UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Hydride, alkyl, and alkenyl complexes of tungsten Debad, Jeff D. 1994

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

Item Metadata


831-ubc_1994-953357.pdf [ 3.59MB ]
JSON: 831-1.0061655.json
JSON-LD: 831-1.0061655-ld.json
RDF/XML (Pretty): 831-1.0061655-rdf.xml
RDF/JSON: 831-1.0061655-rdf.json
Turtle: 831-1.0061655-turtle.txt
N-Triples: 831-1.0061655-rdf-ntriples.txt
Original Record: 831-1.0061655-source.json
Full Text

Full Text

hYDRIDE, ALKYL, AND ALKENYL COMPLEXES OF TUNGSTENbyJEFF D. DEBAI)B.Sc., Simon Fraser University, 1989A THESIS SUBMITTED 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 COLUMBIAJuly 1994© JeffD. Debad, 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__________________________pDepartment of YThe University of British ColumbiaVancouver, CanadaDate ilDE-6 (2188)11AbstractHydrogenation of Cp*W(NO)(CH2SiMe3)affords the 16-electron alkyl hydride complex,Cp*W(NO)(CH2SiMe3) ,as a reactive intermediate. The tungsten-hydride bond of this complexinserts unsaturated reagents such as nitriles, ketones and aldehydes, imines, and acetylenes toform stable products containing azomethine, alkoxide, amide and alkenyl ligands respectively.Hydrogenation of Cp*W(NO)(CH2SiMe3)in the presence of dienes results in the formation ofCp*W(N0)(14hansdiene) species. Hydrogenation in the presence ofPPh3 affordsCp*W(NO)(H)(rI2PPh(C6)).Cp*W(NO)(R)Cl complexes are produced by the reaction of dialkyl- or diarylmagnesiumreagents with Cp*W(NO)(Cl)2.Reaction of Cp*W(NO)(CH2SiMe3)Clwith(p-tolyl)2Mg.Xdioxane results in the formation of the mixed alkyl aryl complex,Cp*W(NO)(CHSiMe3)(ptolyl). Cp*W(NO)(CHMe)(NHtBu) is formed by the reaction ofexcess t-butylamine with the same alkyl chloride compound. The Lewis-base adductsCp*W(NO)(CH2Me3)(L)Cl[L = PMe3,pyridine] are formed upon exposure ofCp*W(NO)(CHSiMe)Clto PMe3 or pyridine. Reaction of the PMe3 adduct with LDA affordsthe tucked-in Cp* complex, (i-i5,r ‘-Ce4CH2)W(NO)(CHMe(PM.The tungsten-alkylbond in Cp*W(NO)(CH2Me3)Clinserts CO and isonitriles to formr2-acyl andr2-iminoacy1compounds. Halide abstraction from the alkyl chloride using AgBF4 in acetonitrile affords theorganometallic cation [Cp*W(NO)(CH2Me3)(NCM )]+BF4_.The bimetallic species[Cp*W(NO)(CH2Me3)]l1N[Cp*W(O)(CHCMe)}is formed upon reduction of the alkylchloride with zinc, but if the same reaction is performed in the presence ofPhSSPh,Cp*W(NO)(CH2Me3)SPhis isolated.A number of mixed alkyl and aryl complexes Cp*W(NO)(R)R1 [R R’; R, R = CH2Me3,CH2SiMe3,CH2MePh, Ph, o-tolyl, Me] have been prepared from Cp*W(NO)(R)Cl species anddialkyl- or diarylmagnesium reagents. These mixed-ligand compounds insert CO into one of theirtungsten-carbon a-bonds. By determining which ligands insert preferentially, the relativemigratory aptitudes of the ligands can be ranked: CH2Me3> CH2SiMe3> o-tolyl > Ph > Me.111Thermolysis ofCp*W(NO)(CH2Me3)Phin benzene orp-xylene in the presence ofPMe3 causesthe formation of Cp*W(NO)(Ph)PMe and Cp*W(NO)(Ph)(2,5-MeC6H4)P3respectively,via a proposed benzyne intermediate.The proposed mechanism of formation ofCp*W(NO)(H)(rI2PPh(C6H4) via hydrogenationof Cp*W(NO)(CH2SiMe3)in the presence ofPPh3 involves the formation of the intermediateCp*W(NO)(CHSiMe)(H)PPh.This species undergoes RH elimination and then metallation ofthe phosphine ligand to afford the observed product. Addition ofPPh3 to a solution ofthemetallated complex induces the formation of Cp*W(NO)(PPh3)2Thermolysis ofCp*W(NO)(H)(rl2PPh(C6H4) in acetone or acetonitrile affords Cp*W(NO)(PPh3)(rl2Lspecies [L = Me2CO, NCMeJ. A kinetic investigation of the acetone addition reaction indicatesthat the reaction is second order overall, first order in each reactant. Thermolysis ofCp*W(NO)(H)(r2PPh(C6H4) in benzene results in solvent activation to formCp*W(NO)(PPh3)(Ph)H. A kinetic isotope effect of 1.9 ± 0.5 has been measured for thisactivation process.Reaction ofCp*W(NO)(CH2SiMe3)C1with (PhC=CH2)MgXdioxane produces the alkylalkenyl species, Cp*W(NO)(CHSiMe)(CPh=C in good yield. Upon thermolysis, thiscomplex loses TMS and activates C-H bonds. Cp*W(NO)(CPhCH2)Phis formed when thethermolysis is performed in benzene. Hexanes, pentane, and Et20 are also activated to form thecompounds Cp*W(NO)(rj4H2HRCHPh)[R = butyl, propyl, OEt respectivelyj. Thesespecies contain a metallacyclic ligand that also exhibits anr3-ben.zy1 interaction between theligand and the metal center. An unsaturatedi2-alkyne complex is proposed as an intermediate inthese thermal reactions. Thermolysis ofCp*W(NO)(CHSIMe3)(CPhCH in the presence ofPMe3 results in the formation of the metallacyclopropane compound,Cp*W(NO)(CH2SiMe)(rI{PMe3}PhCH2).ivTABLE OF CONTENTSAbstract jjTable of Contents ivList of Figures xiList of Tables xiiiList of Schemes xvList of Abbreviations xviAcknowledgements xixCHAPTER 1: General Introduction and Thesis Outline 11.1 Introduction and Background 11.2 Outline of This Thesis 31.3 References and Notes 5CHAPTER 2: Generation and Reactivity of Cp*W(NO)(CH2SiMe3)H:a 16-Electron AlkylHydride Complex 72.1 Introduction 72.2 Experimental Procedures 102.2.1 Methods 102.2.2 Reagents 112.2.3 Synthesis Preparation ofCp*W(NO)(CH2SiMe3)(N=CHMe) (2.1) Preparation ofCp*W(NO)(CHS1Me)(OCHMe (2.2) Preparation ofCp*W(NO)(CHSiMe)(OCHPh)(2.3) Preparation ofCp*W(NO)(CH2Si e3)(OCHMeCH=CHPh) (2.4) Preparation ofCp*W(NO)(CHSiMe)(NHCHPh (2.5) Preparation of Cp*W(NO)(r4diene) Complexes (2.6 and 2.7) 13V2.2.3.7 Preparation ofCp*W(NO)(CH2SIMe3)(CPh=C (2.8) Preparation of Cp*W(NO)(H)(r1PPh6H4(2.9) 142.2.4 NMRMonitoring 152.2.5 Characterization Data for Complexes 2.1 - 2.9 152.3 Results and Discussion 212.3.1 Reaction with Acetonitrile 212.3.2 Reactivity with C=O and C=N Bonds 262.3.3 Reaction with Dienes 312.3.4 Reaction with Phenylacetylene 332.3.5 Reaction with PPh3 422.4 Conclusions 432.5 References and Notes 44CUAPTER 3: Synthesis and Chemistry of Cp*W(NO)(R)C1 Complexes 473.1 Introduction 473.2 Experimental Procedures 483.2.1 Methods 483.2.2 Reagents 483.2.3 Synthesis 493.2.3.1 Preparation of Cp*W(NO)(CH2SiMe3)Cl(3.1) fromCp*W(NO)(CHSiMe)(NCBM ) (2.1) 493.2.3.2 Preparation of Cp*W(NO)(R)CI Complexes 493.2.3.3 Preparation ofCp*W(NO)(CH2Me3)(pto1yl) (3.5) 503.2.3.4 Preparation of Cp*W(NO)(CH2SiMe3)NHCMe (3.6) 503.2.3.5 Preparation ofCp*W(NO)(CHMe3)(PMl(3.7) 513.2.3.6 Preparation ofCp*W(NO)(CH2Me)(py)Cl (3.8) 51vi3.2.3.7 Reaction ofCp*W(NO)(CH2Me3)(PMCl(3.7) with LDA 513.2.3.8 Preparation of Cp*W(NO)(rI { 0 }CH2Me3)Cl(3.10) 523.2.3.9 Preparation of Cp*W(NO)(rl2{NCMe3}CH2Me3)Cl(3.11) 523.2.3.10 Preparation of Cp*W(NO)(1-C { 0 }CHMe)(PMCl(3.12) 533.2.3.11 Preparation of [Cp*W(NO)(CHMe3)(NCM )]BF4(3.13) 533.2.3.12 Preparation of [Cp*W(NO)CH2Mep.N[Cp*W(O)CH2CMe3](3.14) .... 533.2.3.13 Preparation of[Cp*W(NO)CHC e]p.N[Cp*W( ) l] (3.15) 543.2.3.14 Preparation of Cp*W(NO)(CH2Me3)SPh(3.16) 543.2.4 Characterization Data for Complexes 3.1 - 3.16 553.3 Results and Discussion 603.3.1 Initial Synthesis of Cp*W(NO)(CH2SiMe3)Cl(3.1) 603.3.2 Synthesis of Cp*W(NO)(R)Cl Complexes from Cp*W(NO)(Cl)2 613.3.3 Metathesis Reactions of Cp*W(NO)(R)Cl 653.3.4 Reactivity of Cp*W(NO)(CH2Me3)Clwith Lewis Bases 663.3.4.1 Reaction ofCp*W(NO)(CHMe)(PMlwith LDA 683.3.5 Reaction of Cp*W(NO)(CH2Me3)Clwith Unsaturated Lewis Bases 703.3.5.1 Reaction ofCp*W(NO)(1C{O}CHMe) lwith PMe3 733.3.6 Halide Abstraction from Cp*W(NO)(CH2Me3)Cl 743.3.7 Reduction Chemistry of Cp*W(NO)(CHMe)Cl 753.3.7.1 Electrochemistry of Cp*W(NO)(CH2Me3)Cl 753.3.7.2 Chemical Reduction of Cp*W(NO)(CHMe)Cl 773.4 Summary and Future Work 853.5 References and Notes 87CHAPTER 4: Synthesis and Reactivity of the Mixed Alkyl and Aryl ComplexesCp*W(NO)(R)R1 90vii4.1 Introduction 904.2 Experimental Procedures 934.2.1 Methods 934.2.2 Reagents 934.2.3 Synthesis 944.2.3.1 Preparation of Cp*W(NO)(R)Rt Complexes 944.2.3.2 Preparation of Cp*W(NO)(Ph)Me (4.10) 954.2.3.3 Thermolysis Reactions of Complex 4.7 954.2.3.4 NMR Tube Carbonylations of Complexes 4.1 - 4.9 964.2.3.5 Preparation of Cp*W(NO)(1l2{O}R)R? Complexes 4.4’ - 4.9’ 964.2.3.6 Preparation of Complexes 4.1’ - 4.3’ 974.2.3.7 Preparation of Cp*W(NO)(112{ 0 }Ph)Me (4.10’) 974.2.4 Characterization Data for Complexes 4.1- 4.14 and 4.1’ - 4.10’ 994.3 Results and Discussion 1084.3.1 Synthesis of the Mixed Complexes Cp*W(NO)(R)RI 1084.3.2 Characterization of Complexes 4.1- 4.12 1094.3.3 Structural Determination ofCp*W(NO)(CH2Me3)(otoly1) (4.6) 1104.3.4 Thermal Chemistry of Cp*W(NO)(CHMe)Ph(4.7) 1124.3.5 Carbonylation Reactions 1154.3.6 Spectroscopic Characterization of Complexes 4.1’ - 4.3’ 1174.3.7 Spectroscopic Characterization of Complexes 4.4’ - 4.10’ 1184.3.7.1 Crystallographic Analysis ofCp*W(NO)(fl2{O}CHMe3)Ph 1214.3.8 Migratory Aptitudes of the o-Bound Ligands 1244.4 Summary 1264.5 References and Notes 128viiiCRAPTER 5: Mechanism ofFormation and Reactivity of Cp*W(NO)(H)(q2PPh6H4)...1315.1 Introduction 1315.2 Experimental Procedures 1325.2.1 Methods 1325.2.2 Reagents 1325.2.3 Synthesis 1335.2.3.1 Preparation of Cp*W(NO)(D)(rI2P{ CD5}2C6D4)(2.9-d15) 1335.2.3.2 Preparation of Cp*W(NO)PPh3)(5.1) 1335.2.3.3 Preparation of Cp*W(NO)(PPh)NCMe (5.2) 1335.2.3.4 Preparation ofCp*W(NO)(PPh3)(112OCMe(5.3) 1345.2.3.5 Reactions of 2.9 with A.lkenes and Alkynes 1345.2.3.6 Preparation of Cp*W(NO)(H)(Ph)PPh3(5.4) 1355.2.3.7 Preparation of Cp*W(NO)(D)(C6D5)PPh(5.4-d6) 1355.2.3.8 Preparation of Cp*W(NO)(Cl)(q2PPhH4(5.5) 1355.2.3.9 Reactions of Cp*W(NO)(Cl)(rIPPh6)with Hydride Sources 1355.2.4 Kinetic Monitoring of the Formation of Cp*W(NO)(PPh3)(12OCMe 1365.2.5 Kinetic Comparison ofC6D Versus C6H Activation by Complex 2.9 1375.2.6 Characterization Data for Complexes 2.9-d15, 5.1 - 5.5 1385.3 Results and Discussion 1415.3.1 Mechanism of Formation ofCp*W(NO)(H)(rI2PPh6H4)(2.9) 1415.3.2 Reactions ofCp*W(NO)(H)(rl2PPh6H4)(2.9) with Lewis Bases 1445.3.2.1 Kinetic Study ofAcetone Addition to Cp*W(NO)(H)(rIPPh64 1475.3.3 Reaction ofCp*W(NO)(H)(rI2PPh6H4)(2.9) with Benzene 1515.3.3.1 Kinetic Study of Benzene Addition to Cp*W(NO)(H)(r1PPh64 1525.3.4 Reaction ofCp*W(NO)(H)(rl2PPh6H4)with Chloroform 154ix5.4 Summary and Future Work .1555.5 References and Notes 157CHAPTER 6: Reactivity of the Alkenyl Complex, Cp*W(NO)(CH2SiMe3)(CPhCH 1586.1 Introduction 1586.2 Experimental Procedures 1606.2.1 Methods 1606.2.2 Reagents 1606.2.3 Improved Synthesis ofCp*W(NO)(CH2S1Me3)(CPh=C (2.8) 1606.2.4 Preparation of Cp*W(NO)(Ph)(CPh=CH (6.1) 1616.2.5 Thermolysis ofCp*W(NO)(CHSiMe)(CPh=C in C6D 1616.2.6 Measurements of the Rates ofC6D and C6H Activation by Complex 2.8 1616.2.7 Thermolysis Reactions of Complex 2.8 1626.2.8 Preparation of the Metallacyclic Complexes 6.2 - 6.4 1626.2.9 Synthesis of Cp*W(NO)(CH2SiMe3)(fl{PMe3}PhCH2)(6.5) 1636.2.10 Characterization Data for Complexes 6.1 - 6.5 1646.3 Results and Discussion 1676.3.1 Improved Synthesis ofCp*W(NO)(CH2SiMe3)(CPh=C (2.8) 1676.3.2 Thermal Chemistry of Complex 2.8 1686.3.2.1 Activation of Aromatic Solvents 1686.3.2.2 Activation of Aliphatic Solvents by Complex 2.8 1726.3.2.3 Reaction of Complex 2.8 with PMe3 1796.3.3 Insights into the Synthesis of Complex 2.8 via Hydrogenation 1816.4 Summary 1846.5 References and Notes 186xAPPENDIX: Kinetic Data and Calculations for Chapters 4 and 5 188A. 1 Kinetic Data and Calculations of Rates and Activation Parameters for the ReactionBetween Acetone and Complex 2.9 188A.2 Kinetic Data and Calculation of the Kinetic Isotope Effect for the ReactionBetween Benzene and Complex 2.9 194A.3 Kinetic Data for the Reaction BetweenC6H/CDand Complex 2.8 196xiList of FiguresFigure 2.1. 1H NMR ofCp*W(NO)(CH2SiMe3)(N=CHMe)(2.1). 22Figure 2.2. View of the solid-state molecular structure ofCp*W(NO)(CHSiMe3)(N=CNMe) (2.1) 24Figure 2.3. View of the solid state molecular structure ofCp*W(NO)(CHSiMe)(CPh=CH (2.8) 38Figure 2.4. Different bonding descriptions ofr2-alkenyl ligands 39Figure 2.5. ‘3C{’H} N1vIR spectrum of complex 2.8 40Figure 2.6. 1H NMR spectrum of complex 2.8 41Figure 3.1. View of the solid-state molecular structure ofCp*W(NO)(CH2SiMe)C1(3.1) 63Figure 3.2. View of the solid-state molecular structure of(r15,1‘-CMe4CH)W(NO)(CHMeP (3.9) 72Figure 3.3. Cyclic voltammogram of Cp*W(NO)(CH2Me3)Cl 77Figure 3.4. View of the solid-state molecular structure of[Cp*W(NO)CH2Me]tN[Cp*W(O)Cl] (3.15) 83Figure 3.5. Selected examples of known, bridging nitrido complexes 84Figure 4.1. View of the solid-state molecular structure ofCp*W(NO)(CH2SiMe)(otolyl) (4.6) 111Figure 4.2. ‘3C{1H} NMR spectrum ofCp*W(NO)(rj{O}CHMe) (4.5’) 120Figure 4.3. ‘3C{’H} NMR spectrum ofCp*W(NO)(rl2{ 0 }CH2Me3)CSi (4.4’) 120Figure 4.4. H,’3C{’H} COSY spectrum of{ 0 } o-tolyl)Ph (4.9’) 122Figure 4.5. View of the solid-state molecular structure ofCp*W(NO)(i2{O}CHMe3)Ph(4.7’) 123xliFigure 5.1. Sample plot of the kinetic data obtained during the reaction ofcompound 2.9 with acetone at 55 °C in the presence of a ten-fold excess ofacetone 148Figure 5.2. Plot ofk0bs versus acetone concentration for the addition of acetoneto complex 2.9 149Figure 5.3. Plot of ln(kobs/T) versus liT for the addition of acetone tocomplex 2.9 150Figure 5.4. Plot of the kinetic data obtained during the monitoring of thereactions between complex 2.9 and C6H and C6D 153Figure 6.1. View of the solid-state molecular structure ofCp*W(NO)(r4NPhCH2Hh1PrC)(6.2) 174Figure 6.2. Selected examples of metallacyclopropane compounds formed viaphosphine attack on alkenyl ligands 180Figure 6.3. Resonance forms contributing to the bonding in complex 6.5 181Figure 6.4. View of the solid-state molecular structure ofCp*W(NO)(CH2SiMe)(rI{PMe}PhCH2)(6.5) 182Figure A.1. Plot of ln(kobs/T) versus lIT for the reaction of complex 2.9with acetone 192Figure A.2. Plot of ln(kobs/T) versus lIT for the reaction of complex 2.9with acetone 193Figure A.3. Plot of the data contained in Table A.4 197xliiList of TablesTable 2.1. Numbering Scheme, Color, Yield and Elemental Analysis Data forComplexes 2.1 - 2.9 16Table 2.2. Selected Mass Spectral and Infrared Data for Complexes 2.1 - 2.9 17Table 2.3. 1H Nt4R Data for Complexes 2.1 - 2.9 18Table 3.1. Numbering Scheme, Color, Yield, and Elemental Analysis Data forComplexes 3.1 - 3.16 56Table 3.2. Selected Mass Spectral and Infrared Data for Complexes 3.1 - 3.16 57Table 3.3. NMR Data for Complexes 3.1 - 3.16 58Table 4.1. Numbering Scheme, Color, Yield and Elemental Analysis Datafor Complexes 4.1 - 4.14 99Table 4.2. Numbering Scheme, Color, Yield and Elemental Analysis Datafor the Carbonylation Products 4.1’ - 4.10’ 100Table 4.3. Selected Mass Spectral and Infrared Data for Complexes 4.1 - 4.14 101Table 4.4. Selected Mass Spectral and Infrared Data for Complexes 4.1’ - 4.10’ 102Table 4.5. NMR Data for Complexes 4.1 - 4.14 103Table 4.6. NMR Data for Complexes 4.1’ - 4.10’ 106Table 5.1. Numbering Scheme, Color, Yield, and Elemental Analysis Data forComplexes 2.9-d15,5.1 - 5.5 138Table 5.1. Selected Mass Spectral and Infrared Data for Complexes 2.9-d15,5.1-5.5 138Table 5.3. NMR Data for Complexes 2.9-d15,5.1 - 5.5 139Table 6.1. Numbering Scheme, Color, Yield and Elemental Analysis Data forComplexes 6.1 -6.5 164Table 6.2. Selected Mass Spectral and Infrared Data for Complexes 6.1 - 6.5 164Table 6.3. 1H NN’IR Data for Complexes 6.1 - 6.5 165xivTable A.1. Data for the reaction between complex 2.9 and acetone at differentacetone concentrations 188Table A.2. Data for the reaction between complex 2.9 and acetone (25 timesexcess) at different temperatures 191Table A.3. Data for the monitoring of the reactions between 2.9 and C6Hand 2.9 and C6D 194Table A.4. Data for the monitoring of the reactions between 2.8 and C6Hand 2.8 and C6D 196xvList of SchemesScheme 2.1. Insertion chemistry of Cp*W(NO)(CH2SIMe3)H.27Scheme 2.2. Stereochemistry of asymmetric ketone insertion into the W - H bond ofCp*W(NO)(CHSiMe3)H 29Scheme 2.3. Alkyne insertions into metal-hydride bonds: regio- and stereoselectivity 36Scheme 3.1. Reduction of complex 3.2 by zinc 79Scheme 4.1. Benzene activation by complex 4.7 115Scheme 5.1. Hydrogenation of Cp*W(NO)(CH2SiMe3)in the presence ofPPh3 141Scheme 6.1. Benzene activation by complex 2.8 169Scheme 6.2. Pentane activation by complex 2.8 176Scheme 6.3. Hydrogenation of complex 2.8 in the presence of phenylacetylene 176xviList of AbbreviationsThe following list of abbreviations, most of which are commonly used in the chemicalliterature, have been employed in this thesis.A angstrom, 10-10 manal. analysisatm atmosphere(s)br broad (spectral)degree centigradel3 carbon-13‘3C{’H} proton-decoupled carbon-13calcd calculatedC6D benzene-d6CDC13 chloroform-d1CD21 dichloromethane-d2cm wavenumbersCOSY correlation spectroscopyCp i5-CH,cyclopentadienylCp* ii5-CMe,pentamethylcyclopentadienylCp’ both Cp and Cp*CV cyclic voltammogramö chemical shift in ppm referenced to Me4Si at ö 0d doublet (in an NMR spectrum); or day(s)heat (in thermolysis)El electron-impact (in mass spectrometry)Et20 (CH3C2)0,ether, diethyl ethereV electron voltsxviiFAB fast-atom bombardment (in mass spectrometry)g gram(s)AG free energy of activationh Planck’s constant1H protondeuteriumA111 enthalpy of activationHETCOR heteronuclear correlation spectroscopyHz Hertz (s-i)JR infraredJ coupling constant (in NMR spectroscopy)flJp n-bond coupling constant between atoms A and BK degree KelvinkJ kilojoulek Boltzmann’s constantk rate constantkobs observed rate constantLDA lithium diisopropylamideLUMO lowest unoccupied molecular orbitalm multiplet (in NMR spectroscopy)M Mo and W; or molar; or megam/z mass-to-charge ratio (in mass spectrometry)Me CH3, methylmg milligram(s)mm minute(s)mL millilitermmol millimolemol molexviiiMS mass spectrummV millivoltv stretching frequency (in JR spectroscopy)neopentyl CH2Me3neophyl CH2MePhNMI{ nuclear magnetic resonanceORTEP Oak Ridge Thermal Ellipsoid Plotphosphorus-3 1[pj+ parent molecular ion (in mass spectrometry)Ph C6H5,phenylppm parts per million (in NMR spectroscopy)q quartet (in an NMR spectrum)RT room temperatures singlet (in an NMR spectrum)ASI entropy of activationSCE standard calomel electrodet triplet (in an NMR spectrum)tetraglyme tetraethylene glycol dimethyl ether, Me(OCH2CH4OMeo-tolyl C6H4-2-Me, ortho-tolylp-tolyl C6H4-4-Me, para-tolylTHF C4H80, tetrahydrofuranTHF-d8 C4D80UV ultraviolet (in electronic spectroscopy)xixAcknowledgementsFirstly, my thanks go to Peter Legzdins. His unique style of supervision has allowed me topursue my own ideas, while still offering encouragement and guidance when it was needed.Through Peter’s example, I have learned the value of setting high standards for both research andpresentation, which will undoubtedly serve me well in the future. Special thanks also go to RossHill, my undergraduate supervisor, for initiating me into the world of research with a beer on myfirst day in the lab. Without his influence I would not have pursued graduate work in chemistry.I thank my parents for their support and encouragement, and hope to help them understand thiswork someday. Thanks also goes to other members of the Legzdins group (Neil Dryden, EdwardVessey, Eric Brouwer, George Richter-Addo, Mike Shaw, Penny Lundmark, Steve McNeil,Michelle Young, Roser Reina, Kevin Smith, Steven Sayers) for making my time here interestingand enjoyable. Special thanks to John Veitheer for numerous chemistry discussions, and to KevinRoss for his all-too-frequent constructive criticisms. Thanks to Michele Barlow for distracting mefrom my work. Thanks to the members of my guidance committee, Mike Fryzuk, Alan Storr, andRay Andersen for their advice. Special thanks to crystallographers Ray Batchelor and FredEinstein for telling me what I had made, and for their enthusiasm toward my work. Thanks toSteve Rak, Steve Rettig, Peter Borda, Marietta Austria, and Leanne Darge for technical helpalong the way.1CHAPTER 1General Introduction and Thesis Outline1.1 Introduction and Background 11.2 Outline of This Thesis 31.3 References and Notes 51.1 Introduction and BackgroundAt the core of organotransition-metal chemistry is the study of metal-carbon and metal-hydride bond reactivites. These important links have received a great amount of attention sincethe field of organometallic chemistry was founded, 1,2 with the impetus behind such studies being adesire to understand their fundamental chemistry. With this knowledge comes not only anunderstanding of chemical reactivity, but also the ability to predict this reactivity. This predictivepower allows for the planning of reactions such that useful materials can be synthesized throughboth catalytic and stoichiometric reactions. Through this process, organotransition-metalchemistry is finding an ever-increasing number of applications.1 The study of the fundamentalchemistry involved in these processes, and of metal-carbon and -hydride bonds in particular, is farfrom complete, however, and a great amount of work is yet to be performed before completeunderstanding is reached. There is greater interest and pace of research in these fields than everbefore, as an increasing number of research groups strive to uncover and apply the fundamentalchemistry of these important bonds.In the last few years, the contributions that our laboratories have made to this field have beenthe study of the characteristic chemistry of Cp’M(NO)R2complexes [R = alkyl, aryl], whoseV 2chemistry is dominated by the reactivity of their metal-carbon a-bonds. Although we haveperformed a significant amount of work on these complexes, many reactivity studies can still beenvisioned. The ultimate goal in such work is to apply this knowledge such that the reactivity ofthese complexes can be exploited to a useful end. This particular class of complexes,Cp’M(NO)R2,belong to a special subset of organotransition-metal compounds: those that exhibita 16-valence-electron configuration. These Group 6 compounds are able to defy the 18-valence-electron rule because the LUMO, which would normally contain the ‘extra’ two electrons, is non-bonding in nature, and much higher in energy than the HOMO.3 Thus, these compounds do notlose electronic stability by exhibiting a 16-electron versus an 18-electron valence configuration.These compounds are also unique in the respect that they are coordinatively unsaturated.Molecular orbital calculations reveal that the LUMO in these compounds is metal-centered and itslargest lobe is located between the two a-donor ‘legs’ of these three-legged piano-stoolcompounds. The coordinative and electronic unsaturation exhibited by these complexes makesthese species highly reactive and allows for chemistry unattainable by 18-valence-electroncoordinatively saturated compounds.The nitrosyl ligand also helps to impart some unique chemistry to these compounds. Thestrong n-accepting ability of the NO ligand, via backdonation from a metal dn orbital to thenitrosyl n’ orbital, induces a Lewis acidic nature to the metal center.4 This causes the metalcenters to be highly prone to nucleophilic attack, which is the reaction pathway that dominatesmuch of the chemistry of such compounds.5 The nitrosyl ligand behaves as a spectator ligand innearly all of the reactivity studies performed, but recently we have noted some instances whencleavage of the nitrosyl N-O bond has been observed,6with one example of such a reactionincluded in this Thesis.Ever since the discovery of CpW(NO)(CH2SiMe3)in these laboratories in 1985, we havebeen involved with the synthesis and reactivity of this and analogous 1 6-valence-electron species.Great success has been achieved, such that the extension to molybdenum, Cp*, alkyl and arylspecies has been made. In fact, examples of all the possible combinations of complexes containing3alkyl or aryl, Cp or Cp*, tungsten or molybdenum can at least be generated in solution, if notisolated.5 The most unstable and the only unisolable members of this family of compounds are theCpMo(NO)(aryl)2species, which can only be generated in solution at low temperatures.The members of the Cp’M(NO)(R)2class of compounds that have received the mostattention with respect to reactivity studies are the two Cp’W(NO)(CH2SiMe3)species, mainlybecause their syntheses are high-yielding and straightforward, and because the compounds areeasily isolable and stored. The study of the chemistry of these compounds has indeed beenproductive. The Cp compound reacts with Lewis bases to form 1:1 metal-centered adducts, bothspecies form isonitrosyl adducts with Lewis acids, and they react with water to form dioxo alkylspecies.8 The insertion chemistry of the tungsten-carbon a-bonds has been investigated at length,and it has been shown that sulphur,9 selenium, NO, isonitriles, and isocyanates all insert to form avariety of 18-valence-electron products.10 The dialkyl complexes are also prone to react withalcohols to yield alkoxide species.11 The reactivity of both complexes with molecular hydrogenhas been briefly investigated, and it has been shown that hydride species are formed viahydrogenolysis of the tungsten-carbon a-bonds. In addition to the reactivity displayed by the bis(trimethylsilylmethyl) compounds, further reactivity studies including CO’213 andheterocumulene insertions,14and reactions with water and oxygen15 have been performed withother dialkyl and diaryl Cp’M(NO)R2analogues.1.3 Outline of This ThesisThis Thesis begins by discussing the continued investigation of the reactivity of the 16-valence-electron dialkyl species, Cp*W(NO)(CH2SiMe3)with molecular hydrogen. Exposureof this compound to H2 results in the formation of an alkyl hydride species,Cp*W(NO)(CH2SiMe3)H. The 16-valence-electron hydride complex is unstable and cannot beisolated as such, but its existence can be inferred by its reaction with PMe3 to form the isolable18-valence-electron species, Cp*W(NO)(H)(CH2SiMe3)P12The chemistry of the transient16-valence-electron alkyl hydride complex is the focus of the investigations presented in Chapter42 of this Thesis. This hydride species is generated in the presence of a number of reagents, andthe products have been isolated and characterized. When unsaturated reagents are used, thetungsten-hydride bond usually undergoes insertion of reagents, but this is not always the case.Some of the products of these reactions are quite interesting, and in fact, the chemistry presentedin the remainder of the Thesis all derives from the reactivity of the hydrogenation productsdiscussed in this Chapter.Chapter 3 outlines the investigation of the chemistry of the alkyl and aryl chloride complexesCp*W(NO)(R)Cl. The first example of this class of compound is produced via one of thehydrogenation products mentioned in Chapter 2, and a general synthetic methodology isdeveloped such that a range of these alkyl and aryl chloride compounds can be synthesized ingood yield. The neopentyl complex, Cp*W(NO)(CH2Me3)Cl,is used as a representativeexample of this class of compound to investigate their chemistry. These species are highlyreactive, and are extremely useful precursors in the synthesis of many mixed-ligand species of thetype Cp*W(NO)(CH2Me3)X[X = alkyl, aryl, amide, alkoxide]. Other than metathesischemistry, the reactivity of these compounds toward phosphines, CO, isonitriles, reducing agents,and silver salts is also presented. The reactivity of some 18-valence-electron compounds formedin these reactions is also discussed.Chapter 4 presents the synthesis and reactivity of a range of mixed alkyl and aryl complexesderived from the alkyl and aryl chlorides introduced in Chapter 3. These 16-valence-electroncomplexes, Cp*W(NO)(R)R1, exhibit similar reactivity to their symmetric dialkyl and diarylanalogues. However, the advantage of the mixed-ligand system is that the reactivity of differentmetal-carbon bonds are easily compared at the same metal center. To this end, a series of mixedalkyl and aryl complexes has been reacted with CO to determine the relative migratory aptitudesof the different hydrocarbyl ligands. The results of this study are presented, along with someinvestigations into the thermal chemistry exhibited by some of the mixed complexes.A product of one of the hydrogenation reactions presented in Chapter 3,Cp*W(NO)(H)(r12_PPh(C6H4)),is the focus of Chapter 5. This orthometallated complex is able5to undergo demetallation, which allows for the complex to coordinate Lewis bases and to activatearomatic C-H bonds. The mechanisms involved in these transformations are discussed, andkinetic data are presented as support for the proposed mechanisms.Chapter 6 deals with the chemistry of yet another product of the hydrogenation reactions ofChapter 2. The alkyl alkenyl species, Cp*W(NO)(CH2SiMe3)(CPh=C,displays someintriguing thermal chemistry that is initiated by the elimination of TMS from the complex. Theintermediate that is formed from this elimination process is exceedingly reactive, as it is able toactivate both aromatic and aliphatic C-H bonds. The products of both types of bond activationsare discussed, and mechanisms are proposed to account for their structures.An Appendix is included in the Thesis which contains the data and sample calculations for thekinetic studies presented in Chapters 5 and 6.1.4 References and Notes(1) Coliman, J. P.; Hegedus, L. S.; Norton, J. R.; Finke, R. G. Principles and Applications ofOrganotransition Metal Chemistry; University Science Books: Mill Valley, CA, 1987.