Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Dinuclear dinitrogen and mononuclear paramagnetic complexes of zirconium Mylvaganam, Murugesapillai 1994

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

Item Metadata


831-ubc_1994-95373x.pdf [ 5.15MB ]
JSON: 831-1.0059551.json
JSON-LD: 831-1.0059551-ld.json
RDF/XML (Pretty): 831-1.0059551-rdf.xml
RDF/JSON: 831-1.0059551-rdf.json
Turtle: 831-1.0059551-turtle.txt
N-Triples: 831-1.0059551-rdf-ntriples.txt
Original Record: 831-1.0059551-source.json
Full Text

Full Text

DINUCLEAR DINITROGEN AND MONONUCLEAR PARAMAGNETICCOMPLEXES OF ZIRCONIUMbyMURUGESAPILLAI MYLVAGANAMB.Sc. (Hons.), University of Jaffna, Sri Lanka, 1985M. Sc., University of British Columbia, 1989A THESIS SUBMITfED IN PARTIAL FULFILLMENT OFTHE REQUIREMENTS FOR THE DEGREE OFDOCTOR OF PHILOSOPHYinTHE FACULTY OF GRADUATE STUDIES(Department of Chemistry)We accept this thesis as conformingto the required standardTHE UNIVERSITY OF BRITISH COLUMBIAJune 1994© M. Mylvaganam, 1994In presenting this thesis in partial fulfilment of the requirements for an advanceddegree at the University of British Columbia, I agree that the Library shall make itfreely available for reference and study. I further agree that permission for extensivecopying of this thesis for scholarly purposes may be granted by the head of mydepartment or by his or her representatives. It is understood that copying orpublication of this thesis for financial gain shall not be allowed without my writtenpermission.(Signature)___________________________Department of CYt RI? S 7The University of British ColumbiaVancouver, CanadaDate__DE-6 (2/88)IiABSTRACTThe basic focus of this work is the use of the tridentate, mixed donor ligand,[N(SiMe2CHPR)1-abbreviated as PNP, to generate zirconium compounds that containdinitrogen ligands in unusual bonding modes, or to allow the stabilization of the very rarezirconium(III) oxidation state. The approach that is used is to combine synthetic methods inorganometallic chemistry with semi-empirical molecular orbital calculations as a means ofdesigning new complexes and to rationalize bonding.Reduction of ZrCpCl2[N(SiMe2CH2PPr2)2] 2.6, or ZrCl2(OAr*)[N(SiMe2CH2PPr2)] (Ar* = (C6H3Me-2,6)) 2.11, under a dinitrogen atmosphere gave complexes{ [(PPrCHSiMe2)2N]ZrCp } 2(J.t-N2) 2.9, and ([(PPr2CH2SiMe2)2N] Zr(OAr*) } 2(I.t-N2)2.12, respectively. The X-ray structure determination of these complexes confirmed thepresence of an end-on bridging (‘-i1:1-N2) clinitrogen ligand in 2.9 whereas, 2.12 has aside-on bridging (.t-fl2:1-N2) dinitrogen ligand. The observed nitrogen-nitrogen bonddistances in 2.9 was 1.301 (3) A and in 2.12, it was 1.528 (7) A.The resonance Raman spectra of the solid and the solution state samples of 2.9 and 2.2showed isotope sensitive peaks around 1200 and 730 cm1 respectively, assigned to thenitrogen-nitrogen stretching of these complexes. Also, the fact that these resonance Ramanfeatures are the same in the solid and in the solution states strongly suggest that the mode ofbonding of the dinitrogen ligand are same for the respective complexes.The semi-empirical molecular orbital studies performed on the end-on derivative 2.9and on the side-on derivatives 2.2 and 2.12 show that the lu-acceptor interactions involving the7t*orbitals of the dinitrogen ligand are significantly different. In the end-on case theseinteractions give rise to two it-MOs, whereas the side-on cases give rise to one ö-MO and onelu-MO where the lu-MO was found to be much lower in energy than the ö-MO. Mullikenpopulation and Wiberg indices were calculated to show that in the side-on mode there is greaterelectron donation into the dinitrogen ligand than in the end-on cases and also the side-on boundmN2 ligand has a very weak nitrogen-nitrogen bond, which is corroborated by the bond lengthparameters and the resonance Raman data. An analysis of the frontier orbitals of the fragment[(H3P)2(H2N ZrXI, where X = Cl, Cp or OH, shows that the metal-ligand d-p interactionsinfluence the mode of dinitrogen coordination, i.e., end-on vs. side-on.The paramagnetic zirconium(ffl) complex, Zr(15-CH)Cl[N(SiMe2HPPr]4.1,and the corresponding hafnium(III) derivative,Hf(ri-)Cl[N(SiMeP 4.1,were synthesized by the reduction of the respective dichioro precursors M(’fl5-CH)C12-[N(SiMe2-CH2PPr)2j, where M = Zr or Hf. The results of this study show that complex 4.1is a viable precursor to the synthesis of a variety of derivatives such as the first stable examplesof alkyl, aryl and borohydride complexes of zirconium(ffl). X-ray structure elucidation hasbeen carried out for an alkyl 4.8, phenyl 4.4 and borohydride complex, 4.17. Spectroscopicstudies (JR and ESR) indicate that in the case of the alkyl derivatives a weak agostic typeinteraction may be present between one of the x-hydrogens and the metal.Hydrogenolysis of certain alkyl complexes shows a clean conversion to themononuclear zirconium(Ill) hydride complex 4.14. This hydride complex has been shown toundergo an insertion reaction with ethylene. The hydride complex also undergoes hydrogenexchange reactions with H—C(sp2)and H—C(sp3)bonds, presumably by way of s-bondmetathesis. Complexes 4.1 and the zirconium(III) borohydride complex undergo reversibledisproportionation reactions under a CO atmosphere to give zirconium(JV) and zirconium(ll)derivatives. It was also shown that CH3N reacts in a similar fashion with 4.1, however thereversibility of this reaction was not established. In the case of the borohydride complex thereaction with CO proceeds further to give a complex containing a “formyl-ylid” type ligand.ivTABLE OF CONTENTSABSTRACT.iiTABLE OF CONTENTS ivLIST OF TABLES ivLIST OF FIGURES xiiLIST OF ABBREVIATIONS xviiACKNOWLEDGEMENTS xxDEDICATION xxiChapter 1Introduction to Transition Metal Dinitrogen Chemistry1.1 General 11.2 Biological Nitrogen Fixation 21.3 Other Methods of Nitrogen Fixation 41.4 Thermodynamic and Kinetic Factors 61.5 Dinitrogen Complexes 81.5.1 Mononuclear End-On Complexes 91.5.2 Dinuclear End-On complexes 101.5.3 Protonation of Terminally Bonded Dinitrogen Complexes 111.5.4 Polynuclear Dinitrogen Complexes 141.5.6 Mononuclear Side-On Complexes 151.5.7 Dinuclear Side-On Complexes 161.6 Summary 171.7 References 18VChapter 2Dinitrogen Complexes of Zirconium2.1 General 222.2 Synthesis of a Dinitrogen Complex Containing PNP and Cp Ligands 242.2.1 Synthesis of Precursors of the Type ZrCpX2[N(SiMe2CH2PR)] 252.2.2 Synthesis of Precursors of the Type ZrCpX2[N(SiMe2CH2PR 282.3 Attempted Synthesis of a Dinitrogen Complex Containing PNP and Allyl ligands 332.4 Synthesis of a dinitrogen Complex Containing PNP and Aryl Oxide Ligands 352.4.1 Synthesis of Zr(OAr*)C12[N(SiMe2CH2PPri2] 352.4.2 Reduction of Zr(OAr*)Cl2[N(SiMeCH2PPri) 362.5 Synthesis of{[(PrPCHSiMe2)2N]Zr(OBut))2(JIfl: I 452.6 Spectroscopy as a Diagnostic Tool for End-On and Side-On Dinitrogen Complexes 462.6.1 15N NMR Spectroscopy 472.6.2 Resonance Raman Spectroscopy 512.7 Bonding Considerations 572.7.1 General 572.7.2 Bonding in {[(Pr2PCH2SiMe2)2N]Zr(rI5-)}(I-N ) 592.7.3 Bonding in Side-On Complexes 632.7.4 Bonding in {[(Pr2PCH2SiMeN]ZrCl}2(.t-N 642.7.5 Bonding in { [(Pr2PCH2SiMe2)2N] Zr(OAr*) } (p.-N) 692.7.6 Factors Influencing the Side-On Mode of Coordination 772.7.7 Bonding in the Samarium and Lithium Dinitrogen Complexes 822.8 Reactions 842.8.1 Conversion of the Side-On Complex to the End-On Complex 842.8.2 Reaction of Side-On Complexes With LiBEt4 862.8.3 Protonation Reactions 902.8.4 Reactions With Halo Alkanes 93v2.8.4a Reactions of the End-On Complex With BzBr and CH3I 932.8.4b Reactivity With Dihaloalkanes 962.9 Conclusions 1002.10 Experimental 1022.lOa General Procedures 1022.lOb Synthesis of Precursors 1052.lOb.1 ZrBr3{N(SiMe2CH2PPr2)2], 2.4 1052.lOb.2 r(5-H) l[N(SiMeCPPr],2.6 1052.lOb.3 Zr(11-C)Br[N(SiMeHPP 2.7 1062lOb.4 Zr(11-) 1[N(SiMeCPMe,2.8 1062.lOb.5Zr(ri-H) l[N(SiMeHPr] 2.10 1062. lOb.6 Zr(OAr*)C12[N(SiMe2CH2PPrj2)2], 2.11 1072. lOb.7 Zr(OBut)C12[N(SiMe2CH2PPr) ], 2.11 1082.lOb.8 Zr(OBut)C1[N(SiMeCHPMe1,2.13 1082. lOb .9 Zr(OCHPh2)C1[N(SiMe2CHPPr)j,2.16 1082.lOb.10 Zr(NPh) l[N SiMeHP r],2.17 1092.lOb.11Zr(CHSiMe3)Cl[N(SiMeCPrj 2.18 1092.lOc Synthesis of Dinitrogen Complexes 1102.lOc.1 {[(Pr2PCHSiMe)N]ZrCl} 2(1.t-TI:T-N2), 2.2 1102.lOc.2 {[(PrPCHSiMeN]ZrCl}(p.-i:ri-’5),2.2 1102.lOc.3 {[(PrPCHSiMeN]Zr(115-H) .t1:i1-N2), 2.6 1102.lOc.4 { [(Pr2PCH2SiMe)N]Zr(rI-CsH .trI’ :r1-5N2), 2.6 1112.lOc.5 {[(Pr2PCHSiMe]Zr(OAr*) )2(ji-i1:r1-N,2.12 1112.lOc.6 ([(PriPCHSi eN]Zr(OAr*)} 2(.t-11:1-15N2), 2.12 1132.lOc.7 {[(PriPCHSiMe)N]Zr(OBut) }2(.L-T:rI-N2), 2.14 1132.lOc.8 ([(PrPCHSiMeN]ZrBr}(i.-rI:TI-N),2.19 1132.lOc.9 ([(PrPCHSiMeNJZrBr}2(.t-T1:1-’5),2.19 114VII2.lOc.10([(PrPCHSiMe)2NJClBr(.L-1:1-N2),2.20. 1142.lOc.11 2.20 1142.lOd Reactions Involving Dinitrogen Complexes 1152.lOd.1{[(PrPCH2SiMeN]ZrCl}2(.L-r:fl-N )and NaCp.DME 1152.lOd.2 Protonation Reactions 1152.lOd.3{[(PriCHSiMe)N]ZrX}(j.ifl: where X = Cl, Br orOAr*, and LiBEt4 1162.lOd.4{[(PrPCH2SiMeN]ZrX}2(J.t-N2) Where X = Cp, Cl or OAr*, andAlkyl Halides 1172.lOe Molecular Orbital Calculations 1172.11 References 119Chapter 3Introduction to Zirconium(HI) Chemistry3.1 General 1243.2 Coordination Complexes of zirconium(III) 1253.3 Organometallic Complexes of zirconium(1II) 1263.3.1 Dinuclear organometallic complexes of zirconium(ffl) 1263.3.2 Mononuclear organometallic complexes of zirconium(III) 1293.4 Summary 1333.5 References 133Chapter 4Zirconium(III) Complexes: Synthesis and Reactivity41 General 1364.2 Synthesis of Zirconium(llI) and Hafnium(llI) Complexes 1364.3 Synthesis of Zirconium(llI) Hydrocarbyl Complexes 139vifi4.4 Oxidation of Zirconium(ffl) Complexes.1514.5 Hydride Complexes of Zirconium(III) 1534.5.1 General 1534.5.3 Reactions ofZr(5-CH)H[N(SiMe2CHPPr] 1614.5.4 Synthesis of Zirconium(ffl) Borohydride Complex 1634.6 Disproportionation Reactions of Mononuclear Zirconium(I1T) Complexes 1694.7 Further Reaction with CO: Formation of Zirconium Formyl Complex 1774.8 Conclusions 1854.9 Experimental Procedures 1864.9a General 1864.9b Synthesis of Complexes 1874.9b.1Zr(rI5-H)Cl[N SiMe2HPPr2)],4.1 1874.9b.2 f(1-C) l[N(SiMePr4.2 1884.9b.3 Hf(11.)Cl[N(SiMeHPPr],4.3 1884.9b.4Zr(q5-)6[N(SiMeCP rJ 4.4 1884.9b.5Zr(1-C)Me[N SiMe2HPr],4.5 1894.9b.6 Zr(rt-H) HC3[N(SiMeP rj 4.6 1894.9b.7Zr(5-)CH2CD[N(SiMe2CH2PPr)],4.6-d1 1904.9b.8 Zr(rI )CDH[N(SiMeHP r],4.6-cLi 1904.9b.9 r(ri-)C[N(SiMePP1 4.6-d5 1904.9b.1OZr()CHPhN(SiMer],4.7 1914.9b.11 ZrC1(CH2h)[N(SiMeCHPPr4.9 1914.9b.12 r(15-H)Si e3[N(SiMerJ,4.8 1914.9b.13Zr(1-C)OPh[N(SiMeCHPPr] 4.10 1924.9b.14 Zr(rj-)NPh[ (SiMer,4.11 1924.9b.15 Zr(ri-CH)PPh2[N(SiMeCHP r] 4.12 1924.9b.16 Zr(i-)H[N(SiMePPr,4.14 192ix4.9b.17 [ZrHCp[N(SiMe2CHPPr)J{ Cpr[N(MeSidH2HPPr1)(SiMePPr’)]}(i.t-H),4.15 1934.9b.18 Zr(T5-C)BH4[N(SiMeCPPr),4.17 1944.9b.19 Zr(T-)(CCPh)[N(SiMeCH2PPr2)2] 1944.9b.20 Zr(11 )(CCPh)C1[N(SiMe2CHP r],4.16 1954.9b.21 Hf(rl5-)C1(BH4[N(SiMeHPr4.21 1954.9c Disproportination Reactions 1954.9c.1Zr(r1-CSH5)(CO)2[N(SiMe2CH2PPr’2)21, 4.18 1964.9c.2 Zr(T-)(B[N(SiMePPr],4.19 1964.9c.3 Hf(TI-C)( O)[N(SiMeCH2PPrj 4.20 1964.9c.4Zr(fl)(CH3CN)[N(SiMeC 2PPr21,4.22 1964.9c.5Cp(BH)Zr(’HO)[N(SiMeHPPr(SiMe’.B] 4.23.1974.9d Oxidation Reactions 1974.9d.1Zr(rI-C)C1(SPh)[N(SiMe2HP r],4.13 1984.9d.2 Zr(T5-H)C1Et[[N(SiMe2CH2PPr2)2] 1984.9d.3 Zr(rI-C)C1(CH2Ph)[N(SiMeCHPPr)] 1984.9d.4 Zr(fl5-C5H5)(CH2SiMe3)(SPh){N( iMe2CHP r2)] 1984.9d.5 Zr(i5-CH)(CH2CH)(SPh)[N(SiMe2CHPPr )1 1984.10 Refferences 199Chapter 5On-Going and Future Prospects 202Appendix 210xList of TablesTable 2.1 Selected bond distances of complex 2.7 27Table 2.2 Selected bond angles of complex 2.7 27Table 2.3 Selected bond distances of complex 2.9 30Table 2.4 Selected bond angles of the complex 2.9 31Table 2.5 Selected bond distances of complex 2.12a 39Table 2.6 Compilation of nitrogen-nitrogen bond lengths for some selectedcompounds 39Table 2.7 Selected bond angles of complex 2.12a 41Table 2.8 Compilation of 31P{ ‘H} NMR chemical shifts for some zirconiumprecursors and the dinitrogen complexes derived from them. In all thesecomplexes PNP corresponds to [N(SiMe2CH2PPr)2] 46Table 2.9 Compilation of some 15N chemical shifts for some selected compounds 48Table 2.10 Compilation of some nitrogen-nitrogen bond lengths and stretchingfrequencies associated with them 57Table 2.11 Some orbital parameters of le and 2e MOs of model complex A 61Table 2.12 Bond indices and population analysis of model complex A 62Table 2.13 Some orbital parameters oft and ö MOs of complex A 63Table 2.14 Bond indices and population analysis of model complex F 69Table 2.15 SCF energy values and the energies of some MOs of models H, J, K, L,MandF 71Table 2.16 Some orbital parameters of model H 72Table 2.17 Wiberg bond indices and Mulliken orbital populations of model H 73Table 2.18 A comparison of the energies of the lone pair electrons of Cl, NH2 andOH ions. 74Table 2.19 A compilation of chemical shifts of NH protons associated with hydrazidoligands 91Table 2.20 1H(31P} NMR data for complexes 2.12 and 2.12a 112xiTable 2.21 Moles of hydrazine measured from the reaction of ([(Pr2PCHSiMe)-NJZr(r5-C5H5)}(t-r‘:ri’-N) complex in toluene with anhydrous HC1gas. * Expected concentration was calculated by taking into account theamount of pentane present in the complex. 115Table 2.22 Mole percent gas evolved during the reaction of ([(Pr2PCHSiMe)-N]Zr(fl5-C5H5))(t-T1:-N) complex with H20 and HC1 solution. *Percent gas evolutions were calculated by taking into account the amountof pentane present in the complex. ** The data in parenthesis wasobtained after 30 minutes116Table 3.1 Selected zirconium-zirconium distances. (C1OH8 refers to the fulvaleneligand 128Table 4.1 Elemental composition found for some hafnium(ffl) derivatives 139Table 4.2 Micro-analytical data for the hydrocarbyl derivatives ofZr(i5-CH)C1-[N(SiMeCH2PPr1)] 141Table 4.3 Selected bond lengths of complexes Zr(r15-CHS)Ph[N SiMe2CH2Pi2)]4.4, andZr(15-C)CSiMe3[N(S Me2CH2PPr2],4.8 143Table 4.4 Selected bond angles of complexes Zr(fl-H5)Ph[N(SiMe2CHPPr’)2]4.4, andZr(rI-H)CSiMe[N(Si e2CHPPr2],4.8 144Table 4.5 Hyperfine coupling constants (G) for the chioro and the hydrocarbylderivatives ofZr(15-Hs)R[N(SiMe2CHPPr)2].All the values wereobtained from simulations 147Table 4.6 Selected bond lengths of complexZr(T5-C5H)(rjBH4)[N(SiMe2CH2-PPr’2)21, 4.17 167Table 4.7 Selected bond angles of complexZr(r1-)(12BH4)[N(SiMe2CH2-PPr2)], 4.17 167Table 4.8 Selected bond lengths of complexZr(r15-C)(1BH1CHO)-(N(SiMe2CH2PPr)[SiMe2CH2(Pr’2PB3)1),4.23 183Table 4.9 Selected bond angles of complexZr(-)( jH4rCHO)-{N(SiMe2CH2PPr2)[SiMe2CH(Pr2P•BHJ}, 4.23 183xList of FiguresFigure 1.1 Schematic representation of the MoFe nitrogenase and the structures of theMoFe cofactor and the P cluster. 3Figure 1.2 (a) A proposed energy diagram for the reaction of dinitrogen anddihydrogen in gas phase (top), in biological systems (middle) and in theHaber process (bottom). Nad and Had represent these atoms adsorbed on acatalytic surface. (b) An example of a diazine ligand involved inintramolecular hydrogen bonding.8Figure 1.3 Dinitrogen bonding modes. M, M’ and Mt’ can be different metals orsame metals with different ligand environments. 9Figure 1.4 The proposed mechanism for the protonation of end-on bound dinitrogenin molybdenum(O) and tungsten(O) complexes by HC1 in THF. Thechelating phosphine ligand in this case isPh2CHCPh.Theintermediates I and J are proposed for the protonation of bridgingcomplex H.12Figure 1.5 The proposed cycle for the catalytic reduction of hydrazine by complex(fl-C5Me5)WMe3(NNH2), K. 13Figure 2.1 I ORTEP view showing the complete atom labeling scheme of thecomplex ZrCpBr2[N(SiMeCHPr],2.7. II A Chem 3D® viewshowing the pseudo octahedral geometry at the zirconium. 26Figure 2.2 I ORTEP view showing the complete atom labeling scheme of complex{ [(Pr2PCHSiMe2)2NjZr(rI5-CH)} 2(t-11’: ‘-N2), 2.9. II A Chem3D® view of the arrangement of atoms around the zirconium centre. U AChem 3D® view showing the arrangement of the PNP ligand and theZr2N2 core.29Figure 2.3 I ORTEP view showing the complete atom labeling scheme of complex{[(Prj2PCH2SiMe2)2N1Zr(OAr*))2(i1:rI),2.12a. II A Chem3D® view showing the arrangement of the PNP ligand and the Zr2Ncore. III A Chem 3D® view of the arrangement of atoms around thezirconium centre.38xiiiFigure 2.4 Variable temperature 121 MHz 31P{ ‘HI NMR spectra of 2.12a recordedin C7D8. 43Figure 2.5 (right) 31P{1H} NMR spectrum of the sample obtained from the reactiondescribed in Scheme 2.8. (left) 15N NMR spectrum of the same sample(1:1 ratio of THF:C7D8). 50Figure 2.6 The resonance Raman spectra of complex 2.9 recorded at approximately90 K. A and B correspond to the (.i-14N2)and (j.L-’5N2) complexesrespectively. The top two traces were recorded in the solid state and thebottom two were recorded in the solution (THF) state. Each trace is a sumof 5 scans, each recorded at 2 cm-1/s using 40 mW of 514. 5 nm laser52excitation (Appendix A.1).Figure 2.7 The resonance Raman spectra of complex 2.2 recorded at approximately90 K. A and B correspond to the (.t-’4N2) and (j.L-15N2) complexesrespectively. The top two traces were recorded in the solid state (sum of 9scans) and the bottom two were recorded in the solution (THF) state (sumof 5 scans). Each scan was recorded at 1 cm/s using 20 mW of 647.1nm laser excitation (Appendix A.1).Figure 2.8 Depiction of the in-plane normal vibrational modes of the side-on peroxidebridged copper dimer.68 Broken lines increase in length as solid linesdecrease in length. Bonds which do not change in length during a givenvibration are not shown. Thicker lines indicate the dominant motion in thetwo Ag modes.68 The Bi deformation mode is not shown. 54Figure 2.9 The solid state resonance Raman spectra of 2.2, (A) and 2.19, (C) andof the sample (B) obtained from the reaction shown in Scheme 2.6. Allof the samples contain a (L-rI2:1-15N2)ligand. Spectral conditions aresimilar to Figure 2.8. 56Figure 2.10 Qualitative bonding description for mononuclear complexes (top) and endon bridging dinuclear complexes (bottom). 58Figure 2.11 Bonding scheme for the model A. 60Figure 2.12 Bonding scheme (in the box) describes the edge-on bridging dinitrogencomplexes of Co and Ni. 64xivFigure 2.13 Bonding scheme illustrating the types of overlap leading to the formationof the it-MO and the 6-MO. 66• Figure 2.14 Plot showing the change of energy in the model complex F with therotation of the dinitrogen ligand. 67Figure 2.15 Splitting of the it-MO in complex H. 72Figure 2.16 The angular overlap for the side-on bonded dinitrogen with a bent Zr2Ncore. 75Figure 2.17 The orbital interactions for a bent Zr2Ncore of complex 2.12 where thebend angles are greater than 1500. 76Figure 2.18 The change in SCF energy with respect to the bending of the Zr2N coreof model M. The core was bent along the nitrogen-nitrogen axis. 77Figure 2.19 Bonding scheme illustrating the symmetry based orbital requirements forthe dinitrogen binding in bridging side-on and end-on modes involvingML4 fragments. 78Figure 2.20 Important it-type interactions that determine the type of dinitrogen bondingin model complexes F and H. 80Figure 2.21 The frontier orbitals of the fragments G and N arranged in increasingorder of their energy. 81Figure 2.22 I The parent ion peaks of the ‘5N2 analogue of 2.9. II Parent ion peaksof 2.9 formed from the reaction of 15N2 analogue of 2.2 withNaCp’DME under an atmosphere of 14N2. 85Figure 2.23 The 500 MHz1Hf31P} NMR; B 121.4 MHz 31P{1H) NMR; C 30.406MHz 15N NMR and 96.2 MHz 11B { 1H} NMR spectra of the crudereaction mixture obtained from the reaction of 2.2 and LiBEt4in C7D8. 88Figure 2.24 A Variable temperature 1H NMR spectra (300 MHz) of complex 2.26recorded in C7D8. B and C are 31P{1H) NMR (121 MHz) spectra of2.26 and of the crude reaction mixture obtained from the reaction of 2.9and PhCH2Br.94xvFigure 2.25 The models used to generate the intermediate structures for the bending P,and rotation 0, of side-on bound dinitrogen complexes. The phantomatom “A” is placed in the middle of the N—N bond. The line X—Y inmodel P refers to the axis of the hinge about which the Zr2N2 plane wasbent.118Figure 4.1 An overlapping room temperature ESR spectra of a solution (toluene)sample of HfCpCl[N(SiMe2HPPr)]4.3, and ZrCpC1[N(SiMe2-CH2PPr)2], 4.1. 138Figure 4.2 I and III are ORTEP views showing the complete atom labeling schemefor complexes ZrCpPh[N(SiMe2CH2PPr)4.4, and ZrCp(CH2-SiMe3)[N(SiMeCHPPr]4.8, respectively. II and IV are Chem3D® views showing the arrangements of ligands around the zirconium incomplexes ZrCpPh[N(SiMe2CH2PPr2)2] 4.4, and ZrCp(CH2SiMe3)-[N(SiMe2CH2PPr2)]4.8, respectively.142Figure 4.3 The ESR spectra of the ailcyl complexes A 4.6, B 4.6-d1,C 4.5 and D4.8. For each case the observed spectrum is shown on the top and thesimulated spectrum is shown below. 146Figure 4.4 The solution infrared spectra of complexes 4.6, 4.6-d1 and 4.6-d5 149Figure 4.5 The room temperature solution (toluene) ESR spectrum of the phosphidecomplex ZrCpPPh[N(SiMeCH2P ri)j,4.12. 151Figure 4.6 The X-band room temperature solution (toluene) ESR spectrum of, (A)Zr(5-CH)H[N(SiMe2CH2PPr) ] 4.14, and(B) Zr(15-C5H)D[N(SiMe2CH2PPr2)2j,4.14-d 156Figure 4.7 A 500 MHz1H{31P} NMR spectrum of 4.15 showing the resonancesassociated with the PNP ligand. B 500 MHz 1H NMR spectrum of 4.15showing the resonances associated with the Cp ligand. C 500 MHz1H{31P} NMR spectrum of the deuteride analogue 4.15-do showing theresonances associated with the SiMe2 groups. D 76.77 MHz 2H NMRspectrum of 4.15-dn showing the resonances associated with the PNPligand. E 121.4 MHz 31P{1H} NMR spectrum of 4.15.158xviFigure 4.8 (A) Solution (toluene) ESR spectrum of 4.17 recorded in at roomtemperature. (B) The simulated spectrum of 4.17. 164Figure 4.9 I ORTEP view showing the complete atom labeling scheme of complexZrCp(BH4)[N(SiMe2CH2PPr2)2], 4.17. II A Chem 3D® viewshowing the arrangement of the PNP and borohydride ligand around thezirconium center.166Figure 4.10 The Chem 3D® view of the equatorial planes of complexes 4.4, 4.8 and4.17. All numerical values given correspond to the nearest bond angle.The Chem Draw® drawing inside the box is the proposed structure for thealkyl complexes in solution.168Figure 4.11 (top) 1H (300 MHz, C7Dg) spectrum of the diamagnetic species obtainedfrom the reaction of 4.17 and CO. (bottom) 1H NMR (300 MHz, C7D8)spectrum of the diamagnetic species obtained from the reaction of 4.1 andCO.170Figure 4.12 UV-Vis spectroscopic monitoring of the reaction of 4.17 (top right) and4.1 (bottom right) with CO (1 atmosphere). (left) Infrared spectrum(solution in toluene) of the reaction of 4.17 with CO. 171Figure 4.13 A, B Low temperature 1H (300 MHz) and 31P{1H} (121 MHz) NMRspectra (C7D8)of the diamagnetic species obtained from the reaction of4.1 and CH3N. 174Figure 4.14 A 1H NMR (300 MHz) spectrum of 4.23. B, C 1H (300 MHz) andgated decoupled 13C (75.4 MHz) resonances associated with the formylligand of 4.23 labelled with ‘3C. U H{ 11B} NMR (500 MHz)spectrum showing the BH3 unit and E 11B ( 1H} NMR (96.2 MHz)spectrun of 4.23 (C7D8).180Figure 4.15 I ORTEP view showing the complete atom labeling scheme of complexZr(T-CH5)(fl2BH4)(TC O){N(SiMe2CH2PPr2)[SiMeCH -(PriP.BH3)},4.23. II and III are Chem 3D® views of complex 4.23. 182xviiLIST OF ABBREVIATIONSThe following abbreviations, most of which are commonly found in the literature, areused in this thesis.A angstrom(1010m)Anal. analysisatm atmosphereavg averagehr broadBu n-butyl group, -CH2C3Bu iso-butyl group, -CHCH(CH)ButO tertiary butoxy group, -OC(CH3)11B{H} observe boron while decoupling protondegrees Celsius13C{1H} observe carbon while decoupling protonCalcd calculatedChem 3D® molecular modelling program for the Macintoschcm1 wave numberCp cyclopentadienyl group, [C5H]Cp* pentamethylcyclopentadienyl group, (C5CH3)} -Cp’ substituted Cp ligand, e.g., [C5H4Me]x.y-d Complex x.y has n number of 1H atoms replaced by 2H atomsd doubletd sept doublet of septetsd of d doublet of doubletsdeg (or 0) degreesxviiiA heatG gaussAG free energy of activationAH° enthalpy of reactionEl electron ionizationequiv equivalent(s)ESR electron spin resonanceeV electron Volt1H{ 1B } observe proton while decoupling boron1H{31P} observe proton while decoupling phosphorusdeuteriumrn-Ph rneta-hydrogen or rneta-carbono-Ph ortho-hydrogen or ortho-carbonp-Ph para-hydrogen orpara-carbonh Planck’s constant (or hour)HOMO highest occupied molecular orbitalHz Hertz, seconds-1INDO intermediate neglect differential overlapJR infraredJA—B n-bond scalar coupling constant between A and B nucleiK Kelvinkcal kilocaloriesKeq equilibrium constantLUMO lowest occupied molecular orbitalM central metal atom (or “molar”, when refening to concentration)m multiplet (or “medium”, for infrared data)M+ parent ionxixMS mass spectrometryMe methyl group, -CH3mg milligram(s)MHz megaHertzmL millilitremm millimetremmol millimole(s)MO molecular orbitalmol moleNMR nuclear magnetic resonanceORTEP Oakridge Thermal Ellipsoid Plotting ProgramOTf triflate group, -OSO2CF331P{1H} observe phosphorus while decoupling protonPh phenyl group, -C6H5ppm parts per millionPr isopropyl group, -CH(CH3)2ppm parts per millionRT room temperatures singlet (or “strong”, for infrared data)sept septetT temperaturet tripletTHF tetrahydrofuranTMS tetramethylsilaneTHT tetrahydrothiophineTMS tetramethylsilaneVT variable temperaturexxACKNOWLEDGMENTSI would like to thank my supervisor, Professor Mike Fryzuk, for sharing his inspiredmind and for his meticulous guidance and patience (i.e., whenever I failed to make him famous)throughout this project. I enjoyed working with past and present members of Fryzuk’s team;Dave, Guy, Patric, Lisa, Pauline, Tim, D. Berg, Kiran, Cam, Neil, Chas, Randy, Jesse, Craig,Cindy, Brian, Bobbi, Warren, Graham, Jonker, Danny, Garth, Laleh, Martin, Paul and Gao; Iam thankful to all of them.I extend my thanks to Professor Zaworotko, Department of Chemistry, St. Mary’sUniversity, for solving the zirconium(III) structures which were vital to this thesis; also, Dr.Steve Rettig for solving the dinitrogen structures. I greatly appreciate Mr. Jonathan Cohen andProfessor Tom Loehr, Department of Chemistry, Oregon Graduate Institute of Science andTechnology, for their contribution towards the resonance Raman studies. Dr. Peter Wassell isgreatly appreciated for allowing (tolerating) me to use the JR and UV-Vis spectrometers; also, itis a pleasure working with Peter.Many thanks go to Richard Schutte, Ken MacFarlane, Veranja Karunaratne, DannyLeznoff, Garth Giesbrecht and Guy Clentsmith, for proof reading my thesis, hopefully I willspell planer correctly in future.I would also like to acknowledge the assistance provided by the support staff of thechemistry department.I greatly appreciate Yoga’s family for their support during the course of my writing; ifnot for their help I would have been writing Chapter 2 forever.I still admire and appreciate the professors and other support staff from the University ofJaffna for providing an excellent curriculum amidst the turbulent atmosphere.xxiTO MY PARENTSwith love and respectChapter 1Introduction to Transition Metal Dinitrogen ChemistryLi GeneralIn the primitive atmosphere of the earth, the gaseous ammonia and dinitrogen moleculesplayed a crucial role in the formation of more complex nitrogen containing derivatives, such asamino acids and other nitrogen-containing bases.’-3 Formation of such biomonomers led to thechemical evolution of complex polymers such as proteins and nucleic acids, which eventuallyled to the origin of life on our planet.4 After a few billion years of complex evolution, all livingorganisms rely on these essential building blocks for their survival. Presently, mankindconsumes an enormous amount of nitrogen-containing fertilizers, which indirectly supplies theirnutritional needs. In this respect, the synthesis of ammonia from dinitrogen and dihydrogen, theHaber process, could be ranked as one of the most important discoveries. This event led to thecheaper manufacture of fertilizers and thereby has helped to sustain the population explosion ofthis century and to some extent social and political stability.Towards the end of nineteenth century, there was serious concern that the world’s nitratereserves would be exhausted due to the increased consumption by the explosives industry. Thisissue was certainly appreciated by the scientific community of the time. In 1898 WilliamCrookes, in his presidential address to the British association for the advancement of science,said, “it is through the laboratory that starvation may ultimately be turned into plenty”, andadded that, “the fixation of atmospheric nitrogen is one of the great discoveries awaiting theingenuity of chemists”.5Rayleigh and Crookes developed a method to fix atmospheric dinitrogen by electricdischarge of hot air to give nitrous and nitric acids. This so-called arc process consumed suchan enormous amount of energy that it was only feasible in countries where cheap hydroelectricpower was available. In 1895 Frank and Caro discovered that the reaction of calcium carbide,CaC2, with atmospheric nitrogen at very high temperatures gave calcium cyanamide CaCN2.references on page 18Chapterl 2This led to the development of cyanamide process in which the cyanamide produced was thenhydrolyzed to give ammonia.6In 1905 Fritz Haber7 studied the equilibrium reaction of dinitrogen with dihydrogen overan iron catalyst, and determined the thermodynamic data associated with this reaction.Although his findings were disputed by Nernst, Haber within a few years not only confirmed hisexperimental data but also developed a reactor which utilized the heat evolved during theexothermic reaction to heat the reaction gas.8 This convincing demonstration attracted theinterest of the mighty, ambitious BASF company. An engineer from BASF, Carl Bosch,developed the high pressure technology, leading to the opening of a commercial plant in 1911.6At present, most Haber processes operate at pressures between 100 and 350 bar and attemperatures around 530 °C using a Fe-A1203-K20catalyst. Dihydrogen used in the process isobtained by steam reforming or by partial oxidation of hydrocarbons and dinitrogen from air.61.2 Biological Nitrogen FixationPhotosynthesis and nitrogen fixation are the two basic life processes that enable thesurvival of the plant and animal kingdoms. Unlike photosynthesis, nitrogen fixation is onlycarried out by some primitive prokaryotic micro-organisms. Bortels, in 1930, discovered thatthe trace element molybdenum is essential for dinitrogen metabolism.9 It is now known thatnitrogen fixation is carried out by enzymes known as nitrogenases which contain differentcombinations of transition metal elements: Fe-Mo, Fe-V and Fe-Fe.’0 The enzymatic reductionof one mole of dinitrogen requires the break-down of 16 MgATPs, equivalent to the loss of488kJ, and suggests that the process is energy intensive (equation 1.1).”N2 + 8 H + 8 e + 16 MgATP1.1N2- ase2 NH3 + H2 + 16 MgADP + 16 P043 -The nitrogenase enzyme consists of two proteins, the Fe protein and the MoFe protein.The Fe protein is a dimer of two identical subunits bridged by a Fe4S cluster which has areferences on pageChapter I 3cubane type geometry with alike atoms occupying the opposite corners of the cube. The MoFeprotein has four subunits and contains two MoFe cofactors and two P clusters as illustrated inFigure 1.1.12,13 The Fe protein pumps the electrons into the cofactor consuming two moles ofMgATP per electron. The reduction of the nitrogen proceeds in the cofactor. For a long timemolybdenum was thought to be directly involved in the binding and the reduction of dinitrogen.On the basis of spectroscopic and modeling studies many structural models have been proposedfor the cofactor. 14-16 Recently Rees et al. unveiled a crystallographic structure of MoFe proteinfrom Azotobacter vinelandil at 2.7 A resolution. The structure of the cofactor consists of twocuboidal units bridged by two sulfur atoms and another unidentified atom “Y”.’749 Thecoordination geometry around the iron atoms is approximately trigonal planar. The location andthe coordinative saturation of molybdenum seems to diminish the role played by molybdenumin the cofactor.Low-Potential reductante 6H+N2/2NH3[e]2H HFe- Protein MoFe - Protein/FeFe2S\/ “I.. .X \ /Cys S—Fe S Fe Fq 5— Mo—OFLS/ N His S CysMoFe cofactorsFigure 1.1 Schematic representation of the MoFe nitrogenase and the structures of the MoFecofactor and the P cluster.MgATP NMgADP÷Pi} iP clusters Cysreferences on page.iChapter I 4In the newly described cofactor, imaginative ideas have been presented regarding themode of dinitrogen binding. At the observed separation, the two iron atoms bound to “Y”, maybe privileged to accommodate a side-on bound dinitrogen molecule.12 Theoretical studies doneon the Mo-Fe cofactor also suggests that a dinitrogen ligand bound in a side-on fashion will beactivated to a greater extent than the end-on case.20 The most prevalently known end-on modeof binding is also possible at the same site, but accessed only by the expansion of the MoFecage.The role of molybdenum is still alive. The molybdenum centre which is at the +4oxidation level during the resting state of the cofactor could possibly be reduced to the +2oxidation level during its active state. This could lead to the dissociation of certain ligandsattached to the molybdenum centre and thereby create a vacant site for the coordination ofdinitrogen.131.3 Other Methods of Nitrogen FixationLithium and some alkaline earth metals react with dinitrogen at ambient conditions togive metal nitrides, a process referred to as nitriding.” Vol’pin et at. showed that at elevatedtemperatures and pressures, nitriding of some main group metals could be catalyzed bytransition metal salts in aprotic media. The nitrides thus formed are then hydrolyzed to formammonia. A highly active system in this respect is formed by a mixture of TiC14,AlCl3 andmetallic aluminum, where greater than 200 moles of ammonia (after hydrolysis) per mole oftitanium were produced. Circumstantial evidence suggests that reduced species of titaniumformed during the reaction, form Lewis type adducts with A1C13. A species formulated asC6H•TiC12.2A1C3has been isolated in benzene, which is a catalyst for nitriding of aluminum,and also reacts with dinitrogen in the absence of aluminum to give a nitride complex.21’2The aprotic system described in Scheme 1.1 fixes dinitrogen at ambient conditions,where the operation is carried out in cycles. It is presumed that the low valent titanium speciesgenerated by the reaction of Ti(OR)4 and sodium naphthalide reduces dinitrogen to someunknown nitride species. Operation of the cyclic process depends on the sequence involvingreferences on page.iChapter 1 5judicious addition of proton source (isopropyl alcohol); the addition of electron source (sodiummetal) and removal of ammonia at the appropriate stages of the reaction sequence.21’3 Systemssimilar to the ones described above also facilitate in the incorporation of nitrogen fromdinitrogen into organic molecules as amine and nitrile functionalities.21’Ti(OR)4123I/1 +r I4RONaN2 11I”- 6ROH[nitride]4 Na NpScheme 1.1Shilov et a!. demonstrated that dinitrogen could be reduced in protic media. Two potentexamples are the gels formed by the combination of salts, Mo5orV4/Ti3MgCl2/KOH in anaqueous alcoholic medium.25 These systems are catalytic with respect to molybdenum andvanadium ions and fix dinitrogen only in the presence of these ions. Mechanistic investigationssuggest that the dinitrogen is probably activated within a preorganized bimetallic centre wherethe active metal centres are held by a complex arrangement of titanium and magnesium ions(Scheme 1.2).26IIIMo---NN--- Molli IVM0—HN—NH—M0IVI II I• IOH HO 0 0I___________II II H20—Tilil NITI——TiIV IVTi— HN—NHMg MgScheme 1.2Schrauzer et a!. have suggested a different mechanism for the reduction of dinitrogen inaqueous gel matrices. The mechanism involve a side-on bound dinitrogen complex as thereferences on pageTi(OR)2+2 RONa+2NaChapter 1 6intermediate, which is then reduced to give diazine*, HN=NH. The condition of the geldetermines whether diazine undergoes disproportionation to give hydrazineH2N—NH,anddinitrogen or will decompose to give dihydrogen and dinitrogen. (Scheme 1.3).27.28 Other thanthe methods described above, electrochemical29’73and photochemical30studies have also beencarried out. Recently, concerns have been raised regarding the reproducibility of thephotochemical reduction of dinitrogen.3’- N2H4HHO HO N HO ‘N : N2• % iI) IIV% iiv%%j1 _VO + IIH0IHOINHO.N.N+HScheme 1.31.4 Thermodynamic and Kinetic FactorsOn the basis of the discussion presented above, one can gauge the difficulties in fixing**dinitrogen. The extraordinary stability of dinitrogen reflects the relatively high ionizationpotential (15.6 eV)*** making it difficult to oxidize.32’3 A comparison of the dissociationenergies of dinitrogen, 946 kJ mol-1, diazine, 414 kJ mo11, and hydrazine, 159 Id mo11,suggests that the cleavage of the first nitroge-nitrogen bond consumes most of the energy ( 532kJ mol-1-).34 Although the reduction of dinitrogen to ammonia is exothermic the calculated heatof formation of diazine is highly endothermic by 205 kJ mo11. Therefore a reaction pathinvolving a diazine intermediate will require an activation energy not less than 205 Id mold.Such a high activation energy will correspond to a negligible*Dii is also refered to as diimide.**The term fixation will refer to the partial or complete dissociation of the dinitrogen triple bond to give productssuch as hydrazine and ammonia. Activation will refer to the lengthening of the dinitrogen triple bond compared toits gaseous state. Degree of activation can be observed directly in structurally known complexes or indirectly byother spectroscopic methods, e.g., infrared spectroscopy.***Comparable to the ionization energy of argon (15.75 eV)references on pageChapterl 7rate of reaction between gaseous dinitrogen and dihydrogen.26 Biological and industrial meansof surmounting this kinetic barrier are by the use of metal catalysts.N2 + H N2H AH =88 U mo11 1.2N2 ÷ 1 H2 N2H AH = 205 U mo11 1.3N2 + 2 H2 NH AH° =92 U molt 1.4N2 + 3 H2 2 NH3 AH° = -92 U mo11 1.5Comparison of the heats of formation of diazine and hydrazine indicates that asimultaneous reduction of two of the three bonds of dinitrogen seems more favorable than astepwise reduction (equations 1.3 and 1.4). This analysis has been extended to rationalize theactivation of dinitrogen coordinated between two metal centres, M—NN---*M, where it is likelyto undergo a four electron reduction to give a hydrazine type complex, M=N—N=M. Still,these metals would require such high reduction potentials that such reactions could not occur inprotic media.26Interestingly, in some diazine complexes the hydrogen atoms of the diazine ligand werefound to be involved in hydrogen bonding with sulfur atoms of the ancillary ligand (FigureIn biological systems such multiple hydrogen bonds have been shown to exhibit bondenergies up to 70 U mo11. Under such circumstances the formation of diazine intermediatewould possibly become exotherniic. Also for systems that operate in aqueous media, the heat ofhydration* will aid in the stabilization of the intermediates.26In the Haber process, the first step is the dissociative adsorption of dinitrogen andhydrogen to give surface bound nitrogen and hydrogen atoms. Under the conditions of Haberprocess the dissociative adsorption of dinitrogen is the rate-limiting step. The energy associated*\}f (H20) ofN2H4= 34 U mo1--; t\H298K(H3O) ofN2H4= -7.5 U mo11; zH298Kin (H20) of(NH3)= -161 U mo11;IH29gK (H3O) of (NH3)= -168 U mo1.references on pagejChapter I 8with the formation of surface-atom bonds over the catalyst compensates for the relevantdissociation energy, making the overall adsorption reaction exothermic.** The adsorbednitrogen atoms are then hydrogenated in a stepwise manner to give NH, NH2 and NH3respectively. Although these steps are thermodynamically “uphill” they are easily overcome atthe operating temperatures of the Haber process.6N2H200 (a)100-N2(g-i-Hg) N N2HiiN2+H”+M-M-N2H-MM-N2-M. . ---—.. NHBiological M-N2H4-M100 - systems________- Haberrocess- 3ad______‘2ad+Had- 200- / 1ad + 2 HadNad + 3 HadReaction CoordinateFigure 1.2 (a) A proposed energy diagram for the reaction of dinitrogen and dihydrogen; in gasphase (top), in biological systems (middle) and in the Haber process (bottom). Nadand Had represent these atoms adsorbed on a catalytic surface. (b) An example of adiazine ligand involved in intramolecular hydrogen bonding.1.5 Dinitrogen ComplexesIn 1965 Allen and Senoff isolated the first dinitrogen complex, {Ru(NH3)5N2].6Since then a wide variety of dinitrogen complexes have been prepared encompassing nearly allthe transition metals. In most cases the coordinated dinitrogen ligand was derived from gaseousdinitrogen and in some cases was derived from hydrazine, azides, nitrous oxide etc.37’8 Aconvenient way to present an overview of dinitrogen complexes is to group them on the basis of**FOr the adsorption of one nitrogen and three hydrogen atoms the reaction is exothermic by 259 kJ mo11.references on pageiChapter I 9bonding of dinitrogen ligand as shown in Figure 1.3. This figure will form the basis fordiscussion in the following section.M—NNIM-—NN---M’ MN NM’or 4 orM-—NN--’M M=N N=MfiBMM—NN: MEIM M—NN—M’M ,N. MM” M’III IvVM—HI M—WHM MMN NNVI VII VIIIFigure 1.3 Dinitrogen bonding modes. M, M’ and M” can be different metals or same metalswith different ligand environments.1.5.1 Mononuclear End-On ComplexesMononuclear complexes with end-on bonded dinitrogen ligands, I (Figure 1.3),are overwhelmingly represented. The bond distances of the nitrogen-nitrogen bond in thesecomplexes range from 1.0 A to 1.15 A, and have a M—NN bond angle greater than 175°. Thismode of coordination is scarcely known for Group 3, Group 4 and Group 5 metals. Recently ananionic vanadium complex with two terminally bound dinitrogen ligands has been structurallycharacterized, [Na(THF)j[V(N2(PhPCHC172 Monodentate and chelatingphosphine ligands are the common ancillary ligands in mononuclear complexes.Complexes with porphyrin ligands have also been reported.39’4° Some dinitrogencomplexes of iron, ruthenium and cobalt have been prepared by the addition of dinitrogen toreferences on pageChapter 1 10metal hydride complexes (equation 1.6).41 It has been proposed that in nitrogenase, hydrogen isevolved during the coordination of dinitrogen, which is comparable to the aforementionedreaction. Another pertinent reaction is the mutual exchange of hydride and dinitrogen ligands,where the metals involved are also present in nitrogenase (equation 1.7).42[RuH4(PPh3)] +2 [FeH(H)(dmpe) + [Mo(N2)(dppe)J2 [FeH(N)(dmpe)J[RuH2(N)(PPh3] + H2+ [M0H4(dppe)2J1. Dinuclear End-On complexesH3rJ,NH H3yHa]N—Fu—N—N—Ru—NHNH3 H3N NH31.124 (15)A1.11.165 (14)A1.2+In the linearly bridged end-on binding mode, II, the dinitrogen ligand shows a widerange of activation. The observed nitrogen-nitrogen bond lengths range from 1.0 A to 1.3 A,and have a M—NN angle greater than 170°. Case ha (examples 1.1 and 1.4) may beregarded as a Lewis acid-Lewis base type interactions where the nitrogen-nitrogen bond lengthsare close to that of free dinitrogen (1.098 A). In case hib (examples 1.3 and 1.7),4546 thebonded dinitrogen ligand is envisaged as a hydrazido type ligand, (N2)4 where the nitrogen-nitrogen bond lengths are approximately 1.3A. These two cases can be considered as the twolimiting resonance forms for the bridging end-on case II. The intermediate cases (example 1.2and can be understood on the basis of different contributions from the two forms Hareferences on page.2Chapterl 11and JIb. Although the two limiting cases can be reasonably depicted, for the intermediate casesdifferent authors seem to adopt different representations.Interestingly, a close resemblance exists between complex 1.3 and the proposedintermediates in the systems which fix dinitrogen in protic media (Scheme 1.2). In both casesthe metal ions, sodium or magnesium, seem to aid in the stabilization of the bimetallic centre. Arare example of case II having porphyrin as the ancillary ligand (example 1.6) is also known.These complexes are viewed as possible electrode catalysts for the reduction of dinitrogen.48Ph3F ?Et2Ph3—C o—N——N—U•••••’OEt2Ph3Fl.19(4)A1.4Me2Ph JPhMe2 CI I PhMe2 PMe2hCI—R e—N—N—M o—N—N—R e—CIMe2PhF”pphM>\j\I PhMe2F ‘\Me2Ph1.28 (5)A1.51.5.3 Protonation of Terminally Bonded Dinitrogen ComplexesM—NN—MM—NNProton sourceSolventNH3 and/or N2H4 1.8M(O)(d) +and/or N2Protonation reactions of ligated dinitrogen can be summarized as in equation 1.8. Ratiosof dinitrogen, ammonia and hydrazine produced vary widely, depending on the metal, protonsource and solvent. The reaction of [W(N2)2(PMe2Ph)4]with H2S04 in methanol gave 1.86moles of ammonia per mole of complex, which indicates almost a quantitative use of metalPorphyrin ligand1.6 1.7references on pagejChapter I 12electrons. The same complex reacted with HC1 in dimethoxyethane gave 0.5 moles of ammoniaand 0.3 moles of hydrazine per mole of complex.49HH\ H\/HP-I—p-N2 -H HC1HCI./ M \ / I \ _/ rvi \ M \P1 P P1 PN CIH CI CICIII D E /FNN N IHC1[j j // Highacid ÷ 2+I —“ HC1 i’i — p Concentration M M ici1/ri\ /M\.I-cI1 PIP jj IjiHIIIN N N NHAN BN —Cl + I II I—i M M MLowacid N ‘Concentration III : H I I JN :MC12HDihydride \ r1’iComplex H”H’- N2 G N2H4,HydrazineNFigure 1.4 The proposed mechanism for the protonation of end-on bound dinitrogen inmolybdenum(0) and tungsten(0) complexes by HC1 in TEIF. The chelatingphosphine ligand in this case isPh2CHCPh.The intermediates I and J areproposed for the protonation of bridging complex H.Some high oxidation state, dinuclear complexes of tantalum(V), niobium(V) andtungsten(VI), where the dinitrogen ligand is formulated as a hydrazido ligand (N2)4, givequantitative yields of hydrazine upon protonation.50’Conversely, complexes similar to 1.7(example[(rj5-CMe)MMe3]2(i-N,M = Mo or W) also formulated as tungsten(VI) anddinitrogen as hydrazido ligand (N2)4,react with acid to give only 0.3 moles of ammonia permole of complex. The yields of ammonia in these complexes were increased to 1.86 equivalentsreferences on pageChapterl 13when protonation was carried out under reducing conditions.52 It is important to point out that asurvey of numerous protonation reactions suggests no correlation between the degree ofactivation of the dinitrogen ligand and the observed fixation during protonation reactions.53The protonation reactions of complexes M(N2)2Ln where M = W or Mo and L =(PPhMe2)4or L =(Ph2PCHC)have been extensively investigated.37’854 Figure 1.4depicts the intermediates involved during the first two proton transfers to the coordinateddinitrogen. Intermediates B, C, D and E are invoked on the basis of kinetic and spectroscopicstudies. Diazenido (NNH)1 intermediates, similar to D or E are yet to be authenticated, but aLewis acid adduct (N—HN---BPh3)-’, has been structurally characterized.49 The hydrazido(NNH2)- intermediate F has been substantiated by NMR spectroscopy and X-raycrystallographic analysis. An interesting feature in some structurally known examples ofintermediate F (e.g., {[W(NNH2)Br(PMePh)3(C54NMe-3)](Br)}) is the presence of hydrogenbonding between the hydrogens of the hydrazido ligand and the halide ions.37 Differentmechanistic pathways have been proposed for the formation of ammonia and/or hydrazine fromintermediate F.(C5Me)WMe3NNH2Kj H/e•-y!<—NH3w Me’.w..’Me Me’.w”MeMeJNH2 Me’NH Me’’NH2Me Me N2H4 NHj NH2Q L MeH________Me7NH2 Me4CNH Me’4”NHMeMe MeMe NH3 2NH2p 0 NFigure 1.5 The proposed cycle for the catalytic reduction of hydrazine by complex(ri-C5MeS)WMe3NNH2), K.references on page j.Chapter 1 14Schrock et a!. have invoked similar hydrazido (NNH2)-intermediates during theprotonation of dinuclear complexes, for example(r15-CMe)MMe3JJ.t-N,M = Mo or W(Figure 1.5). The hydrazido (NNH2) intermediate formed in this case ((r15-CMe)MMe3(NNH2) M Mo or W) is formally in a higher oxidation state (+6) compared to theintermediates (e.g., F, Figure 1.4) formed from mononuclear complexes. Therefore for thefurther reduction of the hydrazido ligand in K an external electron source is needed. Thesuggested mechanism for the protonation of(ri5CMe)WMe3NNH2,under reducingconditions is shown in Figure 1.5. In the absence of a suitable electron source someintermediates undergo disproportionation reactions to give ammonia and dinitrogen.54’5 Also,intermediates K and M have been shown to catalytically reduce hydrazine (up to 10 equivalentsof hydrazine). It is also noteworthy that some sulfur ligated molybdenum(IV) complexes (e.g.