystal system space group a, A b, A c A u,A3 z radiation no. of unique reflections R Rw C 2 5H55NP 2Si2Hf tetragonal W i 10.2288(5) 10.2288(5) 30.247(2) 3164.7(3) 4 C u K a 1809 0.052 0.046 162 F i n a l p o s i t i o n a l ( f r a c t i o n a l x 1 0 * , Hf x 1 0 5 ) and i s o t r o p i c thermal parameters (U x 1 0 3 A 2 ) with es t imated s tandard d e v i a t i o n s in pa ren theses Atom y 2 17 /[/. eq iso Hf 8645( 9) 39907(10) 25000 40 P(1) 556( 6) 2164( 6) 1833( 2) 46 P(2) 2487( 7) 5548( 6) 3035( 2) 53 s i d ) 3456( 6) 2637( 7) 1976( 2) 45 S i ( 2 ) 3625( 7) 2831( 6) 2962( 2) 48 N 2728(13) 3064(12) 2487( 5) 36 c d ) 2116(18) 2504(20) 1575( 6) 46 C(2) 3117(24) 4194(26) 3341 ( 8) 65 C(3) -687(26) 2058(30) 1402( 10) 92 C U ) 738(24) 446(21 ) 2028( 8) 63 C(5) 2038(33) 6850(32) 3445( 10) 103 C(6) 3954(23) 6180(22) 2767 ( 8) 63 C(7) 4672(27) 3906(28) 1777( 8) 73 C(B) 4363(22) 1045(24) 201 1 ( 7) 66 C(9) 5448(22) 2871(28) 291 8( 9) 85 c d o ) 3160(32) 1197(25) 3220( 8) 92 C( 1 1 ) -2030(27) 2350(31 ) 1570( 9) 86 C( 12) -371(34) 2772(38) 996( 11 ) 127 C(13) -508(34) -76(30) 2252( 12) 133 C(14) 1285(28) -510(26) 1676( 11 ) 97 C(15) 1276(39) 7926(29) 3221 ( 1 1 ) 120 C(16) 1610(41) 6437(32) 3881 ( 12) 157 C(17) 3686(25) 7200(22) 2435( 1 1 ) 91 C(18) 5037(32) 6661 (31 ) 3082( 1 1 ) 126 C(19) -466(22) 2596(24) 2906( 8) 68 C(20) -416(30) 3799(31) 31 47 ( 8) 80 C(21 ) -993(33) 4976(34) 2960( 19) 135 C(22) -1479(25) 4923(30) 2542( 10) 70 C(23) -1684(31) 6084(31 ) 2281 ( 13) 115 C(24) -831(37) 5943(34) 1866( 11 ) 123 C(25) 541(27) 5533(25) 1996( 9) 58 ( 7) ft Bond lengths (A) with estimated standard d e v i a t i o n s i n parentheses Bond Length(A) Bond Length(A) Hf -P(1) 2.767(6) Si ( 2)-N 1.72(2) Hf -P(2) 2.812(6) Si ( 2 ) - C ( 2 ) 1.88(3) Hf -N 2.129(13) Si ( 2 ) - C ( 9 ) 1 .87(2) H f -C(19) 2.32(2) Si ( 2 ) - C d 0 ) 1.90(3) Hf -C(20) 2.36(2) C(3)-C(11) 1.49(4) Hf -C(21) 2.56(3) C(3)-C(12) 1.47(4) Hf -C(22) 2.58(3) C(4)-C(13) 1.54(3) Hf -C(25) 2.22(3) C(4)-C(14) 1.55(3) P(1 ) - c d ) 1.81(2) C(5)-C(15) 1.51(4) P(1)-C(3) 1.83(3) C(5)-C(16) 1.45(4) P (1 ) -C(4) 1.86(2) C(6)-C(17) 1.47(3) P (2 ) -C(2) 1.79(3) C(6)-C(18) 1.54(3) P (2 ) -C(5) 1.8B(2) C(19)-C(20) 1.43(3) P (2 ) -C(6) 1.82(3) C(20)-C(21 ) 1.45(5) S i ( 1 ) - N 1.77(2) C(21)-C(22) 1.36(5) Si (1 ) -C(1 ) 1.83(2) C(22)-C(23) 1.44(4) S i d ) - C ( 7 ) 1.90(3) C(23)-C(24) 1.53(4) Si (1)-C(8) . 