(2) The Chemistry of the Metal-Carbon Bond; Hartley, F. R., Ed.; Wiley: New York, 1987;Vol. 4, Part 2.(3) Legzdins, P.; Rettig, S. J.; Sanchez, L.; Bursten, B. E.; Gatter, M. G. I Am. C’hem. Soc.1985, 107, 1411.(4) Richter-Addo, G. B.; Legzdins, P. Metal Nitrosyls; Oxford University Press: New York,1992; Chapter 1.(5) Legzdins, P.; Veltheër, J. E. Ace. Chem. Res. 1993, 26, 41.6(6) (a) Legzdins, P.; Rettig, S. J.; Ross, K. J.; Veitheer, J. E. J Am. Chem. Soc. 1991, 113,4361. (b) Legzdins, P.; Debad, 3. D.; Young, M. A. Organometallics, submitted forpublication.(7) Legzdins, P.; Rettig, S. 3.; Sanchez, L. I Am. Chem. Soc. 1985, 107, 1411.(8) Legzdins, P.; Phillips, E. C.; Sanchez, L. Organometallics 1989, 8, 940.(9) Legzdins, P.; Sanchez, L. I Am. Chem. Soc. 1985, 107, 5525.(10) Legzdins, P.; Rettig, S. 3.; Sanchez, L. Organometallics 1988, 7, 2394.(11) Phillips, E. C. Ph. D. Dissertation, University of British Columbia, 1989.(12) Legzdins, P.; Martin, J. T.; Einstein, F. W. B.; Jones, R. H. Organometallics 1987, 6,1826.(13) Dryden, N. H.; Legzdins, P.; Lundmark, P. J.; Riesen, A.; Einstein, F. W. B.Organometallics 1993, 12, 2085.(14) Brouwer, E. B.; Legzdins, P.; Rettig, S. J.; Ross, K. J. Organometallics 1993, 12, 4234.(15) Legzdins, P.; Lundmark, P. J.; Phillips, E. C.; Rettig, S. J.; Veitheer, J. E. Organometallics1993, 11, 2991.7CHAPTER 2Generation and Reactivity of Cp*W(NO)(CH2S1Me3)H:a 16-Electron AlkylHydride Complex.2.1 Introduction 72.2 Experimental Procedures 102.3 Results and Discussion 212.4 Conclusions 432.5 References and Notes 442.1 IntroductionMuch work has been done in these laboratories to decipher the characteristic chemistry ofthe unusually stable 16-valence-electron dialkyl and diaryl complexes, Cp*W(NO)R2.Oneprototypal complex in particular, Cp*W(NO)(CH2SIMe3)has received most of the attention inthese reactivity studies. For example, this complex has been shown to form metal-centeredadducts with appropriate Lewis bases, form isonitrosyl adducts with Lewis acids, insert CO,isonitriles, sulfur, and selenium into its tungsten-carbon bonds, and react with oxygen and waterto produce dioxo alkyl compounds.’ The reactivity of Cp*W(NO)(CH2SiMe3)with molecularhydrogen has also been briefly investigated.2 When this complex is placed under a high pressureof hydrogen, both of the alkyl ligands are lost via hydrogenolysis of the tungsten-carbon bonds.The resulting dihydride species dimerizes and is isolated as [Cp*W(NO)HJ2.Hj(equation2.1).84Hoz(s /\/>2.1oNi.{R= CH2SiMe3 0N<1 \PR’3RHWhen the hydrogenolysis is repeated using a much lower pressure of hydrogen, all of thestarting material reacts, but no products can be isolated from the reaction solution. The oneexception that has been observed is when this mild hydrogenation is performed in the presence ofa Lewis base, such as a phosphine (PR3). Under such circumstances, complexes of the typeCp*W(NO)(CH2SiMe3)(H)PR can be isolated (equation 2. 1).2 These products arise when oneof the alkyl groups is lost via hydrogenolysis, and the resulting alkyl hydride is trapped with thephosphine base. The 18-electron products are stable at room temperature, and, at the time thischemistry was first performed, were rare examples of stable alkyl hydride transition-metalcomplexes. The trimethylphosphine-trapped hydride showed some interesting C-H bondactivation chemistry. Upon heating, alkane was reductively eliminated from the complex, andaromatic solvents such as benzene could be activated to yield aryl hydride species (equation 2.2).I 6O°C,CU IW 2.2I \ PMe3 I V’PMe3RH PhilR = CH2SiMe39Although this 18-valence-electron alkyl hydride complex, Cp*W(NO)(CH2SiMe3)(H)PMeshowed interesting thermal chemistry, it proved to be inert to substrates that normally react withhydride links, such as unsaturated bonds. This lack of reactivity was disappointing, but it wasalso realized that the potentially more reactive 16-electron alkyl hydride had been easilygenerated via hydrogenolysis of one of the tungsten-carbon cy-bonds. Even though this speciesneeded to be generated in situ, it was decided that the chemistry of this coordinatively andelectronically unsaturated alkyl hydride species was worth further study. The chemistrypresented in this Chapter consists of an investigation of the chemistry of this transient 16-electronalkyl hydride complex, Cp*W(NO)(CH2SiMe3)H.All reactions presented in this chapter were performed similarly: the hydrogenation of thedialkyl complex Cp*W(NO)(CH2SiMe3)was effected in the presence of reagents that have theability to act as a Lewis base and/or contain an unsaturated link. Only reagents that did not reactwith the dialkyl starting material were chosen for this study. Although the reactions wereperformed similarly, the products testify that the reaction pathways differ for different reagents.When the alkyl hydride is generated in the presence of unsaturated links such as C=O or C=Nbonds, the hydride adds across such bonds to give complexes with tungsten-oxygen or -nitrogenlinks. Similarly, when the hydrogenation is performed in the presence of a substituted acetylene,the acetylene again inserts, this time to form a vinyl species. In the presence of a possible4-electron donor such as butadiene, both of the alkyl groups are lost and the diene coordinates inanii4-fashion to the 14-electron Cp*W(NO) fragment. When the trapping of the alkyl hydride isattempted with triphenylphosphine, both alkyl groups are again lost, and the phosphine orthometalates to form a stable hydride compound.The products of these hydrogenation reactions are interesting both in light of their formationand in the chemistry that they themselves exhibit. The reactivity of a few of these products arediscussed in later chapters, and in fact, all of the work in the following chapters derives in oneway or another from the reactivity of the hydrogenation products presented here.102.2 Experimental Procedures2.2.1 MethodsThe methodologies described in this section apply to the entire thesis. All reactions andsubsequent manipulations involving organometallic reagents were performed under anhydrousconditions in an atmosphere of prepurified argon or nitrogen. Purification of inert gases wasachieved by passing them first through a column containing MnO and then a column of activated4A molecular sieves. Conventional glovebox and vacuum-line Schlenk techniques were utilizedthroughout.3 The gloveboxes utilized for dry chemistry were Vacuum Atmospheres HE-553-2and HE-43-2 models. Some reactions were either completely or partially performed in aInnovative Technology Labmaster double station glovebox equipped for wet chemistry andcontaining a freezer maintained at -35 °C. Many reactions were performed in a thick-walledbomb, here defined as a glass vessel containing a Kontes greaseless stopcock and a side-arm inletfor vacuum-line attachment.All IR samples were either as solutions in NaCl cells or as Nujol mulls sandwiched betweenNaC1 plates. IR spectra were recorded on a Nicolet 5DX FT-W instrument, internally calibratedwith a He/Ne laser. All NMR spectra were obtained on a Varian Associates XL-300spectrometer and are reported in parts per million. 1H NMR spectra (299.94 MHz) arereferenced to the residual proton signal ofC6D (ö 7.15), C7H8 (ö 2.09), CD21 (ö 5.34),CDC13 (ö 7.24) or dioxane-d8(ö 3.53). 31P{’H) NIvIR spectra (121.42 MHz) are referenced toexternal P(OMe)3 set at ö 141.00 ppm relative to 85%H3P04.‘3C{’H} NMR spectra (75.43MHz) are referenced to the natural abundance carbon signals of the solvent employed: C6D(ö 128.00), CDC13 (ö 77.00) or CD21 (ö 53.80). Mrs. M. T. Austria, Ms. L. K. Darge, andDr. S. 0. Chan assisted in obtaining the NMR data. Mass spectra were recorded by Dr. G. K.Eigendorf and the staff of the mass spectroscopy laboratory. Low-resolution mass spectra (El,70 eV) were recorded on a Kratos MS5O spectrometer using the direct-insertion method. Fastatom bombardment (6 kV ion source, 7-8 kV xenon FAB gun) mass spectra were recorded on an11AEI MS 9 spectrometer using 3-nitrobenzyl alcohol as matrix. All elemental analyses wereperformed by Mr. P. Borda of this Department.2.2.2 ReagentsThe organometallic reagents Cp*W(NO)C124and Cp*W(NO)(CH2SiMe3)5were preparedby established procedures. Solvents were freshly distilled from appropriate drying agents under adinitrogen atmosphere and were either purged for 10 mm with argon prior to use or were directlyvacuum transferred from the appropriate drying agent. Dioxane, tetrahydrofhran and diethylether were distilled from sodium/benzophenone; hexanes, benzene, toluene and pentane weredistilled from sodiumlbenzophenone/tetraglyme; dichioromethane was doubly distilled fromP205; CH3N and acetone were doubly distilled from CaH2.6 C6D,dioxane-d8,C7D8,and2,3-dimethyl-1,3-butadiene (Aldrich) were dried over activated 4A molecular sieves anddegassed using 3 freeze-thaw-pump cycles. CDC13 was dried overP205 for two days and thenfiltered through a short column of neutral alumina, activity 1. Hydrogen (Linde, extra dry) wasused as received. Phenylacetylene (MCB) was passed through a short column of activatedalumina before use, benzaldehyde (Fisher) was dried overP205 and filtered through Celite beforeuse, PPh3 (Strem) was recrystallized from hexanes, 4-phenyl-3-penten-2-one,benzophenoneimine, 1 ,4-diphenyl- 1,3 -butadiene (Aldrich), and butadiene (Matheson) were usedas received.2.2.3 SynthesisIsolated yields, physical properties, and spectroscopic data for all complexes are listed inTables 2.1 - Preparation of Cp*W(NO)(CH2S1Me3)(N=CHMe) (2.1)Cp*W(NO)(CHS1Me)(1.90 g, 3.62 mmol) was dissolved in acetonitrile (30 mL) andtransferred to a 500-mL Fisher-Porter vessel. The atmosphere in the vessel was replaced withH2 (12 psig), and the solution was left to stir overnight at room temperature. The resulting red-orange solution was transferred via cannula to a Schlenk tube, pumped to dryness, and theresidue washed with cold pentane repeatedly (3 x 10 mL). The yellow powdery product thatremained (1.33 g, 78% ) was used for subsequent reactions, and could be recrystallized fromhexanes at -3 0°C to obtain analytically pure product. Preparation of Cp*W(NO)(CH2SiMe)(OCHMe (2.2)Cp*W(NO)(CHSiMe)(0.375 g, 0.716 mmol), acetone (0.5 mL, 7 mmol), and hexanes(10 mL) were placed in a thick-walled bomb. The vessel was then charged with H2 (— 18 psig)and the solution was stirred for five hours. After this time, the solvent was removed undervacuum to leave a red powder. The powder was dissolved in a minimum amount of pentane andcooled to -30 °C to induce formation of deep red crystals of the product. These crystals wereisolated by removing the mother liquor via a cannula, and more were obtained from thesupernatant by further concentration and cooling (0.28 g total, 79% yield). Preparation of Cp*W(NO)(CIEI2SiMe3)(OCHPh)(2.3)Cp*W(NO)(CHSIMe)(0.320 g, 0.611 mmol) was dissolved in 15 mL of hexanes in athick-walled bomb. Benzaldehyde (0.067 g, 0.65 mmol) was added, and the atmosphere in thebomb was then replaced with H2 (-18 psig). The solution was stirred for 16 hours, during whichtime its color changed from purple to deep orange. The solvent was removed in vacuo to leave ared-orange powder, which was redissolved in pentane and cooled to -30 °C. Red crystals weredeposited overnight, and were isolated by removing the supernatant with a cannula. Furtherconcentration and cooling of the supernatant yielded more crystalline product (0.28 g total, 83%yield). Preparation ofCp*W(NO)(CH2SiMe3)(OCHMeCH=CHPh) (2.4)Cp*W(NO)(CHSiMe)(0.090 g, 0.17 mmol) was dissolved in 10 mL of hexanes in asmall, thick-walled bomb. 4-Phenyl-3-buten-2-one (0.025 g, 0.17 mmol) was added, and thevessel was charged with H2 (18 psig). After five hours, the solution had become a deep redcolor, and the solvent was then removed under vacuum. The red residue that remained wasdissolved in a minimum amount of pentane and transferred to a small Schlenk tube via cannula.Cooling of this solution overnight at -30 °C resulted in the formation of red crystals. Furtherconcentration and cooling of the supernatant produced more crystalline product (0.80 g total,80% yield). Preparation ofCp*W(NO)(CH2S1Me3)(NIICHPh (2.5)Cp*W(NO)(CHSiMe)(0.280 g, 0.534 mmol) and benzophenoneimine (0.10 g,0.55 mmol) were weighed into a small vial and dissolved in 10 mL of hexanes. This solution wastransferred to a thick-walled bomb which was then charged with H2 (18 psig). After the solutionhad been stirred for 8 hours, the solvent was removed under vacuum to leave a yellow solid.This solid was then washed with cold pentane (2 x 10 mL) and dried under vacuum to obtain ayellow, microcrystalline solid (0.16 g, 49% yield). Recrystallization fromCH21/hexanesafforded analytically pure product. Preparation of Cp*W(NO)(14diene) Compiexes 2.6 and 2.7.Complexes 2.6 and 2.7 were prepared identically, except that the one equivalent of2,3-dimethylbutadiene was weighed into the reaction vessel, whereas butadiene was used inexcess and condensed into the vessel. The preparation of the butadiene complex is given as anexample.Cp*W(NO)(CH2SiMe3)(0.400 g, 0.763 mmol) was placed in a 300-mL thick-walledbomb. The vessel was placed under vacuum and one atmosphere of butadiene gas was added.14The bomb was then cooled to -178 °C to condense the butadiene gas. Pentane (60 mL) was thenadded slowly so that it froze on the sides of the vessel as it was added. The atmosphere wasagain removed, and replaced with H2 (15 psig). The vessel was warmed to room temperature,and the solution was stirred overnight. The solvent was removed in vacuo from the resultingyellow solution to leave an orange powder, which was washed twice with cold pentane(2 x 10 mL). The remaining powder was dissolved in a minimum of hexanes and cooled to-30 °C to induce crystallization. Yellow crystals of 2.6 were isolated by removing thesupernatant by cannulation (0.17 g, 32% yield). In a similar manner, 2.7 was also isolated asyellow crystals in 31% yield. Preparation of Cp*W(NO)(CH2S1Me)(CPh=C (2.8)Cp*W(NO)(CHSiMe)(0.500 g, 0.954 mmol) and phenylacetylene (0.15 g, 1.5 mmol)were weighed together in a small vial in a glovebox. Hexanes (30 mL) were added to dissolvethe solids, and the solution was transferred to a 300 mL thick-walled bomb. H2 (18 psig) wasadded to the vessel, and the solution was stirred for 5 hours, during which time it turned a darkbrown color. The solvent was then removed under vacuum, and the residue was dissolved inEt20 (3 mL) and placed on the top of a chromatography column (Florisil, 10 x 2 cm) made upwith hexanes. Et20/hexanes (1:2) was used to elute the column. A faint yellow band was firstcollected, followed by a red band. Much of the material did not chromatograph, as evidenced bya large brown band that was not eluted from the top of the column. Both the yellow and redeluates were taken to dryness in vacuo, and the residues were dissolved in a minimum of pentaneand cooled to -30 °C to induce crystallization. The solution resulting from the red banddeposited a small amount of red crystals of complex 2.8. The solution resulting from the yellowband deposited a small amount of white crystals that were identified as 1,4-diphenyl-1,3-butadiene by mass spectrometry, 1H and 13C NMR, and infrared spectroscopies, and bycomparisons of these data with those obtained from an authentic sample.15Spectroscopic data for 1,4-diphenyl-1,3-butadiene: IR (Nujol) 1491, 1464, 1073, 993, 912,740, 689 cm1. 1H NMR (CDC13) 7.50 (m, 411, FL3), 7.17 (in, 411, Hme), 7.07 (in, 2H,6.83 (in, 211, C=CHPh), 6.50 (m, 2H, C=CHCH=C). ‘3C{’H} (CDC13)ö 137.44,132.90, 129.33, 128.75, 127.65, 126.46. Low-resolution mass spectrum: m/z 206 [PJ. Preparation of Cp*W(NO)(H)(qPPhH4)(2.9)Cp*W(NO)(CHSIMe)(0.480 g, 0.916 mmol) and PPh3 (0.288 g, 1.10 mmol) werecombined in a large thick-walled bomb (500 mL capacity) and then dissolved in pentane (30 mL).The vessel was then charged with H2 (18 psig) and the solution stirred for one hour. After thistime, pressure in the vessel was vented, and H2 was again added. This procedure was repeatedevery hour for five hours, during which time the product was deposited as a yellow precipitate,and the purple solution became red-brown in color. Approximately one-half of the solvent wasthen removed under vacuum and the precipitate was allowed to settle to the bottom of the vessel.The supernatant was removed via a cannula, and the remaining solid was washed with pentane(20 mL). The bright yellow powder that remained was dissolved in a minimum amount ofEt20and cannulated into a Schienk tube. Cooling of this solution induced the formation of brightyellow crystals of the product (0.240 g from three fractions, 43%).2.2.4 NMR MonitoringExcept for the butadiene reaction, all experiments, as well as being performed on apreparative scale, were also performed in NIvIR tubes in C6D to facilitate monitoring. In eachcase, an equimolar amount of Cp*W(NO)(CH2SiMe3)and the trapping reagent were weighedinto a small vial in a glovebox. C6D (0.6 mL) was added to dissolve the solids, and the solutiontransferred by pipette into an NtvIR tube equipped with a Teflon gas inlet.After an initial spectrum had been recorded, the solution was frozen and the atmosphere inthe tube replaced with H2 (‘-‘10 psig). The tube was warmed to room temperature and thesolution shaken to help dissolve the gas. A 1H NMR spectrum was recorded periodically until16the reaction was deemed to be complete, whereupon a final spectrum was recorded (usually afterapproximately 6 hours).2.2.5 Characterization Data for Complexes 2.1 - 2.9Table 2.1. Numbering Scheme, Color, Yield and Elemental Analysis Data forComplexes 2.1 - 2.9.compd color anal. found (calcd)complex no. (yield, %) C H NCp*W(NO)(CH2SiMe3) 2.1 yellow (78) 40.50(40. 17) 6.34(6.32) 5.99(5.86)(NCHMe)Cp*W(NO)(CH2SiMe3) 2.2 red (79) 40.58(41.21) 6.52(6.71) 2.48(2.83)(OCHMe2)Cp*W(NO)(CHSiMe3 2.3 red (83) 46.38(46.41) 6.15(6.12) 2.40(2.58)(OCH2Ph)Cp*W(NO)(CHSiMe3) 2.4 red (80) 49.43(49.40) 6.45(6.39) 2.32(2.40)(OCHMeCH=CHPh)Cp*W(NO)(CH2SiMe3) 2.5 yellow (49) 52.18(52.43) 6.24(6.19) 4.47(4.53)(NHCHPh2)Cp*W(NO)(114butadiene) 2.6 yellow (32) 41.73(41.70) 5.43(5.25) 3.48(3.47)Cp*W(NO)(q4_2,3 2.7 yellow (31) 44.83(44.56) 5.89(5.84) 3.14(3.25)dimethylbutadiene)Cp*W(NO)(CH2SiM3) 2.8 red (N/A) 49.30(48.98) 6.29(6.17) 2.52(2.60)(CPh=CH2)Cp*W(NO)(H) 2.9 yellow (43) 55.18(55.00) 5.02(4.95) 2.24(2.29)(r12-PPh6II4)17Table 2.2. Selected Mass Spectral and Infrared Data for Complexes 2.1 - 2.9.cmpd MS tempt’ IRno. (°C) (Nujol, cm1)2.1 478 [Pj 150 VNO 1615VCN 16522.2 495 [Pj 180 156815412.3 543 [Pj 120 VNO 15452.4 583 [Pj 150 VNO 1549(br)2.5 618 Pj 120 VNO 1514VNH 32492.6 403 [P] 150 VNO 15632.7 431 [Pj 180 VNO 15742.8 539 [Pj 120 VNO 15392.9 611 [Pj 150 VNO 1545VWH 1929a m/z values are for the highest intensity peak of the calculated isotopic cluster(184V[)b Probe temperatures.18Table 2.3. lB NMR Data for Complexes 2.1 - 2.9 (C6D).compd ilL NMR 13C{’H} NMRno. o o2.1 diastereomer 1: diastereomer 1:7.20 (m, 1H, CH) 158.77 (NCHMe)1.71 (s, 15H, C5Me) 109.55 (CMe)1.45 (d, HH = 54 Hz, 311, NCHMe) 21.57 (Me)0.43 (s, 9H, SiMe3) 19.05 (CR2)0.10 (m, 2H, CH2) 10.00 (C5Me)2.85 (SiMe3)diastereomer 2:6.48 (m, 1H, CR) diastereomer 2:1.66 (s, 15H, C5Me) 158.35 (NCHMe)1.62 (d, HH = 5.5 Hz, 3H, NCHMè) 109.00 (C5Me)0.50 (s, 9H, S1Me3) 20.76 (Me)0.25 (m, 2H, CH2) 19.15 (CH2)9.54 (CcMe5)3.10 (SiMe3)2.2 5.22 (sept, 3Pd = 6.3 Hz, 1H, OCHMe2) 111.83 (C5Me)1.61 (s, 15H, C5Me) 82.29 (OCFIMe1.27 (d, HH = 6 Hz, 3H, Me) 31.88 (wc 109 Hz, CR2)1.21 (d, HH 6 Hz, 3H, Me) 27.94, 26.47 (Me)0.73 (d, 2HH = 12 Hz, 1H, CR2) 9.78 (C5Me)0.52 (d, 2HH = 12 Hz, 1H, CR2) 2.38 (SiMe3)0.35 (s, 9H, SiMe3)2.3 7.41 (d, 3Hfl = 6.9 Hz, 2H, H00) 141.66 (C10)7.1 (m, 3H, H1) 129.62, 128.60, 127.99 (C1)5.97 (d, HH = 10.8 Hz, 111, OCH) 111.88 (C5Me)5.88 (d, HH = 10.8 Hz, 1H, OCH2) 85.48 (OCH2)1.44 (s, 1511, C5Me) 35.05(1=119 Hz, SiCH2)0.82 (s, 211, CH2) 9.46 (C5Me)0.36 (s, 9H, SiMe3) 2.28 (SiMe3)19diastereomer 1:7.4 (m, 5H, H1)6.68 (d, HH = 15.6 Hz, 1H, PhCHCH)6.25 (dd,J = 8.1, 15.6 Hz, 1H,PhCHCJI)5.64 (m, 111, OCH)1.86 (s, 1511, C5Me)1.44(d,J=6.3Hz, 3H,Me)0.84 (d, 2J= 11.7 Hz, 1H, CH2)0.70 (d, HH = 11.7Hz, 1H, CH2)0.11 (s, 9H, SiMe3)diastereomer 2:7.4 (m, 5H, H1)6.48 (d, HH = 15.6 Hz, 1H, PhCHCH)6.23 (dd,JHH=615.6 Hz, 1H,PhCHCF])5.64 (m, 1H, OCH)1.95 (s, 15H, C5Me)1.50 (d, HH = 6.3 Hz, 3H, Me)0.81 (d, HH = 12.0 Hz, 111, CH2)0.66 (d, HH = 12.0 Hz, 1H, CR2)0.06 (s, 9H, SiMe3)(both diastereomers)136.99, 136.83 (PhCHCR)134. 16, 134.03 (PhCHCH)131.18, 129.34, 128.66,128.52, 127.67, 127.60,126.74, 126.63 (C1)112.63, 112.47 (C5Me)87.94, 86.40 (OC)35.22, 35.19 (CR2)26.72, 24.38 (Me)10.10, 10.00 (C5Me)1.98, 1.91 (SiMe3)2.5a 7.75 (br, IH, NH) 146.31, 143.31 (C1)7.6 (m, 6H, H1) 128.76, 128.14, 127.99 (C1)7.4 (m, 4H, H1) 126.92, 126.67, 126.50 (C1)6.72 (d, HH = 11.4Hz, CPh2H) 78.92 (C5Me)1.71 (s, 15H, C5Me) 24.75 (‘wc= 104 Hz, CR2)-0.030 (d, J = 2.7 Hz, 2H, CR2) 9.74 (C5Me)-0.085 (s, 9H, SiMe3) 2.04 (SiMe3)2.6 3.56 (m, 1H, CH) 104.20 (C5Me)3.37 (dd, HH = 4.2Hz, 1H, CH) 91.81 (CH)2.95 (dd, HH = 13.8, 3.6 Hz, 1H, CH) 84.99 (CU)2.40 (dd, HH = 6.3, 3.6 Hz, 1H, CU) 54.32 (CR2)1.63 (s, 15H, C5Me) 51.53 (CH2)1.29 (m, 111, CR) 10.39 (C5Me)1.00 (dd, HH = 12.0, 4.2 Hz, 1H, CR)2.7 3.35 (dd, J 4.5, 0.9 Hz, 1H, CR) 104.58 (C5Me)3.25 (d, J= 3.6 Hz, 1H, CR) 103.45 (=C)2.49 (dd,J= 3.6, 1.2 Hz, 1H, CR) 95.54 (C)2.41 (br, 1H, CII) 54.30 (CR2)2.15 (s, 3H, CH3) 52.28 (CH2)1.71 (s, 15H, C5Me) 21.92 (CH3)1.20 (s, 3H, CH3) 21.74 (CR3)10.56 (C5k1e)2.4202.81) 7.87 (dd, HH = 8.0, 1.2 Hz, 211, H00) 227.89 (WCPh’CH2)7.29 (t,J = 8.0, 2H,1meta) 145.25 (C0)7.11 (t, HH = 8.0, IH, Hpara) 137.44 (C00)3.88 (dd,J = 5.4, 1.2 Hz, 111, CCH2) 129.61, 128.83 (C1)3.56 (dd,J = 5.4, 1.2 Hz, IH, C=CH2) 109.56 (C5Me)1.50 (s, 15H, C5Me) 83.12 (WCPh=CH0.692 (d, HH 12.6 Hz, 1H, CH2) 35.49 (‘‘wc 90.1 Hz, WCH2)0.2 14 (d, HH = 12.6 Hz, 111, CH2) 9.47 (C5Me)3.44 (SiMe3)2.9’ 8.09 (d, HH = 8.0 Hz, 1H, aryl) 171.75 (d, J 5.1 Hz, C0.j)7.95 (m, 2H, aryl) 149.23 (d, J, = 54.6 Hz, C0)7.55 (m, 3H, aryl) 1.44.43 (d, J = 33.7 Hz, C10)7.20 (m, 4H, aryl) 134.39, 134.24, 133.42, 133.28,6.95 (m, 4H, aryl) 132.22, 132.17, 130.70, 130.66,129.903.99 (d,“PH = 10.2 Hz, 1WH = 101 Hz, 129.87, 128.90, 128.86, 128.80, 128.76,1H, WH) 128.69, 128.59, 129.47, 128.36 (Caryl)1.79 (s, 15H, C5Me) 127.41 (d, 9 Hz, Caryl)105.54 (C5Me)10.76 (C5Me)a CDC13. b C7D8. C 31P{1H} ö 49.21, Jp 168 Hz.212.3 Results and DiscussionThe chemistry of the 16-valence-electron alkyl hydride species, Cp*W(NO)(CH2SiMe3)H,isdiscussed in this Chapter. This compound, which can be generated by the reaction ofCp*W(NO)(CH2SiMe3)with molecular hydrogen, has been reacted with a variety of reagents.Each class of reagent is discussed in a separate section, since, although in some cases the hydridesimply inserts the unsaturated reagent, other modes of reactivity are also seen and widelydifferent products are obtained.Each of the reactions discussed below, except for the butadiene reaction, was monitored byNMR spectroscopy to determine the true yields of the processes, and also to help identify byproducts and side-products.2.3.1 Reaction with AcetonitrileThe first example of the reactivity of the transient alkyl hydride was observed during a studyof the synthesis of [Cp*W(NO)H]2I.tHj,the dihydride dimer mentioned in the Introduction.The study involved the hydrogenation of Cp*W(NO)(CH2SiMe3in different solvents todetermine the best solvent for preparative purposes. When acetonitrile was used, none of theexpected dimer was observed, and a new compound was isolated in high yield. This product,complex 2.1, was found to contain an azomethine, or alkylideneamido, ligand resulting fromacetonitrile insertion into the tungsten-hydride bond of the transient alkyl hydride complexformed upon hydrogenation (equation 2.3).I H2,NCMe IW W 23N”IR-TMS NIcA10 R 0 RMeR = CII2SiMe3 2.122An X-ray diffraction study was performed to verifSr the molecular structure of 2.1 since thespectroscopic data recorded for the complex were very confusing in nature. For example, the 1HNMR spectrum of this product is complicated, containing signals of differing integrationsattributable to two Cp* and two trimethylsilylmethyl ligands, two high-field doublets, and twomultiplets at low field (Figure 2.1). The molecular structure reveals why the spectroscopic datawere so confusing: two diastereomers were present, differing in the orientation of the methylgroup on the azomethine ligand.