[( o2C4( --S)(j.t-2-SC5H3NH-3-SiMe2--3 Si].THF) have shown to beeffective in the catalytic disproportionation of hydrazine to ammonia and dinitrogen. However,in the presence of a reductant the conversion of hydrazine to ammonia occurs catalytically (upto 1200 cycles).561.5.4 Polynuclear Dinitrogen ComplexesIn principle, the interaction of dinitrogen with three or more metal centres could generatenumerous modes of bonding (eg. III, IV and V in Figure 1.3). So far only a few differentmodes are known, each uniquely exemplified by 1.8, 1.9, 1.10 and 1.11. In most of these casesthe dinitrogen ligand is bound terminally as well as side-on.These complexes usually show a higher degree of activation of dinitrogen, withnitrogen-nitrogen bond lengths greater than 1.3 A. A few examples of case III (Figure 1.3) areknown, where M is tungsten or cobalt and M’ is either lithium, aluminum or magnesium (e. g.1.8).5758 Complex 1.9 exemplifies case IV. Treatment of the complex 1.9 with acid gave fourequivalents of dihydrogen and one equivalent of dinitrogen, whereas with water, a mixture ofhydrazine and ammonia was obtained (greater than 90% fixation).59 Until recently complex1.10, an example of case V, exhibited the longest nitrogen-nitrogen bond length of 1.36(2) A.60references on pageiChapter 1 15Although this complex showed a high degree of dinitrogen activation, reaction with ethylenesmoothly displaced the dinitrogen ligand.61The alkali-metal-cobalt complexes 1.4 and 1.11 are closely related, and in fact exhibitsimilar nitrogen-nitrogen bond lengths. However, the solid state structures of these twoderivatives differ drastically. In 1.11 the dinitrogen interacts with the alkali metal both in anend-on and a side-on fashion.62 Interestingly this type of coordination of dinitrogen closelyresembles the environment of the isoelectronic carbide ion in the alkaline earth salts (C22-;CaC2, MgC2, BaC2 etc). Calcium carbide has a tetragonally distorted NaC1 structure whereeach carbide ion is coordinated to six calcium ions. Four of these calcium ions interact side-onwith the carbide ion and the other two end-on. 63Co(PMe3) ,Co(PMe3),K-F+---,1-I IN :/ K---- —‘-N—Co(PMe3)(Me3P)Co(Me3P)3Co1.5.6 Mononuclear Side-On ComplexesOn the basis of spectroscopic evidence only one example of a mononuclear side-onbound dinitrogen complex has been reported. The temperature invariant ESR spectrum of thiszirconium(Ill) complex,Cp2Zr[CH(SiMe3)j(r-N,shows coupling due to two equivalentreferences on pageCICI IMe2PhP PPhMe2 ‘ PhMe2I PMe2hALCI—W-—N—N N—N—W—CII\ NAIPy PPhMe Py PMe2hJ CI1.46 (4)A Cland1.25 (3)A1.81.9l.301(12)A— LiO( \LIJNi NiOLI’X)1.36(2) A1.101.16A1.11Chapterl 16nitrogen atoms. Upon protonation, this complex gave hydrazine and dinitrogen (25%fixation).M.65In some cases side-on complexes have been implicated as intermediates. In the case ofcomplex(r1-CMe)Re( O)2(11’N—N’4,the dinitrogen ligand was found to undergo end-to-end rotation in an intramolecular fashion. Inevitably the proposed mechanism involves a side-on bound intermediate (Scheme 1.4).66 Mononuclear dihapto dinitrogen complexes have beenspectroscopically detected with many metals in low temperature matrices.6714N[Re] 14N—5 [Re]— [Re]—15N—415N[Re] =(5-CMe)Re(CO)2’N--N’4Scheme Dinuclear Side-On ComplexesIn the past five years, two examples of case VII and one example of case VIII have beenstructurally characterized. Case VII examples are complexes of samarium (example 1.12) andlithium (example 1.13).Although the complexes 1.12 and 1.13 are formed by different types of metals(lanthanide vs. alkali metal) the properties of the dinitrogen ligand are almost identical. Thenitrogen-nitrogen bond lengths of these two complexes are identical to that of free dinitrogenand have a planar M—N2—M core. Under vacuum the clinitrogen ligand can be stripped fromthe metals coordination sphere.68’9 Reversible binding of dinitrogen in solid and in solutionstates have been demonstrated for complex 1.12.69 Stephan et al. have proposed that forcomplex 1.13, the zirconium-phosphinidene dianion fortuitously favors the Li2—N dicationrather than a fully solvated lithium counter ions.68 Unfortunately, the mode of dinitrogenbonding in solution has not been verified for examples 1.12 and 1.13.The anionic complex 1.14 (example of case VIII), described by Gambarotta et al. has apseudo octahedral Ti(t-N2)2Ti core with the clinitrogen ligands defining the opposite edges ofreferences on pagejChapter I 17an equatorial plane.7° This anion is a mixed oxidation state complex where the titanium centrescan be formulated as either titanium(I) and titanium(ll) or as titanium(III) and titanium(IV)corresponding to neutral N2 and (N2)-ligands respectively. Although the formalism predictsactivation only to the extent of (N2)-the observed nitrogen’-nitrogen bond length of (1.38(2) A)in 1.14, is much longer than a nitrogen-nitrogen double bound; by comparison PhN=NPh has abond distance of 1.255 A.71 The nitrogen-nitrogen distance in 1.14 seems to correlate moreclosely to dinitrogen complexes where the dinitrogen is formulated as hydrazido ligand, (N2)4-.The discussion presented so far on dinuclear side-on dinitrogen complexes seems tosuggest that the degree of activation of dinitrogen as measured by bond lengths falls in twoextreme ends. For the samarium and lithium cases dinitrogen is barely activated whereas in thetitanium case the observed dinitrogen bond distances are much greater than what would beexpected.1.13a1.06 (l)A1.6 SummaryTerminally bound dinitrogen complexes have been synthesized with most transitionmetal complexes with varying degrees of activation. Although the motive for this type ofresearch was eventually to develop complexes that could catalytically reduce dinitrogen, so far1.121.38 (2)A1.142- 2+1references on page.Chapter 1 18mononuclear and dinuclear end-on complexes have been extensively investigated. Yet,compared to the mechanistic understanding of reactions involving other small molecules such ascarbonmonoxide, dihydrogen etc., the protonation reactions of dinitrogen are limited. In thepast five years, few side-on bound dinitrogen complexes have been structurally characterized.Another important milestone is the structural elucidation of the cofactor of the enzymenitrogenase. Ironically, a few researchers have proposed that the dinitrogen might interact in aside-on fashion during its activation in the cofactor. Structural and theoretical studies seem tosuggest that a side-on mode of bonding leads to a greater degree of activation of dinitrogen.In the following chapter new side-on bound dinitrogen complexes of zirconium will bediscussed. Aspects of the bonding both in solution and in the solid state, and some aspects ofthe reactivity of the coordinated dinitrogen will be considered.1.7 References(1) Miller, S. L. J. Am. Chem. Soc. 1955, 77, 2351.(2) Lemmon, R. M. Chem. Rev. 1970, 70, 95.(3) Ponnamperuma, C. Quart. Rev. Biophys. 1971,4, 77.(4) Price, C. C. Synthesis ofLife; Dowden, Hutchingon & Ross, Inc.: Stroudsburg,Pennsylvania, 1974; Vol. 1.(5) Crookes, W. In The Report of the 68th meeting of the British Associationfor theAdvancement ofScience; John Murray, London: Bristol, 1898; p 3.(6) Jennings, I. R. Catalytic Ammonia Synthesis: Fundamentals and Practice; Plenum Press:New York and London, 1991; Vol. 1.(7) Feldman, M. R.; Tarver, M. L. .1. Chem. Ed. 1983, 60, 463.(8) Haber, F.; vonOordt, G. Z. Anorg. Chem. 1905,44, 341.(9) Bortels, H. Arch. Mikrobiol. 1930, 1, 333.(10) Dilworth, M. J.; Glenn, A. R. Biology and Biochemistry ofNitrogen Fixation.; Elsevier:New York, 1991, pp Chapter 4.(11) Chatt, J.; Leigh, G. 3. Chem. Soc. Rev. 1972, 1, 121.(12) Orme-Johnson, W. H. Science 1992,257, 1639.(13) Sellmann, D. Angew. Chem., mt. Ed. Engi. 1993,32, 64.Chapter I 19(14) Ciuril, S.; Hoim, R. H. Inorg. Chem. 1989,28, 1685.(15) Coucouvanis, D. Acc. Chem. Res. 1991,24, 1.(16) Burgess, B. K. Chem. Rev. 1990, 90, 1377.(17) Georgiadis, M. M.; Komiya, H.; Chakrabati, P.; Doo, D.; Komuc, 3. J.; Rees, D. C.Science 1992, 257, 1653.(18) Kim, J.; Rees, D. C. Nature 1992, 360, 553.(19) Kim, J.; Rees, D. C. Science 1992, 257, 1677.(20) Deng, H.; Hoffmann, R. Angew. Chem., mt. Ed. Engi. 1993,32, 1062.(21) vanTamelen, E. E. Acc. Chem. Res. 1970,3, 361.(22) Volpin, M. E.; flatovskaya, M. A.; Kosyakova, L. V.; Shur, V. B. J. Chem. Soc., Chem.Commun. 1968, 1074.(23) vanTamelen, E. E.; Fechter, R. B.; Schneller, S. W.; Boche, G.; Greeley, R. H.;Akermark, B. J. Am. Chem. Soc. 1969, 91, 1551.(24) vanTamelen, E. E.; Rudler, H. J. Am. Chem. Soc. 1970, 92, 5253.(25) Shilov, A.; Denisov, N.; Efimov, 0.; Shuvalov, N.; Shuvalova, N.; Shilova, A. Nature1971,231, 460.(26) Shilov, A. E. Russ. Chem. Rev. 1974,43, 378.(27) Zones, S. I.; Palmer, M. R.; Palmer, J. G.; Doemeny, 3. M.; Schrauzer, G. N. J. Am.Chem. Soc. 1978, 100, 2113.(28) Liu, N. H.; Strampach, N.; Plamer, 3. G.; Schrauzer, G. N. Inorg. Chem. 1984,23, 2772.(29) Haight, G. P.; Scott, R. .1. Am. Chem. Soc. 1964, 86, 744.(30) Edwards, J. G.; Davies, J. A.; Boucher, D. L.; Mennad, A. Angew. Chem., mt. Ed. Engi.1992,31, 480.(31) Edwards, J. G.; Davies, J. A.; Augugliaro, V.; Soria, 3.; Palmisano, L.; Schiavello, M.;Sciafani, A. Angew. Chem., mt. Ed. EngI. 1993, 32, 550.(32) Chatt, J. Pure Appi. Chem. 1970,24, 425.(33) Pelikan, P.; Boca, R. Coord. Chem. Rev. 1984,55, 55.(34) Jolly, W. L. The Inorganic Chemistry ofNitrogen; W. A. Benjamin, INC.: New York,1964.(35) Sellmann, D.; Soglowek, W.; Knoch, F.; Moll, M. Angew. Chem., mt. Ed. Engi. 1989,28, 1271.Chapter 1 20(36) Allen, A. D.; Senoff, C. V. Chem. Commun. 1965, 621.(37) Chatt, J.; Dilworth, J. R.; Richards, R. L. Chem. Rev. 1978, 78, 589.(38) Henderson, R. A.; Leigh, G. J.; Pickett, C. J. Adv. Inorg. Chem. Radiochem. 1983,27,197.(39) Buchier, J. W.; Smith, P. D. Angew. Chem., mt. Ed. Engi. 1974, 13, 745.(40) Camenzind, M. I.; James, B. R.; Dolphin, D. J. Chem. Soc., Chem. Commun. 1986,1137.(41) Knoth, W. H. J. Am. Chem. Soc. 1968, 90, 7172.(42) Leigh, G. J.; Jimenez-Tenorio, M. J. Am. Chem. Soc. 1991, 113, 5862.(43) Treitel, L. M.; Flood, M. T.; Marsh, R. E.; Gray, H. B. J. Am. Chem. Soc. 1969, 91,6512.(44) Yamamoto, A.; Miura, Y.; Ito, T.; Chen, H. L.; In, K.; Ozawa, F. Organometallics 1983,2, 1429.(45) O’Regan, M. B.; Liu, A. H.; Finch, W. C.; Schrock, R. R.; Davis, W. M. J. Am. Chem.Soc. 1990, 112, 4331.(46) Ferguson, R.; Solari, B.; Floriani, C.; Chiesi-Villa, A.; Rizzoli, C. Angew. Chem., mt.Ed. Engl. 1993, 32, 396.(47) Sanner, R. D.; Duggan, D. M.; McKenzie, T. C.; Marsh, R. E.; Bercaw, J. E. J. Am.Chem. Soc. 1976, 98, 8358.(48) Coilman, J. P.; Hutchison, 3. E.; Lopez, M. A.; Guilard, R. J. Am. Chem. Soc. 1992, 114,8066.(49) Talcahashi, T.; Mizobe, Y.; Sato, M.; Uchida, Y.; Hidai, M. J. Am. Chem. Soc. 1980,102,7461.(50) Churchill, M. R.; Li, Y. J.; Theopold, K. H.; Schrock, R. R. Inorg. Chem. 1984,23,4472.(51) Dilworth, J. R.; Henderson, R. A.; Hills, A.; Hughes, D. L.; Macdonald, C.; Stephens,N.; Watson, D. R. M. J. Chem. Soc., Dalton Trans. 1990, 1077.(52) Schrock, R. R.; Kolodziej, R. M.; Liu, A. H.; Davis, W. M.; Vale, M. G. J. Am. Chem.Soc. 1990, 112, 4338.(53) Leigh, G. J. Acc. Chem. Res. 1992,25, 177.(54) Leigh, G. J. J. Mol. Cat. 1988,47, 363.Chapter I 21(55) Schrock, R. R.; Glassman, T. E.; Vale, M. V.; Kol, M. J. Am. Chem. Soc. 1993, 115,1760.(56) Block, E.; Ofori-Okai, G.; Kang, H.; Zubieta, J. J. Am. Chem. Soc. 1992, 114, 758.(57) Glassman, T. E.; Liu, A. H.; Schrock, R. R. Inorg. Chem. 1991,30, 4723.(58) Takahashi, T.; Kodama, T.; Watakabe, A.; Uchida, Y.; Hidai, M. J. Am. Chem. Soc.1983, 105, 1680.(59) Pez, G. P.; Apgar, P.; Crissey, R. K. J. Am. Chem. Soc. 1982, 104, 482.(60) Jonas, K.; Brauer, D. J.; Kruger, C.; Roberts, P. J.; Tsay, Y. H. J. Am. Chem. Soc. 1976,98, 74.(61) Jonas, K. Angew. Chem., mt. Ed. Engi. 1973, 12, 997.(62) Hanimer, R.; Klein, H. F.; Friedrich, P.; Huttner, G. Angew. Chem., mt. Ed. EngI. 1977,16,485.(63) Greenwood, N. N.; Earnshaw, A. Chemistry of The Elements; 1st edition ed.; PergamonPress: Oxford, 1986, pp 318.(64) Jerrery, 3.; Lappert, M. F.; Riley, P. I. J. Organomet. Chem. 1979, 181, 25.(65) Gyane, M. J. S.; Jeffery, J.; Lappert, M. F. J. Chem. Soc., Chem. Commun. 1978, 34.(66) Dengel, A. C.; Griffith, W. P.; O’Mahoney, C. A.; Williams, D. 3. J. Chem. Soc., Chem.Commun. 1989, 1720.(67) Ozin, G. A.; Voet, A. V. Can. J. Chem. 1973,51, 637.(68) Ho, 3.; Drake, R. J.; Stephan, D. W. J. Am. Chem. Soc. 1993, 115, 3792.(69) Evans, W. 3.; Ulibarri, T. A.; Ziller, 3. W. J. Am. Chem. Soc. 1988, 110, 6877.(70) Duchateau, R.; Gambarotta, S.; Bensimon, C. J. Am. Chem. Soc. 1991, 113, 8986.(71) Allen, F. H.; Kennard, 0.; Watson, D. G.; Brammer, L.; Orpen, A. G.; Taylor, R. J.Chem. Soc., Perkin Trans. 1987,11, Si.(72) Rehder, D.; Woitha, C; Priebsch, W; Gailus, H. J. Chem. Soc., Chem. Commun. 1992,365.(73) Pickett, C. J; Ryder, K. S; Talarmin, J. J. Chem. Soc., Dalton Trans. 1986, 1453.22_Chapter 2Dinitrogen Complexes of Zirconium2.1 GeneralOne of the on going research interests of our group involves the synthesis of transitionmetal complexes containing the tridentate ligand [N(SiMe2CHPR)]1,abbreviated as PNP.This hybrid ligand was designed with a central amide unit flanked by two identical armscontaining phosphine donors.1-5 The incorporation of mismatched donor groups, i.e., the hardamide and the soft phosphine donors, has enabled the successful coordination of this ligand tohard, early and soft, late transition metals.6’7 The formation of two fused five-membered chelaterings has also been shown to favour the complexation of this ligand.As a formal six electron donor the PNP ligand is somewhat comparable to theisoelectronic Cp ligand. However, in certain instances, dueMe2 Me2to the availability of lone pair electrons on the nitrogen, PNPF 1 could donate a maximum of eight electrons, which would beR2P M- PR2 manifested by shorter than usual metal-amide bondPNP distances.8 Also, in some late transition metal complexesR = Me, Ph, Pr1, But the nitrogen lone pair has played an important role as anM = metal ionintramolecular Lewis base (e.g., Ir(2-propenyl)(A1Me2)-[N(SiMe2CHPPh)])and as a proton acceptor (e.g., IrHMeBr[NH(SiMe2CH2PPh]).9”°The tridentate ligand usually exhibits meridional coordination, but occasionally facialcoordination has been observed in the solid (e.g.,ZrCl3[N(SiMe2CHPM )])’and in thesolution (e.g.,kH(13-C5)[N(SiMe2CHPPh1)’”states. Also, the ligand could influencethe reactivity at the metal center by changing the coordination from tridentate to bidentate, oreven monodentate, a process that could be compared to the ring slippage of the Cp ligand fromr15 to i3 to q1. The 31P{1H} NMR spectra of PNP derivatives are useful in assigning structuresof these molecules and to monitor the reaction progress. The steric bulk of PNP can be readilyaltered by changing the size of the substituents on the phosphine donors, for example, the Me,Ph Pr1 and But derivatives are all available.references on page j9.Chapter II 23Investigations involving PNP as an ancillary ligand on zirconium led to the discovery ofcarbon-carbon bond formation between butadiene and allylic fragments.’4 Also some zirconiumcomplexes have shown fluxional behavior (e.g.,Zr(r-CH6)Cl[N(SiMe2CH2PPr’)])andintriguing coordination geometries such as a bicapped tetrahedron (e.g., ZrMe3[N(SiMeCH2-PMe2)]).15’6Pr12PMe2Si \ %CI 2 Na/Hg%s.Zr__Cl N2 (4atm)Me Si” CP. (—2NaC1)2.1Scheme 2.1Investigations on the reduction ofZrCl3[N(SiMe2CHPPr)J2.1, under a dinitrogen•Zr—N—N— atmosphere gave rise to a deep blue complex.Structural analysis of the blue complex 2.2, revealedN2.3 it to be a dinuclear derivative containing a bridgingT1nlN2Z .l2A side-on bound dinitrogen ligand (Scheme 2.1).8 Theobserved nitrogen-nitrogen bond distance of 1.548(7)A in 2.2 was significantly longer than that observed for a nitrogen-nitrogen single bond; forexample in hydrazine the nitrogen-nitrogen bond length is 1.46 Prior to this work therewas only one other structurally characterized zirconium dinitrogen complex [(ri 5N—Nt = 1.548 (7) A Zr—P = 2.77AZr—N(1) = 2.175 (3) A Zr—N 2.024 (4) AZr2N = Planar2.2Nreferences on page 1.2.Chapter II 24C5Me5)2ZrN212-(p.-fl’:fl’-N2) 2.3.18-21 The dinuclear complex 2.3 has three dinitrogen ligands,two terminal and one bridging, and the bridging dinitrogen ligand has a nitrogen-nitrogen bondlength of 1.182 A, considerably shorter than that found in 2.2.To elaborate on the chemistry of side-on dinitrogen complexes similar to 2.2 weinvestigated the synthesis of other dinitrogen complexes of zirconium incorporating the PNPligand. We were particularly interested in substituting the chloride ligand of 2.2 with othersubstituents such as Cp, alkoxide, etc. This chapter describes the synthesis of a series ofdinitrogen complexes and aspects of their bonding. The mode of dinitrogen bonding in the solidversus the solution states was of interest along with some aspects of reactivity.2.2 Synthesis of a Dinitrogen Complex Containing PNP and Cp Ligands2.1 R=Pr’,X=Cl2.4 R=Pr,X B2.5 R = Me, X = Cl____________Me2SI\NaCp•DME:j—/Toluene Me2SiR=PX=Y2.6 R=,X=ClNa/Hg2.7 R=Pr1,X BToluene2.8 R = Me, X = Cl(i—N)Pr’222.9Scheme 2.2Initial attempts to metathesize the zirconium-chloride bonds of the side-on dinitrogencomplex 2.2 with alkyl lithium or sodium aryl oxides resulted in decomposition to unidentifiedproducts. Also the synthesis of 2.2 involves a tedious workup procedure, where optimumreferences on page 2.22Chapter II 25conditions give only 40% yields. Therefore a two step reaction sequence starting with complex2.1 was developed as shown in Scheme 2.2. Selective metathesis of a single zirconium halidebond of the trihalo precursor and subsequent reduction did enable the synthesis of a newdinitrogen complex.2.2.1 Synthesis of Precursors of the Type ZrCpX2[N(S1Me2HPR)]Synthesis of ZrCpC12[N(SiMe2CH2PPr’)2] 2.6, was readily achieved by reacting oneequivalent of solid NaCp•DME with a toluene solution ofZrCl3[N(SiMe2CHPPr) ], 2.1.After recrystallization, pure hexagonal shaped crystalline material was obtained in excellentyields (> 75%). The related complexes 2.7 and 2.8 were also synthesized via a similarreactions. A single resonance for the SiMe2 moieties in the 1H NMR spectrum and a singleresonance in the 31P{1H} NMR spectrum suggest a symmetric structure for the complex 2.6 insolution.* The protons of the Cp ligand appear as a triplet, due to the coupling of two equivalentphosphines. These spectroscopic data are consistent with a C2v structure for complex 2.6 insolution, i.e., assuming free rotation of the Cp ligand about the 119axis defined by the zirconium and the centroid of the Cp (Scheme 2.2). The principal C2 axiscoincides with the axis defined by the centroid of the Cp, the zirconium and the nitrogen, andthe zirconium lies in a pseudo-octahedral environment. No evidence of fluxionality wasobserved in the 1H NMR spectrum when a sample of 2.6 was cooled to -90 °C. This couldimply that the Cp ligand is coordinated in an fl5 fashion in solution. The solution spectroscopicdata obtained for the analogous complexes 2.7 and 2.8 also suggest geometries similar tocomplex 2.6.Figure 2.1 shows the X-ray structure of complex ZrCpBr2{N(SiMe2CH2PPr’)],2.7.The Cp ligand is found to be trans to the amide linkage, and the nitrogen, zirconium and thecentroid of the Cp are arranged almost linearly (Table 2.1). The two phosphorus atoms and thetwo bromide atoms are slightly bent away from the Cp ligand. Assuming that in solution the Cp* In the 1H NMR spectrum of complexes containing PM the resonances due to Si(CH3)2group appear between0.00 ppm and 0.75 ppm. The number of resonances observed for the Si(CH3)2moieties, in conjunction with the31P(1H) NMR spectrum enable us to predict the geometry of the molecule with reasonable accuracy.references on pageChapter II 26ligand is undergoing rapid rotation about the zirconium centroid axis the overall geometry of themolecule in solid state agrees with the geometry inferred from solution spectroscopic data. Thestructure also shows the T coordination of the Cp ligand.CZ3C19IIFigure 2.1 I ORTEP view showing the complete atom labeling scheme of the complexZrCpBr2[N(SiMeHPPr)],2.7. II A Chem 3D® view showing the pseudooctahedral geometry at the zirconium.The bond lengths (Table 2.1) and bond angle (Table 2.2) parameters associated with the“Zr[N(SiMe2CHPPr)]”unit are similar to other structurally known complexes ofzirconium(IV) with the PNP ligand.” The distance from zirconium to the centroid of the Cp,2.286(2) A is comparable to the distances observed for the bis(cyclopentadienyl) derivatives (e.g Cp2ZrI2, 2.20 A)22 and the monocyclopentadienyl derivatives (e.g. [CpZrCl3], 2.196 A)23.4czzNBrIreferences on pageChapter II 27Table 2.1 Selected bond distances of complex 2.7ZrCpBr[N(SiMeCH2PPr)].atom atom distance (A) atom atom distance (A)Zr Cp 2.286(2) Zr Br(1) 2.6871(7)Zr P(1) 2.817(1) Zr Br(2) 2.6842 (7)Zr P(2) 2.821(1) Zr C(average) 2.57Zr N 2.227(2) N Si(average) 1.736Table 2.2 Selected bond angles of complex 2.7ZrCpBr[N(SiMeCH2PPr)}.atom atom atom angle (0) atom atom atom angle (0)P(1) Zr P(2) 152.82(4) P(1) Zr Cp 103.02(7)Cp Zr N 179.2(1) P(2) Zr Cp 104.16(7)Br(1) Zr Br(2) 162.95(2) P(1) Zr N 76.18(9)P(1) Zr Br(1) 90.31(3) P(2) Zr N 76.65(9)P(1) Zr Br(2) 86.00(3) Cp Zr Br(1) 99.02(7)P(2) Zr Br(1) 85.08(3) Cp Zr Br(2) 98.03(6)P(2) Zr Br(2) 90.62(3) N Zr Br(1) 81.50(8)Much of what is known about the reactivity of organometallic complexes of zirconiumarises from the work done on dihalo zirconocene derivatives, Cp’2ZrX2 (Cp’ = any ligandcontaining a cyclopentadienyl unit, X = halide).2536 Because complexes 2.6, 2.7 and 2.8 can beconsidered as isoelectronic to Cp’2ZrX2 useful comparisons are possible. An importantdifference in their physical properties is that the PNP derivatives are soluble in aliphaticreferences on page 112Chapter II 28hydrocarbon solvents, whereas the zirconocene derivatives are usually soluble only in polarsolvents.Preliminary studies involving PNP derivatives and aluminoxanes have shown catalyticactivity towards the polymerization of ethylene.37 It is well established that zirconocenederivatives also function as excellent catalysts under similar conditions. Stereo-regularpolymerization involving chiral zirconocene derivatives38’9could also be extended to the PNPsystems by incorporating chiral cyclopentadienyl ligands or by using chiral phosphine donors.40Reactions mediated by Cp’2ZrX2 complexes take place at adjacent sites, i.e., the sitesoccupied by the halide atoms. From a mechanistic viewpoint, the availability of adjacentreactive sites has been conducive to the successful utilization of these complexes (e.g., reductiveelimination, a-H abstraction), particularly in their applications to organic synthesis.36 In thecase of PNP derivatives, the reactive sites, i.e., the sites occupied by the halides, are transrelated. Therefore, for these sites to become involved in a mutual manner they have to bebrought closer, either by the rearrangement of the PNP to a facial type coordination or by thedissociation of one of the phosphine donors.2.2.2 Reduction of Complexes of the Type ZrCpX2[N(S1Me2HPR)]A toluene solution of complex ZrCpCl[N(SiMeCPPr’)J2.6, was reduced withNa/Hg under 4 atmospheres of dinitrogen. The yellow color of the reaction mixture rapidlydisappears to give a deep green solution which turns deep brown within a few days. Theintermediate responsible for the green colour was also formed in the absence of dinitrogen(under an argon atmosphere) implying that it does not contain a dinitrogen ligand. A deepbrown crystalline material was isolated from the brown solution which was formulated as adinuclear dinitrogen complex (Scheme 2.2) on the basis of spectroscopic and micro analyticaldata.references on page 112Chapter IIFigure 2.2 I ORTEP view showing the complete atom labeling scheme of complex([(PrPCHSiMe)2N]Zr(15-CH1JI-1 ’:r’-N,2.9. II A Chem 3D® viewof the arrangement of atoms around the zirconium centre. II A Chem 3D® viewshowing the arrangement of the PNP ligand and the Zr2N2 core.29PNPSiP PIII SiSiISiNIIreferences on page QChapter II 30The X-ray structure of the dinitrogen complex{[(Pr2PCHSiMe)N]Zr(ri5-H}(.t-11 ‘:rl1-N2) 2.9, is depicted in Figure 2.2. The solid state structure unequivocally shows thatthe dinitrogen ligand is bound in an end-on fashion bridging two zirconium centres. Theobserved nitrogen-nitrogen bond distance of 1.30 1(3) A (Table 2.3) is significantly shorter incomparison to the similar distance of 1.548(7) A in the side-on derivative 2.2, but longer than(by 0.1 A) the end-on derivative 2.3, which is the only other zirconium complex with an end-onbound dinitrogen ligand that has been structurally characterized. When a comparison is madewith end-on complexes in general where the nitrogen-nitrogen bond lengths range from 1.1 A to1.3 A, the nitrogen-nitrogen distance in 2.9 falls in the upper limit (Chapter 1, section 1.5.2).Also, in complex 2.9 the zirconium-nitrogen distances associated with the dinitrogen ligand areapproximately 0.1 A shorter as compared to the side-on complex 2.2 or to the similar distance incomplex 2.3*.21 The dinitrogen ligand in 2.9 is slightly offset from the zirconium-zirconiumaxis, having a Zr—N—N angle of 170°.Table 2.3 Selected bond distances of complex 2.9{ [(Pr’PCH2SiMe2)2N]Zr(115-)} 2(L111:r ‘-N2).atom atom distance (A) atom atom distance (A)Zr(1) Cp(1) 2.294 (2) Zr(2) Cp(2) 2.286 (2)Zr(1) P(1) 2.8 16 (1) Zr(2) P(3) 2.778 (1)Zr(1) P(2) 2.776 (1) Zr(2) P(4) 2.828 (1)Zr(1) N(1) 2.306 (3) Zr(2) N(2) 2.303 (3)Zr(1) N(3) 1.920 (3) Zr(2) N(4) 1.923 (3)N(3) N(4) 1.301 (3) Zr C(average) 2.58* In complex 2.3 the zirconium-nitrogen bond distances for the bridging dinitrogen ligand are 2.087 (3) A and2.075 (3) A, and for the terminal dinitrogen ligands is 2.188 (4) A.references on pageChapter II 31Table 2.4 Selected bond angles of the complex 2.9{ [(PrPCH2SiMe2)2N]Zr(rI5-H5)} 2(.LT1’ :r ‘-N2)atom atom atom angle (°) atom atom atom angle (°)P(l) Zr(1) P(2) 145.87 (4) P(3) Zr(2) P(4) 145.65 (3)N(1) Zr(1) N(3) 115.1 (1) N(2) Zr(2) N(4) 115.3 (1)N(l) Zr(l) Cp(l) 127.61 (9) N(2) Zr(2) Cp(2) 127.49 (9)N(3) Zr(1) Cp(1) 117.2 (1) N(4) Zr(2) Cp(2) 117.1(1)P(l) Zr(1) N(1) 71.74 (8) P(3) Zr(2) N(2) 75.37 (7)P(l) Zr(1) N(3) 98.92 (8) P(3) Zr(2) N(4) 86.10 (8)P(1) Zr(1) Cp(l) 102.89 (6) P(3) Zr(2) Cp(2) 104.84 (6)P(2) Zr(1) N(l) 75.66 (8) P(4) Zr(2) N(2) 71.53 (8)P(2) Zr(l) N(3) 85.76 (8) P(4) Zr(2) N(4) 99.76 (8)P(2) Zr(1) Cp(1) 104.84 (6) P(4) Zr(2) Cp(2) 102.41 (6)Zr(1) N(3) N(4) 170.6 (2) Zr(2) N(4) N(3) 170.1 (2)A comparison of the arrangement of the PNP ligand in complexes 2.2 and 2.9 shows thatin the end-on case it is significantly twisted where as in the side-on case it is almost planar,reflecting the increased steric crowding around each metal centre in the end-on derivative. Also,the bond distances associated with the zirconium-amide bonds are approximately 0.1 A longerthan those observed in complex 2.2. In fact, the zirconium-amide bond distances in 2.9 are thelongest of any zirconium-nitrogen distances that have been measured for PNP coordinated tozirconium,8’11 and are just slightly shorter than the distances of 2.443(1) A and 2.412(2) Afound in neutral amine type adducts of zirconium(IV).41’2The zirconium-Cp bond distances ofcomplexes 2.6 and 2.9 are identical. Analysis of the bond angles (Table 2.4) around eachreferences on page i;iChapter II 32zirconium of 2.9 suggests an approximate trigonal bipyramidal geometry, with the phosphinesoccupying the apical positions, slightly bent away from the Cp ligand.The 1H NMR spectrum of 2.9 has only one resonance associated with the Cp ligand andtwo resonances associated with the SiMe2 moieties; the 31P{1H} NMR spectrum consists of asingle resonance at room temperature. These spectroscopic data suggest that complex 2.9adopts a more symmetric structure (C2v) in solution than that observed in the solid state (C2).Variable temperature NMR spectroscopic studies with complex 2.9 show fluxionality. Thevariable temperature 31p{ 1F1 } NMR spectra show broadening and decoalescing of the singletresonance which below -40 C separates into an AB quartet(2Jp..p = 80 Hz). In the 1H NMRspectra the two resonances associated with the SiMe2 groups give rise to four peaks at lowtemperature. These low temperature spectroscopic data are consistent with the structureobserved in the solid state. The 1H and 31P{1H} NMR spectra were virtually unchangedthrough the temperature range from -40 C to -90 C, suggesting that phosphine dissociation isnot involved in this particular fluctional process. Imagining a fluctional process involvingsynchronized pivoting of the zirconium-amide bonds with respect to the Zr2N2 core wouldrender the two phosphines equivalent (Scheme 2.3).Pe2i SiMe2MeSi../ Pr2Me2SiPr’2prScheme 2.3references on page 112Chapter II 33The replacement of a chloride ligand of the side-on dinitrogen complex 2.2 by a Cpwould dramatically increase the steric interactions between the two zirconium centres. By thisrationale, substituting the chloride ligands of 2.2 with Cp ligands would favor the formation ofthe end-on complex over the side-on complex. In the side-on complex 2.2 the two metal atomsare separated by 3.7 A which is significantly closer than the comparable distance of 5.1 A in theend-on complex 2.9. The important difference in the electronic properties of the two complexesis that the Cp ligand in 2.9 is bonded to the metal by a a and two it-bonds whereas in 2.2 the Clligand is mainly bonded via a a-bond. These interactions will significantly alter the type oforbitals available for dinitrogen bonding. Considering electron donor properties, the Cp ligandis likely to make the metal more electron rich. Although this would increase metal to dinitrogenbackbonding interactions it is difficult to predict how this would influence the mode ofdinitrogen bonding.2.3 Attempted Synthesis of a Dinitrogen Complex Containing PNP and AllylligandsConsidering the discussion presented in Section 2.2, replacing the chloride ligand of thedinitrogen complex 2.2 with an allyl ligand would be a suitable extension of our investigation.Sterically, an allyl ligand is less bulky than a Cp ligand. Electronically, it donates less electronsto the metal than a Cp ligand and is bonded to the metal via one a and one it-bond. Synthesis ofthe monoallyl derivative was achieved by a metathesis reaction involving 1 equivalent ofC3H5MgC1 and 2.1 (Scheme 2.4). However, initial attempts to synthesize the allyl derivativeusing C3H5MgBr and 2.1, gave a mixture of products. Spectroscopic and elemental analysissuggests that the mixture consists of two products, the desired complex Zr(i3-CH5) 12[N(SiMe2CHPPr)j} 2.10, andZr(r3-CH5)ClBr[N SiMeHPP] 2.lOa.In the 1H NIVIR spectrum of 2.10 the SiMe2 moieties gives rise to two resonances andthe 31P(1H} NMR spectrum consists of a single peak. The syn, anti and central protons of theallylic group appear as three separate resonances in the 1H NMR spectrum suggesting that theallyl group is rigid within the observed time scale. The 31P{1H} NMR spectrum of areferences on pageChapter II 34recrystallized sample from the reaction ofC3H5MgBr and 2.1 consists of a singlet and a set ofAB quartets(2Jp..p = 7.9 Hz) corresponding to 2.10 and 2.lOa, respectively. Assuming that 2.10and 2.lOa are isostructural in solution the 31P{1H) NMR data can be explained only if the PNPligand adopts a facial geometry. The suggested geometry for 2.10, shown in Scheme 2.4 has Csymmetry, which intuitively suggests that 2.lOa will be asymmetric. The 1H NMR datasuggests that in 2.10 the symmetry plane containing the central proton of the allylic ligand andbisects the molecule into two chemically equivalent halves.Me Sr”\\ C3H5MgX 2 \Me2S:N1rMe2S2.1 X = Cl, Y = Cl, 2.10X = Br, Y = Br, 2.lOa and Cl, 2.10Scheme 2.4NOEDIFF experiments involving the irradiation of the syn or the anti protons of theallylic ligand resulted in the spin saturation of both the syn and the anti protons and showedenhancement of the resonance due to central proton. frradiating the central proton resulted inthe enhancement of the resonances due to syn and anti protons. This seems to suggest that theallyl ligand undergoes syn anti exchange with a time scale comparable to the duration ofradiation during the NOEDIFF experiment.When a toluene solution of complex 2.10 was reduced with Na/Hg or with Mg under 4atmospheres of dinitrogen the yellow colour of the reactant rapidly changed to give a greensolution. Upon further reaction the green colour slowly changed to give a deep brown solution.Attempts to isolate solid materials from these solutions only resulted in impure oils.references on page flChapter II 352.4 Synthesis of a Dinitrogen Complex Containing PNP and Aryl OxideLigandsAfter exploring the effects of substituting the Cl ligand of 2.2 with It-type ligands suchas Cp and allyl, attention was focused on investigating the preparation of the followingmonodentate ligands: amide (NPh2), ailcyl (CH2SiMe3),alkoxides (OCHPh2and OBut) andaryl oxides (OPh and OAr* where Ar* = (C6HMe2-2,6). Of these, OAr* was the mostsuccessful in leading to the synthesis of a dinitrogen complex, hence the chemistry associatedwith OAr* ligand will be discussed in detail.2.4.1 Synthesis ofZr(OAr*)C12[N(SiMeCHPPri)]PP1j2Me2Si ci *\ NaOAr:N—zr—ci/1 Toluenee21\ /2.1TolueneAddition of solid NaOAr* to a toluene solution of the trichloro starting material 2.1 ledto the synthesis of Zr(OAr*)C12[N(SiMeCHPPr1)],2.11 in good yields (Scheme 2.5). It isimportant that toluene be used as the solvent to obtain high yields of the monoaryloxidecomplex. During test runs in THF it was found that the reaction yielded a mixture of mono andPr22.12Scheme 2.5references on pageChapter II 36bis OAr* substituted products in approximately a 3:1 ratio, respectively. When the reaction wascarried out in an NMR tube containing a solvent mixture of THF/C6D(THF:C6D= 4:1), the31P{1H} NMR spectrum showed two resonances: one corresponds to unreacted 2.1 and theother to the uncoordinated phosphine of PNP. A sample prepared in C6D after stripping theTHF solvent did not show any resonances due to uncoordinated phosphines. This spectroscopicevidence seems to suggest that soon after the formation of mono OAr* the phosphines becomelabile in THF solution (Scheme 2.6). The substitution of bulky phosphine donors by THFligands is likely to decrease the steric crowding around the metal center, allowing furthermetathesis reactions. When the reaction was carried out in toluene, after the formation of themono OAr* derivative the phosphine donors remained coordinated. This makes the metal centresterically more crowded and therefore the metathesis of the second zirconium-chloride bond issignificantly slower.R2PMe2Si\CiMe2S:Nzr___oAr*_________*/ /1—- •NZf—QArMe2Si CI / Me2Si” CI” I> ()PIA2The 1H and 31P{ ‘H} NMR spectra of the monoaryloxide 2.11 show that the complexessentially has the same symmetry (C2) as complexes 2.1 or 2.6. As depicted in Scheme 2.5the molecule adopts an approximately octahedral geometry having trans geometry.2.4.2 Reduction of Zr(OAr*)C12[N(S1MeCHPPri)]During the reduction of 2.11 with Na/Hg under a dinitrogen atmosphere, depending onthe purity of the dinitrogen gas, two types of materials were isolated. When the reaction wasScheme 2.6references on pageChapter II 37carried out under normal dinitrogen (refer to experimental for the purification of dinitrogen) thecrude material contained many impurities. The crude material was washed with hexanes andthen recrystallized from a toluene solution to give pure material directly. When the reductionwas carried out under ‘5N2 gas the crude product had fewer impurities, and washing withhexanes afforded pure material. The spectroscopic analysis (31P(1H) NMR spectroscopy) ofthe pure material obtained with normal dinitrogen showed the presence of one compound,whereas the product obtained from the 15N2 reduction showed the presence of two compoundsin approximately 2:1 ratio. In the latter case the chemical shift of the major peak was identicalto that obtained under normal dinitrogen. The elemental analysis of both materials gaveidentical results, and was consistent with the formula{[(Pri2PCHSiMe2)N]Zr OAr*)}(N2),2.12.The above experimental details show that the reduction of 2.11 under dinitrogen givesrise to two isomers of complex 2.12. The major isomer, 2.12a has slightly better solubilityproperties than complex 2.2 whereas the minor isomer 2.12b dissolves readily in hexanes andtoluene. Therefore when reductions were carried out under normal dinitrogen, during workupmost of the minor isomer was removed along with the hexane washings.Single crystals suitable for X-ray analysis were obtained for the major isomer 2.12a.The solid state structure unequivocally shows that the dinitrogen is bound in a side-on fashionbridging the two zirconium centers (Figure 2.3). The observed nitrogen-nitrogen bond length of1.528(7) A (Table 2.5) is identical to the other side-on derivative 2.2, and is significantly longerthan the bridging end-on complexes where, as already mentioned, the nitrogen-nitrogen bondlengths range from 1.12 A to 1.3 A. A compilation of nitrogen-nitrogen bond lengths, presentedin Table 2.6, provides a comparison with other dinitrogen complexes and also to selected othermolecules of interest. The dinitrogen bond lengths in complexes 2.2 and 2.12a are not onlylonger than that of free hydrazine but also longer than that found for coordinated hydrazineligand (N2H4),or hydrazido ligand (NHNH)2 (Table 2.6).references on page 112Chapter II 38Figure 2.3 I ORTEP view showing the complete atom labeling scheme of complex([(PrPCH2SiMe2)2N]Zr(OAr*)}2(IIfl:fN),2.12a. II A Chem 3D® viewshowing the arrangement of the PNP ligand and the Zr2N2 core. III A Chem 3D®view of the arrangement of atoms around the zirconium centre.P PISiSiIllreferences on page 212Chapter II 39Table 2.5 Selected bond distances of complex 2.12a,{ [(PrPCH2SiMe2)2N]Zr(OAr*) }2(JI-’q:fl-N2).atom atom distance (A) atom atom distance (A)Zr(1) 0(1) 2.020(3) Zr(1) N(1) 2.211 (3)Zr(1) P(1) 2.818 (1) Zr(1) N(2) 2.034 (4)Zr(1) P(2) 2.846 (1) Zr(1) N(2)’ 2.082 (4)N(2) N(2)’ 1.528 (7) N(1) Si(1) 1.718 (4)Table 2.6 Compilation of nitrogen-nitrogen bond lengths for some selected compounds.Compound Bond Length (A) ref.N2 1.0975 (2)Ph—N=N—Ph 1.255ON—NO2 2.08 69O2N—NO 1.75 69H2N—NH 1.46 17{[(PrPCHSiMe)N]Z (OAr*)} 2.12a 1.528 (7) this work{[(PrPCH2SiMeN]ZrCl}(,I-1: I- 2.2 1.548 (7) 8,45[Cp*2Smj(I11:1)1.12 1.088 (12) 46W(NPh)Me3](J.I-1’:11‘NHN)(1HNH) 471H1’NH2NI{ 1.434 (14)1.391 (15){ [(PrPCH2SiMe2NJZr(115-H)}i-N2.9 1.301 (3) this work{[(PriPCH2SiMeN]TiCl}QI 1.275 (7) 48{{(PrPCH2SiMeN]VC1}(.L-N 1.257 (8) 48{(Me3)2CCH)Ta= HCMe 12(.t-N) 1.298 (12)references on page 212Chapter II 40The very long nitrogen-nitrogen bond distances of complexes 2.2 and 2.12a vindicatesthe discussion presented for the titanium complex with a “Ti2(I-T12:’ri-N )2” core, 1.14(Chapter 1, Section 1.5.7). In complexes 2.2 and 2.12a the dinitrogen ligand can have amaximum formal charge of 4-, (N2)4-. However, comparison with end-on complexes where thedinitrogen ligand can be formulated as a (N2)4-ligand, for example, as found in complexes withnitrogen-nitrogen bond distances closer to 1.3 A, the side-on complexes of zirconium exhibitextravagant nitrogen-nitrogen bond lengths, in fact they are longer than that found in hydrazine.A conspicuous difference between the side-on complexes 2.2 and 2.12a is that the Zr2Ncore in 2.12a is hinged along the nitrogen-nitrogen axis, whereas in the chloro derivative 2.2 itis planar. In complex 2.12a the triangular planes defined by each zirconium and the nitrogens ofthe Zr2N2 core form an angle of 156.2, and the orientation of the PNP ligands with respect tothe Zr2N2 core can be described as syn. Considering the structure of 2.12a as a whole, thebending of the core perhaps eases any steric interactions that might arise if the core was planar,since with a bent core, PNP ligands are moved away from each other. The calculatedzirconium-zirconium separation of 3.7 A in 2.12a is identical to that of 2.2.The bond distances associated with the phosphine donors and zirconium are comparableto other PNP complexes of zirconium.8’1150 The bond lengths from the zirconium to each of thebridging nitrogens Zr-N(2), are 2.034(4) and 2.082(4) A, and this is shorter than the zirconiumamide bond length, Zr-N(1), 2.211(3) A of the ancillary ligand, PNP; and are comparable to thesame bond parameters of 2.2. Also the zirconium-amide bond, Zr-N(1), 2.211(3) A, issignificantly shorter than the comparable bond length of the cyclopentadienyl dinitrogencomplex 2.9, which is 2.306 (3)A. Other zirconium-nitrogen bond distances in the literaturerange from 1.826(4) A for a zirconium-nitrogen double bond inCp2Zr=NBut(THF)S’ to2.443(1) A in Schiff base chelate derivatives.52The zirconium-oxygen bond separation, Zr-O(1) is 2.020(3) A and the bond angledefined by the atoms Zr, 0(1) and C(19) is 161.5(4)° (Table 2.7). These two interrelatedstructural features are indicators of metal-oxygen d-p interactions where larger bond anglesand shorter bond lengths are correlated with iu-electron donation from the oxygen.53 Byreferences on pagejChapter II 41comparison, parameters associated with the zirconium-oxygen bond for{[(But)3C0]2ZrCl.Li(OEt2)}53are 1.859 A and 169°; for[Cp2Zr(NMe)(OCH3But,6)]4they are 2.056(1) A and 142.7(7)° and for [(Cp2ZrMe)20154,1.945 A and 174.1° are found. Inthe solid state structure of 2.12a the OAr* ligand is oriented such that the methyl substituentsare directed away from the isopropyl substituents of the phosphine donors.Table 2.7 Selected bond angles of complex 2.12a,([(PPrj2CH2SiMe2)2N]Zr(OAr*)12(.t-fl:rI-N2).atom atom atom angle atom atom atom angleP(1) Zr(1) P(2) 145.54 (4) P(1) Zr(1) N(2) 128.10 (10)N(1) Zr(1) N2 114.9* P(2) Zr(1) N(2) 82.89 (10)N(1) Zr(1) 0(1) 123.0 (1) P(1) Zr(1) N(2)1 85.10 (10)0(1) Zr(1) N2 122.3* P(2) Zr(1) N(2)’ 126.19 (10)0(1) Zr(1) N(2) 119.8(1) P(1) Zr(1) 0(1) 87.6(1)0(1) Zr(1) N(2)’ 119.1 (1) P(2) Zr(1) 0(1) 88.1 (1)N(1) Zr(1) N(2) 111.4(1) P(1) Zr(1) N(1) 79.4(1)N(1) Zr(1) N(2)’ 114.7 (1) P(2) Zr(1) N(1) 74.5 (1)N(2) Zr(1) N(2)’ 43.6 (2) Zr(1) N2 Zr(1)’ 156.2*P Zr(1) N2 106.4* Zr(1) N(2) N(2) 69.9 (2)P Zr(1) N2 104.8* Zr(1) N(2) Zr(1)’ 130.7 (2)Zr(1) 0(1) C(19) 161.5 (4)* These data were measured from a molecule constructed by using the structural data of 2.12a ina CAChe work station.The room temperature 1H NMR spectrum of the major isomer shows two resonances forthe SiMe2moieties and a singlet for the protons of the two ortho methyl groups of OAr*; the31P{1H) NMR spectrum consists of a singlet. An averaged solution structure depicted in 2.12areferences on page 212Chapter II 42having C2 symmetry accounts for both 1H and 31P{ ‘H) NMR spectral features. Thespectroscopic analysis of the sample containing a mixture of both isomers suggests that theminor isomer adopts a less symmetric structure, where the SiMe2 protons give rise to fouroverlapping signals and the phosphines give rise to two singlet resonances in the 1H and31P{1H) NIvIR spectra, respectively. On the basis of the assumption that the minor isomer hasthe same type ofZr2(.t- I:11-N)core as the major isomer the structure depicted as 2.12b isproposed, having C symmetry. By having the bent Zr2N core, the ends of the molecule arenow different, consistent with the solution spectral data. Considering the arrangement of theligands with respect to the Zr2N core, the two isomers can be envisaged as the syn 2.12a andthe anti 2.12b isomers.A variable temperature 31P(1H} NMR analysis of 2.12a showed an unsymmetricalbroadening of the singlet resonance at 8.69 ppm. At temperatures below -70 °C a newresonance began to appear around 4.0 ppm which increased in intensity as the temperature waslowered to -90 °C. Upon cooling, it was expected that the singlet resonance would decoalesce ina symmetric manner to give an AB quartet or two singlets which could be related to the solidstate structure of 2.12a. However, none of the low temperature 31P{1H} NMR featurescorrespond to the solid state structure of 2.12a (Figure 2.4). Considering the room temperature1H NMR spectrum of 2.12a, the appearance of a singlet due to the methyl protons of OAr*2.12a 2.12breferences on page QChapter II 43group can result from rotation about the oxygen-carbon bond; but the appearance of only twopeaks for the SiMe2 units, instead of four as predicted by the solid state structure, would requirethe rotation of the OAr* group about the zirconium-oxygen bond, averaging the inequivalentphosphine environments.I I I I I I12 10 8 6 4 2PPIb4 14 12 10 8 6 4PP2Figure 2.4 Variable temperature 121 MHz 31P{1H} NMR spectra of 2.12a recorded in C7D8.Studies associated with zirconium aryloxide complexes (e.g., [Cp2Zr(NMe)(0C6H3-2,6But2)]) suggest restricted rotation about the zirconium-oxygen bond, where the calculatedAGI values were around 73 kJ mol1. It is possible that in complex 2.12a rotation about thezirconium-oxygen bond could be sterically hindered. The complex probably undergoes-55 °C 95 V-80 °Croom temperaturereferences on pageChapter II 44reversible dissociation of one of the phosphine arms of PNP during which the OAr* groupswings from one end of PNP to the other end. The intermediate observed in the variabletemperature 31P{1H} NMR spectrum may be due to the rapid equilibrium between the twointermediates which have one phosphine arm dissociated (Scheme 2.7).‘SiMe2Pr’2 O’ Me2PrZrPN—, ,N.Sj’NIMe2Scheme 2.7Replacing the chloride of 2.1 with OAr* ligand has not altered the mode of dinitrogenbonding nor the degree of activation. The structural parameters associated with the zirconium-oxygen bond indicate that there is t-bonding between the two atoms, which would make themetal centers of 2.12 more electron rich than the metal centers in the side-on derivative with thechloride ligand 2.2. The increased steric crowding caused by the replacement of the chlorideligand of 2.2 with OAr* is overcome by the bending of the Zr2N core. A favorable physical09c0‘II IIreferences on page flChapter II 45property attained by the substitution of the chloride ligand with OAr* is the increased solubilityof the minor isomer 2.12b in hydrocarbon solvents.2.5 Synthesis of {[(PriPCHSiMe)N]Zr(OBut)}Q.irI2:12..N)Selective metathesis of only one chloride of 2.1 with one tert-butoxide (OBut) group toyield the precursor Zr(OBut)Cl2[N(SiMeCHPPr1)2.13, was partially successful. Reactionscarried out under different conditions gave oily mixtures of mono and bis substituted complexeswhich were difficult to separate. The reduction of the crude precursor (containing approx. 80%of the monosubstituted derivative 2.13) with Na/Hg gave a deep purple solution. The 31P{1H}NMR spectrum of the reduced material has a broad peak at 8.16 ppm, which is comparable toother side-on bound dinitrogen complexes 2.2 and 2.12. Unfortunately the complex 2.14 wasextremely soluble in hydrocarbon solvents (pentane) and remains as an impure oil even at-78 °C. This could be partly attributed to the build up of impurities, since the precursor alsocould not be purified. The 1H NMR features were similar to the side-on dinitrogen complexes2.12 having two resonances for the SiMe2group and a singlet for the OBut group.A compilation of 31P{ ‘H} NMR chemical shifts for some zirconium precursors and thedinitrogen complexes derived from them is given in Table 2.8. The data suggest that thephosphorus-3 1 chemical shifts for a given precursor will shift to high-field upon formation of aside-on dinitrogen complex and shift to low-field upon the formation of an end-on dinitrogencomplex. The zirconium tert-butoxy precursor, having a 31P chemical shift of 11.6 ppm, shiftsto a lower field upon its reduction under dinitrogen atmosphere. Therefore, it is presumed thatthe product is a side-on dinitrogen complex and is formulated as {[(PrPCH2SiMe)N](OBut)Zr}2(.t-rI2:rI-N),2.14. In the absence of other corroborating evidence, however, this proposalis speculative.The methyl-PNP derivative Zr(OBut)C12[N(SiMeCHPMe)]2.15, can be synthesizedand isolated as a pure material. The reduction of this complex under dinitrogen atmosphere wasslow and prolonged reaction times afforded many products as detected by 31P ( 1H} NMRspectroscopy. Also, the final reaction mixture was only faintly colored, whereas the reductionreferences on page flChapter II 46of other precursors that yielded dinitrogen complexes always gave deeply colored solutions. Adifferent alkoxide precursorZr(OCHPh2)C12[N(SiMe2C 2PPr’)2] 2.16, upon reduction (Na/Hgor Mg) gave uncoordinated PNP ligand. The mono amide precursor Zr(NPh)Cl2[N(SiMe2-CH2PPr)]2.17, and the mono alkyl precursor Zr(CH2SiMe3)Cl2[N(SiMe CH2PPr)2] 2.18,gave intractable materials upon reduction.Table 2.8 Compilationof31P{1H} NMR chemical shifts for some zirconium precursors andthe dinitrogen complexes derived from them. In all these complexes PNP corresponds to[N(SiMe2CH2PPr’2)2].Precursors Chemical Dinitrogen Complexes ChemicalShift (ppm) Shift (ppm)(PNP)ZrC13 15.4 [(PNP)ZrC1]2(I-rI:fl-N2) 10.0(PNP)ZrBr3 16.5 [(PNP)ZrBr]2(ji-1:r- 10.015.0 [(PNP)Zr(OAr*)12(p.11:12) 8.7, 8.9 &1 1.3(PNP)Zr(OBut)C12 11.6 [(PNP)Zr(OBut)]2(.t11:fl2) 8.2(PNP)ZrCpCl2 16.1 [(PNP)ZrCp12.t-:’1-N2) 20.3(PNP)TiC13 -1.9 [(PNP)TiC1]2(I.L-’il’ :T1-N2) 21.2(PNP)Ta(=CHBut)Cl2 37.5 [(PNP)Ta(=CHBut)]2I.trll:‘-N2) 48.8(PNP)Ta(=CHPh)C12 40.1 [(PNP)Ta(=CHPh)] 2(.t-TI’ :fl -N) 51.12.6 Spectroscopy as a Diagnostic Tool for End-On and Side-On DinitrogenComplexesAlthough X-ray structure determination of dinitrogen complexes will unequivocallyshow the mode of dinitrogen binding in the solid state, it does not necessarily relate to the modeof dinitrogen binding in the solution state. Also, X-ray analysis inevitably requires the isolationof single crystals which ultimately relies on serendipity. Therefore, any spectroscopic techniquethat could differentiate between these two modes of bonding would not only help in theassignment of dinitrogen bonding in complexes that are difficult to crystallize but would alsoreferences on pageChapter II 47enable one to ascertain the mode of dinitrogen bonding both in the solid and the solution states.We were particularly interested in discerning the mode of dinitrogen bonding in the side-oncomplex 2.2 in the solution state.2.6.1 15N NMR SpectroscopyThe nitrogen-15 analogues of complexes 2.2, 2.9 and 2.12 were prepared by introducinggas during the reductions of the corresponding precursors. Originally, we had hoped thatthe measurement of the nitrogen-15 chemical shifts55 would provide a diagnostic handle on themode of binding in these derivatives, that is, side-on bridging versus end-on bridging.However, as is evident from the chemical shifts compiled in Table 2.9, there is no obviouscorrelation between the two modes of coordination. For example, the chemical shift of the side-on complex 2.2, whether in the solid or in the solution state, is in the range of 345 to 351 ppmwhile the end-on derivative 2.9 is at 354 ppm. In fact, the only correlation seems to be related towhich metal the dinitrogen ligand is attached, since the zirconium complexes resonate around350 ppm, with the exception of the bridging dinitrogen ligand of Thedinuclear tantalum complexes also seems to display very similar chemical shifts irrespective ofthe ancillary ligands. But, also notable is that the formally tungsten(VI) derivative,[Cp*WMe3J2Q.LN2), and the mononuclear molybdenum(0) complex, t r a n sMo(N2)(PMePh)4,show similar chemical shifts to the tantalum diniirogen complexes. Thechemical shift of the bridging dinitrogen ligand in the diamagnetic titanium dinitrogen complexhaving PNP as the ancillary ligand appears around 375 ppm. Dinuclear titanium dinitrogencomplexes reported in the literature are paramagnetic complexes.5658 None of the dinitrogencomplexes containing PNP as the ancillary ligand showed any observable coupling to thephosphine donors in their 15N NMR spectra.59’6°references on page flQChapter II 48Table 2.9 Compilation of some 15N chemical shiftsa for some selected compounds.Compound Chemical Shift (ppm) Reference([(2PCH2SiMe2)2N]ZrCl}2Q1-n:r1-N ) 350.9 (solution) this work345.0 (solid){[(PrPCHSiMe2)2N]ZrBr}2(.t-rI:11-N2) 345.8 this work{[(PrPCH2SiMe2)2NJZr(OAr*)}2(J.trIflN 342.9 and 339.1 this work{[(PrPCHSiMe2)2N]Zr(15-CSHS)} 201-1 1:1 --N2) 354.0 this work{[(PrPCHSiMe2)2NjTiCl}2(J.t-fl’:1 ‘-N2) 3755 48[Cp*2Zr(1-N2)12(t-r :rI ‘-N2) 452.01 (bridging) 61354.Ob (terminal){[(PrPCHSiMeNJTa=CHButI 20.1-ri’ :r1-N2) 301.2 62{(Me3P)2C1Ta=CHCMe }2(t-n’:1-N2) 296.2c 63[Cp*WMe](,.Lr:riuN 310.9c 64trans-Mo(1’-N2)2(PMePh4 298.3 & 305•4d 60Afl chemical shifts are referenced to external neat formamide set at 0.0 ppm. boflginallyreferenced to nitric acid. COriginally referenced to liquid ammonia. doriginally referenced tonitromethane.A different approach to probe the binding of the dinitrogen ligand in the binuclearside-on zirconium complexes is to synthesize closely related complexes where the twozirconium centers are chemically different. For example, substituting one of the chlorideligands of complex 2.2 with a different ligand (e.g., Br or OAr*) will render the two zirconiumcenters chemically different, and also change the symmetry of 2.2 from C2h to C. Assumingthat the ancillary ligand arrangements are similar in the solid and in the solution states, thesolution 15N NMR spectrum of the complex corresponding to C symmetry should give a singletresonance, if the dinitrogen ligand is bound in a side-on manner in solution. Conversely, if thereferences on page QChapter II 49dinitrogen is bound in an end-on fashion the 15N NMR spectrum may consist of two singlets ora doublet-of-doublets corresponding to an AB type spin system.Since attempts to selectively metathesize the zirconium-chloride bonds of 2.2 werefutile,65’6 mixtures containing a 1:1 ratio of two different precursors were reduced under 15N2atmosphere. When an equimolar mixture of 2.1 and 2.11 was reduced under dinitrogen it gavemainly complex 2.2 and none of the desired dinitrogen complex containing one chloride and oneOAr* ligand. Conversely, when a similar reduction was carried out with complex 2.1 and thebromide analogue ZrBr3[N(SiMe2CHPPr)]2.4, it gave a mixture of three dinitrogencomplexes. The 31P(1H} NMR spectrum of the mixture had four singlet resonances, whichwere very close in their chemical shifts. Two of these resonances were assigned to complex 2.2and to the dibromo analogue 2.19, and the other two resonances were assigned to the mixedhalogen derivative, 2.20 where each signal is associated with one type of zirconium centre(Figure 2.5). The 1H NMR spectrum of the mixture shows the presence of eight equal intensityresonances (some overlapping) in the SiMe2region. This spectroscopic analyses shows that theabove reaction produces a mixture of three complexes, 2.2, 2.20 and 2.19 in approximately a1:2:1 ratio (Scheme 2.8).ZrC13[N(SiMe2CHP’Pr)j + ZrBr3[N(SiMe2CHP1r)]2.1 N [1 Na/Hg 2.42 TolueneMe2Pr’SL Si22<c-? PP2Me2 Si-”Me22.2Pr’Me2Me2I e22.20 2.19Scheme 2.8references on page 112Chapter II 50NC,,Ti1IlIiIjiO.5 iO.O 9.5 PP9.OFigure 2.5 (right) 31P{1H} NMR spectrum of the sample obtained from the reaction describedin Scheme 2.8. (left) 15N NMR spectrum of the same sampleThe solution 15N NMR spectrum of the sample containing the three dinitrogencomplexes has three singlet resonances in approximately 1:2:1 ratio, of which the two lowintensity peaks were assigned to complexes 2.2 and 2.19 (Figure 2.5). The high intensity peakwas assigned to the mixed halogen derivative, 2.20 which, being a singlet resonance suggeststhat the dinitrogen ligand is bound in a side-on fashion in solution. It is theoretically possiblethat the observed singlet resonance of 2.20 is a time averaged signal resulting from a rapidC)CC,,C)01’,1 1111 I I I II III I35C 355 350 345 340 PPMreferences on page iiChapter II 51fluxional process that involves intermediates containing end-on and side-on bound dinitrogenligand. However, the dinitrogen ligand is highly activated in these complexes, and therefore themetal-nitrogen interactions are likely to be strong, which in turn would imply that the rotation ofthe dinitrogen ligand may have a high activation barrier.To confirm that the bromide ligand also favors the formation of a side-on dinitrogencomplex the pure dibromo derivative 2.20 was synthesized from 2.19. A preliminary X-rayanalysis of this complex showed that in the solid state the Zr2N2 core is identical to that ofcomplex Resonance Raman SpectroscopyFurther evidence for the mode of dinitrogen coordination in the zirconium complexeshas been obtained from resonance Raman spectroscopic analysis.67 Resonance Raman spectraof both isotopomers, nitrogen-15 and nitrogen-14, were obtained for solid samples and forsolution samples made up in THF. The most prominent feature in the solid state resonanceRaman spectrum of the end-on complex 2.9 is the peak at 1211 cm-1,which shifts to 1172 cm-’upon substitution of the 14N2 ligand with 15N2 (Figure 2.6, page 52). The magnitude of thisisotope shift, 39 cm1, is essentially equal to the calculated value of 41 cm. The solutionspectra of the two isotopomers are very similar to the solid state spectra, except for the presenceof some solvent peaks. These spectral data imply that the dinitrogen ligand in 2.9 is bound in anend-on manner both in the solid and in the solution states.The solid state resonance Raman spectrum of the side-on derivative 2.2 has two intensepeaks at 317 cm’ and 731 cm-1 and a few low intensity peaks. Substituting the dinitrogenligand with nitrogen-15 isotopomer the peaks at 731 cm1,991 cm and 1046 cm1 shift to 709cm1, 968 1 and 1024 cnr1 respectively (Figure 2.7, page 53). The isotopic shift of 22 cm1observed for the peak at 731 cm1 is within 2 cm1 of the calculated value, suggesting almostpure nitrogen-nitrogen bond stretching character.references on page i2Chapter II 521000The resonance Raman spectra of complex 2.9 recorded at approximately 90 K. Aand B correspond to the (p.