1.88(2) C(24)-C(25) 1.52(4) Bond angles (deg) with estimated standard deviations in parentheses Bonds Angle(deg) Bonds Angle(deg) P(1)-Hf -P(2) 150.31 [2) N -Si(1)-C(7) 112.6(9) P(1)-Hf -N 77.8 (4) N -S id )-C(8 ) 111.8(8) P(1)-Hf -C(19) 84.4 [6) C(1)-Si(1)-C(7) 109.4(11) P(1)-Hf -C(20) 119.0 [7) C(1)-Si(l)-C(6) 110.0(10) P(1)-Hf -C(21) 125.2 [10) C(7)-Si(1)-C(8) 106.7(12) P(1)-Hf -C(22) 100.3 (7) N -Si(2)-C(2) 105.0(9) P(1)-Hf -C(25) 87.8 [7) N -Si(2)-C(9) 117.9(10) P(2)-Hf -N 74.6 [4) N -SU2)-C(10) 109.3(10) P(2)-Hf -C(19) 112.9 (7) C(2)-Si(2)-C(9) 107.7(12) P(2)-Hf -C(20) 84.1 [7) C(2)-Si(2)-C(l0) 109.5(12) P(2)-Hf -C(21) 84.4 (10) C(9)-Si(2)-C(l0) 107.3(13) P(2)-Hf -C(22) 108.1 (7) Hf -N -Si(1) 120.2(8) P(2)-Hf -C(25) 94.6 (7) Hf -N -Si(2) 121.5(9) N -Hf -C(19) 105.1 [7) Si(D-N -Si(2) 118.1(8) N -Hf -C(20) 118.4 (9) P(1)-C(1)-Si(D 112.8(11 ) N -Hf -C(21) 148.1 (14) P(2)-C(2)-Si(2) 111.1(13) N -Hf -C(22) 174.9 (8) P(1)-C(3)-C(11) 113(2) N -Hf -C(25) 115.9 (8) P(1)-C(3)-C(12) 115(2) C(19)-Hf -C(20) 35.6 (8) C(11)-C(3)-C(12) 113(3) C(l9 )-Hf -C(21 ) 61.3 (10) P(1 )-C(4)-C(13) 113(2) C(l9 )-Hf -C(22) 69.9 (9) P( 1 )-C(4)-C(14) 114(2) C(l9 )-Hf -C(25) 135.4 (9) C(13)-C(4)-C(14) 112(2) C(20)-Hf -C(21) 34.1 (12) P(2)-C(5)-C(15) 110(2) C(20)-Hf -C(22) 58.4 (10) P(2)-C(5)-C(16) 118(3) C(20)-Hf -C(25) 123.0 (12) C(15)-C(5)-C(16) 118(3) C(21)-Hf -C(22) 30.7 (12) P(2)-C(6)-C(17) 114(2) C(21)-Hf -C(25) 89.0 (12) P(2)-C(6)-C(18) 116(2) C(22)-Hf -C(25) 68.5 (9) C(17)-C(6)-C(18) 109(2) Hf -P(1 l-C(1 ) 94.8 (7) Hf -C(19)-C(20) 73.8(13) Hf -P(1 »-C(3) 129.9 (10) Hf -C(20)-C(19) 70.6(13) Hf -P(1 >-C(4) 113.3 (7) Hf -C(20)-C(21) 80(2) C(1)-P(1 >-C(3) 108.5 (12) C(19)-C(20)-C(21) 120(3) C(1)~P(1 )-C(4) 103.2 (11) Hf -C(21)-C(20) 65.5(14) C(3)-P(1 >-C(4) 103.8 (12) Hf -C(21)-C(22) 76(2) Hf -P(2 >-C(2) 94.1 (8) C(20)-C(21)-C(22) 118(4) Hf -P(2 >-C(5) 129.6 (11) Hf -C(22)-C(21) 74(2) Hf -P(2 >-C(6) 115.5 (7) Hf -C(22)-C(23) 114(2) • C(2)-P(2 >-C(5) 107.3 (14) C(21)-C(22)-C(23) 122(4) C(2)-P(2 l-C(6) 102.0 (11) C(22)-C(23)-C(24) 107(2) C(5)-P(2 >-C(6) 104.0 (13) C(23)-C(24)-C(25) 110(3) N -Si( l)-C(1) 106.3 (8) Hf -C(25)-C(24) 121(2) 165 A-4 Xray Crystallographic Analysis of Zr(Ti4:nl-C8Hi3)[N(SiMe 2CH2PPr*2)2] 166 A. Crystal Data Empirical Formula Formula Weight Crystal Color, Habit Crystal Dimensions (mm) Crystal System No. Reflections Used for Unit C e l l Determination (2e range) Omega Scan Peak Width at Half-height L a t t i c e Parameters: Space Group Z value D c a l c F 0 0 0 ^(MoKa) DiffT a c t o m e t e r Radiation Temperature C 2 6 H 5 7 N P 2 S i 2 Z r 593.08 dark, prism 0.200 X 0.320 X 0.480 t r i c l i n i c 25 ( 28 .0 - 3 4 . 