__iLI I I 11 1 liii HI I I LI liii II 111 l I ‘I’! IllJ I [Iii I’7 - 6 5 4 3 2 1 OPPMFigure 2.1. 1H NMR ofCp*W(NO)(CH2SiMe3)(N=CHMe) (2.1) in C6D. The complexity ofthe spectrum is due to the presence of two diastereomers.The two diastereomers could not be separated by crystallization, and in fact crystallized inthe same ratio as present in solution. Consequently, the ratio of integrations for the two distinctCp* signals in the 1H NMR spectrum is approximately 60:40, and the molecular structure is alsodisordered between the two diastereomeric orientations in the same ratio. The ORTEP drawingof the solid-state molecular structure of 2.1 (Figure 2.2) clearly shows the two orientations of theethylideneamido ligand, the major isomer being the one with the methyl group pointing away23from the Cp* ligand. The W-N-C angles of the isomers are nearly linear at 168.4 and 171.7thereby implying that the nitrogen of the ethylideneamido ligand is donating its lone pair ofelectrons to the metal center, resulting in an 18-electron count for the complex. This multiplebonding would create a large rotational barrier about the W=N bond and explains why twodiastereomers are present.When the solution 1H NMR spectrum of complex 2.1 is recorded over the temperaturerange -8 to +40 °C, the isomer ratio is observed to be the same as that at room temperature.This observation tends to rule out the existence of a rotational equilibrium between the twoorientations of the azomethine ligand, as a different ratio is expected at different temperatures.However, this does not preclude the presence of a rotational equilibrium with a large thermalrotational barrier that would make the isomerization very slow. There are reported examples oflinear azomethine ligands that rotate about the M=N=C linkage. For example,CpMo(CO)2(N=CBUt)has been shown to exist as two distinct conformers related by rotation ofthe azomethine ligand, and the ratio of conformers can be changed by varying the temperature.7More recently, Jordan and co-workers8have shown that in solution at room temperature there israpid exchange via rotation between two conformations of the azomethine ligand of[Cp2Zr(N=CMePh)(TIW)J. Cooling of a solution of this compound to -82 °C slows therotation such that both isomers are observed.In the absence of any evidence for interconversion between the orientations of theazomethine ligand in complex 2.1, the explanation for the presence of the two isomers is thatthey must be formed during the hydrogenation process itself their ratio being under kineticcontrol during the reaction. Once formed, the isomers cannot interconvert because of the highrotational barrier. During synthesis, the acetonitrile presumably coordinates to the unsaturatedalkyl hydride when it is formed via hydrogenolysis. Insertion occurs, and the a.zomethine ligandthen adopts its preferred orientations, presumably to accommodate overlap between an emptyorbital on the tungsten center and the orbital containing the lone pair on the nitrogen.Positioning of the methyl group toward the Cp* ligand gives rise to the minor isomer, due to the24Figure 2.2. View of the solid-state molecular structure ofCp*W(NO)(CH2SiMe3)(NCHM )(2.1), including selected bond lengths and angles (with esds in parentheses). Disorder in thecrystal precludes the listing of meaningfi.il esds for the atoms within the azomethine ligands.C12W-N1 L753(7)W-N2 1.87W-N20 1.91N2-C11 1.23N20-C11O 1.25W-N1-O1 171.7(6)W-N2-C11 168.4(9)W-N20-C110 173.0(13)C40C3csoC30C20docli120CsSi 7‘CsBond Lengths (A) Bond Angles (deg)25higher steric crowding in this position. Another possible mechanism that may explain theappearance of the two orientations is one in which the nitrile does not coordinate beforeinsertion, but reacts in a concerted manner. In this mechanism, the direction of approach of thenitrile toward the hydride would determine the final orientation of the azomethine ligand. Theminor isomer would arise when the nitrile approaches from the more hindered Cp* side.Many compounds containing an azomethine ligand have been isolated. There are a varietyof synthetic routes to these types of complexes, including reaction of metal halides withLiN=CR2reagents,9insertion of nitriles into metal-alkyl8J°and metal-hydride bonds,11deprotonation of coordinated imines12 and imido ligands,13 and protonation of coordinatednitriles.14 Although the insertion of acetonitrile into metal-hydride bonds is not a unique one,this appears to be the first example of this type of reaction reported for a tungsten complex. Infact, hydrometalation of nitriles is not that common, possibly due to the fact that unsaturatedhydride complexes are difficult to prepare due to their inherent high reactivity.Exposing a CD3N solution of complex 2.1 to a H2 pressure of 500 psi at 50 °C overnighthad no effect on the complex as judged by 1H NMR spectroscopy. Thus, this 18-electroncomplex is very stable toward further hydrogenation. This reaction was performed in part todetermine if complex 2.1 would catalyze the hydrogenation of acetonitrile. There are variousreports of nitrile hydrogenations occurring on clusters,15bil6aand tn-metallic17systems, and avery few such reactions reported for mono-metallic systems.16 Possibly, the extra electron-donating ability of the a.zomethine ligands acts to stabilize and deactivate mono-metalliccomplexes to further reaction with hydrogen, the success of the metal clusters in these catalysesbeing due to multi-metal cooperation in activating H2.There has been very little work reported on the reactivity of the a.zomethine ligand, which issurprising considering the number of such complexes known. Studies on protonation of thisligand have been reported,18 and other work involving coupling of nitriles with the ligand hasalso appeared.19 In Chapter 3, the reactivity of the a.zomethine complex 2.1 with HC1 will bediscussed.262.3.2 Reactivity with C0 and C=N BondsVarious reagents containing unsaturated carbon-oxygen and carbon-nitrogen bonds wereused as trapping agents to establish the insertion chemistry of the transient alkyl hydride,Cp*W(NO)(H)CH2SiMe3.A ketone, an aldehyde, an a, (3-unsaturated ketone, and an iminewere all used in these reactions. In each case, the heteroatom-carbon unsaturated link insertedinto the tungsten-hydride bond to give alkyl alkoxide or alkyl amide products (Scheme 2.1).When a purple solution of Cp*W(NO)(CH2SiMe3)containing acetone or benzaldehyde isplaced under an atmosphere of hydrogen, the color of the solution changes slowly from purple todeep red. Workup of the reactions yield complexes 2.2 and 2.3, respectively. Both compoundswere characterized as alkyl alkoxide species formed via insertion of the C0 bonds into thetransient hydride. When these reactions were also performed in NMR tubes, the conversionswere observed to be quantitative, even with only one equivalent of the trapping agent present.The only examples of alkyl alkoxide complexes of the type Cp’M(NO)(R)OR’ [M = Mo, WIthat had been reported prior to this work were those containing a benzyl ligand (R =CH26H5).°These complexes were synthesized from the precursor complexesCp’M(NO)(R)Cl, which were available from the reaction of the bis-(benzyl) complexes withHC1.2 This latter reaction was not successful for complexes containing other alkyl groups, andtherefore this class of alkyl alkoxide compounds could not be extended due to the lack ofavailable alkyl halide starting materials, Cp’M(NO)(R)Cl. The isolation of complexes 2.2 and 2.3provides an alternative route to such alkyl alkoxide complexes in high-yields. Since this workwas completed, a newer synthetic methodology to such compounds, starting from the alkylchioro complexes discussed in Chapter 3, has been developed to synthesize compounds similar to2.2 and 2.3.2127The alkyl alkoxide complexes 2.2 and 2.3 are thermally stable, and react only slowly with air.This stability most likely arises from electron donation to the metal center from the lone pair onthe oxygen of the alkoxide ligand. This could conceiveably increase the stability of thecomplexes: (a) by giving the metal center a more stable 18-electron configuration; (b) by usingthe unoccupied orbital on the metal center so that possible decomposition pathways are blocked;(c) and by tying up the lone electron pair on the oxygen, thereby reducing the chance ofelectrophilic attack at that atom. Other similar alkyl and aryl alkoxide complexes have since beenshown to be relatively inert, both chemically and thermally.21NROCHMe2Scheme 2.1/V\N Ro RR = CH2SiMe32.2¶1II KN=C©0N0 R2.42.528Cp*W(NO)(CH2SiMe3)may also be hydrogenated in the presence of4-phenyl-3-butene-2-one, and a complex similar to the acetone and benzaldehyde products canbe isolated in good yield (quantitative by 1H NMR spectroscopy). Again, the transient hydrideinserts the C=O bond to form an alkyl alkoxide species, complex 2.4 (Scheme 2.1). In this case,however, the ketone is prochiral, and isomers are formed due to addition of the hydride to eitherface of the carbonyl group. A 1H NMR spectrum of the reaction mixture indicates two speciesthat are present in a ratio of approximately 3:2, which are assignable to the diastereomers formedfrom addition of the racemic hydride to different faces of the ketone.Two mechanisms, similar to those proposed for the acetonitrile insertion products, can beenvisioned to help explain the observed products and their ratio. One involves the initialcoordination of the ketone before insertion into the tungsten-hydride bond, and the otherinvolves direct attack of the ketone at the hydride in a concerted mechanism. If initialcoordination takes place, then the ketone most likely will coordinate between the alkyl andhydride ligands before insertion takes place. This assumption is based on numerous observations.Molecular-orbital calculations performed on the model complex CpMo(NO)Me2show that thecomplex’s LUMO is a metal-centered, nonbonding orbital with its major lobe located between thetwo alkyl ligands.22 A number of Lewis-base adducts of complexes of this type have beenisolated or observed, and in each case, the Lewis base coordinates in the open site between thetwo one-electron ligands. For example, Cp*W(NO)(ptolyl)2PMe33andCp*M(NO)(CH2e3)(Cl)PMe [M = Mo, Wj24 both have the PMe3 coordinated trans to thenitrosyl ligand. Thus, it is expected that the ketone, if it coordinates to the alkyl hydrideintermediate, will coordinate between the alkyl and hydride ligands. The different isomers wouldthen be caused by addition of the hydride to either face of the ketone. In this case, the stericallypreferred mode of coordination of the ketone would be with the smaller substituent, the methylgroup, pointing upward toward the more sterically demanding Cp* ligand. Insertion in thisconformation would give SS and RR configurations for the major enantiomers formed,depending upon the configuration of the hydride enantiomer (Scheme 2.2).29/r0N RHSSRFor a mechanism involving direct attack of the ketone on the hydride ligand, it is moredifficult to assign an absolute configuration to the major isomers. In this mechanism, the ketonecould approach from either side of the hydride ligand, and with either of its prochiral facestoward the hydride. The difficulty arises when considering which side of the hydride ligand ismore accessible for attack. The alkyl hydride cannot be assumed to be a symmetric three-leggedpiano-stool complex, since compounds that might be expected to have a similar molecularstructure do not. For example, in the complex Cp*W(NO)(CH2SiMe3)Cl(3.1), theCH2 - W - Cl angle is larger than the Cl - W - N angle.25 Another consideration is that the stericScheme 2.2Me-z• N’RR’ MeSSR’Me -o’ MeR’Me)Me0N 11RS— \—H0Me R’NN!R’Me30crowding at the possible sites for attack may favor one position over the other. So, although theangle between the hydride and the alkyl group may be large, the position may be crowded by themethyl groups of the trimethylsilylmethyl ligand, and the site between the hydride and the smallernitrosyl ligand may by more accessible. Since many factors influence the nature of the productsin this mechanism, it is difficult to predict which isomers would be favored.The different integrations of the 1H NMR. data allow the signals due to each pair ofenantiomers of complex 2.4 to be identified, although it is not possible to assign the two sets ofsignals to any absolute configuration. Assuming that the mechanism portrayed in Scheme 2.2 iscorrect and the ketone coordinates between the alkyl and hydride ligands before insertion, themajor pair of enantiomers are assignable as those with the SS and RR configurations. A1H NMR spectrum taken after complex 2.4 was recrystallized revealed that one of theenantiomer pairs had been enriched by approximately 50% from one crystallization. Althoughthe experiment was not performed, further recrystallizations could result, theoretically, in theseparation of the two sets of enantiomers.Hydrogenation of Cp*W(NO)(CH2SiMe3)in the presence of benzophenoneiminedemonstrates that the alkyl hydride intermediate is also prone to insert imine fiunctionalities(Scheme 2.1). Complex 2.5, the product of this reaction, has been characterized as the alkylamide Cp*W(NO)(CHSiMe3)NHCHPh The 1H NMR spectrum of this complex contains abroad peak at 7.6 ppm assignable to the NH proton, and the infrared spectrum of the productshows a strong NH stretching band at 3249 cm-’. The yield of complex 2.5 is fairly low, and theyellow product crystallizes from the brown reaction mixture in an impure form, thus requiringrepeated recrystallizations to obtain analytically pure product. When the reaction is monitored by1H NME. spectroscopy, the reaction is observed not to be quantitative. One possible explanationfor this observation that the imine is not as effective a trapping agent as is a carbonyl group, andsome of the hydride intermediate decomposes before it is trapped. This lower trapping efficiencycould be a result of the larger steric bulk of the imine and the lower nucleophilicity of thenitrogen atom as compared to the carbonyl oxygen.31Recently, many alkyl and aryl amide complexes similar to 2.5 have been prepared by adifferent, and more convenient route than discussed here.26 These preparations involve the useof alkyl chloride or amide chloride precursors in their synthesis. One such compound and itssynthesis will be discussed in Chapter 3 during the discussion of the reactivity of alkyl chloridecomplexes.2.3.3 Reaction with DienesWhen Cp*W(NO)(CH2SiMe)is reacted with hydrogen in the presence of butadiene or1,3-dimethylbutadiene, ri4-diene compounds of the type Cp*W(NO)(r141ransdiene) are formed(i.e. complexes 2.6 and 2.7, respectively, equation 2.4). These compounds have been identifiedby their spectroscopic properties, and by comparison of these properties to analogousmolybdenum compounds synthesized by another route.27 The reaction of the dialkyl startingcomplex with 1,3-dimethylbutadiene was followed by 1H NMR spectroscopy and found to benearly quantitative. Therefore, the low isolated yields for these reactions seems to be amanifestation of the high solubility of the complexes which prevents complete recovery bycrystallization.xs H2,WR’ 2.4N’’I\ -2SiMe4o R R’N0R = CH2SiMe3 2.6 R’ = H2.7 R’MeThe proposed mechanism for the formation of complexes 2.6 and 2.7 is shown in equation2.5. Hydrogenolysis of one of the tungsten-carbon bonds of the starting material generates thealkyl hydride intermediate, similar to the mechanisms of the other reactions discussed in thisChapter. Coordination of one double bond of the diene in an2-fashion would then form an3218-electron complex analogous to the phosphine-trapped alkyl hydride compounds. Eliminationof SiMe4 from this intermediate with concerted coordination of the dangling double bond of thediene would then give the observed products.-ç57-w I/ \ - SiMe0NRR = CH2S1Me3• W 2.5- e4N0The reactions with the dienes show that a cooperativity effect is possible in these systems.That is, if the ligand used to trap the hydride intermediate can act in a bidentate manner, then theligand may do so after the elimination of TMS. Whether TMS is liberated or not depends onwhether it is thermodynamically favorable for the complex to do so. In this case, formation ofthe very stablei4-diene complexes is obviously favored over any intermediates that are formed.The proposed‘r2-diene-trapped intermediate is not observed, hence the elimination of the alkaneis presumed to be fast once this intermediate is formed. The hydrogenation reaction with theunsaturated ketone mentioned in Section 2.3.2 was performed to check for similar cooperativityeffects. It was thought that if the ketone coordinated to the metal center and coordination of thedouble bond induced alkane loss, a similar product to the diene complexes would be observed.However, the ketone inserts into the hydride bond before elimination of the alkane can occur, sothat the stable alkyl alkoxide complex is formed instead.33Another reason diene reagents were chosen in this series of reactions was to determine if thehydride would insert the double bond of the alkene. It was thought that if this insertion occurred,the remaining double bond could possibly bend back toward the metal and coordinate, thusproducing a stable 18-electron complex reminiscent of the phosphine-trapped alkyl hydridecomplexes discussed earlier. When the reaction to form complex 2.7 was monitored by 1H NIVIRspectroscopy, no evidence for this type of product was observed, and no evidence for organicproducts resulting from diene hydrogenation could be identified.Complexes 2.6 and 2.7 represent the first complexes containing a diene coordinated to atungsten center that we have been able to isolate. Other similar compounds containingmolybdenum have been synthesized in these laboratories by the reduction of Cp’Mo(NO)12in thepresence of an excess of diene.28 The synthetic methodology employed to make thesemolybdenum diene complexes was difficult in that the reactions were not clean and requiredrepeated chromatography to separate the products. The reactions also had to be monitored byIR spectroscopy to determine when the solution should be removed from the sodium amalgamreductant, since the diene products were not stable in such an environment. Only molybdenumdiene complexes could be produced by this route; when tungsten starting materials were used, nosimilar products could be isolated. Thus, the hydrogenation route outlined here is the onlyknown route to these complexes for tungsten. Subsequently, it has been shown that thishydrogenation route can also be employed to produce diene complexes of molybdenum startingwith the appropriate dialkyl, Cp*Mo(NO)(CH2SIMe3).9Although the yields of the reactionsare comparable to the reduction method, the methodology employed is much simpler and theproduct is easier to isolate. Using this hydrogenation route, novel chemistry involving thecoordination and coupling of cyclic dienes has recently been investigated on both molybdenumand tungsten centers.29342.3.4 Reaction with PhenylacetyleneHydrogenation of Cp*W(NO)(CH2S1Me3)overnight in the presence of phenylacetyleneresults in the formation of a dark brown solution from which no products could be isolated. A1H NMR spectrum of the reaction mixture revealed a multitude of weak signals in the Cp*region, with no major species present. When the reaction was repeated but only allowed toproceed for five hours, however, 1H NMR spectroscopy revealed that one major Cp*containingproduct was present. Workup of this reaction mixture employing chromatography down aFlorisil column allowed the crystallization of a small amount of red product identified asCp*W(NO)(CH2SiMe3)(CPh=C,complex 2.8 (equation 2.6). An organic product was alsoisolated in small amounts and identified as 1,4-diphenylbutadiene by comparing its spectroscopicproperties to those of an authentic sample (the mechanism of formation of this product will bediscussed in Chapter 6). Complex 2.8 is the expected product of insertion of phenylacetyleneinto the intermediate hydride. The very low yield of the compound is a consequence of two nonproductive processes occurring as the reaction progresses. Firstly, complex 2.8 is thermallyunstable; the chemistry involved will be discussed further in Chapter 6. Secondly, the product isnot stable with respect to hydrogen. This can be demonstrated by dissolving an authentic sampleof the alkenyl complex in C6D in an NMR tube and pressurizing the tube with hydrogen to thesame pressures used for the reaction itself Monitoring of the solution by 1H NMR spectroscopyshows a slow decrease in the starting material so that after approximately 16 hours, no signalsattributable to the alkenyl complex are present. A porcupine-like grouping of peaks in the Cp*region indicate that no major product is formed, and a large peak at 0 ppm indicates that TMS isliberated from the complex upon hydrogenation. Thus, to have any chance of isolating complex2.8, the reaction needs to be stopped after all the starting dialkyl has reacted, but before theproduct decomposes either thermally or by reaction with excess hydrogen. The optimal reactiontime was found to be approximately five hours at a pressure of 18 psig H2.35H2, HCCPh ‘1N”I’’R-SIMe4 N”I2.60 R 0 RçR = CH2SiMe32.8Similar reactions performed with diphenylacetylene and 1-pentyne do not yield any isolableproducts, and monitoring these reactions by NME. spectroscopy shows no signals assignable toalkenyl ligands. Products similar to complex 2.8 may have been produced, but would not bevisible if they were very reactive with hydrogen, or more thermally sensitive than thephenylacetylene product. Another explanation for the lack of products is that these acetylenesmay not have inserted at all. All of the reagents that have been shown to insert into the tungsten-hydride bond are all quite polar in nature, and thus Less polar or non-polar reagents like 1 -pentyneand diphenylacetylene may not react at all with the hydride.The regioselectivity of insertion, i.e. whether the unsymmetrical acetylene undergoes c- orf3-metalation, is difficult to predict for these types of insertion processes. For example,W(H)(CO)2(NO)(PMe)30and Co(H)N(CH2HPPh3,’insert the activated acetyleneHCCC{O}OMe to give cx-metalated alkenyl ligands of the type [MJC(C{O}OMe)=CH2,whereas CpRe(CO)(NO)H32gives the other possible regioisomer [M]CH=CH(C{O)OMe). Onthe other hand, the osmium complex [Os(H)(acetone)(CO)2(P’Pr3]BF43is known to undergoboth modes of insertion. It has been noted that the regiochemistry depends somewhat on theacidity of the hydride ligand, and that early transition-metal hydrides are usually hydridic innature, and thus undergo a-insertions with unsymmetric acetylenes containing electronwithdrawing groups, as outlined in Scheme 2.3.30 This seems to hold true when activatedacetylenes are considered, but does not explain why Cp2Zr(H) reacts with the non-activatedsubstrate phenylacetylene to give the f3-metalated productCpZr(H)CH=CHPh,34whenreactions with later transition-metal hydride complexes of mthenium35’and osmium35bshow the36Scheme 2.3Re2iochemistryPh3+ o_ /M— H M H cx—insertionbasic hydrideo_ 8 HPhCCHH3+ 1M—H Mph j1—insertionacidic hydride /HStereochemistryRM—H MR cis inseion1HRCCRM—H MkB trans insertionRsame regioselectivity. If a simple acidic versus basic argument is used, then complex 2.8 is theexpected product for the insertion reaction based on the reactivity of the intermediate alkylhydride presented so far. That is, the hydride acts in a basic manner, adding to the more“positive” end of the unsaturation, as observed in the reactions of the hydride with carbonyl,imine, and nitrile groups. When exposed to phenylacetylene, the hydridic hydride adds to themore positively charged acetylenic carbon, yielding the ct-insertion product, 2.8.The stereochemistry of acetylene insertions into metal-hydride bonds is also interesting, andoften provides evidence for the mechanism of the reaction. Cis insertion as shown in Scheme 2.3is expected if the acetylene coordinates to the metal before undergoing migratory insertion. In37the reaction to form complex 2.8, labeling studies using deuterated phenylacetylene as thereagent or D2 for hydrogenation could give evidence for the stereoselectivity of the insertionprocess. The low yield of the reaction limited the feasibility of these studies, however, and theywere not performed.Complex 2.8 was subjected to an X-ray crystallographic analysis to confirm its molecularstructure. The ORTEP plot from this analysis is shown in Figure 2.3 along with some selectedbond lengths and angles. The structure clearly shows the regioselectivity of the acetyleneinsertion, such that the phenyl group is attached to the alpha carbon of the alkenyl ligand. Thebond angles about the alpha carbon are not symmetric. The small W-C-C angle of 98.6(7) 0 issomewhat surprising, but may be interpreted as a bending to relieve steric interactions betweenthe nitrosyl and the phenyl ligand. The C=C bond has a normal double bond length of1.336(12) A, and is located above the gap between the two tungsten-carbon bonds of the alkylligands. This site is where Lewis bases tend to coordinate, and where the LUMO of the metal ispresumably located, in similar dialkyl complexes (see Section 2.3.2). The question then arises asto whether the alkenyl ligand can be viewed as anr2-ligand because of this bending towards asite where it could possibly coordinate. Comparisons of the structural parameters of complex 2.8to those of other structurally characterized compounds containing2-alkenyl ligands can easilybe made. When this is performed, however, it becomes clear that complex 2.8 does not contain atrue‘t-2-bonding interaction between the tungsten and the alkenyl double bond. Firstly, for anumber of tungsten complexes containing such ligands, the C-Cp bond length is longer(1 .41-1.45 A)36 than the same distance in 2.8 (1.34 A), and tends to be closer in length to asingle carbon-carbon bond length than a double bond. Another diagnostic feature of truer2-alkenyl ligands is the short M-C bond length that reflects some carbene character in thislinkage, and a C that is within bonding distance of the metal (Figure 2.4). Comparison of thesebond lengths in complex 2.8 to reported values36 reveals that the M-C bond of 2.066(7) A islonger than knownr2-alkenyl lengths of 1.89 - 1.96 A, and the M-C length is longer at2.624(9) A than literature values of 2.18 - 2.33 A.38Figure 2.3. View of the solid state molecular structure ofCp*W(NO)(CH2SiMe3)(CPh=C(2.8), including selected bond lengths and angles (with esd’s in parentheses).C(11)C( IC( 14) C(l3)C(6)wC(8) C(19) C(22)Si 2O)C(2C(9)C(7)C(25)Bond Lengths (A) Bond Angles (deg)W.