-14N2)and (i-15N2)complexes respectively. The toptwo traces were recorded in the solid state and the bottom two were recorded in thesolution (THF) state. Each trace is a sum of 5 scans, each recorded at 2 cm1/susing40 mW of 514. 5 nm laser excitation (Appendix A.1).The resonance Raman studies done on the oxyhemocyanin model complex {Cu[HB(3,5-Ph2pz)3]} ( 112:ri-0)were used as the basis for the interpretation of the resonance Ramanfeatures of complex 2.2.68 It is important to note that both complexes have planar M2X (whereM = Zr and X = N or M = Cu and X = 0) type cores with the X2 moiety bound in an identicalside-on fashion.(N(‘4AC’Lft4S —500Figure 2.61500 cm-i 2000references on pagejChapter II 53The resonance Raman spectra of complex 2.2 recorded at approximately 90 K. Aand B correspond to the (p,-14N2) and (J.t-’5N2) complexes respectively. The toptwo traces were recorded in the solid state (sum of 9 scans) and the bottom two wererecorded in the solution (THF) state (sum of 5 scans). Each scan was recorded at 1cm’/s using 20 mW of 647.1 nm laser excitation (Appendix A.1).The vibrational modes derived from normal coordinate analysis of the Cu20core aredepicted in Figure 2.8. The peak observed at 731 cm1 for the complex 2.2 has been assigned tothe symmetric mode Ag where the dominant stretch involves the nitrogen-nitrogen bondC.’A500—cm-1 1000 1400CoV)THFcqo-I IFigure 2.7I 5O0 8O0 cm-ireferences on page flChapter II 54(Figure-2.8). By comparison, the similar stretching mode in the copper peroxide complex givesa value of 763 cm-1 for the 1602 complex which shifts to 723 cm-’ upon 1802 substitution.CCu,,,_,Cu Cu1 Cu_.Cu CuCu U.Ag Ag B2 B3 BigFigure 2.8 Depiction of the in-plane normal vibrational modes of the side-on peroxidebridged copper dimer.68 Broken lines increase in length as solid lines decrease inlength. Bonds which do not change in length during a given vibration are notshown. Thicker lines indicate the dominant motion in the two Ag modes.68 TheB, deformation mode is not shown.114The other two isotope sensitive peaks of the side-on derivative 2.2, 991 cm-1 and 1046cm4 shift by 23 cm1 and 22 cm-1 respectively, upon substitution with the 15N isotope. Thesetwo peaks are assigned as combination bands where the former is due to the peaks at 258 cm’and 731 cm-’ and the latter is due to the peaks at 317 cm-1 and 731 cnT1. Since the peaks at 258cm1 and 317 cm1 are not isotope sensitive, the shifts in the combination bands can beattributed to the isotope sensitive band at 731 cm1. Overall the spectrum consists of a fewfundamental bands and the rest of the peaks are a result of overtones or combination bands.The solution resonance Raman spectrum of 2.2 was different in some aspects. The peakdue to the nitrogen-nitrogen stretching at 731 cm-1 in the solid state was split into tworesonances at 745 cm1 and 733 cm-1. These two peaks shift to 719 cur’ and 709 cmrespectively, upon substitution with nitrogen-15 dinitrogen ligand. The magnitude of the shifts(26 cm-1 and 24 cm-’) are similar to the shifts observed in the solid state spectra. Also thefeature at 317 cm-1 of the solid state spectrum was split into two resonances at 331 cm4 and 321cur1. The fact that the peaks corresponding to the nitrogen-nitrogen stretching are very close inthe solid and in the solution state spectra strongly suggests that the mode of dinitrogen bindingis very similar in the solid and in the solution states.references on page QChapter II 55It is possible that the appearance of two peaks associated with the nitrogen-nitrogenstretching of the side-on complex 2.2 in the solution (THF) state is due to some solvent effects.However, recording the spectra in a relatively non-coordinating solvent like pentane alsoshowed two peaks. Therefore it is believed that the molecule is undergoing some fluxionalprocess, for example, a process that involves the bending of the planar Zr2N core along thenitrogen-nitrogen axis (Equation 2.1). An isomer with a bent core is comparable to the OAr*derivative, 2.12.zrNzrzrNIIh1zr 2.1Planar Core Bent CoreCuriously, the solution spectrum of the 15N isotopomer of the side-on complex 2.2 had apeak at 1028 cm-1 which was comparable to the dinitrogen stretching of the end-on complex2.9. To ascertain the origin of this peak, polarization studies were carried out on the nitrogen-15isotopomer in solution. It was found that the peaks at 719 cnr1 (p, 0.32), 709 cm1 (p, 0.32),327-i (p, 0.40), 318 cm1 (p, 0.40) and 260 cm1 (p. 0.35) were polarized and the peak at 1028cm1 (p. 0.76) was depolarized. The lower polarization associated with the peaks at 700 cm1and the peak at 260 cm1 suggest that they are due to the Ag modes, where the former peakcorresponds to the nitrogen nitrogen stretching mode and the latter to the breathing mode of theZr2N2 core (Figure 2.8). The peaks around 300 cm’ also share some Ag character. Also thedepolarized peak at 1028 cm1 is not associated with any totally symmetric mode and isprobably due to solvent (THF).The resonance Raman spectra of the sample containing a mixture of three complexes2.2, 2.21 and 2.20 gives only a single peak at 709 cm1 (Figure 2.9). The absence of a center ofsymmetry in the mixed halogen complex 2.18 does not seem to affect the Ag stretching modeassociated with the dinitrogen ligand. This is in agreement with our previous suggestion that thepeaks around 700 wave numbers are almost purely of dinitrogen stretch in character. Thereferences on pageChapter II 56resonance Raman spectra of the aryloxy derivative 2.12 also gives a strong Raman feature at730cm-1.1400Figure 2.9 The solid state resonance Raman spectra of 2.2, (A) and 2.19, (C) and of the sample(B) obtained from the reaction shown in Scheme 2.6. All of the samples contain a(.i-12:1i-’5Nligand. Spectral conditions are similar to Figure 2.8.A compilation of various bond lengths and stretching frequencies associated with anitrogen-nitrogen bond are shown in Table 2.10. The stretching frequencies associated with thedinitrogen ligand in the side-on complexes are significantly lower than what has been observedfor hydrazine, whereas the comparable feature of the end-on derivative, 2.9 is higher; i.e., inthese zirconium complexes the stretching frequencies seems to corroborate well with theobserved nitrogen-nitrogen bond distances. Also these data suggest that the end-on zirconiumderivative has a bond order higher than that of hydrazine whereas the side-on derivatives have abond order of less than one.-I0 SC’%00C’00 AB500 1000 cm-ireferences on page i;iChapter II 57Table 2.10 Compilation of some nitrogen-nitrogen bond lengths and stretching frequenciesassociated with them.Compound Bond length A V cm-1 referencesNN 1.0975(2) 2331 115Me—N=N—Me l.247° 1576 71Ph—N=N—Ph 1.255 1442 71H2N—NH 1.46 1111 71FN—NF 600, 588 71{Ru2(NH3)lo(.t-1’:‘q1-N2)]4 1.124(15) 2050-2 100 72{[(PrPCHSiMeN]Zr(rI5-CHI 2(P-fl1:T -N) 1.301(3) 1211 this work( [(PrPCH2SiMe2)2N]Zr(X) }(.t-r2:T-N2)X=Cl 1.548(7) 731 this workXOAr 1.51(1) 7322.7 Bonding Considerations2.7.1 GeneralThe Dewar-Chatt-Duncanson bonding model developed for the qualitativeunderstanding of metal-olefin interactions has been extended to explain the bonding oftransition metal dinitrogen complexes.73’4 This synergic bonding formalism involves electrondonation from a filled hgand orbital into a vacant metal orbital, and concomitant back donationof electrons from a filled metal orbital into a vacant ligand orbital. In mononuclear end-oncomplexes, the lone-pair orbitals of the end-on bound dinitrogen ligand act as the donor orbitalsand the vacant 7t*orbitals act as the acceptor orbitals.75 The bonding in mononuclear side-ondinitrogen complexes can be rationalized by invoking the filled it-orbitals as the donor orbitalsand the vacant t*orbitals as the acceptor orbitals.76To explain the bonding in dinuclear metal complexes having an end-on bridgingdinitrogen unit, M—NN—M, Chatt et a!. proposed the qualitative molecular orbital (MO)analysis depicted in Figure 2.10.77,78 For this four-center M—N—N—M interaction, the atomicreferences on page U.QChapter II 58orbitals (AO) of each center (i.e., of both the metals and the two nitrogens) contribute equallytowards each MO.NJDonor-Acceptor interactions for Donor-Acceptor interactions forthe mononuclear end-on mode the mononuclear side-on mode‘3’’9 4e8;i8 j. 3jj1i8!3.Ij 2eieA qualitative MO bonding scheme A more realistic representation forfor the end-on bridging mode the case described on the leftFigure 2.10 Qualitative bonding description for mononuclear complexes (top) and end-onbridging dinuclear complexes (bottom).Molecular orbital calculations performed on structurally known complexes give aslightly different picture.79 Fenske-Hall type calculations performed on the end-on dinuclearcomplexes, { [Ru(NH3)S]2(i-N2} 4, { [Nb(CH2)(C3(PH3)2]2(.L-N2) } and (Cp*2Zr N2)2-(p.-N2), 2.3, suggest that the MOs labelled 3e for all of the .I-11:fl-N2 complexes are eithernon-bonding or weakly bonding in character so that their occupation cannot affect the nitrogennitrogen bond length.79 Thus, the character of the 2e MOs must control the strength of thereferences on pageChapter II 59metal-nitrogen and nitrogen-nitrogen bonds. The AO composition of the 2e MOs varyconsiderably for different complexes. It was found that the participation of the dinitrogen*bital in the 2e MOs was greater for the zirconium and niobium cases and was less for theruthenium case, which also corroborates with the longer nitrogen-nitrogen bond distancesobserved in the zirconium and niobium complexes. For the niobium and zirconium complexes,{ [Nb(CH2)(CH3)(PH3)2]2(p-N2) } and (Cp*2ZrN2)2(.LN2), a push-pull type of bondingmechanism has been postulated that parallels the synergic bonding in mononuclearcomplexes.21’79 In these examples, of the two d-orbitals (one from each metal) that contributetowards the 2e MOs, only one is filled and the other is vacant. As a result, metal to dinitrogenM—*N2,and dinitrogen to metal N2—*M, electron flow can occur without significant chargebuild up on the dinitrogen ligand.21’792.7.2 Bonding in {[(Pr’2PCHSiMe)2N]Zr(15-C5H)}2(J.t-NINDO/l-MO calculations were performed on the idealized metal complex[(II3P)(HN)Zr(r5-CsHs)]Q.t-ri’:ri’-N A, which was restricted to C2, symmetry.45 Thefragment MO analysis was carried out on the idealized fragment[(H3P)2(H2N)Zr(T15-CS S)] Bwith an imposed C symmetry. In these models, the relative spatial arrangement of the atomsdirectly bonded to zirconium was kept very close to that of the X-ray structure.45 In order tominimize the total number of AOs associated with the model the isopropyl groups on thephosphine donors were replaced with hydrogen atoms. The isopropyl substituents are importantin view of their steric contribution but in a semi-empirical level of calculations they are unlikelyto have a significant influence in the electronic properties associated with the metal.H3 I HZNN—Z” H ....Zrr 4piø’H H3P H3P A H HA Breferences on page flChapter II 60Figure 2.11 Bonding scheme for the model A.From the dinitrogen ligand the lone-pair orbital 3Gg, the bonding tu-orbita1s and theantibonding 1t*orbitals are available for bonding. MO analysis of the fragment B indicated thatthere are two metal-based orbitals a’ and a” which are available for bonding with dinitrogen.These frontier orbitals a’ and a” mainly consist of d2 (52%) and dx2y (68%) respectively(Table 2.11).* The d2 lies along the axis containing the metal and the centroid of the Cp ligand,and is therefore higher in energy than thed22, which lies between the phosphine and theamide ligands.** The and dyz orbitals are involved in bonding with the Cp ligand and thedxy-orbital is directed towards the ancillary ligand atoms. This formalism is consistent with the* The total d-orbital contribution towards the frontier orbital a’ (-0.674 eV) is 63% and a” (1.556 eV) is 85%.Orbital a’ has a significant zirconium p-orbital (Pz47%) contribution as well.** If the symmetry of the major d-orbital contributing towards a frontier orbital agrees with the adopted coordinatesystem then that frontier orbital will be referred to by the symmetry symbol of the d-orbital.xy: o..4% 88á,::,88N2—it,2 2LiX y2(H3P)N)CpZr4S44%4%4%444%4%4%44%4SS8 8w”ereferences on page 112Chapter II 61isoelectronic and qualitatively isolobal fragment[(C5Me)WMe3]*SO where the two isolobal a’and a” orbitals mainly consist of d2 and orbitals.M,SlTable 2.11 Some orbital parameters of le and 2e MOs of model complex A.Zr d-orbital contri- N2 p-orbital contriMolecular orbital Energy (eV)bution per metal bution per nitrogenHOMO (2e) -5.824 10% 23% (Py = 18%)HOMO-1 (2e) -6.594 24% 21% (P = 14%)HOMO-9 (le) -11.065 8% 26% (Pz 17%)HOMO-li (le) -12.363 9% 24% (Py= 19%)The bonding scheme for model A is shown in Figure 2.11. The symmetric combinationof thed2..-orbita1 with one of the ic-orbitals of dinitrogen gives rise to HOMO-9 and thesymmetric combination of thed2-orbitals with the second ic-orbital of dinitrogen gives rise toHOMO-il. These two MOs correspond to the le MOs depicted in Figure 2.11 and consistmainly of ic-N2 in character.* In a similar fashion the anti-symmetric combinations of the d2and orbitals can interact with the it*orbitals of dinitrogen to give the 2e MOs, whichcorrespond to HOMO and HOMO-1 respectively. The relative energies and orbitalcontributions (for one Zr and one nitrogen atom of the dinitrogen ligand) of le and 2e MOs aregiven in Table 2.11. With regard to the AO contributions towards the 2e MOs, the HOMO has asignificantly greater dinitrogen p-orbital character than the d-orbital character of zirconium,which in turn suggests that the occupation of the HOMO would lead to a charge build up on thedinitrogen ligand. The le and 2e MOs are occupied by eight electrons, four of which are fromthe dinitrogen and the remainder from the metal orbitals where each metal formally contributes* The CpZr(PH3)2(NH fragment deviates from being ideally isolobal to the (C5Me5)2WMe3fragment due to theit interactions of the amide lone pair with the metal d-orbital. Both fragments belong to point group.* Since these two MOs have different energy they are not strictly le MOs. They are comparable to le MOs only onthe basis of the number of nodes and the type of interactions between the metal and N2. Similar reasoning isadopted for the 2e, 3e and 4e MOs.references on page 212Chapter II 62two electrons. The unoccupied 3e MOs (LUMO+2, 0.602 eV and LUMO+4, 1.392 eV) aremainly metal d-orbital in character, i.e., these MOs can be considered as virtual non-bondingorbitals. In principle therefore, addition of electrons into the 3e MOs (i.e., further reduction of2.9, e.g., electrochemical reduction) is unlikely to lead to any significant activation ofdinitrogen.The values for the Wiberg indices82-4 (Table 2.12) clearly show a weak nitrogen-nitrogen bond in A; by comparison, the calculated bond indices of diazenido (N2)2- andhydrazido (N2)4-ligands give rise to bond orders of 2 and 1, respectively.85’6 Therefore thecalculated bond order of 1.15 for the nitrogen-nitrogen bond in A, agrees well with theformalism of the dinitrogen ligand as a hydrazido (N2)4 ligand and is also in line with theresonance Raman data where the observed nitrogen-nitrogen stretch for the end-on derivative2.9 was approximately 100 cm1 higher than that of hydrazine. Also the calculated bond orderof 2.18 for the zirconium-nitrogen bond of the Zr2N2 core supports the formulation of the Zr2N2core as a Zr=N—N=Zr unit. The population analysis shows a net donation of 0.22 electronsfrom the dinitrogen lone pairs (on the basis of a decrease of electron density of the nitrogen sand Px orbitals). It also shows a net gain of 0.61 electrons into the 1*orbitals The calculatedtotal charge on the dinitrogen ligand is -0.76 (-0.38x2).Table 2.12 Bond indices and population analysis of model complex A.Wiberg Bond Index Mullilcen Orbital Populations of N2 per nitrogenBond Index Orbital N2 complex free N2(N—N = l.096A)Zr—N (N2) 2.18 s 1.684 1.906Zr—N (arnide) 1.00 Px 1.086 1.096N—N 1.15 P,, 1.276 1.000Zr—P 0.63 P 1.333 1.000references on page 2iChapter II 632.7.3 Bonding in Side-On ComplexesDue to the lack of experimental evidence for side-on bound dinitrogen complexes, only afew publications have addressed the issue of bonding in these complexes. Theoretical studieson some mononuclear dinitrogen complexes have shown that the nitrogen-nitrogen bond isweakened (i.e., on the basis of longer nitrogen-nitrogen distances) to a greater extent in theside-on mode of coordination as compared to the end-on mode.76’87 For example, the simpleinteraction of a CH2 fragment with dinitrogen in an end-on (diazomethane) fashion versus thecorresponding side-on form (diazirine) showed a greater weakening of the dinitrogen bond inthe latter case.87 Also, calculations done on model complexes [Ru(NH3)5(N2)]2and[CoH(PH3)(N2]indicate that the metal nitrogen bond strengths are stronger for side-on cases,whereas the nitrogen-nitrogen bond strengths (i.e., calculated bond energies) were greater forthe end-on cases.It has been suggested87 that the paucity of side-on complexes may be attributed to theoverwhelming destabilization of the nitrogen-nitrogen bond upon complexation which in turn isnot stabilized adequately by the formation of the metal-nitrogen bond. In the case of the modelcomplex [Ni(PH3)2( ]the SCF binding energies are calculated to be -15.1 kcal mol1 and-12.6 kcal mo11 for side-on and end-on cases respectively, favoring the side-on mode. It wasalso shown that the nitrogen-nitrogen bond was more elongated in the side-on case and also thenegative charge on the nitrogens of the dinitrogen ligand was greater (0.30 electrons for side-onversus 0.02 electrons for end-on), implying increased ic-backdonation for the former.76Conversely, calculations done on a different nickel complex, [Ni(02)(N2)Jshowed that the endon mode of coordination is favored.88Hoffmann et a!. have performed MO calculations on dinuclear complexes containingside-on bound dinitrogen unit where the M—N2--M core adopts a bent geometry (Figure 2.12).They considered the cobalt complex C as a possibility and the nickel complex D as a model forexample 1.10 (Chapter 1).89 The two important It-acceptor interactions of the bridgingdinitrogen are shown in Figure 2.12. The in-phase and out-of-phase combinations of thefragment orbitals overlap with the two 7t*orbita1s to give two It MOs, bi and a. Also, thereferences on page flChapter II 64EH-MO calculations on C and D gave a substantial energy separation (3 eV) between theHOMO and LUMO for the side-on mode, whereas similar energy separation for the parallelend-on case E was only 0.02 eV. These energy values clearly indicate higher thermodynamicand kinetic stability for the side-on modes.89OC ,N=N\ /CO.Co Co.,oc”J V”coOC E COFigure 2.12 Bonding scheme (in the box) describes the edge-on bridging dinitrogen complexesof Co and Ni.2.7.4 Bonding in{[(Pr’2PCHS1Me)N}ZrCI}(ii-NH IP— Zr—RH’ f HGH HUsing the INDO 1/MO level, semi-empirical calculations were performed on this side-oncomplex using the idealized form {(H3P)2(N)ZrCl1.t-N),F, with an imposed C2hsymmetry.45’62 The fragment MO analysis was carried out on the species [(H3P)2(N)ZrC1],G, with an imposed C symmetry.45’62 In these models, the relative spatial arrangement of theatoms directly bonded to zirconium was kept very close to what was observed in the X-rayreferences on page flCit-MO b1it-MO a2Ni.D HH3PFChapter II 65structure.8 The isopropyl substituents are important in view of their steric contribution but in asemi-empirical level of calculations they are unlikely to have a significant influence in theelectronic properties associated with the metal. The SiMe2 groups associated with the amidewere also replaced with hydrogens, and a comparison of the theoretical electronic data for F andthe silyl substituted analogue,{(H3P)2[SiN)]ZrC1}(j.t-Ngave identical results.The fragment orbitals la” and 2a” of G have the appropriate symmetry to form a it-MOand a s-MO, respectively. The frontier orbital la’ has 84% d-character (55%, d2; 23%,d22and 6%, d) and 2a” has 80% d-character (38%, d,,; 25%, and 16%, d). In the case of G,due to the extensive mixing of the d-orbitals the frontier orbitals la” and 2a” will be referred toas virtual-d, and virtual-d orbitals* respectively (Table 2.13).Table 2.13 Some orbital parameters of it and MOs of complex F.Molecular orbitalHOMO () -6.937 12%HOMO-9 (it)-10.329 20%Zr d-orbital contri- p-orbital contribution N2Energy (eV)bution per metal per nitrogen34%(Px 12.4%;pz= 15%)27%(Pz=5.7%;Py21)From a symmetry point of view the acceptor interactions of the dinitrogen 7c*orbitalswith the metal in the side-on case F differ considerably from that of the end-on model A. In Fthe 7c*orbitals in the plane of Zr2N core overlap with a virtual d-orbita1, which is alsocontained in the same plane to give a ic-MO (Figure 2.13). This MO has 27% zirconium-d and20% dinitrogen-p character, which is similar to the orbital contributions calculated for theHOMO-1 of the model complex A (Table 2.13). But, an important difference is that the it-MOof F is much lower in energy than the it-MOs of A. This is probably due to the fact that thedinitrogen it*orbitals overlap better with the metal d-orbitals in the side-on mode than in the* These frontier orbitals are referred to as virtual orbitals because they are assigned according to the adoptedcoordinate system and not on the basis of which atomic orbital contributes the most.references on pageChapter II 66end-on mode. By comparison, the overlap values calculated for the t-type interaction betweenthe 7t*orbital of dinitrogen and titanium d-orbitals give a value of 0.227 for the side-on bondingand 0.155 for the end-on bonding.9°The out of plane 1t*orbitals of the dinitrogen in F overlap with a virtual d-orbital,which is contained in a parallel plane to form a 6-MO (Figure 2.13). This MO has 12%zirconium-d and 34% nitrogen-p character and suggests that there is considerable ionic characterwith electron density mainly located on the nitrogens of the dinitrogen ligand. A survey of thelow lying LUMOs of F (up to LUMO+5) showed that they are mainly non-bonding metal-basedorbitals.Figure 2.13 Bonding scheme illustrating the types of overlap leading to the formation of theit-MO and the 6-MO.references on page ii2Chapter II 67The large energy gap (6.4 eV) between the HOMO and LUMO is consistent with theobserved stability of this complex. When the hypothetica:1 end-on complex* having identicalligand arrangements around the zirconium core, as that of F, was subjected to INDO/1-MOanalysis the HOMO-LUMO gap (6.3 eV) was found to be very close to the gap for F.Therefore, an analysis on the basis of HOMO-LUMO gap does not favor the side-on mode overthe end-on.89 However, the SCF energy value does favor the side-on case F, by greater than1 eV.-103.15-103.35-103.55-103.75-103.95-104.15-104.35Figure 2.14 Plot showing the change of energy in the model complex F with the rotation of thedinitrogen ligand.To evaluate the barrier for the conversion of the side-on complex to the end-oncomplex, the dinitrogen ligand was rotated in a stepwise manner on the plane of the Zr2N coreuntil the dinitrogen ligand became coordinated end-on. The angle of rotation 9 is depicted in* TheZr2N core of F was changed to a linear end-on mode, Zr—N—N—Zr; Zr—N = 1.92 A, N—N = 1 .30A. Bondlength parameters of the ligands were kept identical to F.0 20 40 60 80Angle of rotation (°)references on pageChapter II 68Figure 2.14 where it was increased from 0° to 90°. During rotation the closest distances betweena zirconium center and a nitrogen of the dinitrogen ligand were kept constant. The curveobtained from the plot of e vs. energy (at the SCF level) suggests that the barrier for the rotationfrom an end-on to side-on coordination is greater than 1 eV. More importantly, from athermodynamic viewpoint the shape of the curve suggests that this particular end-on complexshould spontaneously convert to the side-on complex. It is of interest to point out that ab initiocalculations done on naked dinuclear side-on dinitrogen complexes of the type M2(t-i1:fl-N2),where M = Ti, Zr and Y, suggested that zirconium and yttrium favor a side-on bound dinitrogenligand. The calculated nitrogen-nitrogen bond distances for the ligand-free zirconium case(Zr2N2) was 1.48 A, which was only 0.07 A shorter than the experimental value 1.55 A foundfor complex 2.2. 1In the model F, the calculated value of 0.85 (Table 2.14) for the nitrogen-nitrogen bondorder, is less than the value of 1.15 found for the end-on case A, and this corroborates well withthe resonance Raman data of complex 2.2, where the observed nitrogen-nitrogen stretchingfrequency was approximately 300 cm4 lower than that of hydrazine. But, comparison with theobserved differences in the nitrogen-nitrogen bond lengths of end-on and side-on complexes(e.g., 2.2 versus 2.9) the differences in bond indices are not significant. By the definition ofbond order, i.e., the net bonding electrons are equal to twice the value of bond order, in goingfrom models A to F the number of bonding electrons associated with the dinitrogen ligand havedecreased by 0.6. This would imply greater backbonding in the side-on complexes and isconsistent with other theoretical studies done on side-on complexes. The overall bond orderbetween the dinitrogen and zirconium in the side-on case was found to be 2.38 (2x1.19), andsuggests stronger metal dinitrogen interactions than in the end-on case. The calculated bondorder between the amide nitrogen and zirconium was found to be greater than 1, implyingsignificant t interactions. The significance of this interaction will be discussed later.references on page 212Chapter II 69Table 2.14 Bond indices and population analysis of model complex F.Wiberg Bond Index Mulliken Orbital Populations of N2 per nitrogenBond Index Orbital N2 complex free N2(N-N = 1.548 A)Zr—N (N2) 1.19 s 1.845 1.937Zr—N (amide) 1.13 Px 1.152 1.063N—N 0.85 P,, 1.284 1.000Zr—P 0.61 Pz 1.260 1.000Population analysis of F shows only a small decrease in the s-orbital population ofdinitrogen, but the populations of all the p-orbitals have been increased. The calculated formalcharge of -1.08 on the dinitrogen ligand is significantly higher than what was found for theend-on species A (-0.76). It is possible that the increased coulombic repulsion (due to the highercharge) between the two nitrogens of the dinitrogen ligand and weak bonding interactions (onthe basis of Wiberg indices) are important factors contributing to the long nitrogen-nitrogendistances observed in the side-on complex 2.2. With regard to this coulombic repulsion, thechanges in nitrogen-nitrogen bond lengths in neutral hydrazine and its protonated derivatives areimportant. The observed nitrogen-nitrogen bond length decreases consistently upon successiveprotonation of hydrazine, i.e., the nitrogen-nitrogen bond lengths for H4N2, (H5N2)Br, and(H6N2)S04are 1.47 A, 1.45 A and 1.40 A respectively. This decrease in bond length has beenattributed to the removal of the highly repulsive lone-pair interactions by protonation.692.7.5 Bonding in {[(Pri2PCHSiMe)N]Zr(OAr*)}(LNUsing the INDO 1/MO level, semi-empirical calculations were performed on this side-oncomplex using the idealized form ((H3P)2N)Zr(OH) }2(.t-N2), H, with an imposed C2symmetry. The relative positions of the ligands around each zirconium are within 10 of theactual parameters found in the solid state structure. When model H was subjected toINDO/l-MO analysis, the MOs changed considerably from those found for F. Mostreferences on page flChapter II 70importantly, the it-MO associated with the dinitrogen ligand was split into two separate MOs,which also showed some d-p type interactions with the lone-pairs of the oxygen. To evaluatethe cause of this change the model J was constructed as follows: the bent core of model H wasmade planar and one zirconium was rotated by 1800 with respect to the other and then the OHgroups were replaced by chloride (Zr—Cl = 2.5A). Model J now mimics model F and they differonly in their angle parameters associated with the zirconium and the ancillary ligands.HH 2.5A o’Step 1 \ PHa /..PH3Zr__NZrReplacement of -N/\ClwithOHH3P / H3P0H KHI-I H0PH3 0H \H3P N...,H PH3H HScheme 2.9Then model J was changed to model H in a stepwise manner, as described in Scheme2.9; in each case the MOs were compared to the previous case. The results of this step-wiseanalysis are shown in Table 2.15. The SCF-energy values of the models J and K were withinHH\ PH3JStep 2Adjustment of Zr—Obond to 2.02ABending of the2N coreStep 4HPH3H \:N/H PH3HStep 3N.., Rotation of oneH Zrcenterby 180°MN..,HLreferences on page 222Chapter II 710.02 eV, and the bonding interactions associated with the dinitrogen ligand, i.e., the it and theMOs were similar to F.Table 2.15 SCF energy values and the energies of some MOs of models H, J, K, L, M and F.Model Energy (SCF) s-MO it-MOJ -104.070 eV -6.830 eV (HOMO) -10.229 eV (HOMO-9)K -104.046 eV -6.640 eV (HOMO) -10.049 eV (HOMO-7)L -104.597 eV -5.924 eV (HOMO) -9.213 eV (HOMO-3)-11.070 eV (HOMO-8)M -104.597 eV -5.930 eV (HOMO) -9.190 eV (HOMO-3)-11.001 eV (HOMO-9)H -104.616 eV -6.093 eV (HOMO) -9.019 eV (HOMO-3)-11.041 eV (HOMO-9)F -104.306 eV -6.937 eV (HOMO) -10.329 eV (HOMO-9)The shortening of the zirconium-oxygen bond of K gave model L, which was stabilizedby 0.6 eV than models J and K. Analyzing the contours of the MOs corresponding to L showsthat the zirconium-dinitrogen bonding had significantly changed. The it-MO has been split intotwo separate MOs and their contours suggest a d-p type interaction between the zirconiumorbitals and the oxygen lone-pair orbitals. These interactions are depicted in Figure 2.15, whichalso qualitatively explain the relative contributions of different AOs to each MOs (also seeTable 2.16). The lower energy it-MO was largely oxygen p-orbital in character with a bondinginteraction with the metal orbital. The higher energy it-MO was mainly a metal dinitrogenbonding interaction with antibonding character between the metal and oxygen orbitals. Furthermodifications from L to H do not seem to cause any major changes in the type of overlaps or inthe total energies of the models.references on pageChapter II 72Figure 2.15 Splitting of the it-MO in complex H.Table 2.16 Some orbital parameters of model H.Molecular Energy (eV) Zr d-orbital N2 p-orbital 0 p-orbitalorbital contribution contribution per contribution perper metal nitrogen oxygenHOMO (6)-6.098 9% 36% (Px = 21%; < 1%Py= 14%)HOMO-3 (it)-9.018 17% 22% (Py = 18%) 4%HOMO-9 (it)-11.041 5% 3% 31% (Pz = 23%;Px8%)H2NH3P H3PHOMO-3H3PH3PH3P H3P HOMO-8Oxygen2p-orbitalH2N,H3Preferences on pageChapter II 73The metal and dinitrogen orbital contributions of HOMO and HOMO-3 are similar tothat in model F (Table 2.17). Comparison of bond orders associated with similar bonds ofmodels F and H show that the zirconium-amide bond and the zirconium-phosphine bonds areslightly weaker in H than in F.* A bond order of 1.16 between the zirconium-oxygen bond,comparable to the bond orders between the zirconium and nitrogens of the dinitrogen ligand,confirms the presence of p-d bonding, which was previously suggested on the basis ofstructural features. The bond indices associated with the Zr2N2 core are also comparable tomodel F. For H, the indices indicate a weakening of the zirconium-aniide bond compared to thesame indices of model F. This could probably be due to the increased it interactions of theoxygen lone pairs since the bond order between the zirconium and oxygen was greater than one.Table 2.17 Wiberg bond indices and Mullilcen orbital populations of model H.Wiberg Bond Index Mulliken Orbital Populations of N2 per nitrogenBond Index Orbital N2 Complexfree N2(N-N = 1.548 A)Zr—N(N2) 1.17 s 1.85 1.937Zr—N (NH2) 1.07 Px 1.25 1.063N—N 0.87 Py 1.22 1.000Zi—P 0.55 Pz 1.23 1.000Zr—O 1.16Alkaline salts of chloride and hydroxide ions were analyzed to estimate the relativeenergies of the lone-pair electrons of the anions. It was assumed that in these alkaline salts therewould be no it-type interactions, and therefore this enables the estimation of the energies of thelone-pairs. Calculations suggest that the oxygen lone-pairs are relatively lower in energy thanthose on the chloride. Comparison of these energies with the it-MO of F shows that the oxygen* The ZINDO output files gives the values of Mulliken populations and Wiberg bond indices to six significantfigures.references on page flChapter II 74lone pairs are the closest in energy (Table 18). The calculated it overlap values for F were alsogreater for oxygen p-orbitals than those of the chloride porbitals* . These factors are likely tocontribute towards the mixing of the it-MO with the oxygen lone pair resulting in the formationof two it-MOs. When the hydroxyl ligand of model L was replaced with a chloride ligandwithout altering the zirconium-chloride distance (2.02 A), analysis gave almost identical resultsto those obtained for L. This could be explained on the basis of the lowering of the energy ofthe chloride lone pair (with the shortening of distances) and also increasing the overlap with thed-orbitals.Table 2.18 A comparison of the energies of the lone pair electrons of Cl, NH2 and OH ions.Salt Distance Lone pair Salt Distance Lone pairEnergy (eV) Energy (eV)LiNH2 2.20 A -7.789 LiC1 2.50 A -8.548LiOH 2.02 A -9.344 NaCl 2.50 A -8.793LiC1 2.02 A -9.002 NaOH 2.02 A -9.391The angular overlap model92 suggests that the frontier orbital requirements for a side-onbound dinitrogen ligand with a bent core can be different from those of a planar core. Withreference to the planar case depicted in Figure 2.16, the frontier orbitals shown can only overlapwith one 7t*orbital of the dinitrogen, which is also contained on the same plane, whereas theother 7c*orbital of dinitrogen is perpendicular to the plane and will require a different metalbased orbital for bonding, for example see the bonding scheme for the planar side-on model Fdepicted in Figure 2.13. It is also important to note in Figure 2.13, that each set of frontierorbitals (i.e., the set corresponding to the it-MO and the set corresponding to the ö-MO), canhave a symmetric and an anti-symmetric combination, of which only the anti-symmetriccombination is involved in bonding. In the absence of bonding, i.e., in the absence of a* Calculated overlap integerals for Zr—Cl, 2.50 A; Zr—Cl, 2.02 A and Zr—OH, 2.02 A are 0.087,0.192 and 0.110respectivly.referenced on pageChapter II 75dinitrogen ligand in between the two metal centers, these two combinations corresponding toeach set will be degenerate (i.e., in Figure 2.13). In fact, during the MO analysis of models Fand M the symmetric combinations were found to be non-bonding, because they do not have asuitable dinitrogen based orbital for bonding.Figure 2.16 The angular overlap for the side-on bonded dinitrogen with a bent Zr2N core.Anti-symmetric combinationof the frontier orbitals Planar coreSymmetric combinationof the frontier orbitalsH2N NHBonding interactions with N2 Non-bonding interactions with N2Bendingof,ll,Zr2NcoreAnti-symmetric combinationof the frontier orbitalsH2iH3P’H3P’Symmetric combinationof the frontier orbitalsH2N N4> NH2H3Pfl.:...11PH3Bonding interactions with N2 Bonding interactions with N2Open three centre bond I IL jreferences on page 112Chapter II 76Figure 2.16 illustrates the bonding of a side-on bound dinitrogen ligand where the Zr2N2core is bent. It can be seen that the bonding in this case can be rationalized by using only oneset of frontier orbitals, where both, the symmetric and anti symmetric combinations are used inbonding with the dinitrogen ligand.89 The interaction of the anti-symmetric combination (i.e., inthe bent mode shown on the right) with the dinitrogen *orbita.l is comparable to an openthree-center bond whereas the interaction of the symmetric combination is comparable to aclosed three-center bond.To evaluate this hypothesis the model M was bent in a step-wise manner using thenitrogen-nitrogen bond as a hinge. To perform this calculation, a phantom atom X was placedin the middle of nitrogen-nitrogen bond and the bend angle was measured by taking the Zr—X—Zr angle, e. Calculations were done for different models which were constructed by decreasingthe Zr—X—Zr angles by approximately 5° from that of the planar case M (i.e., decreased from180°), and for each case the SCF energy and the contours of 13 HOMOs (from HOMO toHOMO-12) were analyzed. The contours of the models with values of 0 greater than 150° showthe involvement of two different frontier orbitals (i.e., similar to the bonding described formodel H) with the 7u*orbitals of dinitrogen. These interactions are shown in Figure 2.17. It isimportant to note that the &-interaction is weakened above the Zr2N2 core and is strengthenbelow the core. For smaller values 0, i.e., less than 120°, the contours seem to suggest theinvolvement of only one frontier orbital (i.e., from each metal) as described in Figure 2.16.Because of extensive mixing of atomic orbitals during the ZINDO/1-MO analysis a moredefinite evaluation was not possible.Figure 2.17 The orbital interactions for a bent Zr2N2 core of complex 2.12 where the bendH2NPH3HOangles are greater than 150°,HO OHreferences on pageChapter II 77The change in the SCF energy values of the models with different values of 8 is shownin Figure 2.18. The curve suggests that the bending of the planar Zr2N2 core, even by a smallangle (5°) increases the energy of the model by approximately 0.1 eV, but upon further bendingthe energies of the model decreases. The energies become comparable to the planar model Hwhen the core is bent by about 135°. In fact, the shape of the curve suggests that continuedbending produces even more stable geometries, although presumably steric factors will becomedominant as the angle decreases.-104.655:’I-104.75-104.85 •115 135 155 175Bending Angle 9 (°)Figure 2.18 The change in SCF energy with respect to the bending of the Zr2Ncore of modelM. The core was bent along the nitrogen-nitrogen axis.2.7.6 Factors Influencing the Side-On Mode of CoordinationOn the basis of the discussion presented on dinuclear dinitrogen complexes in Chapter 1it is clear that the precedence for the formation of end-on bonded complexes is overwhelming.references on page 112Chapter II 78The bonding description that has been presented so far has been mainly concerned with theanalysis of the MOs of complexes 2.2, 2.9 and 2.12 and how they relate to their structuralfeatures. So far no attempts have been made to address the question that why complexes 2.2and 2.12 did not form the end-on complexes, despite the overwhelming experimental evidencefor their formation.Figure 2.19 Bonding scheme illustrating the symmetry based orbital requirements for thedinitrogen binding in bridging side-on and end-on modes involving ML4fragments.In an attempt to evaluate the electronic factors which favored the formation of side-oncomplex 2.2 a simple, back of the envelope type MO analysis was considered (Figure 2.19).During analysis two ML4 fragments having C4v symmetry, were arranged to satisfy an overall:. •3ed LM L4MN2 \_______END-ON •.,/ d,dlt*lt*7tpN2SIDE-ON___1ti 2t:.: •d2o‘flifItreferences on page hfChapter II 79D4h symmetry, with ligands pointing away from the center of metal-metal axis (Figure 2.19).The metal-centers were separated by sufficient distance, such that a dinitrogen ligand could beinserted in an end-on or side-on manner.Thed2-and d2 orbitals which point toward the ligands are used for a-bonding withligands L of ML4 and also with the clinitrogen. The remaining three d-orbitals, the andpointing in between the ligands, are available for bonding.8° Since it is already establishedthat the it acceptor interactions of dinitrogen are the most crucial, only the bonding ofit*orbitals of dinitrogen with the metal d-orbitals are considered.79’90 Simple analysis showsthat the end-on mode results in the formation of two degenerate it-MOs which correspond to the2e MOs described previously (Figure 2.10). In the side-on mode the dinitrogen 1t*orbital in theplane of M—N2—M will form a it-MO; however, the it*orbital perpendicular to the plane willform the S-MO, with the it-MO is placed lower in energy.90 Although this relative positioningof the MOs agree with the conventional wisdom that it-orbitals are more stable than 5-orbitals, acomparison of the it-MOs of the end-on case, A and the side-on case, F show that the it-MO ofside-on is much lower in energy. A rationale for the formation of lower energy it-MOs in theside-on case is that the respective overlaps of the AOs leading to the formation of the it-MO isgreater for the side-on mode than that of the end-on mode. By comparison, the overlap valuescalculated for the interaction of naked titanium metal with dinitrogen ligand are 0.227 and 0.155for side-on and end-on cases respectively.90In the two analyses described in Figure 2.19 only two of the three available d-orbitals areused in bonding. In the end-on case the two dy-orbitals and in the side-on case the twod,-orbitals are non-bonding. Therefore, any it-type interactions with one of these three metald-orbitals with an appropriate ligand orbital (excluding the dinitrogen hgand) will leave onlytwo d-orbitals available for bonding with the 1t*orbital of dinitrogen. In the case where theorbital is not available then only the end-on complex can be formed. If the or orbital isnot available then the result would be the formation of a side-on complex.When the HOMOs of F were carefully analyzed at the INDO 1/MO level, the orbitalHOMO-8 (Figure 2.20) was intriguing in that one of the d-orbitals on the metal was involved inreferences on page flChapter II 80it-bonding with the aniide lone pair and a-bonding with the chloride ligand. Principally, this isone of the non-bonding d-orbitals described in Figure 2.19, which according to the symmetrybased analysis can only form a it-MO. Therefore, due to the formation of HOMO-8 theremaining d-orbitals can only form one it-MO and one s-MO resulting in the formation of theside-on complex. In the model complex H the d-orbital that is required for the second it-MOwas it-bonded to oxygen and a-bonded to the amide.PH3 H3Pg:iN$:dPH3PH3 HsPNJ PH3YHH 4’HH HHOMO-8 in F HOMO-12 in HFigure 2.20 Important it-type interactions that determine the type of dinitrogen bonding inmodel complexes F and H.To vindicate this claim the MOs of the fragment G and those of the fragment N,obtained by rotating the plane of the amide by 90°, were analyzed. In Figure 2.21 the fragmentMOs corresponding to G (on the left) and those of N (on the right) are arranged such that theirenergies decrease in going from top to bottom. For both cases the set of orbitals la” and 2a”will favor the side-on mode and the set la” and la’ will favor the end-on mode of bonding. Thecalculated energy for the Ia’ orbital of G was found to be much higher (37 kJ moll) than thenearest orbital 2a”. The la orbital of G is raised in energy partly due to the amide d-pinteractions and also due to the interactions with the chloride ligand. Therefore, on the basisof symmetry the available orbitals that will lead to the side-on complex are la” and 2a”.When the plane of the amide in model G was rotated by 90°, the model N was obtained.Calculations on model N show that, due to the removal of the d-p interactions, the orbital la’was lowered in energy. Also, due to the newly imposed d-p interactions, the frontier orbital2a” is raised in energy. It was found that orbitals la’ and 2a” were only separated by 0.5 kJ0H3Preferences on page flChapter II 81mo11. Therefore la’ can form two sets of orbitals (i.e., la and la” or la’ and 2att) which areenergetically similar, where each set will lead to different mode of bonding. The case leading tothe end-on mode will generate two it-MOs, while the other will generate a t-MO and a 6-MO,where the formation of two t-MOs will be energetically more favorable. Therefore in theorythe fragment N should lead to an end-on complex.Figure 2.21 The frontier orbitals of the fragments G and N arranged in increasing order oftheir energySupport for this analysis can be found in two recent X-ray structures of relatedisoelectronic dinuclear titanium dinitrogen complexes, shown in 2.27 and 2.28, both of whichdisplay the bridging end-on mode.57’8 In 2.27, having cis disposed neutral amine donors ratherthan trans disposed phosphine ligands, the major difference is that the bis(trimethylsilyl)amideancillary ligand is arranged perpendicular to the plane defined by the donor atoms; this is alsofound in 2.28 but now the neutral pyridine donors are trans disposed. Both of these complexesare similar to that proposed in model conformer N and their structures are a reasonableconsequence of the fact that both 2.27 and 2.28 have no conformational restrictions to force theHGcv4 H1 a’HNCi NHP2 a” fr1 a’2 a”1 a”NH21 a”ci—1Y NH2Preferences on page flChapter II 82amide ligand to bond in a manner as is found in the chelating amido-diphosphine ligand of theside-on derivative 2.2.SiMe3Me3Si/CII’pyN :-=- NCI N....SiMMe3Si2.28This rationale could be also extended to explain the formation of the end-on complex2.9. The replacement of the chloride ligand of 2.2 by a cyclopentadienyl group has altered theavailable d-orbitals for dinitrogen bonding. Analysis of the molecular orbitals of the fragment Bshows that the d-orbital that would be used to generate the 6-molecular orbital for a side-ondinitrogen fragment overlaps with the ic-orbitals of the cyclopentadienyl ligand. Also animportant structural feature that correlates well with the electronic description is the zirconiumamide bond distances. Since the fragment orbital that has the correct symmetry to overlap withthe amide p-orbital is involved in it-bonding with the dinitrogen ligand, one would expect a longzirconium amide bond length (see section 2.2.2). It is noteworthy that in the solid-statestructure, the NSi2 plane in the tridentate ligand is twisted with respect to the plane of the threedonors (NP2) and this would be expected to further minimize overlap between the lone pair onthe amide and the d orbital.Me3Si../Me2N/NMe211NN11Me2N’’7 \N—SiMe3NMe2 Me3Si2.27references on page flChapter II 832.7.7 Bonding in the Samarium and Lithium Dinitrogen ComplexesThe dinuclear samarium and lithium dinitrogen complexes discussed in Chapter 1 have avery weak, side-on bound dinitrogen ligand.93’4 This is in stark contrast to the side-ondinitrogen complexes of zirconium where the dinitrogen ligand is highly activated and isstrongly bound to the metal. For comparative reasons some suggestions can be made withregard to the weak bonding interactions of the samarium complex, 1.12, and the lithiumderivative, 1.13.To understand the origin of molecular interactions in electron donor-acceptor complexes(e.g.,H3N—*BH),the total interaction energy of these complexes can be decomposed into fivecomponents; electrostatic (ES), polarization (PL), exchange repulsion (EX), charge transfer(CT) and coupling (MIX).95 This ab initio SCF method has been extended to the donor-acceptor interactions of a dinitrogen ligand with a transition metal, for example, in the complex[Ru(NH3)5(N2)]2having the dinitrogen end-on, the calculated energy for the metal dinitrogeninteraction was -15.52 eV of which -0.68 eV has been attributed to the ES interaction, whereasfor the side-on mode of bonding, the calculated energy for the metal dinitrogen bond was -18.64eV of which -1.15 eV has been atthbuted to the ES interaction.87’8It is therefore possible that in the samarium and lithium complexes 1.12 and 1.13 theprimary interaction between the dinitrogen and the metals is wealdy electrostatic in nature. Onthe basis of this assumption, the bonding in these complexes can be compared to the bonding indilithioacetylide, Li2C where the interactions between the lithium and the acetylide are ionic innature. X-ray structure analysis and theoretical studies done on dilithioacetylide have shownthat the favored mode of coordination is side-on: Li2(J.t-T1-C2).29’3°Therefore, in complexes1.12 and 1.13 if the metal dinitrogen interactions are mainly ionic in nature the favored mode ofdinitrogen coordination could be side-on. However, it is important to note that the mode ofdinitrogen coordination in the solution state has not been determined for these complexes. Also,in the case of the lithium complex it has been suggested that the solid state structure could beattributed to solid state effects.94references on page flChapter II 842.8 Reactions2.8.1 Conversion of the Side-On Complex to the End-On ComplexWhen the side-on complex 2.2 was reacted with NaCp•DME in THF the deep bluecolour of the solution changed to give a deep brown solution over a period of 1 week. The 1Hand 31P{1H} NMR spectra of the final product showed only the presence of the end-onderivative 2.9 (Equation 2.2). Monitoring the reaction shown in Equation 22 by NMRspectroscopy showed only the depletion of the side-on complex 2.2, and concomitant formationof the end-on derivative, 2.9. This seems to suggest that the mono cyclopentadienylintermediate formed by the metathesis of only one zirconium-chloride bond of 2.2 is rapidlyconverted to the bis cyclopentadienyl derivative 2.9.NaCp’DME2.22.2NaCpDME15 15 15 15(PNP)CpZr N— N= ZrCp(PNP) (PNP)CpZr= N— N ZrC1(PNP)2.9 2.22aRapidI I I IsomerizationNaCp•DME(PNP)C1ZrjZrCl(PNP)2.2 2.22“(PNP)CpZr” “ZrC1(PNP)(N2)”Scheme 2.102.91:*references on page 112Chapter II 85One possible mechanistic proposal for the reaction shown in Equation 2.2 is given inScheme 2.10. The elusive mono cyclopentadienyl intermediate, { [(Pr2PCH2SiMe2)2N] Zr} 2-ClCp(i-N2)2.22, could dissociate into mononuclear zirconium species and then recombine toform complexes 2.2 and 2.9 In accordance with this postulate, before or during therecombination of the monomers the coordinated dinitrogen may be exchanged with freedinitrogen from the surroundings.Control reaction under Ar Reaction under ‘4N2I IIFigure 2.22 I The parent ion peaks of the 15N2 analogue of 2.9. II Parent ion peaks of 2.9formed from the reaction of 15N2 analogue of 2.2 with NaCp•DME under anatmosphere of‘4N2.Aforementioned hypothesis was tested by performing the reaction with the nitrogen-15labelled material of 2.2, under an atmosphere of normal dinitrogen (‘4N2). When the product ofthis reaction was analyzed by mass spectroscopy, the relative intensities of the peaks associatedwith the parent ions were virtually identical to the parent ion peaks found for an authenticsample of the 15N2 analogue of complex 2.9 (Figure 2.22). This experiment suggests that there1E831126 11261075 1125 1175 1110 1125references on page flChapter II 86is no exchange between the coordinated dinitrogen and the dinitrogen from the surroundings.Therefore, it is possible that the side-on bound dinitrogen ligand in the intermediate 2.22 rotatesin an intramolecular fashion to give an end-on intermediate 2.22a (Scheme 2.10). Also, the end-on isomer 2.22a is sterically less hindered than the side-on isomer, 2.22, and thereforekinetically more favored to react with the second equivalent of NaCp•DME to give complex Reaction of Side-On Complexes With LiBEt4Earlier work in our laboratory has shown that in the case of the dinuclear palladiumcomplex 2.23 the Pd2(j.i-H) core interacts with a LiBEt4 unit, where the lithium ion is foundsandwiched between the borate moiety and the Pd2(1.L-H)2 core.96 It is postulated that theinteraction between thePd2(-H)2 core and the lithium cation is a Lewis acid, Lewis base type.By comparison, complex 2.2 has a planar Zr2N2 core, with considerable negative charge on thenitrogen atoms. Therefore reacting complex 2.2 and LiBEt4 may lead to the synthesis of acomplex that would resemble 2.23, for example a complex having a core similar to T.Furthermore, an X-ray structure elucidation of such an adduct of LiBEt4 could provide someexperimental evidence for the suggestion that the very long nitrogen-nitrogen bond length of 2.2is partly attributed to the higher charge densities on the nitrogens of the dinitrogen ligand.