0 ° ) 0.51 a •= 11.028 ( 2)A b - 15.796 (2)A c - 10.596 (2)A a - 104.59 (1)° P - 108.46 (2)° Y - 100.07 (1)° V - 1628 (1)A 3 Pi (#2) 2 1.210 g/cm3 636 5.15 cm~* B. Intensity Measurements Rigaku AFC6S MoKa (X « 0.71069 A) 21°C Take-off Angle 6.0 ' D e t e c t o r A p e r t u r e 167 6.0 sun h o r i z o n t a l 6.0 sun v e r t i c a l C r y s t a l t o D e t e c t o r D i s t a n c e Scan Type Scan Rate Scan W i d t h 26. max No. of R e f l e c t i o n s Measured 285 mm co-26 32.0°/min ( i n omega) (8 r e s c a n s ) (1.21 + 0.35 tan6)° 55.0° T o t a l : 7841 Unique: 7449 (R i n t - .030) C o r r e c t i o n s L o r e n t z - p o l a r i z a t i o n A b s o r p t i o n ( t r a n s , f a c t o r s : 0.60 - 1. Decay ( -3.00% d e c l i n e ) C. S t r u c t u r e S o l u t i o n and Refinement S t r u c t u r e S o l u t i o n Refinement F u n c t i o n M i n i m i z e d L e a s t - s q u a r e s Weights p - f a c t o r Anomalous D i s p e r s i o n No. O b s e r v a t i o n s ( I> 3 . 0 0 e ( I ) ) No. V a r i a b l e s R e f l e c t i o n / P a r a m e t e r R a t i o R e s i d u a l s : R; R w Goodness of F i t I n d i c a t o r Max S h i f t / E r r o r i n F i n a l C y c l e Maximum Peak i n F i n a l D i f f . Map Minimum Peak i n F i n a l D i f f . Map Patterson Method Full-matrix least-squares I w (|Fo| - |Fc | ) 2 4Fo 2/o 2(Fo 2) 0.05 A l l non-hydrogen atoms 4775 307 15.55 0.058; 0.071 1.81 0.03 1.77 e~/k\ -1.40 e"/A 3 168 F i n a l atomic c o o r d i n a t e s ( f r a c t i o n a l ) and B atom X y z B i eq Z r ( l ) 0 . 4 7 9 8 7 ( 4 ) 0 . 2 0 7 2 5 ( 3 ) 0 . 2 7 3 9 0 ( 5 ) 3 . 6 4 ( 2 ) P ( D 0 . 7 4 4 1 ( 1 ) 0 . 2 6 2 5 7 ( 8 ) 0 . 4 5 9 6 ( 1 ) 4 . 2 0 ( 4 ) P ( 2 ) 0 . 2 7 1 6 ( 1 ) 0 . 2 3 6 5 7 ( 8 ) 0 . 0 6 3 1 ( 1 ) 4 . 4 0 ( 4 ) S i d ) 0 . 7 0 5 2 ( 1 ) 0 . 3 9 9 0 4 ( 8 ) 0 . 3 0 4 9 ( 1 ) 4 . 2 8 ( 4 ) S i ( 2 ) 0 . 5 3 1 0 ( 1 ) 0 . 2 7 3 2 2 ( 8 ) 0 . 0 1 3 5 ( 1 ) 3 . 8 6 ( 4 ) N(l) 0 . 5 7 9 9 ( 4 ) 0 . 3 0 1 6 ( 2 ) 0 . 1 9 4 3 ( 4 ) 3 . 7 ( 1 ) Cd) 0 . 7 7 4 5 ( 5 ) 0 . 3 8 4 1 ( 3 ) 0 . 4 8 1 8 ( 5 ) 4 . 6 ( 2 ) C (2 ) 0 . 3 4 7 1 ( 5 ) 0 . 2 1 1 0 ( 4 ) - 0 . 0 6 7 1 ( 5 ) 5 . 0 ( 2 ) C (3 ) 0 . 8 0 9 5 ( 6 ) 0 . 2 5 2 3 ( 4 ) 0 . 6 3 8 6 ( 6 ) 6 . 1 ( 2 ) C (4) 0 . 8 6 7 7 ( 5 ) 0 . 2 3 0 6 ( 4 ) 0 . 3 8 5 0 ( 6 ) 5 . 1 ( 2 ) C (5) 0 . 0 9 1 9 ( 5 ) 0 . 1 7 5 3 ( 4 ) - 0 . 0 3 3 2 ( 7 ) 6 . 3 ( 2 ) C (6) 0 . 2 7 3 6 ( 5 ) 0 . 3 5 7 9 ( 4 ) 0 . 0 8 7 6 ( 7 ) 6 . 0 ( 2 ) C (7) 0 . 8 4 6 5 ( 5 ) 0 . 4 2 4 0 ( 4 ) 0 . 2 4 4 4 ( 7 ) 6 . 3 ( 2 ) C (8) 0 . 6 4 4 6 ( 7 ) 0 . 5 0 2 6 ( 4 ) 0 . 3 3 3 9 ( 7 ) 6 . 9 ( 2 ) C (9) 0 . 5 5 0 6 ( 6 ) 0 . 3 7 1 8 ( 4 ) - 0 . 0 5 3 3 ( 6 ) 5 . 5 ( 2 ) C (10 ) 0 . 6 2 0 0 ( 6 ) 0 . 