-C(19) 2.066(7)W - C(20) 2.624(9)C(19) - C(20) 1.336(12)W - C(19) - C(20) 98.6(7)W-C(19)-C(21) 136.2(7)C(6)- W - C(19) I 14.2(4)39Figure 2.4. Different bonding \___—M—ffdescriptions of2-a1kenyl ligands.NMR spectroscopic evidence, however, cannot definitely rule out an ‘q2-like interaction forthe alkenyl ligand of 2.8 in solution. Since2-alkenyl ligands can also be described asmetallacyclopropenes that have carbenoid character in the alpha carbons, they exhibit downfieldshifts of these carbon resonances in the 13C NMR spectrum. Templeton and co-workers havenoticed that for most tungsten2-alkenyl compounds, the chemical shift of the alpha carbonsfalls within the range of approximately 230 - 270 ppm, and the beta carbon signals are foundmuch further upfield between 20 and 40 ppm.37’8 Complex 2.8 displays the analogousresonances at 228 and 83 ppm, respectively. These values can be compared to knownq1-alkenylcomplexes, that usually display C resonances around 160 ppm and C resonances around 130ppm39’4° (It should be noted here that a report of a transition-metal complex containing analkenyl ligand analogous to the one in 2.8 could not be found in the literature for a directcomparison, although similar ligands containing a phenyl group bound to C were used for thecomparisons above).A careful look at the NMR data for complex 2.8 reveals some interesting tungsten couplingthat may imply an interaction between the =CH2 group of the alkenyl and the metal center. The13 NMR spectrum of the complex in C6D is shown in Figure 2.5. 184W satellites are clearlyvisible on the alpha carbon ( = 99 Hz) and also on the beta carbon (‘wc = 13 Hz) signals.It is unusual to see this latter two-bond coupling to tungsten, presumably because of the smallcoupling constant. No such coupling has been evident in any complexes containing beta carbonsthat we have yet characterized, and therefore we have no basis for comparison.The 1H NMR spectrum of complex 2.8 in Figure 2.6 also shows some interesting couplingpatterns. Both of the alkenyl protons are weakly (J= 1.4 and 1.1 Hz) coupled to one of the40Jwcpl3HzJwca=99Hz////;a___I H II III ill [ I I iII IIij Iii II Ij 11111 1111111 I ‘iii ii Iii jill I IIIj 11111 11111220 200 180 160 140 120 100 80 60 40 20 PPMFigure 2.5. ‘3C{’H} NIvER spectrum of complex 2.8 in C6D. Insets are enlargements of theC and C alkenyl resonances.methylene protons of the trimethylsilylmethyl ligand. This is very surprising, because thiscoupling is through five bonds and four atoms. When this coupling is removed from the alkenylspectrum by decoupling the methylene signal, the alkenyl signals appear as they do in the insert ofFigure 2.6. There are tungsten sateffites observable on both of the signals, and one couplingconstant is observed to be much greater than the other = 10 and 5.1 Hz). This difference incoupling to the tungsten is not surprising, given that the protons are inequivalent and in differentgeometries, but the fact that they show a three-bond coupling to the tungsten, and a five-bondcoupling to the CH2SIMe3proton is remarkable.One explanation for the odd NMR couplings is that the tungsten is interacting somewhatwith the alkenyl ligand in something more than an1-fashion. The nature of this interaction isunknown, but probably involves electron donation from the it-electrons in the alkenyl double41WH = 10 HZ 3WH 5.1 Hz8 7 6 5 4 3 2 1F’PMOFigure 2.6. ‘H NMR spectrum of complex 2.8 in C6D. Insets show enlargements of thealkenyl proton resonances when decoupled from the CH2S1Me3methylene proton at 0.20 ppm.bond to the unsaturated metal. This would help explain the bending of the alkenyl ligand towardthe metal center that is observed in the solid-state structure, and the positioning of the doublebond over the site where the tungsten LUMO is situated. A variable-temperature 1H NMR studyof this complex did not show any evidence for a more pronounced interaction of the alkenyl CH2group with the tungsten center at lower temperatures.The evidence thus points to an interaction of the unsaturated tungsten center with theelectron density of the alkenyl ligand. The interaction is not strong enough to label the alkenyl asa classicrj2-alkenyl ligand, as evidenced by the molecular structure in which the beta carbon isnot within bonding distance of the tungsten center. An interaction, albeit weak, can beinterpreted from the NMR data, which shows a coupling of the =CH2 to the tungsten center andalso to the methylene proton of the trimethylsilylmethyl ligand.422.3.5 Reaction with PPh3An attempt was made to trap the intermediate hydride generated upon hydrogenation ofCp*W(NO)(CH2SiMe3)with triphenyiphosphine so that the reactivity of the 18-electronphosphine-trapped alkyl hydride could be compared to previously prepared analogouscompounds.2 Triphenyiphosphine was chosen because it was hoped that the ligand’s larger stericbulk would promote the reductive elimination of alkane from the complex at lower temperaturesthan needed to induce the same reactivity for the trimethylphosphine analogue. When thehydrogenation of Cp*W(NO)(CH2SiMe3)is performed in the presence of one equivalent ofPPh3, a yellow precipitate forms that can be isolated easily by collection on a fit. Analysis ofthis precipitate reveals that it is not the expected phosphine-trapped alkyl hydride. Instead, theH2, PPh32.7N’ I ‘R -2S1Me4o RR=CH2SiMe3compound has been characterized as the product of alkane elimination and ortho-metalation ofthe phosphine ligand (equation 2.7). The product, Cp*W(NO)(H)(rI2PPh5H4)(2.9), containsa hydride ligand and an ortho-metalated triphenyiphosphine ligand in place of the two originalalkyls. The JR spectrum clearly shows a metal-hydride stretching band at 1929 cm’, and the 1HNMR spectrum exhibits the hydride resonance as a doublet at 6 3.99 ppm(2Jp = 10 Hz). 183Wcoupling, in the form of small satellite peaks(1jwH = 101 Hz), are also observable on the lattersignal, which is indicative of a proton directly bonded to the tungsten. The ortho-metalatednature of the phosphine ligand is evident from the large number of peaks in the aryl region ofboth the 1H and 13C NMR spectra, and by comparison of these spectra to a similar compound,CpW(NO)H[ri2-P(OPh)(064)j.Complex 2.9 is likely formed via a phosphine-trapped alkyl2.943hydride intermediate, which loses alkane and orthometalates the phosphine ligand to form theisolated product. A more detailed discussion of the proposed mechanism of formation ofcomplex 2.9, together with some of its reactivity, is presented in Chapter 5.2.4 ConclusionsIt has been demonstrated by the work presented in this Chapter that hydrogenation ofCp*W(NO)(CH2SiMe3)at low pressures of hydrogen results in the hydrogenolysis of one ofthe tungsten-trimethylsilylmethyl bonds to form the 16-electron alkyl hydride intermediate,Cp*W(NO)(CH2SiMe3)H. However, this product is unstable under the reaction conditions andcannot be isolated as such, and has not been observed spectroscopically.The chemistry of the transient alkyl hydride has been investigated by way of reaction with avariety ofunsaturated substrates. The products of these reactions testify that polar unsaturatedsubstrates undergo insertion into the tungsten-hydride bond, presumably via initial coordinationof the substrate to the coordinatively unsaturated metal center. Nearly all of the reactions havebeen found to be quantitative, even in dilute solution, indicating that the intermediate hydride hasa definite lifetime in solution.The hydride shows typical hydndic behavior towards polar unsaturated bonds, as indicatedby the regioselectivity of the insertion reactions, and does not show any desire to insert non-polaror only slightly polar substrates such as dienes, 1 -alkynes, or diphenylacetylene. When ligandsthat have the potential of functioning in a bidentate fashion coordinate to the intermediatehydride, the molecule eliminates alkane and the ligand becomes bidentate. This cooperativityeffect has been illustrated in the diene and triphenyiphosphine reactions.A variety of new products has been synthesized using the hydrogenation methodologypresented in this Chapter, some of which are unattainable by other synthetic means. Study ofthese new products has provided some interesting chemistry, some of which is presented in theremainder of this Thesis.442.5 References and Notes(1) Legzdins, P.; Rettig, S. 3.; Sanchez, L. Organometallics 1988, 7, 2394.(2) Legzdins, P.; Martin, J. T.; Einstein, F. W. B.; Jones, R. H. Organometallics 1987, 6,1826.(3) Shriver, D. F.; Drezdzon, M. A. The Manipulation ofAir-Sensitive Compounds, 2nd ed.;Wiley-Interscience: New York, NY, 1986.(4) Dryden, N. H.; Legzdins, P.; Batchelor, R. J.; Einstein, F. W. B. Organometallics 1991,10, 2077.(5) Legzdins, P.; Veitheer, J. E. Acc. Chem. Res. 1993, 26, 41.(6) Perrin, D. D.; Armarego, W. L. F.; Perrin, D. R. Pur,f1cation ofLaboratory Chemicals,3rd ed., Pergamon Press: Oxford, 1988.(7) Kilner, M.; Midcalf C. Chem. Comm. 1970, 552.(8) Alelyunas, Y. W.; Jordan, R. F.; Echols, S. F.; Borkowsky, S. L.; Bradley, P. K.Organometaiics 1991, 10, 1406.(9) (a) Kilner, M.; Midcalf C. .J Chem. Soc., Dalton Trans. 1974, 1620. (b) Cetinkaya, B.;Lappert, M. F.; McMeeking, 3. J Chem. Soc., Dalton Trans. 1973, 1975.(10) (a) Bercaw, J. E.; Davies, D. L; Wolczanski, P. T. Organometallics 1986, 5, 443. (b)Bochmann, M.; Wilson, L. M. Organometallics 1988, 7, 1148. (c) den Haan, K. H.;Luinstra, G. A.; Meetsma, A.; Teuben, 3. H. Organometallics 1987, 6, 1515. (c)Richeson, D. S.; Mitchell, I. F.; Theopold, K. H. Organometallics 1989, 8, 2570.(11) (a) Evans, W. 3.; Meadows, J. H.; Hunter, W. E.; Atwood, J. L. J Am. Chem. Soc. 1984,106, 1291. (b) Churchill, M. R.; Wasserman, H. J.; Belmonte, P. A.; Schrock, R. R.Organometallics 1982, 1, 559. (c) Mays, M. 3.; Prest, D. W.; Raithby, P. R. J Chem. Soc.,Chem. Commun. 1980, 171. (d) Fromberg, W.; Erker, G. .1 Organomet. Chem. 1985, 280,343.45(12) Daniel, T.; Muller, M.; Werner, H. Inorg. them. 1991, 30, 3118.(13) Chatt, J.; Dosser, R. J.; Leigh, G. J. Chem. Comm. 1972, 1243.(14) Pombeiro, A. J. L.; Hughes, D. L.; Richards, R. L. I Chem. Soc., Chem. Commun. 1988,1052.(15) Band, E.; Pretzer, W. R.; Thomas, M. G.; Muetterties, E. L. I Am. Chem. Soc. 1977, 99,7380.(16) (a) Grey, R. A.; Pez, G. P.; Wallo, A. I Am. Chem. Soc. 1981, 103, 7536. (b) Yoshida,T.; Okano, T.; Otsuka, S. I Chem. Soc., Chem. Commun. 1979, 870.(17) Bernhardt, W.; Vahrenkamp, H. Angew. Chem. mt. Ed Engi. 1984, 23, 381.(18) Feng, S. G.; Templeton, J. L. Organometallics 1992, 11, 1295.(19) Doxsee, K. M.; Farahi, J. B. I Am. Chem. Soc. 1988, 110, 7239.(20) Legzdins, P.; Lundmark, P. L.; Rettig, S. J. Organometallics 1993, 12, 3545.(21) Dryden, N. H.; Legzdins, P.; Trotter, J.; Yee, V. C. Organometallics 1991, 10, 2857.(22) Legzdins, P.; Rettig, S. J.; Sanchez, L.; Bursten, B. E.; Gatter, M. G. I Am. Chem. Soc.1985, 107, 1411.(23) Dryden, N. H.; Legzdins, P.; Batchelor, R. 3.; Einstein, F. W. B. Organometallics 1991,10, 2077.(24) Debad, 3. D.; Legzdins, P.; Rettig, S. J.; Veltheer, 3. E. Organometallics 1993, 12, 2714.(25) Debad, 3. D.; Legzdins, P.; Batchelor, R. 3.; Einstein, W. F. B. Organometallics 1991, 1],6.(26) Legzdins, P.; Rettig, S. J.; Ross, K. 3. Organometallics 1993, 12, 2103.(27) Christensen, N. 3.; Legzdins, P.; Einstein, F. W. B.; Jones, R. H. Organometallics 1991,10, 3070.(28) Footnote 27 and references therein.46(29) Debad, J. D.; Legzdins, P.; Young, M. A.; Batchelor, R. J.; Einstein, F. W. B. .1 Am.Chem. Soc. 1993, 115, 2051.(30) van der Zeijden, A. A. H.; Bosch, H. W.; Berke, H. Organometallics 1992, 11, 563.(31) Bianchini, C.; Innocenti, Paolo, Masi, D.; Meli, A.; Sabat, M. Organometallics 1986, 5,72.(32) Labonova, I. A.; Zdanovich, V. I.; Kolobova, N. E.; Kalinin, V. N. Metallorg. Khim.1990, 3, 916.(33) Esteruelas, M. A.; Fernando, J. L.; Lopez, J. A.; Oro, L. A.; Schiunken, C.; Valero, C.;Werner, H. Organometallics 1992, 11, 2034.(34) McGrady, N. D.; McDade, C.; Bercaw, 3. E. Organometallic Compounds. Synthesis,Structure, andReactivity; B. L. Shapiro, ed.; Texas A + M University Press, CollegeStation, Texas, 1983, Vol. 1.(35) (a) Romero, A.; Santos, A.; Vegas, A. Organometaiics 1988, 7, 1988. Romero, A.;Santos, A.; Lopez, J.; Echavarren, A. M. .1 Organomet. Chem. 1990, 391, 219. (b)Werner, H.; Esteruelas, M. A.; Otto, H. Organometallics 1985, 5, 2295.(36) Feng, S. G.; Gamble, A. S.; Templeton, J. L. Organometallics 1989, 8, 2024.(37) Feng, S. G.; Templeton, 3. L. Organometallics 1992, 11, 2168.(38) van der Zeijden, A. A. H.; Bosch, H. Q.; Berke, H. Organometallics 1992, 11, 563.(39) (a) Herberich, G. E.; Mayer, H. Organometallics 1990, 9, 2655. (b) Werner, H.;Esteruelas, M. A.; Otto, H. Organometallics 1986, 5, 2295.(40) (a) Reger, D. L.; Belmore, K. A.; Mintz, E.; Charles, N. G.; Griffith, E. A. H.; Amma, E.Organometallics 1983, 2, 101. (b) Reger, D. L.; Mintz, E.; Lebioda, L. .1 Am. Chem.Soc. 1986, 108, 1940.47CHAPTER 3Synthesis and Chemistry of Cp*W(NO)(R)Cl Complexes.3.1 Introduction 473.2 Experimental Procedures 483.3 Results and Discussion 613.4 Summary and Future Work 843.5 References and Notes 873.1 IntroductionIn the past few years, work in our laboratories has revolved around establishing thereactivity of symmetric dialkyl and diaryl complexes of the type Cp’M(NO)(R)2[M = Mo, WI.We have compiled a large data base of the characteristic chemistry of these 16-electroncompounds, dealing primarily with the reactivity of the metal-carbon bonds of the alkyl and arylgroups (see Section 2.1 for examples). The synthesis of the dialky and diaryl compounds is nowconsidered routine in our laboratories and involves the alkylation of the dichioride precursorsCp’M(NO)C12with Grignard or dialkyllarylmagnesium reagents. In all the years of working withthese complexes, however, we have not been able to find a general synthetic route tomonoalkylated complexes of the type Cp’M(NO)(R)X [X halide; R = alky, aryl]. The onlyexamples of such compounds that we have prepared are those with R benzyl, and they areproduced by reaction of the bis-(benzyl) complex with HC1.1 All attempts to repeat this reactionwith compounds containing other alkyl or aryl ligands have failed. So too have the attempts ofalkylating the dihalide precursors, Cp’M(NO)(X)2with one equivalent of Grignard reagents.48The outcome of such reactions is the formation of one-half an equivalent of the dialkylatedproduct, with one-half an equivalent of starting dihalide remaining unreacted.1In this Chapter, a new, general route to the alkyl and aryl chlorides Cp*W(NO)(R)Cl ispresented. The isolation of these complexes allow us to explore the chemistry of a number ofnew compounds that were previously unattainable, especially in the field ofunsymmetric speciesof the type Cp*W(NO)(R)(X) [X = amide, alkoxide, alkyl, aryl, etc.]. This chapter is concernedmainly with the synthesis and reactivity of the alkyl and aryl chloride compounds and some oftheir 18-electron adducts. The chemistry of the other unsymmetric complexes is presented inlater Chapters.3.2 Experimental Procedures3.2.1 MethodsThe synthetic methodologies employed throughout this thesis are described in detail inSection 2.2.1.Electrochemical methods used in this Chapter are described in full detail elsewhere.23.2.2 ReagentsThe Mg.X(dioxane) reagents [R = Me, CH2Me3CH2SiMe3,p-tolyl, phenyl, o-tolyl]were prepared by published procedures,2and their potency determined by titration of ahydrolyzed sample before use.3 PMe3 (Strem) was vacuum transferred from Nalbenzophenone.CO (Matheson), CNCMe3AgBF4,LDA, PhSSPh (Aldrich), and zinc powder (Fisher) wereused as received. Pyridine and acetone (BDH) were doubly distilled from CaH2. THF-d8(Aldrich) was used as received.49Filtrations were performed through Celite 545 diatomaceous earth (Fisher), Florisil (60-100mesh, Fisher), or alumina (80 - 200 mesh, Fisher neutral, Brockman activity I) that had beenoven-dried and cooled in vacuo.3.2.3 SynthesisIsolated yields, physical properties, and spectroscopic data for all complexes are listed inTables 3.1 - 3.3. The alkyl and aryl chloride complexes prepared in this Chapter weresynthesized using similar methodology. Therefore, only one representative example is describedin detail to outline the general procedure. Preparation of Cp*W(NO)(CEI2SiMe3)C1(3.1) fromCp*W(NO)(CHSiMe)(NCHM ) (2.1)Addition of 2.5 equivalents of HC1 in Et20 (2.4 mL, Li M) to a dilute solution of complex2.1 in Et20 (0.41 g, 0.86 mmol, 20 mL Et20) resulted in the formation of a deep blue solutionand a white precipitate. The solvent was removed under vacuum and the solid extracted withpentane (40 mL). The extract was filtered through Celite (1 x 2 cm) supported on a fit, and theresulting solution was then concentrated. Cooling the solution to -30° C afforded deep bluecrystals of complex 3.1 (0.35 g, 88% yield). Preparation of Cp*W(NO)(R)C1 Complexes [R = CH2SiMe3(3.1), CH2Me3(3.2),o-tolyl (3.3), Ph (3.4)]Cp*W(NO)(CH2SiMe)Cl(3.1) and Cp*W(NO)(CH2Me3)Cl(3.2) were prepared in asimilar manner. The synthesis of 3.2 is described as a representative example. THF (20 mL)was vacuum transferred onto a mixture of Cp*W(NO)(CI)2(2.10 g, 5.00 mmol) and(Me3CCH2)Mg.X(dioxane) (0.635 g, 2.50 mmol) at -196 °C. The stirred reaction mixturewas allowed to warm until all the solids dissolved and the solution became purple. The50solvent was removed in vacuo, and the resulting residue was left under vacuum for one hour.The purple solid remaining was extracted with pentane (100 mL), and the extract was filteredthrough Celite (2 x 4 cm) supported on a fit. The filtrate was concentrated in vacuo andtransferred to a freezer (-30 °C). This cooling induced the formation of violet needles thatwere isolated from the mother liquor by cannulation and were dried in vacuo. Additionalproduct was obtained by further concentrating and cooling the supernatant solution.Cp*W(NO)(otolyl)Cl (3.3) and Cp*W(NO)(Ph)Cl (3.4) were prepared as described inthe preceding paragraph except that CH21 instead of pentane was used as the extractionsolvent. Preparation of Cp*W(NO)(CH2Me3) toIyI(3.5)Cp*W(NO)(CHMe)Cl(2.0 mmol) was generated as above in cold THF. The solventwas removed under vacuum, and the residue was kept under dynamic vacuum for two hours.Et20 (40 mL) was then added, and the resulting solution was cooled to -68 °C. A solution of(p-tolyl)MgX(dioxane) (0.35 g, 2.0 mmol) in Et20 (15 mL) was slowly added via cannula. Thesolution was then allowed to warm slowly to room temperature while being stirred, during whichtime it turned from purple to red. Approximately one-half of the solvent was removed undervacuum, and the remaining solution was filtered through Celite (2 x 3 cm) supported on a fit.The solvent was removed from the filtrate under vacuum, and the remaining red oil was dissolvedin a minimum amount of pentane (10 mL). Cooling of this solution caused the deposition ofcomplex 3.5 as maroon rosettes (0.51 g, 49% yield). Preparation of Cp*W(NO)(CH2SiMe3)NHCMe (3.6)An excess ofNH2CMe3(—1 mL) was vacuum transferred onto a frozen solution ofCp*W(NO)(CH2SiMe3)Cl(0.15 g, 0.32 mmol) inEt2O (15 mL) at -198 °C. The stirred solutionwas allowed to warm to room temperature, during which time its color changed from blue toyellow and a white precipitate formed. The solution was then taken to dryness under vacuum,51and the remaining yellow powder was extracted with pentane (20 mL) and filtered through Celite(1 x 2 cm) supported on a fit. Cooling of the solution at -30 °C overnight induced the formationof rosette-shaped clumps of crystals (0.11 g, 68% yield). This solid was recrystallized fromhexanes to obtain analytically pure Preparation of Cp*W(NO)(CH2Me)(PMCI(3.7)Onto a frozen solution of Cp*W(NO)(CHMe3)Cl(0.075 g, 0.165 mmol) in pentane(10 mL) at -196 °C was vacuum transferred an excess ofPMe3. Upon warming, the mixturebecame yellow and a yellow precipitate formed. The reaction mixture was taken to dryness invacuo, and the residue was then redissolved in a minimum ofEt20. Slow cooling of the ethersolution to -30 °C yielded orange prisms of 3.7. The crystals were isolated by removing themother liquor via cannulation. A second fraction of crystals was obtained by additionalconcentration and cooling of the mother liquor (0.068 g, 78% yield). Preparation of Cp*W(NO)(CEI2Me3)(py)CI(3.8)A sample of Cp*W(NO)(CHMe)Cl(0.200 g, 0.440 mmol) in Et20 (10 mL) was treatedwith pyridine (0.50 mL, excess) at room temperature with no noticeable change in color. Uponcooling to -30 °C, the purple solution turned orange-red and the desired product began tocrystallize. The reaction mixture was stored in a freezer (-30 °C) overnight to complete thecrystallization. 3.8 (0.12 g, 53% yield) was isolated as orange blocks which were dried in vacuoat low temperature (-10 °C). Prolonged drying at room temperature caused the complex to losepyridine slowly and reform Cp*W(NO)(CH2Me3)Cl.523.2.3.7 Reaction of Cp*W(NO)(CH2Me3)(PMCI(3.7) with LDA: Preparation of(q5,n1-Me4CH)W(NO)(CHMeP (3.9)Into a rapidly stirred solution ofCp*W(NO)(CH2Me3)(PMCl(0.500 g, 0.940 mmol) inTFTF (20 mL) (vNO 1566 cm-l) was cannulated a slurry of LDA (0.100 g, 0.93 5 mmol) in THF(10 mL). The mixture was stirred for 2 h at room temperature during which time the reactionmixture darkened from orange to red-orange (vNO 1557 cml). The solvent was removed fromthe final reaction mixture in vacuo, and the residue was extracted with Et20 (5 x 20 mL). Thecombined extracts were filtered through alumina (2 x 3 cm) supported on a flit. The yellowfiltrate was concentrated and maintained at -30 °C for 1 week to induce the deposition of 3.9 as ayellow powder. More product was isolated from the mother liquor by repeated concentrationsand coolings (0.21 g, 46% yield). X-ray quality crystals of 3.9 were obtained by recrystallizationof the yellow powder from benzene/hexanes (1:1). Preparation ofCp*W(NO)(q2{O}CBMe3)C1(3.10)A sample of Cp*W(NO)(CHMe)C1(0.100 g, 0.220 mmol) was dissolved in pentane(10 mL), and an atmosphere of CO was then introduced into the flask. Over the course of 2 h, ayellow precipitate formed. The supernatant solution was then removed by cannulation anddiscarded, and the precipitate was dissolved in a minimum ofEt20 (20 mL). Cooling the Et20solution in a freezer (-30 °C) for several days induced the precipitation of yellow,microcrystalline 3.10 (0.076 g, 71% yield) that was isolated by removing the mother liquor witha pipette. Preparation of Cp*W(NO)(12_C{NCMe3} II2CMe)C1(3.11)In a glovebox, a sample ofCp*W(NO)(CHMe3)Cl(0.227 g, 0.500 mmol), contained in a100-mL reaction vessel, was dissolved in Et20 (10 mL). To this solution was added CNCMe3(0. 