[Li]ItIII SI‘[Zr] [Zr]T[Li] = LiEt4B[Zr] = “ZrX[N(SiMe2CHPPr’)where X = Cl or OArReacting the side-on derivative, 2.2 with LiBEt4 in toluene over a period of 7 days gavea yellow solution. The 31P{1H} NMR spectrum of the yellow material had a spectral patternp1J2Pr22.23references on page flChapter II 87which is consistent with an ABMX spin system (Figure 2.23). The1H(31P} NMR spectrumconsists of 8* resonances in the SiMe2 region, which implies that the complex is anunsymmetrical dimeric species. The resonance at 4.81 ppm is attributed to a hydride ligand.The multiplet at 0.03 ppm is probably an overlapping doublet of quartets with small coupling tophosphorus nuclei. This suggest that the peak at 0.03 ppm probably corresponds to one of themethylene protons of an ethyl ligand, and the other methylene proton is probably beneath theSiMe2 resonances. The 15N NMR spectrum has two doublets centered at 250.5 ppm and 185.9ppm, with 2JNN = 12 Hz. The llB{ lH} NMR spectrum has a broad resonance** centeredaround -41.6 ppm. On the basis of the aforementioned spectroscopic evidence the complex canbe formulated as {f[(Pri2PCHSiMe)N]Zr }{[(Pr2PCHSiMe)N]ZrEt} } (.t-N.BEt),2.24.The reaction of LiBEt4 with other side-on derivatives, 2.12, 2.20 and 2.21 also gave ayellow material which had identical spectral features to that of 2.24. This seems to suggest thatduring the reaction the zirconium-X bond, where X = Cl, Br or OAr*, undergoes a metathesisreaction to give a diethyl intermediate,“{[(Pr2PCH2SiMe)N]ZrEt}2(p.-N”where one of theethyl group undergoes a 3-hydride elimination reaction. Monitoring the reaction (inside asealed NMR tube) by 1H and 31P{ 1H} NMR spectroscopy failed to provide any usefulinformation; however, the 1H NMR spectrum showed the formation of ethylene. Also, thepresence of free Et3B in the 11B { 1H } NMR suggests that only one of the BEt3 is associated withthe complex 2.24.* One of the methyl group of a SiMe2 seems to have shifted into the region where the isopropyl methyl resonancesappear. By comparison with the 1H and1H(31P) NMR spectra the resonance at 0.95 ppm was assigned to onemethyl group of a SiMe2moiety.** The broad resonances due to the glass of the NMR tube also appears from -20 to -140 ppm. The peak at -41.6was found by subtracting the spectra due to the glass.references on page flChapter II 8896.2 MHz 11B{H} NMR30.406 MHz 15N NMRCII III IiI0 — 0 —40—60—0 PP.+1Q0121.4 MHz 31P{1H} NMRB111111 III IIj I liii!20 15 10 PPM 00.0Figure 2.23 AThe500MHz1H(31P} NMR;B 121.4 MHz31P{H} NMR; C30.406 MHz15N NMR and D 96.2 MHz 11B ( 1H} NMR spectra of the crude reaction mixtureobtained from the reaction of 2.2 and LiBEt4 in C7D8.D/Jdo do ao 20 20 i0Si(CH3)2iV Si(CH3)2500 MHz1H{31P} NMRA[ZrJ_Jj•-I, I I •iii I I UI 1111115 42.5 ppm 2.0 isP[CH(CH3)2] 2N0.5references on page 1iChapter II 89The 15N NMR chemical shifts of the nitrogens of the dinitrogen ligand are centeredaround 250.5 and 185.9 ppm, which by comparison with the side-on complexes are shifted bymore than a 100 ppm. Also, comparison of the chemical shifts of the two nitrogens, which areseparated by 65 ppm, suggests that the BEt3 is probably interacting with one of the nitrogens ofthe dinitrogen ligand. In the 31P(1H} NMR spectrum the two high field singlet resonances areprobably due to a PNP ligand that is facially coordinated*, whereas the two low field doubletsare due to a meridionally coordinated PNP ligand where the observed coupling constant2A-B =72 Hz is comparable to other zirconium complexes containing a meridional PNP ligand.’5 Thestructure depicted as 2.24 seems to fit most of the NIVIR spectroscopic data.A structure involving an2-dinitrogen ligand, 2.24a with one of the nitrogensinteracting with BEt3 is also possible. The coordination around the dinitrogen ligand in 2.24a iscomparable to the tetranuclear titanium dinitrogen complex 1.9. Although the complex 2.24gave good crystals at -40 C, upon warming to room temperature they melted, which hamperedattempts to obtain an X-ray structure./—...p p1Me2Si 1/cMeSiZL__p” \ SiMe2N_zNH\/ SiMe2Et3B pJ22.24 2.24a* The chemical shift for uncoordinated phosphine is usually between -4.5 and -4.8 ppm. It is believed that facialcoordination of the PNP ligand with isopropyl substituents is not sterically favored. However, it could approachfacial coordination by lengthening the zirconium-phosphorus bond which would enable the bulky isopropylsubstituents to be further away. Therefore, the31P (1H1 NIvIR chemical shifts of such phosphines are also likely toshift towards that of the uncoordinated phosphines.Ph2 Me2N.... Se2Phi2references on page iiChapter H 902.8.3 Protonation ReactionsWhen a toluene solution of 2.9 was reacted with anhydrous HC1 gas 1 equivalent ofhydrazine was produced. By comparison, the side-on derivative 2.2 also gave 1 equivalent ofhydrazine under similar conditions. However, when the end-on derivative 2.9 was reacted with1M HC1(aq) or with water, approximately 0.8 equivalents of gas was produced, and a qualitativetest of the aqueous solutions after the reaction showed the presence of hydrazine. Also, when2.9 was decomposed with D20 inside a NMR tube, the 1H NMR spectrum showed theformation of only a small amount of cyclopentadine (<2%), suggesting that the hydrolysis ofthe zirconium-Cp bonds is minimal during protonation in aqueous media.* The gaseous mixtureprobably consists mainly of dinitrogen and dihydrogen in approximately a 1:1 ratio. Whencomplex 2.9 is decomposed to give dinitrogen it would also generate a zirconium(ll) specieswhich, under aqueous conditions, is likely to reduce hydrogen ions to dihydrogen(Scheme 2.11). Therefore depending on the strength of the acid, complex 2.9 could either giveonly hydrazine, or a mixture of hydrazine, dinitrogen and dihydrogen.[R2 NllZr=-N=[Zr][R2 N]___________ ____ _ __ _Reduced to Hydrazine[R2 N][Zrj=N—N=[Zr][R N]H3O[R2 N][Zr]=N—N=[Zr][R N] > [R N][Zr]=N—N=[Zr][R2N]H 1DinitrogenDihydrogen + Zr(IV) Species < Zr(ll) SpeciesScheme 2.11* This reaction was done to show that the amount of gas measured did not contain any volatile cyclopentadiene.references on pageChapter II 91Under anhydrous conditions and in a non-polar solvents like toluene, HC1 acts as a weakacid,*97 and therefore is likely to react at the most basic site of the molecule which is probablythe reduced dinitrogen ligand, (N2)4. The reaction in toluene can also be envisaged as anaddition of HC1 across the zirconium-dinitrogen bonds (Scheme 2.11). Under aqueousconditions the strength of HC1 is many orders of magnitude greater than its strength in tolueneand therefore protonation at different sites of the complex becomes possible. When a C7D8solution of 2.9 was reacted with pyridinium hydrochloride inside a sealed NMR tube, the31P(1H} NMR spectrum showed a clean conversion of 2.9 to the uncoordinated amine of PNP,and no signals were found around 4 ppm that would correspond to hydrazine. This observationsuggests that the protonation probably takes place at the nitrogen of the amide ligand which inturn could result in the decomposition of the complex.Table 2.19 A compilation of chemical shifts of NH protons associated with hydrazido ligands.Compound ö (NH) ppm referencesCp*Me3W=NNH2 4.48 98Cp*MeW=NNH(SiMe) 5.12 98[Cp*Me3W=NNHLi]x 6.10 98[Cp*MeW=NNH]2ZrCp 7.81 98Cp*MeW=NNH(ZrCpMe) 8.03 98[W(C5e4But)(CCMe3)Clj(J.trl:‘q1N}TJJ}i) 10.7 99,100W(NPh)Mej2(-1’:i1NH2)(p,-rrjHNH) 47L-fl’:r’NB2NH2 4.261:1NHN}I 5.70When the side-on complex 2.2 was reacted with pyridinium hydrochloride in C7D8 the31P{1H} NMR spectrum showed the presence of one intermediate, 2.25 (amounts to approx.* The Ka for the dissociation of HBr in CH3N solvent was found to be 3.16x106HBr + CH3N = CH3NH + Br.references on page 212Chapter II 9230%) along with the trichioro complex 2.1 (20%) and the uncoordinated amine of the ligandPNP, [(Pr2PCH2SiMe2)2N}1] (40%). Repeating the reaction with the ‘5N2 analogue of 2.2 the1H NMR spectra showed that the resonance at 4.99 ppm was split into a doublet(2JN-H 59.7Hz), suggesting that the intermediate has a hydrazido type ligand, (N2H)-. Comparison of thechemical shifts (Table 2.19) of different types of hydrazido ligands suggest that the peak at 4.99ppm probably corresponds to a hydrazido ligand of the type, (NNH2)-.The mononuclear structures, for example 2.25a are excluded on the basis of theintegration of (NNH2)-protons with respect to the SiMe2 resonances. The structure depicted in2.25 (Scheme 2.12) is proposed for the intermediate which has a bridging hydrazido (N2H2)2ligand. Although the structure is drawn with an overall C2 symmetry, a fluxional process thatwould enable the chloro ligands trans to the amide nitrogen to exchange positions would give atime averaged C2v symmetry to the molecule. This would be consistent with a broad 31P(1H}NMR signal and two singlets for the SiMe2 groups in the 1H NMR spectra.Me2eCi— Zr—.2.2j \,i’LC5HN•HClMe2SI\ \ CI Me2Si\ \ci. / SiMe2:N—zr :N—Zr zr—N:I’ I.-..-.I IN / N \Me2Si / ‘ Me2Si / I SIMe2NH2 \%___. NH2Pr’2 p1J22.25a 2.25Scheme 2.12references on page flChapter II 932.8.4 Reactions With Halo AlkanesAlthough the chemistry associated with dinitrogen is mainly directed towards theprotonation of the coordinated dinitrogen ligand, occasionally attention has been focusedtowards incorporating nitrogen from dinitrogen into organic molecules. Reactions involving N2complexes and ailcyl or aryl halides have led to the synthesis of alkyl or aryl amines.101”02Recently, molecular nitrogen has been incorporated into N-heterocycles by using titanium andpalladium as catalysts.’°3In some instances, the reactivity of the coordinated dinitrogen has been helpful inunderstanding the nature of metal-dinitrogen interactions, e.g., some tantalum dinitrogencomplexes react with ketones to form hydrazones, where the (formulated) Ta=N—N=Ta unithas been transformed intoR2C=N—N=CR.492.8.4a Reactions of the End-On Complex With BzBr and CH3IWhen the reactivity of 2.9 was investigated with CH3I or C6H52Br (BzBr), evenunder carefully controlled conditions, a complex mixture of products was obtained. On thebasis of 31P { 1H} NMR spectral analysis the only identifiable product was the dihalidederivative ofZrCpX2[N(SiMeCHPPr’)](X = Br or I), which corresponds to a resonancearound 15 ppm. Fortunately, when the reaction of 2.9 with approximately 1 equivalent of BzBrwas monitored by variable temperature 31P { 1H} NMR spectroscopy from -78 °C to roomtemperature, the formation of two products was observed. The crude material from the reactionof 2.9 with BzBr consisted of three compounds: ZrCpBr2[N(SiMe2CHPPr)j2.7, theintermediate 2.26, along with some unreacted 2.9 (Figure 2.24). Persistent attempts, by repeatedfractional crystallization, eventually led to the isolation of the intermediate 2.26 as pale yellowcrystals. Elemental analysis of this intermediate was consistent with a product formed by theaddition of one equivalent of BzBr to the dimer 2.9. It was found that in solution theintermediate 2.26 slowly reacts to give the end-on complex 2.9 and other unidentified products.references on pageChapter II 94*2.7 ZrCpBr[N(SiMe2CH2PPr)2J2.9 { [(PrPCHMe2Si)2NjZrCp } (p-N)* 2.26f (PNP)HFigure 2.24 A Variable temperature 1H NMR spectra (300 MHz) of complex 2.26 recorded inC7D8. B and C are 31P(1H} NMR (121 MHz) spectra of 2.26 and of the crude reactionmixture obtained from the reaction of 2.9 and PhCH2Br.I III1 I III I I I I 1111 II II III I I 111111111 III I lIlt I I I 111111 I III I1 I I 11111111 III III 11111 lii I8 4 3 2 1. OPPMreferences on page iiChapter II 95The complexity of the variable temperature 1H and 31P{1H} NMR spectra of 2.26 isconsistent with the presence of more than one isomer in solution, and also that these isomersinterconvert with one another. The SiMe2 resonances in the 1H NMR spectrum were too broadto be interpreted. However, some of the resonances associated with the Cp ligand wereinformative. The appearance of two broad doublets at 6.05 and 4.65 ppm were assigned to theolefinic protons of an r1-Cp ligand. The variable temperature 1H NMR spectra show that theresonances associated with the r1-Cp ligand decrease in intensity with a concomitant increase ofthe peak at 5.22 ppm. This spectral feature could be due to the Cp ring-slip between a TI1 and T5coordination, where the latter corresponds to the resonance at 5.22 ppm. Comparison of the 1HNMR chemical shifts of the Cp ligand in zirconium(IV) PNP derivatives, which appear around 6ppm, the peak at 5.22 ppm is significantly shifted to higher field. In fact, in the zirconium(II)dicarbonyl complex,Zr(11SCH)(CO)[N SiMeCHPPri],the resonances due to Cp appearat 4.7 ppm.The fluxional process shown in Scheme 2.13 can rationalize some of the spectralfeatures of intermediate 2.26. The species 2.26a is generated by the addition of one BzBr unitacross one of the zirconium-nitrogen bonds of the Zr2Ncore, which probably undergoes rapidisomerization to give other species. The species depicted as 2.26b, 2.26c and 2.26d aregenerated by incorporating a zirconium(ll) fragment, “ZrCp[N(SiMeCHPPr)]”into abenzyldiazinide (PhCH2N=N) ligand of a zirconium(IV) monomer, “ZrCpBr(N=NCH2Ph)-[N(SiMe2CHPPr)1”. All of the structural types associated with the benzyldiazinide ligandhave precedence in the literature.104’5 The intermediates 2.26a and 2.26b are formulated asZr(IV)-Zr(IV) complexes whereas 2.26c and 2.26d are formulated as Zr(ll)-Zr(IV) complexes.In 2.26b one of the zirconium centers has a Cp ligand bound in an r1 fashion and also interactswith the diazinide ligand in an 112 fashion.106 Of the two Zr(II)-Zr(IV) complexes 2.26d issterically more congested than 2.26c and could dissociate into Zr(II), “ZrCp[N(SiMe2-CH2PPr)J” and zirconium(IV), “ZrCpBr(N=NCHPh)[N(SiMe2CHPPr’)J”monomers.The former, in the presence of dinitrogen, may proceed to form the end-on dinitrogen complex2.9 whereas the latter is probably unstable and undergoes decomposition. It is important to notereferences on page 2.12Chapter H 96that in Scheme 2.13 all of the intermediates are drawn with phosphine donors coordinated, buteach case can involve reversible dissociation of the phosphine donors. Such a process wouldexplain the broad resonances observed in the 31P{ ‘HI NMR spectrum.prj22.26cComplex 2.6Scheme Reactivity With DihaloalkanesInitial investigations involved the reaction of the simplest dihalo alkanedibromomethane, CH2Br and complexes 2.2 and 2.9. It was envisaged that the reaction of 0.5equivalents of CH2Br2 with 2.2 or 2.9 would lead to the formation of an isodiazomethane or adiazomethane complex respectively (Scheme 2.14). It is also known that reactions involvingPt2Pr’2Me2 SiMe2Si2.26aPr’22.26bUIPr’2Pr’2Me2Me2 Si\SIMe2ISIMe 2.26dreferences on page 112Chapter II 97diazomethane and low valent metal complexes (e.g., [Ru(NO)(PPh3)2Cll)lead to the formationof methylidene intermediates which then proceed to form ethylene complexes.’072.2CH2BrCH2BrPr’2Me2\: N—Zr/Me2S1 / NNPr2CH2Pr’2MeSi \ 0:N—Zr”/ %S44Me2Si N’L_N\-CH2Scheme 2.14Reacting the end-on complex 2.9 with approximately 0.5 equivalent of CH2Br inside asealed NMR tube and monitoring by 31P(1H} NMR spectroscopy showed a clean conversion ofthe dinitrogen complex 2.9 to the zirconium(IV) derivativeZrCpBr2[N(SiMeCH2PPr’)2], 2.7.The only peak unaccounted for in the 1H NMR spectrum was at 5.2 ppm, which is identical tothe chemical shift of ethylene. Also, at the end of the reaction (i.e., when all the CH2Br wasconsumed) approximately 0.5 equivalent of the dinitrogen complex 2.9 was left unreacted andno NMR features that would correspond to an oxidative addition product (e.g.,ZrCpBr(CH2B )[N(SiMe2CHPPr)])were observed.The above results seems to suggest that the proposed diazomethane intermediate,ZrCp(NCH)[N(SiMePPr]2.27, was formed during the course of the reaction. Lowtemperature NMR studies did not show any features that would correspond to free diazomethaneCH2N,or CH2NNCH. Scheme 2.15 shows a possible mechanistic pathway to generate2.9S[ pPr’2Me2‘SiMe2references on pageChapter II 98ethylene from the proposed diazo intermediate. The diazomethane intermediate probablyundergoes a bimolecular decomposition pathway to yield ethylene and 0.5 equivalent ofdinitrogen complex. However, the literature precedence seems to suggest that the diazomethanecomplex could form a methylidene intermediate, ZrCp(CH)[N(SiMe2CHPPri)]which thenproceeds to yield ethylene and the dinitrogen complex 2.9.1,b07[Zr]Br2[ZrN—N[Zr]- [ZrN—N=CH2N21 rC2H4[Zr](II) Species - [Zr]CH[Zr] = “ZrCp[N(S1Me2HPPr1)]”Scheme 2.15H Me H MeMeHNOC1 [MeH]N0MeH2.3Br[Zr]N—N=[Zrj [Zr]N—N4 )[Zr]BrN2C2H4 [Zr]Br2 d-’:[Zr](II) Species [Zr]N—N[Zr] = “ZrCp[N(SiMe2CHPP?)]”Scheme 2.16references on page 212Chapter H 99The reaction involving 0.5 equivalents of 1,2-dibromoethane and the end-on derivative2.9 also gave exclusively ethylene and the zirconium(IV) complex,ZrCpBr2[N(SiMeCH2PPr’2)2]. In this case a three-membered azacyclic intermediate,ZrCp[NN(CH2)2][N(SiMe2CH2PPr)2] is probably formed during the reaction (Scheme 2.16).The elimination of ethylene from this intermediate may be comparable to the reaction of someN-nitrosyl aziridine derivatives which decompose to give olefins and N20 (Equation 2.3).b08Br BrCH2Br N Br I[Zr] I [Zr] - [Zr] I Br [Zr]...[Zr][Zr] NCH2[Zr]( + [Zr](I)CH2 EthrleneBr N 4Br Br[Zr][Zr]-j . [ZrkN[Zr]BrCH2CHr N[Zr] = “ZrX[N(SiMe2CHPPr1)]”Scheme 2.17The reactions involving side-on complexes 2.2 or 2.12 and CH2Br2 or BrCH2CH2Bralso gave ethylene and the zirconium(IV) trihalo derivative ZrBr2Cl[N(SiMeCPPr)].Dueto the poor solubility of the side-on dinitrogen complexes the solutions always contained anexcess of the dihalo alkane and therefore hampered the NMR experiments to detect theintermediates involved in the reaction. The possible intermediates involved during the reactionswith CH2Br could be either anr2-diazirine complex or a bridging t-r1-diazomethane complex(Scheme 2.17). For the reaction of the side-on derivatives with BrCH2CH2Br the intermediatereferences on page 1.12Chapter II 100{(Pri2PCHSiMe)N]ZrXBr}[j.LNN(CH,where X = Cl or OAr*, with a bridging[NN(CH)2]-ligand is possible, which is similar to the proposed intermediate for the end-oncase (Scheme 2.17).2.9 ConclusionsThe synthesis of dinitrogen complexes similar to {[(Pr2PCHSiMe)N]ZrC1}2-(i.-112:-N2) 2.2 was successful in two cases where the chloride ligand was replaced with theCp or with an aryloxy ligand, ((0C6H3-2,6-Me)’,OAr*). The complex with Cp ligands,{[(PriPCHSiMe)NZrCp} 2(I.t-111:1-N2) 2.9 has a bridging end-on bound dinitrogen ligandwhereas the complex with OAr’ ligands ([(PrjPCHSiMeN]ZrOAr*}2(i.rI:12)2.12has a side-on bound dinitrogen ligand with a bent Zr2N2 core. However, attempts to incorporateallyl, alkyl or amide ligands or the incorporation of the less bulky tridentate ligand“Zr[N(SiMe2CH2PMe2)21” were not successful. The solid state X-ray structures of the twocomplexes show that in the end-on case 2.9, the nitrogen-nitrogen bond distance of thedinitrogen ligand was 1.301(3) A whereas in the side-on case 2.12, the bond distance was1.528(7) A which was identical to the chloride analogue 2.2 (1.548(7) A). The tert-butoxyanalog of complex the 2.12,{[(PriPCHSiMeN]ZrOBut)(p,N) 2.14, shows certain NMRspectroscopic features which tend to suggest that it is similar to other side-on derivatives,although the lack of suitable crystals hampered full charecterization.Solid and solution state resonance Raman spectroscopic studies on the side-on 2.2 andon the end-on 2.9 derivatives show that the mode of clinitrogen coordination observed in thesolid state X-ray structures of these complexes is also preserved in the solution state. Theobserved nitrogen-nitrogen stretching frequencies for the end-on and side-on derivatives were1211 cm-1 and 731 cm-1 respectively, which correlate well with the observed bond distancesassociated with the same atoms. The 15N NMR spectroscopic evidence also shows that theside-on derivative has the same mode of dinitrogen coordination in both the solid and thesolution states. The mixed halogen derivative gives riseto a singlet in the 15N NMR spectrum showing that nitrogens of the dlinitrogen ligand arereferences on page 212Chapter II 101chemically equivalent and therefore, it is inferred that the bonding mode of the N2 ligand in thesolution state is side-on, (.t-r2:rI-N2).The ML4fragment (C4)has three available orbitalsthat are suitable for dinitrogen bonding11In the case of[(H3P)2(N)ZrX], X = Cl, Cp or OAr*the it-interactions leads to two cases,it-interactions on the planedefined by the ligandsLEnd-On Complex Side-On ComplexScheme 2.18The semi-empirical molecular orbital studies performed on the end-on derivative 2.9 andon the side-on derivatives 2.2 and 2.12 show that the it-acceptor interactions involving the itorbitals of the dinitrogen ligand are significantly different. In the end-on case these interactionsgive rise to two it-MOs (HOMO and HOMO-l) whereas, the side-on cases gives rise to oneö-MO (HOMO) and one it-MO (HOMO-9 for 2.2 and HOMO-3 for 2.12), which were found tobe much lower in energy than the ö-MO. The Mulliken population analysis shows greaterelectron donation into the dinitrogen ligand in the side-on cases than in the end-on case, whichit-interactions perpendicular to theplane defined by the ligands/L11The remaining two metal based orbitals areperpendicular to the plane and can formonly two it-type bonds with the dinitrogenit*orbitalsiiOf the remaining two metal based orbitalsone is perpendicular to the plane and theother is on the plane and can form oneit-type and one ö-type bond respectivelywith the dinitrogen 1t*orbitalsJireferences on page 112Chapter II 102in turn is reflected in the formal charge on the dinitrogen ligand which was greater for theformer than in the latter. The Wiberg indices suggest a weaker nitrogen-nitrogen bond for theside-on mode than the end-on mode, which is consistent with the observed nitrogen-nitrogenbond distances of these complexes.An analysis of the fragment orbitals of the metal unit, [(H3P)2(H2N)ZrX], where X = Cl,Cp or OH, show that the metal-ligand d,1-p interactions influence the mode of dinitrogencoordination, i.e., end-on vs side-on. Scheme 2.18 summarizes how the metal-ligand it-interactions influence the mode of dinitrogen bonding.An intramolecular rotation of the dinitrogen ligand within a dinuclear center wasdemonstrated by the conversion of the side-on complex 2.2 to the end-on complex 2.9 by thereaction of NaCp.DME. The reaction of the side-on derivative 2.2 and 2.12 with LiBEt4 formsan unusual complex, ([(Pr2PCHSiMe)N]Zr }{[(PrPCHSiMe)N]ZrEt}Qi-N2.BEt3)where the dinitrogen ligand seems to be bonded to a BEt3 unit. Reactions of the dinitrogencomplex 2.9 with alkyl halides were very complex, but in the case of benzyl bromide it wasfound that the reaction proceeds through an intermediate containing a benzyl diazenido,(NNBz)-1,ligand. The side-on and end-on derivatives react with methylene dibromide or with1,2-dibromoethane to give ethylene.2.10 Experimental2.lOa General Procedures All manipulations were performed under prepurified nitrogen in aVacuum Atmospheres HE-553-2 work station equipped with a MO-40-2H purification systemor in Schienk-type glassware.109”0Pentane, diethyl ether and hexanes were dried anddeoxygenated by distillation from sodium-benzophenone ketyl under argon. Toluene waspredried by refluxing over CaH2 and then distilled from sodium under argon. Tetrahydrofuranand hexanes were predried by refluxing over CaH2 and then distilled from sodiumbenzophenone ketyl under argon. Deuterated benzene (C6D,99.6 atom % D) and deuteratedtoluene (C7D8,99.6 atom %D), purchased from MSD Isotopes, were dried over activated 4Amolecular sieves, vacuum transferred and freeze-pump-thawed three times before use.references on page j2Chapter II 103Dinitrogen gas was purified by purging through a column containing manganese oxide,MnO, and activated 4A molecular sieves. The 99% 15N2 was obtained from CambridgeIsotopes Ltd., and was used as supplied.The 1H, 31P and 13C NMR spectra were recorded on a Varian XL-300, a Bruker AC200, a Bruker WH-400 or a Bruker AM 500 spectrometer. Proton spectra were referenced usingthe partially deuterated solvent peak as the internal reference, C6D5H at 7.15 ppm andC6D5D2Hat 2.09 ppm relative to Me4Si. The 31P{ ‘H) NMR spectra were referenced toexternal P(OMe)3 set at +141.00 ppm relative to 85% H3P04. The ‘3C{ 1H} NMR spectra werereferenced to the C6D6 signal at 128.0 ppm or to the CD36D5signal at 20.4 ppm. Solution15N{1H} NMR spectra were recorded on the Varian XL-300, referenced to external formamideset at 0.00 ppm. Solid-state 15N NMR spectra were run on a Bruker MSL-400, referenced tosolid NH4C1 set at -73.39 ppm with respect to neat formamide at 0.00 ppm. 11B{IH} NMRspectra were recorded in C7D8 on the Varian XL-300 referenced to external Et3B set at 0.00ppm. The final 11B (‘H) NMR spectra were obtained by subtracting the spectra of glassobtained using identical acquisition parameters as those of the sample spectra. 1H { 31P) NIVIRspectra were recorded on the Bruker AM 500 spectrometer.Resonance Raman* spectra were recorded on a modified Jarrel-Ash model 25-300Raman spectrometer. Excitation radiation was supplied by Spectra Physics LAJ+ and Kr+ lasersoperating at 514.5 nm and 647.1 nm respectively. Scattered radiation was collected by a backscattering geometry. Multiple scans were collected and calibrated against indene, toluene andTHF as external standards for peak positions. All wave number assignments were estimated tobe accurate within 2 cm1. Spectra for polarization studies of solution samples were collected ina 90° scattering geometry at 278 K on a Dilor Z-24 Raman spectrophotometer. Typical slit-width settings were between 5 cm-1 and 9 cm-’. All samples were sealed under N2 in 1.5 -1.8x90 mm Kimax glass capillary tubes.* The resonance Raman spectra were recorded at the Department of Chemistry, Biochemistry and MolecularBiology of the Oregon Graduate Institute of Science and Technology by Prof. Thomas M. Loehr and Jonathan D.Cohen.references on page flChapter II 104UV-Vis spectra were recorded on a Perkin Elmer 5523 UV/Vis spectrophotometerstabilized at 20CC. Mass spectral studies were carried out on a Kratos MS 50 using an Elsource. JR spectra were recorded on a Bomem MB-l00 spectrometer. Solution samples wererecorded in a 0.1mm KBr cell and solid samples were recorded as KBr pellets. Carbon,hydrogen and nitrogen analyses were performed by the microanalyst of this department.The complexes ZrC13[N(SiMe2 H2PR2)2] where R = Pr and Me, were preparedaccording to the published procedures.1’NaCp•DME,” where DME = l,2-dimethoxyethane,was prepared by the reaction of sodium metal with freshly cracked cyclopentadiene in dry DMEand crystallized directly from the reaction mixture. Mercury was purchased from BDH andpurified as foilows:2in a separatory funnel mercury (500g) was washed with 2M HC1 (2x25mL) acid then with distilled water (2x50 mL) and finally rinsed a few times with Et20 (25 mL)until no further gray colour was present in the ether washings. During washings a slag wasformed on the surface of the Hg which was separated from the shiny Hg. Pure Hg was driedunder vacuum for 12 hours. Amalgam was made under a nitrogen atmosphere and washed withtoluene (2x25 mL) until the washings showed no gray coloration.Sodium ailcoxide, NaOR’, where R’ = CHPh2 and aryloxides, NaOAr where Ar = Ph orC6H3Me2-2,6 were prepared by reacting (3 hours at room temperature and then refluxed for 3hours) a toluene solution of the alcohol with sodium. The resulting white solid was filtered,extracted with THF (to remove finely dispersed sodium) and stripping off THF under vacuumgave alcohol free NaOAr. Prior to the reaction the alcohol was dissolved in toluene and stirredwith Mg turnings (0.1 equivalent) for 12h and then filtered. The above procedure is likely tominimize the presence of NaOH or unreacted alcohol in the final product. KOBut waspurchased from Aldrich and was sublimed prior to use. The sodium amide, NaNPh2 wasprepared by reacting the amine, HNPh2 with NaN(SiMe3)2 in toluene. The NaNPh2 precipitatesout of toluene, which was then collected on a frit and washed with hexanes to obtain purematerial.references on pageChapter II 105CH3I,CH2Br,C2H4Br2 andC6H5H2Br were purchased from Aldrich and were driedover activated 4A molecular sieves for 24h. ZrBr4 was purchased from Strem and used assupplied.2.lOb Synthesis of Precursors2.lOb.1ZrBr3[N(SiMeCHPP 2)2J, 2.4. To a suspension of ZrBr4 (2.03 g, 4.94 mmol) intoluene (50 mL) was added a solution of Li[N(SiMe2CHPPr2) j, (1.00 g, 2.50 mmol) intoluene (10 mL) at room temperature. The reaction mixture was stirred for 16 hours and thenthe salt (LiCl) was removed by filtering the solution through Celite. The filtrate wasconcentrated to 15 mL and hexanes was added until the solution turned turbid; cooling at -30 °Cgave a colorless crystalline product (1.12 g, 73%). 1H NMR (6, 200. 123 MHz, C6D6): 0.41 (s,12H, Si(CH3)2; 1.11 (m, 28H, P[CH(CH3)2]2, SiCH2P); 2.07 (sept of triplets, 4H,P[CH(CH3)J2JH-P = 2.7 Hz, 3JHH = 7.2 Hz). 31P{1H) NMR (6, 81.015 MHz, C6D): 16.52(s). 13C{H} NMR (6, 50.323 MHz, C6D): 4.50 (s, Si(CH3)2; 11.93 (s, SiCH2P); 18.85 (s,P[CH(CH32j);19.12 (s, P[CH(CH)J;25.14 (t, P[CH(CH]2Jc.p = 6.8 Hz). Anal.Calcd for l8HBr3NP2Si2Zr: C, 29.88; H, 6.13; N, 1.94. Found: C, 30.00; H, 6.30; N, 1.99.2.lOb.2Zr(i5-C) 1[N(S1MeCHP r] 2.6. To a solution ofZrCl[N(SiMe2CH-PPr)2] 2.1, (4.040 g, 6.84 mmol) in toluene (150 mL) was added solid NaCp.DME (1.340 g,7.52 mmol) at room temperature. The reaction mixture was stirred for 16 hours and then the salt(NaCl) was removed by filtering through Celite. The filtrate was concentrated to 15 mL andhexanes was added until the solution turned turbid; cooling at -30 °C gave yellow hexagonalshaped crystals (3.16 g, 75%). 1H NMR (6, 300 MHz, C6D6): 0.64 (s, 12H, Si(CH3)2); 1.00 (m,28H, P[CH(CH3)2]2, SiCHP); 2.12 (m, 4H, P[CH(CH)2j;6.15 (t, -5H, C5H3Jp 1.4Hz). 31P{1H} NMR (6, 121.421 MHz, C6D): 16.11 (s). 13C{H} NMR (8, 50.323 MHz,C6D): 7.82 (s, Si(CH3)2; 8.71 (s, SiCH2P); 18.74 (s, P[CH(CH)2J; 18.96 (s,P[CH(CH32]);24.72 (t, P[CH(CH3)2]2,2J..p = 4.5 Hz); 116.00 (s, C5H). Anal. Calcd forC2491NPSiZr:C, 44.56; H, 7.97; N, 2.26. Found: C, 44.72; H, 8.20; N, 2.32.references on pageChapter II 1062.lOb.3Zr(5-CH)Br[N(SiMeCHPP1J,2.7. This complex was synthesized by aprocedure similar to the one described above in section 2.2b.2, usingZrBr3[N(SiMeCHPR)](0.180 g, 0.270 mmol) and NaCp.DME (0.053 g, 0.300 mmol). 1H NMR spectrum of the crudereaction mixture showed the presence of 2.7 in greater than 90% yield. 1H NMR (ö, 300 MHz,C6D): 0.69 (s, 1211, Si(CH3)2;0.98 (m, 24H, P[CH(CH3)2];1.06 (t, 4H, SiCH2P,3JpH =5.4 Hz); 2.13 (sept, 4H, P[CH(CH]3JH-H = 7.2 Hz); 6.62 (br, 511, C5H). 31P{1H} NMR(, 121.421 MHz, C6D6): 16.57 (s). Anal. Calcd for C23H49Br2NP2Si2Zr: C, 38.97; H, 6.97; N,1.98. Found: C, 39.00; H, 7.08; N, 1.98.2.lOb.4Zr(ri5-) 1[N(SiMePMe] 2.8. To a solution of ZrC13[N(SiMe2CH-PMe2)21 (1.45 g, 3.03 mmol) in toluene (150 mL) was added solid NaCp.DME (0.541 g, 3.04mmol) in two portions at 1 hour interval at room temperature. The reaction mixture was stirredfor 16 hours and then the salt (NaC1) was removed by filtering through Celite. The filtrate wasconcentrated to 20 mL and hexanes was added until the solution turned turbid; cooling at -30 °Cgave a yellow crystalline material (1.08 g, 70%). 1H NMR (, 300 MHz, C6D6): 0.58 (s, 12H,Si(CH3)2); 0.92 (br, 411, SiCH2P); 0.95 (t, 1211, P(CH3)2,23H-P = 3.6Hz); 6.33 (t, 5H, C5H3H-P 1.5Hz). 31P{1H} NMR (, 121.421 MHz,C6D6): -21.73 (s). ‘3C(1H} NMR (, 50.323MHz, C6D): 7.24 (s, Si(CH3)2); 14.96 (t, P(CH3)213c..p = 7.8Hz); 20.64 (s, SiCH2P); 115.50(s, CH5). Anal. Calcd for1533C1NPSiZr:C, 35.49; H, 6.55; N, 2.76; Cl, 13.98. Found:C, 35.53; H, 6.38; N, 2.76; Cl, 14.20.2.lOb.5Zr(ri-)C1[N(SiMeCHPP ’] 2.10. To a cooled solution (-78 °C) of ZrC13-[N(SiMe2CH2PPr’)j(2.50 g, 4.24 mmol) in THF (70 mL) was added a THF solution ofC3H5MgC1 (23.3 mL of 0.2 M solution, 4.66 mmol) and stirred at -78 °C for 10 minutes. Afterallowing the reaction mixture to warm to room temperature, dioxane (4.5 mL, 50 mmol) wasadded and stirred for 3 hours. A copious white precipitate of MgC12’xTHF was removed byfiltering through Celite. Solvent was stripped under vacuum and redissolved the crude productin hexanes and cooled to -30 °C to give a bright yellow crystalline product. (1.62 g, 64%). 1HNMR (6, 300 MHz, C6D)*: 0.58 (s, 6H, Si(CH3)2;0.59 (s, 6H, Si(CH3)2;0.96 (m, 4H,* H0, Ha and H refer to central, anti and syn protons of the allylic ligand respectivly.references on pageChapter II 107SiCH2P);1.08 (m, 24H, P[CH(CH3)2];2.18 (sept, 2H, P[CH(CH3)2j; 2.25 (sept, 2H,P[CH(CH3]);3.32 (d of triplets, 2Ha, C5H33JHa-Hc = 15.2 Hz,3JHa-P = 3.4 Hz); 4.46 (d oftriplets, 2H5C5H3,3’Hs-Hc = 8.8 Hz,3JHs-p = 4.4 Hz); 6.08 (t of t, lH, C5H33Jj-ja = 15.2Hz,3JHc-Hs = 8.8 Hz). 31P(1H} NIVIR (6, 121.421 MHz, C6D): 21.13 (s). 13C{H} NMR (6,50.323 MHz, C6D): 78.20 (s, CH2HC 2); 139.25 (s, CH2HCH). NOEDIFF Experiments(6, 400 MHz, C6D): Irradiating the resonance at 6.08 ppm resulted in the enhancement of peaksat 3.32 and 4.46 ppm. Attempts to radiate peaks at 3.32 or 4.46 ppm resulted in spin saturationof both resonances and enhanced the peak at 6.08 ppm. Anal. Calcd forC21H49NPSiZr:C, 42.33; H, 8.29; N, 2.35. Found: C, 42.67; H, 8.47; N, 2.34.The 31P(1H) NMR (6, 8 1.015 MHz, C6D6) data forZr(1-CH5)C1B [N(S MeHPPr2)]2.lOa, an AB quartet, 21.08 (1P, 2Jp..p = 7.9 Hz) and 21.65 (1P, 2Jp.p = 7.9 Hz). Anal.Calcd forC21H49IBrNPSiZ :C, 39.39; H, 7.71; N, 2.19. Found: C, 38.90; H, 7.58; N, 2.10.2. lOb.6 Zr(OAr*)C1[N(SIMeHPPri] 2.11. To a solution of ZrCl3[N(SiMe2CH-PPr’)2] 2.1, (4.00 g, 5.08 mmol) in toluene (150 mL) was added solid Na(OAr*) (1.03 g, 6.35mmol) in three portions at 1 hour intervals at room temperature. The reaction mixture wasstirred for 16 hours and then the salt (NaC1) was removed by filtering through Celite. Thefiltrate was concentrated to 25 mL, an equal volume of hexanes added, and the mixture allowedto stand at room temperature for 24 hours. A colorless crystalline product slowly separatedfrom the solution (2.41 g, 70%). 1H NMR (6, 300 MHz, C6D): 0.48 (s, 12H, Si(CH3)2); 0.96(d of d, 16H, 12H of P[CH(CH3)212and 4H of SiCH2P, 3H.H = 6.1 Hz, 3JP-H = 14.0 Hz); 1.06(d of d, 12H, P[CH(CH3)]3JwH = 6.1 Hz, 3Jp..H = 14.0 Hz); 2.00 (sept, of t, 4H,P[CH(CH3)2]3H-H = 6.1 Hz, 2JpH = 2.0 Hz); 2.80 (s, 6H, 2, 6-Me2Ph); 6.82 (t, 1H, p-Ph,3H-H = 7.4 Hz); 7.05 (d, 2H, rn-Ph, 3JwH = 7.4 Hz). 31P{ ‘H) NMR (ö 121.421 MHz, C6D):14.98 (s). ‘3C{1H} NMR (6, 50.323 MHz, C6D): 5.41 (s, Si(CH3)2); 10.70 (s, SiCH2P); 19.07(s, P[CH(CH32);19.15 (s, P[CH(CH)1;19.86 (s, 2, 6-Me2-Ph); 24.09 (t, P[CH(CH3)2J= 6.2 Hz); 120.94 (s, p-Ph); 128.79 (s, rn-Ph). Anal. Calcd forC26H531NOPSiZr:C,46.20; H, 7.90; N, 2.07. Found: C, 46.07; H, 8.10; N, 2.03.references on page 112Chapter II 1082.lOb.7 Zr(OBut)C1[N(S1MeCHPPri)],2.11. To a solution of ZrC13[N(SiMe2CH-PPri)2j (1.50 g, 2.54 mmol) in Et20 (60 mL) was added a solution of KOBut (285 mg, 2.54mmol) in Et20 (10 mL) at room temperature and stirred for 3 hours. The solvent was strippedunder vacuum and the residues were extracted with pentane (40 mL) and then filtered through alayer of Celite. Stripping off the solvent gave a colorless oil containing >80% (by 1H NMRspectroscopy) of the desired product. 1H NMR (6, 300 MHz, C6D): 0.45 (s, 12H, Si(CH3)2;1.05 (d, 4H, SiCH2P2Jp..H = 5.7 Hz); 1.32 and 1.35 (each d of d, 24H, P[CH(CH3)2J3PH =2.9 Hz, 3H-H = 7.6 Hz); 1.50 (s, 9H, OC(CH3);2.04 (t of sept, 4H, P[CH(CHj2Jp..H =1.9 Hz,3JH..H =7.6Hz). 31P{1H) NMR (6, 81.015 MHz, C6D): 11.60 (s).2.lOb.8 Zr(OBut)C1[N(SiMeCPMe)],2.13. The complex was prepared by a proceduresimilar to the one described above in section 2.lOb.7, usingZrC13[N(SiMe2CHPMe)J(440mg, 0.92 mmol) and KOBut (114 mg, 1.01 mmol). The product was crystallized from a solventmixture containing Et20 and pentane (0.38 g, 7 1%). LH NMR (6, 300 MHz, C6D): 0.26 (s,12H, Si(CH3)2;0.87 (d, 4H, SiCH2P,2Jp..H = 9.6 Hz); 1.01 (d, 12H, P(CH3)2 = 6.3 Hz);1.50 (s, 9H, OC(CH3)3). 31P{1H} NMR (6, 121.421 MHz, C6D6): -32.33 (s). 13C(H} NMR(6, 50.323 MHz, C6D): 5.92 (s, Si(CH3)2; 12.53 (d, P(CH3>21J..p = 10.6 Hz); 17.48 (s,SiCH2P); 32.14 (s, OC(CH3). Anal. Calcd forC14371NOPSiZr:C, 32.61; H, 7.23; N,2.72. Found: C, 32.72; H, 7.13; N, 2.80.2.lOb.9 Zr(OCHPh) 1[N(SiMePPr] 2.16. The complex was prepared by aprocedure similar to the one described above in section 2.lOb.6, using ZrC13[N(SiMe2CH-PPr)2j (1.25 g, 2.12 mmol) andPh2CHONaTHF (589 mg, 2.12 mmol). The product wascrystallized from a solvent mixture containing Et20 and pentane (1.23 g, 7 8%). 1H NMR (6,200.132 MHz,C6D): 0.50 (s, 12H, Si(CH3)2;0.94 (m, 28H, SiCH2Pand P[CH(CH2]);1.73(sept, 4H, P[CH(CH]3JH-H = 6.8 Hz); 6.82 (s, 1H, CHPh); 7.00 (211, t, p-Ph, 3JH-H = 7.6Hz); 7.16 (4H, t, rn-Ph, 3JH..H = 7.6 Hz); 7.71 (4H, d, o-Ph,3JH..H = 7.6 Hz). 31P{1H} NMR (6,81.015 MHz, C6D): 13.53 (s). Anal. Calcd forC31552ONPSiZr:C, 50.45; H, 7.51; N,1.90. Found: C, 51.25; H, 7.67; N, 1.76.references on page 2iChapter II 1092.lOb.1O Zr(NPh)C1[N SiMeCHPPr],2.17. To a solution ofZrC13[N(SiMe2CH-PPr’)21 (1.25 g, 2.12 mmol) in THF (60 mL) was added a solution of NaNPh2 (199 mg, 2.12mmol) in THF (20 mL), at room temperature and stirred for 2 hours. The solvent was strippedundervacuum and the residues were extracted with toluene (20 mL) and then filtered through alayer of Celite. The product was crystallized from a solvent mixture containing toluene andhexanes (1.15 g, 75%). 1H NMR (6, 400 MHz, C6D): 0.47 (s, 12H, Si(CH3)2;1.16 and 1.10(m, 28H, SiCH2Pand P[CH(CH32]);2.12 (sept, 4H, P[CH(CH3)2]3HH = 4.0 Hz); 6.96(2H, t, p-Ph,3JH-H = 8.0 Hz); 7.23 (4H, t, rn-Ph,3JH.H = 8.0 Hz); 7.30 (4H, d, o-Ph,3JH..H = 8.0Hz). 31P{1H} NMR (6, 81.015 MHz, C6D): 15.58 (s); in a solvent mixture containing THFand C6D:2.70 (br.); -1.20 (br.). 13C(H} NMR (6, 50.323 MHz, C6D): 5.04 (s, SiCMe2);9.69 (s, CH2Si); 18.85 and 19.59 (s, CH(CH3)2;24.28 (t, CH(CH3)21Jp.. = 5.6 Hz); 123.41(s, Ph); 126.98 (s, Ph); 128.26 (s, Ph). Anal. Calcd for3054C1NPSiZr:C, 49.83; H, 7.53;N, 3.88. Found: C, 50.09; H, 7.56; N, 4.00.2.lOb.11Zr(CHS1Me) 1[N(S1MeCPPr] 2.18. To a cooled (-78 °C) solution ofZrCl[N(SiMeCHPPr)J,(1.25 g, 2.12 mmol) in THF (60 mL) was added a solution ofLiCH2S Me3 (199 mg, 2.10 mmol) in toluene (15 mL) over a period of 30 minutes. Theresulting yellow colored reaction mixture was stirred at -78 °C for lh, warmed to roomtemperature and stirred for another 1 hour. The solvent was stripped off under vacuum andextracted with pentane (20 mL) and then filtered through a layer of Celite. It is important toremove THF as completely as possible to ensure complete precipitation of LiC1. Gradualstripping of pentane (5 mL at a time) and cooling at -30 °C gave yellow solid (960 mg, 7 1%).1H NMR (6, 300 MHz, C6D): 0.33 (s, 12H, CHSi(CH3);0.52 (s, 9H, Si(CH3)3);0.97 (d,2H, SiCH2P,2JpH = 9.2 Hz); 1.10 (d of d, 12H, P[CH(CH]3Jp.H = 13.0 Hz, 3JHH = 7.3Hz); 1.18 (d of d, 12H, P[CH(CH3J3Jp-H = 13.4 Hz, 3JH..H = 7.3 Hz); 1.35 (hr s, 2H,CH2Si(CH3);2.01 (d of sept, 4H, P[CH(CH3)2]2Jp.H = 4.1 Hz, 3JH-H = 7.3 Hz). 31P{1H}NMR (6, 121.421 MHz, C6D): 8.48 (s).references on page 2iChapter II 1102.lOc Synthesis of Dinitrogen Complexes2.lOc.1{[(Pr’PCHSiMe)N1ZrCI}(t-ri: i-4,2.2. This complex was made accordingto the published procedure.8 Resonance Raman (cm-1): Solid, 258, 317, 393, 518, 579, 636,731, 991,1046; Solution (THF), 262,321,331,585,733,745, 1030. MS (El) m/z: 1066 (Mj,1023, 981, 935, 803, 350, 262.2.lOc.2{[(PrPCHSiMe)N]ZrC1}2(J.-rl:-52),2.2. The nitrogen-15 analogue of 2.2was prepared by introducing 15N2 gas into the flask containing the degassed reaction mixture.Workup was carried out under normal N2. 15N{1H} NMR (8, 30.406 MHz, C7D8): 350.92 (s).15N MAS NMR (6, 40.60 MHz): 345 (s). Resonance Raman (cm-1): Solid, 258, 317, 339, 518,579, 636, 709, 968, 1024; Solution (THF), 260, 318, 327, 710, 718, 1028. MS (El) m/z: 1068(M), 1025, 983, 937, 805, 350, 262.2.lOc.3{[(PrPCHSiMeN]Zr(1SCH}(.LrIl: LN2.9. A solution of Zr(r5-C5H5)-C1[N(SiMeH2PPr)2] (0.980 g, 1.58 mmol) in toluene (60 mL) was transferred into a thick-walled reaction flask (300 mL) containing Na/Hg (85 g of 0.33% amalgam, 12.17 mmol of Na).The flask was then cooled to —196 °C, filled with 1 atmosphere of N2, sealed, and allowed towarm slowly to room temperature with stirring. Upon warming up to room temperature thereaction mixture quickly turned green (1 hour) and then slowly turned deep brown. After thedisappearance of the green colour (3 days) the solution was decanted from the amalgam andfiltered through a layer of Celite. The amalgam was extracted with 15 mL portions of toluene(total of 60 mL), until the extracts showed no brown colour. The filtrate and the extracts werecombined and stripping off the solvent gave a dark brown powder (0.860 g, 96%) in high purity(>95% by NMR spectroscopy). The product was crystallized from a 1:1 mixture of toluene andpentane. 1H NMR (6, 300 MHz, C13D6): 0.28 (s, 12H, Si(CH3)2); 0.33 (s, 12H, Si(CH3)2;1.18(m, 48H, P[CH(CH3)2J;1.54 (broad, 8H, SiCH2P); 1.88 (m, 4H, P[C[I(CH)212); 2.54 (m,4H, P[CH(CH3)2]2); 6.11 (broad, —10H, C5H). 31P{1H} NMR (6, 121.421 MHz, C7D8): at 20°C 20.26 (s); at -40 °C, AB quartet at 19.95 (d, 2Jp..p = 80.7Hz) and 20.57 (d, 2Jp.p 80.7Hz).13C{H} NMR (6, 50.323 MHz, C6D): 6.46 (s, Si(CH3)2; 6.71 (s, Si(CH3)2; 11.63 (s,SiCH2P); 18.40 (s, P[CH(CH)2];20.94 (s, P[CH(CH3)2J;21.35 (s, P[CH(CH]);101,37references on pageChapter II 111(s, C5H). Anal. Calcd forC46H98N4P4Si4Zr2•0.5C512:C, 50.13; H, 9.02; N, 4.82. Found: C,50.11; H, 8.94; N, 4.90. Resonance Raman (cm-1): Solid, 226, 277, 295, 325, 1126, 1211;Solution (toluene), 225, 290, 320, 1125, 1201. MS (El) m/z: 1124 (Mj, 1081, 993, 949, 732,688, 562, 519, 497, 475, 445, 431, 350, 262.2lOc.4 {[(Pr2PCHSiMe2)2N1Zr(fl-Hs)}(p-fl:1-5N), 2.9. The nitrogen- 15 analoguewas prepared by a procedure similar to the one described in section 2.lOc.3, by introducing 15N2gas into the flask containing the degassed reaction mixture. Workup was carried out underunlabelled N2. 15N{1H} NMR (8, 30.406 MHz, C7D8): 353.95 (s). Resonance Raman (cm4):Solid, 228, 277, 294, 323, 1125, 1172; Solution (toluene), 227, 290, 319, 1124, 1165. MS (El)m/z: 1126 (M), 1083, 995, 951, 734, 690, 563, 520, 498, 476, 446, 432, 350, 262.2.lOc.5{[(PriPCHSiMeN]Zr OAr*)}(irIl: ll2.12. A solution of Zr(OAr*)C12[N(SiMeCHPPr)](1.05 g, 1.48 mmol) in toluene (100 mL) was transferred into a thick-walled reaction flask (300 mL) containing Na/Hg (80 g of 0.17% amalgam, 5.74 mmol of Na).The flask was then cooled to —196 C, filled with 1 atmosphere of N2, sealed, and allowed towarm slowly to room temperature with stirring. The colorless solution slowly takes on the deepblue colour of the product. The reaction mixture was stirred for 5 days the solution wasdecanted from the amalgam and then filtered through a layer of Celite. The amalgam-containing residue was extracted with several 50 mL portions (approx. 400 mL) of toluene, untilthe extracts showed no blue colour. The filtrate and the extracts were combined and strippingoff the solvent gave a deep blue solid which was washed with hexanes (2x25 mL). Pure productwas obtained by slow evaporation of a toluene solution of the crude product at roomtemperature (0.36 g, 40%). 