1 9 5 0 ( 4 ) -0 . 0 5 5 7 ( 6 ) 5 . 5 ( 2 ) C(ll) 0 . 8 0 8 4 ( 8 ) 0 . 3 3 0 3 ( 6 ) 0 . 7 5 3 6 ( 7 ) 8 . 4 ( 3 ) C (12) 0 . 7 3 8 0 ( 7 ) 0 . 1 5 9 3 ( 5 ) 0 . 6 3 7 6 ( 7 ) 7 . 2 ( 3 ) C (13 ) 0 . 8 6 9 9 ( 6 ) 0 . 1 3 2 3 ( 4 ) 0 . 3 6 8 8 ( 7 ) 6 . 7 ( 2 ) C (14 ) 1 . 0 0 8 8 ( 6 ) 0 . 2 9 2 6 ( 4 ) 0 . 4 6 0 8 ( 7 ) 6 . 6 ( 2 ) C (15 ) 0 . 0 6 6 4 ( 7 ) 0 . 0 8 0 8 ( 5 ) - 0 . 1 2 6 3 ( 8 ) 8 . 6 ( 3 ) C ( 1 6 ) 0 . 0 1 5 9 ( 6 ) 0 . 1 7 7 0 ( 5 ) 0 . 0 6 3 7 ( 8 ) 7 . 4 ( 3 ) C (17 ) 0 . 2 3 6 ( 1 ) 0 . 3 9 9 8 ( 5 ) 0 . 2 1 1 ( 1 ) 9 . 7 ( 4 ) C ( 1 8 ) 0 . 1 9 8 4 ( 8 ) 0 . 3 7 7 6 ( 5 ) -0 . 0 4 6 ( 1 ) 9 . 5 ( 3 ) 169 Final atomic coordinates (fractional) and B atom X y z B e q C(19) 0 .4454(6) 0 .2849(3) 0 .4692(6) 5.5(2) C(20A) 0 .374(2) 0 .233(1) 0 .542(1) 5 .0(4) C(20B) 0 .343(3) 0 .245(2) 0 .486(3) 5(1) C(21A) 0 .2807(9) 0 .1411(6) 0 .448(1) 4.4(3) C(21B) 0 .347(2) 0 .150(2) 0 .509(2) 4 .4(7) C(22) 0 .3599(7) 0 .0981(4) 0 .3680(7) 6 .4(2) C(23) 0 .3032(6) 0 .0664(3) 0 .2232(8) 6 .3(2) C(24) 0 .3787(6) 0 .0521(3) 0 .1390(6) 5.1(2) C(25) 0 .5190(6) 0 .0646(3) 0 .2061(6) 5.2(2) C(26) 0 .2247(7) 0 .0881(5) 0 .5184(8) 7.5(3) 170 Bond lengths (A) w i t h estimated standard d e v i a t i o n s i n parentheses. atom atom distance atom atom dis t a n c e Z r ( l ) P d ) 2 .782(2) S i ( 2 ) C(10) 1 .857(5) Z r ( l ) P(2) 2.850(2) C(3) C d l ) 1 .503(9) Z r ( l ) N(l ) 2.190(4) C(3) C(12) 1 .54(1) Z r ( l ) C(19) 2.305(5) C(4) C(13) 1 .524(8) Z r ( l ) C(22) 2 .573(5) C(4) C(14) 1 .519(8) Z r ( l ) C(23) 2.500(5) C(5) C(15) 1 .49(1) Z r ( l ) C(24) 2.358(4) C(5) C(16) 1 .515(9) Z r ( l ) C(25) 2.345(5) C(6) C(17) 1 .52(1) P d ) C d ) 1.831(5) C(6) C(18) 1 .53(1) P d ) C(3) 1.864(6) C(19) C(20A) 1 .54(2) P d ) C(4) 1.860(5) C(19) C(20B) 1 .28(4) P(2) C(2) 1.823(5) C(20A) C(21A) 1 .51(2) P(2) C(5) 1.860(5) C(20B) C(21B) 1 .59(4) P(2) C(6) 1.864(6) C(21A) C(22) 1 .52(1) S i d ) N(l) 1.724(4) C(21A) C(26) 1 .44(1) S i d ) C d ) 1.879(6) C(21B) C(22) 1 .57(2) S i d ) C(7) 1.881(6) C(21B) C(26) 1 .57(2) S i d ) C(8) 1.865(6) C(22) C(23) 1 .376(9) Si (2 ) N( l ) 1.730(4) C(23) C(24) 1 .405(8) Si (2) C(2) 1.895(5) C(24) C(25) 1 .437(8) Si (2) C(9) 1.873(6) 171 Bond a n g l e s (deg) w i t h p a r e n t h e s e s . atom atom atom a n g l e P d ) z r ( l ) P(2) 149 .26(4) P d ) Zr (1) N ( l ) 76 .1(1) P d ) Z r ( l ) C(19) 83 .4(2) P d ) Zr (1) C(22) 104 .4(2) P d ) Z r ( l ) C(23) 128 • 4(2) P d ) Z r d ) C(24) 119 .6(1) P d ) Zr (1) C(25) 84 .5(1) P(2) Z r d ) N ( l ) 74 .6(1) P(2) Z r d ) C(19) 98 .5(2) P(2) Z r ( l ) C(22) 104 .8(2) P(2) Z r ( l ) C(23) 82 .3(2) P(2) Z r ( l ) C(24) 84 .1(1) P(2) Zr {1) C(25) 114 .5(1) N ( l ) Z r ( l ) C(19) 111 .1(2) N( 1) Zr (1) C(22) 179 .0(2) N ( l ) Z r ( l ) C(23) 148 .7(2) N ( l ) Zr (1) C(24) 121 .5(2) N ( l ) Z r ( l ) C(25) 110 .5(2) C(19) Z r ( l ) C(22) 68 .2(2) C(19) Z r ( l ) C(23) 92 .9(2) C(19) Zr (1) C(24) 125 .8(2) C(19) Zr (1) C(25) 132 .2(2) C(22) Zr (1) C(23) 31 .4(2) C(22) Z r ( l ) C(24) 59 .1(2) C(22) Zr (1) C(25) 70 .4(2) C(23) Z r ( l ) C(24) 33 .5(2) e s t i m a t e d s t a n d a r d d e v i a t i o n s i n atom atom atom angle C(23) Z r ( l ) C(25) 60 .6(2) C(24) Z r ( l ) C(25) 35 .6(2) Z r ( l ) P d ) C ( l ) 97 .4(2) Z r ( l ) P d ) C(3) 127 .1(2) Z r ( l ) P d ) C(4) 116 .7(2) C ( l ) P d ) C(3) 107 .4(3) C d ) P d ) C(4) 102 .4(2) C(3) P d ) C( 4 ) 102 .7(3) Z r ( l ) P(2) C(2) 93 .0(2) Z r ( l ) P(2) C(5) 132 .6(2) Z r ( l ) P(2) C(6) 115 .4(2) C(2) P(2) C(5) 105 .2(3) C(2) P(2) C(6) 102 .9(3) C(5) P(2) C(6) 102 .9(3) N ( l ) S i d ) C ( l ) 107 .6(2) N ( l ) S i d ) C(7) 114 .2(2) N ( l ) S i d ) C(8) 112 .9(2) C d ) S i ( l ) C(7) 107 .9(3) C ( l ) S i ( l ) C(8) 106 .6(3) C(7) S i ( l ) C(8) 107 .3(3) N ( l ) Si (2) C(2) 107 .2(2) N ( l ) Si(2) C(9) 115 .4(2) N ( l ) Si(2) C(10) 111 .6(2) C(2) Si(2) C(9) 106 .6(3) C(2) S i d ) C(10) 107 .7(2) C(9) S i d ) C(10) 108 .0(3) 172 atom atom atom angle 2 r ( l ) N d ) S i d ) 122.3(2) Z r ( l ) N d ) Si(2) 117.1(2) S i d ) N d ) Si(2) 120.5(2) P d ) C d ) S i d ) 109.5(2) P(2) C(2) Si(2) 110.4(3) P d ) C(3) C d l ) 113.9(5) P d ) C(3) C(12) 110.2(4) C d l ) C(3) C(12) 112.6(6) P d ) C(4) C(13) 112.3(4) P d ) C(4) C(14) 115.9(4) Cd3) C(4) C(14) 109.4(5) P(2) C(5) C(15) 113.6(5) P(2) C(5) C(16) 112.7(4) Cd5) C(5) C(16) 111.1(5) P(2) C(6) C(17) 112.4(5) P(2) C(6) C(18) 115.5(5) Cd7) C(6) C(18) 112.5(6) Z r d ) Cd9) C(20A) 120.8(6) Z r d ) Cd9) C(20B) 115(1) Cd9) C(20A) C(21A) 114.2(9) C(19) C(20B) C(21B) 114(2) C(20A) C(21A) C(22) 104.0(8) C(20A) C(21A) C(26) 114.7(8) C(22) C(21A) C(26) 116.3(7) C(20B) C(21B) C(22) 100(2) C(20B) C(21B) C(26) 121(2) atom atom atom angle C(22) C(21B) C(26) 107(1) Z r ( l ) C(22) C(21A) 113 .2(4) Z r ( l ) C(22) C(21B) 112 .1(8) Z r ( l ) C(22) C(23) 71 .4(3) C(21A) C(22) C(23) 117 .8(7) C(21B) C(22) C(23) 147(1) Z r ( l ) C(23) C(22) 77 .2(3) Z r ( l ) C(23) C(24) 67 .7(3) C(22) C(23) C(24) 122 .5(6) Z r ( l ) C(24) C(23) 78 .8(3) Z r ( l ) C(24) C(25) 71 .7(3) C(23) C(24) C(25) 118 .9(5) Z r ( l ) C(25) C(24) 72 .7(3) 173 A - 5 Xray Crystallographic Analysis of {ZrCl [N(SiMe 2 CH 2 PPr l 2)2 ] }2 ( l t-'n 2 :T l 2 -N2) 174 A. Crystal Data Empirical Formula Formula Weight Crystal Color, Habit Crystal Dimensions (mm) Crystal System No. Reflections Used for Unit C e l l Determination (26 range) Omega Scan Peak Width a t H a l f - h e i g h t L a t t i c e P a r a m e t e r s : Space Group Z va lue D c a l c F o o o "(CuKa) D i f f T a c t o m e t e r R a d i a t i o n Temperature T a k e - o f f A n g l e D e t e c t o r A p e r t u r e C 4 3 H 9 6 C l 2 N 4 P 4 S i 4 Z r 2 1 1 5 8 . 8 4 dark, plate 0 . 0 5 0 X 0 . 2 0 0 X 0 . 2 0 0 monoclinic 2 5 ( 6 5 . 0 - 8 7 . 3 ° ) 0 . 3 8 B a - 14.103 (3)A b - 16.233 (3)A c - 14.678 (3)A 0 - 114.24 ( l ) e V - 3064 (1)A 3 P2 1/c (#14) 2 1.256 g/cm3 1224 56.60 cm"1 Intensity Measurements Rigaku AFC6S CuKa (X - 1.54178 A) 21°C 6.0e 6.0 mm horizontal 6.0 mm v e r t i c a l 175 Crystal to Detector Distance Scan Type Scan Rate Scan Width 26 max No. of Reflections Measured Corrections 285 mm u>-2 6 16.0*/min (in omega) (8 rescans) (1.00 + 0.20 tan6)° 155.5° T o t a l : 7046 Unique: 6783 (* i n t - .035) Lorentz-polarization Absorption (trans, f a c t o r s : 0.43 - 1.00) Secondary Extinction ( c o e f f i c i e n t : 0.10352E-05) C. Structure Solution and Refinement Structure Solution Refinement Function Minimized Least-squares Weights p-factor Anomalous Dispersion No. Observations (l>3.00o(I)) No. Variables Reflection/Parameter Ratio Residuals: R; R w Goodness of F i t Indicator Max S h i f t / E r r o r i n F i n a l Cycle Maximum Peak i n Final D i f f . Map Minimum Peak in Final D i f f . Map Patterson Method Full-matrix least-squares I w (|Fo| - |F c | ) 2 2 2 2 4FO / a (FO^) 0.03 A l l non-hydrogen atoms 4158 303 13.72 0.040; 0.049 1.71 0.01 0.47 e~/k\ -0.50 e"/A3 176 Final atomic coordinates and B e g (A 2) atom y z occ. Z r d ) 0 . 3 6 8 8 1 ( 3 ) 0 . 5 2 5 4 3 ( 2 ) 0 . 5 3 3 5 2 ( 3 ) 3 . 7 4 ( 1 ) C l d ) 0 . 4 0 9 1 ( 1 ) 0 . 6 5 8 4 1 ( 8 ) 0 . 6 2 4 6 ( 1 ) 6 . 8 6 ( 6 ) P d ) 0 . 2 4 0 6 ( 1 ) 0 . 6 0 5 1 7 ( 8 ) 0 . 3 7 1 1 ( 1 ) 5.11(5) P ( 2 ) 0 . 4 1 5 7 ( 1 ) 0 . 4 5 5 3 4 ( 8 ) 0 . 7 1 4 7 ( 1 ) 5 .42 (6 ) S i d ) 0 . 1 9 5 0 ( 1 ) 0 . 4 1 9 7 ( 1 ) 0 . 3 5 4 7 ( 1 ) 6 . 1 7 ( 6 ) S i (2 ) 0.3128(5) 0 . 3 2 9 9 ( 4 ) 0.5400(5) 5 . 7 ( 2 ) 0 . 565 S i ( 2 A ) 0 . 2 7 8 5 ( 6 ) 0 .3426(5 ) 0 . 5 5 2 3 ( 7 ) 5 . 8 ( 3 ) 0 . 435 N d ) 0 . 2 7 8 2 ( 3 ) 0 . 4 2 4 5 ( 2 ) 0 . 4 7 8 5 ( 3 ) 4 . 3 ( 1 ) N U ) 0 . 4 7 2 2 ( 3 ) 0 . 5 2 2 5 ( 2 ) 0 . 4 4 9 4 ( 3 ) 4 . 7 ( 1 ) C d ) 0 . 1 9 9 8 ( 4 ) 0 . 5 1 7 8 ( 3 ) 0 . 2 8 8 6 ( 4 ) 6 . 1 ( 2 ) C ( 2 ) 0 . 3 9 1 8 ( 6 ) 0 . 3 4 9 6 ( 4 ) 0 . 6 7 7 7 ( 4 ) 8 . 