1 mL, 0.9 mmol) via a pipette. The solution instantly changed from purple to amber.Within seconds the amber solution began to turn bright yellow and a yellow powder precipitated.53The reaction vessel was sealed, removed from the drybox, and reattached to a vacuum line. Thesolvent and excess isocyanide were removed in vacuo. The yellow residue was then dissolved ina minimum ofEt20 (80 mL) and placed in a freezer (-30 °C). Overnight, the desired productcrystallized (0.23 g, 85% yield) as microcrystalline yellow needles. Preparation of Cp*W(NO)(qlC{O}CHMe)(PMCI(3.12)To a rapidly stirred solution ofCp*W(NO)(fl2{O}CHMe3)Cl(0.120 g, 0.248 mmol) inEt20 (20 mL) was introduced excess PMe3 (1 atm) whereupon an immediate color change fromyellow to orange occurred. The final orange solution was taken to dryness in vacuo. Theremaining residue was extracted with Et20 (3 x 30 mL) and filtered through Celite (2 x 2 cm)supported on a fit. Cooling of the filtrate to -30 °C overnight afforded orange crystals ofcomplex 3.12 (0.11 g, 76% yield). Preparation of jCp*W(NO)(CH2Me3)(NCM )1BF4(3.13)A solution ofCp*W(NO)(CHMe)Cl(0.400 g, 0.880 mmol) in MeCN (10 mL) at -10 °Cwas cannulated into a stirred slurry of AgBF4 (0.170 g, 0.870 mmol) in MeCN (5 mL). Themixture became orange immediately upon mixing. The reaction mixture was filtered throughCelite (2 x 2 cm), and the filtrate was concentrated in vacuo to 5 mL. Et20 (5 mL) was addedto the MeCN solution, and the resulting mixture was placed in a freezer overnight to inducecrystallization. Orange crystals of 3.13 were isolated by removing the mother liquor with acannula (0.34 g, 66% yield). Preparation of [Cp*W(NO)CH2Me3Iu-NICp*W(O)CH2Me3](3.14)Cp*W(NO)(CHMe)Cl(0.15 g, 0.33 mmol) and excess Zn powder (0.5 g) were weighedinto a Schienk tube containing a magnetic stirbar, and THF (-40 mL) was then vacuumtransferred onto the solids. The vessel was warmed to room temperature, and the reaction54mixture was stirred for 5 h, during which time its color changed from purple to green-brown.The solvent was removed in vacuo, and the residue was extracted with Et20 (2 x 10 mL). Thecombined extracts were filtered through Florisil (2 x 2 cm) supported on a flit, and the columnwas washed with Et20 until the washings were colorless. The solvent was removed from thecombined filtrates in vacuo, and the brown powder remaining was dissolved in a minimum ofhexanes. Cooling of the hexanes solution at -30 °C overnight induced the deposition of green-brown crystals of complex 3.14. These crystals were recrystallized from hexanes to obtainanalytically pure product (0.80 g, 58% yield based on tungsten). Preparation of [Cp*W(NO)CH2Me3]it-N ICp*W(O)CII (3.15)Cp*W(NO)(CHMe)Cl(0.45 g, 0.99 mmol), Cp*W(NO)(Cl)2(0.21 g, 0.50 mmol), andexcess Zn powder (0.5 g) were weighed into a Schienk tube containing a magnetic stirbar. THF(—10 mL) was then vacuum transferred onto the solids. The vessel was warmed to roomtemperature, and the solution then stirred for 20 mm, during which time the solution changedfrom purple to brown. Solvent was removed in vacuo, the residue was extracted with Et20(2 x 10 mL), and the extracts were filtered through a column of Florisil (2 x 2 cm) supported ona fit. The column was washed with Et20 until the washings were colorless. The solvent wasremoved from the combined filtrates under reduced pressure, and the tan-colored powderremaining was dissolved in a minimum of hexanes. Cooling of the hexanes solution for 3 days at-30 °C resulted in the deposition of brown crystals of complex 3.15 (0.10 g, 24% yield based onCp*W(NO)Cl2). Single crystals of 3.15 suitable for X-ray diffraction studies were obtained byslow recrystallization of this material from hexanes. Preparation of Cp*W(NO)(CH2Me3)SPh(3.16)Cp*W(NO)(CHMe)C1(0.040 g, 0.088 mmol), PhSSPh (0.020 g, 0.09 mmol) and excessZn powder (0.3 g) were weighed into a small Schienk tube containing a magnetic stirbar. THF(5 mL) was added, and the solution was stirred overnight. The solvent was then removed under55vacuum, and the residue was extracted withEt20/hexanes (4 mL of 1:1 mixture). The violetcolored solution was filtered through a small pad of Celite (2 x 0.5 cm) and then placed in afreezer overnight at -30 °C. Complex 3.16 crystallized as small, violet crystals which wereisolated by pipetting off the supernatant and were then dried under vacuum (0.03 5 g, 75% yield).563.2.4 Characterization Data for Complexes 3.1 - 3.16Table 3.1. Numbering Scheme, Color, Yield, and Elemental Analysis Data for Complexes3.1 - 3.16.compd color anal. found (calcd)complex no. (yield, %) C H NCp*W(NO)(CH2SiMe3)Cl 3.1 blue (62) 35.64(35.97) 5.56(5.59) 2.97(2.92)Cp*W(NO)(CHMe) l 3.2 purple (63) 39.25(39.54) 5.80(5.75) 2.89(3.07)Cp*W(NO)(otolyl)Cl 3.3 violet (40) 42.92(43.20) 4.66(4.90) 2.95(2.86)Cp*W(NO)(Ph)Cl 3.4 blue (49) 41.63(42.10) 4.37(4.47) 3.03(3.10)Cp*W(NO)(CH2Me3)(ptolyl) 3.5 purple (49) 51.72(51.67) 6.52(6.50) 2.63(2.74)Cp*W(NO)(CHSi e)NHCMe 3.6 yellow (68) 42.80(42.52) 7.32(7.14) 5.30(5.51)Cp*W(NO)(CHMe)(PMCl 3.7 orange (78) 40.91(40.66) 6.83(6.63) 2.56(2.63)Cp*W(NO)(CH2Me3)(py)Cl 3.8 orange (53) 44.82(44.92) 5.92(5.84) 5.24(5.24)(ri5, j‘-CMe4CH)W(NO) 3.9 orange (46) 43.33(43.65) 6.90(6.92) 2.79(2.83)(CH2CMe3)PCp*W(NO) 3.10 yellow (71) 39.91(39.73) 5.60(5.42) 2.86(2.90)12-C{O}CHMe3)ClCp*W(NO) 3.11 yellow (85) 44.41(44.58) 6.44(6.55) 5.10(5.20)12-C{NCMe3}CH2CMe3) lCp*W(NO) 3.12 orange (76) 40.86(40.76) 6.37(6.30) 2.46(2.50)(ii‘-C{O}CHCMe)(PMl[Cp*W(NO)(CH2Me3 3.13 orange (66) 38.56(38.73) 5.59(5.48) 7.09(7.13)(NCMe)]BF4[Cp*W(NO)CH2Me3.t-N 3.14 green (58) 42.59(42.87) 6.44(6.24) 3.29(3.33)[Cp*W(O)CHMeJ3.15 green (24) -- -- --[Cp*W(O)Cl]Cp*W(NO)(CH2Me3)SPh 3.16 violet (75) 47.56(47.64) 5.87(5.90) 2.68(2.65)57Table 3.2. Selected Mass Spectral and Infrared Data for Complexes 3.1 - 3.16.compd MS tempb ER (Nujol mull)no. (°C) VNO other bands3.1 471 [p+] 120 15993.2 455 [Pj 100 15823.3 475 [Pj 100 15883.4 461 [Pj 200 15913.5 511 [Pj 120 15553.6 508 [Pj 100 1520 3220 (VNH)3.7 455 [P-PMe3] 180 15533.8 455 [P- NC5H] 120 15493.9 495 [Pj 180 1524 953 (m,PMe)3.10 483 [Pj 180 1584 1564 (s, VCO)455 [P-CO]3.11 538 [Pi 200 1537 l68O(m,vCN)508 [P- NO]3.12 483 [P - PMe3] 180 1580 1634 (s, VCO)3.13 589 [p+]C 1589 2319, 2291 (VCN)1060 (br, BF4)3.14 840 [Pj 200 1545 9628913.15 804 [Pj 120 1549 9628893.16 529 [P] 150 1580a m/z values are for the highest intensity peak of the calculated isotopic cluster, i.e. 1’W.b Probe temperatures. ‘ FAB-MS (matrix: 3-nitrobenzyl alcohol).58Table 3.3. NMR Data for Complexes 3.1 - 316 (CD).compd 111 NMR 13C{1} NMRno. o3.1 2.00 (m, 2H, CH2) 113.48 (C5Me)1.94 (s, 15H, C5Me) 63.62 (CR2)0.12 (s, 9H, SiMe3) 10.30 (C5A’fe)1.38 (SiMe3)3.1 3.42 (d, 2J= 12.0 Hz, 1H, CR2) b1.61 (s, 15H, C5Me)1.37 (s, 9H, CMe3)-0.03 (d, HH = 12.0 Hz, 1H, CR2)3.2 2.92 (d, 2HH = 11.1 Hz, 1H, CR2) 112.63 (C5Me)1.89 (s, 1511, C5Me) 96.59 (CH2)1.13 (s, 9H, CMe3) 39.02 (CMe3)-0.02 (d, 2HH = 11.1 Hz, 1H, CR2) 33.72 (CMe3)9.75 (CMe)3•3 7.50 (m, 2H, m/p-ArH) 130.67, 129.65, 122.31 (C1)7.37(m, 1H,m-ArH) 114.54 (C5Me)7.2 (br, 1H, 0-ArM) 25.41 (Me)2.74 (br, 3H, Me) 10.13 (C5Me)2.26 (s, 1511, C5Me)3.4 7.43 (m, 2H, m-ArH) 132.53, 128.84, 127.80 (C1)7.25 (m, 3H, o/p-ArH) 114.52 (C5Me)1.91 (s, 15H, C5Me) 10.07 (C5Me)3.5 7.68 (d, 3HH = 7.2 Hz, 2H, H00) 179.91 (C10)7.03 (d, HH = 7.2 Hz, 2H, Hme) 137.29 (Cpa)4.03 (d, HH = 11.7 Hz, IH, CR2) 137.21 (crtho)2.09 (s, 3H,p-Me) 128.69 (Cme)1.59 (s, 15H, C5Me) 118.50 (CR2)1.25 (s, 911, CMe3) 110.82 (C5Me)-1.87 (d, 2HH 11.7 Hz, 1H, CH2) 40.80 (CMe3)34.02 (CMe3)21.76 (p-Me)10.08 (C5Me)3.6 6.87 (br, 1H, NH) 109.79 (C5Me)1.58 (s, 15H, C5Me) 100.44 (NCMe31.34 (s, 9H, NCMe3) 33.48 (CMe3)0.44 (s, 9H, SiMe3) 17.87 (CR2)-0.12 (s, 2H, CR2) 9.81 (C5Me)3.06 (SiMe3)37C 1.83 (s, 15H, C5Me) 109.03 (C5Me)1.49 (d, 2HP = 9.9 Hz, 9H, PMe3) 61.70 (d, 2J = 9.8 Hz, CR2)1.26 (dd, HH = 11.4 Hz, HP = 4.2 Hz, 1H, 37.59 (CMe3)CH2) 34.92 (CMe3)1.04 (s, 9H, CMe3) 12.03 (d, 1J 30.1 Hz, PMe3)0.92 (d, HH 11.4 Hz, 1R, CH2) 9.97 (C5Me)593gd 9.02 (br s, 2H, py) b8.07 (brs, 1H,py)7.62(brs,2H,py)2.03 (br s, 1H, CH2)1.59 (br s, 1511, C5Me)1.09 (br s, 911, CMe3)0.26 (br s, 1H, CH2)39c,e 2.75 (vt, J= 3.6 Hz, 1H,C5Me4H2) 131.06 (C5Me4CH2)2.60 (dd, J 3.6, 13.2 Hz, 1H,C5Me4H2) 108.66 (CMeCH1.98 (d, HP 2.1 Hz, 3H,C5Me4H) 105.15 (d, J = 8.0 Hz,C5Me4H2)1.86 (s, 3H,C5Me4H) 103.28 (CMe4CH)1.74 (s, 3H, MeCH) 102.49 (C5MeCH1.53 (d,“HP = 8.4 Hz, 9H, PMe3) 47.51 (d, J, = 8.0 Hz,C5Me4H2)1.31 (d, HP = 3.3 Hz, 3H,C5Me4H) 37.23 (d, J, = 3.0 Hz, CHMe31.01 (s, 9H, CMe3) 36.74 (d, J = 8.1 Hz, CH2Me)0.87 (dd, J= 23.7, 13.8 Hz, 111, CH2Me3) 35.37 (CMe3)0.34 (dd, J= 13.8, 2.4 Hz, 1H, CHMe 16.54 (d, J = 30.0Hz, PMe3)11.24 (CMeCH)11.15 (CMeCH9.39 (CMeCH2)8.62 (CMe4H3.10 3.40 (d, 2HH = 15.0 Hz, JH, CH2) 288.00 (CO)2.97 (d, 2HH = 15.0 Hz, 1H, CH2) 111.56 (C5Me)2.01 (s, 1511, C5Me) 54.41 (CH2)1.09 (s, 9H, CMe3) 33.69 (CMe3)29.84 (CMe3)10.11 (C5Me)3.11 3.52 (d, 2HH = 12.6 Hz, 1H, CH2) 207.53 (CN)2.85 (d, 2Hfl = 12.6 Hz, 1H, CH2) 110.73 (C5Me)1.90 (s, 15H, C5Me) 63.44 (C!vIe3)1.48 (s, 9H, CNMe3) 41.66 (CH2)1.08 (s, 9H, CH2Me 32.99 (CMe3)30.10 (CMe3)29.77 (CMe3)10.12 (C5Me)3.12c,e 2.90 (d, 2HH = 16.7 Hz, 111, CH2) 264.6 (d, = 8.0 Hz, CO)2.44 (d, 2HH 16.7 Hz, 1H, CH2) 111.4 (C5Me)1.94 (s, 15H, C5Me) 69.82 (CH2)1.51 (d, 2HP = 9.9 Hz, 9H, PMe3) 54.36 (CMe3)0.90 (s, 9H, CMe3) 29.96 (CMe3)11.71 (d,‘1p 32.0 Hz,PMe3)10.23 (C5Me)3. 13f 3.02 (br, 3H, MeCN) MeCN not observed2.70 (br, 3H, MeCN) 111.64 (C5Me)2.22 (s, 15H, C5Me) 56.62 (CH2)1.53 (d, 2HH = 12.0 Hz, 1H, CH2) 37.79 (CMe3)142 (s, 9H, CMe3) 34.91 (CMe3)0.52 (d, HH = 12.0 Hz, 1H, CH2) 9.23 (C5Me)4.88 (br, MeCN)3.55 (br,MeCN)603.14e 2.00 (s, 15H, C5Me) 115.11, 111.34 (C5Me)1.93 (s, 15H, C5Me) 64.80, 49.99 (CH2)1.78 (d, 2HH = 14.1 Hz, 1H, CR2) 37.43, 33.99 (CMe3)1.49 (d, 2J = 13.2 Hz, 111, CR2) 33.60, 33.40 (CMe3)1.31 (d, 2J = 14.1 Hz, 1H, CR2) 10.94, 10.37 (C5Me)1.13 (s, 911, CMe3)1.01 (s, 911, CMe3)remaining CR2 signal obscured.3.15e 2.10 (s, 15H, C5Me) 118.32, 112.35 (CMe)1.92 (s, 15H, C5Me) 69.14 (‘wc = 107 Hz, CH2)1.73 (d, 2J = 14.1 Hz, 1H, CH2) 37.69 (CMe3)1.40 (d, 2HH = 14.1 Hz, 1H, CR2) 33.72 (CM’e3)1.12 (s, 9H, CMe3) 11.06, 10.16 (C]i6e5)3.16 7.42 (d, 3HH = 7 Hz, 2H, o-ArH) 142.8 (C1 )7.32 (t, 3HH = 7 Hz, 2H, m-ArH) 130.877.18 (t, HH = 7 Hz, 1H, p-ArH) 128.26 (Cme)1.94 (s, 15H, C5Me) 126.57 (Cpa)1.60 (d, 2HH 14 Hz, 1H, CR2) 111.43 (CMe5)1.02 (s, 911, SiMe3) 74.47 (CH)0.47 (d, 2HH = 14 Hz, 1H, CH2) 36.84 (CMe3)34.03 (CMe)10.29 (C5Me)a In THF-d8. b Not recorded. C 31P{’H} NMR (, ppm): 3•7e = -4.34 (s), = 191 Hz; 3•9e = -21.88 (s),= Hz; 3•12e= -4.33 (s), = 187 Hz. d1 THF-d8at -100 °C. e In CDC13. fin CD21.613.3 Results and DiscussionAs presented in the Introduction, we have previously been unable to synthesize complexes ofthe type Cp*W(NO)(R)X [X halide], and therefore questioned the stability of these 16-electronspecies. Cp*W(NO)(CH2S1Me3)Cl,which was synthesized via the reaction ofCp*W(NO)(CHSiMe)(N=CHMe) (2.1) with HC1, was the first such complex that we had beenable to isolate.4 Once the stability of this complex was recognized, a new, more general syntheticroute to a range of similar monoalkyl and aryl complexes was derived. This Chapter reports thesynthesis of a series of such alkyl and aryl chloride complexes and their reactivity toward avariety of substrates.Although the synthesis of four alkyl or aryl chloride complexes of the type Cp*W(NO)(R)Clis presented here, the neopentyl complex (R = CH2Me3)was chosen as a representativeexample, and most of the reactivity studies have been performed with this compound. There isno reason to believe that the observed chemistry of this neopentyl analogue is not general for thewhole class of compounds, and indeed in some cases, reactions were repeated with other alkyl oraryl chlorides and similar results were obtained.3.3.1 Initial Synthesis of Cp*W(NO)(CB2S1Me3)Cl(3.1)When a solution of complex 2.1, Cp*W(NO)(CHSiMe)(N=CHM ), in Et20 is exposed toat least two equivalents of HC1, the initial yellow solution turns royal blue with the formation of awhite precipitate. The precipitate can be removed by filtration, and complex 3.1,Cp*W(NO)(CH2SiMe3)Cl,can be isolated from the deep blue solution (equation 3.1). No2HCJW 3.1N” I ‘N ,.H -LMedfl=NHzfa N” I0 R 0 RMe3.1R = CH2SiMe362attempt has been made to characterize the air-sensitive white precipitate, which is presumably theimine salt [MeCH=NH2j l.Complex 3.1 is the first of its type that we have been able to isolate, and therefore it hasbeen subjected to an X-ray crystallographic analysis to determine, mainly, if the complex is amonomer or a dimer in the solid state.4 This feature was uncertain originally because the dialkylanalogue is a monomer, whereas Cp*W(NO)(Cl)2is known to be a dimer in the solid state.5 Thecrystallographic study confirmed the monomeric nature of complex 3.1, and an ORTEP drawingof its solid-state molecular structure is shown in Figure 3.1. The bond lengths and anglesresemble those of the dialkyl analogue, Cp*W(NO)(CH2SiMe3).6As expected, the largestangle between the three mono hapto ligands is that between the alkyl and the chloride ligand (seeSection 2.3.2). The metal-nitrosyl linkage is linear, thereby providing the metal center with a16-valence-electron configuration.3.3.2 Synthesis of Cp*W(NO)(R)Cl Complexes from Cp*W(NO)(Cl)2Once the stability of complex 3.1 had been recognized, a more convenient route to thisgeneral class of alkyl and aryl chlorides was desired. It has been shown that the reaction of thedihalo complexes Cp’M(NO)(X)2with Grignard reagents does not result in the production of themonoalkylated species. Also, the reaction of the dialkyl and diaryl derivatives with HC1 does notproduce any isolable products, and so a new route to alkyl and aryl chloride complexes wasneeded.1 It was decided that the reaction between Cp*W(NO)(Cl)2and one-half an equivalent ofdialkylmagnesium reagents,R2Mg.X(dioxane), would be attempted. These reagents are ideal forthis type of alkylation reaction because the stoichiometry can be precisely controlled by weighingout the exact amount of alkylating equivalents desired, and the solid reagents, theR2Mg.X(dioxane) and the dichioride, can be mixed thoroughly before the addition of solvent.Another reason for the choice of these reagents is that the presence of the dioxane63Figure 3.1. View of the solid-state molecular structure of Cp*W(NO)(CH2SiMe3)CI(3.1),inc1udinc celected bond lengths and angles (with esds in parentheses).C(15)C(14)C(13)Bond Lengths (A)W-Cl 2.321(3)W-N 1.804(9)W-C6 2.111(9)N-O 1.07(1)Bond Angles (deg)W-N-O 169.4(10)N-W-C6 94.4(4)N-W-Cl 102.0(3)C6- W - Cl 109.9(3)C(11)C(6)ClN0C(7)C(9)C(8)64prevents the magnesium halide salts produced in the reaction from forming isonitrosyl adductswith any products or starting materials. In the past, such adducts have been cleaved with water,but this can lead to decomposition of the alkylated product if it is moisture-sensitive.7The reaction between dialkyl- and diarylmagnesium reagents and Cp*W(NO)(Cl)2issuccessful in producing mono-alkylated products (equation 3.2), but carefi.il control of thereaction conditions is needed for success. THF is used as the solvent, and is vacuum transferredfrom sodium!benzophenone onto the solid mixture of dichioride and alkylating reagent at-196 °C. The reaction is then left to warm to room temperature very slowly while being stirred.If the reaction is attempted with other solvents such as Et20, or if it is allowed to warm tooquickly, the major product of the reaction is the dialkylated product, with one-half an equivalentof the starting dihalide left unreacted.1/2RMgdioxanew w/ \ THF / 3.2N ci N CiCi 0 RR = CH2S1Me3(3.1)= CBMe (3.2)= o-toiyl (3.3)= phenyl (3.4)The success of these reactions that useR2Mg.X(dioxane) reagents instead of Grignardreagents is due to a combination of many factors. The use of a highly polar solvent like THFhelps to solubilize the reactants at low temperatures. This is important because if the dichlorideis not all in solution, the more soluble alkylating reagent may attack any alkyl chloride complexthat has already formed, giving a mixture of products including the dialkylated one. The failureof the same reactions in Et20 is a result of this solubility problem, since the dichioride is onlyslightly soluble in Et20, even at room temperature.65Temperature control during the alkylation reaction is also critical. The first alkylation ofCp*W(NO)(Cl)2begins at around -100 °C, while the second alkylation, indicated by anothercolor change, is observed to occur around -30 °C. Therefore, the temperature of the reactionshould be controlled such that all of the alkylating reagent reacts before the solution is allowed towarm above approximately -50 °C. This also stresses the need for a low-freezing-point solventsuch as THF that is also polar enough to dissolve the reagents at such low temperatures. Theslowness of the second alkylation at low temperatures is presumably a kinetic effect. It has beenshown that the first step in the mechanism of alkylation reactions of Cp’M(NO)(X)2complexesusing Grignard reagents is reduction of the dihalide by the Grignard reagent followed by transferof the alkyl ligand.8 If the same type of mechanism is operative when dialkylmagnesium reagentsare used, then this could help explain the slow second alkylation. The reduction potentials of thealkyl chlorides are much greater than that of the dichioride, and thus the second alkylation, if itproceeds via reduction, will be much less favored, and will require higher temperatures (seeSection 3.3.7 for a comparison of the electrochemistry of the tungsten compounds involved).This mechanism involving initial reduction could also help explain the fact that if the Cp* ligand isreplaced by a Cp group, then a clean mono-alkylation is not possible using dialkylmagnesiumreagents, no matter how carefully the temperature of the reaction is controlled.9 The Cp liganddoes not donate as much electron density to the metal center as does the Cp* ligand, and thus theCp alkyl chloride species are more easily reduced, so that the major product is the dialkylatedspecies.Reaction 3.2 is thus a high yielding, general preparation for the alkyl and aryl halidecomplexes Cp*W(NO)(R)Cl. Two alkyl and two aryl complexes have been isolated andcharacterized, as shown in equation 3.2. All are deeply colored, either purple or blue, and areassumed to be monomeric by comparison with the structurally characterized trimethylsilylmethylchloride complex. Both alkyl chlorides are soluble in pentane, while the aryl chlorides are slightlysoluble in Et20. It should be noted here that the use ofaryl2MgX(dioxane) reagents in thepreparation of these complexes also helps in their isolation, since methylene chloride can be used66to extract the aryl chloride products away from the insoluble MgCI2.(dioxane) byproduct. Thiswould be impossible if Grignard reagents were used due to the solubiity ofMgX2 salts inmethylene chloride.All of the alkyl and aryl complexes isolated are thermally stable once they have beencrystallized. They are stable enough to be handled briefly in air, but they are very air-sensitivewhen in solution. When these compounds are exposed to air, either as solids or solutions, theyeventually form the well-known dioxo alkyl or dioxo aryl compounds, Cp*W(O)2R,in varyingyields.3.3.3 Metathesis Reactions of Cp*W(NO)(R)CIThe isolation of the alkyl and aryl chloride complexes 3.1 - 3.4 allows for the reactivity ofthese compounds to be investigated. One type of reaction that can be performed is themetathesis of the chloride for a variety of different ligands. This is, in fact, why the synthesis ofthese compounds was first attempted. It was hoped that the mixed alkyl speciesCp*W(NO)(R)R1 could be synthesized and their chemistry compared to the symmetric dialkylcompounds already known. To some extent, this has been accomplished,10and the synthesis andreactivity of a series of mixed alkyl and aryl complexes is discussed in Chapter 4. One example isgiven here to introduce the synthesis of these mixed species.When Cp*W(NO)(CH2Me3)Clis reacted with one-half an equivalent of(p-tolyl)2Mg.X(dioxane , the mixed alkyl-aryl complex Cp*W(NO)(CH2Me3) tolyl (3.5) isformed. This compound is isolated in good yield, and can be made either using isolated1/2 (p-to1yI),Mgdioxane0N”lCITHF3.5R = CH2Me367Cp*W(NO)(CH2Me3)Clor from the addition of the arylating agent to a solution of the alkylchloride generated at low temperatures. The latter route is a more convenient one-pot synthesis,and is used to make all of the mixed alkyl and aryl complexes described in Chapter 4.The chloride in the Cp*W(NO)(R)Cl complexes can also be metathesized for ligands otherthan alkyls or aryls. For example, Cp*W(NO)(CH2SiMe3)Clhas been reacted with alkoxide saltsto generate a series of alkyl alkoxide complexes, Cp*W(NO)(R)OR, similar to thosesynthesized by the hydrogenation route discussed in Chapter 2. The reaction of the alkyl chloridecomplexes with alkylamines to form alkyl amide compounds of the type Cp*W(NO)(R)NHR hasalso been investigated.12 One such compound is given here as an example of this type ofreactivity.s NH2CMe3W W 34/\ EtO /\.N CI - [Me3CNI1JCt N NHCMe30 R 0 RR = CH2SiMe33.6When Cp*W(NO)(CH2SiMe3)Clis reacted with an excess of t-butylamine in Et20, a whiteprecipitate is fonned, and the color of the reaction changes almost instantaneously from blue toyellow. The product of the reaction is characterized as Cp*W(NO)(CH2SiMe3)N}{CMe (3.6),the precipitate being [Me3CNTI3]C[ (equation 3.4). The spectroscopic properties of 3.6 aresimilar to those of the analogous compounds mentioned above, and include a diagnostic N-Hstretching band in the JR spectrum at 3220 cm4, and a broad signal at 6.87 ppm in the 1H NMRspectrum assignable to the NH proton.3.3.4 Reactivity of Cp*W(N0)(CH2Me3)CIwith Lewis BasesSimilar to the 16-electron Cp*W(NO)R2compounds, the monoalkyl complexes react readilywith Lewis bases to form coordinatively and electronically saturated I 8-electron adducts. Thus,68Cp*W(NO)(CH2Me3)Clreacts instantaneously with PMe3 to form the yellow 1:1 adduct,Cp*W(NO)(CHe)(PMl(3.7, equation 3.5). The phosphine ligand is believed to becoordinated trans to the NO ligand, as expected (see Section 2.3.2). Consequently, the twomethylene proton signals of the alkyl group show very different couplings to the phosphorusnucleus, as observed in the 1H N1*iR spectrum of this compound. In fact, one diastereotopicmethylene proton signal shows a 4.2 Hz coupling to the phosphorus, while the other shows nophosphorus splitting at all. If the phosphine was coordinated trans to the alkyl ligand, thecoupling of each methylene proton to the phosphorus would be more similar. Upon warming,complex 3.7 does not isomerize, so it is assumed that the thermodynamically more stable isomerhas been isolated.Upon coordination ofPMe3 to Cp*W(NO)(CH2Me3)Cl, the nitrosyl-stretching frequencydrops by 29 cm1, a feature consistent with an increase of electron density at the metal center.The 18-electron configuration of the complex has a great stabilizing effect, so that although theneopentyl chloride complex is air-sensitive, the phosphine adduct is stable to air for at least anumber of days in the solid state.o PMe3/V\ 3.5RR = CH2Me3o’ /\NNRCIL23.8Cp*W(NO)(CH2Me3)Cl(3.2) also forms an adduct with pyridine, but only at lowtemperatures (equation 3.5). The adduct, complex 3.8, can be crystallized by cooling an Et2069solution of 3.2 containing an excess of pyridine to -30 °C. The adduct crystallizes as orangeblocks and can be separated from the mother liquor as such, but when the crystals are placedunder vacuum, the pyridine is slowly lost to reform the alkyl chloride starting material. This is incontrast to the vacuum-stable trimethylphosphine adduct, and indicates that the pyridine is boundmuch more weakly than the phosphine.Dissolution of the isolated pyridine adduct in both coordinating and noncoordinatingsolvents such as CDC13,C6D and THF-d8 causes quantitative loss of the pyridine and formationof the alkyl chloride precursor. Thus, the only way to characterize the adduct by 1H NMRspectroscopy is by performing a low-temperature NMR study. When CDC13 is used as a solvent,the adduct does not form above the solvent’s freezing point. THF-d8can be used with moresuccess because of its much lower freezing point. When a sample of the pyridine adductdissolved in this solvent is cooled, a number of coalescent temperatures for the different peaksare observed, until at -100 °C all peaks sharpen into a spectrum consistent with the existence ofthe pyridine adduct. At this temperature, the solution is not the purple of the alkyl chloride, butorange, the characteristic color of the adduct. Reaction of Cp*W(NO)(CH2Me3)(PMCIwith LDARecently, members of our research group have reported the synthesis of alkylidenecomplexes of molybdenum, CpMo(NO)(=CHCMe3)L[L = PR3, pyridine].13 These compoundsare formed by thermal alkane elimination from the bis-(neopentyl) compound and trapping of theunsaturated alkylidene intermediate with a Lewis base. The analogous reaction with the tungstensystem does not lead to any isolable products, although due to the similarity ofthe two metals,the alkylidene product, if formed, is expected to be stable. In an attempt to synthesize suchalkylidene compounds of tungsten, the PMe3 adduct of the neopentyl chloride complex, 3.7, wasreacted with a strong base to try to dehydrohalogenate the compound and form an alkylidenespecies.70The reaction ofCp*W(NO)(CH2Me3)(PMClwith LDA is performed in THF. Thecolor of the solution does not change drastically, but when the reaction is monitored by IRspectroscopy, the nitrosyl-stretching band of the starting complex is seen to disappear and a newband grow in during the two-hour reaction. Chromatography of the final reaction mixture allowsthe isolation of a compound whose elemental analysis matches the composition of the expectedalkylidene product. However, the 1H NMR spectrum of the compound reveals that the expectedproduct has not formed, but instead the metalated Cp* compound,( i5,ri’-CMe4CH2)W(NO)( HMe3P (3.9, equation 3.6) has been isolated. Thus,instead of deprotonating the (1-position of the neopentyl ligand, the base deprotonates one of theCp* methyl groups, which then undergoes an intramolecular metathesis of the chloride ligand.An alternative mechanism for the reaction is the initial formation of the alkylidene productfollowed by rearrangement to form the observed product. Bercaw and co-workers haveobserved such a rearrangement in a hafnium system.’4 This latter pathway, however, is unlikelybecause the methylene protons of the starting material are much less accessible by the bulky basethan the Cp* methyls, and so complex 3.9 should be the kinetic product. Also, if the alkylidenecomplex is formed, it should be very thermally stable, by analogy to the stability observed for themolybdenum system, and would therefore not be expected to rearrange to the metalated Cp*complex.LDA3.60N’NpMe3 THF R2..RC1 N PMe303.9The 1H and 13C NMR spectra of 3.9 each contain four signals assignable to the fourinequivalent Cp* methyls. In addition to the PMe3 and CMe3 signals, four different methyleneproton resonances are observed in the 1H NMR spectrum due to the two sets of diastereotopic71methylene groups. These protons show markedly different2Jp couplings characteristic of theirchemical environments. The neopentyl methylene protons are bound to a purely sp3 methylenecarbon and exhibit a coupling constant of 12.9 Hz whereas the methylene protons of the tucked-in Cp* ring are bound to a carbon with considerable sp2 character and are weakly (or geminally)coupled (3.6 Hz) to one another.15An X-ray crystallographic analysis was performed on suitable crystals of complex 3.9 grownfrom a benzene/hexanes solution. The molecular structure, in the form of an ORTEP drawing, isshown in Figure 3.2 along with selected bond lengths and angles. The diagram clearly shows thepuckering of the deprotonated Cp* ring. The ring carbons of this ligand are in a plane, but themethylene carbon is bent down toward the tungsten by 23.38° from this plane. The ring showssome tetramethylfulvene-like character, as there is evidence for localized bonding within the Cpring, and the C(1)-C(6) bond length of 1.44(1)A is intermediate between that for single anddouble C-C bonds. For comparison, an iridium complex with a purelyii4-C5MeCH2ligandshows a ring-methylene C-C bond length of 1.3 77(10) A and the methylene is in the planedefined by the Cp ring.16 Consistent with this tetramethylfulvene character in complex 3.9 is theobservation of sp2 character in this methylene group, as mentioned above.Many transition-metal compounds containing the ligand have been reported.Some, like(5,ri’-Ce4CH)W(NAr)l’7and(ri5,ii’-CMe4CH2)Ru(COD)’8have beenmade by direct deprotonation of the Cp* ring with various strong bases. Another common routeto such complexes is metalation of the Cp* ring with concomitant alkane loss from metal alkylcomplexes. For example, both Cp*Hf(rj5,rIlCMe42)(CHPh)landCp*Zr(15,lCMe4H2)(Ph)l9are produced when the dialkyl complexes Cp*Hf(CH2Ph) andCp*Zr(Ph)2,respectively, are heated to induce alkane elimination.72Figure 3.2. View of the solid-state molecular structureof(r5,r1-CMe4CH2)W(NO)-(CH2CMe)P (3.9), including selected bond lengths and angles (with esds in parentheses).C18C?Bond Lengths (A)W-C(6) 2.319(8)W - C(14) 2.242(7)C(1)-C(6) 1.44(1)C(2)-C(3) 1.39(1)C(1)-C(2) 1.42(1)Bond Angles (deg)W-N-O 171.7(6)C(14)-W-N 99.6(3)C(14)-W-P 82.9(2)W - C(6)-C(l) 66.0(4)P-W-N 90.3(2)COC3 C2ClC4 CswC14C16C611C15C170C13Cl2NR/W\ 3.10 37N Ro ci cEDR = CR CMe .