1H{31P) NMR (8, 500 MHz, C7D8): See Table 2.2031P(1H} NMR (6, 121.421 MHz, C7D8): at 20 °C, major isomer 8.69 (s); minor isomer 8.85 (s)and 11.26 (s). NOEDIFF experiments (6, 400 MHz, C7D8): radiating the resonances at 7.00 or6.98 ppm showed enhancements at 2.34 and 2.32. Variable temperature 31P(1H} NMR (6,121.42 MHz, C7D8): upon cooling a sample of pure major isomer, the resonance at 8.69 ppmbroadens and below -40 begins to show shoulders at 7.60 ppm and 9.10 ppm. Below -78 °Ca broad peak begins to appear at 4.00 ppm which increases in intensity with decreasingreferences on pagejChapter H 112temperature (up to -93 °C). Anal. Calcd for a sample containing only the major isomer,C26H53ON2P2SiZr: C, 50.44; H, 8.63; N, 4.53. Found: C, 50.70; H, 8.87; N, 4.33. Anal.Calcd for a sample containing a mixture of major isomer and minor isomer (major:minor = 2:1),C26H53ON P2SiZr:C, 50.44; H, 8.63; N, 4.53. Found: C, 50.24; H, 8.71; N, 4.29. ResonanceRaman (cm-1): Solid, 258, 278, 314, 350, 595, 732, 751, 1046. MS (El) m/z: 1236, 1193, 1107,1063, 1007, 975, 695, 350, 262.Table 2.20 1H(31P} NMR data for complexes 2.12 and 2.12a.Groups Major isomer (8) Minor isomer(8)Si(CH3)2 0.37 (s, 12H) 0.29 (s, 6H), 0.32 (s. 6H)0.41 (s, 12H) 0.32 (s, 6H), 0.35 (s, 6H)SiCH2P 1.30 (d, 4H,2JHH = 1.6 Hz) obscured1.32 (d, 4H,2JH-H = 1.6 Hz)P[CH(CH3)2] 1.04 (d, 12H, 3JH..H = 6.7 Hz) 1.13 (d, 6H), 1.26 (d, 6H)1.05 (d, 12H, 3HH = 6.7 Hz) 1.34 (d, 6H), 1.44 (d, 6H)1.17 (d, 12H, 3JH.H = 7.3 Hz) rest of the resonances were1.23 (d, 12H, 3H-H = 7.3 Hz) obscuredP[CH(CH3)2] 2.03 (sept. 4H,3JH-H = 7.3 Hz) 2.04 (sept. 2H, JH.H = 7.4 Hz)2.39 (sept. 4H,3JH..H = 6.7 Hz) 2.20 (sept. 2H,3JH..H = 7.4 Hz)2.48 (sept. 2H,3JH-H = 7.4 Hz)2.53 (sept. 2H,3JH-H = 7.4 Hz)2, 6-MePh 2.34 (s, 12H) only assignable resonance was abroad peak at 2.32p-Ph 6.66 (t, 2H, 3JH..H = 7.1 Hz) 6.62 (t, 2H,3JH.H = 6.9 Hz)rn-Ph 7.00 (t, 4H, 3JH..H = 7.1 Hz) 6.98 (t, 4H,3JH-H = 6.9 Hz)references on pageChapter II 1132.lOc.6 {[(Pr12PCHSiMe)N]Zr(OAr*}2(1:rIJSN),2.12. The nitrogen- 15 analoguewas prepared by a procedure similar to the one described in section 2.lOc.5, but by introducing‘5N2 gas into the flask containing the degassed reaction mixture. Workup was carried out underunlabelled N2. 15N{1H} NMR (6, 30.406 MHz, C7D8): minor isomer, 342.91 (s); major isomer,339.06 (s). Resonance Raman (cm-1): Solid, 278, 310, 350, 577, 595, 725, 751, 1046. MS (El)m/z: 1238, 1195, 1111,977,696,350,262.2.lOc.7{[(PriPCHSiMeN]Zr(OBut)}(J.Ii1Z:112.12. A solution of crude Zr(OBut)Cl2[N(SiMe2CH2PPr)2] (approx. 1.05 g, 1.48 mmol) from the reaction described in section2.lOb.7 was dissolved in toluene (100 mL) and transferred into a thickwalled reaction flask (300mL) containing Na/Hg (80 g of 0.30% amalgam, 10.4 mmol of Na). The flask was then cooledto —196 °C, filled with 1 atmosphere of N2, sealed, and allowed to warm slowly to roomtemperature with stirring. The colorless solution slowly takes on a deep purple colour of theproduct. The reaction mixture was stirred for 5 days and the solution was decanted and filteredthrough a layer of Celite. Stripping off the solvent from the filtrate gave a dark purple oil.Attempts to crystallize the product were not successful. 1H NMR (6, 300 MHz, C6D): 0.46and 0.42 (s, 12H, Si(CH3)2; 1.24 (br m, 28H, SiCH2Pand P[CH(CH3)2j;1.42 (s, 9H,OC(CH3); 1.97 (br sept. 2H, P[CH(CH3)3JH-H = 7.2 Hz); 2.25 (br sept. 2H,P[CH(CH2],3JH-H = 6.6 Hz). 31P{1H) NIVIR (6, 81.015 MHz, C6D): 8.16 (s). 15N{’H}NIVIR (6, 30.406 MHz, C7D8): 346.41 (s); 334.35 (s); 319.64 (s); 254.87 (s); 248.86 (s).2.lOc.8 {[(PrP HSiMeN]ZrB }( -i:1), 2. 19. A solution ofZrBr3[N(SiMe2CH2-PPr’)}(0.628 g, 0.943 mmol) in toluene (80 mL) was transferred into athick-walled reaction flask (300 mL) containing 0.35% Na/Hg (24 g, 3.59 mmol). The flaskwas then cooled to —196 °C, filled with 1 atmosphere of N2, sealed, and allowed to warm slowlyto room temperature with stirring. The colorless solution slowly takes on the deep blue colourof the product. The reaction mixture was stirred for 7 days and the solution was decanted andfiltered through a layer of Celite. The amalgam-containing residue was extracted with 80 mLportions (approx. 1L) of toluene, until the extracts showed no blue colour. Stripping off thesolvent from the combined filtrate and extracts gives a dark blue crystalline material (0.23greferences on page Z;LChapter II 11442%). 1H NMR (6, 300 MHz, C7D8): 0.30 (s, 12H, Si(CH3)2;0.34 (s, 12H, Si(CH3)2;1.02(m, 8H, SiCH2P); 1.44 (d of d, 12H, P[CH(CH]3JH..H = 7.3 Hz, 2J..p = 13.7 Hz); 1.24 (m,36H, P[CH(CH3)2];2.18 (d of sept, 4H, P[CH(CH3)2]3JH.H = 7.5 Hz,2JH-P = 4.2 Hz); 2.42(d of sept, 4H, P[CH(CH3)2]3JH.H = 7.6 Hz, 2JH-P 3.9 Hz). 31P{1H} NMR (6, 121.4 MHz,C7D8): 9.96 (s). Anal. Calcd forC18H44BrN2P2Si2Zr: C, 37.42; H, 7.68; N, 4.85. Found: C,37.84; H, 7.84; N, 3.89. According to the microanalysts comments, the complex shows lowernitrogen content than expected due to the formation of nitrides during the combustion.2.lOc.9{[(Pr1PCHS1MeN]ZrBr}(p.-ri:i-’5),2.19. The nitrogen-15 analogue wasprepared by a procedure similar to the one described in section 2.lOc.8, but by introducing ‘5N2gas into the flask containing the degassed reaction mixture. Workup was carried out underunlabelled N2. ‘5N(1H} NMR (6, 30.406 MHz, THF / C6D6): 345.75 (s). Resonance Raman(cm-1): Solid, 295, 589, 709, 880, 1009, 1167, 1293. MS (El) mlz: 1158, 1113, 1025, 941, 895,350, 262.2.lOc.10{[(PrPCHSiMeN]Zr}C1Br(.t-i:-N),2.20. A solution containing a mixtureofZrBr3[N(SiMe2CH2PPr’)](0.160 g, 0.221 mmol) and ZrCl3[N(SiMe2CH2PPr’)2J (0.131g,0.222 mmol) was treated with 0.34% Na amalgam (12.2 g, 1.79 mmol) as described in thesynthesis of ([(Pr2PCH2SiMe)2N]ZrBr} ($.t-1:fl- 1H(31P} NMR (6, 500 MHz, C7D8):0.28 (s, Si(CH3)2; 0.29 (s, Si(CH3)2; 0.31 (s, Si(CH3)2; 0.32 (s, Si(CH3)2; 0.33 (s,Si(CH3)2;0.34 (s, Si(CH3)2;0.65 (m, SiCH2P); 1.00 (m, SiCH2P); 1.28 (m, P[CH(CH3)2]2);1.41 (m, P[CH(CH)21;2.40 (m, P[CH(CH3)2J.1P{1H} NMR (3, 121.4 MHz, C7D8): 10.30(s); 10.06 (s); 9.96 (s); 9.89 (s). MS (El) mlz: 1111, 1067, 1023, 979, 262.2.lOc.11{[(Pr2PCHS1MeN]Zr}2C1B (.i-rI:-5N2.20. The nitrogen- 15 analogue wasprepared by a procedure similar to the one described in 2.lOc.10, but by introducing 15N2 gasinto the flask containing the degassed reaction mixture. Workup was carried out underunlabelled N2. 15N{1H} NMR (6, 30.406 MHz, C4H80/ C6D): 352.50 (s, approx. iN);349.26 (s, approx. 2N); 345.75 (s, approx. iN). Resonance Raman (cm): Solid, 264, 300, 313,579, 709, 884, 976, 1021. MS (El) m/z: 1113, 1069, 1025, 981, 895, 849, 350, 262.references on page 112Chapter II 1152.lOd Reactions Involving Dinitrogen Complexes2.lOd.1{[(PrPCHSiMe)N]ZrCI}(j.t-ii:-5and NaCp.DME. To a solution of([(PPCHSiMeNjZrC1)t-r:1-’5(30 mg, 0.028 mmol) in THF (4 mL) was added asolution of NaCp.DME (11 mg, 0.062 mmol) dissolved in TFIF (2 mL), under an atmosphere ofunlabelled dinitrogen. A control experiment was set up under an atmosphere of argon. Bothreactions were stirred at room temperature until they turned brown (6 days). The solvent wasremoved in vacuo, and extraction of the crude product with pentane and striping off the solventgave a brown powder. The brown powder was shown to be { [(Pr2PCH2SiMe2)N]Zr-(r15-CH)1 (.i.-N) by NMR. A similar reaction carried out in an NMR tube showed cleanconversion of the side-on complex to the end-on complex. MS (El) mlz: 1126 M, 1083, 995,951,734, 690.2.lOd.2 Protonation Reactions. The zirconium dinitrogen complex { [(PrPCH2SiMe)N]Zr-(ri5-C5H5)}.t-1’:T’-N),2.9 was protonated by the addition of excess dry HC1 gas to atoluene solution of the complex. Extraction of the organic phase with distilled water andcalorimetric analysis to determine the amount of hydrazine produced by the method of Watt andChrisp”3gave the following results:Table 2.21 Moles of hydrazine measured from the reaction of {[(Pr2PCHSiMe)N]Zr-(fl5-CsHs)}i-rI’:r’-N) complex in toluene with anhydrous HC1 gas. * Expectedconcentration was calculated by taking into account the amount of pentane present in thecomplex.Sample Weight *Expected Concentration Measured Concentration RatioofNH2—NH of NTI2—NH2 Measured/Expected0.023 g 5.09 10-6 M 5.43. 10-6 M 1.070.037g 6.33•10M 7.08•106M 1.12references on pageChapter II 116The reactions with 1M HC1 solution and with water were carried out with a gas uptakeapparatus. An appropriate amount of sample was weighed into the glass bucket and wascovered with KBr. The proton source (HC1aq) or water) was syringed into the flask andallowed to equilibrate at a known pressure of dinitrogen. Then the bucket was dropped and theflask was vigorously shaken and the pressure change was measured after 15 minutes. Aqualitative test with the colour developer used in the Watt and Chrisp method showed thepresence of hydrazine.Table 2.22 Mole percent gas evolved during the reaction of {[(Pr2PCHSiMe)N]Zr-(15-C5H5)-r1:TIN)complex with H20 and HC1 solution. * Percent gas evolutions werecalculated by taking into account the amount of pentane present in the complex. ** The data inparenthesis was obtained after 30 minutes.Sample Weight Proton Source Measured Pressure % Gas evolution*0.066 g 1 M HC1 351 mniHg 90%0.037 g H20 349 mmHg 81% (114%)**Protonation reactions with pyridinium hydrochloride were carried out inside a NMRtube. Complex 2.6 or 2.2 (approx. 30 mg) was dissolved in C7D8 (1 mL) and added an excessofC5HN•HC1 (5 mg) and shaken occasionally until the colour of the N2 complex dissappeared.1H NMR data for complex 2.25: 0.40 and 0.55 (s, Si(CH3)2,1.48 and 2.35 (m, P[CH(CJzT3)2]),4.99 (s, NNH2). 31P( ‘H) NMR has a broad peak centered at 14 ppm.2.lOd.3 {[(Pr2PCH2SiMe2)2N]ZrX}2(J.t-fl: 1-N )where X = Cl, Br or OAr*, and L1BEt4.The side-on dinitrogen complexes were reacted with LiBEt4 in an approximately 1:2 mole ratioin toluene. The deep blue colour of the reaction mixture slowly decreased in intensity to give ayellow solution over a period of 7 days. Stripping off the toluene and extracting the residuewith pentane and subsequent stripping off pentane gave a yellow oily substance which seemedto consist of a single complex. Monitoring the reaction by NMR spectroscopy suggested anreferences on page 112Chapter II 117almost quantitative conversion of the N2 complex to the product. The following spectroscopicdata is for a reaction carried out in a sealed NMR tube.1H(31P} NMR (6, 500 MHz, C7D8): 0.05 (d of quartet, 1H, 2JH.H = 10.2 Hz, 3H-H = 7.1 Hz);4.81 (s, 1H); 0.19, 0.31, 0.33, 0.44, 0.48, 0.52, 0.54 and 0.95 (s, each 3H, Si(CH3)2;1.62, 1.70,1.78, 1.82, 1.85, 1.95, 2.15 and 2.25 (sept, P[CH(CH3)212, each 1H, 3H..H 7.0 Hz); 1.51, 1.37and 1.31 (t, B(CH2CH3)3JHH = 8.3 Hz); 0.6 to 0.85 (overlapping singlets, SiCH2); 0.9 to 1.3(P[CH(CH3)2]2 and B(CH2CH3).31P{1H} NMR (6, 121.4 MHz, C7D8): Spectrum consistedof an ABMX spin system; 22.28 (d, 1P, 2Jp..p 71.7 Hz); 18. 56 (d, 1P,2Jp.p = 71.7 Hz); 12.56(s, 1P); 0.06 (s, 1P). 15N{1H} NMR (6, 30.406 MHz, C7D8): 250.5 (d, approx. iN,2JNN = 12Hz); 185.9 (d, approx. iN, 21N-N = 12Hz). “B(’H) NMR (6, 96.2 MHz C7D8): 0.00 (s, freeBEt3); -41.6 (br., coordinated BEt3); -104.1 (s, unreacted LiBEt4).2.lOd.4{[(PrPCHSiMeNJZrX}(j.t-N Where X = Cp, Cl or OAr*, and AlkylHalides. A typical reaction involved a toluene solution of the dinitrogen complex and varyingequivalents of the alkyl halides. Appropriate equivalents of the volatile alkyl halides Mel,CH2Br2 and BrCH2CH2rwere transfered into the reaction flask by using a constant volumegas bulb. The constant volume bulbs were filled with the vapor of the alkyl halide at roomtemperature at the vapor pressure of the pure alkyl halide and then condensed into the reactionflask. Benzyl bromide was made up into a standard toluene solution and then an appropriatevolume of the solution was used.The isolation of {[(Pr2PCH2SiMe2)2N]ZrCp}2(N2).C7Br,2.26. Anal. Calcd forC53H1OBrN4P4Si4Zr2: C, 49.08; H, 8.16; N, 4.32. Found: C, 49.44; H, 8.19; N, 4.21.2.lOe Molecular Orbital Calculations. All molecular orbital calculations were performed onthe CAChe Worksystem, a product developed by Tektronix. The parameters used in theINDO/i and Extended Huckel semi-empirical molecular orbital calculations were used asprovided by the CAChe Work system, which in turn were taken from the literature.45’934 Thestructural parameters for the model [(H3P)2(NH)ZrC112(I-1:fl- 2),F were taken from theliterature.8 The structural parameters for the models[(H)(N)ZrCpJj.t-:i- ,A andreferences on pageChapter II 118H were obtained from the X-ray crystal structureanalysis of complexes { [(Pri2PCH2SiMe2)2N]Zr(r15C5H5)} 2(JI11l :i1-N2), 2.9 and([(Pri2PCH2Si)21Zr(OAr*)}2(.ifl:N ),2.12a. The symmetry of the models H andA were restricted to C2v and C2 respectively. The hydrogen atoms were aligned to satisfy theoverall symmetry of the model. For all the models the following standard bond lengths wereused: P—H = 1.38 A, C—H = 1.09 A, N—H = 1.07 A. The fragment model(H3P)2NH2)ZrCl, Gwas obtained from the literature,45 and the model (H3P)2(NH2)ZrCp, B was constructed fromthe model A.The model 0 was used to obtain the intermediate cases for the rotation of the side-onbound dinitrogen to the end-on bonding mode. The dinitrogen ligand was rotated in the planedefined by the Zr2N2 core (which also contains the phosphine ligands) without altering therelative positions of the ligands on the metal centers. A phantom atom “A” was placed in thecentre of the dinitrogen ligand and the rotation angle was measured as showed in Figure 2.16.The shortest zirconium-nitrogen bond distance between the zirconium and dinitrogen was keptidentical to that in model F.— S— — —Figure 2.25 The models used to generate the intermediate structures for the bending P, androtation 0, of side-on bound dinitrogen complexes. The phantom atom “A” isplaced in the middle of the N—N bond. The line X—Y in model P refers to theaxis of the hinge about which the Zr2N2plane was bent.H3P H3PHPH3PH3PHPH30references on page flQChapter II 119The model P was used to obtain the intermediate cases for the bending of the side-onbound dinitrogen complex. The bending was carried out by moving the clinitrogen ligand on aplane perpendicular to the zirconium-zirconium axis while having the zirconium-”A” distanceconstant. Also the dihedral angle H—N—Zr—O, where the hydrogen and nitrogen corresponds tothe amide, was set to 900. The angle e was measured by taking Zr—”A”--Zr bond angles. Thebond angle parameters associated with “A”—Zr—Z, where Z P, N or Cl were maintainedconstant. The process can be imagined as bending of the Zr2N core using the nitrogen-nitrogen axis as a hinge.2.11 References.(1) Fryzuk, M. D. Can. J. Chem. 1992, 70, 2849.(2) Fryzuk, M. D.; MacNeil, P. A. .1. Am. Chem. Soc. 1981, 103, 3592.(3) Fryzuk, M. D.; MacNeil, P. A.; Rettig, S. 3.; Secco, A. S.; Trotter, J. Organometallics1982, 1, 918.(4) Fryzuk, M. D.; Montgomery, C. D. Coord. Chem. Rev. 1989, 95, 1.(5) Fryzuk, M. D.; Haddad, T. S.; Berg, D. 3.; Rettig, S. 3. Pure Appi. Chem. 1991, 63, 845.(6) Pearson, R. G. J. Chem. Ed. 1968,45, 581.(7) Pearson, R. G. J. Chem. Ed. 1986,45, 643.(8) Fryzuk, M. D.; Haddad, T. S.; Rettig, S. 3. J. Am. Chem. Soc. 1990, 112, 8185.(9) Fryzuk, M. D.; McManus, N. T.; Rettig, S. J.; White, G. S. Angew. Chem., mt. Ed. Engi.1990,29, 73.(10) Fryzuk, M. D.; MacNeil, P. A.; Rettig, 5. 3. J. Am. Chem. Soc. 1987, 109, 2803.(11) Fryzuk, M. D.; Carter, A.; Westerhaus, A. Inorg. Chem. 1985,24, 642.(12) Fryzuk, M. D.; Gao, X.; Joshi, K.; MacNeil, P. A.; Massey, R. L. J. Am. Chem. Soc.1993, 115, 10581.(13) Fryzuk, M. D.; Gao, X. 1993, Unpublished work.(14) Fryzuk, M. D.; Haddad, T. S.; Rettig, S. J. Organometalllcs 1988, 7, 1224.(15) Fryzuk, M. D.; Haddad, T. S.; Rettig, S. I. Organometallics 1989,8, 1723.(16) Fryzuk, M. D.; Carter, A.; Rettig, 5. 3. Organometallics 1992, 11, 469.(17) Sutton, L. E. Tables ofInteratomic Distances and Configurations in Molecules and Ions;Chemical Society: London, 1958; Vol. No. 11.(18) Manriquez, 3. M.; Bercaw, J. E. J. Am. Chem. Soc. 1974, 96, 6229.Chapter II 120(19) Manriquez, I. M.; Sanner, R. D.; Marsh, R. E.; Bercaw, J. E. J. Am. Chem. Soc. 1976,98, 3042.(20) Manriquez, I. M.; McAlister, D. R.; Sanner, R. D.; Bercaw, J. E. J. Am. Chem. Soc.1976, 98, 6734.(21) Sanner, R. D.; Manriquez, J. M.; Marsh, R. E.; Bercaw, J. E. J. Am Chem. Soc. 1976,98, 8351.(22) Bush, M. A.; Sim, G. A. .1. Chem. Soc. (A) 1971, 2225.(23) Engelhardt, L. M.; Papasergio, R. I.; Raston, C. L.; White, A. H. Organometallics 1984,3, 18.(24) Cardin, D. I.; Lappert, M. F.; Raston, C. L. Chemistry of Organo-Zirconium andHafnium Compounds; 1 ed.; Ellis Horwood Limited: Toronto, 1986.(25) Buchwald, S. L.; King, S. M. J. Am. Chem. Soc. 1991, 113, 258.(26) Grossman, R. B.; Davis, W. M.; Buchwald, S. L. J. Am. Chem. Soc. 1991, 113, 2321.(27) Houri, A. F.; Didiuk, M. T.; Zu, X.; Horan, N. R.; Hoveyda, A. M. J. Am. Chem. Soc.1991, 1993, 6614.(28) Lipshutz, B. H.; Ellsworth, E. L. J. Am. Chem. Soc. 1991, 112, 7441.(29) Negishi, E. Pure Appi. Chem. 1981,53, 2333.(30) Negishi, E.; Takahashi, T. Aidrichimica Acta 1985, 18, 31.(31) Nigishi, E.; Takahashi, T. Synthesis 1988, 1.(32) Takagi, K.; Rousset, C. 3.; Negishi, E. J. Am. Chem. Soc. 1991, 113, 1440.(33) Takahashi, T.; Hasegawa, M.; Suzuki, N.; Subari, M.; Rousset, C. J.; Fanwick, P. E.;Negishi, E. J. Am. Chem. Soc. 1991, 113, 8564.(34) Takahashi, T.; Seki, T.; Nitto, Y.; Saburi, M.; Rousset, C. 3.; Negishi, E. J. Am. Chem.Soc. 1991, 113, 6266.(35) Tidwell, 3. H.; Senn, D. R.; Buchwald, S. L. J. Am. Chem. Soc. 1991, 113, 4685.(36) Yoshida, T.; Nlgishi, E. J. Am. Chem. Soc. 1981, 103, 4985.(37) Fryzuk, M. D.; Erker, G.; Mylvaganam, M. Private Communication.(38) Erker, G.; Wilker, S.; Kruger, C.; Goddard, R. J. Am. Chem. Soc. 1992, 114, 10983.(39) Erker, G.; Temme, B. .1. Am. Chem. Soc. 1992, 114, 4004.(40) Burk, M. J. J. Am. Chem. Soc. 1991, 113, 8518.(41) Peterson, E. 3.; Dreele, R. B. V.; Brown, T. M. Inorg. Chem. 1979, 15, 309.(42) Archer, R. D.; Day, R. 0.; llhingsworth, M. L. Inorg. Chem. 1976, 18, 2908.(43) Wilkinson, P. G.; Hounk, N. B. J. Chem. Phys. 1956,24, 528.Chapter II 121(44) Allen, F. H.; Kennard, 0.; Watson, D. G.; Brammer, L.; Orpen, A. G.; Taylor, R. J.Chem. Soc., Perkin Trans. 1987,11, Si.(45) Fryzuk, M. D.; Haddad, T. S.; Mylvaganam, M.; McConville, D. H.; Rettig, S. J. J. Am.Chem. Soc. 1993, 115, 2782.(46) Evans, W. J.; Ulibarri, T. A.; Ziller, I. W. J. Am. Chem. Soc. 1988, 110, 6877.(47) Blum, L.; Williams, I. D.; Schrock, R. R. J. Am. Chem. Soc. 1984, 106, 8316.(48) Chow, P. C. Ph. D. Thesis, UBC, 1993.(49) Turner, H. W.; Fellmann, 3. D.; Rocklage, S. M.; Schrock, R. R.; Churchill, M. R.;Wasserman, H. 3. J. Am. Chem. Soc. 1980, 91, 6512.(50) Fryzuk, M. D.; Haddad, T. S.; Berg, D. 3. Coord. Chem. Rev. 1990, 99, 137.(51) Walsh, P. J.; Hollander, F. 3.; Bergman, R. G. J. Am. Chem. Soc. 1988, 110, 8729.(52) Archer, R. D.; Day, R. 0.; lllingsworth, M. L. Inorg. Chem. 1976, 18, 2908.(53) Lubben, T. V.; Wolczanski, P. T.; vanDuyne, 0. D. Organometallics 1984,3, 977.(54) Hunter, W. E.; Hrncir, D. C.; Bynum, R. V.; Penttila, R. A.; Atwood, 3. L.Organometallics 1983,2, 750.(55) Mason, 3. Chem. Rev. 1981, 81, 205.(56) Sanner, R. D.; Duggan, D. M.; McKenzie, T. C.; Marsh, R. E.; Bercaw, 3. E. J. Am.Chem. Soc. 1976, 98, 8358.(57) Duchateau, R.; Gambarotta, S.; Beydoun, N.; Bensimon, C. J. Am. Chem. Soc. 1991,113, 8986.(58) Beydoun, N.; Duchateau, R.; Gambarotta, S. J. Chem. Soc., Chem. Commun. 1992, 244.(59) Mtunzi, S. D.; Richards, R. L. J. Chem. Soc., Dalton Trans. 1984, 469.(60) Lazarowych, N. 3.; Morris, R. H.; Ressner, 3. M. Inorg. Chem. 1986,25, 3926.(61) Manriquez, J. M.; McAlister, D. R.; Rosenberg, E.; Shiller, A. M.; Williamson, K. L.;Chan, S. I.; Bercaw, 3. E. .1. Am. Chem. Soc. 1978, 100, 3078.(62) McConville, D. H. Ph. D. Thesis, UBC, 1991.(63) Rocklage, S. M.; Turner, H. W.; Fellmann, 3. D.; Schrock, R. R. Organometallics 1982,1,703.(64) O’Regan, M. B.; Liu, A. H.; Finch, W. C.; Schrock, R. R.; Davis, W. M. J. Am. Chem.Soc. 1990, 112, 4331.(65) Haddad, T. S. Unpublished results.(66) Mylvaganam, M. Unpublished results.(67) Resonance Raman Spectroscopy As An Analytical Tool; Melveger, A. 3., Ed.; TheFranklin Institute Press: Philadelphia, 1977.Chapter II 122(68) Baldwin, M. J.; Root, D. F.; Pate, 3. E.; Fujisawa, K.; Kitajima, N.; Solomon, E. I. .1. Am.Chem. Soc. 1992, 114, 10421.(69) Jolly, W. L. The Inorganic Chemistry OfNitrogen; W A Benjamin, Inc.: New York,1964.(70) Dung, J. R.; Pate, C. B.; Harris, W. C. J. Chem. Phys. 1972,52, 5652.(71) Dollish, F. R.; Fateley, W. G.; Bentley, F. F. Characteristic Raman Frequencies ofOrganic Compounds; John Wiley and Sons: Toronto, 1974.(72) Treitel, I. M.; Flood, M. T.; Marsh, R. E.; Gray, H. B. .1. Am. Chem. Soc. 1969, 91, 6512.(73) Chatt, J.; Duncanson, L. A. J. Chem. Soc. 1953, 2939.(74) Dewar, M. J. S. Bull. Soc. Chim. Fr. 1951, 79.(75) Chatt, 3. Pure Appi. Chem. 1970,24, 425.(76) Rosi, M.; Sgamellotti, A.; Tarantelli, F.; Floriani, C.; Cederbaum, L. S. J. Chem. Soc.,Dalton Trans. 1989, 33.(77) Sellmann, D. Angew. Chem., mt. Ed. EngI. 1974, 13, 639.(78) Chatt, J.; Leigh, G. 3. Chem. Soc. Rev. 1972, 1, 121.(79) Powell, C. B.; Hall, M. B. Inorg. Chem. 1984,23, 4619.(80) Albright, T. A.; Burden, 3. K.; Whangbo, M. H. Orbital Interactions in Chemistry; AWiley-Interscience Publication: New York, 1985.(81) Kubacek, P.; Hoffmann, R.; Havias, Z. Organometallics 1982, 1, 180.(82) Schleyer, P. R.; Maslak, P.; Chandrasekhar, J.; Grey, R. S. Tetrahedron Lett. 1993,34,6387.(83) Sannigrahi, A. B.; Kar, T. J. Chem. Ed. 1988, 65, 674.(84) Wiberg, K. A. Tetrahedron 1968,24, 1083.(85) Boca, R.; Pelilcan, P.; Haring, M. J. Mo!. Cat. 1981, 11, 41.(86) Pelikan, P.; Boca, R. Coord. Chem. Rev. 1984,55, 55.(87) Yamabe, T.; Hori, K.; Mlnato, T.; Fukui, K. Inorg. Chem. 1980, 19, 2154.(88) Hori, K.; Asai, Y.; Yamabe, T. Inorg. Chem. 1983,22, 3218.(89) Goldberg, K. I.; Hoffmann, D. M.; Hoffmann, R. Inorg. Chem. 1982,21, 3863.(90) Hoffmann, R.; Chen, M. M. L.; Thorn, D. L. Inorg. Chem. 1977, 16, 503.(91) Blomberg, M. R. A.; Siegbahn, P. E. M. J. Am. Chem. Soc. 1993, 115, 6908.(92) Richardson, D. E. J. Chem. Ed. 1993, 70, 372.(93) Evans, W. 3.; Ulibarri, T. A.; Ziller, J. W. J. Am. Chem. Soc. 1988, 110, 6877.(94) Ho, 3.; Drake, R. 3.; Stephan, D. W. J. Am. Chem. Soc. 1993, 115, 3792.(95) Morokuma, K. Acc. Chem. Ses. 1977, 10, 294.Chapter II 123(96) Fryzuk, M. D.; Lloyd, B. R.; Clentsmith, G. K. B.; Rettig, S. J. J. Am. Chem. Soc. 1991,113, 4332.(97) Henderson, R. A.; Oglieve, K. E. J. Chem. Soc., Dalton Trans. 1990, 1093.(98) Glassman, T. E.; Liu, A. H.; Schrock, R. Inorg. Chem. 1991,30,4723.(99) SmithIll, M. R.; Cheng, T. Y.; Hilihouse, G. L. J. Am. Chem. Soc. 1993, 115, 8638.(100) Churchill, M. R.; Li, Y. 3.; Blum, L.; Schrock, R. Organometallics 1984,3, 109.(101) vanTamelan, E. E. Ace. Chem. Res. 1970, 11, 361.(102) Coiquhoun, H. M. Ace. Chem. Res. 1984, 17, 23.(103) Kawaguchi, M.; Hamaoka, S.; Mori, M. Tetrahedron Lett. 1993,34, 6907.(104) Coilman, 3. P.; Hegedus, L. S.; Norton, J. R.; Finice, R. G. Principles and Applications ofOrganotranstion Metal Chemistry; second edition ed.; University Science Books: MillValley, CA., 1987.(105) Henderson, R. A.; Leigh, G. 3.; Pickett, C. J. Adv. Inorg. Chem. Radiochem. 1983,27,197.(106) Nakamura, A.; Yoshida, T.; Cowie, M.; Ibers, S. 0. A. J. Am. Chem. Soc. 1977, 99,2108.(107) Burrell, A. K.; Clark, G. R.; Rickard, C. E. F.; Roper, W. R. J. Chem. Soc., DaltonTrans. 1991, 609.(108) Joule, 3. A.; Smith, G. F. Heterocyclic Chemistry; second ed.; Van Nostrand Reinhold(UK) Co.Ltd: Cambridge, 1984, pp 345.(109) Wayda, A. L.; Darensbourg, M. Y. In ACS Symp. Ser. Vol. 357: ExperimentalOrganometallic Chemistry; American Chemical Society: Washington, D. C; 1987.(110) Shriver, D. F.; Drezdzon, M. A. The Manipulation ofAir-Sensitive Compounds; 2nd ed.;John Wiley and Sons: New York, 1986.(111) Smart, 3. C.; Curtis, C. 3. Inorg. Chem. 1977, 16, 1788.(112) Perrin, D. D.; Armarego, W. L. F. PurWcation ofLaboratory Chemicals; 3 ed.;Pergoman Press: Toronto, 1988.(113) Watt, G. W.; Chrisp, J. D. Anal. Chem., 1954, 24, 2006.(114) Wang, C; Bodenbinder, M; Wiliner, H; Rettig, S; Trotter, J; Aubke, F. Inorg. Chem.1994, 33, 779.(115) Rasetti, F. Proc. Nati. Acad. Sci. U. S. A. 1929, 15, 431.LqChapter 3Introduction to Zirconium(III) Chemistry3.1 GeneralIt is well established in transition metal chemistry that the first row elements show somedeviations from their second and third row counterparts.1’2For example, if one considers formaloxidation states, then for complexes of the group 4 elements, titanium exhibits +3 and +4 formaloxidation states predominantly,2which is in stark contrast to the heavier members of the group,namely zirconium and hafnium, for which the +4 formal oxidation state dominates.3 In fact,organometallic and coordination compounds of zirconium(III) and hafnium(llI) are very poorlydocumented in the literature and still remain a scarcity.3The rarity of the +3 formal oxidation state in zirconium is partly attributed to therelatively low potential at which it is reduced from the +4 to the +3 state. Reduction potentialsfor various zirconocene derivatives range from B0 = -1.6 V to E0 = -2.1 v.3’4 In comparison totheir titanium counterparts, for example the first reduction potential for CpTiC12 is E0 = -0.75V, zirconium derivatives are reduced approximately 1 V more negative.3’4This large differencein reduction potential is in part due to the coordinative solvation (by THF) of the zirconiumcenter.3’4 Electrochemical studies involving complexes containing bulky ligands, for examplecomplexes of the typeCp2M[C6H4(CHSiMe3)-l,21,gave E0 values of -1.46, -2.02 and -2.26 Vfor M = titanium, zirconium and hafnium respectively, where the differences between thereduction potentials of titanium and zirconium are only 0.56 V.5,6A comparison of the ionic radius of titanium (for 4, 5, 6 and 8 coordination numbers theeffective ionic radii are 56, 65, 74.5 and 88 ppm respectivly)7and zirconium (for 4, 5, 6, 7, 8and 9 coordination numbers the effective ionic radii are 73, 80, 86, 92, 98 and 103 ppmrespectivly)7 in their +4 formal oxidation state suggests that for complexes with samecoordination number and identical ligands, a titanium center in +3 formal oxidation state will bemore effectively shielded than the a zirconium(llI) center. Therefore, it has been suggested thatthe zirconium(ffl) centers could be kinetically activated to undergo further reactions.3’4references on pace 133Chapter III 1253.2 Coordination Complexes of Zirconium(Ill)A number of adducts of ZrX3 (X Cl, Br and I) with nitrogen containing donor ligandssuch as ammonia, pyridine, acetonitrile, bipyridyl and phenanthroline have been prepared.Although these can be considered as zirconium(ffl) derivatives the evidence is circumstantial.These compounds are highly colored and have unusual stoichiometries; for example, the metalto-ligand ratio is 1:2 for ZrCI3(py)2, 1:1.5 for Zr2C16(bipy)3and 1:2.5 for Zr2C16(MeCN)5.8”The magnetic properties of these complexes are lower than what would be expected for acomplex, and suggest dinuclear or oligonuclear structures with some magnetic exchangebetween neighboring zirconium atoms. Due to the low solubility of these complexes molecularweight measurements were not possible and therefore any structural proposals are necessarilyspeculative.ZrC14 + 2PR3 toluene [Zr2C16(PR3)4]Na/Hg3.1PR3 = PMe3,PEt3,PPr3,PBuT133.2[Zr2C16(PR3)4] [Zr2C16(PR3)] + PR3____+PR30.5 [Zr2C16(PR3)4]— [ZrC13(PR)2] [ZrC13(PR)1 3.3-PR3The class of complexes [ZrC12(p,-Cl)(PR3)2]2, (R Me, Et, Pr or Bus) prepared by thereduction of ZrC14 in the presence of an alkyl phosphine was the first well characterizedcoordination complexes of zirconium(Ill), with complex [ZrCl2(p-Cl)(PBu3)]3.1, beingstructurally characterized.12’3 The solid state structure of [ZrCl(.t-Cl)(PBu’,3.1, iscentrosymmetric with the zirconium atoms positioned in a pseudo octahedral environment. Theshort zirconium-zirconium separation of 3.182(1) A in 3.1 is attributed to a metal-metal bondand is close to that of zirconium metal, 3.1789 A. The NMR spectroscopic studies with thecomplexes analogous to 3.1, [ZrC12(L-C1)(PR3)2]2, indicate that they undergo reversiblereferences on pace 133Chapter III 126phosphine dissociation (Equation 3.2). Also the ESR spectra of a toluene solution of thesecomplexes show a weak signal which becomes more intense when excess phosphine is added,implying the break-down of the dimer into monomers (Equation 3.3).Recently, complexes analogous to 3.1 have been synthesized; addition of 4-tert-butyl-pyridine to complex 3.1 gave [ZrCl2(.tCl)(C5H4NBut4)j,which upon reaction with ButCNgave a dinuclear complex containing a bridging ButCN ligand where the bonding of the nitrileligand could be described as (L.112:11 1), 3.2.’CMe3 CI(Bu)3Pss,.\ .ssCL,, /0Py(Bun)aP/%ScV PY0/CI’” \cI3.3 Organometallic Complexes of Zirconium(Ill)3.3.1 Dinuclear Organometallic Complexes of Zirconium(Ill)Most of the attempted syntheses of organometallic zirconium(III) derivatives haveresulted in the formation of dinuclear and diamagnetic complexes.’3’26 One class of dinuclearzirconium(ffl) complexes that contain fulvalene type ligands were synthesized by the reductionof dihalozirconocene derivatives17’20or by disproportionation reactions involving Zr(II) andzirconium(IV) precursors (Scheme 3.1). During the course of the reaction, coupling of two Cpligands leads to the formation of the fulvalene ligand which bridges the two metal centers. Thesecond class of complexes are of the type (Cp’2Zr)Q.t-X),where Cp’ = C5H orC5H4Me andthe bridging ligands are either halide groups’9 or dialkyl phosphide21 groups. The bridginghalide derivatives are prepared by the photochemical reaction of the monoalkyl derivatives,Cp’2ZrXBui (Cp’ =C5H4Me) and the corresponding phosphide derivatives were prepared by theCIreferences on pane 133Chapter III 127reduction of Cp2ZrC12 (Cp’ = C5H5) in the presence of Me2P—PMe2. The hafnium derivativesof the latter case (i.e., with bridging phosphides) have also been synthesized.Cp2ZrC1 Na/HgCp2ZrC1 + Cp2Zr(PMe3)(ri5-CH4Me)2ZrX(Bu)hvristeneX = Cl, Br and I, 3.5(fl5-CH)2M 1 + Me2P—PMeScheme 3.1Considerable attention has been focused on the metal-metal separations in the dinuclearzirconium(III) derivatives and their relationship to the possible existence of metal-metalbonding.29 In fact, the incorporation of a fulvalene dianion as a bridging ligand allows one toconstrain the two metal centers within short distances i.e., in the range of zirconium-zirconiumbonds.29 The compilation of metal-metal separations given in Table 3.1 shows that thezirconium-zirconium separations range from 3.099 A to 4.171 A. Of all the known dimers, onlythe complexes of the type[C5H3(SiMe)-l,3]2Zr(i-X)]2 where X = Cl, 3.7 (Scheme 3.4), Brand I, show paramagnetic behavior.26 The diamagnetic behaviour of the complexes with metal-metal separations less than 3.4 A is attributed to metal-metal bonding, whereas for the caseswith metal-metal separations greater than 3.4 A ligand mediated superexchange is invoked.However, ab initio calculations done with zirconium dimers having metal-metal separationranging from 3.6 A to 4.0 A, for example,[(C5Hs)2Zr(t-PMe2)j2, 3.6, have shown that there is3.3M = Zr, 3.6 or Hfreferences on pace 133Chapter III 128significant d-orbital interactions between the two zirconium centers of the zirconocenefragments (i.e., up to 4 A separation between “Cp2Zr”), suggesting the existence of the “superlong” metal-metal bonds.30Table 3.1 Selected zirconium-zirconium distances. (C10H8refers to the fulvalene ligand).Complex metal-metal separation referencef-Zr 3.1789 A 2713-ZrCl 3.07 A 28ZrBr3 3.16 A 28Zn3 3.32 A 28[Zr(.t-Cl)(Cl2)(dppe)] 3.099(2) A 16[Zr(.t-Cl)Cl(PBut’3],3.1 3.182 A 13[(C10H8)52r(t- l)3.3 3.233(1) A 23[(ClOH)(C2Zr(.L-I)2, 3.4 3.472(1) A 19[(C5H4Me)2Zr(p.-I)],3.5 3.649 (1) A 19[(CH5)2Zr(.L-PMe)]3.6 3.653 (2) A 21[C5H3(SiMe3)-1 ,312Zr(p.-X)12X = Cl, 3.7 3.90 AX=Br 4.101(1)A 26X= I 4.171(2)AMost of the chemistry associated with the dinuclear zirconium(ffl) complexes has beenrestricted to their synthesis and X-ray structure elucidation. The oxidation of the fulvalenederivatives (e.g., complex 3.3 in Scheme 3.1) gives Zr(IV)-Zr(IV) dimers having cis (i.e., bothzirconium centers on the some side of the fulvalene ligand) or trans geometries.29 It wasexpected that during the oxidation of the fulvalene bridged Zr(III)-Zr(III) dimers toZr(IV)-Zr(IV) dimers the cis geometry would be retained, however, experimental evidencereferences on page 133Chapter III 129shows that the cis geometry is retained only when good bridging ligands (e.g., S2 or 02) arepresent, whereas trans geometry was found for poorly bridging ligands (e.g., methyl group).The thermal or photochemical disproportionation reactions of the complexes of the type[Cp2Zr(p-X)],where X = Cl, Br and I, giving zirconium(IV), (Cp2ZrX and Zr(ll), (“Cp2Zr”)species is probably the only reactivity know for bimetallic zirconium(llI) complexes.’9 Studiesshow that the complexes with bridging chloride ligands were the most susceptible towardsdisproportionation whereas the iodide and bromide analogues were significantly slower. Animportant feature of these reactions is that their disproportionation was accelerated in thepresence of coordinating solvents and in the presence of ligands such as CO and butadiene. Forexample, reacting [Cp2Zr(I.L-I)] with CO gives an equimolar ratio of Cp2ZrI2 andCp2Zr(CO).’93.3.2 Mononuclear Organometallic Complexes of Zirconium(Ill)In most instances mononuclear zirconium(ffl) species were generated in situ underphotochemical or reducing conditions; typically a common synthetic procedure involvedperforming reactions inside the cavity of an ESR spectrometer.31-42 Photochemical reactions ofzirconium(IV) zirconocene derivatives leads to the homolysis of a zirconium-carbon bond,forming the corresponding organic radical and the zirconium(III) species. The photolysis ofCp2ZrC1 led to the formation of CpZrCl2 and a Cp radical, whereas the photolysis ofCp2ZrMeC1 gave Cp2ZrC1 and a Me radical.4’ When the dialkyl or diaryl derivatives, Cp2ZrR(where R = alkyl or Ph) were irradiated, the ESR spectral features consisted of a doublet whichhas been attributed to the formation of a mononuclear hydride species.38The reduction of dialkyl zirconocene derivatives with sodium dihydronaphthylide haveshown the formation of an anionic zirconium(llI) species, some of which were stable up to 12hours at room temperature. The anionic species, formed by the reduction ofCp2Zr(CHSiMe3),gave a binomial quintet, coupling due to the methylene protons of the twoalkyl groups, suggesting that it is a mononuclear species formulated as [Cp2Zr(CHSiMe3)1[Na(THF)x].4 By comparison, the ESR spectrum of a similar titanium species, for example,references on page 133Chapter III 130[Cp2TiClJ[Na(THF),J, shows coupling due to the interaction of the d1 metal center with asolvated sodium ion (23Na, I = 3/2, The absence of such features in the ESRspectrum of the zirconium species is attributed to the steric bulk of the alkyl groups whichprobably hinders the formation of ion pairs.Mononuclear neutral zirconium(III) intermediates have been synthesized by thereduction of the corresponding monoalkyl zirconocene derivatives, Cp2ZrC1R, with sodiumamalgam. Among these neutral zirconium(III) derivatives, complexes containing adiphenyiphosphinomethyl group, CH2PPh are particularly interesting because of their abilityto form a three membered chelate ring upon formation of the zirconium(III) species (Scheme3.2).374045 The complex 3.7 was formed by reducing the precursor,Cp2Zr(CHPPh)Cl,withNa/Hg whereas, complexes 3.8 and 3.9 were formed by electrochemical reduction (Scheme3.2).3740.46ElectrochemicalR=R =Electrochemical 0c0CH2PPhR = R= reductionCH2PPh-Cp03.8Scheme 3.2It was found that a toluene solution ofCp2Zr(CHPPh)3.7 slowly decomposes to givea diamagnetic dinuclear species, the formation of which is accelerated in THF solvent. On thebasis of NMR spectroscopy, the mechanistic pathway shown in Scheme 3.3 is proposed for theNa/HgTHFR = ClR = CH2PPh3.73.9references on pane 133Chapter III 131decomposition of 3•7•42,47 This study is important because it shows that the instability of thezirconium(II1) complexes is at least partly due to their ability to undergo bimoleculardisproportionation reactions which in turn sets off a series of other reactions.‘PPh2 Ppph2Zr(IV)—Zr(IV)PPh2Zr(ll)—Zr(IV)Scheme 3.3The first step involves a bimolecular disproportionation of 3.7 to form a Zr(II)—Zr(IV)dimeric intermediate, where the zirconium(II) center oxidatively adds a C-H bond of a Cpligand to give a Zr(IV)—Zr(TV) intermediate with one bridging (it-ri1:fl5) Cp ligand.22’3 ThisZr(IV)—Zr(IV) intermediate reductively eliminates one MePPh2 unit to form a Zr(III)—Zr(III)species which in turn disproportionates to give another Zr(H)—Zr(IV) intermediate. The Zr(II)Zr(II)—Zr(IV)Zr(llI), Zr(llI)/-PPh2MeZr(Ill)—ZrWI)(1) -PPh2MePPPh2Zx(IV)—Zr(IV)(2) +2 PPhMeZr(TII)—Zr(Ill)references on pace 133Chapter III 132center of this intermediate can C—H activate a second Cp ring and eventually lead to theformation of a second Zr(IV)—Zr(IV) dimer with two bridging (t-fl1:115)Cp ligands. Thereductive elimination of the second mole of MePPh2 gives the final product which is aZr(llI)-Zr(Ill) dimer.[N(n-Bu)4]3.12 [N(n-Bu)4fScheme 3.4ri‘S3.10 3.11 ITo our knowledge there are only four mononuclear zirconium(III) complexes that havebeen well characterized (i.e., including elemental analysis):(15-CMe)Zr(r8H8),4 3.10,(rI5CH3But21,3)2ZrC1,49 3. 1 1, [B u4N]{ {(ri5-CH3SiMe)21,3)2ZrC1],2 6 3• 12, andZr(1-)2(rN )[CH(SiMe];6of these, only three, 3.10, 3.11 and 3.12, have beencharacterized by X-ray crystallography. The anionic complex 3.12 was prepared by reacting thedinuclear Zr(III)—Zr(llI) dimer 3.7, with a tetraalkyl ammonium salt (Scheme 3.4), whereas theother three were made by the reduction of the corresponding zirconium(IV) halide precursor.Si3.7references on pace 133Chapter III 133These examples illustrate that with sufficiently sterically demanding ligands, the tendency of thezirconium(III) state to dimerize and possibly disproportionate42can be avoided. However, thechemistry of these complexes has not been explored. In fact, there are no examples of simple,well characterized hydrocarbyl or hydride complexes of zirconium(ffl) known.3.4 SummaryThe chemistry of zirconium(III) complexes is rather poorly developed, with most of theattention being focused on the synthesis of such complexes. The dinuclear Zr(III)—Zr(III)complexes seem to be favored over the corresponding mononuclear counterparts. It appears thatthe disproportionation of zirconium(llI) complexes to give Zr(II) and zirconium(IV) species isthermodynamically favored, which in part can be attributed to the stability of the zirconium(IV)complexes. The incorporation of bulky ligands around zirconium(III) centers can providekinetic stability, which in turn has led to the isolation of some mononuclear zirconium(llI)complexes.In the following chapter the preparation of a series of mononuclear zirconium(III)complexes which also incorporate hydrocarbyl and hydride ligands will be discussed. Theseparamagnetic zirconium(Ill) derivatives provide the first opportunity to examine the reactivityof Zr—C and Zr—H bonds in d1 complexes.3.5 References(1) Cotton, F. A.; Wilkinson, G. Advanced Inorganic Chemistry; Third Edition ed.; WileyInterscience: New York, 1972.(2) Greenwood, N. N.; Earnshaw, A. Chemistry of the Elements; Pergamon Press: Oxford,1984; Vol. p. 1116.(3) Cardin, D. J.; Lappert, M. F.; Raston, C. L. Chemistry of Organo-Zirconium andHafnium Compounds; 1 ed.; Ellis Horwood Limited: Toronto, 1986.(4) Lappert, M. F.; Pickett, C. J.; Riley, P. I.; Yarrow, P. I. W. J. Chem. Soc., Dalton Trans.1981, 805.(5) Dumond, D. S.; Richmond, M. G. .1. Am. Chem. Soc. 1988, 110, 7547.Chapter III 134(6) Gladysz, 3. A.; Tam, W. J. Am. Chem. Soc. 1978, 100, 2545.(7) Shannon, R. D. Acta Crystallogr. 1976, A32, 751.(8) Fowles, G. W. A.; Russ, B. 3.; Willy, G. R. Chem. Commun. 1967, 646.(9) Fowles, G. W. A.; Willey, G. R. J. Chem. Soc. (A) 1968, 1435.(10) Larsen, E. N.; Hehzler, T. E. Inorg. Chem. 1974, 13, 581.(11) Nicholls, D.; Ryan, T. A. Inorg. Chim. Acta 1977,21, L17.(12) Wengrovius, J. H.; Schrock, R. R. J. Organomet. Chem., 1981,205, 319.(13) Wengrovius, J. H.; Schrock, R. R.; Day, C. S. Inorg. Chem. 1981,20, 1844.(14) Hoffman, D. M.; Lee, S. Inorg. Chem. 1992,31, 2675.(15) Ho, 3.; Drake, R. J.; Stephan, D. W. J. Am. Chem. Soc. 1993, 115, 3792.(16) Cotton, F. A.; Diebold, M. P.; Kibala, P. A. Inorg. Chem. 1988,27, 799.(17) Ashworth, T. V.; Agreda, T. C.; Herdtweck, E.; Herrmann, W. A. Angew. Chem.., mt.Ed. Engi. 1986,25, 289.(18) Wielstra, Y.; Gambarotta, S.; Meetsma, A.; Boer, 3. L. d. Organometallics 1989, 8, 250.(19) Wielstra, Y.; Gambarotta, S.; Meetsma, A. Organometallics 1989, 8, 2948.(20) Cuenca, T.; Herrmann, W. A.; Ashworth, T. V. Organometallics 1986,5, 2514.(21) Chiang, M. Y.; Gambarotta, S.; Boihuis, F. v. Organometallics 1988, 7, 1864.(22) Gell, K. I.; Harris, T. V.; Schwartz, 3. Inorg. Chem. 1981,20, 481.(23) Gambarotta, S.; Chiang, M. Y. Organometallics 1987, 6, 897.(24) Gambarotta, S.; Wielsira, Y.; Spek, A. L.; Smeets, W. J. 3. Organometallics 1990, 9,2142.(25) Edwards, P. G.; Howard, J. A. K.; Perry, 3. S.; Al-Soudani, A. R. J. Chem. Soc., Chem.Commun. 1991, 1385.(26) Hitchcock, P. B.; Lappert, M. F.; Lawless, G. A.; Olivier, H.; Rayn, E. 3. .1. Chem. Soc.Chem. Commun. 1992,474.(27) Pearson, W. B. Handbook ofLattice Spacing Structures ofMetals and Alloys;Pergomon: New York, 1976; Vol. 2, p 90.(28) Dahi, L. F.; Chiang, T.; Seabaugh, P. W.; Larsen, E. M. Inorg. Chem. 1964,3, 1236.(29) Wielstra, Y.; Gambarotta, S.; Spek, A. L.; Smeets, W. J. 3. Organometallics 1990, 9,2142.(30) Benard, M.; Rohmer, M. M. J. Am. Chem. Soc. 1992, 114, 4785.Chapter III 135(31) Choukroun, R.; Gervais, D. J. Chem. Soc., Chem. Commun. 1985, 224.(32) Choukroun, R.; Basso-Bert, M.; Gervais, D. J. Chem. Soc., Chem. Commun. 1986, 1317.(33) Choukroun, R.; Dahan, F.; Larsonneur, A. M.; Samuel, E.; Peterson, J.; Meunier, P.;Somay, C. Organometallics 1991, 10, 374.(34) Jones, C. B.; Petersen, 3. F. J. Am. Chem. Soc. 1983, 105, 5503.(35) Jones, C. B.; Petersen, J. F. Inorg. Chem. 1981, 20, 2889.(36) Samuel, E. Inorg. Chem. 1983,22, 2967.(37) Schore, N. E.; Hope, H. J. Am. Chem. Soc. 1980, 102, 4251.(38) Hudson, A.; Lappert, M. F.; Pichon, R. J. Chem. Soc., Chem. Commun. 1983, 374.(39) Gynane, M. J. S.; Jeffery, J.; Lappert, M. F. J. Chem. Soc., Chem. Commun. 1978, 34.(40) Etienne, M.; Choukroun, R.; Gervais, D. J. Chem. Soc., Dalton Trans. 1984, 915.(41) Atkinson, J. M.; Brindiley, P. B.; Davis, A. G.; Hawaii, I. A. A. J. Organomet. Chem.1984,264, 253.(42) Blandy, C.; Locke, S. A.; Young, S. J.; Schore, N. E. .1. Am. Chem. Soc. 1988, 110,7540.(43) Kenworthy, 3. G.; Myatt, J.; Todd, P. F. Chem. Commun. 1969, 263.(44) Olive, H.; Olive, S. Angew. Chem., mt. Ed. Engi. 1968, 7, 386.(45) Manzer, L. E. J. Am. Chem. Soc. 1978, 100, 8068.(46) Raoult, Y.; Choukroun, R.; blandy, C. Organometallics 1992, 11, 2443-2446.(47) Schore, N. B.; Young, S. 3.; Olmstead, M. M. Organometallics 1983,2, 1769.(48) Rogers, R. D.; Teuben, 3. H. J. Organomet. Chem. 1989,359, 41.(49) Urazowski, I. F.; Ponomaryev, V. I.; Nifantev, I. E.; Lemenovskii, D. A. J. Organomet.Chem. 1989,368, 287.13Chapter 4Zirconium(llI) Complexes: Synthesis and Reactivity4.1 GeneralA survey of the literature on zirconium(III) complexes in Chapter 3 has shown thatalmost all of the stable mononuclear organometallic derivatives contain bulky ligands such as[15-CH3(CMe3)2-1,3]-1, [r15-CH(SiMe)21,31 1 or cyclooctatetraenyl ligand, [TI8-CH]2(Chapter 3). By comparison the tridentate ligand [N(SiMe2CHPR)j-PNP is somewhatisoelectronic to the Cp ligand and, in addition, the steric bulk of PNP can be considerablyaltered by changing the substituents on the phosphorus, for example R = Me, Ph, Pr or But.Therefore by incorporating bulky PNP ligand systems that are comparable to the bulky Cpbased ligands such as[T15-CH3(CMe)21,31-’, it should be possible to stabilize mononuclearzirconium(III) complexes. It is also important to note that the syntheses of dinitrogencomplexes (e.g., 2.2, 2.9, 2.12 etc.) were only successful with the isopropyl version of the PNPligand, [N(SiMe2CH2PPr) ]’ whereas reactions involving complexes containing the lessbulky PNP ligand [N(SiMeCHPMe]-1,gave intractable material (Chapter 2). This lastresult seems to suggest that incorporating the isopropyl substituents on PNP could provideadequate kinetic stability for stabilizing reactive intermediates.4.2 Synthesis of Zirconium(Ill) and Hafnium(llI) ComplexesIt has been previously shown (Section 2.2.2) that the reduction of the zirconium(IV)complexZr(r15-CH)[N(SiMeCH2PPr1],2.6 with excess Na/Hg under dinitrogen leadsto the formation of the dinuclear dinitrogen complex {[(PPrCSiMeN]Zr(T5-C} 2-(-N2), 2.9. During this process there is a color change from yellow to deep green whichpersists for hours followed by a slow change to deep brown, the color of the dinucleardinitrogen derivative. By control of reaction times, the intermediate was intercepted in highyield (88%) as dark green crystals having the formulaZr(ri5-CH)Cl[N SiMe2HPPr],4.1(Scheme 4.1). This highly colored (?max 324 nm, Emax 2600 M1cm)zirconium(III)references on pageChapter IV 137chioro derivative, in contrast to other unstable (see Chapter 3) zirconium(III) derivatives,exhibits high thermal stability.1’2Pr’2Me2Si Na/Hg\ tolueneM=Zr,2.6 M=Zr,4.1M=Hf,4.2 M=Hf,4.3Scheme 4.1The 1H NMR spectrum of 4.1 is consistent with it being a paramagnetic system sinceonly a broad resonance is observed and no signals were observed in the 31P{1H} NMRspectrum. Moreover, the solution (in toluene) ESR spectrum recorded at room temperatureshows a binomial triplet at g = 1.955 with coupling to two phosphorus-31 nuclei and satellitesdue to one magnetically dilute zirconium nucleus (a(91Zr) = 37.2 G, 91Zr, 11.23%, I = 5/2),which is consistent with a mononuclear species in solution. The ESR spectrum of 4.1 wassimulated using the following parameters: a(31P) = 21.1 G, 2P; a(14N) = 2.9 G, iN; a(1H) = 1.8G, 5H; linewidth = 2.3 G. The solution magnetic measurements of 4.1 gave a value of ff =1.57 B. M.*The reduction of the hafnium(IV) precursorHf(q5-C5)Cl2[N(SiMe2CH2PPr’)2] 4.2,with Na/Hg gave a brown material, from which deep brown crystalline cubes were isolated byrecrystallization. The ESR spectrum of the recrystallized material consisted of a triplet and abroad singlet, where the triplet resonance was identical to the resonance of the zirconium(III)complex 4.1 (Figure 4.1). The broad signal is assigned to the hafnium(III) speciesHf(ri5-C)C1[N(SiMe2HPP],4.3. An intriguing feature of the spectrum shown in* This value is lower than the spin only value of 1.73 B. M. The lower magnetic moment could be atthbuted tomagnetic exchange, i.e., some degree of association between mononuclear species; or to spin-orbit coupling. Thecalculated spin-orbit coupling constant ofZr3,500 cm4,should give a magnetic moment of about 1.2 B.M.74references on pageChapter IV 138Figure 4.1 is that the signals due to the zirconium(III) and the hafnium(III) species haveapproximately equal intensity. However, it will be shown later that the sample consists mainlyof hafnium(ffl) complex in approximately 10:1 ratio. This suggests that the ESR sensitivity ofthe zirconium(Ill) complex is approximately ten times greater than that of hafnium(III).21.1 GCommercially available hafnium precursors (e.g., HfCl4)contain zirconium impurities inthe range of 1 to 4%. Therefore, most organometallic complexes of hafnium are likely to have avery small percentage of the zirconium analogue as an impurity. During the reduction of thehafnium(IV) derivatives the hafnium intermediates possibly undergo decomposition at a higherrate than the zirconium counterparts, which would eventually lead to the enrichment ofzirconium analogue in the product. The presence of zirconium impurities would also hamperthe isolation of analytically pure hafnium(III) samples (Table 4.1). To our knowledge theHfCpC1[N(SiMe2HPPr1),4.3‘,ZrCpCl[N(SiMe2CHPPr’)],4.1Figure 4.1 Overlapping room temperature ESR spectra of a solution (toluene) sample ofHfCpCl[N(SiMe2HPPr’)]4.3, and ZrCpCl[N(SiMe2HPPri)],4.1.references on pageChapter IV 139hafnium(HI) complex 4.3 is the only complex that has been characterized by satisfactorymicroanalysis.Table 4.1 Elemental composition found for some hafnium(Ill) derivatives.Complex C H Other referencesHfCpCl[N(SiMe2CH2PPr’)J, 4.3 NCalculated 41.13 7.35 2.09 this workFound 41.49 7.49 1.96[HfC13(dippe)2] ClCalculated 30.7 5.89 19.4 4Found 31.7 6.05 18.3Hf(fl5-C5Me5)(r7C) HfCalculated 50.43 5.48 44.09 3Found 51.86 5.63 44.554.3 Synthesis of Zirconium(ffl) Hydrocarbyl ComplexesRMTolueneRM = MeMgC1, EtLi, PhLi,PhCH2K,Me3SiCH2LiR=Me4.5, Et4.6, Ph 4.4,CH2Ph 4.7, CH2SiMe34.8Scheme 4.2The metathesis reaction of Zr(T15-CH)C1[N(SiMe2HPPri],4.1 with PhLi at -78 Cin toluene gives an instantaneous color change from a green to a deep red solution.Recrystallization from the crude product gave diamond shaped crystals; these were shown to be4.1p1J2references on pageChapter IV 140Zr(15-CH)Ph[N(SiMe2CH2PPr2],4.4. The solution (in toluene) ESR spectrum of complex4.4 recorded at room temperature shows a broad binomial triplet at g = 1.981 indicatingcoupling to two phosphorus-31 (a(31P) = 20.3 G) nuclei, however no satellites due to thecoupling of the zirconium-91 nucleus were observed.The zirconium(llI) alkyl complexesZr(fl5-CH)R[N(SiMeCH2PPr) where R = Me,4.5; Et, 4.6; PhCH2, 4.7 and Me3SiCH2,4.8, were synthesized using different alkyl transferreagents (Scheme 4.2). A typical reaction was carried out with a toluene solution of the chioroderivative 4.1 and an alkyl transfer reagent, usually at room temperature. Unlike the phenylderivative, all of the ailcyl derivatives are deep green in colour. Isolation of micro-analyticallypure samples of the complexes with small alkyl groups, for example methyl, 4.5 and ethyl, 4.6was hampered by the incorporation of LiCl or MgC12 in the samples. Table 4.2 gives theelemental compositions found for the hydrocarbyl derivatives of 4.1 and it can be noted that thevalues for complexes 4.5 and 4.6 are lower than expected. For the ethyl derivative, 4.6 theanalytical data for the crystals from the first recrystallization and from the secondrecrystallization seem to fit for 0.6 and 0.2 equivalents of LiCl respectively. Also, for themethyl derivative 4.5 the crystals from the first recrystallization seem to contain 0.2 equivalentsof MgC12, which is further corroborated by the determination of the chloride content of thesample.It has been established that some zirconium(IV) complexes form adducts with LiC1,5 aphenomenon that could be attributed to the Lewis acidity and to the coordinative unsaturation ofthe zirconium(IV) center. Similar factors may be responsible for the incorporation of LiC1 andMgC12 into the zirconium(III) complexes 4.6 and 4.5 respectively. However, in thesezirconium(llI) derivatives, the interactions between the zirconium and the halide ion of the salts(i.e., LiCl or MgC12) are probably weaker and become insignificant as the steric bulk of thealkyl group is increased. The room temperature solution ESR spectra of the samples formulatedas 4.6•(LiCl)06 and 4.6•(LiC1)02were virtually identical. It is noteworthy that the ESRspectrum of the anionic zirconium species [Cp2Zr(CH2SiMe3]iNa(T F)J did not showreferences on pageChapter IV 141hyperfine interaction due to sodium;2 in contrast, the phosphide derivative [Cp2Zr(PEt)][Na(THF)x]+ shows hyperfine features due to the coupling of one sodium nucleus.6Table 4.2 Micro-analytical data for the hydrocarbyl derivatives ofZr(r15-CH)Cl[N(SiMeHPPr’)].Complex C H N ClZr(rj5-CH)Me[N(SiMe2CH2PPr)2], .4.5Calculated for 4.5 51.11 9.29 2.48Calculated for 4.5.