0 ( 3 ) C( 3) 0 . 2 8 8 5 ( 6 ) 0 . 6 8 1 8 ( 4 ) 0 . 3 0 7 3 ( 5 ) 7 . 8 ( 3 ) C(4 ) 0 . 1 2 4 8 ( 6 ) 0 . 6 4 7 4 ( 6 ) 0 . 3 8 1 4 ( 6 ) 9 . 8 ( 4 ) C ( 5 ) 0 . 5 4 8 7 ( 6 ) 0 . 4 5 6 9 ( 4 ) 0 . 8 1 6 8 ( 5 ) 8 . 7 ( 3 ) C (6 ) 0 . 3 2 7 7 ( 8 ) 0 . 4 8 8 5 ( 7 ) 0 . 7 7 0 1 ( 7 ) 1 2 . 1 ( 6 ) C ( 7 ) 0 . 0 5 7 8 ( 6 ) 0 . 4 0 0 9 ( 7 ) 0 . 3 3 5 6 ( 7 ) 1 6 . 1 ( 6 ) C ( 8 ) 0 . 2 2 8 ( 1 ) 0 . 3 3 9 7 ( 5 ) 0 . 2 8 4 0 ( 6 ) 1 6 . 1 ( 7 ) C ( 9 ) 0 . 1 9 7 ( 1 ) 0 . 2 6 5 ( 1 ) 0 . 5 2 6 ( 1 ) 1 0 . 4 ( 9 ) 0 . 565 C O A ) 0 . 1 5 4 ( 1 ) 0 . 3 2 9 ( 1 ) 0 . 5 6 9 ( 1 ) 1 0 ( 1 ) 0 . 435 C d O ) 0 . 3 9 4 ( 1 ) 0 . 2 6 3 6 ( 8 ) 0 . 4 9 8 ( 1 ) 9 . 5 ( 7 ) 0 . 565 C d O A ) 0 . 3 0 1 ( 2 ) 0 . 2 4 3 8 ( 8 ) 0 . 5 0 6 ( 1 ) 9 . 8 ( 9 ) 0 . 435 C d l ) 0 . 3 2 9 0 ( 8 ) 0.7577(5) 0 . 3 6 8 5 ( 6 ) 1 2 . 9 ( 5 ) C ( 1 2 ) 0 . 2 1 9 1 ( 7 ) 0.7014(5) 0.2013(5 ) 1 1 . 8 ( 5 ) C ( 1 3 ) 0 . 0 2 7 2 ( 7 ) 0 . 6 5 1 2 ( 7 ) 0 . 2 8 6 3 ( 8 ) 1 5 . 0 ( 7 ) 177 F i n a l a tomic c o o r d i n a t e s and B (A ) ( c o n t . ) atom X y z o c c . C(14) 0. 109(1) 0. 618(2) 0. 468(2) 15(1) 0.540 C(14A) 0. 134(2) 0. 709(2) 0. 445(2) 15(1) 0.460 C(15) 0. 5846(7) 0. 5419(5) 0. 8524(6) 11 .6 (4 ) C(16) 0. 5696(8) 0. 3978(5) 0. 9015(6) 12 .8 (5 ) C(17) 0. 239(1) 0. 5264(9) 0. 708(1) 21(1) C(18) 0. 302(1) 0. 429(1) 0. 832(1) 21(1) C(19) - 0 . 040(4) 0. 056(2) 0. 522(2) 28(2) C(20) 0. 033(3) 0. 074(2) 0. 507(2) 27(2) C(21) 0. 075(3) 0. 022(4 ) 0. 476(2) 25(2) C(22) - 0 . 111(2) 0. 076(2) 0. 548(2 ) 17(1) 0. 500 178 Bond lengths (A) with estimated standard deviations.* atom atom di stance atom atom distance Z r ( l ) C l ( l ) 2.493(1) Si(2) C(10) 1.86(2) Z r ( l ) P d ) 2.764(1) Si(2A) N(l) 1.71(1) Z r ( l ) P(2) 2.772(1) Si(2A) C(2) 1.88(1) Z r ( l ) N(l) 2.175(3) Si(2A) C(9A) 1.88(2) Z r ( l ) N(2) 2.024(4) Si(2A) C(10A) 1.62(2) Z r ( l ) N d ) ' 2.027(4) N(2) N(2) ' 1.548(7) P d ) C(l) 1.800(5) C(3) C ( l l ) 1.49(1) P d ) CO) 1.844(6) C(3) C(12) 1.494(9) P d ) C(4) 1.834(7) C(4) C(13) 1.51(1) P(2) C d ) 1.790(6) C(4) C(14) 1.45(2) P(2) C(5) 1.859(7) C(4 ) C(14A) 1.34(2) P d ) C(6) 1.621(8) C(5) C(15) 1.49(1) S i d ) N(l ) 1.713(4 ) C(5) C(16) 1.50(1 ) S i d ) C d ) 1.880(5) C(6) C(17) 1.36(1 ) S i ( l ) C(7) 1.863(9) C(6) C(18) 1.47(1 ) S i d ) C(8) 1.840(8) C(19) C(20) 1.18(6) S i d ) N d ) 1.746(8) C(19) C(21)" 1.36(5) S i d ) C d ) 1.89(1) C(19) C(22) 1.26(6 ) S i d ) C(9) 1.88(1) C(20) C(21) 1.23(7) * Here and elsewhere the symbols ' and " r e f e r to symmetry operations: l - x d - y , l - z ; and -x»-y_#l-z; respectively. 