—NCMe32 3 /W\//N Co Ci\R3.113.3.5 Reaction of Cp*W(NO)(CR2Me3)CIwith Unsaturated Lewis Bases: Formation ofInsertion ProductsTo determine if the tungsten-carbon bonds of the alkyl halides are prone to insertion ofpolar, unsaturated links like their dialkyl and diaryl analogues, Cp*W(NO)(CH2Me3)Clhasbeen reacted with carbon monoxide and t-butylisocyanide. Both reagents react very quickly withthe organometallic complex to form the acyl and iminoacyl insertion products 3.10 and 3.11,respectively (equation 3.7). The complexes are isolated in high yield, and the insertion reactionsare assumed to be quantitative.Complex 3.10 contains a classicr2-acyl ligand. This is easily deduced by inspecting the ll.spectrum, which contains two strong bands at 1584 and 1564 cm1 attributable to the terminalnitrosyl and thei2-acyl ligand. Unambiguous assignment of these bands is impossible without alabeling study. However, a low value for the vCO in this range is consistent with other ‘r2-acylcomplexes that we have isolated in the past, 10,20 and similar to other reported complexes withsimilar ligands.2’ The solid-state molecular structure of the molybdenum analogue of complex3.10 has been determined by X-ray crystallography, and as expected, shows the acyl bound to themolybdenum center through both the carbonyl carbon and oxygen atoms.22The iminoacyl ligand in complex 3.11 is also bound to the tungsten in anri2-fash on. This isalso evident from the IR. spectrum which contains a sharp band at 1680 cm1 assignable to VCN74and a broader band at 1537 cm1 assignable to the nitrosyl stretch. This value of vCN ischaracteristic ofr2-iminoacy1 complexes; 1-iminoacyls typically display this vibration at higherenergy.21 It is interesting that the nitrosyl-stretching frequency for the iminoacyl complex is47 cm lower than that of the acyl complex 3.10. This leads to the conclusion that the iminoacylligand is a better source of electron density for the metal center than is the acyl group. Reaction of Cp*W(NO)(fl2{O}CHMe3)CJwith PMe3To determine if the oxygen atom of thet2-acyl ligand in 3.10 could be displaced from themetal center to give an1-acyl, complex 3.10 has been exposed to PMe3. A reaction occursinstantaneously upon PMe3 addition, and the product can be crystallized in good yield aftertaking the reaction mixture to dryness and redissolving the residue in Et20.PMe3N”I’C 0N/\PMe33.80 C1\ CIC==0R /RR = CH2Me3 3.12The product of the reaction is shown in equation 3.8. As predicted, complex 3.12 containsan1-acyl ligand, and the PMe3 is coordinated directly to the metal center, presumably betweenthe acyl and the chloride ligands. The evidence that the acyl is now bound to the metal center byonly the acyl carbon comes from JR and 13C NME. data. The JR spectrum shows a vCO band at1634 cm1, at much higher energy than the analogous band for ther2-complex discussed above(1564 - 1584 cm1). The carbonyl carbon resonance in the 13C NMR spectrum has also beenshifted upfield to 264.6 ppm from a value of 288.0 ppm for the starting complex. The PMe3 isdirectly attached to the metal center, as evidenced by the strong coupling of the phosphorusnucleus to the spin one-half isotope of tungsten(1Jp = 187 Hz). This information is important75in determining the true structure of complex 3.12, since nucleophilic attack at the carbonylcarbon of acyl ligands has been observed in similar systems. For example, treatment ofCpMo(NO)(2-C{O}p-tolyl)I with PMe3 leads to the formation of the carbon-based phosphineadduct, CpMo(NO)(C { 0 } {PMe }p-tolyl)I,23 and the acyl carbon of Cp*Ta(q2{ 0 } SiMe3)C1is prone to attack by nucleophiles such as pyridine and PMe3.243.3.6 Halide Abstraction from Cp*W(NO)(CE2Me3)CI: Reaction with AgBF4To determine if the halide ligands of the alkyl and aryl chlorides could be abstracted to formstable alkyl cationic complexes, Cp*W(NO)(CH2Me3)Clhas been reacted with AgBF4 inacetonitrile. The reaction is almost instantaneous, even at low temperatures, and is indicated by acolor change from purple to orange and the formation of a flocculent white precipitate,presumably AgCl. The product can be crystallized from CH3N/Et20orCH2IIEtOasorange blocks that are only soluble in polar solvents such as acetonitrile or methylene chloride.The product, 3.13, has been characterized as the organometallic salt[Cp*W(NO)(CH2Me)(CH)]+BF4_.The presence of two acetonitrile molecules bound tothe tungsten center is indicated by the elemental analysis and also by the NME. data. Two signalsassignable to the methyl groups of two distinct acetonitriles are observed in the 1H NMR and13C NMR spectra. This implies that the nitriles are cis to one another in the molecule, since ifthey were trans they would be magnetically equivalent and thus exhibit a single resonance. ThevCN bands appear at 2319 and 2291 cm1 in the IR spectrum, and are higher in energy than freeacetonitrile (2230 cm-’), a feature diagnostic of the presence of N-bound nitrile ligands.25 Thetwo coordinated acetonitrile ligands give the metal center an 18-electron configuration, althoughthe cation is still reactive and decomposes readily when exposed to air either in the solid state orin solution to form the well-known dioxo complex, Cp*W(O)2(CHMe3).76AgBF,—z;:z +3.8/ \ MeCNN Cl 0” ‘ e0 R RNCMeR = CH2Me33.13Complex 3.13 is related to other Mo and W cations that our group has previouslydescribed.26 The isoelectronic [Cp*Mo(NO)(X)(NCMe)2]+PF6_[X = I, Br, Cl] series ofcomplexes have previously been prepared, and are probably isostructural to the tungsten cationprepared here. Complex 3.13 is also isoelectronic with Jordan’s cations, [Cp’2Zr(R)(L)]BX4[R = H, Me, CH2Ph, Ph; L = NCMe, THF; X = F, Ph].27 The latter cations display a wide rangeof interesting polymerization and insertion chemistry, including insertion of acetonitrile into thezirconium-phenyl bond of[Cp2Zr(Ph)(NCMe)]BPh4to form the a.zomethine complex[Cp2Zr(NC{Me}Ph)(NCMe)]BPh4.8The analogous reaction does not occur for the tungstencation prepared here, even when a solution of the complex in acetonitrile is heated to 40 °C.Halide abstraction from compound 3.2 in a non-coordinating solvent such asdichioromethane does not lead to any tractable products, presumably because the unsolvated14-electron [Cp*W(NO)(CH2Me3)]+,if formed, would be exceedingly reactive. Similarly, thesame reaction performed in THF yields no isolable products, presumably because the THF is nota strong enough coordinator to stabilize the cationic species.3.3.7 Reduction Chemistry of Cp*W(N0)(CR2Me3)ClThe reduction behavior ofCp*W(NO)(CHMe)Clhas been investigated first byelectrochemical, and then by chemical means. It was thought initially that reduction of thiscomplex would cause the formation of a metal-metal bonded dimer. This is not the case,however, and a much more complex and interesting reaction takes place. The mechanism of thisreaction has also been briefly probed by chemical means.773.3.7.1 Electrochemistry of Cp*W(NO)(CH2Me3)C1The redox properties ofCp*W(NO)(CHMe)Clhave briefly been studied using cyclicvohammetry in THF. The complex shows no oxidation features out to the solvent limit, but doesshow a reduction feature at approximately -1300 mV, as shown in Figure 3.3. The reductionpotential of the complex can be compared to those exhibited by the dichioride (E°’ = -750 mV)and the bis-(neopentyl) analogues (E°’ = -1720 mV). This trend is not unexpected, sincereplacement of a chloride ligand with a more electron-donating alkyl ligand should, and does,cause the metal center to become more electron rich, and consequently more difficult to reduce.This electron-density trend is also apparent when considering the nitrosyl-stretching frequenciesof these three complexes. The more electron-rich center in Cp*W(NO)(CH2Me3)(vNO =1568 cml) has a lower vNO than the alkyl chloride (vNO = 1595 cm1)which in turn has a lowerstretching frequency than the dichioride (vNO = 1630 cm1).806040E20UE0-20400 -500 -1000 -1500mV (scan rate = 2.0 V/s)Figure 3.3. Cyclic voltammogram of Cp*W(NO)(CH2Me3)Cl,measured in THF containing0.10 M [Bu4N]PF6,at a Pt-bead working electrode. Potentials are measured vs SCE.78As is evident from the cyclic voltammogram shown in Figure 3.3, the reduction of complex3.2 is not completely reversible. This is obvious from the small return oxidation peak, and alsofrom other parameters that suggest a quasi-reversible behavior. Thus, p,a’p,c is lower than theexpected unity value (being only 0.4 at a scan rate of 1 V/s) and increases with increasing scanrate. Also, the reduction peak and the return oxidation peak shift further apart when the scanrate is increased, indicating that there is either a very slow electron-transfer process occurring, oran irreversible reaction is occurring when complex 3.2 is reduced.29 To determine if a reactionoccurs upon reduction, it was decided that a chemical reduction of this complex would beattempted. Chemical Reduction of Cp*W(NO)(CH2Me3)ClExposure of Cp*W(NO)(CH2Me3)Clto zinc powder in THF overnight leads to a changein the solution color from purple to a deep green-brown. An IF. spectrum of the green solutionshows the presence of one new nitrosyl-stretching band and the absence of the analogous bandfor the starting material. Workup of the reaction mixture leads to the isolation of a deep greencompound in moderate yield. Mass spectral and elemental analysis data for the product indicatethat it is a dimer of composition [Cp*W(NO)(CH2Me3)]The spectroscopic data for thecomplex, however, do not support the presence of a symmetric dimer, since the 1H NMRspectrum contains distinct sets of resonances for two Cp* and two neopentyl ligands.Furthermore, the JR spectrum of the complex, as well as exhibiting a nitrosyl stretch, indicatesthe presence of a terminal oxo ligand. The complex is thus formulated as the bridging nitridospecies, [Cp*W(NO)CH2Me3jIIN[Cp*W(O) (3.14), based on comparison of itsspectroscopic data to those of a molybdenum analogue that has been structurally characterized 3079The complex contains a Cp*W(NO)(CH2Me3)fragment bound via a nitride bridge to aCp*W(O)(CH2Me3)fragment. It appears that a nitrosyl ligand has been cleaved to form thebridging nitride and the terminal oxo ligands. No labeling studies have been performed, but theScheme 3.1RNIN \/R0.-WcI,N”IC1o RR = CH2Me3ZAITIIF3.103.14NICI0 R+ e0N‘XCI0N“CINcR”Jj+ e-a-80reaction was repeated under rigorously anaerobic and anhydrous conditions to confirm that theoxo ligand did not arise from adventitious water or molecular oxygen. Therefore, a mechanismfor this reaction must include a step or series of steps in which the nitrosyl N-O bond is broken aswell as account for the reduction of two equivalents of starting material,A proposed mechanism for the transformation in equation 3.10 is shown in Scheme 3.1. Thefirst step of the reaction is assumed to be reduction by zinc of 3.2 to form the radical anion,[Cp*W(NO)(CH2Me3) l]._. This complex loses C1 and the nitrosyl ligand then adopts an2-coordination mode. The next step of the proposed mechanism is the formation of an adductbetween the Lewis-acidic startingmaterial and the nitrogen atom of thei2-NO to form anintermediate containing a nitrogen bridge between the metals. The final step would then bereduction with loss of chloride, and rearrangement of ther2-nitrosyl to terminal oxo and bridgingnitride ligands. Although this simple series of reactions shows a logical progression, manysubtleties have been purposefully avoided because of the lack of evidence to support such amechanism.One aspect of the mechanism, namely that the species which is reduced first is the one tohave its nitrosyl ligand cleaved, can be proven using a type of crossover experiment. Fromelectrochemical data, it is known that Cp*W(NO)(C1)2is reduced much easier than the alkylchloride complex, Cp*W(NO)(CH2Me3)Cl.2It is also known that this former reduction,when performed with powdered zinc, is many times faster than the latter. Thus, when a mixtureof the dichloride and the alkyl chloride is exposed to zinc, the dichioride will be reduced first.This mixed reaction was performed in THF with a two-fold excess of the alkyl chloride, such that+ xs ‘ 3.11N”lC1 N”ICI0 Ci 0 RR = CB2Me3 3.1581when the dichloride was reduced, it would preferentially encounter the alkyl chloride. Thereaction was allowed to proceed for 20 minutes, during which time the nitrosyl-stretching bandof the dichloride disappeared, and a new product band grew in. The intensity of the excess alkylchloride nitrosyl band did not change much during the course of the reaction. The solution wasworked up similarly to the reaction to form 3.14 above, and was assisted by the fact that theunreacted alkyl chloride did not pass through Florisil while the product did. A compound of thesame type as 3.14 was isolated in moderate yield, and exhibited the expected spectroscopicproperties. In other words, the presence of one neopentyl and two Cp* ligands was inferred fromthe NMR data, and the presence of a terminal oxo as well as the bridging nitrido ligand wasevident from lB. data. However, it could not be determined if the oxo ligand was attached to thetungsten having the alkyl or the chioro ligand, and therefore complex 3.15 was subjected to anX-ray crystallographic analysis, which established that the oxo was attached to the tungstencontaining the chloride ligand.Since it is known that the dichloride is the species that is reduced first and that the oxoligand in the final product is attached to the metal derived from the dichioride, it can beconcluded that the species that is reduced first is the one that has its nitrosyl ligand cleaved. Thisfact is consistent with the proposed mechanism shown in Scheme 3.1. If the same mechanism isassumed to be operative in the crossover experiment, the dichloride, once reduced, is trappedwith the alkyl chloride that is in excess. Further reduction and NO bond cleavage leads to theobserved product.An ORTEP plot of complex 3.15 is shown in Figure 3.4. The structure is similar to thatproposed for 3.14, and also to the structurally characterized [Cp*Mo(NO)CH2Si e]pN{Cp*Mo(O)CH2Si e3].The W(1)-N-W(2) angle is almost linear at 177.4(3) , implying thatthe nitrogen is using its lone pair for multiple bonding in this bridge. The bridge is notsymmetric, as is expected because of the different substituents on the two tungstens, but bothbond lengths are similar, and in the range between double and single tungsten-nitrogen bonds.The two bond lengths of 1.911(4) and 1.808(4) A can be compared to the multiple tungsten-82nitrogen bond length in Cp*W(NO)(CH2SiMe3)(N=CHMe) (2.1) of 1.9 A. A comparison canalso be made to the symmetrically bridged nitrido dimer [Cp*WMe3j(l.LN) reported bySchrock’s group, that has W-N bond lengths of 1.8475(8) A.3’The two Cp* ligands in complex 3.15 are orthogonal, with a dihedral angle about the W-NW link of 101 0 The tungsten-oxo link is ofa typical length (1 .713(4) A) and can be comparedto similar bonds in CpW(O)2HSIMe3(1.723(5) and 1.716(5) A).32 The dimer’s structuralparameters can also be compared to the molybdenum analogue discussed above. In this case, theMo-N-Mo angle is 169.6(2)°, this bridge is again asymmetric (Mo-N bond lengths of 1.908(3)and 1.8 12(3) A), and the Cp* ligands are likewise orthogonal (99.4 0)Most nitride-bridged transition-metal complexes that have been structurally characterizedcontain asymmetric bonds to the nitride from each metal (for example, compound A33 in Figure3.5). Many others are known that contain nitride bridges between two different metal centers(compound B34). A rare example of a complex containing a symmetric bridge between twometals is compound C in Figure 3.5, a cationic coordination compound of ruthenium.35Schrock’s group has structurally characterized a tungsten complex containing a totally symmetricand linear nitrido bridge (compound D).31A discussion of the IR spectra of the two nitrido-bridged complexes 3.14 and 3.15 is meritedsince the IR data are very useful in assigning the structure of these and similar compounds. Bothcomplexes exhibit two sharp bands in their IR spectra at 962 and 890 cm1. One of these isassignable to vWO. Examples of tungsten oxo complexes, all of which normally show sharpbands in this region, include Cp*W(O)(Me)2HSIMe3(vWO 933 cm-’) andCp*W(O)(=CHPh)CH2SiMe3(vWO = 939 cm’).36 The other band exhibited by the imidodimers is av mode, although it is not clear exactly what vibration this band represents. It hasbeen attributed to an asymmetric stretch of the W-N-W link in complexes such asW2NX10383Figure 3.4. View of the solid-state molecular structure of [Cp*W(N0)CH2Me3].I[Cp*W(O)Clj (3.15), including selected bond lengths and angles (with esds in parentheses).C(12)C(52)Bond Lengths (A)W(1)-N(2) 1.911(4)W(2) - N(2) 1.808(4)W(2) - 0(2) 1.713(4)W(1)-N(l) 1.766(5)Bond Angles (deg)W(2)-N(2)-W(1) 177.4(3)W(1) - N(1)- 0(1) 170.3(6)Cp(l) - W(2) - W( 1) - Cp(2) 101C(13)C(53)C(54)C(3)C(43)0(1)C(11)N(2) C(55)C(6)C(51)0(2)C(9)C(10)84Ci Ci—NCI’, PEt3/ MC3SIO\ IVN —‘V—N— ...• V=N—Pt—PEt3/ I I Me3SiO’ /Cl Me3S1OBMe(en)2Ru =N==Ru(en)2H2N Nil2en ethylenediamineCFigure 3.5. Selected examples of known, bridging nitrido complexes.[X = Cl, Br],37 and labelling with 15N shifts this band to lower energy in [Cp*WMe3]2(j.tN)(800 to 789 cml).3’ The band is nonetheless assigned to some M-N vibration which makes it adiagnostic tool for the identification of these types of complexes.The mechanism shown in Scheme 3.1 for the reduction of Cp*W(NO)(R)Cl complexescontains a proposed radical intermediate. In an attempt to trap this intermediate, or any otherradical species that may form, the reaction has been repeated in the presence of a radical trap,diphenyl disulfide. When the reaction of Cp*W(NO)(CH2Me3)CIwith zinc is performed in thepresence of an excess ofPhSSPh, no nitride-bridged dimer is formed. Instead, the reactionsolution turns violet instead of the green-brown color characteristic of the dimer.Cp*W(NO)(CH2Me3)SPh(3.16) is isolated from the final reaction mixture in good yield. Theproduct is not unexpected, since it is the species formed when the radicalCp*W(NO)(CH2Me3),formed by reduction, is trapped with PhS. It had been hoped that anD85intermediate containing an2-NO ligand would be trapped and possibly provide insite on themechanism of the reduction reactions. This result, however, does not preclude the existence ofsuch ani2-NO intermediate, since diphenyl disulfide is an efficient radical scavenger and mayhave trapped the Cp*W(NO)(CH2Me3).species before it had a chance to form anyintermediates.3.4 Summary and Future WorkIn this Chapter, the synthesis and some chemistry of Cp*W(NO)(R)C1 complexes has beenpresented. It has been shown that these complexes are stable, and can be synthesized in usefulyields from the readily available dichioride precursor. These 16-electron alkyl halide species havebeen shown to be very reactive toward a variety of substrates. Many sites of reactivity have beenobserved in these complexes, including the tungsten-alkyl bonds, the chloride ligand, and also theCp* and NO ligands.The tungsten-carbon bonds of the alkyl or aryl ligands are prone to insertion of CO andisonitriles, formingi2-ligands that can be converted to ri’-bound ligands by addition of a strongLewis base. The alkyl and aryl chlorides are coordinatively and electronically unsaturated, andthus form 18-electron complexes when exposed to bases that cannot undergo insertion, likePMe3 or pyridine. The phosphine adduct has been reacted with a strong base, whichdeprotonates the Cp* ligand to form a tucked-in Cp* complex. These various 18-electroncompounds may themselves have some interesting chemistry. For example, the q’-acylcomplexes could possibly be converted to Fischer carbenes by attack of an electrophile on theacyl ligand. The chloride of the phosphine adduct may be metathesized for an alkyl groupcontaining 3-hydrogens, or be abstracted to form a cation. Other metathesis reactions can beenvisioned for both the alkyl and aryl chloride complexes and their 18-electron Lewis baseadducts. For example, it may be possible to metathesize the chloride ligand with thiolates,phosphides, complex alkyl groups, silyls, or other more exotic ligands.86The synthesis of the salt containing an 18-electron alkyl cation,[Cp*W(NO)(CH2Me3)(CH)]+BF4_,has been presented here, but no reactivity studieswere performed on this compound. It is evident from NMR data that the nitriles are labile, thusthe coordination of other bases, possibly unsaturated ones, may be successful. This could lead topolymerizations, condensations, or other such useful reactions known to occur on Lewis-acidictransition-metal cations. The halide abstraction step in the production of this cation could also beperformed in the presence of unsaturated substrates to determine if this was a viable route tocationic complexes containing unsaturated ligands.The neopentyl chloride complex shows very interesting reduction chemistry. Themechanism of the NO bond cleavage reaction that occurs upon reduction of this compound is notclear, and more work needs to be done to clarif,r the mechanism. For example, the reduction ofthe chloride could be performed in the presence of other radical traps, in the presence ofLewisbases, or with other reducing agents to try to trap intermediates. The nitrido-oxo product alsomerits further attention. An understanding of the chemistry of this dimer and others like it couldlead to new types of complexes, and possibly some conclusions as to the nature of the bonding inthese bimetallic species.It has been shown that the alkyl and aryl chloride complexes are very useful precursors to anumber of new types of compounds, some of which have interesting chemistry of their own. Thepreparations of these complexes are now considered standard in our laboratories, and are used tosynthesize many mixed alkyl-alkyl, alkyl-aryl, alkyl-amide, and alkyl-alkoxide compounds that areused for the investigation of the fundamental chemistry of tungsten-carbon and tungstenheteroatom bonds.873.5 References and Notes(1) Dryden, N. H.; Legzdins, P.; Trotter, 3.; Yee, V. C. Organometallics 1991, 10, 2857.(2) Dryden, N. H.; Legzdins, P.; Rettig, S. J.; Veitheer, 3. E. Organometallics 1992, 11,2583.(3) The italicized X in the formula for the dialkyl and diarylmagnesium reagents,R2Mg.X(dioxane), refers to a variable amount of dioxane contained in the differentreagents. The potency of the reagents is usually consistent with there being between oneand two dioxane molecules for each magnesium atom.(4) Debad, J. D.; Legzdins, P.; Batchelor, R. 3.; Einstein, F. W. B. Organometallics 1992, 11,6.(5) Dryden, N. H.; Legzdins, P.; Batchelor, R. J.; Einstein, F. W. B. Organometallics 1991,10, 2077.(6) Legzdins, P.; Rettig, S. J.; Sanchez, L. Organometallics 1988, 7, 2394.(7) Legzdins, P.; Veitheer, J. E. Acc. Chem. Res. 1993, 26, 41.(8) Herring, F. G.; Legzdins, P.; Richter-Addo, G. B. Organometallics 1989, 8, 1485.(9) The attempted synthesis ofCpW(NO)(CH2SiMe3)Clfrom CpW(NO)(Cl)2and(CH2SiMe3MgX(dioxane) was performed analogous to the Cp* preparation presented inthe Experimental Section. The only product isolated was CpW(NO)(CH2SiMe3).(10) Debad, 3. D.; Legzdins, P.; Batchelor, R. J.; Einstein, F. W. B. Organometallics 1993, 12,2094.(11) Legzdins, P.; Lundmark, P. J.; Rettig, S. J. Organometallics 1993, 12, 3545.(12) Legzdins, P.; Ross, K. 3.; Brouwer, E. B.; Rettig, S. 3. Organometallics 1994, 13, 0000.(13) Legzdins, P.; Rettig, S. J.; Veitheer, J. E. J Am. Chem. Soc. 1992, 114, 6922.(14) Bulls, A. R.; Schaefer, W. P; Serfas, M.; Bercaw, J. E. Organometallics 1987, 6, 1219.88(15) The exact assignments of the methylene protons were eludicated from a two-dimensional1H:-COSY spectrum of 3.9.(16) Glueck, D. S.; Bergman, R. G. Organometallics 1990, 9, 2862.(17) Hyber, S. R.; Baldwin, T. C.; Wigley, D. E. Organometallics 1993, 12, 91.(18) Kolle, U.; Kang, B-S.; Thewalt, U. I Organomet. Chem. 1990, 386, 267.(19) Schock, L. E.; Brock, C. P.; Marks, T. J. Organometallics 1987, 6, 232.(20) Dryden, N. H.; Einstein, F. W. B.; Legzdins, P.; Lundmark, P. J.; Reisen, A.Organometallics 1993, 12, 2085.(21) Durfee, L. D.; Rothwell, I. P. Chem. Rev. 1988, 88, 1059.(22) Debad, J. D.; Legzdins, P.; Rettig, S. J.; Veitheer, J. E. Organometallics 1993, 12, 2714.(23) Bonnesen, P. V.; Yau, P. K. L.; Hersh, W. H. Organometaiics 1987, 6, 1587.(24) Arnold, J.; Tilley, T. D.; Rheingold, A. L.; Geib, S. J.; Arif A. M. I Am. Chem. Soc.1989, 1]], 149.(25) Nakamoto, K. Infrared and Raman Spectra ofInorganic and Coordination Compounds,4th ed.; Wiley-Interscience: New York, 1986; pp 280-281.(26) Chin, T. T.; Legzdins. P.; Trotter, J.; Yee, V. C. Organometallics 1992, 11, 913.(27) Jordan, R. F.; Bradley, P. K.; LaPointe, R. E.; Taylor, D. F. New.J Chem. 1990, 14, 505and references therein.(28) Borkowsky, S. L.; Jordan, R. F.; Hinch, G. D. Organometallics 1991, 10, 1268.(29) For moderate charge-transfer rates, quasi-reversible electrochemical behaviour is oftenobserved. The slowness of electron transfer is often ascribed to subtle structural changesthat occur upon reduction (or oxidation). For a general description of this phenomenon,see: Geiger, W. E. Frog. Inorg. Chem. 1984, 275. For more detailed descriptions ofelectrochemical quasi-reversibility, see: (a) Matsuda, H.; Ayabe Y. Z. Elektrochem. 1955,8959, 494. (b) Bard, A. J.; Faulkner, L. F. ElectrochemicalMethods; John Wiley and Sons:New York, NY, 1980; Chapter 6.(30) Young, M. A.; Legzdins, P.; Batchelor, R. J.; Einstein, F. W. B. Manuscript inpreparation.(31) Glassman, T. E.; Liu, A. H.; Schrock, R. R. Inorg. Chem. 1991, 30, 4732.(32) Legzdins, P.; Rettig, S. J.; Sanchez, L. Organometallics 1985, 4, 1470.(33) Sorensen, K. L.; Lerchen, M. E.; Ziller, J. W.; Doherty, N. M. Inorg. Chem. 1992, 31,2678.(34) Doherty, N. M.; Critchiow, S. C. .1 Am. Chem. Soc. 1987, 109, 7906.(35) Griffith, W. P.; McManus, N. T.; Skapski, A. C. .L Chem. Soc., Chem. Commun. 1984,434.(36) Legzdins, P.; Phillips, E. C.; Sanchez, L. Organometallics 1989, 8, 940.(37) Homer, M.; Frank, K-P.; Strähle, J. Z. Naturforsch. 1986, 41b, 423.90CHAPTER 4Synthesis and Reactivity of the Mixed Alkyl and Aryl ComplexesCp*W(NO)(R)R.4.1 Introduction 904.2 Experimental Procedures 934.3 Results and Discussion 1084.4 Summary 1264.5 References and Notes 1284.1 IntroductionSystems containing transition metal-carbon cs-bonds, especially those with coordinativelyand/or electronically unsaturated metal centers, have received the attention of many workers. 