(MgC12)o.2 49.28 8.96 2.39 2.67Found 48.90 9.08 2.39 2.52Zr(15-CH)Et[N(SiMeCHPPr],4.6Calculated for 4.6 51.95 9.42 2.42Calculated for 4.6•(LiCl)06 49.76 9.02 2.32found for 1St recrystallization 49.81 9.14 2.30Calculated for 4.6•(LiC1)02 51.20 9.28 2.39found for 2nd recrystallization 51.21 9.27 2.40Zr(15-H)Ph[N(SiMeCHPr],4.4Calculated 55.63 8.69 2.24Found 55.51 8.68 2.35Zr(rI-CH5)CH2Ph[N(SiMe2CHP r’)],4.7Calculated 56.29 8.82 2.19Found 56.58 8.86 2.25Zr(11-CH5)CHSiMe3[N(SiMeCHPPr)],4.8Calculated 50.97 9.51 2.20Found 50.68 9.46 2.21Figure 4.2 shows the X-ray structures of the phenyl derivative Zr(rj5-CH)Ph[N(SiMe2CH2PPr)]4.4, and the alkyl derivativeZr(1-CH5)CH2SiMe3[N(SiMe2CH2PPr)],4.8.These two examples are the first structurally characterized, mononuclear hydrocarbylderivatives of zirconium(III).references on page iChapter IV 142I IllI and III are ORTEP views showing the complete atom labeling scheme forcomplexesZr(r5-CH)Ph[N(SiMe2CH2PP )]4.4, andZr(fl5-CH)CH2SiMe3-[N(SiMe2CH2PPr)2] 4.8, respectively. II and IV are Chem 3D® views showingthe arrangements of ligands around the zirconium in complexes Zr(rj5-CH)Ph-{N(SiMe2CH2PPr)21 4.4, and Zr(15-H)CSiMe3[N(SiMeCHPPr)]4.8,respectively.IISiIvNSiSiCl 4c3Figure 4.2references on pageChapter IV 143Table 4.3 Selected bond lengths of complexesZr(T5-C5H)Ph[N SiMe2CH2PPr’)J4.4,andZr(fl5-CH)CH2SiMe[N(Si2HPPr’)],4.8.Zr(115-CH)Ph[N(SiMe2CHPr’)] Zr(i-C)CH2SiMe3[N(SiHPPr]4.4 4.8Atom Atom Distance (A) Atom Atom Distance (A)Zr P(l) 2.7819 (14) Zr P(l) 2.7923 (16)Zr P(2) 2.7843 (16) Zr P(2) 2.8563 (17)Zr N 2.231 (3) Zr N 2.216 (4)Zr C(31) 2.272 (4) Zr C(31) 2.337 (5)Zr Cp 2.225 Zr Cp 2.2162 (5)The structural parameters (Table 4.3 and 4.4) associated with the PNP ligand inboth complexes are similar to other zirconium(IV) derivatives.79 A comparison of bond anglesaround the zirconium centers suggests that the coordination environment can be envisaged as adistorted trigonal bipyramid with the phosphine donors of the meridionally coordinated PNPligand occupying the axial positions. The distances from the zirconium to the centroid of the Cpligand in the phenyl, 4.4 (2.225 A) and the alkyl, 4.8 (2.2162 (5) A) derivatives are slightlyshorter (the difference is greater than 0.05 A) than the distances observed for the zirconium(IV)derivatives 2.7 (2.286 A) and 2.9 (2.286 and 2.294 A). Shorter zirconium—Cp distances reflectthe decreased steric constrains around the zirconium(III) centers of 4.4 and 4.8 than thezirconium centers of 2.7 and 2.9.The zirconium-carbon bond distances associated with the phenyl, 4.4 (2.272 (3) A)and the alkyl, 4.8 (2.337 (5) A) derivatives are within the observed range for zirconiumcomplexes containing bulky alkyl groups, for example in Cp2ZrPh[CH(SiMe3)]thezirconium-Cp0and zirconium-C(alkyl) distances are 2.324 (7) and 2.329 (6) A respectively.10For zirconocene derivatives the zirconium-C(sp3)bond distances range from 2.25 1 (6) A to2.388 (12) A where the longer distances are associated with larger alkyl groups.2”°’ In thePNP derivativeZr(r4-CH6)Ph[N(SiMeCHP],the zirconium-Cjp0distance of 2.3 17 (7)A is slightly longer than what was observed in the zirconium(Ill) phenyl derivative, 4.4.references on page 122.Chapter IV 144Table 4.4 Selected bond angles of complexesZr(rl5-CH)Ph[N(SiMe2CH2PPr)2) 4.4,andZr(115-CH)CH2SiMe3[N(SiMe2CH2PPr)2], 4.8.Zr(5-CH)Ph[N(SiMe2CH2PPr)2] Zr(15-CH)CHSiMe3[N(SiMe2CH2PPr)2]4.4 4.8Atom Atom Atom Angle () Atom Atom Atom Angle (°)P(l) Zr P(2) 150.09 (4) P(1) Zr P(2) 151.17 (5)P(l) Zr N 79.47 (9) P(l) Zr N 77.83 (11)P(2) Zr N 75.01 (10) P(2) Zr N 74.37 (11)P(1) Zr C(31) 87.04 (11) P(l) Zr C(31) 85.13 (13)P(2) Zr C(31) 88.57(11) P(2) Zr C(31) 93.47 (13)P(1) Zr Cp 104.27 P(1) Zr Cp 103.95 (4)P(2) Zr Cp 104.82 P(2) Zr Cp 102.08 (4)N Zr C(3l) 112.60 (13) N Zr C(31) 102.47 (16)N Zr Cp 135.57 N Zr Cp 139.47 (10)C(31) Zr Cp 111.81 C(31) Zr Cp 118.05 (13)Zr C(31) C(36) 132.5 (5) Zr C(31) Si(3) 134.0 (3)Zr C(31) C(32) 114.9 (3) Si(3) C(31) C(33) 114.9 (3)Si(1) N Si(2) 119.15 (17) Si(1) N Si(2) 120.18 (23)Cp refers to the centroid of the Cp ligand. All structural parameters associated with the Cpligand for the complex 4.4 were taken from the Chem 3D® structure.The bond angles between the zirconium, ipso-carbon and the ortho-carbons inZr(4-CH6)Ph[N(SiMe2CHPr]are 121.1° (6) and 125.3° (6)’; however, the comparableangles in the zirconium(III) phenyl complex 4.4, Zr—C(31)—C(36) and Zr—C(31)—C(32) are132.5° (5) and 114.9° (3) A respectively. These angle parameters seems to suggest that in 4.4the phenyl ligand is twisted about the ipso-carbon towards the nitrogen of the PNP. The phenylgroup in complex 4.4 is coplanar with the equatorial plane and therefore is expected toreferences on page jQChapter IV 145experience significant steric interactions with the Cp ligand, where the centroid of the Cp ligandalso lies on the equatorial plane.It is noteworthy that attempts to metathesize the zirconium-chloride bond of 4.1 with theGrignard reagent BzMgC1 gave a diamagnetic compound (in approximately 20% isolated yield)as one of the products. Spectroscopic and elemental analysis of this compound are consistentwith the formulaZrBz2Cl[N(SiMeH2PPr)],4.9. It seems that the reaction of 4.1 withBzMgC1 or BzK involves more than one reaction pathway, one of which being the exchange ofthe Cp ligand of 4.1 with the benzyl group. This exchange reaction is less significant when BzKwas used, where the zirconium(III) benzyl derivative was the main product. It is noteworthythat the 31P{1H} NMR spectrum of the mother liquor of the aforementioned reactions showsthat one of the diamagnetic impurities is the end-on dinitrogen complex 2.9. This observationseems to suggest the formation of a zirconium(TI) species (which subsequently binds dinitrogen)during the reaction of 4.1 with benzyl transfer reagents.A possible mechanism to rationalize the formation of the zirconium(IV) dibenzylderivative 4.9 could involve the formation of the zirconium(III) benzyl intermediateZrBzCl[N(SiMe2CHPPr)]by the exchange of the Cp ligand of 4.1 rather than thereplacement of the chloride ligand. This intermediate, by comparison with the stablezirconium(ffl) complex ZrCpCl[N(SiMe2CH2PPr)2j 4.1, is sterically less crowded. Therefore,ZrBzCl[N(SiMe2CH2PPr)2] is likely to be kinetically labile towards disproportionation to givethe zirconium(IV) complex 4.9 and some unknown zirconium(II) species.Solution magnetic studies (Evans Method) on the CH2SiMe3 derivative gave a value of.teff. = 1.73 B.M., which is close to the calculated spin only value of a d1 zirconium(llI) center.The room temperature solution ESR spectra of the alkyl derivatives are more complicated thanthe spectrum of the phenyl derivative 4.4. With the exception of the benzyl derivative, thehyperfine interaction of the unpaired electron with the two phosphorus nuclei are in the range of20 to 21 G. In the case of the Et and the Bz derivatives, the satellites due to zirconium-91 nucleiwere not observed.references on page jChapter IV 146[Zr].CH2CH3ObservedSimulated[Zr]-CH2CHD[Zr]-CH2S1Me3ObservedSimulated[Zr] = “Z rCp[N(SiMe2HPPr’)]”Figure 4.3 The ESR spectra of the alkyl complexes A 4.6, B 4.6-d1,C 4.5 and D 4.8. For eachcase the observed spectrum is shown on the top and the simulated spectrum isshown below.A2OGBD[Zr]-CH3C2OGreferences on page 122Chapter IV 147Table 4.5 Hyperfine coupling constants (G) for the chloro and the hydrocarbyl derivatives ofZrCpR[N(SiMe2CH2PPr)2j.All the values were obtained from simulations.R g a(91Zr) a(31P) a(14N) a(’Ha) a(1H) a(’Hcp)Cl 1.955 37.2 21.1 2.9 1.8C6H5 1.981 20.7 20.3 1.8 1.1CH3 1.963 28.0 21.1 2.1 6.6CH23 1.962 20.9 2.6 9.3 3.2 1.8CH26H5* 1.956 18.6 3.4 3.2 1.21.973 30.5 21.4 2.0 9.3 6.2CH2Si(CH3)* The hyperfine values given for the benzyl derivative are only an approximate value.Simulation studies of the room temperature solution (toluene) ESR spectrum of themethyl derivative 4.5 suggest that the three cc-hydrogens of the methyl group attached tozirconium are isotropic. It is important to note that the minor features of the simulated ESRspectrum of the methyl complex 4.5 (Figure 4.3) using the values given in Table 4.5 are slightlydifferent from that of the actual spectrum. It is believed that this could be a result of theunsophisticated ESR simulation program, where only one overall line broadening value could beused. It is possible that a more accurate spectral simulation could be obtained using differentvalues of line broadening for each type of spin system, for example, a different line broadeningvalue for the two phosphorus nuclei and a different value for the three ct-hydrogens.An intriguing feature in the ESR spectra of the ethyl and CH2SiMe3derivatives is thatthe cc-hydrogens have different hyperfine coupling constants. This difference in hyperfineinteraction could be due to restricted rotation about the zirconium-carbon bond associated withthe alkyl substitution. If the rotation is slow on the ESR time scale, some rotational isomersreferences on pageChapter IV 148would have a longer lifetimes that the others.14’5 Scheme 4.3 shows the two types ofconformational isomers for the ethyl and the CH2SiMe3 derivatives, where type A hasdiastereotopic cc-hydrogens, and type B has enantiotopic cc-hydrogens. The diastereotopica-hydrogens of conformer A will have different hyperfine coupling constants.14’5Pr’2 Pr’ Pr’2 2Me2S(’\0 Me2S(’\0Me2Sç,/ Me2S1/RPr2 Pr’2A BR=Me, SiMe3Scheme 4.3An alternate explanation for the different coupling constants for the c-hydrogens wouldbe that, in solution, one of the a-hydrogens weakly interacts with the metal center, C (Scheme4.3). Infrared spectroscopic experiments were carried out to elucidate the presence of agosticC—H bond interactions. Any such bonding interactions of the c-hydrogens should be reflectedin the stretching frequencies associated with the carbon-hydrogen bonds of the a-hydrogens.The carbon-hydrogen stretching frequencies associated with the PNP ligand give rise to a verycomplicated pattern from 3000 to 2700 cm1. Therefore the solution infrared spectra ofdeuteride derivatives of the ethyl complex,Zr(rI5CH)CH2CH2D[N(SiMe2CHPP )],4.6-d1 andZr(15-CH)CD2C3[N(SiMeHPPr],4.6-d5 were analyzed. The infraredspectra of the monodeuterio derivative 4.6-d1 enabled the assignment the stretching frequenciesassociated with the -deuterides of the ethyl group and then comparison with the infraredspectra of the perdeuterio ethyl derivative 4.6-d5 enabled to assign the stretching bandsassociated with the a-deuterides.Creferences on pageChapter IV 1493200 2720. 2480Figure 4.4 The solution infrared spectra of complexes 4.6, 4.6-d1 and 4.6-d5A comparison of the infrared spectra of complexes 4.6-d1 and 4.6-d5 show that the latterhas an absorption band at 2043 cm-’ which can be assigned as an x-deuterium stretch. Also,this band is 85 cm1 lower than the next nearest carbon-deuterium stretching band, suggestingsome weakening of the C—D bond associated with the band at 2043 cm-1. The carbondeuterium stretching band at 2043 cm-’ will correspond to a carbon-hydrogen stretching bandaround 2783 cm4. For comparison, the agostic carbon-hydrogen stretching frequencies ofvarious transition metal complexes range between 2700 cm1 and 2350 cm1.6’8 However, insome rarer occasions higher C—H stretches have been reported for agostic C-H bonds. Forexample in the cationic complex [frH2(PPh3)L],where L = 8-methylquinoline, an agosticbond is formed between the iridium and the one of the C—H bonds of the methyl group, and anreferences on page i2[Zr]-CH2[Zr] = “ZrCp[N(SiMe2CHp ri)”[Zr]-CD2CD32960 2240. 2000.Chapter IV 150infrared band at 2848 cm-1 has been assign to the agostic C—H stretch.19 Therefore it is possiblethat one of the cc-hydrogen of the ethyl derivative could be weakly interacting with the metalcenter which in turn could be attributed to different coupling constants observed for the cchydrogens in the ESR spectra of the alkyl derivatives.The solution ESR spectrum of the benzyl zirconium(III) derivative 4.7, does not have acenter of symmetry, suggests that more than one species exist in solution. It is possible that thebenzyl ligand is coordinated in an 12 fashion, with a geometry similar to C shown in Scheme4.3. Such a geometry would make the two benzylic protons enantiotopic, and this proposal isconsistent with the hyperfine values obtained (by simulation) for the cc-hydrogens.LiPPh2,NaNPh2or PhONaY = OPh 4.10, NPh24.11 or PPh24.12Scheme 4.4Metathesis reactions with the zirconium(Ill) chloride complex 4.1 with one equivalent ofNaOPh, NaNPh2 or LiPPh2led to the synthesis ofZr(15-CH)OPh[N(SiMe2CH2PPr)2]4.10,Zr(15-CH)NPh2[N(SiMe2CHPPr1 4.11, and Zr(15-H)PPh2[N(SiMeHP r]4.12, respectively (Scheme 4.4). The phenoxy derivative was deep green in colour whereas theamido and the phosphido derivatives gave dark brown solutions. These complexes wereextremely soluble in hydrocarbon solvents and only the phenoxy derivative was isolated in asolid form amenable to elemental analysis. However, the ESR spectra of the crude sampleswere symmetric, and the 1H NMR spectra showed only broad peaks, suggesting that thesynthesis of these complexes proceeds in virtually quantitative yields. The room temperaturesolution ESR spectra of the amido derivative, 4.11 (g = 1.953, a(31P) = 11.2 G), and thephenoxy derivative, 4.10 (g = 1.955, a(31P) = 18.7 G), consist of a broad binomial tripletreferences on page4.1Chapter 1V 151indicating coupling of two equivalent phosphorus nuclei. The phosphido derivative, 4.12displays of a doublet of triplets (g = 1.965; a(31P) = 29.8 G, 1P; a(31P) = 18.6 G, 2P) which isconsistent with a mononuclear zirconium species (Figure 4.5). It is important to note that otherphosphido derivatives of zirconium(III) cited in the literature are reported to be binuclear anddiamagnetic containing bridging phosphido ligands (e.g., 3.6).Figure 4.5 The room temperature solution (toluene) ESR spectrum of the phosphide complexZr(rj-CH5)PPh2[N(SiMe2CH2PPr)2j, Oxidation of Zirconium(Ill) ComplexesTreating a solution of zirconium(llI) chioro derivative 4.1 with solid TiC13 or PbC12 ledto the oxidation of the zirconium complex to give the dichloro zirconium(IV) complex 2.6 inalmost quantitative yield. Also, 0.5 equivalents of Ph2S react with complex 4.1 to give thezirconium(IV) thiolate derivative,Zr(r-CH)(SPh)Cl[N(SiMeCHPPr)],4 . 13 inquantitative yields (Scheme 4.5). These reactions have been useful in quantifying the purity ofsome zirconium(Ill) derivatives. For example, during the synthesis of the alkyl complexZr(5-CHS)CH2SiMe3[N(SiMeCH2PPr)],4.8 an aliquot of the crude reaction mixture wastitrated against a solution of Ph2S,where a color change occurs from deep green to yellow.The resulting solution is then analyzed by 31P{1H} NMR spectroscopy. Such analyses havereferences on page i2Chapter IV 152shown that the metathesis reactions involving 4.1 and alkyl transfer reagents proceed in almostquantitative (>95%) yields.Oxidation of a crude sample of the hafnium(ffl) chioro derivative, 4.3 with TiC13showed the presence of hafnium(IV) and zirconium(IV) species in approximately 10:1 ratio.This result is consistent with the previous suggestion that during the reduction of thehafnium(IV) precursor the ratio of the zirconium impurity increases.TiC13 or Phd2TolueneMe2SiMe2Si2.6R=Cl 4.1, Me 4.5,Et 4.6, CH2SiMe34.8Slow DimerizationFor R = Me, Et, CH2SiMe3[Zr] = ZrCp[N(SiMe2CHPPr’JScheme 4.54.1Me2SiMe2SIPh2STolueneMe2Si\Me2SiRPh Ph R[Zr]Q[Zr] . [Zr][Zr]R Phreferences on pageChapterlV 153It is noteworthy that the oxidized monoallcyl derivatives, i. e., allcykhiolate and thealkylchloro complexes,Zr(15-C5H5)XR[N(SiMe2CH2PPr2)2], where X SPh and R = Me, Etor CH2SiMe3 or X = Cl and R = Me, Et or Bz, show broad resonances in their NMR spectra,suggesting the molecules are fluxional. The alkylchloro derivatives could be isolated as solidmaterials and were found to be thermally labile. The alkyithiolate derivatives are oils at roomtemperature and undergo further reaction to give a material which has complex 1H NMRspectra. Also, the 31p ( 1H } NMR spectra of this materials were virtually identical for all threederivatives, consisting of two low field and two high field singlets around 19 and -.2 ppmrespectively. The NMR spectral features are possibly a result of an unsymmetrical dimericspecies, containing bridging sulfido ligands (Scheme 4.5).20 However, more investigation isrequired to comfirm the nature of this diamagnetic complex.4.5 Hydride Complexes of Zirconium(llI)4.5.1 GeneralHydride derivatives of zirconocene dihalides; for example Schwartz’s reagent,Cp2ZrHC1, are among the transition metal hydrides that have been found useful instoichiometric and catalytic reactions.21 Although there are a number of reports ofzirconium(llI) hydride derivatives, they are all poorly characterized (only by ESR) and all ofthese hydride complexes have been generated in situ by different reduction methods (Scheme4.5)2228 Also, there are no unambiguously understood reactions of zirconium(III) hydrides,even the migratory insertion of the hydride into an unsaturated organic functionality such asolefm is not well established.’26’8[(15-CH4CMe3)2ZrH(p.-H)] [015-CH4CMe3)2ZrH]-ElectrochemicalReduction g = 1.996, a(1H) = 8.4 G, tripletNaorKa(91Zr)=16.2GC10H8g = 1.992, a(1H) = 8.3 G, tripleta(23Na) = 10.1 G or a(39K) = 2.0 Ga(91Zr) = 10.4 GScheme 4.6 continuedreferences on page i2Chapter IV 154[(115-CsH4Me)2ZrH(t- )]hv (T5-CH4Me)2ZrH + H2.orAg= 1.9854 A(1H)=6.8G/CPh \PPh3(115-CH4Me)2ZrD(15-CH4Me)2ZrH.(PhCCPh)g = 1.9931, A(1H) = 5.6 G, (rL5-CH4Me)2ZrHPPh3A(91Zr) = 26.0 G g = 1.9977, A(31P) = 24.3 G,A(91Zr) = 11.3 GNo evidence for hydrideH2(rj5-CH)2Zr HPPh “(it -C5H)2ZrH”g = 1.985, a(31P) = 19.5 G g = 1.987, a(’H) = 8 GMg PhCCPh(ri5-CH)2ZrC11 “(T15-CH)2ZrH” Insertion Productsor StyreneScheme 4.6The hydride species (r15-CH4Me)2Z H, generated by thermolysis or photolysis of{(n5-CH4Me)2ZrHI(s.t-H),is suggested to form an adduct with diphenyl acetylene; however,there is no evidence for the insertion of the triple bond into the zirconium-hydrogen bond.22’6However, the zirconocene monohydride, Cp2ZrH, generated either by magnesium reduction28ofCp2ZrC12 or by the hydrogenolysis24ofCpZrCH2PPh is suggested to insert carbon-carbontriple and double bonds into the zirconium-hydrogen bond. In an experiment monitored by ESRspectroscopy, Cp2ZrH was generated in situ by magnesium reduction and then, upon addition ofstyrene or diphenyl acetylene, there was a disappearance of the doublet resonance due to thehydride species and the formation of a singlet resonance, which has been attributed to theformation of the insertion product.28 Also, hydrolysis of the reaction mixture shows thepresence of hydrogenated products.28 However, it is important to note that ESR spectra of theadduct formed from the reaction of(r15-C4Me)2ZrH and PPh3 consists only of a doublet,references on page i2Chapter IV 155coupling that has been atthbuted to one phosphorus-31 nucleus; no coupling was observed dueto the hydride ligand.26Hydrogenolysis of Cp2ZrCH2PPh2 in the presence of a substrate shows catalytichydrogenation of the substrate. For example, 1-hexene was hydrogenated to hexane at 80 °Cand 40 bar hydrogen pressure in THF using a catalyst/substrate/solvent ratio of 1:300:1000where 100% conversion was observed in 20 minutes. However, when the reaction mixture wasanalyzed by ESR spectroscopy only two doublets were observed, one of which corresponded toCp2ZrCH2PPh2 and the other to the hydride species Cp2ZrH; no evidence for the zirconium(llI)alkyl intermediate was found.244.5.2 Synthesis ofZr(15-CH)H[N(SiMeCHPPr]Pr2H21 atm(-RH)Pr’2R = CH2SiMe3,4.8R=CH3,4.SScheme 4.7The reaction of the zirconium(Ill) alkyl complexes such as the methyl 4.5 or theCH2SiMe3 4.8 derivatives with one atmosphere of dihydrogen at room temperature proceedssmoothly over a period of 10 to 12 hours to generate the mononuclear zirconium(III) hydridecomplexZr(T5-C)H[N(SiMe2CH2PPr)21, 4.14 (Scheme 4.7). The ESR spectrum of thehydride complex consists of a 1:1:2:2:1:1 doublet of triplets centred at g = 1.988, due to thecoupling to two phosphorus-31 nuclei (a(31P) = 21.7 G) and one hydride (a(1H) = 8.7 0) ligand.The same reaction with molecular deuterium gas produces the corresponding deuterideZr(15-CHS)D[N(SiMe2CHPPr)2], 4.14-d; the triplet observed in the ESR spectrum of this4.14references on page QChapter IV 156material is consistent with the expected deuterium hyperfine interaction of approximately oneseventh of a(1H) (Figure 4.6). For comparison, ESR data reported for other d’ hydridecomplexes are as follows: (g, a(1H)); Cp2NbH2, (2.0097, 11.7 G);29 [Cp2TiHJ, (1.992, 7 G(approximate));30-2TaC12H2(PMe3,(1.960, 6 to7 G (estimated)).3321.7G 21.7 GZrCpH[(NSiMe2PP1)JAZrCpD[(NSiMe2HPP’r)]BFigure 4.6 The X-band room temperature solution (toluene) ESR spectrum of,(A) Zr(rI5-CH)H[N(SiMe2CHPPr)]4.14, and(B) Zr(Ti-HS)D[N(SiMe2CHPPr)2],4.14-dMonitoring the hydrogenolysis reaction of the CH2SiMe3derivative 4.8 by ESR and1H NMR spectroscopy shows a clean conversion to the hydride complex with concomitantformation of tetramethyl silane. Also, the reaction of the methyl derivative 4.5 with moleculardeuterium shows the formation of CH3D. However, we have not been able to isolate thehydride complex as a solid due to its high solubility in hydrocarbon solvents. In the absence ofdihydrogen, complex 4.14 is thermally stable for 48 hours at RT as evidenced by monitoring atoluene solution of 4.14 by ESR spectroscopy. The hydrogenolysis of the benzyl derivative,references on page8.7GChapter IV 1574.12 was very slow at room temperature whereas the phenyl derivative was inert underdihydrogen.An attempted one-pot synthesis of the zirconium(Ill) hydride by reacting the zirconiumchloro derivative 4.1 with LiCH2SiMe3and then, without removing the LiCl, stirring under anatmosphere of dihydrogen resulted in the formation of a diamagnetic zirconium species 4.15.This reaction reaches completion in approximately 5 hours. However, when a pure sample ofthe alkyl derivative was sealed under 4 atmospheres of dihydrogen and monitored by 1H and31P{1H} NMR spectroscopy, there was observed a slow formation (approximately 5 days) of adiamagnetic species, 4.15. Also, the reduction of the zirconium(IV) dichloro derivativeZr(15-C)Cl2[N(SiMeCH2PPr)2] 2.6, under dihydrogen gave the identical diamagneticspecies, 4.15. The micro-analytical data of this compound was consistent with the formulaZr(-CHs)H[N(SiMeCHPPr).The NMR and infrared spectroscopic data strongly suggest that this diamagnetic hydridecomplex is not a mononuclear species. The solid state infrared spectra of 4.15 show a broadband at 1435 cm4 which shifts to 1160 cm4 upon substitution of the hydrides with deuterium,4.15cL1.* Although, this infrared spectral feature could be assigned to a Zr—H bond stretch, bycomparison with other known zirconium hydride complexes it is difficult to assign whether it isdue to a terminal or a bridging hydride ligand. For example, in (Cp’2ZrH).t- the terminaland the bridging hydrides give rise to absorptions at 1565 and 1330 cm’ respectively.34The 1H { 31P} and 31P { 1H } NMR (Figure 4.7) data suggest that the complex is anunsymmetrical dinuclear species and in solution it exists in two isomeric forms in approximatelya 9:1 ratio. The 1H NMR spectrum shows 7 separate resonances for the SiMe2 protons of themajor isomer and two resonances, 6.15 (triplet) and 6.57 (singlet) ppm corresponding to the Cpligand. Occasionally, when the hydride complex was precipitated from a Et20 solution it gaveonly the major isomer, which within a few hours isomerized to give the mixture of isomers.* It will be more appropriate to designate the deuteride as 4.15-d.references on page 122Chapter IV 158Facial-PNPFigure 4.7 A 500 MHz1H{31P} NMR spectrum of 4.15 showing the resonances associatedwith the PNP ligand. B 500 MHz 1H NMR spectrum of 4.15 showing theresonances associated with the Cp ligand. C 500 MHz1H{31P} NMR spectrum ofthe deuteride analogue 4.15-d11 showing the resonances associated with the SiMe2groups. D 76.77 MHz 2H NMR spectrum of 4.15-do showing the resonancesassociated with the PNPligand. E 121.4 MHz 31P{1H} NMR spectrum of 4.15.Meridional-PNP121 MHz 31P{1H} NMR100 076.77 MHz 2H NMRppm 20BBI I10 0500 MHz 1H NMR‘SISiMe2 region of 4.15..dCpLI 14 I.360C0•5[Zr]— -‘ I.);,N—SiMeP[CH(CH3)2]ASi(CH3)22.5 ppm 2:0 1.5 0.5references on page 122Chapter IV 159The 1H{31P} NMR spectrum of the major isomer seems to suggest that the hydrideresonances are probably obscured by the resonances due to the ligand. Also, the two doublets(23H-H = 10.7 Hz) around 1.48 and 2.18 ppm may correspond to a cyclometalated CH2 unit ofthe missing methyl of the SiMe2 unit. The 31P{ ‘HI NMR spectrum consist of an ABMX spinsystem where the AB spins appear at lower field and the MX spins appear at higher fields. Infact the 31P( ‘HI NMR spectral pattern of complex 4.15 was very similar to the product 2.24,obtained from the reaction of the side-on dinitrogen complexes with LiBEt4. It is possible thatthe low field phosphorus resonances, that is the AB spins, correspond to the meridionally boundPNP ligand of one zirconium center and the high field resonances, that is the MX spins,correspond to a facially bound PNP ligand of the second zirconium center. Selective 31Pdecoupled 1H NMR spectra show that the triplet resonance of the Cp ligand at 6.15 ppm is dueto the coupling of the two low field 31p nuclei.To locate the resonances due to the hydride ligands, the deuteride analogue 4.15-d wassynthesized by reducing the zirconium(IV) precursor 2.6 under molecular deuterium. However,the 1H NMR spectrum of this material was more complicated than the spectrum of the hydrideanalogue; in particular, the SiMe2 region showed numerous broad peaks. Also, the 2H NMRspectrum showed overlapping resonances between 0.0 and 0.7 ppm, which clearly indicate thatsome of the protons of the SiMe2moieties are replaced by deuterium. The 2H NMR showed noevidence for the incorporation of deuterium into the isopropyl groups of PNP.4.15 4.15areferences on page iChapter IV 160On the basis of the available NMR spectroscopic data the structures depicted as 4.15 and4.15a have been proposed for the diamagnetic dinuclear hydride complex. Because the majorand the minor isomers show similar spectroscopic features they are likely to have closely relatedstructures; for example 4.15a could be the minor isomer.[(PP’r2CHSiMeNjCpZr —HZr(ffl)‘-CI[(PPLr2CHSiMe)N]CpZZr(IV) CINa/Hg Toluene“ZrCp[N(SiMe2CHPP’r)j”Zr(II)H[(PP’r2CHSiMeN] CpZr( )ZrCpH[N(S1Me2PP’rJH4-i-H2 -H2[(PP1r2CHSiMe)N]HCpZr ZrCp { N[(SiMeCH2)CHPI%][SiMe2CHPPrj}HScheme 4.8With regard to the mechanism of formation of the diamagnetic hydride complex, thereduction of the zirconium(IV) complex 2.6 could possibly generate the zirconium(ll) species,“Zr(15-CH)[N(SiMe2CHPPr)J”which then oxidatively adds dihydrogen and forms thedihydride intermediate, Zr(15-CH)H2[N(SiMeCPPrj(Scheme 4.8). Dimerization ofLiCI1k[(PP1r2CHSiMe)N1CpZrBinucleardisproportionationzH]LiC1Zr(IV) H2references on pageChapter IV 161this intermediate and a a-bond metathesis reaction with one of the carbon-hydrogen bonds ofthe SiMe2 unit would lead to the formation of complex 4.15. In the presence of dihydrogen thezirconium-carbon bond of the metallacycle could undergo reversible hydrogenolysis and a-bondmetathesis; such a process, under molecular deuterium would lead to the exchange of hydrogensof the SiMe2 unit with deuterium.As mentioned before the formation of the diamagnetic hydride 4.15 from thezirconium(III) hydride 4.14 under dihydrogen is influenced by the presence of LiC1. On thebasis of the discussion presented for the ailcyl derivatives, particularly the methyl and ethylderivatives, it is reasonable to presume that the hydride derivative could also interact with LiCl(Scheme 4.8). In the presence of LiC1 the neutral hydride complex 4.14 may establish anequilibrium with an anionic zirconium(Ill) hydride species. This anionic species could interactwith a neutral zirconium(III) hydride species, for example formation of a dimeric species havingbridging hydride ligands. Such a species could undergo disproportionation to give azirconium(II) and a zirconium(IV) dihydride species (Scheme 4.8). It is possible that even apure sample of the zirconium(llI) ailcyl precursor 4.8, may contain small amounts of LiCl whichcould be catalyzing the formation of the zirconium(IV) hydride, 4.15 during hydrogenolysis.4.5.3 Reactions of Zr(15-CH)H[N(SiMe2CHPPr]The paramagnetic hydride 4.14 reacts almost instantaneously with 1 atmosphere ofethylene at room temperature. The ESR spectrum of the product was identical to thezirconium(III) ethyl derivative 4.6, generated by the metathesis reaction of EtLi and 4.1.Reacting the hydride complex with ethylene-d4 gaveZr(T5-C5H5)CD2CD2H[N(SiMeHPPr’)2] 4.6-cLi, and similarly the deuteride complexesZr(r1-C5HS)CH2CHD[N(SiMePPr)2]4.6-d1, and Zr(rI5-CHS) D2CD3[N(SiMe2CHPPr]4.6-d5 were synthesized byreacting 4.14-d with ethylene and ethylene-d4, respectively. The comparison of the hyperfinecoupling information obtained form the three deuterio ethyl derivatives and the non-deuterloethyl derivative correlates well with the transfer of the hydride or the deuteride ligand to the J3-position of the ethyl group. Solution monitoring by ESR spectroscopy of the deuteridereferences on pageChapter IV 162derivatives 4.6-d1 and 4.6-d4 shows that over a period of 72 hours there is no scrambling of the13-deuterium or the 3-hydride into the ce-position, consistent with no 3-hydride elimination forthese zirconium(III) ailcyl derivatives. In this respect, these zirconium(Ill) alkyls mirror relatedzirconocene alkyl derivatives for which f3-elimination is a slow process for unhindered alkylderivatives.35Pr’2 Pr’Me2Si(\\O2Me2SiPr’2[Cl = C6H5,C6D5,C6H11,C5H9Proposed transition statefor hydride exchangePr’2D2 or C6D or C6D12Me2SPr112 or C6H orC5H10H2 A90°C A900CI DCH6 C6DMe2SI\ \N— ZCMe2Si))c%/ ‘)HHPr2Pr’2/PMe2SI\/HMe2S1Pr2Proposed transition states for theexchange of hydride with phenylScheme 4.9references on page i2Chapter IV 163The hydride ligand of complex 4.14 was found to undergo exchange reactions withdeuterium from molecular deuterium gas or from the carbon-deuterium bonds of both aromaticand aliphatic compounds (Scheme 4.9). Also, when the hydride complex 4.14 is heated inbenzene at 90 C the deep green color due to the hydride complex slowly changes to give a redsolution. The ESR and UV-Vis. spectrum of this red solution indicates the formation of thephenyl derivative 4.4.It is presumed that these reaction involve a four centered transition state similar to theone proposed for a-bond metathesis reactions in d0 metal complexes.36 According to thedepictions shown in Scheme 2.8, and considering only the steric factors, the exchange ofhydrogen atoms between a metal-hydride bond and a carbon-hydrogen bond should be favored.The proposed mechanism is consistent with the experimental observation that the conversion ofthe zirconium(III) hydride to the phenyl derivative only takes place at elevated temperatures andwith longer reaction times.The reaction of the zirconium(III) hydride complex with a tenninal acetylene did notshow any evidence for the formation of the insertion product. For example, reacting Ph—CC--Hwith complex 4.14 and oxidizing the product with PbC12 gave a product which did not show anycharacteristic olefinic signals in the 1H NMR spectrum. However, this crude sample containedmainly of one complex (>80 %) and the 1H and 31P[’H} NMR features were consistent with theformula Zr(15-C5H)C1(CCPh)[N(SiMe2CHPPr)2J, 4.16. This result suggests that thezirconium-hydride bond probably undergoes a rapid a-bond metathesis with the terminalcarbon—hydrogen bond of the acetylene to give the zirconium(III) acetylide derivativeZr(rj5-CH)(CECPh)[N(SiMe2CH2PPr’)2j.4.5.4 Synthesis of Zirconium(llI) Borohydride ComplexA variety of tetrahydroborate derivatives of early and late transition metals have beensynthesized.37 These complexes are potentially important as precursors to solid metal borides,38to metal hydrides,39’40to metal alkyls4’ and they may also function as olefin polymerizationcatalysts.42 From a structural point of view, they are intriguing due to the versatility displayedreferences on pageChapter IV 164by the tetrahydroborate group in the mode of its coordination to the metal center, which mayinvolve triple, double and single hydrogen bridges;37 all of these ligation modes are known forgroup 4 transition metals.Figure 4.8 (A) Solution (toluene) ESR spectrum of 4.17 recorded at room temperature.(B) The simulated spectrum of 4.17.Pr’2 Pr’24.1LiBH4Toluene 4.1The reaction of the chioro derivative 4.1 with LiBH4 in toluene slowly changed colorfrom deep green to give a dark brown solution (Equation 4.1). Recrystallization of the crudematerial from an Et20 solution gave a pure crystalline material in greater than 80% yield withan elemental composition that was consistent with the formulaZr(r15-CH)(BH[N(SiMe2CH2PPr)2],4.17. This zirconium(III) borohydride complex exhibits good thermal stability.For example, heating a toluene solution of the complex at 90 °C for 12 hours shows nodecomposition by ESR spectroscopic analysis.20 GAObserved SimulatedBreferences on pageChapter IV 165The solution ESR spectrum of 4.17 consists of a broad binomial triplet due to couplingto two equivalent phosphorus nuclei; no satellites due to the magnetically dilute zirconium-91nuclei were observed (Figure 4.8). The ESR spectrum of 4.17 was simulated using thefollowing parameters: g = 1.958; a (31P) = 20.7 G, 2P; a(14N) = 2.0 0, iN and a(1H) = 2.8 g,4H; this last hyperfine coupling parameter suggests that all four hydrogens of the borohydrideligand have similar interactions with the metal center. The borohydride ligand may beundergoing some fluctional process that exchanges the positions of the terminal and bridginghydrides (Scheme 4.1O).The solid state infrared spectrum of the complex consists of bands at 2401 and 2374 cm1 and strong absorptions at 2126 and 1460 cm-1, consistent with a BH4 ligand bound in abidentate fashion. The bands at 2401 and 2392 cm-1 were assigned to the terminal B—H bondstretches and the absorption at 2124 cm-’ was assigned to the bridging B—H bond stretch. Thesolution (toluene) infrared spectrum of complex 4.17 was identical in the region associated withthe B—H stretches, however the feature at 1460 cm-’ observed in the solid state was absent.H2 H2 H14H1’... /‘______•14H2’... /‘— Zr—.’H1—Bç”-H2 H2 H2H2 1 H1,H1,.. / ,1H2’.. /- ZrHB_H2H2 H2H2 H2 H1H1’....• /.H1’... I.H2.,.. /Zr....,,,,1 B2Zr -‘H2Zr....,,,,H1B\H H1 H2Scheme 4.10A fluctional process that would exchange the terminal and bridging hydrogens couldinvolve a monodentate or a tridentate or an agostic type intermediate (Scheme 4.9). On thebasis of the solution infrared spectra the tridentate intermediate seems unlikely since this wouldreferences on pageChapter IV 166significantly alter the B—Ht stretching region. In the literature it has been stated that the infraredspectra of monodentate and bidentate borohydride derivatives are likely to give similar features,particularly for the B—H stretches.43 It is possible that the fluctional process could involve anagostic type intermediate. Such agostic type* bonding of the BH4 ligand is known for thecomplex (PMe3)2Ti(BH4where two of the BH4 ligands are bound in an agostic fashion,involving the B—H bond.C3I III ORTEP view showing the complete atom labeling scheme of complexZr(15-CH)(l2BH4[NSiMeCPPrJ,4.17. II A Chem 3D® view showingthe arrangement of the PNP and borohydride ligands around the zirconium center.The X-ray structure of the borohydride derivative 4.17, is depicted in Figure 4.9. Thestructure clearly shows the complex to be mononuclear with the borohydride ligand bound in abidentate fashion. All of the hydrogen atoms associated with borohydride ligand were locatedduring the refinement of the X-ray structure. By comparison with the structures of the phenyl* These interactions are also refered to as side-on bound borohydride ligandC15C13C14IC2SiCl Cl 2 SiC8Figure 4.9Hreferences on pageChapter IV 1674.4 and the CH2SiMe34.8 derivatives the borohydride complex 4.17 is somewhat isostructural.Analysis of the bond angles around the zirconium center suggests a pseudo trigonal bipyramidalgeometry with the phosphine donors occupying the axial positions. The centroid of the Cpligand, the nitrogen and the two bridging hydrogens of the BH4 ligand, all lie in the equatorialplane.Table 4.6 Selected bond lengths of complexZr(15-CH)(T2BH4[N(SiMe2CH2PPr)],4.17Atom Atom Distance (A) Atom Atom Distance (A)Zr P(1) 2.7933 (15) Zr P(2) 2.7849 (15)Zr N 2.224 (4) Zr B 2.617 (7)Zr Cp 2.204 (3) Zr HB(1) 2.19(5)N Si(1) 1.714 (4) Zr HB(2) 2.09(4)N Si(2) 1.726 (4) B HB(1) 1.12(5)Table 4.7 Selected bond angles of complexZr(ri5-CH)(rBH[N(SiMeCPP )},4.17Atom Atom Atom Angle (0) Atom Atom Atom Angle ()P(1) Zr P(2) 147.66(5) P(1) Zr N 78.80(10)P(1) Zr Cp 105.55 P(2) Zr N 74.00 (10)P(2) Zr Cp 104.74 P(l) Zr B 84.04Cp Zr N 127.21 P(2) Zr B 90.29Cp Zr B 120.48 N Zr B 112.32HB(l) Zr 1113(2) 47.0 (20) Zr 1113(1) B 99 (3)Cp Zr HB(2) 98.2 Cp Zr 1113(1) 145.1Cp refers to the centroid of the Cp ligand. All structural parameters associated with the centroidof Cp and B for the complex 4.17 were taken from the Chem 3D® structure.references on page 222Chapter IV 168A comparison of the X-ray structures of the phenyl 4.4, CH2SiMe3 4.8, and theborohydride 4.17, show that the centroid of the Cp ligand the nitrogen of PNP and the ligatingatoms of the third group, that is the phenyl or the CH2SiMe3or the BH4, lie in the pseudoequatorial plane. Also, the plane of the phenyl ring and the chelating ring of the BET4 ligand lieon the equatorial plane. A Chem 3D® illustration of the equatorial planes of the three moleculesare shown in Figure 4.10.Figure 4.10 The Chem 3D® view of the equatorial planes of complexes 4.4, 4.8 and 4.17. Allnumerical values given correspond to the nearest bond angle. The Chem Draw®drawing inside the box is the proposed structure for the ailcyl complexes insolution.It seems that in the borohydride complex the site occupied by the hydrogen atom trans tothe Cp ligand is probably the vacant site in the phenyl and the alkyl complexes. The unusualbond angles around the ipso-carbon of the phenyl group may be due to a weak interaction of thereferences on page98.22°127.2°Zr120.5°NH 139.18.05°ZrH135.6°ZrNPr2/PMe2Sç:NMe2S4 H—L’-jPr’2R=H,Me,SiMe3Chapter IV 169ortho-hydrogen of the phenyl group with the metal centre. Also, for the alkyl derivatives thesolution structure could be depicted as shown inside the box in Figure 4.10, where one of the xhydrogen is involved in a weak agostic type interaction in the equatorial plane.The zirconium-Cp bond distance of 2.209 A in the borohydride complex is comparableto the phenyl, 4.4 and the alkyl, 4.8 complexes and is shorter than the distances observed incomplexes 2.7 and 2.9. The zirconium-hydride distances, 2.19 (5) and 2.09 (4) A are identicaland the zirconium-boron distance was 2.617 (7) A, all of which is consistent with the bidentatecoordination of the BH4 ligand. The calculated zirconium-boron distance for the tridentate andbidentate modes are 2.35 and 2.60 A respectively.45 The metal-BH4 interactions in complex4.17 are similar to those inZrH(BH4)PMeand Zr(BHt)437’964.6 Disproportionation Reactions of Mononuclear Zirconium(Ill) ComplexesTreating a solution of the zirconium(ffl) chloro derivative 4.1 with an atmosphere of COresulted in a rapid color change from deep green to dark brown. Monitoring the reaction by 1Hand 31P{1H} NMR spectroscopy showed the formation of diamagnetic species; the ESRspectrum of the same sample showed only a very weak signal.* However, upon removing COfrom the aforementioned reaction, the color changed from dark brown to deep green and theESR and NMR spectra of the resulting product were identical to the paramagnetic chioroderivative, 4.1. This result clearly suggests a reversible reaction of complex 4.1 with CO. Uponmonitoring the reaction by UV-Vis spectroscopy, a band at 450 nm was observed whichdisappears upon the removal of CO from the reaction mixture. The solution infrared spectrumof a CO saturated toluene solution of 4.1 shows a weak and a strong band at 1965 and 1871cm-1, respectively. A similar reaction involving the borohydride complex 4.17 with CO alsogave diamagnetic species, for which some of the spectroscopic features (JR. UV-Vis and NMR)were identical to those observed for the diamagnetic species generated by the reaction of COand 4.1. As mentioned above, the reaction could be reversed by the removal of CO to give theparamagnetic borohydride derivative, 4.17.* The signal could be observed only at very high gain.references on page jChapter IV 170ZrCp(BH4)[N(SjMe2CfJppri1+ oC7D8P[CH(CH3)]aZrCpCJ{N(SiMe2Hp r’)] + CO 78)P[CH(C113)2]11111 11111 II!! liii 1111111 I III I I 11111111!! 1111111 III 111111 1111116 5 4 3 2 IPPM 0Figure 4.11 (top) 1H (300 MHz, C7Dg) spectrum of the diamagnetic species obtained from thereaction of 4.17 and CO. (bottom) 1H NMR (300 MHz, C7D8) spectrum of thediamagnetic species obtained from the reaction of 4.1 and CO.references on page 222Chapter IV 171— ‘ ‘ ‘ IZOO. 2000.cmlJZrCp(BH4)[N(SiMe2-1lCHPPr] + Co Jtoluene Wave length (nm)Figure 4.12 UV-Vis spectroscopic monitoring of the reaction of 4.17 (top right) and 4.1(bottom right) with CO (1 atmosphere). (left) Infrared spectrum (solution intoluene) of the reaction of 4.17 with CO.These results indicate that complexes 4.1 and 4.17 react with CO to give two complexes,with one of the products common for both reactions. On the basis of the spectroscopic features,the common species is formulated as the zirconium(II) dicarbonyl derivative, Zr(r15-CH)-(CO)2[N(SiMeCHPPr],4.18. The 1H NMR spectrum of 4.18 has only a single resonancefor the SiMe2 groups and a triplet for the Cp ligand due to coupling to two equivalentphosphorus-31 nuclei. The singlet resonance at 37.91 ppm in the 31P{1H) NMR spectrum,which was common for both reactions, was assigned to the dicarbonyl complex 4.18. TheAbsorbanCe(ZH)[N(SiMezCHzPPzj + CO) toluene(UB-HtB—Hb(ZrCpC1[N(S1Me2CHPPr>]+ CO) tolueneicmreferences on page i2Chapter IV 172complex could be envisaged as having a pseudooctahedral geometry with a meridionally boundPNP ligand and trans disposed CO ligands.After assigning the NMR resonances associated with the dicarbonyl complex 4.18 theremaining spectroscopic features from the reaction of 4.1 with CO shows it to be thezirconium(IV) dichloro derivativeZr(r15-C5H)Cl2[N(SiMeCH2PPr)],2.6. Therefore, it ispresumed that the second diamagnetic species obtained from the reaction of the borohydridecomplex 4.17 is the zirconium(IV) bis(borohydride) derivative, Zr(15-C5H5)(BH4)2[N(SiMe2-CH2PPr)2j, 4.19.The hafnium(Ill) chioro complex 4.3 also reacts with CO to give the hafnium(IV)dichloro complexHf(fl5-C)Cl2[N(SiMe2CHPPr) ]4.2, and the hafnium(II) dicarbonylderivativeHf(rI-)(CO)2[N(SiMeCH2PPr)j,4.20. Therefore, these disproportionationreactions could be given in a general form as shown in Scheme 4.11.Me2 MeSj2____Me2,M=Zr;X=Cl, 4.1;X=BH,4.17 M=Zr,4.18 M=Zr;X=Cl,2.6;BH,4.19M = Hf; X Cl, 4.3 M Hf, 4.20 M = Hf; X = Cl, 4.2Scheme 4.11The solution infrared spectrum of the bis(borohydride) complex 4.19 shows strongabsorptions due to the terminal boron-hydrogen stretches at 2429 and 2381 cm-1 and a band at2134 cm-1 corresponding to the bridging boron-hydrogen stretch. These boron-hydrogenstretching frequencies are comparable to that observed for the zirconium(Ill) borohydridecomplex, 4.17. Therefore it is presumed that the BH4 ligands in complex 4.17 are also bound ina bidentate fashion.M(ffl) M(ll) M(IV)references on pageChapter IV 173The NMR spectroscopic features of the bis(borohydride) derivative 4.19 suggest that it isisostructural with the dichioro derivative 2.6 with trans disposed BIT4 ligands. Also, the 1H and31P{1H} NMR spectra of 4.19 recorded at room temperature and at -85 °C were virtuallyidentical; the resonances due to the BIT4 ligands were not located in the 1H NMR spectrum.Although the disproportionation of the zirconium(III) borohydride complex 4.17proceeds quantitatively to give the bis(borohydride) complex 4.19, attempts to synthesize it byother routes were unsuccessful. For example, reacting the dichioro complex 2.6 with LiBH4 orattempts to oxidize complex 4.17 with PbCl2 to makeZr(fl5-C5H5)(BH4)Cl[N(SiMe2CH2PPr)21 4.17, only produced intractable materials. Also, the reaction of the hafnium dichiorocomplex Hf(15-C) l[N(SiMeCHPPr 4.2, with LiBH4 gave only themonoborohydride complex f(fl5-C5H5)(BH4)Cl[N(SiMePPr)J 4.21.*24.2The room temperature reaction of the zirconium(III) chloro derivative 4.1 with CH3CNgave a deep red solution; the 1H and 31P{ hH} NMR spectral analysis showed the presence ofzirconium(IV) dichioro complex 2.6 and some unidentified products, and no paramagneticspecies were found by ESR spectroscopy. Monitoring the reaction by 31P{ 1H} NMRspectroscopy at low temperatures showed the formation of a second product, 4.22 whichcorrelated to a resonance at 19.78 ppm. By comparison with the reaction of CO with 4.1, thecomplex 4.22 could possibly be an acetonitrile adduct of a zirconium(ll) species. Spectroscopicmonitoring also showed that the zirconium(II) complex is only stable at lower temperatures and* The solid state infrared spectrum of the complex 4.21 shows that the BH4 ligand is bound in a bidentate fashionreferences on pageZr(llI) Zr(IV) Zr(ll)4.1 2.6 4.22L=CH3CNChapter IV 174rapidly decomposes at room temperature. Although the low temperature experiments show thatthe zirconium(ffl) chloro derivative 4.1 does undergo disproportionation reaction in the presenceof acetonitrile, due to the instability of the zirconium(II) species it was difficult to establishwhether the reaction is a reversible process.III III) IlillIlil liii II II)IIIII1IIII I III IIIIII liii 11111111111)116 4 iPPM 0Figure 4.13 A, B Low temperature 1H (300 MHz) and 31P{1H} (121 MHz) NMR spectra(C7D8)of the diamagnetic species obtained from the reaction of 4.1 and CH3N.On the basis of the 1H NMR spectrum of the zirconium(II) acetonitrile adduct 4.22, themolecule appears to be less symmetric than the zirconium(ll) dicarbonyl species, 4.18. The 1HNMR spectrum of 4.22 has two resonance in the SiMe2 region and integration values suggestthat the molecule has only one acetonitrile ligand. Considering that the complex gives only a* decomposition products4.2221 20 19*18 17 16 15ZrCpCI[N(SiMe2HPPr1)]14+CpCH3NC7D8-10 CCp zc)references on pageChapter IV 175singlet in the 31P(1H} NMR spectrum, the geometry around the zirconium center in 4.22 couldbe described as a pseudo trigonal bipyramidal, with the phosphine donors occupying the apicalpositions (Equation 4.2).Some dinuclear zirconium(III) complexes (Chapter 3) undergo thermally orphotochemically induced disproportionation to give zirconium(IV) and zirconium(II) specieswhere the zirconium(ll) species can be trapped with ligands such as CO* and butadiene (top ofScheme 4.12, also see Chapter 3).48 It was also shown that these dinuclear zirconium(III)complexes are stable in noncoordinating solvents like toluene and, in polar solvents like THF,they undergo decomposition. However, the mononuclear zirconium(Ill) complexes 4.1 or 4.17do not shown any sign of such a disproportionation even at elevated temperatures or in thepresence of ligands such as THF, ethylene or PMe3.The dicarbonyl titanium(II) complexCp2Ti(CO),reacts with the titanium(IV) complexCp2TiX2, where X = Cl or SPh, to give the dinuclear titanium(III) species [Cp2TiCl](.t-X)49A similar reaction has also been reported between Cp2Zr(CO) and Cp2ZrX,where X = Cl orSPh.5° These reactions are somewhat comparable to the reverse reactions shown in Scheme4.11.The proposed mechanism for the reaction for the chloro complex 4.1 with CO is shownin Scheme 4.12. It is believed that the initial step involves the formation of an adduct with CO,intermediate A, where, as shown in Scheme 4.11, the CO could adopt a linear or a bentstructure. Intermediates having bent CO ligands have been proposed for the hydrogenabstraction reactions by 19-electron metal carbonyl complexes, for example the reaction of[Cr(CO)6] with HSnBu3 (bottom of Scheme 4.12).51 It could be argued that the formation of abent CO ligand in 19-electron complexes enables the metal center to attain an 18-electronconfiguration. However, intermediate A is a 17-electron complex and therefore bending the COligand would give the metal center a 16-electron configuration. For electropositive metals likezirconium, a bent CO could also donate electrons from the oxygen. The intermediate A coulddisproportionate via a dimeric intermediate as shown in Scheme 4.12.* j is important to note that the disproportionation reactions in these dinuclear complexes are not initiated by CO.references on pageChapter IV 176It is noteworthy that a solution of the dicarbonyl complex under a dinitrogen atmosphereslowly converts to give the dinitrogen complex 2.9. This suggests that the CO ligands of thedicarbonyl complex are labile, a process that is important for the reverse disproportionationreaction to occur. Removal of CO from the reaction could accelerate the dissociation of COfrom 4.18 and lead to the formation of complex 4.1. Acetonitrile and CO being similar ligands,ZN_____L[Zr]7 + “[Zr]” [Zr]L11[ r]7.