179 Bond angles (deg) with estimated standard deviations. atom atom atom .angle atom atom atom angle C l ( l ) Z r ( l ) P d ) 86 .39(5) C(5) P(2) C(6) 106 .5(4) CHI) Z r ( l ) P(2) 84 .27(5) N(l) S i ( l ) C d ) 110 .8(2) C l ( l ) Z r ( l ) N(l) 139 .3(1) N(l) •Sid) C(7) 112 .6(3) C l ( l ) Z r ( l ) N d ) 112 .3(1) N(l) S i d ) C(8) 114 .1(4) C l ( l ) Z r d ) N d ) ' 111 .9(1) C d ) S i ( l ) C(7) 108 .4(3) P d ) Z r ( l ) P d ) 141 .00(5) C(l) S i ( l ) C(8) 103 .6(3) P(l) Z r d ) N(l) 81 .9(1) C(7) S i d ) C(8) 106 .8(6) P d ) Z r ( l ) N d ) 84 .7(1) N(l) Si(2) C(2) 108 .7(4) P d ) Z r ( l ) N d ) ' 129 .6(1) N(l ) S i d ) C(9) 113 .0(6) P d ) Zr (1) N(l ) 80 .9(1) N(l) S i d ) CdO) 114 .6(5) P d ) Z r ( l ) N(2 ) 133 .7(1) C(2) Si(2) C(9) 108 .4(6) P d ) Zr (1) N d ) ' 88 .8(1) C(2) S i d ) CdO) 106 .4(6) N ( l ) Z r ( l ) Nd) 105 .2(1) C(9 ) Si(2) C(10) 105 .4(8) N d ) Z r ( l ) N( 2 ) ' 105 .5(2) N(l) Si(2A) C(2) 110 .6(5) Nd) Zr (1) Nd) ' 44 .9(2) N(l ) Si(2A) C(9A) 114 .4(8) Z r ( l ) P d ) C d ) 97 .5(2) N(l) Si(2A) CdOA) 113 .5(8) Z r d ) P d ) C(3) 116 .7(2) C(2) Si(2A) C(9A) 110 .0(7) Z r d ) P d ) C(4) 119 .2(3) C(2) Si(2A) C(lOA) 102 .2(8) C d ) P d ) C(3) 105 .8(3) C(9A) Si(2A) C(10A) 105(1) C d ) P d ) C(4) 107 .5(4) Z r ( l ) N(l) S i ( l ) 120 .2(2) C(3) P d ) C(4) 108 .3(4) Z r ( l ) N(l) S i d ) 117 .7(3) Z r ( l ) P d ) C(2) 99 .7(2) Z r ( l ) N(l ) Si(2A) 122 .3(4) Z r d ) P d ) C(5) 117 .5(2) S i d ) N(l ) S i d ) 115 .6(3) Z r ( l ) P d ) C(6) 117 .2(3) S i d ) N(l ) Si(2A) 117 .2(4) C d ) P d ) C(5) 105 .1(3) Z r ( l ) N( 2 ) Z r ( l ) ' 135 .1(2) C d ) P(2) C(6) 109 .9(4) Z r ( l ) N(2) N(2) ' 67 .6(2) 180 Bond angles (deg) with atom atom atom angle Z r ( l ) N d ) ' N(2) 67 .5(2) P(l) C(l) S i d ) 112 .9(3) P(2) C(2) Si ( 2 ) 116 .1 (4) P(2) C(2) Si ( 2 A ) 109 .4(4) P d ) C(3) C ( l l ) 112.4(5) P d ) C(3) C(12) 116 .9(5) C d l ) C(3) C(12) 112.1(6) P d ) C(4) C(13) 116.4(6) P d ) C(4) C(14) 114.2(8) P d ) C(4) C(14A) 120(1) C(13) C(4) C(14) 115(1) Cd3) C(4) C(14A) 114(1) P(2) C(5) C(15) 112.4(5) P(2 ) C(5) C(16) 116.6(6) C(15) C(5) C(16) 112.2(6) P(2) C(6) C(17) 116.2(8) P(2) C(6) C(18) 117.9(8) C(17). C(6) C(18) 109(1) C(20) C(19) C(21)" 125(4) C(20) C(19) C(22) 149(6) C d l ) " C(19) C(22) 85(6) C(19) C(20) C(21) 120(3) C(19) n C(21) C(20) 114(4) estimated standard deviations, atom atom atom angle 181 W H Y ? 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

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"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0059690/manifest

Comment

Related Items