1This fact is due not only to the synthetic utility of such complexes because of their involvement ina variety of stoichiometric and catalytic transformations,2but it also reflects the desire of manyresearchers to continue elucidating the fundamental chemistry of the transition-metal-carbonbond. Recent work in these laboratories has centered on the characteristic chemistry of the 16-valence-electron bis-(hydrocarbyl) complexes Cp’M(NO)R2,especially in the area of reactivity oftheir metal-carbon CT-bonds.3 Some of this chemistry is discussed in Chapter 2. A ranking of therelative reactivities of different metal-carbon bonds may be established by comparing thereactivities of the different symmetric dialkyl or diaryl complexes. Some comparisons have beenmade, but these have been limited mainly to empirical observations. For example, we havenoticed that for the series CpM(NO)R2,the complexes containing aryl ligands are less thermally91stable than their alkyl counterparts. An accurate comparison between different complexes,however, would require a more quantitative approach, such as the gathering and comparing ofkinetic data. Even then, steric and electronic factors unrelated to the metal-carbon bonds wouldpossibly affect the reactivity of one complex over another, even though the bond reactivities aresimilar. One simple solution to this problem is the direct comparison of different metal-carbonbonds on the same metal center. A determination of which bond is more reactive toward a givensubstrate would then be easily made by observing which bond reacts preferentially. Such asystem would be more convenient than comparing two metal complexes with different ligands bymeasuring how fast each reacts with the substrate. This is the advantage of mixed-ligandsystems, and it is the main reason we endeavored to find a synthetic route to complexes of thistype.Before the work presented in this Chapter was performed, there was no general route tosuch complexes, and in fact, during our previous work we had been able to synthesize only onemixed alkyl species. This complex, CpW(NO)(CH2SiMe3)CHPhwas prepared by thereaction of the symmetric alkyl CpW(NO)(CHSiMewith trityl cation (equation 4. i).4However, this synthetic route could not be extended to produce the range of mixed complexesneeded for a comprehensive study. Nevertheless, the thermal chemistry of this complex turnedout to be quite interesting, as it eliminated TMS upon heating to form a metallacycle that couldbe trapped with Lewis bases (equation 4.1).Ph3CPF6 A, PR3w w/ / -TMS R3P 4.1RR RCH2Ph3 PhR = CH2SiMe3The obvious route to the desired mixed complexes is through the alkyl or aryl chlorideintermediates, Cp*W(NO)(R)Cl. Until recently, however, a general preparation of these92desirable intermediates was not known, and all attempts at their synthesis resulted in failure (seeSection 3.1). The successful synthesis of a series of alkyl and aryl chlorides presented in Chapter3 has given us the potential of creating a variety of mixed alkyl and aryl complexes. This Chapterdescribes the detailed synthesis of a series of mixed complexes Cp*W(NO)(R)R [R R’; R, R’ =alkyl, aryl] and discusses their characterization and properties. Some thermal chemistry of onederivative, Cp*W(NO)(CH2Me3)Ph,is also investigated.Previously, members of our group have studied the reactions of the symmetric alkyl and arylspecies with CO.5 The products of these reactions depend somewhat on the alkyl or aryl ligand,but in most cases, the mono-inserted product is isolated (equation 4.2). Occasionally, COadducts of the starting materials can be observed spectroscopically, implying that the insertionreactions occur after prior coordination of carbon monoxide to the metal center. The products ofthese reactions contain2-acyl ligands, which give them an 18-valence-electron configuration.Nevertheless, some of these complexes react further with a higher pressure of CO to yield bisinserted products (equation 4.2).S7I CO (1 atm) I CO (6 atm) IM M°- 0 M 42R”I”R R”I\N N\ /N\0 OR R ORThis reactivity with CO has been extended to a series of mixed alkyl and aryl complexesprepared in this Chapter, and the results of this study are presented and discussed here. Thedetermination of which alkyl or aryl ligands insert preferentially leads to a relative ranking of themigratory aptitudes of the different ligands involved in the study. These results are compared toliterature reports of similar insertion studies, and also to reported trends derived from kineticstudies of carbonylation reactions.934.2 Experimental Procedures4.2.1 MethodsThe synthetic methodologies employed throughout this thesis are described in detail inSection 2.2.1.Many of the complexes described in this Chapter were synthesized using similarmethodology. The syntheses are therefore grouped accordingly, with a representative examplebeing described in detail for each procedure. All solid reagents such as Cp*W(NO)(Cl)2andR2MgX(dioxane) were weighed into a Schlenk tube contained in a glovebox. A magnetic stirbarwas added, the stoppered tube was removed from the box, and subsequent chemistry wasperformed on a vacuum line. Unless indicated otherwise, solvents were added via syringe to thereaction vessel. Isolated yields, and the spectroscopic and physical properties of all complexesare collected in Tables 4.1 - ReagentsThe MgX(dioxane) reagents [R = CH2Me3,CH2MePh,CH2SiMe3Ph, o-tolyl, Me]were prepared by published procedures.6 PMe3 (Strem) was vacuum transferred fromNalbenzophenone. CO (Matheson), benzophenone (BDH), and 1-pentyne (Aldrich) were usedas received. p-Xylene and cyclohexanone (BDH) were vacuum transferred directly fromP205.Phenylacetylene (Baker) was passed down an alumina column and degassed prior to use.944.2.3 Synthesis4.2.3.1 Preparation of Cp*W(NO)(R)R ComplexesCp*W(NO)(CHS1Me)M (4.1), Cp*W(NO)(CH2SiMe3)Ph(4.3),Cp*W(NO)(CHSiMe)CH(4.4), Cp*W(NO)(CHMe) (4.5),Cp*W(NO)(CH2Me3)Ph(4.7), Cp*W(NO)(CHSi eCHPh(4.11) andCp*W(NO)(CHe)C (4.12) were prepared in a similar manner. The synthesis of4.4 is described as a representative example. THF (15 mL) was vacuum transferred onto amixture of Cp*W(NO)(Cl)2(0.840 g, 2.00 mmol) and(CH2CMe3Mg•X(dioxane) (0.254 g,1.00 mmol) at -196 °C. The reaction mixture was then allowed to warm to room temperaturewhile being stirred. The solvent was removed under reduced pressure, and the purple residuewas left under dynamic vacuum for 2 h. Et20 (30 mL) was added to the residue, and theresulting solution was cooled to -30 °C. A solutionof(CH2SiMeMgX(dioxane) (0.286 g,1.00 mmol) in Et20 (20 niL) was slowly cannulated into the reaction vessel. The resultingmixture was stirred as it was allowed to warm to room temperature. Approximately one-half ofthe solvent was removed under vacuum, and the remaining red solution was filtered throughFlorisil (100-200 mesh, 2 x 3 cm) supported on a fit. More Et20 (20 mL) was used to wash theFlorisil column. The combined filtrates were taken to dryness, and the solid was dissolved in aminimum amount of pentane. Cooling of this solution to -30 °C overnight induced thecrystallization of maroon needles ofCp*W(NO)(CH2SiMe3)CH(4.4) (0.80 g, 79% yield)which were isolated by cannulation.Cp*W(NO)(CH2SiMe3)(otolyl) (4.2), Cp*W(NO)(CHMe)(otolyl) (4.6) andCp*W(NO)(otolyl)Me (4.8) were prepared in a manner identical to that described for complex4.4 above. The only difference was that the product was crystallized directly from the Et20filtrate after concentration and subsequent cooling.Cp*W(NO)(otolyl)Ph (4.9) was also prepared similarly to complex 4.4 above. However,after filtration of the Et20 solution through Florisil, the solvent was removed from the filtrate in95vacuo, and the remaining solid was crystallized fromCH21/hexanes to obtain analytically pure4. Preparation of Cp*W(NO)(Ph)Me (4.10)THF (15 mL) was vacuum transferred onto a mixture of Cp*W(NO)(Cl)2(0.840 g,2.00 mmol) andPh2MgX(dioxane) (0.300 g, 1.00 mmol) at -196 °C. Warming of the stirredreaction mixture to room temperature produced a deep blue solution. This solution was thencooled to -30 °C, and a solution ofMe2gX(dioxane) (0.50 mmol, 0.080 M, 6.3 mL) in Et20was slowly cannulated into it. Warming of this solution to room temperature resulted in theformation of a deep violet solution of 4.10. All attempts to isolate 4.10 from this solution failed.For instance, removal of the solvent from the violet solution under vacuum resulted in theformation of a brown intractable solid. Nevertheless, the complex thus generated may be utilizedin situ for iijrther chemistry. Thermolysis Reactions of Complex 4.7The thermolyses of Cp*W(NO)(CH2Me3)Phin C6D,C6H,and p-xylene in the presenceofPMe3 were all performed similarly. The reaction in C6H is given as a representative example.Cp*W(NO)(CH2Me)Ph(0.15 g, 0.30 mniol) was dissolved in benzene (10 mL) andtransferred to a small, thick-walled bomb. The vessel was attached to a vacuum line, cooled to-196 °C, and evacuated. PMe3 (excess) was added by vacuum transfer, and the bomb was thensealed. The solution was heated at 110 °C for 6 hours, during which time it changed in colorfrom deep purple to yellow-orange. The solvent and excess PMe3 were removed under vacuum,and the remaining yellow powder was dissolved in a minimum ofEt20. Cooling of this solutionto -30 °C overnight induced the crystallization of complex 4.13 (0.13 g, 71 % yield).Similar thermolysis inp-xylene instead of benzene led to the isolation of 4.14 (32% yield).96Monitoring of the above reactions, and others involving different trapping ligands, wasperformed by 1H NIvIR spectroscopy. Cp*W(NO)(CH2Me3)Ph(0.030 g) and approximatelytwo equivalents of the trapping agent were dissolved in C6D in an NMR tube equipped with aTeflon stopcock. A 1H NMR spectrum was recorded, and the solution was then heated at110 °C until a color change was observed (usually 3 - 5 hours). An NMR spectrum was againrecorded and compared to the initial spectrum. If a large amount of starting material was stillpresent, the thermolysis was continued until all of it was consumed, as judged by NMRspectroscopy. Volatiles, if present, could then be removed by attaching the NMR tube to avacuum line and removing them under vacuum. The residue was then redissolved in C6D,andthe spectrum was again recorded. In this manner, the following reagents were tested as possibletrapping agents: cyclohexanone, 1-pentyne, benzophenone, and phenylacetylene. Reactions withCD3N and acetone-d6were carried out in neat solvent. None of the reactions produced onlyone product, and most lead to total decomposition of the organometallic reactant. NMR Tube Carbonylations of Complexes 4.1 - 4.9Approximately 0.03 0 g of each complex was dissolved in 0.5 mL ofC6D contained in anNrvIR tube bearing a Teflon stopcock. The solution was freeze-pump-thaw degassed twice, andits 1H NMR spectrum was recorded. Carbon monoxide (1 atm) was then introduced into thetube through the stopcock, and the solution was shaken until the reaction with CO was complete,as evidenced by the disappearance of the color of the starting material. The final solution wasthen degassed once, and its 1H NMR spectrum recorded. Preparation of Cp*W(NO)(q2{O}R)Rt Complexes 4.4’ - 4.9’The synthesis ofCp*W(NO)(r1_C{O}CHMe3) (4.5’) is described as a representativeexample. Cp*W(NO)(CH2Me3) (0.20 g, 0.46 mmol) was dissolved in benzene (10 mL) in aSchlenk tube to produce a deep red solution. Carbon monoxide (1 atm) was introduced into thevessel, whereupon the solution turned yellow within minutes. The solvent was removed in97vacuo, and the remaining orange powder was washed with cold pentane (2 x 5 mL). The crudeproduct was then crystallized fromEt20/hexanes to obtain yellow crystals of 4.5’ (0.19 g, 88%yield).Cp*W(NO)(q2{ 0 }CH2Me3)(o-tolyl) (4.6’), Cp*W(NO)(fl2{ 0 }CH2Me3)Ph(4.7’),Cp*W(NO)(rl{O}otolyl)Me (4.8’), and Cp*W(NO)(1l{O}otolyl Ph (4.9’) were preparedin a maimer identical to that described for 4.5’ in the preceding paragraph. The sole differencewas thatCH2lIhexanes rather thanEt20/hexanes were used for the final crystallizations.Cp*W(N0)(fl{O}CHMe3) HSi (4.4’) was prepared similarly to 4.5’ with theexception that the crude product was not washed with pentane. Instead, it was dissolved in aminimum of pentane and was placed in the freezer (-30 °C) to induce crystallization. Preparation of Complexes 4.1’ - 4.3’The syntheses of Cp*W(NO)(OC { CH } SiMe3)M (4.1’),Cp*W(NO)(OC { CH }SiMe3)(o-tolyl) (4.2’), and Cp*W(NO)(OC { CH } SiMe3)Ph (4.3’) wereperformed in a similar manner to those of the acyl complexes in the preceding paragraphs.However, one difference was that a minimum of CO was used during these reactions.Specifically, the atmosphere of CO was removed in vacuo as soon as the color of the startingmaterial had disappeared. Solvent removal from the final reaction mixture afforded in each casea white powder which was washed once with cold pentane (5 mL). Complexes 4.1’ and 4.3’were crystallized from hexanes, while 4.2’ was crystallized fromCH21/hexanes. All threecomplexes formed clear, colorless crystals. Preparation of Cp*W(NO)(q2{O}Ph)Me (4.10’)Carbon monoxide (1 atm) was introduced into a Schlenk tube containing a solution of 4.10generated as described above. The solution quickly changed from deep purple to orange. The98solvent was removed under reduced pressure, and the residue was extracted with pentane(100 mL). The extracts were filtered through Celite (2 x 4 cm) supported on a fit, and thefiltrate was concentrated and cooled (-30 °C). The orange powder that precipitated was isolatedby cannulation, and was redissolved in pentane. Cooling of this solution to -30 °C overnightinduced the deposition of 4.10’ as an orange powder.994.2.4 Characterization Data for Complexes 4.1 - 4.14 and 4.1’ - 4.10’Table 4.1. Numbering Scheme, Color, Yield and Elemental Analysis Data forComplexes 4.1 - 4.14, Cp*W(NO)(R)R.R R’compd color anal. found (calcd)no. (yield %) C H NCH2SiMe3 CH3 4.1 violet (38) 39.92(39.76) 6.48(6.49) 3.10(3.11)CH2SiMe3 o-tolyl 4.2 violet (31) 47.82(48.08) 6.3 1(6.36) 2.66(2.73)CH2SiMe3 Ph 4.3 purple (38) 46.79(47.11) 6.09(6.24) 2.73(2.56)CH2SiMe3 CH2Me3 4.4 maroon (79) 44.97(44.72) 7.35(7.29) 2.76(2.87)CH2Me3 CH3 4.5 red (53) 44.15(43.67) 6.72(6.54) 3.22(3.36)CH2Me3 o-tolyl 4.6 purple (44) 51.67(51.93) 6.50(6.55) 2.74(2.72)CH2Me3 Ph 4.7 red (55) 50.72(50.68) 6.28(6.38) 2.82(2.78)o-tolyl CR3 4.8 violet (13) 47.49(47.60) 5.54(5.40) 3.08(2.99)o-tolyl Ph 4.9 blue (20) 53.40(53.43) 5.26(5.31) 2.71(2.80)Ph CH3 4.10 purple (-) - - -CH2SiMe3 CH2MePh 4.11 maroon (61) 50.85(50.62) 6.90(6.99) 2.46(2.50)CH2Me3 CH2MePh 4.12 maroon (40) 54.28(54.25) 7.13(7. 10) 2.42(2.53)Ph Ph 4.13a yellow (71) 52.06(51.82) 5.92(5.92) 2.48(2.42)Ph 2,5-MeC6H3 4.14a yellow (34) 52.98(53.38) 6.30(6.3 1) 2.28(2.3 1)a these complexes have the molecular formula Cp*W(NO)(R)(R’)PMe.100Table 4.2. Numbering Scheme, Color, Yield and Elemental Analysis Data for theCarbonylation Products 4.1’ - 4.10’, Cp*W(NO)(R)(X).Rcompd color anal. found (calcd)no. (yield %) C H NCH3 OC(CH2)SiMe3 4.1’ colorless (45) 40.09(40. 14) 6.10(6.17) 2.92(2.97)o-tolyl OC(CH2)SiMe3 4.2’ colorless (24) 47.57(47.27) 5.99(5.83) 2.52(2.56)Ph OC(CH2)SiMe3 4.3’ colorless (20) 46.58(46.83) 5.77(5.85) 2.59(2.54)CH2SiMe3 C(O)CH2Me3 4.4’ orange (46) 44.69(44.67) 6.94(7.07) 2.61(2.54)CH3 C(O)CH2Me3 4.5’ yellow (88) 44.07(43.9 1) 6.3 1(6.48) 3.02(2.97)o-tolyl C(O)CH2Me3 4.6’ yellow (50) 5 1.21(50.95) 6. 17(6.24) 2.60(2.55)Ph C(O)CH2Me3 4.7’ yellow (95) 50.29(50.53) 5.95(5.90) 2.67(2.74)CH3 C(O)o-tolyl 4.8’ orange (52) 47.22(46.98) 5.21(5.32) 2.90(3.01)Ph C(O)o-tolyl 4.9’ orange (75) 52.86(52.59) 4.99(4.98) 2.57(2.54)CH3 C(O)Ph 4.10’ orange (10) 45.16(46.07) 4.90(4.94) 2.88(2.99)101Table 4.3. Selected Mass Spectral and Infrared Data for Complexes 4.1 - 4.14.compd MS Tempb ER (Nujol mull).m’Za (°C) UNO (cm’)4.1 451 [Pj 100 15414.2 527 [Pj 100 153943 513 [Pj 180 154744 507 [Pj 100 1555477 fP-NO]4.5 435 [Pj 180 15844.6 511 [Pj 100 15454.7 497 [Pj 100 15494.8 455 [Pj 150 153649 517 [Pj 150 15494.11 569 [P-NO] 100 15534.12 553 [Pj 180 15704.13 503 [P-PMe] 150 15374.14 607 [Pj 80 1538a m/z values are for the highest intensity peak of thecalculated isotopic cluster, i.e. ‘84Wb Probe temperatures102Table 4.4. Selected Mass Spectral and Infrared Data for Complexes 4.1’ - 4.10’.compd MS Temp1’ IR (Nujol mull)no. n,JZa (°C) (cnr1)4.1’ - - 16441576 (°No)4.2’ - - 1645 (u)1574 (uNo)4•3’ - - 1642 (t,-)1576,155144’ 535 [Pj 150 1563, 1537507 [P-CO]45’ 463 [Pj 120 1570, 1560 1537435 [PCO]4.6’ 539 [Pj 180 1576, 1564511 [P-CO] 153747’ 525 [Pj 180 1570, 1557497 [F1C0]4.8’ 483 [Pj 120 1549, 1526455 [PCO]49’ 545 [Pj 120 1574, 1553517 [P-CO] 15284.10’ 470 [P+Hic - 1549, 1520a rn/i values are for the highest intensity peak of the isotopic cluster, i.e. 184Wb Probe temperaturesC FABMS, 3-mtrobenzylalcohol matrix103Table 4.5. NMR Data for Complexes 4.1 - 4.14 in C6D.compd 1H NMR 3C {1H} NMRno.4.1 2.03 (d, HH = 12 Hz, 1H, CH2) 109.91 (C5Me)1.47 (s, 15H, C5Me) 72.47 (Me)1.02 (d, 2HH = 12 Hz, IH, CH2) 41.31 (CH2)0.57 (s, 3H, Me) 9.67 (C5Me)0.36 (s, 9H, SiMe3) 2.58 (SiMe3)4.2 7.30 (m, 4H, ArH) 129.96, 128.97, 122.87 (C1)2.78 (br, 3H,C64Me) 112.17 (C5Me)1.56 (s, 18H,5Me+M ) 26.05 (C6H4Ivfe)0.23 (s, 9H, SiMe3) 10.25 (CMe)2.37 (SiMe3)4.3 7.71 (d, 3HH = 6 Hz, 2H, o-ArH) 188.93, 134.45, 128.47, 127.77 (C1)7.2 (m, 3H, m/p-ArH) 111.45 (CMe)2.25 (d, 2HH = 15 Hz, 1H, Cl2) 71.76 (CH2)1.60 (s, 15H, C5Me) 9.96 (C5Me)0.30 (s, 9H, SiMe3) 2.85 (SiMe3)-0.55 (d, 2HH = 15 Hz, 11, CH2)4.4 3.28 (d, 2HH = 15 Hz, 1H, CH2) 109.91 (C5Me)1.54 (s, 15H, C5Me) 105.77 (CH2CMe31.35 (s, 9H, CMe3) 51.13 (CHSiMe)1.07 (d, HH = 15 Hz, 1H, CH2) 39.82 (CMe3)0.38 (s, 9H, SiMe3) 34.24 (CMe3)-1.32 (d, 2HH = 15 Hz, 1H, CH2) 9.87 (C5Me)-2.08 (d, 2HH = 15 Hz, 1H, CH2) 2.86 (SiMe3)4.5 4.13 (d, 2HH = 18 Hz, 1H, CH2) 120.91 (CH2)1.52 (s, 15H, C5Me) 109.56 (C5Me)1.34 (s, 9H, CMe3) 40.27 (CMe3)0.54 (s, 3H, Me) 33.93 (CMe)-3.55 (d, 2HH = 18 Hz, 1H, CH2) 29.23 (Me)9.70 (C5Me)4.6 7.16 (d, 3HH = 6 Hz, 111, o-ArIT) 130.61, 128.17, 123.0 (br, C1)7.0 (m, 3H, m/p-ArH) 110.93 (C5Me)2.7 (br, 3H,C6H4Me) 40.9 (br, CMe3)1.52 (s, 15H, C5Me) 34.17 (CMe3)1.43 (s, 9H, CMe3) 27.2 (br,6H4Me)9.86 (C5Me)4.7 7.69 (d, 3HH = 6 Hz, 2H, o-ArH) 182.14, 137.24, 128.02, 127.64 (C1)7.1 (m, 3H, rn/p-ArM) 122.25 (CH2)4.48 (d, 2HH = 15 Hz, 1H, CH2) 110.90 (C5Me)1.53 (s, 15H, C5Me) 41.28 (CMe3)1.22 (s, 9H, CMe3) 33.99 (CUe3)-2.07 (d, 2HH = 15 Hz, 1H, CH2) 10.00 (C5Me)1044.8 7.25 (m, 311, o/m-ArH) 149.20, 130.06, 129.336.72 ( HH = 6 Hz, 1H, p-ArH) 129.28, 122.63, 115.98 (C1)2.68 (s, 3H,C64Me) 110.81 (C5Me)1.44 (s, 15H, C5Me) 41.06 (Me)1.13 (s, 3H, Me) 26.10 (CH4Mre)9.85 (CMe)4.9 8.16 (d, 3HH = 6 Hz, 2H, ArH) 150.43, 141.54, 130.067.97 (d, 3HH = 6 Hz, 1H, ArH) 135.41, 129.98, 129.567.1 (m, 611, ArH) 127.93, 122.84, 119.81 (C1)2.83 (s, 311, C64Me) 112.14 (C5Me)1.58 (s, 15H, C5Me) 26.80 (C6HMe)10.01 (C5Me)4.11 7.60 (d, 3HH = 6 Hz, 211, o-ArH) 153.31 (C10)7.27 (t, HH = 6 Hz, 2H, m-ArH) 128.34, 126.36, 125.54 (C1)7.13 (t, HH = 6 Hz, 1H, p-ArH) 110.04 (C5Me)3.41 (d, 2HH = 13 Hz, 1H, CH2) 104.48 (CMe2Ph)1.88 (s, 3H, Me) 55.04 (CHCMePh)1.84 (s, 3H, Me) 46.07 (CH2SiMe3)1.48 (s, 15H, C5Me) 32.82 (CMePh)1.24 (d, 2HH = 11 Hz, 1H, CH2) 9.64 (C5Me)0.43 (s, 9H, SiMe3) 2.84 (SiMe3)-1.46 (Vt, HH 12 Hz, 2H. two CH2)4.12 7.51 (d, 3HH 8.4 Hz, 2H, o-ArH) 153.61 (C10)7.22 (t, HH = 7.8 Hz, 2H, m-ArH) 128.41, 126.30, 125.45 (C1)7.06 (t, HH = 7.8 Hz, 1H,p-ArH) 109.70 (C5Me)2.96 (d, 2HH = 12.6 Hz, 1H, CH2) 102.01 (CH2CMePh)2.72 (d, HH = 13.2 Hz, 1H, CH2) 92.58 (CHSiMe3)1.84 (s, 3H, Me) 45.41 (CMe2Ph or CMe3)1.75 (s, 3H, Me) 39.63 (CMePh or CMe3)1.44 (s, 15H, C5Me) 34.22 (CMe3)1.27 (s, 9H, CMe3) 34.05, 32.65 (Clvfe2Ph)0.77 (dd, HH 13.2, 3 Hz, 1H, CH2) 9.68 (C5Me)-1.92 (dd, HH 12.6, 3 Hz, 1H, 2.84 (SiMe3)CH2)4.13a 759 (d, HH = 8 Hz, 211, ArH) 168.24 (d, = 30 Hz, C0)7.13 (m, 4H, ArH) 140.80, 139.68, 128.437.02 (m, 4H, ArH) 126.59, 123.60 (C1)1.51 (s, 1511, C5Me) 110.76 (C5Me)0.68 (d, PH = 9 Hz, 9H, PMe3) 15.78 (d, J = 29 Hz, PMe3)10.03 (C5Me)7.57(d,JHH8Hz,2H, ArH)7.1 (m, 4H, AiR)6.66 (d, HH =8 Hz, 1H, ArH)6.56 (s, 1H, ArH)2.39 (s, 3H,C63Me2)2.24 (s, 3H, Me1.49 (s, 15H, C5Me)0.72 (d, PH = 9 Hz, 9H, PMe3)167.85 (cL, Jp = 32 Hz, C10)166.87 (d,‘c = 32 Hz, C0)140.995, 140.79, 140.56140.31, 140.23, 140.16130.99, 128.01, 126.45124.42, 123.51 (C1)111.26 (C5Me)26.50, 20.95 (C6H3Me2)9.68 (C5Me)15.20 (d, Jp =29 Hz, PMe3)10.17 (C5Me)414a,b105a In CDCI3. b Major isomer.106Table 4.6. NMR Data for Complexes 4.1’ - 4.10’ in C6D.compd 1fl N!VER 13C{H} NM.R4.37 and 5.80 (s, 1H each, CH2) 182.82 (C=CH)1.73 (s, 1511, C5Me) 111.02 (C5Me)1.15 (s, 311, Me) 66.35 (CH2)0.82 (s, 911, SiMe3) 15.40 (Me)10.28 (C5Me)-1.21 (SiMe3)7.77 (d, 3HH 6 Hz, 1H, o-ArH) 184.85 (CCH2)7.2 (m, 311, m/p-ArH) 171.43 (C10)4.30 and 5.93 (s, 1H each, CH2) 147.05, 139.56, 129.512.63 (s, 3H, Me) 125.72, 124.93 (C,1)1.68 (s, 15H, C5Me) 112.66 (C5Me)0.04 (s, 9H, SiMe3) 66.26 (CH2)28.03 (CH4Me)10.70 (C5Me)-1.18 (SiMe3)8.03 (t, 3HH = 8 Hz, 2H, o-ArH) 183.76 (CCH2)7.37 (t, HH = 8 Hz, 2H, m-ArI{) 169.5 118 Hz)7.2 (t, HH = 8 Hz, 1H, p-ArH) 139.8, 127.8, 125.5 (C1)4.37 and 5.97 (s, 1H each, CH2) 112.50 (C5Me)1.69 (s, 15H, C5Me) 67.18 (CR2)0.00 (s, 9H, SiMe3) 10.72 (C5Me)-1.20 (SiMe3)2.64 (d, 2HH = 13 Hz, 1H, CH2Me3) 294.8 (C0, = Hz)3.35 (d, HH = 13 Hz, 1H, CHMe 108.32 (C5Me)1.58 (s, 15H, C5Me) 53.48 (CHCMe1.03 (s, 9H, CMe3) 33.63 (CMe)0.48 (s, 9H, SiMe3) 30.04 (CMe3)-0.07 (d, 2HH 13 Hz, 1H, 14.76 (CH2SiMe,‘J = 87 Hz,CH2SiMe3) = 51 Hz)0.36 (d, 2HH = 13 Hz, 1H, CH2SiMe3) 9.88 (C5Me)3.46 (SiMe3)2.78 (d, 2HH = 13 Hz, 111, CH2) 296.22 (CO)3.06 (d, HH = 13 Hz, 1H, CH2) 108.90 (C5Me)1.67 (s, 1511, C5Me) 53.44 (CH2)1.27 (s, 3H, Me) 33.33 (CMe)1.09 (s, 9H, CMe3) 29.98 (CUe)11.04 (Me, wc = 88Hz)9.74 (C5Me)1074.6u2 7.16 (d, 3HH = 6 Hz, 1H, o-ArH) 295.2 (br, C=O)7.0 (m, 3H, m/p-AtH) 148.0 (br), 138.85 (br), 129.983.07 (d, 2HH = 12 Hz, 1H, CH2) 125.48, 123.79 (C1)3.57 (d, 2HH 12 Hz, 1H, CH2) 110.28 (C5Me)2.34 (br, 3H,C64Me) 53.67 (CH2)1.87 (s, 15H, C5Me) 33.98 (CMe3)1.13 (s, 9H, CMe3) 30.08 (CAfe3)10.24 (C5Me)4•-pa 7.65 (d, 3J = 6 Hz, 2H, o-ArH) 297.0 (0=0)7.26 (t, 3J = 6 Hz, 2H, m-ArH) 171.52 (C10)7.12 (t, HH = 6 Hz, 1H, p-ArH) 139.18, 127.87, 125.02 (C1)3.11 (d, HH = 12 Hz, 1H each, CH2) 110.31 (C5Me)3.46 (d, 2HH 12 Hz, 1H each, CH2) 53.47 (CH)1.84 (s, 15H, C5Me) 33.88 (CMe3)1.14 (s, 9H, CMe3) 30.04 (CMe3)10.22 (C5Me)4•8’a 7.97 (d, 3HH = 7 Hz, 1H, o-Ar}{) 282.8 (0=0)7.62 (t, 1H, m-ArH) 181.0 (C180)7.51(t3,JHH=7Hz, 1H,p-ArH) 141.0, 136.53, 135.487.41 (d, HH = 7 Hz, 1H, m-ArH) 131.85, 127.06 (C1)2.65 (s, 311, C64Me) 109.30 (C5Me)1.85 (s, 15H, C5Me) 20.36 (CH1M’e)0.73 (s, 3H, Me) 10.50 (Me, Jwc 90 Hz)9.99 (C5Me)4•9va 8.07 (d, 3HH = 8 Hz, 1H, ArH) 281.9 (C”O)7.79 (d, 3HH = 8 Hz, 2H, ArH) 153.0(C10-6H4Me)7.66 (t, 3HH 8 Hz, 111, ArH) 149.2 (_C5)7.55 (t, 3HH 8 Hz, 1H, ArH) 139.63 (0,..-H7.44 (d, 3HH 8 Hz, 1H, ArH) 136.38(C00-4Me)7.29 (t, 3HH 7 Hz, 2H, o-C6H5) 135.00, 131.48, 127.697.12 (t, HH 7 Hz, 1H, o-C4Me) 126.67, 125.26 (C1)2.74 (s, 3H,C64Me) 109.61 (C5Me)1.79 (s, 15H, C5Me) 20.23 (CH4Me)9.87 (C5Me)4.10’ 7.91 (d, 3HH = 7 Hz, 2H, o-ArH) 280.63 (C=O)7.1 (m, 311, rn/p-ArE) 135.13, 131.96, 129.28, 127.68 (C1)1.62 (s, 1511, C5Me) 108.38 (C5Me)1.30 (s, 3H, Me) 11.95 (Me, 1Jwc = Hz)9.84 (C5Me)a CD211084.3 Results and Discussion4.3.1 Synthesis of the Mixed Complexes Cp*W(NO)(R)RtWe initially undertook this study in order to synthesize a series of mixed alkyl and arylcomplexes, Cp*W(NO)(R)R, to be used in an investigation of the relative reactivities of thedifferent metal-carbon a-bonds at the same metal center. The choice of alkyl groups was limitedto those that did not contain 3-hydrogens, because compounds containing these ligands readilydecompose, presumably via f3-H elimination pathways.7 Consequently, methyl, neopentyl,neophyl and trimethylsilylmethyl ligands were chosen as representative alkyl groups. Phenyl ando-tolyl were chosen as representative aryl ligands because of their differing steric properties, afeature that could possibly be an important factor in comparative reactivity studies.The Cp*W(NO)(R)Rt complexes 4.1 - 4.12 can be prepared in a one-pot synthesis bysequential metathesis reactions using two different alkylating or arylating reagents, or they can beprepared from isolated alkyl or aryl chlorides (see Chapter 3). The most convenient route, theone-pot synthesis, is used to make all of the mixed species presented in this Chapter (equation4.3). The choice of which alkyl or aryl chloride intermediate should be generated first in anygiven synthesis is somewhat arbitrary. Although all of the complexes containing a methyl groupwere synthesized by introducing it in the second step, all of the mixed complexes can presumablybe made from either sequential route.I R2Mghoane I R’2Mgdioxane IW W W 4.3N” I ‘‘Cl THF N” I Cl THF N” I R’0 Cl 0 R 0 R4.1 - 4.12The mixed complexes 4.1 - 4.12 have all been fully characterized, except for 4.10,Cp*W(NO)(Ph)Me, which has defied isolation. Nevertheless, this latter compound can be109generated in situ using the same methodology as for the other mixed species. Evidence for thepresence of 4.10 in such solutions comes from the reaction of these solutions of 4.10 with Co.As discussed later, the expected acyl product Cp*W(NO)(rl2{O}Ph)Me (4.10’) can beisolated, thereby implying the existence of 4.10.All of the isolable 16-electron mixed complexes are air- and moisture-sensitive to varyingdegrees. The most stable members of this class of complexes are those with bulky alkyl ligands,a property previously observed for the related symmetric dialkyl and diaryl species,Cp*W(NO)R2.3This observed trend may help explain why the phenyl methyl derivative, 4.10,cannot be isolated; the small bulk of the two ligands render the complex prone to decomposition.The presence of the aryl group may add to the instability of the complex. We have noticed in thepast that the symmetric diaryl species are the most difficult to synthesize because of theirinstability, a feature which has been associated with the greater electron-withdrawing ability ofthe aryl ligands.3No systematic synthetic study of a series transition-metal complexes containing two differentmetal-carbon a-bonds has appeared in the literature. There are a few isolated examples of suchcomplexes, although in most cases the synthesis reported is not completely general and does notlead to a series of similar compounds. For example, Jordan and co-workers have synthesized thezirconocene complexes, Cp2Zr(Ph)R [R = Me or CH2Ph], and then investigated the comparativeprotonolysis of their metal-carbon bonds.8 Other selected examples includeCp2Zr(CH(SiMe3)R[R = Me, Et, “Pr, CH2SiMe3],9Cp*Re(O)(Me)R[R = Et, CH2SiMe3CH2Me3I,lOa(r5-indenyl)Ir(PMe)(Me)Ph,lob and Cp*2Ti(Me)R[R = Et, CH=CH2Ph, CH2Phj.lOC4.3.2 Characterization of Comp’exes 4.1 - 4.12All of the mixed complexes, excluding 4.10, have been characterized