[Zr]Zr(ffl) Zr(ffl) Zr(IV) Zr(ll) Zr(ll)x Cl, Br, L = CO, butadieneMe2Sj\AZr(IV)[Zr] Cl22.6Dinitrogen Zr(II)Complex [Zr]CO +2.9+CO-CO1•[Zr] (CO)24.18Disproportionation ci c::::: [Zl [Zr]cz0Dimeric intermediate[Zr] = “ZrCp[N(SiMe2CHPP1)]”HSnBu3[Cr(CO)6f- [Cr(CO)5( HO)F— 00..t##Scheme 4.12references on pageChapter IV 177the disproportionation reactions involving acetonitrile may also proceed via an intermediatesimilar to A. Outer sphere electron transfer type process might also be important; however, it isdifficult to rationalize the exchange of ligands that must occur to generate the product.4.7 Further Reaction with CO: Formation of Zirconium Formyl ComplexThe Fischer-Tropsch process involves the reaction of CO with dihydrogen over aheterogeneous metal catalyst to form a mixture of hydrocarbons and oxygenated products.Considerable efforts have been made to model different intermediates that are involved duringthe Fischer-Tropsch process. It is believed that the CO is reduced in a stepwise manner, whereat the initial stages an intermediate containing a formyl type ligand has been invoked.52’3(1) K[HB(OP’r)3f[P(OPh)3Fe(CO)(CHO)f(Et4N(2) (Et4NBr‘H NMR, 14.82 ppm (pH = 44.4 Hz)‘H NMR, 16.48 ppm (JP-H = 44.4 Hz)‘3C NMR, 251.3 ppm (Jp- = 11 Hz)PPh3oc00/H2Ph3PP(OPh)3Fe(CO4Li[Et3BH1-or NaBH4[CpRe(NO)(PPh3)(CO)](BF4CH2OOs(CO)2(PPh3)—- CpRe(NO)(PPh3)(CHO)Ph3Solid75°C %%%%,HII0Ph3‘H NMR, 7.44 ppm (P-H = 4.0 Hz)Scheme 4.13Most of the known transition metal formyl complexes are synthesized by the addition ofa hydride ligand to a coordinated carbonyl ligand (Scheme4.l3)M58, Only a few cases areknown where the formyl ligand is formed by the reaction of a metal hydride complex with CO,where the reaction is presumed to take place by the insertion of CO into the metal-hydride bond(Scheme 4.14).59.60 However it is important to note that ford0-metal complexes, the formylreferences on pageChapter IV 178ligand is only observed at low temperatures or was found to be involved in some type ofsecondary interaction (Scheme 4.14).60.61Rh(OEP)H + Co Rh(OEP)(CHO)[Rh(OEP)]2 ++ H2 2 CO 2Rh(OEP)(CHO)OEP = Octaethylporphyrin‘H NMR, 2.90 ppm (ic-H = 200 Hz)13C NMR, 194.4 ppmObserved at -80 °C‘H NMR, 15.2 ppm (ic-H = 114 Hz)‘3CNMR,372ppm(C5Me4Et)2TaC1H‘H NMR, 6.08 ppm (ic-H = 168 Hz)‘3c NMR, 168 ppm1H NMR, 1.8 ppm‘3C NMR, 94 ppm(3c- = 44 Hz)Scheme 4.14The bis(borohydride) complex 4.19 formed from the disproportionation reaction reactsslowly with CO to give a complex which has an elemental composition that is consistent withreferences on pageCOOR4.26CO4.24Cl7PMe3Chapter IV 179the formulaZr(15-CH5)(BH4)2(CO)[N(SiMe2CH2PPr)2], 4.23. The solid state infraredspectrum of this complex did not show any characteristic peak that would correspond to aterminal carbonyl stretch and also the boron-hydrogen stretching region clearly suggested thateach BH4 ligand is bound in a different fashion. The terminal B—H bond stretches at 2423 and2378 cm-’ and the only bridging B—H band stretch at 2146 cmt suggests that one of the BH4ligands is bound in a bidentate fashion to zirconium.The 1H NMR spectrum (Figure 4.14) of this complex shows four resonances for theSiMe2 protons and the 31P{1H} NMR spectrum consists of a sharp peak at 39.72 ppm and abroad peak at 31.20 ppm, which are consistent with an unsymmetrical mononuclear complex.The mukiplet underneath the SiMe2protons (around 0.25 ppm) and the broad resonance around1.2 ppm could be attributed to the boron hydrides. The intriguing feature in the 1H NMRspectrum is the doublet at 3.02 ppm which corresponds to one proton. Selective 31P decoupled1H NMR experiments suggest that the resonance at 3.02 ppm is coupled to the 31P resonance at39.72 ppm with a coupling constant of Jp..H = 27.4 Hz. The B {1H} NMR spectrum of thiscomplex has a singlet at -101.3 ppm and a doublet at -127.2 ppm, the latter is probably due tothe coupling with one phosphorus-31 nucleus (JB-P = 47.4 Hz).The carbon-13 analogue of complex 4.23 was synthesized using 13C0 and shows thatthe resonance at 3.02 ppm in the 1H NMR spectrum has an unusually large coupling constant tocarbon, JCH 161.8 Hz; in addition, the 31P{1H} resonance at 39.72 ppm was now split into adoublet with Jp. = 55 Hz. In the 13C { 1H} NMR spectrum the signal due to the enrichedcarbon was found to be a doublet at 60.73 ppm, Jp.c =55 Hz. One possibility, suggested by thespectroscopic data, is that the complex 4.23 has a formyl type ligand; a proposed structure isdepicted in Scheme 4.15. However, the main problem with this proposed structure is that the13C and 1H NMR chemical shifts associated with the formyl ligand in complex 4.23 aresignificantly shifted to higher field than the usual shifts observed for a formyl ligand (Scheme4.13 and 4.14). Fortunately, we were able to grow single crystals of complex 4.23 and subjectthem to X-ray analysis.references on pageChapter IV 180CBCpi rTrrFi I tIII I 11111111111116.5 6.0 5.5 5.0 4.5 4.0 3.5 MP[CH(CH3)2]11111111 Jil till liii IL lU ill 1111111111111111 JIl 11111 Iij 111111111111111111111111111111111111! lijI 1111111 lj 111112.0 1.8 1.6 1.4 1.2 1.0 0.8 0.6 0.4 0.2 PF1.0Figure 4.14 A 1H NMR (300 MHz) spectrum of 4.23. B, C 1H (300 MHz) and gateddecoupled 13C (75.4 MHz) resonances associated with the formyl ligand of 4.23labelled with 13C. D 1H(1B) NMR (500 MHz) spectrum showing the BH3 unitand E B(’H} NMR (96.2 MHz) spectrun of 4.23 (C7D8).96.2 MHz 11B{H} NMR°UUUL1I1111111Il I IIIII J ll I llI64 63 62 61 60 59 PPM[1E1111111111! 111111111111111111 III If II—100 —110 —120 PPM-130500 MHz1H{1B} NMRD1.5 ppmIII 1111111111 lIuII 1)1111 1111111)13.2 3.0 2.8 PPM—0Zr1 /r°’ [[HC.)Cl)references on page 122Chapter IV 181Pr’2 Pr’2 P2Me2Si’’P“ \ p0Me2Si \BH4 Me2SI+ 2C0 :N_zr_(C1I + :N—zr2 :N—Me2SI/ I 1” -b / /1.4P.BH Me2S1 H4B / V Me2S1Pr’2 Pr’24.19 4.18The X-ray structure (Figure 4.15) of the complex 4.23 agrees with most but not all of thefeatures determined from the spectroscopic analysis. However, the original proposal has one ofthe phosphine donors of the PNP ligand coordinated to the zirconium whereas, in the X-raystructure, this phosphine ligand is bound to the carbon atom of the formyl ligand. Such aphosphine interaction would probably explain the relatively high field proton and carbon-13NMR signals observed for the formyl ligand. In fact the coordination around the formyl carbonatom is somewhat comparable to the proposed structure of the tantalum formyl complex 4.24(Scheme 4.14) where a second tantalum atom is interacting in a similar fashion to the phosphinein 4.23. Also, the NMR spectroscopic features associated with the formyl ligand of bothcomplexes are comparable.61 The structure also shows that the the second phosphine donor ofthe PNP ligand is coordinated to a BH3 unit.4.17 CoSlowHH‘Pr2pBH3 H/H 4.231H NMR, 3.02 ppm, 3C-H = 161.8 Hz3P-H = 27.4 HZ13C NMR = 161.8 ppmScheme 4.15references on pageChapter IV 182HHCP3HB1DC4C6Figure 4.14 I ORTEP view showing the complete atom labeling scheme of complexZr(’rI5CH)(12-BH4(q2-CHO) {N(SiMe2CHPPr)[SiMe2C(Pr•BH3]},4.23. II and III are Chem 3D® views of complex 4.23.0HHII‘PifiC19C15doC8C3CP4ClC7BiHB1CIreferences on pageChapter IV 183Table 4.8 Selected bond lengths of complexZr(15-CH)(12-BH4)(r1CHO){N(SiMe2CHPPr)[SiMe2CH2(PrP•B3]}, 4.23Atom Atom Distance (A) Atom Atom Distance (A)Zr N 2.135 (3) Zr Cp 2.2855 (9)Zr 0 1.987 (3) Zr C(1) 2.254 (4)Zr HB1A 2.09(4) Zr HB1B 2.31 (4)P(1) C(8) 1.795 (4) Zr B(1) 2.594 (6)P(1) C(1) 1.765 (4) P(2) B(2) 1.927 (7)C(1) 0 1.429 (5) Si(1) N 1.731 (4)Table 4.9 Selected bond angles of complexZr(5-CH)(i12BH4TCHO){N(SiMe2CHPPr)[SiMe2CH(Pr2P•BH3)j),4.23Atom Atom Atom Angle () Atom Atom Atom Angle (0)0 Zr N 104.63 (13) C(1) Zr N 95.70 (14)0 Zr Cp 105.68 (9) C(1) Zr Cp 141.84(12)N Zr Cp 110.30 (9) 0 Zr C(1) 38.73 (14)Zr 0 C(1) 80.78 (21) 0 C(1) Zr 60.49 (19)P(1) C(1) 0 116.1 (3) P(1) C(1) Zr 129.48 (22)C(1) P(1) C(8) 114.15 (20) P(1) C(8) Si(1) 120.78 (23)N Si(1) C(8) 110.55 (18) Zr N Si(1) 115.11 (17)Si(1) N Si(2) 122.11 (20) Zr N Si(2) 122.76 (18)In Scheme 4.16 the bonding parameters associated with the formyl ligand in complex4.23 and a molybdenum complex containing an analogous phosphine stabilized acyl ligand 4.25are shown.62 The phosphorus-carbon bond distance of 1.831(5) A in 4.25 is very similar to asingle bond. By comparison, in the zirconium complex the distance was 1.765(4) A and issignificantly shorter. In fact it is comparable to a related tantalum formyl complex 4.24,61references on page iChapter IV 1841.750(18) A, and to the dinuclear iron complex,Fe2(CO)6[C(CH )P(PhEL1)],1.748(6) A.63In these complexes the bonding between the phosphorus and the carbon of the “R3P—C” moietyis regarded as an ylid. The carbon-oxygen bond distance, 1.429(5) A of the metallaoxirane ringof the zirconium complex, 4.23 is longer than the molybdenum complex, 1.367(6) A; this isconsistent with greater electron donation from oxygen to zirconium (Chapter 2, Section 2.5.2).The zirconium-carbon distance of 2.254(4) A is within the range of 2.25 1(6) to 2.3 88(12) A.H/R3P.•Zr—O80.78(21)’4.23I“a‘—-—flipMe3P4.2574.96(25)’boP PScheme 4.16H HI /CZr”O Zr’OH H> R3PCIR3F,’\ \Zr 0 Zr 0I > R3PC/Zr”\Scheme 4.17references on pageChapter IV 185The bonding of the formyl ligand can be described in terms of two limiting resonanceforms as shown on the top of Scheme 4.17, where one is an2-formyl and the other is an112 oxycarbene canonical structure.61,M,5 In the case of d0 transition metal complexes, theoxycarbene canonical form has a significant contribution towards the nature of the bonding ofthe formyl ligand. For example, the synthesis of complex 4.26 from the thorium formylcomplex could be envisaged as a coupling of two thorium oxycarbene units (Scheme 4.14). Asimilar rational could be extended to describe the derivation of the “formyl-ylid” ligand incomplex 4.23 (Scheme 4.17).4.8 ConclusionsThe reduction of the zirconium(IV) or the hafnium(IV) complex M(15-CH) l2[N(SiMe2CH2PPr)],where M = Zr or Hf, has enabled the synthesis of the zirconium(Ill) andthe hafnium(III) complexes. The zirconium(III) complex, Zr(115-CSH5)Cl[N(SiMe2CH2PPr2)2], 4.1 was obtained in excellent yields, and was used as a precursor to synthesizezirconium(III) alkyl, aryl and borohydride complexes via simple metathesis reactions. Mostintriguingly, all of the zirconium(III) hydrocarbyl complexes, Zr( 5CH)R[N(SiMe2-CH2PPr)],where R = Ph, Me, Et, Bz and CH2SiMe3,show very good thermal stability. TheX-ray structure determination of the aryl (Ph), ailcyl (CH2SiMe3)and the borohydride (BH4)derivatives showed that all of the complexes are mononuclear and have closely relatedstructures. These examples constitute the first well characterized zirconium(III) complexescontaining hydrocarbyl and borohydride ligands.The ESR spectroscopic studies indicate that the x-hydrogens of the alkyl ligands showdifferent hyperfine coupling constants. Although this could be explained on the basis ofrestricted rotation about the zirconium-carbon bond, infrared studies and some structuralfeatures seem to suggest a weak agostic type interaction may be present between one of theoc-hydrogens and the metal.Hydrogenolysis of certain alkyl complexes shows a clean conversion to the mononuclearzirconium(Ill) hydride complex. This hydride complex undergoes an insertion reaction withreferences on pageChapter IV 186ethylene. By means of labeling studies with the ethyl zirconium(Tll) complex it was shown thatno scrambling of the 13-hydrogens occurred which mirrored the behavior of unhinderedzirconium(IV) alkyls. The hydride complex also undergoes hydrogen exchange reactions withH—C(sp2)and H—C(sp3)bonds, presumably by way of a-bond metathesis. Reactions withterminal acetylene predominantly gave the acetylide complex and no definitive evidence wasfound for the formation of the insertion product.ComplexesZr(15-CH)Cl[N(SiMe2CH2PPr)2], 4.1 and Zr(r15-CH)(BH2 )[N(SiMe2-CH2PPr1)2j, 4.17 undergo a reversible disproportionation reactions under a CO atmosphere togive zirconium(IV) and zirconium(ll) derivatives. It was also shown that CH3N reacts in asimilar fashion with 4.1; however the reversibility of this reaction was not established. In thecase of the borohydride complex the reaction with CO proceeds further to give a complexcontaining a “formyl-ylid” type ligand.Finally it could be stated that the isopropyl version of the PNP ligand“[N(SiMe2CH2PPr’)2j” was very effective in stabilizing the zirconium(ffl) complexes. Also,these complexes have shown many different types of reactivity, which is in stark contrast to thebulky Cp ligand based zirconium(III) complexes where no reactivity has been reported.4.9 Experimental Procedures4.9a GeneralUnless otherwise stated all procedures were as described in section 2.9a. Hydrogen gas,purchased from Matheson, was purified by passage through a column of activated 4A molecularsieves and MnO supported on vermiculite. Deuterium (99.5 atom %D) gas and ethylene-d4(99.5 atom %D) were obtained from MSD Isotopes. Ethylene and CO were purchased fromMatheson and were used as supplied. Phenyllithium,66ethyllithium,67benzylpotassium,68LiCH2SIMe3,69NaCp•DME7°and dibenzyl-magnesium7’were prepared according to literatureprocedures. Methylmagnesiumbromide (1.4 M solution in 75% toluene and 25% THF),benzylmagnesiumchloride (1 M solution in Et20), CH3N, TiCl3,Ph2S and PbCl2 werepurchased from Aldrich and LiBH4 was purchased from BDH. Diphenyldisuiphide wasreferences on page jChapter IV 187purified by sublimation, acetonitrile was dried over 4A activated molecular sieves for 12 hoursand then vacuum transferred and degassed by three freeze-pump-thaw cycles. Lead(II)dichioride and LiBH4were pumped under vacuum for 24 hours prior to use.The ESR spectra were recorded on Varian E-3 spectrometer calibrated with a sample ofVO(acac)2, g = 1.965.72 ESR simulations were done on a Macintosh lix using ESR II fromCalleo Scientific Software. The values of line broadening and coupling constants were obtainedfrom the simulated spectra (see appendix Bi). NMR, UV/Vis, JR and MS spectra were recordedon the instruments described in section 2.9a.The complexZr(fl5-C5H5)C12[N(SiMe2CH2PPr’)],2.6 was prepared according to theprocedure described in section 2.9b.2. HfCl3[N(SiMe2CH2PPr)jwas synthesized accordingto the published procedure.734.9b Synthesis of Complexes4.9b.1Zr(15-H)C1[N(SiMe2HPPr’],4.1. A solution ofZr(15-CH)C12[N SiMeCH2PPr)]2.6, (1.50 g, 2.42 mmol) in toluene (80 mL) was transferred into a thick-walledreaction flask containing Na/Hg (100 g of 0.3% amalgam, 13.0 mmol). The flask was thenevacuated under vacuum (3 minutes) and sealed. Upon stirring the reaction mixture turns deepgreen. After 48 hours the reaction mixture was decanted from the amalgam and filtered througha layer of Celite®. The amalgam was extracted with 20 mL portions of hexanes (total of 60 mL)until the washings showed no green color. Upon removal of the solvent from the combinedfiltrate and extracts a dark green solid was obtained (95%). ESR spectrum of this green solidshows only the presence of 4.1 and 1H NMR shows <5% (estimated relative to the amount ofC6D5H present in C6D6) diamagnetic impurities. Recrystallization from hexanes gaveanalytically pure material (1.25 g, 88%). ESR (toluene): g = 1.955; a(91Zr) = 37.2 G, lZr;a(31P) = 21.1 G, 2P; a(14N) = 2.9 G, iN; a(1H) = 1.8 G, 5H; linewidth used for simulation, 2.3G. .teff (Evans method) = 1.57 B. M. UV-Vis (toluene, 1 cm quartz cell): 2max = 324 nm, 6max= 2600 Lmo11c; 360 nm, e = 2500 Lmol1c . Anal. Calcd forC23H491NPSiZr:C, 47.27; H, 8.45; N, 2.40. Found: C, 46.99; H, 8.99; N, 2.20.references on page i2Chapter IV 1884.9b.2Hf(i5-C) 1[N(S1MeCHPPr’],4.2. To a solution of HfC13[N(SiMe2CH2-PPr’)2] (6.00 g, 8.89 mmol) in toluene (150 mL) was added solid NaCp•DME (1.736 g, 9.74mmol) at room temperature. The NaCp•DME was added in three portions at 1 hour intervalsand the resulting mixture was stirred for 12 hour. The salt (NaC1) was removed by filteringthrough Celite®, the filtrate concentrated to 15 mL, hexanes was added until the solution turnedturbid and cooling at -30 C gave a pale white crystalline material (5.52 g, 88%). 1H NMR (6,300 MHz, C6D6): 0.60 (s, 12H, Si(CH3)2); 1.6 to 0.96 (m, 28H, P[CH(CH3)2]SiCH2P); 2.17(sept, 4H, P[CH(CH3)2]3JwH = 8.9 Hz); 6.46 (br s —5H, C5H). 31P{1H) NMR (6, 121.421MHz, C6D): 20.04 (s). Anal. Calcd forC2349NC1PSiHf:C, 39.07; H, 6.98; N, 1.98.Found: C, 39.36; H, 7.16; N,[N(SiMeHPPr] 4.3. A solution ofHf(ri5-C) 12[N(SiMeCH2PPr)J(2.00 g, 2.83 mmol) in toluene (80 mL) was transferred into a thick-walled reactionflask containing Na/Hg (100 g of 0.33% amalgam, 14.1 mmol). The flask was then evacuatedunder vacuum (3 minutes) and sealed. Upon stirring the reaction mixture turns deep greenishbrown. After 48 hours, the reaction mixture was decanted from the amalgam and filteredthrough a layer of Celite®. The amalgam was extracted with 20 mL portions of hexanes (totalof 60 mL), until the extracts show no color. Upon stripping off solvent from the combinedfiltrate and extracts a dark brown solid was obtained. Pure material was obtained byrecrystallization from a pentane solution (0.8 g, 47%). ESR spectrum of this brown solid showsthe presence of 4.1 and 4.3. ESR (toluene): g = 1.9 16. Anal. Calcd forC23H49NC1PSi2Hf: C,41.13; H, 7.35; N, 2.09. Found: C, 41.49; H, 7.49; N,[N(SIMeCHPPr’] 4.4. To a solution of 4.1 (240 mg, 0.41mmol) in toluene (5 mL) was added PhLi (1.6 mL of 0.25 M solution in Et20, 0.40 mmol) at-78 °C and stirred for 5 minutes. After stirring the reaction at room temperature for 3 hours thesolvent was removed under vacuum. The deep brown residue was extracted with pentane (20mL) and the extract was filtered through Celite®. The filtrate was concentrated (2 mL) and thencooled at -40 °C to give hexagonally shaped crystals (190 mg, 74%). ESR (toluene): g = 1.981;a(91Zr) = 20.7 G, lZr a(31P) = 20.3 G, 2P; a(14N) = 1.8 G, lN; a(1H) = 1.1 G, 5H; linewidthreferences on pageChapter IV 189used for simulation, 2.8 G. UV-Vis (toluene, 1 cm quartz cell): = 342 nm; = 478 nm.Anal. Calcd forC29H54NSi2P2Zr: C, 55.63; H, 8.69; N, 2.24. Found: C, 55.51; H, 8.68; N;[N(SiMeHPPr’)],4.5. To a solution of 4.1 (825 mg, 1.46 mmol)in toluene (25 mL) was added a solution of MeMgX (X = Br, 1.1 mL of 1.4 M solution, 1.54mmol; X = Cl, 0.55 mL of 3.OM solution in THF, 1.6 mmol) at room temperature. The reactionmixture was stirred for 4 hours and then the solvent was stripped off under vacuum. Theresulting solid was extracted with pentane (10 mL) and the extract was filtered through a layerof Celite®. The filtrate was concentrated (-- 3 mL) and then cooled at -40 °C to give a deepgreen crystalline product (480 mg, 58%). ESR (toluene): g = 1.963; a(91Zr) = 28.0 G, lZr;a(31P) = 21.1 G, 2P; a(1H) = 6.6 G, 3H; a(14N) = 2.1 G, iN; linewidth used for simulation, 3.0G. Anal. Calcd forC245NPSiZr(MgC1)o.2:C, 49.28; H, 8.96; N, 2.39; Cl 2.67. Found:C, 48.90; H, 9.08; N, 2.39; Cl,[N(S1MePPr’1,4.6. Method 1: To a solution of 4.1 (1.00g, 1.71 mmol) in toluene (25 mL) was added a solution of CH3CH2Li(70 mg, 1.94 mmol) intoluene (10 mL) at -78 ‘C. The reaction mixture was allowed to warm to room temperature andstirred for 8 h. The solvent was stripped off under vacuum, the residue was extracted withpentane (10 mL) and the extract was filtered through a layer of Celite®. The filtrate wasconcentrated (3 mL) and then cooled to -40 ‘C to give a deep green crystalline product (560 mg,56%). ESR (toluene): g 1.962; a(31P) = 20.9 G, 2P; a(’Ha) = 9.3 G, 1H; a(’Ha) = 3.2 G, 1H;a(’Hp) = 1.8 G, 3H; a(14N) = 2.6 G, iN; linewidth used for simulation, 2.2 G. Anal. Calcd forC254NP2SiZr: C, 51.95; H, 9.42; N, 2.42. Found: (a) C, 49.76; H, 9.02; N, 2.32; (b) C,51.20; H 9.28; N 2.39. Case (a) was obtained from the first recrystallization and case (b) wasobtained from the second recrystallization and the data seems to fit for 0.6 equivalents and 0.2equivalents of LiCl respectively.Method 2: To a solution 4.1 (300 mg, 0.51 mmol) was added a solution of MeMgBr (0.4 mL of1.4 M solution) and this was stirred at room temperature for 4 hours. The solvent was thenstripped off under vacuum and the resulting solid was extracted with pentane. Stripping off thereferences on pageChapter IV 190pentane gave a green solid containing 4.5. The solid was dissolved in toluene (15 mL) andtransferred into a thick walled reaction flask (150 mL). The solution was freeze-pump-thawedtwice and stirred under 1 atmosphere of H2 for 12 hours at room temperature to give 4.14. TheH2 was removed under vacuum and ethylene was introduced into the reaction flask and stirredfor 0.5 hours. After removing the ethylene under vacuum, dlioxane (1.0 mL) was added andstirred for lh. The solvent was stripped under vacuum and the resulting solid was extractedwith pentane. ESR spectrum of the pentane shows only the presence of[N SiMePPr)],4.6-d1 A solution of 4.14-d (— 0.5mmol) was prepared as described in the synthesis of 4.14 using D2 gas. After removing thedeuterium gas and stirring the solution under ethylene for 0.5 hours gave 4.6-d1 ESR (toluene):g = 1.962; a(31P) = 20.9 G, 2P; a(’Ha) = 9.3 G, 1H; a(’Ha) = 3.2 G, 1H; a(1H) = 1.8 G, 2H;a(2H) = 0.3 G, 1D; a(14N) = 2.6 G, iN; linewidth used for simulation, 1.7 G. JR (cm-1,toluene): 2169 (w, f3C-D of Et ligand); 2143 (m, 13C-D of Et ligand); 2109 (s, j3C-D of Etligand).4.9b.8Zr(5-CsHs)CDDH[N(SiMeCHPPr],4.6-d4. A solution of 4.14 (-0.5mmol)was prepared as described in the synthesis of 4.14. After removing the hydrogen gas andstirring the solution under ethylene-d4 for 0.5 hours gave 4.6-d4. ESR (toluene): g = 1.962;a(31P) = 20.9 G, 2P; a(2Ha) = 1.5 G, 1D; agH) = 0.5 G, 1D; a(2H) = 0.3 G, 2D; a(1H) = 1.8G, 1H; a(14N) = 2.6 G, iN; linewidth used for simulation, 2.2 G.4.9b.9Zr(i5-C)CDC[N(SIMeHPPr],4.6-d5 A solution of 4.14- d1 (‘—0.5mmol) was prepared as described in the synthesis of 4.14 using D2 gas. After removing thedeuterium gas and stirring the solution under ethylene-cL for 0.5 hours gave 4.6-d5. ESR(toluene): g = 1.962; a(91Zr) = 28.4 G lZr; a(31P) = 20.9 G, 2P; a(2Ha) = 1.5 G, 1D; a(2Ha) =0.5 G, 1D; a(2H) = 0.3 G, 3D; a(14N) = 2.6 G, iN; linewidth used for simulation, 2.2 G. JR(cm1, toluene): 2192 (m, f3C-D and possibly an ctC-D of Et ligand); 2164 (m, C-D of Etligand); 2128 (m, C-D of Et ligand); 2043 (s, ctC-D of Et ligand).4.9b.1OZr(’-) HPh[N(SiMeCPr],4.7. To a solution of 4.1 (315 mg, 0.57mmol) in THF (20 mL) was added a solution of PhCH2K(70 mg, 0.54 mmol) in THF (5 mL) atreferences on page i2.Chapter IV 191-40 DC. The reaction was warmed to room temperature and stirred for 5 hours. The solvent wasstripped off under vacuum, and the resulting solid was extracted with hexanes (10 mL). Theextract was filtered through a layer of Celite®, concentrated (— 3 mL) and cooled at -40 DC togive a dark green crystalline product (275 mg, 75%). ESR (toluene): g = 1.956; a(31P) = 18.6G, 2P; a(1Ha) = 3.2 G, 2H; a(14N) = 3.4 G, iN; a(1H) = 1.2 G, 5H linewidth used forsimulation, 1.6 G. Anal. Calcd forCOH56NP2SiZr:C, 56.29; H, 8.82; N, 2.19. Found: C,56.58; H, 8.86; N, ZrC1(CHPh)[N(SiMe2CHPr )],4.9. To a solution of 4.1 (350 mg, 0.6 mmol) intoluene (30 mL) was added a solution of PhCH2MgC1 (0.6 mL of 1 M solution in Et20, 0.6mmol) at room temperature and stirred for 6 hours. The solvent was stripped off under vacuum,the resulting solid was extracted with toluene (10 mL) and filtered through a layer of Celite®.The filtrate was concentrated (1 mL), and added hexanes (10 mL) and cooling at -40 °C gave apale yellow material (63 mg, 15%). Upon further recrystallization the zirconium-(Ill) benzylderivative 4.7 was isolated. 1H NMR (, 300 MHz, C6D): 0.18 (s, 6H, Si(CH3)2;0.45 (s, 6H,Si(CH3)2; 0.6 (m, 4H, SiCH2P); 0.8 to 1.0 (m, 24H, P[CH(CH3)2];1.60 (sept, 2H,P[CH(CH)2]2,3H-H = 6.3 Hz); 1.70 (sept, 2H, P[CH(CH3)3JH-H = 7.5 Hz); 1.98 (s, 2H,CH2Ph); 2.02 (s, 2H, CH2Ph); 6.87 (t, 2H, p-Ph,3JH..H = 6.6 Hz); 7.23 (m, 4H, rn-Ph); 7.37 (d,4H, o-Ph, 3JH..H = 6.6 Hz). 31P(1H} NMR (8, 121.421 MHz, C6D): -11.87 (s). Anal. CalcdforC32581NSi2Zr:C, 54.78; H, 8.33; N, 2.00. Found: C, 55.08; H, 8.56; N; r(i-)CSiMe[N(SiMePPr’] 4.8. To a solution of 4.1 (250 mg,0.43 mmol) in toluene (10 mL) was added a solution ofMe3SiCH2Li(42 mg, 0.45 mmol) intoluene (3 mL) at room temperature. After stirring for 12 hours the solvent was stripped offunder vacuum, and the resulting solid was extracted with pentane (10 mL). The filtrate wasfiltered through a layer of Celite® and concentrated (‘- 3 mL) and then cooling this solution at-40 °C gave a dark green crystalline product (165 mg, 60%). ESR (toluene): g = 1.973; a(91Zr)= 30.5 G, lZr; a(31P) = 21.4 G, 2P; a(’Ha) = 9.3 G, 1H; a(’Ha) = 6.2 G, 1H; a(14N) 2.0 G,iN; linewidth used for simulation, 2.6 G. ji (Evans method) = 1.71 B. M. UV-Vis (toluene,1 cm quartz cell): 2max = 334 nm, Emax = 2700 Lmol1c ; ? = 430 nm, C = 1700 Lmolcm.references on pageChapter IV 192Anal. Calcd forC27H60NPSi3Zr:C, 50.97; H, 9.51; N, 2.20. Found: C, 50.68; H, 9.46; N,’fl5-CSHS)OPh[N(SiMeHPP ’2)2], 4.10. To a solution of 4.1 (200 mg, 0.34mmol) in THF (10 mL) was added a solution of PhONa (40 mg, 0.34 mmol) in THF (5 mL) at-10 C and stirred for 10 minutes. The reaction was then warmed to room temperature andstirred for 12 hours to give a deep green solution. The solvent was stripped off under vacuum,and the resulting solid was extracted with hexanes (10 mL). The extract was filtered through alayer of Celite®, concentrated (- 3 mL) and cooled at -40 °C to give a dark green crystallineproduct (130 mg, 60%). ESR (toluene): g = 1.955; a(31P) = 18.7 G, 2P. Anal. Calcd forC29H54NOPSiZr:C, 54.25; H, 8.48; N, 2.18. Found: C, 54.04; H, 8.50; N,[ (SiMeCP r] 4.11. To a solution of 4.1 (200 mg, 0.34mmol) in THF (10 mL) was added a solution of NaNPh2 (66 mg, 0.34 mmol) in THF (5 mL) at-10 C and stirred for 10 minutes. The reaction was then warmed to room temperature andstirred for 12 hours to give a deep brown solution. The solvent was stripped off under vacuum,and the resulting solid was extracted with hexanes (10 mL). The extract was filtered through alayer of Celite® and stripping off the solvent gave an oil. ESR (toluene): g = 1.953; a(31P) =11.2 G, 2P.4.9.15Zr(i-CH)PPh[N(SiMeCHr],4.12. To a solution of 4.1 (200 mg, 0.34mmol) in THF (10 mL) was added a solution of LiPPh2 (66 mg, 0.34 mmol) in THF (5 mL) at-10 °C and stirred for 10 minutes. The reaction was then warmed to room temperature andstirred for 12 hours to give a deep brown solution. The solvent was stripped off under vacuum,and the resulting solid was extracted with hexanes (10 mL). The extract was filtered through alayer of Celite® and stripping off the solvent gave an oil. ESR (toluene): g = 1.965; a(31P) =18.6 G, 2P; a(31P) 29.8 G.4.9b.16Zr(i5-CH)H[N(S1MeCHPPr’],4.14. A solution of 4.5 or 4.8 (- 0.1 mmol) intoluene (5 mL) was degassed by freeze-pump-thaw cycles and then stirred under 1 atm of H2 for10 to 12 hours at room temperature. Spectroscopic (ESR, NMR) analysis of the resultingsolution shows clean formation of the hydride complex. A solution of 3 (0.05 mmol, 0.6 mL ofreferences on pageChapter IV 193C7D8) in a sealable NMR tube was treated with H2 as described above, and its 1H NMRspectrum shows only the formation of Me4Si and the ESR spectrum of the same sample showsonly 7. Treatment of 4.5 in a similar fashion with D2 shows the formation of CH3D (0.17 ppm,1:1:1 triplet, 23H-D = 1.9 Hz) and the formation of the deuteride 4.14. ESR (toluene): g = 1.988;a(91Zr) = 25.1 G, lZr; a(31P) 21.7 G, 2P; a(1H) = 8.7 G, 1H or a(2H) = 1.4 G, 1D; a(14N) =1.4 G, iN; linewidth used for simulation, 1.6 G.4.9b.17 [ZrHCp[N(SiMePPr’)]{CiZr[N( eSkH(Si]}-(jt-H)2,4.15.Method 1: A solution of Zr(15-CH)Cl2[N(SiMe2CH2PPr)2] 2.6, (300 mg, 0.48 mrnol) intoluene (40 mL) was transferred into a thick-walled reaction flask (300 mL) containing Na/Hg(56 mg of 0.1% amalgam, 2.4 mmol of Na). The dinitrogen in the flask was removed undervacuum (approximately 5 mL of solvent was removed) and then the reaction mixture wasdegassed once by a freeze-pump-thaw cycle. The flask was then cooled to —196 °C, filled with1 atmosphere of dihydrogen, sealed, and allowed to warm slowly to room temperature withstirring. Upon warming up to room temperature the reaction mixture quickly turned green (1hour) and slowly the intensity of the green color decreased. After the disappearance of thegreen colour (7 days) the solution was decanted from the amalgam and filtered through a layerof Celite®. The amalgam was extracted with 15 mL portions of toluene (total of 60 mL), untilthe extracts showed no yellow colour. The filtrate and the extracts were combined and strippingoff the solvent gave a yellow powder (0.20 g, 76%). The product was crystallized from a 1:1mixture of toluene and Et20.Method 2: Approximately 30 mg of the zirconium(III) alkyl derivativeZr(r5-CH)R[N(SiMe2CHPPr)],where R = Me, CH2SiMe3 orC6H5H2was dissolved in0.5 mL of C6D6 inside a NMR tube and the resulting solution was degassed by three freezepump-thaw cycles. Then the NMR tube was cooled to -196 C, filled with 1 atmosphere ofdihydrogen and sealed. Monitoring the reaction by NMR spectroscopy shows only theformation of the diamagnetic hydride complex.references on pageChapter IV 194Major isomer: 1H {31P} NMR (6, 500 MHz, C7D8): 0.17, 0.25, 0.29, 0.39, 0.46, 0.50 and 0.68(s, each 3H, Si(CH3)2); 0.84 to 0.92 (m, SiCH2P); 1.0 to 1.30 (m, (P[CH(CH3)J;1.48 (d, 1H,SiCH2Zr, 2JH-H = 11.2 Hz); 1.6 to 1.9 (overlapping sept, 7H, P[CH(CH213JH-H = 7.0 Hz);2.16 (d, 1H, SiCH2Zr,23H-H = 11.2 Hz); 2.45 (sept, 1H, P[CH(CH3)2]3H-H = 7.0 Hz); 6.15(t, —5H, C5H5,3Jp..11 = 1.4 Hz); 6.56 (s, —5H, C5H). 31P(1H} NMR (6, 121.4 MHz, C7D8):Spectrum consisted of an ABMX spin system; 22.31 (d, 1P, 2Jp..p = 64.7 Hz); 17.45 (d, 1P,2Jpp64.7 Hz); 0.48 (s, 1P); -1.90 (s, 1P). 13C{H} N?vIR (6, 50.324 MHz, C7D8): 111.67 (s, Cp);104.94 (s, Cp). Minor isomer: 1H{31P} NMR (6, 500 MHz, C7D8): 6.12 (t, —5H, C5H3Jp..H1.3 Hz); 6.39 (s, —5H, C511). 31P{1H} NMR (6, 121.4 MHz, C7D8): Spectrum consisted ofan ABMX spin system; 22.00 (d, 1P, 2Jp.p = 65.1 Hz); 18.38 (d, 1P, 2Jp.p = 65.1 Hz); 0.90 (s,1P); -1.34 (s, 1P). JR (cm1, KBr): 1435 (br s, Zr—Ht), 1159 (Zr—Dt). Anal. Calcd forC23H51NP2SiZr:C, 50.13; H, 9.33; N, 2.54. Found: C, 50.35; H, 9.15; N;[N(SiMePr] 4.17. To a solution of 4.1 (800 mg, 1.37mmol) in toluene (40 mL) was added solid LiBH4 (150 mg, 6.84 mmol) and was stirred at roomtemperature. After stirring for 12 hours the reaction mixture was filtered through a layer ofCelite® and the filtrate was concentrated to give a black oil. The oil was dissolved in Et20 (5mL) and cooled at -40 C to give black rectangular crystals (650 mg, 84%). ESR (toluene): g =1.958; a(31P) = 20.7 G, 2P; a(14N) = 2.0 G, iN; a(1H) = 2.8 G, 4H; linewidth used forsimulation, 2.8 G. UV-Vis (toluene, 1 cm quartz cell): 2’max = 332 nm; Em 2800 Lmol1c1 JR (cm-1,KBr): 2953, 2871, 2401 (s, B-1I), 2379 (s, B-Ht), 2126 (s, B-Hb), 1457 (m, Zr-H),1373, 1244, 1124. Anal. Calcd forC23HS3NESiP2Zr: C, 49.00; H, 9.47; N, 2.48. Found: C,48.82; H, 9.46; N;[N(SiMeHP r].A solution of 4.14 (—0.5 mmol) wasprepared as described in the synthesis of 4.14. The hydrogen gas was removed under vacuum,the solution cooled to -78 C and then a solution (toluene) of PhCCH (1 equivalent) was added.Instantly the color changed from deep green to purple. The simulated spectrum wasapproximatly close to the observed spectrum: ESR (hexanes): g = 9.8; a(31P) = 15 G, 2P; a(14N)=3.5G, JN;a(1H)=1G,5H.references on pageChapter IV 1954.9b.20Zr(5-CH)(C—=CPh)C1[N SiMeHP r],4.16. Complex Zr( 5.C5H)-(CCPh)[N(SiMeCH2PPr’)2] was oxidized with PbC12. 1H NMR (6 300 MHz, C7D8): 0.55(br s, 6H, Si(CH3)2); 0.60 (br s, 6H, Si(CH3)2); 0.86 (br s, 4H, SiCH2P); 0.98 to 1.1 (m, 24H,P[CH(CH3)2];2.13 (brm, 4H, P[CH(CH2]);6.53 (s, Cp, 5H); 6.95 (br, 1H,p-Ph); 7.08 (m,2H, rn-Ph); 7.35 (br, 2H, o-Ph). 31P{1H} NMR (6, 121.421 MHz, C6D): 16.32 (s).4.9b.21Hf(iC)C1(BH[N(S1MePPr’],4.21. To a solution of 4.2 (250 mg, 0.35mmol) in toluene (25 mL) was added solid LiBH4 (60 mg, 2.74 mmol) and was stirred at roomtemperature. After stirring for 3 days the reaction mixture was filtered through a layer ofCelite® and the filtrate was concentrated to give an oil. The oil was dissolved in hexanes andcooled at -40 C to give a white crystalline material (170 mg, 7 1%). 1H NMR (6, 300 MHz,C6D6): 0.56 (s, 6H, Si(CH3)2;0.59 (s, 6H, Si(CH3)2); 0.8 to 1.1 (m, 28H, P[CH(CH)2]SiCH2P); 1.96 (sept, 2H, P[CH(CHJ3H-H = 6.8 Hz); 2.24 (sept, 2H, P[CH(CH3)]3JHH= 6.8 Hz); 6.25 (br s —5H, C5H). 31P{1H} NMR (6, 121.421 MHz, C6D): 14.07 (s). 13C{1H}NMR (6, 50.323 MHz, C6D): 110.05 (s, C5H). JR (cm1,KBr): 2488 (s, B-Hi), 2414 (s, BHt), 2143 (s, B-Hb), 1456 (m, Zr-H). Anal. Calcd forC23H57NBC1SiP2Hf: C, 40.24; H, 7.78;N, 2.04. Found: C, 40.02; H, 7.79; N; Disproportionation ReactionsA solution of ZrCp(BH4)[N(SiMe2CH2PPr)2] 4.17, (30 mg, 0.05 mmol) in C7D8 (0.5 mL) wastransferred into an NMR tube and degassed three times by freeze-pump-thaw cycles. The NMRtube was then filled with 1 atmosphere of CO and sealed. The NMR spectroscopic analysis(within 15 minutes) shows the formation of two diamagnetic complexes,ZrCp(BH4)2[N(SiMe2CHPPr) J4.19, and ZrCp(CO)2[N(SiMe2CH2PPr)]4.18 in 1:1 ratio.Upon standing (3 days) the peak associated with 4.19 decreases in intensity and two new peaksappear which correspond to the fomiyl complex, 4.23.The reactions with CO and the zirconium(III) chloro derivative 4.1, or with the hafnium(ffl)chioro derivative 4.3, were carried out as described for the borohydride complex 4.17.The reaction of the zirconium-(Ill) chioro derivative, 4.1 with CH3N was carried out atlow temperature. A solution of 4.1 (30 mg in 0.5 mL of C7D8)was transferred into a NMRreferences on pageChapter IV 196tube, degassed three times by freeze-pump-thaw cycles, and using a constant volume gasapparatus 2 equivalents of CH3N vapor was condensed at -196 °C, and sealed. The samplewas transferred into a -60 C bath and allowing to stand for 1 hour gave a deep red solution.The NMR spectroscopic analysis shows the presence of two complexes,ZrCpC12[N(SiMeCHPPri)]2.6, and ZrCp(CH3CN)[N(SiMe2CH2PPr’)2], 4.22. Warmingthe sample to room temperature shows a rapid decrease in the intensity of the resonancesassociated with complex ZrCp(CO)2[N(SiMeCHPPr)],4.18. 1H NMR (8, 300 MHz, C6D): 0.31 (s, 12H,Si(CH3)2;0.68 (m, 4H, SiCH2P); 0.85 to 1.0 (m, 24H, P[CH(CH)2j;1.70 (br sept, 4H,P[CH(CH],3J11.. = 7.4 Hz); 4.75 (t —5H, C5H3p-H = 1.5 Hz). Gated decouplled CNMR (6, 75.4 MHz, C7D8): 92.12 (s, Cp, 13C..H = 171.5 Hz); 249.0 (s, CO). 31P(1H} NMR (6,121.421 MHz, C6D): 37.91 (s). JR (cm4, toluene): 1965 (w, CO); 1871 (s, CO); 1845(shoulder, CO). Anal. Calcd forC25H49NC1OPSiZr:C, 49.63; H, 8.16; N, 2.32. Found: C,49.72; H, 8.35; N, ZrCp(BH[N(S1MeCHPPr1],4.19. 1H NMR (6, 300 MHz, C6D): 0.44 (s, 12H,Si(CH3)2;0.89 (br m, 4H, SiCH2P); 1.0 to 1.1 (m, 24H, P[CH(CH)2];2.00 (br sept, 4H,P[CH(CH1,3H.H = 7.4 Hz); 6.28 (t —5H, C5H3p..H = 1.5 Hz). ‘3C NMR (6, 75.4 MHz,C6D): 109.53 (s, Cp). 31P{1H} NMR (6, 121.421 MHz, C6D): 5.80 (s). 11B(H} NMR(6,96.2 MHz, C7D8): -92.00 (s, BH4). JR (cm-1, solution): 2427 (s, B-He), 2381 (s, B-Ht), 2130 (s,B-Hb).4.9c.3 HfCp(CO)[N(SiMeCHPPr],4.20. 1H NMR (6, 300 MHz, C6D): 0.30 (s, 12H,Si(CH3)2;0.75 (m, 4H, SiCH2P); 0.55 to 1.1 (m, 24H, P[CH(CH)2];1.70 (br sept, 4H,P[CH(CH],3J11.. = 8.0 Hz); 4.68 (t, 5H, C5H5, 3p.H = 1.5 Hz). 31P{ H} NMR (6, 121.421MHz, C6D): 32.92 (s).4.9c.4 ZrCp(CHCN)[N(SIMeHPPr],4.22. 1H NMR (6, 300 MHz, C6D): at 10 °C,0.02 (s, 6H, Si(CH3)2;0.15 (s, 6H, Si(CH3)2;1.12 to 1.26 (m, 24H, P[CH(CH)2];2.00 (hrsept, 2H, P[CH(CH]3J-p = 6.5 Hz); 2.36 (hr sept, 2H, P[CH(CH3)]3JwH = 6.5 Hz);references on pageChapter IV 1972.78 (s, 3H, CH3N); 5.66 (s, 5H, C5H). 31P(1H} NMR (6, 121.421 MHz, C6D): at -18 °C,32.92 (s).4.9c.5 Cp(BH)Zr(13H0)[N(SiMe2CHPPr’)(S1r•BH],4.23.1H NMR (6, 500 MHz, C7D8): 0.23, 0.35, 0.61, 0.69 (s, each 3H, Si(CH3)2;0.60 (overlappingd of d, 6H, (P[CH(CH3)2],3JH-H = 7.6 Hz, 3JP-H = 15.3 Hz); 0.94 (overlapping d of d, 6H,(P[CH(CH3)2j,3H-H = 7.1 Hz,3JpH = 15.3 Hz); 1.03 (two overlapping d, 2H, SiCH2P); 0.98to 1.11 (m, 14H, (P[CH(CH)2]and SiCH2P); 1.54 (sept, 1H, P[CH(CH3)2]3JwH = 7.1 Hz);1.77 (sept, 1H, P[CH(CH33JH.H = 7.1 Hz); 1.80 (sept, 2H, P[CH(CH3H-H = 7.6 Hz);3.10 (d of d, 1H, CHO, 2JpH = 27.4 Hz, 1JCH = 161.9 Hz); 6.41 (s, 5H, C5H). 31P{1H} NMR(8, 121.4 MHz, C7D8): 31.20 (br, 1P); 39.72 (d, 1P, 2Jp..c = 55.1 Hz). Gated decoupled ‘3CNMR (8, 75.4 MHz, C7D8): 60.73 (d of d, CHO, 2Jp.. = 55.7 Hz, UC-H = 162.2); 113.5 (d,C5H,1JC-H = 170.8 Hz). 11B{H} NMR (8, 96.2 MHz, C7D8): -101.33 (br s, BH4); -127.23(hr d, (BH3).P[CH(CH2j13P-B = 47.4 Hz). 1H(1B} NMR (6, 500 MHz, C7D8): 1.22 (s,3H, probably (BH3)•P[CH(CH2j;also showed some changes in the minor resonanceslocated around 0.2 to 0.4 ppm (probably, BH4). JR (cm4,KBr): 2423 (s, B-Ha, 2378 (s, B-Ht),2355 (s, B-Ht); 2345 (s, B-Ht); 2146 (s, B-Hb), 1460 (m, Zr-H). Anal. Calcd forC24H57BNOPSiZr:C, 47.52; H, 9.47; N, 2.31. Found: C, 47.93; H, 9.45; N, Oxidation ReactionsOxidation reactions with TiC13 or PbC12 was carried out by adding an excess of solid oxidant(approximately 10 equivalents) to a cooled (approximately -10 °C) toluene solution of thezirconium(III) complex. The resulting yellow or orange solution was then decanted, toluenewas striped off, and the residue was extracted with pentane and filtered through a layer ofCelite®.Oxidation with Ph2S was carried out by adding a toluene solution of the oxidant (0.5equivalents) to a solution of the zirconium(III) complex at -78 °C. Analytically purechlorothiolate complex, 4.13 was obtained (75%) by recrystallization from Et2O. Thealkylthiolate complexes gave only an oily material.references on pageChapter IV 1984.9e.1 ZrCpC1(SPh)[N(SiMe2HP r)],4.13. 1H NMR (6, 300 MHz, C6D): 0.71 (s, 6H,Si(CH3)2;0.74 (s, 6H, Si(CH3)2;0.93 to 1.15 (m, 24H, P[CH(CH3)2j;1.15 to 1.26 (m, 4H,SiCH2P); 2.28 (m, 4H, P[CH(CHJ);6.33 (s, 5H, Cp); 7.05 (t, 1H, p-Ph, 3JH-H = 7.4 Hz);7.24 (m, 2H, in-Ph); 7.47 (d, 2H, o-Ph, 3JwH = 7.4 Hz). 31P(1H} NMR (6, 121.421 MHz,C6D): 13.74 (s). 13C(H} NMR (6, 50.324 MHz, C6D): 113.34 (s, Cp). Anal. Calcd forC29H54NC1Si2P2SZr:C, 50.22; H, 7.85; N, 2.02. Found: C, 50.14; H, 7.87; N; ZrCpC1Et[N(SIMeHPPr] Complex 4.6 was oxidized with PbC12. 1H NMR (6,300 MHz, C6])): 0.40 (br s, 6H, Si(CH3)2, Avia = 40 Hz); 0.55 (s, 6H, Si(CH3)2, Avia = 40Hz); 0.75 to 0.90 (br, 6H, SiCH2Pand CH23);0.95 to 1.10 (m, 24H, P[CH(CH>]);1.78(br m, 4H, P[CH(CH3)2j2); 1.78 (t, 3H, CH233JwH = 7.6 Hz); 6.33 (s, Cp, 5H). 31P(1H}NIvIR (6, 121.421 MHz, C6D): -0.5 (br s, iXvl/2 = 370 Hz); 8.0 (br s, tWl/2 = 370 Hz). Anal.Calcd forC25HC1NSi2P2SZr: C, 48.94; H, 8.87; N, 2.28. Found: C, 48.30; H, 8.64; N; ZrCpC1(CHPh)[N(SIMeHr].Complex 4.7 was oxidized with TiC13. 1HNMR (6, 300 MHz, C6D): 0.40 (br s, 6H, Si(CH3)2); 0.50 (s, 6H, Si(CH3)2;0.70 (br, 4H,SiCH2P); 0.90 to 1.0 (m, 24H, P[CH(CH3)2j2); 1.70 (br m, 4H, P[CH(CH1);2.68 (d, 1H,CH2Ph,2JH..H = 11 Hz); 2.98 (d, 1H, CH2Ph,23H..H = 11 Hz); 6.20 (s, Cp, 5H). 31P(1H} NMR(6, 121.421 MHz, C6D6): -0.5 (br s, AV1 = 370 Hz); 8.0 (br s, 1XV12 = 370 Hz).4.9e.4ZrCp(CHSiMe3)(SPh)[N(SiMeCHPPr] Spectra were recorded within 2 hoursafter the reaction. 1H NIvIR (6, 300 MHz, C6D): 0.32 (s, 9H, CH2Si(CH3);0.42 (s, 6H,Si(CH3)2;0.52 (s, 6H, Si(CH3)2;0.87 (m, 4H, SiCH2P); 0.98 to 1.12 (m, 26H, P[CH(CH3)2]2andCHSi(CH); 1.80 (m, 4H, P[CH(CH3)2];6.40 (s, 5H, Cp); 6.98 (t, 1H, p-Ph, 3H-H =7.3 Hz); 7.18 (m, 2H, rn-Ph); 7.83 (d, 2H, o-Ph, 3JHH = 7.4 Hz). 31P(’H} NMR (6, 121.421MHz, C6D6): 2.50 (br s).Complicated spectra were obtained after 7 days. 1H NMR (6, 300 MHz, C6D): 6.08 (s, 5H,Cp); 6.28 (s, 5H, Cp). 31P{1H) NMR (6, 121.421 MHz, C6D6): -1.9 (s), -2.2 (s), 18.6 (s), 18.8(s).4.9e.5 ZrCp(CHCH3)(SPh)[N(SiMePPrJ.Spectra were recorded within 2 hoursafter the reaction. 1H NMR (6, 300 MHz, C6D): 0.45 (br s, 6H, Si(CH3)2,Av112 = 35 Hz); 0.68references on page QChapter IV 199(br s, 6H, Si(CH3)2Avia = 35 Hz); 0.85 to 0.90 (br m, 6H, SiCH2Pand CH23);0.95 to 1.15(m, 24H, P[CH(CH]);1.75 (m, 4H, P[CH(CH3)21;1.81 (t, 3H, CH233H-H = 6.7 Hz);6.13 (s, 5H, Cp); 7.00 (t, 1H,p-Ph, 3JH..H = 7.0 Hz); 7.18 (m, 2H, rn-Ph); 7.70 (d, 2H,o-Ph, 3JH..H = 7.0 Hz). 31P{1H) NMR (8, 121.421 MHz, C6D): -1.0 (br s, Av112 = 250 Hz); 11.0 (br s,Avl/2 = 250 Hz).Complex spectra were obtained after 7 days. 1H NMR (6, 300 MHz, C6D): 6.05 (s, 5H, Cp);6.24 (s, 5H, Cp). 31P{1H} NN’IR (6, 121.421 MHz, C6D): -1.8 (s), -2.1 (s), 18.6 (s), 18.8 (s).4.10 Refferences(1) Ashworth, T. V.; Agreda, T. C.; Herdtweck, E.; Herrmann, W. A. Angew. Chem.,Jnt. Ed.Engi. 1986,25, 289(2) Lappert, M. F.; Pickett, C. 3.; Riley, P. I.; Yarrow, P. I. W. J. Chem. Soc., Dalton Trans.1981, 805.(3) Blenkers, 3.; Bruin, P.; Teuben, 3. H. J. Organomet. Chem. 1985,297, 61.(4) Morse, P. M.; Wilson, S. R.; Girolami, G. S. Inorg. Chem. 1990,29, 3200.(5) Lubben, T. V.; Wolczanski, P. T.; vanDuyne, G. D. Organometallics 1984,3,977.(6) Baker, R. T.; Whitney, J. F.; Wreford, S. S. Organometallics 1983,2, 1049.(7) Fryzuk, M. D.; Haddad, T. S.; Berg, D. 3. Coord. Chem. Rev. 1990, 99, 137.(8) Fryzuk, M. D.; Montgomery, C. D. Coord. Chem. Rev. 1989, 95, 1.(9) Fryzuk, M. D. Can. J. Chem. 1992, 70, 2849.(10) Jeffery, 3.; Lappert, M. F.; Luong-Thi, N. T.; Atwood, J. L.; Hunter, W. E. J. Chem.Soc., Chem. Commun. 1978, 1081.(11) Lappert, M. F.; Riley, P. I.; Yarrow, P. I. W.; Atwood, J. L.; Hunter, W. E.; Zaworotko,M. J. J. Chem. Soc., Dalton Trans. 1981, 814.(12) Atwood, 3. L.; Barker, G. K.; Holton, 3.; Hunter, W. E.; Lappert, M. F.; Pearce, R. J. Am.Chem. Soc. 1977, 99, 6645.(13) Fryzuk, M. D.; Haddad, T. S.; Rettig, S. J. Organometallics 1989, 8, 1723.(14) Fitzgerald, R. J.; Drago, R. S. J. Am. Chem. Soc. 1968, 90, 2523.(15) Fessenden, R. W.; Schuler, R. H. J. Chem. Phys. 1963,39, 2147.(16) Brookhart, M.; Green, M. L. H.; Wong, L. Prog. Inorg. Chem. 1988,36, 1.(17) Brookhart, M.; Green, M. L. H. J. Organomet. Chem. 1983,250, 395.(18) Crabtree, R. H.; Hamilton, D. G. Adv. Organomet. Chem. 1988,28,299.(19) Crabtree, R. H.; Holt, E. M.; Lavin, M.; Morehouse, S. M. Inorg. Chem. 1985,24, 1986.(20) Wielstra, Y.; Gambarotta, S.; Spek, A. L.; Smeets, W. J. J. Organometallics 1990, 9,2142.Chapter IV 200(21) Cardin, D. J.; Lappert, M. F.; Raston, C. L. Chemistry of Organo-Zirconium andHafnium Compounds; 1 ed.; Ellis Horwood Limited: Toronto, 1986.(22) Bajgur, C. S.; Jones, C. B.; Petersen, J. F. Organometallics 1985,4, 1929.(23) Choukroun, R.; Gervais, D. J. Chem. Soc., Chem. Commun. 1985, 224.(24) Choukroun, R.; Basso-Bert, M.; Gervais, D. .1. Chem. Soc., Chem. Commun. 1986, 1317.(25) Choukroun, R.; Dahan, F.; Larsonneur, A. M.; Samuel, E.; Peterson, J.; Meunier, P.;Sornay, C. Organometallics 1991, 10, 374.(26) Jones, C. B.; Petersen, J. F. J. Am. Chem. Soc. 1983, 105, 5503.(27) Raoult, Y.; Choukroun, R.; blandy, C. Organometallics 1992, 11, 2443.(28) Samuel, E. Inorg. Chem. 1983,22,2967.(29) Elson, I. H.; Kochi, J. K.; Kiabunde, U.; Manzer, L. E.; Parshall, G. W.; Tebbe, F. N. J.Am. Chem. Soc. 1974,96, 7374.(30) Bercaw, J.; Brintzinger, H. J. Am. Chem. Soc. 1969,91, 7301.(31) Brintzinger, H. J. Am. Chem. Soc. 1967,89, 6871.(32) Kenworthy, 3. G.; Myatt, J.; Todd, P. F. J. Chem. Soc. (B) 1970, 791.(33) Luetkens, M. L.; Huffman, J. C.; Sattelberger, A. P. .1. Am. Chem. Soc. 1983, 105, 4474.(34) Jones, C. B.; Petersen, 3. F. Inorg. Chem. 1981,20, 2889.(35) Schwartz, J.; Libinger, 3. A. Angew. Chem., mt. Ed. Engi. 1976, 15, 333.(36) Coilman, J. P.; Hegedus, L. S.; Norton, J. R.; Finke, R. G. Principles and Applications ofOrganotranstion Metal Chemistry; second edition ed.; University Science Books: MillValley, CA., 1987.(37) Marks, T. 3.; Kolb, 3. R. Chem. Rev. 1977, 77, 263.(38) Jensen, J. A.; Gozum, J. E.; Pollina, D. M.; Girolami, G. S. J. Am. Chem. Soc. 1988, 110,1643.(39) Fryzuk, M. D.; Rettig, S. 3.; Westerhaus, A.; Williams, H. G. Inorg. Chem. 1985,24,4316.(40) Luetkens, M. L.; Huffman, J. C.; Sattelberger, A. P. .1. Am. Chem. Soc. 1985, 107, 3361.(41) Jensen, J. A.; Wilson, S. R.; Schultz, A. J.; Girolami, G. S. J. Am. Chem. Soc. 1987, 109,8094.(42) Raum, A. L. 3.; Frazer, D. A. British Patent. 1958, 801401.(43) Marks, T. J.; Kennelly, W. 3.; Koib, J. R.; Shimp, L. A. Inorg. Chem. 1972, 11, 2540.(44) Jensen, 3. A.; Girolami, G. S. J. Chem. Soc., Chem. Commun. 1986, 1160.(45) Edelstein, N. Inorg. Chem. 1981,20, 297.(46) Gozum, 3. E.; Girolan,i, G. S. J. Am. Chem. Soc. 1991, 113, 3829.(47) Plato, V.; Hedberg, K. Inorg. Chem. 1971, 10, 590.(48) Wielsira, Y.; Gambarotta, S.; Meetsma, A. Organometallics 1989, 8, 2948.(49) Floriani, C.; Fachinetti, G. J. Chem. Soc., Dalton Trans. 1973, 1954.Chapter IV 201(50) Fochi, G.; Guidi, G.; Floriani, C. J. Chem. Soc., Dalton Trans. 1984, 1253.(51) Tyler, D. R. Prog. Inorg. Chem. 1988,36, 125.(52) Masters, C. Adv. Organomet. Chem. 1979, 17, 61.(53) Herrmann, W. A. Angew. Chem., mt. Ed. Engl. 1982,21, 117.(54) Dumond, D. S.; Richmond, M. G. J. Am. Chem. Soc. 1988, 110, 7547.(55) Gladysz, 3. A.; Tam, W. J. Am. Chem. Soc. 1978, 100, 2545.(56) Casey, C. P.; Neumann, S. M. J. Am. Chem. Soc. 1978, 100, 2544.(57) Casey, C. P.; Andrews, M. A.; McAlister, D. R.; Rinz, 3. E. J. Am. Chem. Soc. 1980,102, 1925.(58) Tam, W.; Lin, G. Y.; Wong, W. K.; Kiel, W. A.; Wong, V. K.; Gladysz, 3. A. J. Am.Chem. Soc. 1982, 104, 141.(59) Wayland, B. B.; Woods, B. A.; Pierce, R. .1. Am. Chem. Soc. 1982, 104, 302.(60) Fagan, P. 3.; Moloy, K. G.; Marks, T. 3. .1. Am. Chem. Soc. 1981, 103, 6959.(61) Belmonte, P.; Schrock, R. R.; Churchill, M. R.; Youngs, W. 3. J. Am. Chem. Soc. 1980,102, 2858.(62) Bonnesen, P. V.; Yau, P. K. L.; Hersh, W. H. Organometallics 1987,6, 1587.(63) Churchill, M. R.; Rotella, F. 3. Inorg. Chem. 1978, 17, 2614.(64) Wolczanski, P. T.; Bercaw, 3. E. Acc. Chem. Res. 1980, 13, 121.(65) Wolczanski, P. T.; Threlkel, R. S.; Bercaw, J. E. J. Am. Chem. Soc. 1979, 101, 218.(66) Schlosser, M.; Ladenberger, V. J. Organomet. Chem. 1967, 8, 193.(67) Bryce-Smith, D.; Turnrt, E. E. J. Chem. Soc. 1953, 861.(68) Schiosser, M.; Hartmann, J. Angew. Chem., mt. Ed. Engi. 1973, 12, 508.(69) Schrock, R. R.; Fellmann, 3. D. J. Am. Chem. Soc. 1978, 100, 3359.(70) Smart, 3. C.; Curtis, C. 3. Inorg. Chem. 1977, 16, 1788.(71) Schrock, R. R. J. Organomet. Chem. 1976, 122, 209.(72) Stewart, C. P.; Porte, L. A. J. Chem. Soc., Dalton Trans. 1972, 15, 1661.(73) Fryzuk, M. D.; Carter, A.; Westerhaus, A. Inorg. Chem. 1985,24, 642.(74) Figgis, B. N.; Lewis, 3.; Prog. Inorg. Chem. 1964,6, 37.Chapter 5On-Going and Future ProspectsDinitrogen ChemistrySynthesisThe synthesis of side-on dinitrogen complexes having favorable solubility properties inhydrocarbon solvents is very important in order to investigate their stoichiometric reactions.The following possibilities could be considered: (a) to investigate the synthesis of side-ondinitrogen complexes analogous to 2.12 by incorporating aryloxy ligands having substituents onthe para positions; (b) to reduce ditriflate precursors of the type ZrR(OSO2CF3)2[N SiMe H-PPr)2J, where R = alkyl or aryl ligands (Scheme 5.1). In comparison with the reduction of thedichioro precursors, for example reduction of ZrC12(OAr*)[N(SiMe2CH2PPIr2)2 1 4.11, thesecond possibility may give better yields of the dinitrogen complex.[Zn = Zr[N(SiMe2CHPPr1)]Scheme 5.1Electrochemical studies with concomitant ESR measurements on the side-on dinitrogencomplexes could provide valuable information. ESR spectra of the oxidized species could yieldMe2Me2-CI2.1(1)NaOTf(2) RLiReductionDinitrogenComplexX=Y=MeX = Me, Y = ButMe2 sç, \ ph Reduction Hydrogenolysis:N—zr—R [Zn I [Zr]Me2 s(’ TfO”/Chapter V 203some information about the 6-MO (HOMO) which in turn can be compared with the results ofthe MO analysis.Reactivity StudiesReactions that would lead to the formation of Lewis-acid Lewis-base type “adducts” willprovide useful information regarding the nature of the dinitrogen ligand, particularly in theside-on dinitrogen complexes. The usefulness of such investigation will depend on the X-raystructure determination of the final product. The reactions shown in Scheme 5.2 are similar tothe reaction of LiBEt4with the side-on dinitrogen complexes discussed in chapter 2.[(C6F5)3BEtrLi.. Product similar to 2.24:{[C6H3( F)2-2,614B) H[Zr] = Zr[N(SiMe2CHPPr’)-{[C6H3( F)2-2,6]4B}Scheme 5.2Some reactions with ailcyl halides that could be investigated are shown in Scheme 5.3.The motive for such reactions would be to incorporate the activated dinitrogen ligand intoorganic molecules. Such reactions could also be useful for incorporating nitrogen- 15 intoorganic molecules. It is important to note that the end-on complex 2.6 is synthesized in greaterthan 90% yields and therefore is an ideal candidate to explore its reactivity with alkyl halides.The suggestions given in Scheme 5.3 take into consideration what has been discussed under thesection on reactivity of dinitrogen complexes in Chapter 2. Preliminary reactivity studies withBr(CH2)3rand Cl(CH2)4land dinitrogen complexes show a clean reaction, but the organicproducts are not yet identified.+On-Going and Future ProspectsChapter V 204Controlled reaction of the end-on dinitrogen complex with two equivalents of ketonecould provide a route to synthesize a zirconium oxo complex (Scheme 5.3).In the case of the side-on dinitrogen complexes the dinitrogen ligand forms a threemembered ring with each zirconium. The possibility of expanding these three memberedmetallacycles by the insertion of other unsaturated functionalities may be investigated(Scheme 5.3).Br2CR2Side-On [Dinitroen Complexes1-- End-On and Side-On_JBrCR2CRrRCp[Zr]N—NCRBr2CREnd-ON/0R2- -X[Zr]N—N=C IBrCR2CRr CR2xI—[Zr]MeO2CC eThe mechanism for the formation of the end-on dinitrogen complex could be envisagedas follows: during the reduction of the zirconium(IV) complexZrCpCl2[N(SiMeCHPPri),aX[Zr]()CR2RICp[Zr]=CR OrX[Zr] /[zrlxNR2CZNCR2Br(CH2)3r End-ON andCI(CH41 Side-On(CH2) Or (H2C) (CH(H2C)N—R2C=OCp[Zr]N—N=[Zr]Cp Cp[Zr]0 + R2CN—N=CRn = 1 or 2F2CCF2x;O2MeCO2Me[Zr] = Zr[N(SiMe2CHPPr’)Scheme 5.3On-Going and Future ProspectsChapter V 205zirconium(II) species of the type “ZrCp[N(SiMe2CH2PPr)2]”is probably formed which reactsfurther with dinitrogen to yield the end-on dinitrogen complex 2.9. This suggested mechanismcould be corroborated by the fact that stable zirconium(II) complexes, for exampleZrCp(CO)2[N(SiMe2CHPPr’) ], have been isolated. However, the reduction ofZrCl3[N(SiMeCHPPr)]2.1, may involve a different pathway. Attempts to isolate anyzirconium(H) species formed during the reduction of 2.1, for example reductions carried out inthe presence of PMe3,gave a brown material which contained mainly one product. Although itwas important to reduce 2.1 in the presence of PMe3 the 1H and 31P{1H} NMR spectra of thisbrown product did not show any resonances that could be attributed to a PMe3 ligand. Thisproduct is formulated as a dimeric species {ZrCl[N(SiMe2CHPPr)2]),which did not showany reactivity with dinitrogen at room temperature.ZrCI3[N(SiMe2CHPPr’)1L fl Reduction NZrClL(SiMe2CHPPr1)]L = N2[Zr] [Zr] [Zrr(R ci” bi ci” SL=PMe3jJ, L=THF’JRed solutionbrown probably decomposition N{ZrCI[N(SiMe2CHPPr)]12 by ring opening of [Zr [Zr]N x[Zr] = Zr[N(SiMe2CHPPr’)Scheme 5.4It is noteworthy that the reduction ofZrCl3[N(SiMe2CH2PPr2)]2.1 in toluene undervacuum gave uncoordinated PNP ligand whereas, repeating the same reduction in THF(Haddad, T. S.; UBC, unpublished work) gave a deep red solution. These results suggest thatduring the reduction of 2.1 an intermediate of the type “ZrClL[N(SiMe2CH2PPr)]”,R,where n<3 and L = dinitrogen, PMe3 or THF is produced. A speculative but intriguingmechanism for the formation of the side-on complex is shown in Scheme 5.4. Tn the case of theOn-Going and Future ProspectsChapter V 206intennediate S if the metal-metal bond has It symmetry then the insertion of the 7t*orbital (thatis contained in a plane parallel to the it-orbital associated with the metal-metal bond) of the“N2” unit should be symmetry allowed. It is important to point out that when model F (Section2.7.4) was modified by removing the dinitrogen ligand and then subjecting to INDO 1/MOanalysis showed the presence of one a and one it bond between the two zirconium centers.Investigating the reduction of ZrC13[N(SiMe2CH2-PPr)]with different substrates (e.g.,butadiene or R—CC—R) may provide some useful information about the reduced zirconiumspecies formed during the reduction of 2.1.Zirconium(III) ChemistrySynthesis and CharacterizationSince complexes of the type “ZrCp[N(SiMe2CH2PPr)2]” gave stable zirconium(IlI)complexes electrochemical measurements on these systems would be valuable information. Thechoice of the supporting electrolyte during electrochemical measurements is critical; forexample, attempts to use [NBu4J[PF6]-resulted in rapid decomposition of the zirconiumcomplex, presumably due to the attack by fluoride ions on the PNP ligand. The zirconiumcomplexes may be stable in the presence of electrolytes like [NBu4][BPh4]-.Low temperature solution and frozen glass ESR spectra with better resolution may beobtained for the alkyl and the borohydride derivatives.Disproportionation reactions could be extended to other ligands comparable to CO, forexample benzonitrile and tert-butylisocyanide. Monitoring the disproportionation reactions ofthe zirconium(ffl) chloro and the borohydride derivatives by UV-Vis spectroscopy shows thepresence of two isobestic points. By monitoring the band associated with the dicarbonylcomplex useful kinetic data could be obtained. The hydrogenolysis of the zirconium(III) ailcylsalso warrants kinetic measurements. Importantly, measurement of the kinetic isotope effect forthe hydrogenolysis would show whether the reaction proceeds by a-bond metathesis. Attemptsto investigate such studies using NMR techniques did show that the hydrogenolysis obeysOn-Going and Future ProspectsChapter V 207overall second order kinetics. However, more accurate (and reliable) measurements could beobtained by performing the experiments in a gas uptake apparatus.Excessive solubility of the zirconium(Ill) hydride complex has hindered the isolation ofthis complex as a solid material. Incorporating the phenyl version of PNP may lower thesolubility of the hydride complex and thereby enable the isolation of the complex as a pure solidmaterial. It important to note that previous attempts in our laboratory has shown that reactingLi[N(SiMe2CHPh)]with ZrC14 results in the formation of the bis ligand complexZrC1{N(SiMeCPPh However, using a different zirconium precursor may lead to thedesired product; for example, reaction involving Li[N(SiMe2CH2Ph2)2] and CpZrC13.(DME)could yield complex ZrCpCl2[N(SiMe2CHPh2) ] (Scheme 5.5). It is known that the reactionof CpZrC13•(DME) with the less bulky PN ligand Li[N(PhCH)(SiMe2PPh )j forms themono PN derivative of zirconium,ZrCpCl[N(Ph H)(SiMeCH2PPhJ.Me2Si (1) ReductionCpZrC13•(DME) :N— Zr—JL Me2Si” CI (2) Me3SiCHLi\___pLi[N(SiMe2CHPPh)] Ph2 (3) H2CpZrCl3•(DME)Li[N(CH2Ph)(SiMeCHPh)]Scheme 5.5ReactivityReactions involving some zirconium(Ill) complexes (e.g., 4.1) and B(C6F5)3gave adeep blue solution which is stable for approximately 10 minutes at room temperature. Theformation of the blue color seems to suggest the formation of the radical anion, [B(C6F5)3].On-Going and Future ProspectsChapter V 208For comparison the radical anion {B[C6H3(CF)2-2,6]}is stable at room temperature and hasan intense blue color in solution. Therefore reactions with B[C6H3(CF)2-2,6j andzirconium(III) complexes could be investigated.Cationic ComplexesPi2Zr( /\CIMe2Si—.N CICationic Species(1) Alkylation(2) B(C6F5(3) H{B[C(CFj4Dicationic SpeciesScheme 5.6Species generated by treating ZrCpCl2[N(SiMe2HPPr)jand alumoxanes have beenshown to be catalytically active towards ethylene polymerization (see Section 2.2.1, page 28).Many of the precursors that have been synthesized during the investigation of the dinitrogenchemistry are potential candidates for generating cationic complexes. Spectroscopic evidenceLi[N(But)(SiMe2CHPPr1](1) Alkylation(2) AgBPh4orB(C6F5)CpZrC13•(DME)ZrCl4.(THT)2-.-‘i) Alkylation(2) AgBPh4orMonocatiomc SpeciesLi[N(But)(SiMe2CHPPr1]LIDMECIOn-Going and Future ProspectsChapter V 209(1H and 31P{1H} NMR) suggests that the methyl derivative ZrCpMe2[N(SiCHPPr)]reacts with AgBPh4 in THF to yield two isomeric cationic species. Also, the dibenzylderivative Zr(OAr*)(PhCH2[N(SiMeCPPri]reacts with B(C6F5)3to generate specieswhich could be formulated as Zr(OAr*)(PhCH)[PhCH. B653][N(SiMeCHr).Oue efforts in ligand design have so far concentrated on the tridentate PNP systems.Very recently, we have developed an isopropyl version of the PN ligand,Li[N(SiMe2CHPPr) But)] that has been incorporated on to zirconium (Scheme 5.6). Somesuggested reactions that could lead to the synthesis of mono and dicationic species are shown inScheme 5.6.On-Going and Future ProspectsAppendixA.1 UV-Vis Spectra of Complexes 2.2 and 2.91.9000 304nm4’ {[(PPri2C11SiMe)N1ZrC1}2(N2) 2.21. 5200 646 nm, excitation wave lengthduring resonance Raman experiment.1.1400•‘ 0.760000.38000 402 nm 576 nm 646 nm4. 4.. 4.0.0000300 400 500 600 700L)VELNTHII0.950000.760000.57000.356nm {[(PPri2CHSiMe)NZr(fl5H5)}2 P’2) 2.90.38000.514 nm, excitation wave lengthduring resonance Raman experiment.0.19000‘I,300 400WRVELEHGTH500 600AppendixAppendix 211B..1 ESR SimulationsThe solution ESR spectra of the zirconium(III) complexes, i.e., complexes incorporatingPNP ligand, consist mainly of a triplet due to the coupling of two equivalent phosphorus nucleiof the PNP ligand. Therefore the a(31P) values used for the simulations were measured directlyfrom the actual solution spectra. The coupling constants due to other nuclei were obtained byiteration and the values reported gave the best possible fit between the observed and simulatedspectra. Simulations were carried out in such a manner that a lowest possible value is obtainedfor line broadening. When the coupling constants were lower than the value obtained for linebroadening it could be considered as the upper or the lower limit for the given nucleus wherethe limits are within 0.5 G. For example, in 4.17 the a(14N) value is the upper limit of theestimated coupling constant due to nitrogen of the PNP. It is important to note that in the caseof the ethyl derivative the values obtained from simulation correlate well with the deuterioderivatives, i.e., a(2H) 1/6 a(1H).(II) different from I is(I) line broadening = 2.8 G; a(1H) = 8.0 MHz, 4H; a(14N) = 4.5 MHz,1N.a(14N) = 5.5 MHz, iN; a(31P) = 58.0 MHz, of the parameters were identical for (Ill) thfferer!t from I issimulations 1,11 and III. hne broadening = 2.5 G.2’(I)(II) (Ill)AppendixAppendiXAppendixC.1 The Cartesian Coordinates for the Model Complexes Used inINIJO/1-MO AnalysisC.1.1 The Atom Numbering Scheme and the Cartesian Coordinates for the ModelComplex AH13212Atom X Y Z Atom X Y ZZr(1) 2.832 -0.227 0.138 P(38) -2.440 -0.722 -2.597C(7) 3.879 1.743 1.435 H(39) -1.204 -1.426 3.251C(8) 2.612 2.182 1.035 H(40) -3.233 -2.285 2.778C(9) 2.614 2.295 -0.360 H(41) -3.056 -0.180 3.562C(10) 3.881 1.924 -0.822 H(42) -1.175 -0.931 -3.110C(11) 4.664 1.583 0.287 H(43) -3.025 0.350 -3.240N(12) 4.096 -2.141 -0.021 H(44) -3.208 -1.851 -2.793P(15) 3.024 -0.746 -2.606 N(45) 0.934 -0.498 0.116P(16) 3.000 -1.214 2.753 N(46) -0.366 -0.486 0.118H(17) 1.760 -0.926 -3.130 H(2) 4.261 1.739 2.487H(18) 3.772 -1.888 -2.807 H(3) 1.826 2.595 1.716H25C351128C9c8H431141HiN341136H19H211120H35 1114Ni2AppendixAppendix 213-0.929H(20) 1.731 H(5) 4.269 2.098 -1.857H(21) 3.603 H(6) 5.772 1.432 0.274H(22) 3.747 H(13) 4.390 -2.510 -0.981Zr(23) -2.258 H(14) 4.391 -2.663 0.866C(29) -3.267 H(24) -3.648 2.170 -1.848C(30) -1.993 H(25) -1.198 2.843 -0.951C(31) -1.995 H(26) -1.173 2.622 1.694C(32) -3.270 H(27) -3.655 1.811 2.496C(33) -4.056 H(28) -5.168 1.531 0.287N(34) -3.557 H(35) -3.859 -2.588 0.875P(37) -2.464 H(36) -3.860 -2.436 -0.972C.1..2 The Atom Numbering Scheme and the Cartesian Coordinates for the ModelComplex BH20H18H(19) 3.634 0.320 -3.235 H(4) 1.800 2.811-1.481 3.226-0.273 3.562-2.374 2.760-0.180 0.1441.994 -0.8 102.338 -0.3462.221 1.0491.802 1.4461.662 0.297-2.072 -0.012-1.140 2.766H6C7H22H21H15 N13 H14AppendixAppendix 214Atom X Y Z Atom X Y ZZr(1) -0.031 -0.734 -0.065 H(20) 3.408 -0.100 0.236C(7) -1.127 1.553 -0.536 H(21) -3.252 -2.126 -0.138C(8) -0.757 1.044 -1.786 H(22) -3.408 0.023 0.521C(9) 0.641 1.010 -1.842 H(23) -2.757 -1.561 1.986C(10) 1.135 1.496 -0.627 H(2) -2.147 1.706 -0.182C(11) 0.042 1.832 0.181 H(3) -1.438 0.730 -2.577N(13) 0.047 -1.404 2.134 H(4) 1.240 0.665 -2.685P(16) 2.674 -1.249 0.444 H(5) 2.186 1.595 -0.356P(17) -2.700 -1.153 0.669 H(6) 0.093 2.240 1.191H(18) 3.121 -2.240 -0.406 H(14) 0.990 -1.582 2.606H(19) 2.826 -1.661 1.752 H(15) -0.861 -1.537 2.685C.1.3 The Atom Numbering Scheme and the Cartesian Coordinates for the ModelComplex HH14H13SiH19 1111H15H38114’AppendixAppendix 215Atom X Y Z Atom X Y ZZr(1) 1.900 0.018 -0.632 P(37) -3.290 0.107 -1.556P(2) 1.634 0.176 -3.468 0(40) -2.190 1.834 1.168P(3) 3.664 -0.492 1.507 N(41) -2.884 -1.820 0.9210(6) 2.605 1.911 -0.666 N(42) 0.414 -0.495 0.700N(7) 3.063 -1.725 -1.344 H(61) -2.607 2.792 1.068N(8) -0.132 -0.307 -0.737 H(4) 2.584 -2.348 -2.070H(9) 0.324 -0.074 -3.821 H(5) 4.059 -1.952 -1.027H(10) 2.457 -0.760 -4.061 H(15) -0.167 -1.009 3.803H(11) 1.992 1.439 -3.894 H(16) -1.741 0.555 4.198H(12) 2.948 -0.945 2.596 H(17) -2.347 -1.580 3.809H(13) 4.582 -1.451 1.129 H(18) -2.553 -0.108 -2.703H(14) 4.327 0.671 1.844 H(19) -3.874 1.356 -1.613H(27) 3.075 2.794 -0.349 H(20) -4.273 -0.856 -1.454Zr(35) -1.595 -0.037 0.692 H(38) -2.473 -2.631 1.485P(36) -1.443 -0.589 3.486 H(39) -3.877 -1.896 0.530C.1.4 The Atom Numbering Scheme and the Cartesian Coordinates for the ModelComplex MH61 h27HilN42H38H4 H5 1113AppendixAppendix 216Atom X Y Z Atom X Y ZZr(1) 1.815 -0.023 -0.502 P(37) -3.263 0.262 -2.011P(2) 1.820 0.400 -3.323 0(40) -2.869 1.494 1.142P(3) 3.238 -1.084 1.688 N(41) -2.862 -2.079 0.0810(6) 2.930 1.651 -0.322 N(42) 0.148 -0.272 0.685N(7) 2.605 -1.932 -1.294 H(61) -3.432 2.380 1.159N(8) -0.224 0.143 -0.761 H(4) 2.057 -2.371 -2.102H(9) 0.520 0.489 -3.777 H(5) 3.495 -2.406 -0.938H(10) 2.455 -0.655 -3.946 H(15) -0.752 -1.530 3.486H(11) 2.488 1.573 -3.611 H(16) -2.617 -0.335 3.900H(12) 2.352 -1.437 2.685 H(17) -2.776 -2.398 3.010H(13) 3.942 -2.200 1.282 H(18) -2.350 0.411 -3.034H(14) 4.117 -0.135 2.168 H(19) -4.046 1.395 -1.924H(27) 3.560 2.376 0.102 H(20) -4.064 -0.833 -2.264Zr(35) -1.892 -0.121 0.423 H(38) -2.395 -2.925 0.541P(36) -2.027 -1.240 3.043 H(39) -3.763 -2.212 -0.480AppendixD.1X-rayCrystallographicAnalysisDataD.1.1X-rayCrystallographicAnalysisof Zr(15-CH)Br2[N(S1MeCHPPr’],2.7EmpiricalFormulaFormulaWeightCrystalColor,HabitCrystalDimensions(mm)CrystalSystemNo.ReflectionsUsedforUnitCellDeterminatiOn(20range)OmegaScanPeakWidthatHalf—heightLatticeParameters:C23H49BrNPSiZr708.79yellow,hexagonal0.080X0.300x0.300monoclinic25(28.5—37.6)a—16.310(2)Ab—10.678(2)Ac—18.306(3)A5—100.59(l)°V—3133.9(9)A3P21/a(*14)4 1.502g/cm31448ScanWidth20maxNo.ofReflectionsMeasuredCorrectionsA.CrystalData0.41CrystaltoDetectorDistance285mmScanType—29ScanRate16.0/min(inomega)(8rescans)(1.05+0.35tano)°SS.0Total:7845Unique:7582(Ri0t—.045)Lorentz—polarizationAbsorption(trans.factors:0.54—1.00)Decay(—3.10%decline)SecondaryExtinction(coefficient:0.30238E—06)C.StructureSolutionandRefinementStructureSolutionPattersonMethodRefinementFull—matrixleast—squaresFunctionMinimized£w(Fol—FcILeast—squaresWeights4Fo2/o(Fo)p—factor0.02AnomalousDispersionAllnon—hydrogenatomsNo.Observations(I>3.OOq(I))3645No.Variables281Reflection/ParameterRatio12.97Residuals:R;R0.031;0.030GoodnessofFitIndicator1.27MaxShift/ErrorinFinalCycle0.01MaximumPeakinFinalDiff.Map0.37eiA3MinimumPeakinFinalDiff.Map—0.41e/A3SpaceGroupIvalueF000‘(MoKa)DiffractometerRadiationTemperatureTake—offAngleDetectorAperture30.S9cm4B.IntensityMeasurementsRigakuAFC6SMOKa(X—0.71069A)21°C6.0°6.0mmhorizontal6.0mmverticalID.1.1X-rayCrystallographicAnalysisofZr(rj5-CHs)Br2[N(SiMeCHPPr2)2],2.7continuedFinalatomiccoordinates(fractional)andBegA2*Table.Finalatomiccoordinates(fractional)andBeq(2)*(cont.)atomxYatomxYzBegZr(l)0.29705(2)0.37118(4)0.25013(2)2.25(2)C(17)0.5541(3)0.5806(5)0.2841(3)4.6(2)Br(l)0.44783(3)0.30222(5)0.32472(3)3.71(2)C(18)0.4565(3)0.7092(5)0.1517(3)4.9(3)Br(2)0.17026(3)0.50463(5)0.17361(3)3.86(2)C(19)0.2555(3)0.1477(4)0.2039(3)4.2(2)P(l)0.26513(7)0.4991(1)0.37654(7)2.84(5)C(20)0.2910(3)0.1344(5)0.2774(4)5.1(3)P(2)0.36988(7)0.3501(1)0.12231(6)2.79(5)C(21)0.2398(4)0.1941(5)0.3197(3)5.1(3)Si(l)0.32572(8)0.6925(1)0.27709(7)3.00(5)C(22)0.1720(3)0.2407(5)0.2708(4)4.8(3)Si(2)0.45539(8)0.5683(1)0.21340(7)3.05(5)C(23)0.1812(3)0.2127(5)0.1998(3)4.3(2)N(1)0.3629(2)0.5532(3)0.2469(2)2.3(1)C(l)0.2S03(3)0.6570(4)0.3415(2)3.2(2)2*Be—(8/3)£tVjjaj*a*(ai.aj)C(2)0.4684(3)0.4274(4)0.1531(2)3.2(2)qC(3)0.3600(3)0.5045(5)0.4501(3)3.9(2)C(4)0.1738(3)0.4598(5)0.4203(3)4.4(2)C(S)0.3721(4)0.3798(6)0.4930(3)5.5(3)C(6)0.3693(3)0.6174(6)0.5029(3)5.7(3)C(7)0.1737(4)0.5150(6)0.4964(4)6.7(3)C(8)0.0931(3)0.4956(6)0.3687(4)6.1(3)C(9)0.3118(3)0.4465(5)0.0453(3)3.8(2)C(l0)0.3940(3)0.1985(4)0.0811(3)3.8(2)C(l1)0.2345(3)0.3758(6)0.0068(3)5.2(3)C(12)0.3606(4)0.5018(6)—0.0106(3)5.7(3)C(13)0.4445(3)0.1140(5)0.1406(3)5.2(3)C(14)0.4367(4)0.2063(6)0.0143(3)6.8(3)C(15)0.2702(3)0.7997(4)0.2028(3)4.6(2)C(16)0.4098(3)0.7913(5)0.3330(3)4.8(3)00D.1.1X-rayCrystallographicAnalysisof Zr(15-CH)Br2[N(SiMeCHPPr’],2.7continuedTable.Bondangles(deg)withestimatedstandarddeviations.Table.Bondlengths(A)withestimatedstandarddeviations.*atomatomatomangleatomatomatomangleatomatomdistanceatomatomdistanceBr(l)Zr(1)Br(2)162.95(2)C(9)P(2)C(l0)106.9(2)zr(l)Br(l)2.6871(7)Si(l)C(15)1.879(5)Br(l)Zr(l)P(l)90.31(3)N(l)Si(l)C(l)109.4(2)Zr(l)Br(2)2.6842(7)Si(l)C(16)1.877(5)Br(l)Zr(l)P(2)85.08(3)N(l)Si(l)C(15)116.2(2)Zr(l)P(l)2.817(1)Si(2)N(l)1.738(3)Br(l)Zr(l)N(l)81.50(8)N(l)Si(1)C(16)113.1(2)Zr(l)P(2)2.821(1)Si(2)C(2)1.900(5)Br(l)Zr(1)Cp(l)99.02(7)C(1)Si(1)C(15)106.9(2)Zc(l)t4(1)2.227(3)Si(2)C(17)1.876(5)Br(2)Zr(1)P(l)86.00(3)C(1)Si(1)C(16)105.5(2)Zr(1)C(19)2.581(5)Si(2)C(18)1.884(5)Br(2)Zr(1)P(2)90.62(3)C(15)Si(l)C(16)105.0(2)Zr(1)C(20)2.583(5)C(3)C(5)1.541(7)Br(2)Zr(1)N(1)81.46(8)N(l)Si(2)C(2)109.2(2)Zr(1)C(21)2.551(5)C(3)C(6)1.535(7)Br(2)Zr(1)Cp(1)98.03(6)N(1)Si(2)C(17)117.0(2)Zr(1)C(22)2.555(5)C(4)C(7)1.514(8)P(1)Zr(1)P(2)152.82(4)(1)Si(2)C(18)113.0(2)Zr(1)C(23)2.578(5)C(4)C(8)1.522(8)p(l)Zr(l)N(1)76.18(9)C(2)Si(2)C(17)106.4(2)Zr(1)Cp(1)2.286(2)C(S)C(ll)1.527(7)P(1)Zr(1)Cp(1)103.02(7)C(2)Si(2)C(18)105.6(2)P(1)C(1)1.804(4)C(9)C(12)1.526(7)P(2)Zr(1)r1(1)76.65(9)C(17)Si(2)c(18)104.9(2)PCi)C(3)1.856(5)C(10)C(13)1.532(7)P(2)Zr(1)Cp(1)104.16(7)Zr(1)N(1)Si(l)122.8(2)P(1)C(4)1.863(5)C(10)C(14)1.517(7)N(1)Zr(1)Cp(1)179.2(1)Zr(l)N(1)Si(2)123.0(2)P(2)C(2)1.802(4)C(19)C(20)1.371(7)Zr(l)P(1)C(1)101.2(1)Si(l)14(1)Si(2)114.2(2)P(2)C(9)1.858(5)C(19)C(23)1.386(7)Zr(1)P(1)C(3)110.8(2)P(l)C(1)Si(1)110.4(2)P(2)C(10)1.859(5)C(20)C(21)1.393(8)Zr(1)P(l)C(4)121.8(2)P(2)C(2)Si(2)110.9(2)si(1)14(1)1.734(3)C(21)C(22)1.382(8)C(1)P(l)C(3)105.6(2)P(1)C(3)C(S)110.8(4)Si(1)C(1)1.893(5)C(22)C(23)1.369(7)C(1)P(1)C(4)107.3(2)P(1)C(3)C(6)117.0(4)C(S)P(1)C(4)108.7(2)C(S)C(3)C(6)111.6(4)*Hereandelsewhere,Cp(1)referstotheunweightedcentroidZr(1)P(2)C(2)101.0(1)P(l)C(4)C(7)116.4(4)oftheC(19—23)cyclopentadienylring.Zr(l)P(2)C(9)110.6(2)P(i)C(4)C(8)110.3(4)Zr(1)P(2)C(lO)124.0(2)C(7)C(4)C(8)109.0(4)C(2).P(2)C(9)106.5(2)P(2)C(S)C(11)109.8(4)C(2)P(2)C(10)106.4(2)P(2)C(9)C(12)117.9(4)AppendixzInInIn5220AppendixD.1.2X-rayCrystallographicAnalysisof {{[(PPr12CHSiMe)N]Zr(r5-CH}(j.i-N,2.9continuedEXPERIMENTALDETAILSA.CrystalDataEmpiricalFormulaFormulaWeightCrystalColor,HabitCrystalDimensions(mm)CrystalSystemNo.RefleCtionSUsedforUnitCellDetermination(28range)OmegaScanPeakWidthatHalf—heightLatticeParameters:SpaceGroupZvalueDcalcF000DiffractometerRadiationTemperatureTake—offAngleDetectorAperture8 1.2369/cm34968ScanWidth28maxNo.ofReflectionsMeasuredCorrectionsC.StructureSolutionandStructureSolutionRefinementFunctionMinimizedLeast—squaresWeightsp—factorAnomalousDispersionNo.Observations(I>3.OOq(I))No.VariablesReflection/ParameterRatioResiduals:R;GoodnessofFitIndicatorMaxShift/ErrorinFinalCycleMaximumPeakinFinalDiff.MapMinimumPeakinFinalDiff.Map32.0°/mm(inomega)(8rescans)(0.91+0.35tan9)55.0Total:15389Unique:15197(Rit.025)Lorentz—polarizationAbsorption(trans.factors:0.97—1.00)RefinementPattersonMethodFull—matrixleast—squaresIw(IFOL—Fc()24Fo2/a(Fo) 0.02Allnon—hydrogenatoms7287S7312.720.034;0.0351.40C49H103NPSiZr2 1167.06red—brown,prism0.120x0.250X0.450monoclinic25(29.2—33.8°)CrystaltoDetectorDistanceScanTypeScanRate285mm0.36a— b— c— 6— V C2/c48.755(4)A10.226(7)A25.987(4)A104.582(9)°12539(9)A3(#15)5.34B.IntensityMeasurementsRigakuAFC6SMOKa(X—0.71069A)21°C6.0°6.0mmhorizontal6.0mmvertical0.210.37e/A3—0.27e/A3D.1.2X-rayCrystallographicAnalysisof{{[(PPr12CHSiMe)NIZr(115-CSHS)}20.L-N2),2.9continuedTable.Finalatomiccoordinates(fractional)andBeg(A2)*(cont.)Table.Finalatomiccoordinates(fractional)andBeg(2)*atomxYzBeg0CC.atomxzBeg0cc.C(ll)0.1526(1)0.4564(4)0.1775(2)5.5(2)Zr(l)0.113703(7)0.09216(3)0.13299(1)2.83(1)C(12)0.1522(1)0.5547(4)0.0885(2)4.6(2)Zr(2)0.150099(7)0.11657(3)0.33438(1)2.77(1)C(13)0.1909(1)0.1769(5)0.0516(2)5.7(2)P(l)0.14629(2)0.2757(1)0.09337(4)3.12(4)C(14)0.20304(9)0.2428(5)0.1470(2)5.7(2)P(2)0.06178(2)0.0400(1)0.15426(4)3.71(5)C(15)0.0346(1)—0.0340(5)0.2368(2)6.7(3)P(3)0.20307(2)0.0731(1)0.31595(4)3.52(4)C(160.0723(1)—0.1851(5>0.2188(2)6.5(3)?(4)0.11478(2)0.2988(1)0.36856(4)3.12(4)C(17)0.0011(1)0.0208(6)0.1086(2)7.9(3)Si(l)0.08718(2)0.2905(1)0.02207(4)3.79(5)C(18)0.0324(2)—0.144(1)0.0781(5)7.9(6)0.565Si(2)0.05801(2)0.3162(1)0.10898(4)3.71(5>C(18A)0.0331(3)—0.051(2)0.0548(5)7.9(8)0.435Si(3)0.20215(2)0.3570(1)0.35039(4)3.49(5)C(l9)0.20888(8)0.2406(4)0.2986(1)3.7(2)Si(4)0.17371(2)0.3367(1)0.43869(4)3.75(5)C(20)0.13492(8)0.3229(4)0.4371(1)3.8(2)N(1)0.08336(6)0.2446(3)0.0834(1)3.2(1)C(21)0.2051(1)—0.0219(4)0.2564(2)4.6(2)N(2)0.17794(6)0.2841(3)0.3784(1)3.0(1)C(22)0.2346(1)0.0304(6)0.3700(2)7.3(3)N(3)0.12671(6)0.1315(3)0.2074(1)2.7(1)c(23)0.11695(8)0.4624(4)0.3389(1)3.6(2)N(4)0.13675(6)0.1382(3)0.2587(1)2.6(1)C(24)0.07689(8)0.2869(4)0.3714(2)4.2(2)C(l)0.12611(8)0.2856(4)0.0241(1)3.7(2)c(25)0.23721(9)0.4001(5)0.3984(2)5.6(2)C(2)0.05336(8)0.2081(4)0.1654(2)3.9(2)C(26)0.1900(1)0.5132(4)0.3149(2)4.9(2)C(3)0.14174(8)0.4433(4)0.1175(1)3.4(2)C(27)0.1937(1)0.2362(5)0.4961(2)6.0(2)C(4)0.18427(8)0.2707(4)0.0920(1)3.7(2)C(28)0.1837(1)0.5119(5)0.4555(2)5.6(2)C(5)0.0621(1)—0.0427(4)0.2175(2)4.8(2)C(29)0.1977(1)—0.1658(5)0.2604(2)6.8(3)C(6)0.0305(1)—0.0207(6)0.1030(2)7.5(3)C(30)0.2326(1)—0.0088(5)0.2381(2)6.5(3)C(7)0.0685(1)0.1792(5)—0.0327(2)5.8(2>C(31)0.2630(1)0.0808(5)0.3648(2)7.8(3)C(8>0.0750(1)0.4617(5)0.0007(2)5.6(2)c(32)0.2326(3)0.013(2)0.4203(4)8.1(7)0.510C(9)0.0223(1)0.3373(5)0.0602(2)6.0(2)c(32A).0.2336(3)—0.092(1)0.4002(6)8.1(7)0.490C(10)0.0674(1)0.4827(4)0.1380(2)4.8(2)C(33)0.1058(1)0.4622(4)0.2789(2)5.4(2)t’Jt.JD.1.2X-rayCrystallographicAnalysisof{[(PPriZCH2SiMe)N1Zr(T15-C5H5)}2(.t-N2),2.9continuedIntramolecularBondAnglesatomatomatomangleatomatomatomangleTable.Finalatomiccoordinates(fractional)andBeq2)*(coot.)P(l)Zr(l)P(2)145.87(4)zr(l)P(2)c(2)96.0(1)PU)Zr(l)N(1)71.74(8)Zr(1)P(2)c(5)117.5(1)atomXYZBegoCC•PCI.)Zr(l)N(3)98.92(8)Zr(1)P(2)C(6)123.1(2)C(34)0.1049(1)0.5753(4)0.3644(2)5.0(2)P(1)Zr(1)Cp(l)102.89(6)C(2)P(2)C(5)104.1(2)C(35)0.0718(1)0.2046(5)0.4160(2)6.8(3)P(2)Zr(l)N(l)75.66(8)C(2)P(2)C(6)104.7(2)C(36)0.0579(1)0.2439(6)0.3183(2)7.0(3)P(2)Zr(l)N(3)85.76(8)C(S)P(2)C(6)107.9(3)C(37)0.1081(1)—0.1573(4)0.1142(2)4.9(2)P(2)Zr(1)Cp(1)104.84(6)Zr(2)P(3)C(19)95.9(1)C(38)0.1343(1)—0.1375(4)0.1494(2)4.9(2)N(l)Zr(1)N(3)115.1(1)zr(2)P(3)c(21)118.4(1)C(39)0.1514(1)—0.0698(4)0.1232(2)4.8(2)N(l)Zr(1)Cp(1)127.61(9)Zr(2)P(3)C(22)122.4(2)C(40)0.1357(1)—0.0498(4)0.0706(2)5.0(2)N(3)Zr(l)Cp(1)117.2(1)C(19)P(3)C(21)104.1(2)C(41)0.1089(1)—0.1025(4)0.0656(2)5.0(2)P(3)Zr(2)P(4)145.65(3)C(19)P(3)C(22)105.0(2)C(42)0.1297(1)—0.0177(4)0.4008(2)5.3(2)P(3)Zr(2)N(2)75.37(7)c21P(3)c(22)107.5(2)C(43)0.1572(1)—0.0620(5)0.4090(2)5.7(2)P(3)Zr(2)N(4)86.10(8)Zr(2)P(4)C(20)100.1(1)C(44)0.1596(1)—0.1245(4)0.3630(2)5.7(2)p(3)Zr(2)Cp(2)104.84(6)Zr(2)P(4)c(23)111.1(1)C(45)0.1336(1)—0.1200(4)0.3262(2)5.4(2)P(4)Zr(2)N(2)71.53(8)1r2)P(4)C(24)130.9(1)C(46)0.1148(1)—0.0518(4)0.3491(2)5.1(2)p4Zr(2)N(4)99.76(8)C(20)P(4)C(23)102.2(2)C(47)0.4857(2)0.107(1)0.1510(4)15.4(3)P(4)Zr(2)Cp(2)102.41(6)C(20)P(4)C(24)105.4(2)C(48)1/20.055(3)1/421.3(8)0.500N(2)zr(2)N(4)115.3(1)c(23)P(4)c(24)103.6(2)C(49)0.4873(7)0.115(3)0.216(2)23(1)0.500N(2)lr(2)Cp(2)127.49(9)N(1)Si(1)C(1)108.1(2)C(50)0.4944(3)0.035(2)0.1934(7)22.7(6)N(4)Zr(2)Cp(2)117.1(1)N(1)Si(1)C(7)113.2(2)Zr(1)P(1)C(1)100.6(1)N(1)Si(1)C(8)115.3(2)Zr(l)PCI)C(3)111.4(1)C(1)Si(1)C(7)107.4(2)(8/3)n2EU1a*a*(aj.aj)Zr(1)P(1)C(4)130.1(1)C(1)Si(1)C(8)105.8(2)eqC(1)P(1)C(3)101.7(2)C(7)Si(1)C(8)106.5(2)C(l)p(1)C(4)105.9(2)N(1)Si(2)C(2)106.9(2)C(3)P(1)C(4)103.7(2)N(1)si(2)C(9)115.0(2)Anglesareindegrees.Estimatedstandarddeviationsintheleastsignificantfigurearegiveninparentheses.t’-)D.1.2X-rayCrystallographicAnalysisof {[(PPr’2CH2SiMe2)N]Zr(1&CsHs)}(p.N2.9continuedIntramolecularDistancesIntramolecularBondAngles(cont.)atomatomdistanceatomatomdjtanceatomatomatomangleatomatomatomanglezr(1)P(l)2.816(1)9(3)C(19)1.812(4)N(1)Si(2)C(10)114.8(2)9(1)C(3)C(ll)112.8(3)Zr(l)9(2)2.776(1)9(3)C(21)1.852(4)C(2)Si(2)C(9)108.4(2)9(1)C(3)C(12)116.1(3)zr(1)N(l)2.306(3)p(3)C(22)1.854(5)C(2)Si(2)C(10)106.5(2)C(11)C(3)C(12)111.5(3)zr(l)N(3)1.920(3)9(4)C(20)1.821(4)C(9)Si(2)C(l0)104.9(2)P(1)C(4)C(13)114.7(3)zr(l)C(37)2.599(4)9(4)C(23)1.857(4)N(2)Si(3)C(19)107.0(2)P(1)C(4)C(14)110.7(3)Zr(1)C(38)2.547(4)P(4)c(24)1.871(4)N(2)Si(3)C(25)115.3(2)C(13)C(4)C(14)110.5(4)Zr(1)C(39)2.536(4)Si(1)N(1)1.716(3)N(2)Si(3)C(26)114.7(2)P(2)C(5)C(l5)116.6(3)Zr(1)C(40)2.600(4)Si(1)C(l)1.887(4)C(19)Si(3JC(25)109.0(2)9(2)C(5)C(16)112.6(3)Zr(l)C(41)2.624(4)si(1)C(7)1.870(4)C(19)Si(3)C(26)106.1(2)C(15)C(5)C(16)110.4(4)Zr(l)Cp(1)2.294(2)si(1)C(8)1.887(5)C(25)Si(3)C(26)104.3(2)9(2)C(6)C(17)117.1(4)Zr(2)9(3)2.778(1)si(2)N(l)1.709(3)N(2)si(4)C(20)107.8(2)9(2)C(6)C(l8)118.9(6)zr(2)P(4)2.828(1)Si(2)C(2)1.895(4)N(2)Si(4)C(27)113.4(2)9(2)C(6)C(l8A)119.9(7)zr(2)N(2)2.303(3)si(2)C(9)1.890(4)N(2)Si(4)C(28)115.3(2)C(17)C(6)C(18)117.0(6)zr(2)N(4)1.923(3)si(2)C(10)1.872(4)C(20)Si(4)C(27)107.2(2)C(17)C(6)C(l8A)117.7(8)Zr(2)C(42)2.591(4)Si(3)N(2)1.706(3)C(20)Si(4)C(28)106.0(2)9(3)C(19)31(3)110.6(2)Zr(2)C(43)2.622(4)si(3)C(19)1.887(4)C(27)Si(4)C(28)106.6(2)9(4)c(20)Si(4)108.8(2)Zr(2)C(44)2.583(4)Si(3)C(25)1.897(4)Zr(1)N(1)31(1)120.0(1)9(3)C(21)C(29)112.7(3)Zr(2)C(45)2.541(4)Si(3)C(26)1.865(4)Zr(l)N(1)81(2)119.8(1)9(3)C(21)C(30)116.5(3)Zr(2)C(46)2.529(4)si(4)N(2)1.718(3)Si(l)N(1)Si(2)120.2(2)C(29)C(21)C(30)110.1(4)Zr(2)Cp(2)2.286(2)Si(4)C(20)1.886(4)Zr(2)N(2)Si(3)119.9(1)9(3)C(22)C(31)117.0(4)9(1)C(1)1.825(4)Si(4)C(27)1.870(4)Zr(2)N(2)Si(4)119.7(1)9(3)C(22)C(32)121.3(6)9(1)C(3)1.857(4)Si(4)C(28)1.879(5)Si(3)N(2)Si(4)120.5(2)9(3)C(22)C(32A)117.8(6)9(1)C(4)1.862(4)N(3)N(4)1.301(3)Zr(1)N(3)N(4)170.6(2)C(31)C(22)C(32)115.0(7)9(2)C(2)1.808(4)C(3)C(11)1.520(5)Zr(2)N(4)N(3)170.1(2)C(31)C(22)C(32A)119.1(6)P(2)C(S)1.845(4)C(3)C(12)1.523(5)9(1)C(l)31(1)108.5(2)P(4)C(23)C(33)112.5(3)9(2)C(6)1.860(5)C(4)C(13)1.515(5)9(2)•C(2)Sj(2)110.8(2)9(4)C(23)C(34)115.7(3)Distancesareinangstroms.EstimatedstandarddeviationsinP.nglesareindegrees.Estimatedstandarddeviationsintheleasttheleastsignificantfigurearegiveninparentheses.significantfigurearegiveninparentheses.D.1.3X-rayCrystallographicAnalysisof{[(PPri2CH2SiMe2)2N]Zr(OAr*}2(iN2),2.12aEXPERIMENTALDETAILSA.CrystalDataEmpiricalFormulaFormulaWeightCrystal Color,HabitCrystalDimensionsCrystalSystemLatticeTypeNo.ofReflectionsUsedforUnitCellDetermination(20range)OmegaScanPeakWidthatHalf-heightLatticeParametersCs2H1o6N4P4Si4OZr2 1238.11dark,prism0.15X0.25X0.25mmorthorhombicPrimitive25(46.4-72.70.38a=14.384(1) Ab=17.754(1)A=25.997(1) ASpaceGroupZvalueF000o(CuKo)V=6638.8(7)APbcn(#60)4 1.239g/cm3263245.53cmB.lnten8ityMeasurementsDiffractometerRediationTake-offAngleRigakuAFC6SCuKo(A=1.54178A)graphitemonochromated6.0I. t..J cMD.1.3X-rayCrystallographicAnalysisof{[(PPri2CHSiMe)N]Zr(OAr*}(p.N,2.12acontinuedTableIT.AtomiccoordinatesandBqDetectorAperture6.0mmhorizontal6.0mmverticalatomYCrystal toDetectorDistance285mmZr(1)0.02851(3)0.25905(2)0.17979(1)2.990(7)Temperature21.0CCP(1)0.22304(9)0.27162(8)0.17013(5)3.85(3)ScanTypew-29P(2).0.14513(9)0.26760(8)0.12761(5)4.01(3)ScanRate16.05/min(inu)(upto9scans)Si(1)0.1306(1)0.37463(9)0.08792(6)4.34(4)ScaoWidth(0.84+020tan9)5Si(2).0.0524(1)0.42082(8)0.13069(5)3.92(3)2Omar155.50(1)0.0407(2)0.1522(2)0.1539(1)437(9)No.ofReflectionsMeasuredTotal:6948N(1)0.0391(3)0.3585(2)0.1289(1)3.38(9)CorrectionsLrentz-polarizationAbsorptionN(2).0.0514(2)0.2812(2)0.2426(1)3.49(9)(trans.factors:0.78-1.00)Decay(2.01%decline)CCI)0.2335(4)0.3145(3)0.1066(2)4.5(1)SecondaryExtinction(coefficient: 9.2(5)x_5)C(2).0.1621(3)0.3659(3)0.1441(2)4.3(1)C(3)0.1714(4)0.4750(3)0.0843(2)6.7(2)C.StructureSolutionandRefinementC(4)0.1029(4)0.3477(3)0.0201(2)6.0(2)C(5)-0.0394(4)0.4923(3)0.1825(2)6.0(2)StructureSolutionPattersonMethods(DIRDIFO2 PATI’Y)C(6)-0.0739(4)0.4746(3)0.0695(2)5.9(2)RefinementFull-matrixleast-squaresC(7)0.2788(4)0.3375(3)0.2157(2)5.2(2)FunctionMinimizedEw(IFoI—IFcD2C(8)0.2961(4)0.1859(4)0.1871(3)6.4(2)LeastSquaresWeights;‘1p-y=C(9)0.2276(4)0.4113(3)0.2172(2)6.1(2)p-factor0.000cOo)0.3822(5)0.3496(5)0.2096(3)10.3(3)AnomalousDispersionAllnon-hydrogenatomsC(11)0.2974(6)0.1423(4)0.2137(3)10.4(3)No.Observations(1>3.OOcT(I))3444C(12)0.3853(4)0.1886(4)0.1366(3)6.9(2)No.Variables308C(13)-0.1389(4)0.2706(3)0.0568(2)5.1(1)Reflection/ParameterRatio11.18C(14)-0.2518(4)0.2174(3)0.1481(2)5.4(2)Iesiduals:R;Rw0.0370.034C(15)-0.1117(5)0.1954(4)0.0323(2)8.9(2)GoodnessofFit Indicator1.54C(16)-0.2208(5)0.3072(4)0.0282(2)7.0(2)MaxShift/ErrorinFinalCycle0.00030D.1.3X-rayCrystallographicAnalysisof{[(PPri2CH2SiMe2)2N]Zr(OAr*}Z(tN),2.12acontinuedTable11.AtomiccoordinatesandB,(continued)atomxyzC(17)-0.2603(5)0.1400(4)0.1246(3)7.7(2)C(18)-0.3433(4)0.2603(4)0.1430(3)6.4(2)C(19)0.0193(4)0.0845(3)0.1354(2)4.3(1)C(20)-0.0447(4)0.0387(3)0.1614(2)5.1(2)C(21)-0.0655(5)-0.0318(4)0.1412(3)7.1(2)C(22).0.0240(6)-0.0571(4)0.0964(3)7.8(2)C(23)0.0371(5)-0.0118(4)0.0705(2)6.2(2)C(24)0.0593(4)0.0599(3)0.0895(2)4.4(1)C(25)-0.0951(5)0.0653(4)0.2091(2)6.9(2)COG)0.1240(4)0.1096(3)0.0594(2)6.2(2)I.TableIV.RoadAugles()atomatomatomangleatomatomatomanglePU)Z(1)P(2)145.54(4)PU)Zr(l)0(1)87.6(1)PU)Z(1)NO)79.4(1)PU)Zt(1)14(2)128.50(10)PU)Zr(1)N(2)’85.10(10)P(2)Zr(1)0(1)88.1(1)P(2)Zr(S)N(S)74.5(1)P(2)Zr(S)14(2)82.89(10)P(2)Zr(S)14(2)’126.19(10)0(1)Zr(l)N(S)123.0(1)0(1)Zr(S)14(2)119.8(1)0(1)ZrU)N(S)’119.5(1)N(l)Zr(S)N(2)111.4(1)14(1)Zr(S)N(2)’114.7(1)14(2)Zr(S)14(2)’43.6(2)Zr(S)P(1)C(1)101.3(2)Zr(1)P(l)C(7)115.1(2)Zr(l)P(i)C(8)120.1(2)CU)PU)C(7)106.3(3)CU)PU)C(8)104.9(3)C(7)PU)C(8)107.6(3)Zr(1)PU)CU)93.3(2)ZrO)PU)C(t3)115.8(2)Zr(S)PU)C(14)124.4(2)C(S)PU)CU3)102.3(2)P(2)CU4)106.5(3)CU3)PU)C(14)109.9(3)SIU)CU)110.3(2)NU)Si(1)CU)115.4(2)SiU)C(4)112.3(2)CU)Si(l)CU)107.7(3)SU)C(4)105.4(2)CU)Si(l)C(4)105.1(3)Si(S)C(S)108.2(2)N(S)Si(2)C(5)112.4(2)Si(2)C(S)115.4(2)C(2)SI(S)C(S)107.7(2)Si(S)C(6)106.3(2)C(S)Si(S)C(S)106.4(3)0(1)C(19)161.5(4)Zr(l)N(S)Si(S)123.9(2)N(S)Si(S)116.4(2)SI(S)14(1)Si(S)119.7(2)14(2)Zr(l)’130.7(2)Zr(1)14(2)14(2)’69.9(2)Zr(S)’14(2)NO)’86.5(2)B,=,r2(Ui5(aa)2+(.122(06’)’+U,s(cc’)’+2U1,aa’bb’cony+2Uj5aa’cccoefl+2Umbbce’roan)Table111.BondLengths(A)atomatomdistanceatomatomdistanceZr(S)PU)2.8180)Zr(S)P0)2.8460)Zr(1)0(5)2.020(3)Zr(S)NO)2.211(3)ZrU)14(2)2.034(4)Zr(S)NO)’2.082(4)PU)C(S)1.824(5)PU)C(7)1.848(6)PU)C(8)1.8510)P0)C(S)1.813(5)P0)C(13)1.844(5)PU)CU4)1.852(6)Si(1)NO)1.718(4)Si(S)C(S)1.890(5)SI(S)CU)1.878(6)Si(S)C(4)1.870(6)Si(S)N(S)1.720(4)Si(2)C(S)1.887(5)Si(S)C(5)1.861(6)Si(S)C(6)1.881(5)00)COO)1.331(6)N(S)N(S)’1.528(7)CU7)C(S4)CU8)110.5(5)00)COO)C(24)119.8(5)C(S)14(1)N(S) CU)NO)NO)CU)Zr(S)Zr(1)Zr(S)PU)C(14)C(18)116.7(4)00)C(19)COO)120.1(5)D.1.4X-rayCrystallographicAnalysisof Zr(r)5-CSHS)Ph[N(SiMe2CH2PPr’2)2],4.4SpaceGroupandCellDimensionsTriclinic,P-la9.0970(10)b10.512(3)c19.079(5)alpha90.90(3)beta95.890(10)gamma107.390(20)Volume1729.8(7)A3Empiricalformula:ZrSi2PNC9H54Celldimensionswereobtainedfrom24reflectionswith2Thetaangleintherange28.00—36.00degrees.Crystaldimensions:0.30X0.40X0.50mmZ=2F(000)=Dcalc1.202Mg.m—3,mu0.49mm—i,lambda0.70930A,2Theta(max)44.9TheintensitydatawerecollectedonaNoniusdiffractometer,usingtheomegascanmode.Theh,k,1rangesare:———99,0ii,—2020No.ofreflectionsmeasured7929No.ofuniquereflections4512No.ofreflectionswithmet>3.Osigma(Inet)3361Absorptioncorrectionsweremade.Theminimumandmaximumtransmissionfactorsare0.831and1.00.Thelastleastsquarescyclewascalculatedwith89atoms,316parametersand3361outof4512reflections.Weightsbasedoncounting—statisticswereused.TheweightmodifierKinKF02is0.000050Theresidualsareasfollows:——Forsignificantreflections,RF0.036,Rw0.038G0F2.67Forallreflections,RF0.036,Rw0.038.whereRF=Sum(Fo—Fc)/Sum(Fo,Rw=Sqrt[Sum(w(Fo-Fc))/Sum(wFo2]andG0F=Sqrt[Sum(w(Fo—Fc)2)/(No.ofrefins—No.ofparams.)]Themaximumshift/sigmaratiowas0.008.InthelastD—map,thedeepestholewas—0.280e/A3,andthehighestpeak0.240e/A3.FW=626.08012010cliC907ClC8C3P1C22SilC5Zr02NC21C31C32S1204C6P2Cl3016C14018C17Cl5660t-)00D.1.4X-rayCrystallographicAnalysisof Zr(15-CH)Ph[N(S1Me2CHPPrI],4.4continued-Int.ratoidcDistanc.sandng1.eDiatanc.a(A)AtomicParametersx,y,zandB0xyzB0Zr—P(1)2.7819(14)Si(2)—C(5)1.846(6)Zr—P(2)2.7843(16)Si(2)—C(6)1.885(5)Zr—N2.231(3)C(7)—C(8)1.490(8)Zr—C(21)2.540(4)C(7)—C(9)1.498(8)Zr—C(22)2.549(5)C(10)—C(11)1.508(7)Zr—C(23)2.526(5)C(10)—C(12)1.527(8)Zr—C(24)2.491(4)C(13)—C(14)1.509(7)Zr—C(25)2.497(4)C(13)—C(15)1.350(8)•Zr—C(31)2.272(4)C(16)—C(17)1.537(7)P(1)—C(1)1.817(5)C(16)—C(18)1.512(7)P(1)—C(7)1.868(5)C(21)—C(22)1.378(7)P(1)—C(10)1.868(5)C(21)—C(25)1.390(7)P(2)—C(6)1.821(4)C(22)—C(23)1.373(8)P(2)—C(13)1.846(5)C(23)—C(24)1.397(9)P(2)—C(16)1.861(5)C(24)—C(25)1.420(7)Si(1)—N1.726(3)C(31)—C(32)1.405(6)Si(1)—C(1)1.883(5)C(31)—C(36)1.412(6)Si(1)—C(2)1.878(5)C(32)—C(33)1.378(6)Si(1)—C(3)1.878(5)C(33)—C(34)1.355(7)Si(2)—N1.725(3)C(34)—C(35)1.352(8)Sj(2)—C(4)1.871(5)C(35)—C(36)1.380(7).nqle5()Zr0.92623(5)0.33970(4)0.26186(3)3.716(20)P11.15726(14)0.35823(12)0.37245-(7)4.38(6)P20.78042(13)0.23576(11)0.12882(7)4.45(6)Sil1.12160(14)0.10076(11)0.28202(8)4.29(7)Si20.79039(15)0.00660(12)0.22164(8)4.78(7)N0.9521(3)0.1357(3)0.25364(18)3.75(18)Cl1.2555(5)0.2491(4)0.3364(3)4.8(3)C21.2305(5)0.0727(4)0.2077(3)5.6(3)C31.0972(6)—0.0516(5)0.3352(3)6.8(3)C40.8312(6)—0.1360(4)0.1759(3)7.2(3)C50.6687(6)—0.0675(5)0.2911(3)7.6(3)C60.6673(5)0.0731(4)0.1552(3)4.8(3)C71.3134(6)0.5149(5)0.4041(3)6.2(3)C81.4275(6)0.5681(5)0.3529(3)7.3(3)C91.2491(7)0.6189(6)0.4319(4)10.1(4)ClO1.0984(6)0.2909(5)0.4587(3)6.4(3)Cli0.9545(7)0.1719(5)0.4509(3)7.6(3)C121.2291(8)0.2657(8)0.5079(4)11.9(5)C130.6458(7)0.3020(5)0.0723(4)10.7(4)C140.7095(6)0.4480(5)0.0582(3)6.4(3)C150.4989(7)0.2321(6)0.0474(4)10.1(4)C160.9058(6)0.1930(5)0.0669(3)5.7(3)C170.9991(6)0.3133(5)0.0288(3)7.3(3)C180.8272(7)0.0788(5)0.0136(3)8.4(4)C210.8565(6)0.5348(4)0.3143(3)5.8(3)C220.8293(6)0.4390(5)0.3639(3)6.7(3)C230.7098(6)0.3301(5)0.3366(3)7.7(4)C240.6606(5)0.3568(5)0.2681(3)6.8(3)C250.7543(5)0.4861(5)0.2539(3)6.0(3)C311.1072(4)0.4931(4)0.20838(22)3.63(21)C321.2228(5)0.4462(4)0.18375(25)4.62(25)C331.3475(5)0.5248(5)0.1532(3)5.6(3)C341.3640(5)0.6560(5)0.1454(3)6.4(3)C351.2556(6)0.7081(4)0.1664(3)6.4(3)C361.1309(5)0.6299(4)0.1972(3)5.3(3)1.P(1)—Zr—P(2)150.09(4)C(4)—Sj(2)—C(6)106.63(23)P(1)—Zr—N79.47(9)C(5)—Si(2)—C(6)107.32(23)P(1)—Zr—C(31)87.04(11)Zr—N—Si(1)123.21(16)9(2)—Zr—N75.01(10)Zr—N—Si(2)117.54(16)P(2)—Zr—C(24)78.50(13)Si(1)—N—Si(2)119.15(17)P(2)—Zr—C(25)85.80(12)P(1)—C(1)—Si(1)113.90(22)P(2)—Zr—C(31)88.57(11)P(2)—C(6)—Si(2)109.65(22)N—Zr—C(24)117.57(14)P(1)—C(7)—C(8)114.5(4)N—Zr—C(25)149.07(14)P(1)—C(1)—C(9)112.0(4)N—Zr—C(31)112.60(13)C(8)—C(7)—C(9)112.6(4)Zr—P(1)—C(1)99.87(16)P(1)—C(10)—C(11)113.1(4)Zr—P(1)—C(7)124.83(17)P(1)—C(10)—C(12)114.3(4)Zr—P(1)—C(10)118.31(17)C(11)—C(10)—C(12)112.0(4)C(1)—P(1)—C(7)105.65(22)P(2)—C(13)—C(14)114.4(4)C(1)—P(1)—c(10)106.44(22)P(2)’—C(13)—C(15)125.0(4)C(7)—P(1)—C(10)100.10(23)C(14)—C(13)—C(15)120.5(5)Zr—P(2)—C(6)97.73(16)P(2)—C(16)—C(17)113.7(3)Zr—P(2)—C(13)125.97(24)P(2)—C(16)—C(18)116.1(4)Zr—P(2)—C(16)115.96(16)c(17)—C(16)—C(18)109.3(4)C(6)—P(2)—C(13)106.77(22)C(22)—C(21)—C(25)108.7(4)C(6)—P(2)—C(16)103.08(21)C(21)—C(22)—C(23)109.1(5)C(13)—P(2)—C(16)104.4(3)C(22)—C(23)—C(24)108.0(5)N—Si(1)—C(1)108.72(17)C(23)—C(24)—C(25)107.5(4)N—Si(1)—C(2)113.18(20)C(21)—C(25)—C(24)106.7(5)N—Si(1)—C(3)115.19(20)Zr—C(31)—C(32)114.9(3)C(1)—Sj(1)—C(2)105.86(21)Zr—C(31)—C(36)132.5(3)C(1)—Si(1)—C(3)108.66(23)C(32)—c(31)—C(36)112.5(4)C(2)—Si(1)—C(3)104.75(23)C(31)—C(32)—C(33)124.1(4)N—Si(2)—C(4)115.13(20)C(32)—C(33)—C(34)120.0(4)N—Sj(2)—C(5)112.86(22)C(33)—C(34)—C(35)119.5(4)N—Si(2)—C(6)108.72(17)C(34)—C(35)—C(36)120.7(4)C(4)—Si(2)—C(5)105.7(3)C(31)—C(36)—C(35)123.2(4)BistheMeanofthePrincipalaxesoftheThermalEllipsoidD.1.5X-rayCrystallographicAnalysisof Zr(rI5C5Hs)CH2SiMe3[N(SiMe2CH2PPri2)2],4.8SpaceGroupandCellDimensionsMonoclinic,P21/cC5a10.0813(19)b17.7883(18)C20.414(4)C6C4beta100.243(22)C15C17Volume3602.5(10)A3C13SMEmpiricalformulaZrSi3P2NCHC16Celldimensionswereobtainedfrom24reflectionswith2Thetaangleintherange37.00—44.00degrees.P2NC3Crystaldimensions:0.40X0.60X0.90mmC14FW=636.21Z=4F(000)=1364C25C18S12Di.l7Mg.m—3,mu0.50mm—i,X0.70930A,2Theta(max)45.0C21TheintensitydatawerecollectedonaNoniusCAD-4diffractometer,C2usingtheomegascanmode.Theh,k,irangesusedduringstructuresolutionandrefinementare:——C24(ZrC8Allnin,max-1010;Kmin,max019;Lmin,nax021-ThNo.ofreflectionsmeasured4831C7)No.ofuniquereflections4686C22No.ofreflectionswithI>3.0sigma(I)3647C23C31MergingR—valueonintensities0.026CiC88Absorptioncorrectionsweremade.P1Theminimumandmaximumtransmissionfactorsare0.887and0.999.Thelastleastsquarescyclewascalculatedwith88atoms,319parametersand3647outof4686reflections.S3jC34Weightsbasedoncounting—statisticswereused.TheweightmodifierKinKF02is0.000006TheresidualsareasfollowsForsignificantreflections,RF0.039,Rw0.043G0F4.37Forallreflections,RF0.039,Rw0.043.C32CliwhereRF=Sum(Fo-Fc)/Sum(Fo,C12RwSqrt[Sum(w(Fo-Fc))/Sum(wFo2))andG0F=Sqrt[Sum(w(Fo—Fc)2)/(No.ofrefins—No.ofparams.)]Themaximumshift/sigmaratiowas0.000.InthelastD-map,thedeepestholewas-0.330e/A3,andthehighestpeak0.400e/A3.CD.1.5X-rayCrystallographicAnalysisof Zr(i5-CH)CH2S1Me3[N(S1MeCHPPr’],4.8continuedInteratomicDistances(A)andAngles()Pr2PCH,SiMe)N)Cp(CH2SiMe3)ZrDistancesAtomicParametersx,y,zandBxyIZr—Pl2.7923(16)Si2—C21.864(7)Zr—P22.8563(17)Si2—C31.870(6)Zr—N2.216(4)Si3—C311.843(5)Zr—C2].2.505(6)Si3—C321.858(7)Zr—C222.561(6)Si3—C331.848(7)Zr—C232.549(5)Si3—C341.871(7)Zr—C242.475(5)C7—C8A1.396(22)Zr—C252.463(6)C7—C8B1.571(13)Zr—C312.337(5)C7—C8C1.353(23)P1—Cl1.821(5)C8A—C8C1.71(5)P1—C71.842(8)ClO—Cil1.483(10)P1—dO1.877(7)C10—C121.416(10)P2—C61.826(5)C13—C141.463(11)P2—C131.866(7)C13—C151.308(13)P2—C161.876(7)C16—C171.477(11)Sil—N1.726(4)C16—C181.497(11)Sil—C41.879(6)C21—C221.385(13)Sil—C51.858(6)C21—C251.398(12)Sil—C61.889(6)C22—C231.375(12)Si2—N1.713(4)C23—C241.385(11)Si2—C11.890(5)C24—C251.393(11)Zr0.10188(5)0.78205P1—0.05298(14)0.71414P20.30818(15)0.77622Sil0.14827(17)0.62939Si20.15093(18)0.59864Sj30.31782(17)0.89647N0.1386(4)0.66093Cl0.0631(5)0.6390C20.3286(7)0.5826C30.0751(8)0.5041C40.2493(6)0.5413CS—0.0214(6)0.6121C60.2265(6)0.7061C7—0.2113(9)0.6666C8A—0.2234(23)0.6417CBB—0.2560(10)0.5977C8C—0.324].(23)0.7018ClO—0.0813(8)0.7629Cli—0.1079(8)0.7163C12—0.1504(9)0.8325Ci)0.3682(9)0.8488C140.4163(11)0.9197C150.3390(13)0.8486C160.4630(6)0.7252Cl70.5365(8)0.6820C180.5526(7)0.7767C21—0.1028(7)0.8296C22—0.1120(7)0.8626C23—0.0059(7)0.9115C240.0705(6)0.9094C250.0108(8)0.8591C3l0.2534(5)0.8088C320.1950(7)0.9403C330.3617(8)0.9709C340.4712(7)0.8801z3)0.162399)0.051099)0.277989)0.264459)0.1211110)0.0602121)0.183763)0.03914)0.11063)0.13013)0.28653)0.28513)0.32277)0.059815)0.12347)0.013012)0.07104)—0.03184)—0.09305)—0.03585)0.34245)0.31896)0.40245)0.26315)0.32006)0.23274)0.20314)0.14103)0.14263)0.20614)0.24443)0.09124)—0.00774)0.12354)0.02233) 7) 8) 8) 9) 9) 20) 3) 3) 3) 3) 3) 3) 6) 9) 5)20) 3) 3) 4) 4) 5) 5) 4) 5) 4) 5) 4) 4) 4) 3) 3) 4) 4) 4)2.9874.024.484.274.454.873.353. 7) 7) 8) 9) 8)21) 3) 4) 5) 3) 3) 3) 6) 15)10)24) 4) 5) 5) 5) 7) 9) 4) 6) 6) 5) 4) 4) 4) 4) 3) 4) 5) 5)Zr—Cent2.2162(5)Angles3,istheMeanofthePrincipalAxesoftheThermalEllipsoidP1—Zr—P2151.17(5)Cl—Si2—C3107.0(3)P1—Zr—N77.83(11)C2—Si2—C3107.0(3)P1—Zr—C2l92.39(22)C31—Si3—C32112.5(3)P1—Zr—C24126.89(17)C31—Si3—C33114.9(3)Pl—Zr—C25124.85(19)C3l—Sj3—C34112.2(3)Pl—Zr—C3l85.13(13)C32—Si3—C33105.4(4)P2—Zr—N74.37(11)C32—Si3—C34104.3(3)P2—Zr—C2l105.06(24)C33—Si3—C34106.7(4)P2—Zr—C2481.75(17)Zr—N—Sil118.88(21)P2—Zr—C2576.33(17)Zr—N—Si2120.90(21)P2—Zr—C3193.47(13)Sil—N—Sj2120.18(23)N—Zr—C21112.41(20)P1—C1—Si2111.8(3)N—Zr—C24147.80(22)P2—C6—Sil110.7(3)N—Zr—C25118.45(20)P1—C7—C8A117.1(9)N—Zr—C3].102.47(16)Pl—C7—C8B117.3(6)C21—Zr—C2453.55(22)P1—C7—CBC124.9(11)C21—Zr—C2532.7(3)C8A—C7—C8B104.7(12)C21—Zr—C31143.69(23)C8A—C7—C8C77.0(23)C24—Zr—C2532.8(3)C8B—C7—C8C107.8(12)C24—Zr—C3l100.21(21)C7—C8A—C8C50.4(14)N—Zr—Cent139.47(10)C31—Zr—Cent118.05(13)P1—Zr—Cent103.95(4)P2—Zr—Cent102.08(4)D.1.6X-rayCrystallographicAnalysisof Zr(i5-C5H5)(1i2-BH4)[N(SiMe2CH2PPr’2)2],4.17SpaceGroupandCellDimensionsMonoclinic,I2/aa17.1426(19)b10.8752(14)C33.754(4)beta90.804(9)Volume6292.1(l3)A3Empiricalformula:ZrP2Si2NCH53Celldimensionswereobtainedfrom25reflectionswith2Thetaangleintherange30.00—40.00degrees.Crystaldimensions:0.20X0.30X0.80mmFW553.02Z8F(000)2366Dcalc1.168Mg.m—3,mu0.53mm—l,lambda0.70930A,2rheta(max)47.9TheintensitydatawerecollectedonaSiemensP3/PCdiffractometer,usingtheomegascanmode.Theh,k,lrangesare:———1919,012,038No.ofreflectionsmeasured5096No.ofuniquereflections4903No.ofreflectionswithmet>3.Osigma(Inet)3370C3AnempiricalabsorptioncorrectionwasmadeC21Thelastleastsquarescyclewascalculatedwith83atoms,287parametersand3370outof4903reflections.Weightsbasedoncounting—statisticswereused.TheweightmodifierKinKF02is0.000100Theresidualsareasfollows:——Forsignificantreflections,RF0.041,Rw0.045GoF2.00Forallreflections,RF0.041,Rw0.045.whereRFSum(Fo—Fc)/Sum(Fo,RwSgrt[Sum(w(Fo—Fc))/Sum(wFo2)]andG0F=Sqrt(Sum(w(Fo_Fc)Z)/(No.ofreflns—No.ofparams.))Themaximumshift/sigmaratiowas0.001.InthelastD—map,thedeepestholewas—0.460e/A3,andthehighestpeak0.470e/A3.Hydrogenatomsweretreatedasfollows:Locatedandrefinedwithisotropicthermalparameters:BH4Locatedandfixedwithisotropicthermalparameters:CHgroupsCalculated(DCK1.08A)andfixedwithisotropicthermalparameters:CH2,CHgroups.C17Cl5Cl3C2ClC22C1209C8Appendix 233B0 is the Mean of the Principal Axes of the Thermal EllipsoidD.1.6 X-ray Crystallographic Analysis ofZr(15-CH)(r2BH4[N(SiMeCPPr’],4.17continuedAtomic Parameters x,y,z and B10x y z B0Zr 0.61108( 3) 0.31118( 4 0.152689(14) 3.402(19)P1 O.55982( 9) 0.2l150(14 0.08063 ( 4) 4.70 ( 7)P2 0.63355( 8) 0.52057(l4 0.19895 ( 4) 4.24 ( 6)Sil 0.58831( 9) 0.48788(15 0.06837 ( 4) 4.44 ( 7)Si2 0.72058( 8) 0.55335(15 0.12248 ( 4) 4.61 ( 7)N 0.64155(21) 0.4612 ( 4) 0.11086 (11) 3.61 (17)Cl 0.5213 ( 3) 0.3529 ( 5) 0.05972 (15) 4.8 ( 3)C2 0.6489 ( 4) 0.5102 ( 7) 0.02310 (17) 7.5 ( 4)C3 0.5216 ( 3) 0.6245 ( 6) 0.07138 (19) 6.4 ( 3)C4 0.8147 ( 3) 0.4838 ( 7) 0.10542 (19) 6.9 ( 4)C5 0.7164 ( 4) 0.7129 C 6) 0.10101 (20) 7.6 ( 4)C6 0.7254 ( 3) 0.5741 ( 5) 0.17823 (15) 4.64 (25)C7 0.4839 ( 4) 0.0908 ( 7) 0.07280 (23) 8.3 ( 4)C8 0.4825 ( 5) —0.0037 ( 8) 0.10478 (25) 10.3 ( 5)C9 0.4064 ( 4) 0.1322 ( 8) 0.0618 ( 3) 11.3 ( 6)dO 0.6388 ( 4) 0.1675 ( 6) 0.04443 (17) 6.6 ( 3)Cli 0.6146 ( 5) 0.1785 ( 8) 0.00115 (22) 9.7 ( 5)C12 0.6717 ( 5) 0.0368 ( 8) 0.05227 (21) 9.4 ( 5)C13 0.6524 ( 5) 0.5156 ( 7) 0.25340 (18) 8.7 ( 4)C14 0.5872 ( 4) 0.4572 ( 7) 0.27576 (18) 7.6 ( 4)C15 0.6991 ( 4) 0.6005 ( 8) 0.27316 (18) 8.2 ( 4)C16 0.5645 ( 3) 0.6473 ( 5) 0.18681 (18) 5.6 ( 3)C17 0.5930 ( 4) 0.7775 ( 6) 0.19446 (23) 7.5 ( 4)C18 0.4843 ( 4) 0.6311 ( 6) 0.2042 ( 3) 9.1 ( 5)C21 0.7285 ( 4) 0.1789 ( 6) 0.14958 (19) 5.9 ( 3)C22 0.6684 ( 4) 0.0957 ( 6) 0.15739 (19) 5.9 ( 3)C23 0.6403 ( 4) 0.1179 ( 6) 0.19428 (20) 6.0 ( 3)C24 0.6799 ( 4) 0.2159 ( 6) 0.21001 (17) 6.2 ( 3)C25 0.7365 ( 3) 0.2544 ( 6) 0.18292 (23) 6.1 ( 3)B 0.4627 ( 4) 0.3106 ( 8) 0.16984 (23) 5.0 ( 3)HB1 0.491 ( 3) 0.382 ( 5) 0.1515 (14) 5.8 (13)HB2 0.508 ( 3) 0.252 ( 4) 0.1788 (13) 4.8 (12)HB3 0.434 ( 3) 0.354 ( 5) 0.1927 (15) 6.5 (15)HB4 0.420 ( 3) 0.246 ( 6) 0.1494 (17) 8.2 (15)AppendixD.1.6X-rayCrystallographicAnalysisof Zr(15-CH)(12BH4[N(S1MeCHPPr’],4.17continuedcC(9)—H(9A)1.126(8)H(12A)—N(12C)1.71165(19)C(9)—H(9B)1.100(10)H(14A)—H(14B)1.50581(15)DistancesC(9)—H(9C)1.099(9)N(14A)—H(14C)1.47993(13)C(10)—C(11)1.518(10)H(14B)—H(14C)1.46254(18)Zr—P(1)2.7933(15)C(12)—H(12B)1.117(9)c(lo)—c(12)1.551(10)H(15A)—H(15C)1.66048(17)Zr—P(2)2.7849(15)C(12)—H(12C)1.110(7)C(10)—U(10)1.011(7)H(15B).-H(15C)1.66109(14)Zr—N2.224(4)C(13)—C(14)1.499(9)C(10)—H(12B)1.782(7)H(17A)’-H(178)1.70590(17)Zr—C(21)2.477(5)C(13)—C(15)1.386(10)C(11)—H(11A)0.999(8)H(17A)—H(17c)1.58234(18)Zr—C(22)2.545(6)C(13)—H(13)1.042(9)c(11)—N(11B)1.158(8)li(17B)—H(17C)1.37468(14)Zr—C(23)2.573(6)C(14)—N(14A)0.986(6)C(11)—H(11C)1.100(9)H(18A)—H(1SB)1.70246(17)Zr—C(24).2.479(5)C(14)—H(14B)0.974(7)c(12)—H(12A)0.936(7)H(18B)—H(18C)1.64798(15)Zr—C(25)2.446(5)C(14)—H(14C)1.147(8)Zr—B2.617(7)C(15)H(13)1.848(8)AnglesZr—HB(1)2.19(5)C(15)—H(15A)1.032(7)Zr—HB(2)2.09(4)C(15)—H(15B)1.040(6)P(1)—Zr—P(2)147.66(5)H(9A)—c(9)—M(9c)114.5(7)P(1)C(1)1.813(6)C(15)H(15C)1.134(8)p(1)..zrN78.80(10)H(9B)—C(9)—1(9C)113.5(7)P(1)C(7)1.866(7)C(16)C(17)1.519(8)P(1)—Zr—C(21)89.06(17)P(1)—C(j0)—C(j1)114.3(5)P(1)C(10)1.898(6)C(16)—C(18)1.513(9)p(1)Zrc(24)131.14(15)P(1)—C(10)—C(12)112.4(5)P(2)—C(6)1.826(5)C(16)M(16)1.019(6)P(1)ZrC(25)121.96(18)P(1)—C(10)—H(10)106.0(4)P(2)—C(13)1.862(6)C(17)—H(17A)1.122(8)P(1)—Zr—HB(1)80.7(12)C(11)—C(l0)—C(12)109.3(5)p(2)C(16)1.859(6)C(17)H(178)0.970(7)P(1)—zr—HB(2)89.5(13)C(11)—C(10)—H(10)106.9(6)Si(1)N1.714(4)c(17)H(17c)0.755(7)P(2)—Zr—N74.00(10)c(12)—C(10)—H(10)107.5(6)Si(1)—C(1)1.884(6)C(18)—H(18A)1.119(8)P(2)—Zr—C(21)113.10(18)C(10)—C(11)—R(11A)105.2(6)Si(1)—C(2)1.876(6)C(18)—H(18B)0.865(7)P(2)—Zr—C(24)81.00(16)c(lo)—c(11)—H(11B)98.0(6)si(1)c(3)1.878(6)C(18)H(18C)1.184(8)P(2)—Zr—C(25)81.86(17)C(10)—C(11)—H(11C)109.2(7)Si(2)—N1.726(4)C(21)—C(22)1.399(10)p(2)jr.NB(1)81.0(13)fl(11A)—C(11)—H(11E)112.7(8)Si(2)C(4)1.879(6)C(21)C(25)1.399(10)P(2)—Zr—HB(2)97.2(13)N(11A)—c(11)—14(11c)115.8(8)Si(2)—C(5)1.882(7)C(21)—H(21)1.014(6)N—Zr—C(21)101.58(18)H(11B)—C(11)—H(11C)113.9(6)Si(2)C(6)1.896(5)C(22)C(23)1.363(10)N—Zr—C(24)133.50(22)C(10)—C(12)—H(12A)108.0(7)Si(2)H(4C)2.2679(15)c(22)H(22)1.036(6)N—Zr—C(25)103.71(20)C(10)—C(12)—H(12B)82.1(5)C(1)14(1A)1.009(5)C(23)C(24)1.367(10)N—Zr—14B(1)87.6(14)C(10)—C(12)—H(12C)105.5(6)C(1)—H(1B)1.021(5)C(23)—H(23)1.046(7)N—Zr—HB(2)134.5(14)H(12A)—C(12)—H(128)132.8(8)C(2)—14(2A)0.940(7)C(24)—C(25)1.406(11)c(21)—zr—c(24)54.00(21)H(12A)—C(12)—H(12C)113.3(8)C(2)—H(2B)1.017(7)C(24)—H(24)1.038(6)c(21)—zr—c(25)33.00(24)H(128)—C(12)—H(12C)107.5(6)C(2)—H(2C)1.066(6)C(25)—H(25)1.031(6)C(21)—Zr—HB(1).164.8(14)P(2)—C(13)—C(14)112.9(5)C(3)H(3A)1.010(6)BHB(1)1.12(5)c(21)—zr—BB(2)122.2(14)P(2)—C(13)—C(15)123.1(5)c(3)N(3B)1.136(7)BMB(2)1.04(5)C(24)—Zr—C(25)33.2(3)P(2)—C(13)—H(13)99.0(4)C(3)—H(3C)0.986(6)B—HB(3)1.04(5)C(24)—Zr—HB(1)126.6(13)c(14)—c(13)—c(15)118.0(5)C(4)—H(4A)0.969(7)B—HB(4)1.22(6)C(24)—zr—HB(2)86.5(14)c(14)—c(13)—H(13)97.3(6)C(4)H(48)1.076(6)HB(1)—HB(2)1.71(7)C(25)—Zr—HB(1)156.0(12)C(15)—C(13)—H(13)98.1(7)C(4)B(4C)0.969(6)H(1A)R(1B)1.63295(18)C(25)—Zr—HB(2)119.4(14)C(13)—C(14)—H(14A)129.4(6)c(5)H(5A)0.879(6)H(2A)H(25)1.36590(17)HB(1)—Zr—HB(2)47.0(20)C(13)—C(14)—H(14B)129.8(6)C(5)—N(5B)1.028(7)H(2A)—H(2C)1.62532(15)Zr—P(1)—C(1)96.79(17)C(13)—C(14)—H(14c)89.2(5)C(5)—fl(SC)1.140(7)H(2B)—H(2C)1.66291(16)Zr—P(1)—C(7)127.37(25)H(14A)—C(14)—H(14B)100.4(5)C(6)H(6A)1.007(5)H(3A)N(3B)1.64914(19)Zr—P(1)—C(10)116.08(20)B(14A)—C(14)—H(14C)87.5(5)C(6)—H(6B)1.023(5)N(3A)—H(3C)1.59300(14)C(1)—P(1)—c(7)107.0(3)H(14B)—C(14)—H(14C)86.8(6)C(7)—C(8)1.491(12)14(4A)—N(4B)1.55129(18)C(1)—P(1)—C(10)102.8(3)C(13)—C(15)—H(15A)114.8(6)C(7)—C(9)1.446(11)H(4B)—H(4C)1.56897(17)C(7)—P(1)—C(10)103.6(3)C(13)—C(15)—H(158)110.9(6)C(7)H(7)1.017(8)H(5A)H(5B)1.47341(15)Zr—P(2)—C(6)99.15(17)C(13)—C(15)—H(15C)97.4(6)C(7)H(9B)1.853(8)H(6A)H(6B)1.63295(18)zr-P(2)—c(13)123.4(3)N(15A)—C(15)—N(15B)126.9(6)C(8)H(8A)1.001(7)H(8A)H(8B)1.46359(12)Zr—P(2)—C(16)113.58(19)H(15A)—C(15)—fl(15C)99.9(6)C(8)—H(8B)1.059(8)H(8A)—H(SC)1.63738(16)C(6)—P(2)—C(13)104.4(3)H(158)—C(15)—H(15C)99.5(6)C(8)—H(BC)1.161(10)H(8B)—H(8C)1.68244(18)C(6)—P(2)—C(16)103.2(3)P(2)—C(16)—C(17)116.8(4)C(13)—P(2)—C(16)109.9(3)P(2)—C(16)—C(18)114.1(4)-) 4.1.D.1.7X-rayCrystallographicAnalysisof Zr(15-CH)(12BH4(r-CHO){[N(SiMe2CH2PPr’2)-[S1Me2CH(Pr1P•BH3)J),4.23SpaceGroupandCellDimensionsMonoclinic,P21/ca13.303(3)b9.5813(7)c27.518(7)beta103.42(5)Volume3411.8(12)AEmpiricalformulaZr2Si20NC,B2HCelldimensionswereobtainedfrom24reflectionswith20angleintherange36.00—40.00degrees.Crystaldimensions:0.40X0.50X0.60mmFW=606.67Z=4F(000)1296D1.l8Mg.m-3,p0.4lmni—l,.0.70930A,20(max)45.0’TheintensitydatawerecollectedonaNoniusCAD—4diffractometer,usingtheomegascanmode.Theh,k,lrangesusedduringstructuresolutionandrefinementareHmin,max—1413;Kmin,max010;Lmin,max029No.ofreflectionsmeasured4680No.ofuniquereflections4447No.ofreflectionswithI,>3.Osigma(I,,,)3626MergingR—valueonintensities0.018Absorptioncorrectionsweremade.Theminimumandmaximumtransmissionfactorsare0.843and0.992.Thelastleastsquarescyclewascalculatedwith90atoms,314parametersand3626outof4447reflections.Weightsbasedoncounting—statisticswereused.TheweightmodifierKinKFo2is0.000004Theresidualsareasfollows:——Forsignificantreflections,RF0.038,Rw0.042G0F4.53Forallreflections,RF0.038,Rw0.042.whereRF=Sum(Fo_Fc)/Sum(Fo)Rw=Sqrt[Sum(w(Fo—Fc))/Sum(wFo2]andGoF=SqrtSum(w(Fo—Fc)2)/(No.ofreflns—No.ofparams.)Themaximumshift/sigmaratiowas0.005.Inthelast0—map,thedeepestholewas—0.430e/A’,andthehighestpeak0.420e/A3.Afteranisotropicrefinementofallnon—hydrogenatoms,methyl,formyl,andboranehydrogenatomswerelocatedandfixedviainspectionofadifferenceFouriermap,temperaturefactorsbeingbasedupontheatomtowhichtheyarebonded.Boratehydrogenatomswerelocatedandrefined.Methyleneandsp2hydrogenatomswerefixedincalculatedpositions(d41=l.O8)withtemperaturefactorsbeingbaseduponthecarbontowhichtheyarebonded.AllcrystallographiccalculationswereconductedwiththePCversionoftheNRCVAXprogrampackagelocallyimplementedonanIBMcompatible80486computer.C15C18019010 C8.C3CP4 CP3•1 UiC4BCl:1C7HB1DBiHB1CC6X-rayCrystallographicAnalysisof Zr(T5-C5H5)(12BH4)(r1CHO){[N(SiMeCHPPri)-[SiMe2CH(PrIP.BH3)J},4.23continuedInteratomicDistance.(A)andAngl.s()AtomicParametersx,y,zandBDistancesxyzZr—01.987(3)Sil—ClO1.878(5)Zr0.59612(4)0.85878(5)0.138626(16)3.334(20)Zr—N2.135(3)Si2—N1.733(4)P10.68673(9)1.11441(11)0.06067(4)2.98(5)Zr—Cpl2.534(6)sil—c111.871(5)P20.86911(10)0.92905(15)0.35646(5)3.90(6)Zr—Cp22.527(5)Si2—C121.868(5)Sil0.69846(10)1.16080(13)0.17736(4)3.33(6)Zr—Cp32.555(6)Si2—C131.906(5)Si20.80563(11)0.91100(14)0.23448(5)3.66(6)Zr—Cp42.573(6)0—Cl1.429(5)Zr—CpS2.560(6)Cpl—Cp21.325(10)00.52305(22)0.9919(3)0.08702(11)3.94(15)Zr—Cl2.254(4)Cpl—Cp51.330(11)N0.7087(3)0.9827C3)0.18801(12)3.19(18)Zr—B12.594(6)Cp2—cp31.355(11)CP10.4318(5)0.8835(6)0.1695(3)8.6(5)P1-Cl1.765(4)Cp3—Cp41.332(13)CP20.4185(4)0.7701(8)0.14117(22)7.9(4)P1C21.832(4)Cp4—Cp51.297(13)CP30.4837(7)0.6694(6)0.1645(3)12.0(6)P1—C51.818(5)C2—C31.520(7)P1—C81.795(4)C2—C41.543(7)CP40.5338(5)0.7264(9)0.2075(3)12.2(6)P2—C131.828(5)C5—C61.535(7)CP50.5037(6)0.8542(9)0.21052(23)11.3(5)P2—C141.828(7)C5—C71.525(7)Cl0.6127(3)0.9665(4)0.06808(15)321(21)P2—Cl?1.827(5)C14—C151.504(10)C20.6108(4)1.2365(4)0.01540(15)3.49(22)P2—821.927(7)C14—C161.283(11)C30.5224(4)1.2977(5)0.03501(18)4.43(24)SL1N1.731(4)C17—C181.546(8)S11—C81.896(4)C17—C191.487(9)C40.6778(4)1.3520(5)—0.00019(19)5.1(3)Sil—C91.868(5)C50.7936(4)1.0547(5)0.03513(18)4.07(24)C60.7575(5)0.9896(7)—0.01698(20)6.7(3)Zr—Cent2.2855(9)Zr—HB1B2.31(4)C70.8637(4)0.9551(6)0.07098(22)6.0(3)Zr—HB1A2.09(4)Ca0.7387(3)1.2072(4)0.11764(16)3.32(21)C90.5652(4)1.2287(5)0.17330(18)4.73)anglesClO0.7844(4)1.2739(5)0.22539(18)5.6(3)Cli0.8093(4)0.7158(5)0.23357(20)5.9(3)0—Zr—N104.63(13)Zr—0—C180.78(21)C120.9364(4)0.9669(6)0.22802(19)5.4(3)0—Zr—Cl38.73(14)Zr—N—Sil115.11(17)C130.7767(4)0.9597(5)0.29700(16)4.39(25)N—Zr—Cl95.70(14)Zr—N—Sj2122.76(18)Cl—P1—C2110.90(21)Sii—N—Si2122.11(20)C140.9470(5)1.0882(7)0.3665(3)9.6(5)C1P1C5107.41(21)Cp2—Cp1—Cp5107.3(6)C150.8917(6)1.2176(7)0.3771C3)10.0(5)C1P1C8114.15(20)Cpl—Cp2—Cp3108.8(6)C161.0457(7)1.0849(11)0.3820(5)19.5(9)c2—P1—C5107.18(21)Cp2—Cp3—C94105.2(6)C170.7898(4)0.9284(7)0.40251(19)6.3(3)C2—P1C8108.51(20)Cp3—Cp4—Cp5110.2(6)C180.8558(6)0.9172(8)0.45659(21)9•3(5)C5—P1—C8108.43(22)Cpi—CpS—C94108.5(7)C13—P2—C14103.9(3)Zr—C1—P1129.48(22)C190.7094(5)0.8175(8)0.3927(3)10.05)C13—P2—C17104.17(23)Zr—Ci—060.49(19)B10.6696(5)0.6441(6)0.10097(24)5.5(4)C13—P2—82117.7(3)P1—C1—0116.1(3)820.9545(6)0.7647(8)0.3621(3)7.0(4)C14—P2—C17107.2(3)P1—C2—C3110.8(3)C14—P2—B2111.5(3)P1—C2—C4112.6(3)BistheMeanofthePrincipalAxesoftheThermalEllipsoidC17-P2B2111.4(3)C3-C2-C4111.5(4)N—Sjl—C8110.55(18)P1—C5—C6112.7(3)N—Sj1—C9112.93(20)Pl—C5—C7111.3(3)N—Sil—C10116.00(20)C6—C5—C7111.2(4)C8—Sil—C9108.77(21)P1—C8—Si1120.78(23)C8—Sil—C10103.08(22)P2—C13—Si2122.3(3)SydrogenParametersx,y,aandC9—Sil—C10104.81(24)P2—C14—C15115.5(5)N—Si2—C11113.80(21)P2—C14—C16122.0(7)xN—Si2—C12111.47(21)C15—C14—C16117.7(7)N—Si2—C13107.30(19)P2—C17—C18112.2(4)HB1A0.718(3)0.731(4)0.1292(13)3.9(9>C11—Si2—C12104.9(3)P2—C17—C19112.5(5)RB1B0.577(3)0.671(5)0.0846(17)7.0(13)C11—Sj2—C13105.55(24)C18—C17—Cl9110.4(5)881C0.703(4>0.646(5)0.0734(19)9.2(16>C12—Si2—C13113.78(23)HB1D0.678(3)0.540(4)0.1207(14)5.3(11)H82A1.0060.7490.4017.90—Zr—Cent105.68(9)11818—Zr—Cent94.1(11)1B2S0.9030.6660.35479N—Zr—Cent110.30(9>Zr—Cl—Ni115.2(3)NB2C1.0010.7700.3327.9Cl—Zr—Cent141.84(12)P1—Cl—Hi110.0(3)H81h—Zr—Cent121.6(10)0—Cl—Ni116.6(4)C’


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



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


Related Items