by standardspectroscopic techniques, and all data are consistent with the Cp*W(NO)(R)R1 species beingmonomeric, 16-electron complexes.110One notable characteristic of the 1H and 13C NMR data of the complexes containing ano-tolyl ligand is that evidence of hindered rotation of the o-tolyl group about the ipso carbon-tungsten bond is present. Resonances due to both the ortho and methyl protons of the o-tolylligand appear broadened at room temperature. This phenomenon is mirrored in the 13C NIVIRspectra where the ortho carbon and the ortho methyl carbon signals are also broadened, even tothe extent of not being observable in the spectrum of complex 4.6,Cp*W(NO)(CH2Me3)(otolyl). This broadening is attributed to slow rotation (on the NMRtimescale) about the ipso carbon-tungsten bond so that the atoms affected are exchangedbetween different environments. Consistently, the broadening of these peaks becomes morepronounced as the steric congestion at the metal center increases. Thus, while the signalassignable to the methyl protons on the o-tolyl ligand in the 1H NMR spectrum ofCp*W(NO)(otolyl)Cl (3.3) is sharp, the same peak in Cp*W(NO)(CH2e3)(otolyl) (4.6) is sowide that it is barely discernible from the baseline. To try to determine if this rotation could beincreased to effectively coalesce the spectrum, complex 4.6 has been subjected to a high-temperature NMR study. Warming of a solution of this compound to 80 °C, however, has littleeffect on the 1H NMR spectrum, so that the high-temperature spectra are almost identical tothose taken at room temperature.4.3.3 Structural Determination of Cp*W(NO)(CH2Me3)(otolyl) (4.6)Cp*W(NO)(CH2Me)(otolyl) (4.6) has been subjected to an X-ray diffraction study as arepresentative example of the mixed-ligand complexes. This specific compound was chosen forthe study because of the odd NMR behavior mentioned above. We had initially thought that thisbehavior could possibly be a result of an agostic interaction of the o-tolyl methyl group with themetal center. The solid-state molecular structure of 4.6 is shown in Figure 4.1. It is obviousfrom the structure that there are no agostic interactions present in the solid state. In fact, themethyl group of the o-tolyl ligand is pointing toward the nitrosyl and away from the largest gapbetween the three legs of the piano-stool molecule. This orientation of the o-tolyl111Figure 4.1. View of the solid-state molecular structure ofCp*W(NO)(CH2SiMe3)(otolyl)(4.6), including selected bond lengths and angles. Disorder in the crystal precludes the listingof meaningful esds at present.Bond Lengths (A)W-N 1.75W-C(H2) 2.12W-C1,0 2.16N-O 1.23Bond Angles (deg)W-N-O 170N-W-C(H2) 97N-W-C10 99- W - C(H2) I 1 1112ligand has also been observed in other similar compounds that have been structurallycharacterized, including Cp*M(NO)(otolyl)2[M = Mo, WI.64.3.4 Thermal Chemistry of Cp*W(NO)(CH2Me3)Ph(4.7)Recently, our laboratories have investigated the thermal chemistry of the symmetric bis(neopentyl) compound, CpMo(NO)(CH2Me3)and found that upon heating the complex losesneopentane to form a transient alkylidene complex. This alkylidene species can be trapped with anumber of Lewis bases to give complounds of the type CpMo(NO)(CHCMe3)L[L = PMe3,PMePh2,pyridine].ll Similar reactions have been performed usingCp*W(NO)(CH2Me)in the hope of trapping a tungsten alkylidene complex. Thermolysis ofthe tungsten dialkyl in the presence ofPPh3 at 80 °C leads only to decomposition, as determinedby spectroscopic monitoring of the reaction.It was then hoped that some of the mixed complexes, Cp*W(NO)(R)Rt, might thermallydecompose cleanly to form an alkylidene species. A few such reactions have been carried out inNMR tubes to determine if any interesting products are formed in good yield. Most of thereactions have been performed both with and without the addition of a trapping base, and all ofthe experiments have been perfomed in C6D so that they could be monitored by 1H NMRspectroscopy. Thus, complexes 4.2, 4.4, 4.6, 4.12, and also 3.5 have been thermolyzed in thisway in the presence of various Lewis bases. No major products are formed during any of thesethermolyses reactions. It should be noted here that this is in no way a comprehensive survey ofthe thermal chemistry of the mixed alkyls and is not meant as such. Further investigation couldquite possibly uncover a high-yielding route to a tungsten alkylidene complex or other interestingthermal chemistry.One of the mixed species, however, does show some interesting thermal chemistry. WhenCp*W(NO)(CH2Me3)Ph(4.7) is thermolysed in C6D at 80 °C, neopentane elimination isobserved, but no major organometallic product is formed, judging by the plethora of peaks in theCp* region of the 1H NIvIR spectrum. The appearance of neopentane, with no benzene113liberation, is encouraging, however, and so the reaction has been repeated in the presence of anexcess ofPMe3 in an attempt to trap the product of alkane elimination. No reaction occurs at80 °C, but when the NMR tube is heated to 110 °C, the maroon-colored solution slowly turnsyellow-orange. A ‘H NMR spectrum of the final solution reveals that only one new Cp*resonance is present, that neopentane has been liberated, and that some of the excess PMe3 isnow coordinated to the tungsten center.I 1100c,c6H IW 44PMe3 N PMe3-CMe4 0(3 )413When the reaction is attempted on a preparative scale in C6H,a yellow crystalline productcan be isolated in good yield which has been identified as Cp*W(NO)(Ph)2PMe3(4.13) (equation4.4). The NMR data confirm that the complex contains two phenyl groups and a coordinatedPMe3, and the elemental analysis also confirms the molecular formula. Only one isomer of 4.13is present, and it is assumed to be the trans isomer based on the presence of two equivalentphenyl group resonances in the NIvIR spectra.Some evidence for the mechanism of formation of complex 4.13 has been gathered. Firstly,the thermolysis of 4.7 has also been performed in C6D,and the product observed by 1H NMRspectroscopy. The spectrum appears nearly identical to that for 4.13, the only difference beingthe integrations of the aryl region. Whereas complex 4.13 exhibits signals due to 10 aryl protons,theC6D-thermolysis product exhibits only 4 aryl protons. The integrations of the ortho, metaand para signals indicate that there is only one ortho proton in this latter product. Thus itappears that one of the phenyl ligands in 4.13 originates as solvent, and that an ortho proton onthe original phenyl group is replaced with a proton originating from the solvent.114More evidence for the ability of this system to activate aromatic solvents comes from thethermolysis performed in p-xylene. This solvent was chosen to prevent the possible formation ofisomers due to ortho, meta, or para activation, since all of the aromatic protons in p-xylene areequivalent. The thermolysis product, Cp*W(NO)(Ph)(2,5Me26H3)PMe(4.14), does indeedcontain an equivalent ofp-xylene, as is evident from the characterization data for this compound.The mass spectral and elemental analysis data confirm the molecular formulation, and the NMRspectra indicate the presence of two different aryl substituents as well as two different resonancesfor the ortho and meta methyl groups on the xylyl ligand. The NMR data also indicate thepresence of a minor isomer in approximately 10% of the concentration of the major isomer. Thisminor isomer is presumably the other possible stereoisomer, the cis configuration, that is notobserved for 4.13.Given that Cp*W(NO)(CH2Me)Ph(4.7) loses neopentane and activates aromatic solventsupon thermolysis, a mechanism for this transformation can be proposed (Scheme 4.1). The firststep of the thermolysis reaction is loss of neopentane to form a proposed benzyne intermediate.This intermediate and its PMe3 adduct are obviously unstable since neither one is isolated or evenobserved during the reaction. The intermediate benzyne complex then activates the solvent toform the bis-(phenyl) derivative, which is trapped by PMe3 to afford the very stable final product.Whether PMe3 coordinates before or after the benzyne activates solvent is unknown, but it isassumed that this activation would be easier prior to phosphine coordination, and thus is drawnthis way in Scheme 4.1. It is clear that a Lewis base is needed to isolate any product, since it isknown that the bis-(phenyl) complex is itself thermally sensitive.’2Many transition-metal benzyne complexes are known and have been reported.’3 Some ofthese complexes contain a discrete benzyne ligand.14 In other systems, the generated benzyne isreacted with unsaturated substrates, and the intermediacy of a benzyne ligand is implied from thestructure of the products.15 Attempts have been made to trap the proposed benzyne115Scheme 4.1A— CMe4PMe34C6Hintermediate formed upon thermolysis of 4.7. Acetonitrile, acetone, 1-pentyne, cyclohexanone,benzophenone, and phenylacetylene have all been used as potential trapping agents, but with nosuccess. In each case, NMR spectroscopic monitoring of the reactions reveals that no majorproducts are formed, and so the reactions have not been performed on a preparative scale. Inretrospect, this result is not that surprising, since a temperature of 110 °C is needed to initiate thealkane elimination reaction. Any trapped product that is formed, therefore, would need to bestable at this temperature in order to be observed or isolated. Experience with related dialkyl,alkyl alkoxides, and alkyl amides, which would be structurally similar to the expectedmetallacycle products, suggest that these products would not be stable at such a hightemperature.1164.3.5 Carbonylation ReactionsIsolation of the mixed compounds Cp*W(NO)(R)R1 presents the opportunity of comparingthe relative reactivity of the different metal-carbon bonds at the same metal center. From ourexperience with the symmetric dialkyls and diaryls, it is known that CO insertion into the metal-carbon bonds of these compounds is a straightforward reaction, and the products are easilyisolable. Knowing this, we decided that the initial reactivity study to be performed on the mixedcomplexes would be a comparison of the metal-carbon bond reactivities with CO. Ten of themixed compounds were chosen for the study, containing all possible combinations of the fiveligands CH2Me3,CH2SiMe3,Me, phenyl, and o-tolyl (complexes 4.1 - 4.10).The carbonylation reactions are performed by exposing benzene solutions of complexes 4.1 -4.9 to an atmosphere of CO. As discussed above, complex 4.10 could not be isolated, and so thecarbonylation of this compound has been performed by exposing to CO a solution of this mixedalkyl that had been generated in THF. In each reaction, the insertion of one equivalent of COinto one of the metal-carbon a-bonds occurs within minutes at room temperature. As indicatedCOw 4 5/\ NJi /\//N R’ 0 i’-0 N Co R RR’ 0 RR’4.4-4.10 4.4’-4.10’in equation 4.5, these processes probably proceed via initial adduct formation by the CO at themetal center’s vacant orbital which bisects the R - M - R’ angle (Section 2.3.2). Aftercoordination, the CO then undergoes competitive insertion into either metal-carbon bond to formtherj2-acyl products.Each of the carbonylation products are numbered so as to reflect the mixed complex fromwhich they originate; so for example, compound 4.5’ results from treatment of 4.5 with CO. All117of the carbonylation reactions have also been performed in NMR tubes to determine true yields.In most instances, the reactions are quantitative by NT4R spectroscopy.In each case, one equivalent of CO is incorporated into the organometallic molecule to forman ‘q2-acyl species, which is the isolated product for the reactions of complexes 4.4 - 4.10 withCO. In some cases, however, this acyl rearranges and is not isolated as such, and this is the casefor the carbonylation products of 4.1 - 4.3. The next two sections discuss the different casesseparately, and examine the spectroscopic data for each class of product.4.3.6 Spectroscopic Characterization of Complexes 4.1’ - 4.3’When the mixed alkyl complexes 4.1 - 4.3 are reacted with CO, the expected2-acylproducts are not observed. Instead, enolate products arising from a 1 ,2-silyl shift rearrangementof the initially-formed acyl species are isolated (equation 4.6). There is ample precedent for thisrearrangement.16 These types of rearrangements commonly occur with acyl intermediates atelectron-deficient and oxophilic metal centers that give thei2-acyl carbon a high degree ofcarbene-like character. 16bd Mechanistic speculation involving the proposed intermediacy of anacyl species has been discussed for these reactions,l6,d,eand in one case the observedrearrangement of an isolated acyl complex to an enolate species provides support for thismechanism. 16c No intermediate acyl compounds have been observed during the formation of4.1’ - 4.3’ since the insertion and rearrangement processes are complete within seconds atambient temperatures.Co. W 46/\ /N// /\RCH2SiMe3 R RCH2SiMe3 I4.1 - 4.3 SiMe34.1’-4.3’118The enolate species 4.1’ - 43’ are not spectroscopically or physically similar to otherr2-acyl compounds we have isolated or to the acyl products discussed in the next section. Theyare colorless and thermally unstable such that they need to be stored cold, even in the solid state.The enolate complexes also have several unique spectroscopic features that distinguish themfrom simplei2-acyls, namely (1) their lB. spectra show a sharp band at approximately 1640 cm’assignable to the C=C stretch of the enolate ligand, (2) the signal of the inserted carbon in the13C NMR spectrum appears at approximately 180 ppm while analogous ‘q2-acyl carbons resonatenear 300 ppm, and (3) the methylene proton signals originating from the trimethylsilylmethylligand appear in the vinyl region in the 1H NMR spectrum as two singlets (Jgem = 0 Hz). Sincethese enolate complexes are thermally sensitive, satisfactory mass spectral data could not beobtained. Nevertheless, comparisons of the spectroscopic data of complexes 4.1’ - 4.3’ withthose of related complexes reported in the literaturel6b,cleave little doubt as to their molecularstructures.4.3.7 Spectroscopic Characterization of Comp(exes 4.4’ - 4.10’All of the mixed alkyl and aryl complexes 4.4 - 4.10 react with CO to affordr2-acylproducts. These compounds, 4.4’ - 4.10’, are thermally stable, are yellow or orange in color, andcan be handled in air for short periods of time without decomposition. These data are consistentwith the complexes being 18-electron species possessing molecular structures analogous to thatestablished forCpW(NO)(i2-C{O}CHMePh)Ch.5It should be noted, however,that for complexes 4.4’ - 4.10’, the nitrosyl and carbonyl stretching bands in their IR spectra havenot been assigned unambiguously. Most of the complexes have at least two strong absorptions inthe region 1520 to 1560 cm1 due to the nitrosyl andi2-acyl ligands. The definitive assignmentof TB. bands is further complicated by the presence of aryl C=C stretching bands in the sameregion for those complexes containing either an o-tolyl or a phenyl group. Nevertheless, theabsence of any vCO bands at higher frequencies supports the description of these complexes as‘n2 and not1-acyl species (see also Section 3.3.5).119When complex 4.4, Cp*W(NO)(CH2Me3)CHSi is reacted with CO in a NMR tube, aspectrum recorded following the reaction indicates the presence of two products. Along withsignals attributable to the product of CO insertion into the tungsten-neopentyl bond, i.e. complex4.4’, one other Cp* signal is observable along with some other minor peaks. This minor product,present in about 20% yield, is identifiable as the product of CO insertion into the tungstentrimethylsilylmethyl bond. Its 1H NMR data17 suggest that this minor product has undergone thesame rearrangement as the other trimethylsilylmethyl-containing complexes to form an enolatespecies. The major acyl product, complex 4.4’, crystallizes preferentially from the final reactionmixture, and no attempt has been made to isolate the minor enolate product.Although it is generally easy to determine by mass spectroscopy and elemental analysis thatone carbon monoxide molecule has been incorporated into the Cp*W(NO)(R)R complexes, itproves to be difficult in some cases to determine which alkyl or aryl ligand has undergone theinsertion. Fortunately, 13C{ ‘H} NMR data can be used to answer this question. The presenceof 183W satellites on the carbon signals in the 13C{1H} NMR spectra of the carbonylationproducts indicate which carbon atoms, and therefore which ligands, are still attached directly tothe metal center, thereby identifying the other ligand as the one which has undergone COinsertion. For example, in the‘3C{’H} NMR spectrum ofCp*W(NO)(r12C{O)CHMe3)(4.5’), the signal due to the methyl carbon contains 183W satellites (‘Jwc = 88 Hz) while themethylene carbon signal of the neopentyl group does not (Figure 4.2). In some cases,satellites on the carbonyl carbon signal are also discernible, as is the case in the latter example forwhich ‘Jwc 76 Hz.Unfortunately, determining which ligand has inserted CO in complexes 4.4’ and 4.9’ is notquite as straightforward. This is due to the difficulty in unambiguously assigning the signals forthe carbons attached to the metal center. A 13C{1H} NMR spectrum with a very good signal-to-noise ratio is used to determine the structure ofCp*W(NO)(12{O}CHMe)CHSi(4.4’). The observation of 29Si satellites (‘cs = 51Hz) on one of the signals in this spectrum120( I I I I I I I I I I I j I I I I I I I300 2C 200 150 100 50 PPM CFigure 4.2. ‘3C{’H} NMR spectrum ofCp*W(NO)(r){O)CHMe3) (4.5’) in C6D.The insets show expansions of the carbonyl and methyl resonances.-.__1111111 11W I II III III IlI I H I I I I I I I100 80 60 40 20 PPM CFigure 4.3. ‘3C{’H} NI{R spectrum ofCp*W(NO)(q2C{O}CHMe3) HSi (4.4’) inC6D. Expansions of the two methylene carbon resonances are shown in the insets. Thecarbonyl resonance at 295 ppm is not shown.Jwc76Hz Jwc88HzJJ121is used to identify the resonance assignable to the methylene carbon of the trimethylsilylmethylligand. The presence of another set of sateffites due to coupling with 183W (wc 87 Hz) onthe same methylene carbon signal provides evidence that the trimethylsilylmethyl ligand has notundergone insertion and is still attached directly to the metal center. Some confirmation of thisfact comes from the NMR tube carbonylation reaction of complex 4.4, where a minor enolateproduct due to insertion into the tungsten-trimethylsilylmethyl bond has been observed.For (4.9’) it is impossible to differentiate the two ipsocarbon signals in its 13C{H} NMR spectrum. This prevents the use of the above criteria oftungsten satellite observation to determine which aryl group is still attached to the metal. Toovercome this problem, a ‘H,’3C{1H} COSY spectrum has been recorded to allow for thecorrelation of the ortho carbon signals to the ortho proton signals (Figure 4.4). These signals areeasily assignable by their differing integrations (one for o-tolyl and two for phenyl) in the 1HNMR spectrum. A‘3C{’H} NMR spectrum of this compound prepared using 13C-enriched COhas also been recorded. In this spectrum the ortho carbon signal of the o-tolyl ligand appears asa doublet(2j 7.3 Hz) due to two-bond coupling to the labeled acyl carbon, therebyindicating that it is the o-tolyl group that has undergone insertion, and not the phenyl group(Figure 4.4). Crystallographic Analysis of Cp*W(NO)(r12_C{ }CII2Me3)Ph (4.7’)In an attempt to confirm the bonding mode of the acyl ligand in the acyl complexes, oneexample has been chosen for an X-ray diffraction determination of its molecular structure.Cp*W(NO)(rj2{O}CHMe3)P (4.7’) was chosen because suitable single crystals could begrown fromCH21/hexanes. The solid state structure of this compound is shown in Figure 4.5as an ORTEP diagram.As expected, and consistent with the spectroscopic data discussed in the preceding sections,the acyl ligand is clearly bound to the metal in anii2-fashion, the carbonyl oxygen‘U.C.1HE--‘U0-IiLi‘i’’1’j’’’I”’’ IllijiBi iii1ijjiii4o 13s 136 1 4 132 130 1 6 1 6 l4 PPM13c122Figure 4.4. 1H.’3C{’H} COSY spectrum ofCp*W(NO)(rI2C{O}otolyJ)Ph (4.9’) in CD2J,aryl region only. The ortho signals of the phenyl group (.) and of the o-tolyl group (.) areindicated. The o-tolyl ortho signal appears as a doublet when the sample is prepared using 13C0.©©123Figure 4.5. View of the solid-state molecular structure of Cp*W(NO)(ri2_C{ 0)CH2Me3)Ph(4.7’) including selected bond lengths and angles (esds in parentheses).C(82)C(83)W-0(2) 2.202(3)W-C(6) 2.046(4)W-N 1.779(4)W-C(21) 2.183(4)N-O(1) 1.212(5)C(6) - 0(2) 1.236(5)Bond Angles (deg)W-N-0(1) 170.0(3)0(2)-C(6)-W 80.2(2)0(2)-W-N 103.9(1)C(6)-W-C(21) 116.5(2)C(6)-W-N 95.5(2)N-W-C(21) 93.6(2)C(11)C(2) C(1)C(15)C(13) C(14)C(25)C(26)0(2)C(7)Bond Lengths (A)C(81)124being within bonding distance of both the tungsten (2.202(3) A) and the carbonyl carbon(1.236(5) A). Not surprisingly, this oxygen donor is positioned trans to the NO group which is astrong it-acid ligand. The intramolecular dimensions of theq2-C{O}CHMe3ligand resemblethose determined previously for related systemsl6aand indicate that this ligand is acting as aformal three-electron donor to the metal center in this 1 8-valence-electron complex.4.3.8 Migratory Aptitudes of the a-Bound LigandsThere has been a vast amount of work performed in the areas of CO insertions and ligandmigrations on transition-metal complexes.18’9 The great majority of migratory insertion studieshas dealt with comparing the migratory aptitudes of different alkyl or aryl groups on a series ofmono-alkylated transition-metal complexes. Almost all of these studies, and in particular all ofthe more comprehensive investigations, have relied on kinetic data to compare the rates of COinsertions and deinsertions into the various metal-carbon bonds. 16c,19,20The work presented here is different from the kinetic studies done in the past not onlybecause ligand reactivities are being directly compared at the same metal center, but also becausethe acyl ligands formed areii2-bound in the product complexes. The direct formation of the 2.acyl ligand simplifies the reaction considerably over those that rely on incoming ligands either topromote insertion or fill the coordination site left at the metal center by the migrating ligand. Themechanism of the CO insertions discussed here is also unique among these kinetic studies. Inmost cases, the CO is already coordinated to the metal center and an incoming ligand thenpromotes insertion. With the series of mixed alkyls Cp*W(NO)(R)W, the CO is added to theunsaturated organometallic complex to form a saturated, 18-electron product. It is believed thatthe CO coordinates between the alkyl ligand before insertion (equation 4.5), and oncecoordinated, a true competition between the two alkyl ligands for migration is then achieved.The carbonylation products are then the result of a kinetic competition between the two metalcarbon bonds. Many of the carbonylation products have been heated in an attempt to induce125isomerization, but no changes which would indicate that the insertions are reversible, and that athermodynamic product ratio could be achieved have been observed.By examining the products of carbonylation of all the mixed alkyl and aryl species, 4.1 -4.10’, to determine which ligands insert preferentially, a relative ranking of the migratoryaptitudes of the a-bound ligands can be established. In order of decreasing migratory aptitude,this ranking is:CH2Me3> CH2SiMe3> o-tolyl > phenyl > MeIn other words, in all of the mixed complexes containing a neopentyl ligand, this group migratespreferentially over all other ligands upon treatment of the complex with CO. Conversely, if thecomplex contains a methyl ligand, CO does not undergo insertion into the methyl-tungsten bond,but rather into the other metal-alkyl or -aryl bond. The complexes containing the other ligandsfill in the rest of the trend. For example, CO inserts into the tungsten-o-tolyl bond inCp*W(NO)(otolyl)Ph, thereby indicating the preference to migration of o-tolyl over phenyl. Onthe other hand, the o-tolyl ligand in the complex Cp*W(NO)(CH2SiMe3)(otolyl) does not insert,thereby indicating that the migratory aptitude of this ligand is less than that of thetrimethylsilylmethyl ligand.Although the differing M-C bond strengths of the inserting groups undoubtedly play a role,the above trend seems primarily to reflect steric effects, the most sterically demanding ligandmigrating preferentially to decrease steric crowding at the metal center. The trend agrees withresults found from kinetic studies where similar ligand choices allow some comparisons to bemade. For example, in Cp3ThR systems the rate of carbonylation to formq2-acyl productsdecreases as: isopropyl > sec-butyl >> CH2Me3> n-butyl>> CH2SiMe3> methyl > benzyl. 16cA similar trend has been noted for the insertion rates in CpFe(CO)2Rin Me2SO solutions to formi1-acyls, i.e.: CH(SiMe3)2>>CH2Me3> sec-butyl > isopropyl > CH2SiMe3> n-butyl>CH23>Few examples of competitive carbonylation of two ligands attached to the same metal centerexist in the chemical literature, but those that do generally corroborate the results found with the126Cp*W(NO)(R)RI system. Thus, CO inserts into the more sterically demanding groups ofCp2Zr(CH(SiMe3)Me9andCp2Zr(Ph)Me1 to relieve steric crowding at the metal center.Evidently such steric factors are not dominant in the reaction of CO withCp2Ti(C6F5)Me, and itinserts preferentially into the presumably weaker metal-methyl bond.22 Interestingly, CO insertsinto the Jr-Me bond in the complexes(i-5-indenyl)Ir(PMe3)(Me)R [R = Ph, o-tolyl],23 a trendopposite to the one reported here. However, a complicated mechanism involving indenyl ringslippage and removal of coordinated CO with trimethylamine oxide is involved in this reaction, afact which may well have some effect on determining the preferred reaction products.The rearrangement of the trimethylsilylmethyl ligand to enolate complexes upon COinsertion does not appear to interfere with the ranking of migratory aptitudes established by thisstudy. Nevertheless, it should be noted that the neopentyl and the trimethylsilylmethyl ligandshave very similar migratory aptitudes. This feature is clearly evident during the carbonylation ofCp*W(NO)(CH2Me3)CHSi (4.4), where a 4:1 ratio of products was observed, the majorproduct arising from CO insertion into the tungsten-neopentyl bond, and the minor one fromtungsten-trimethylsilylmethyl insertion. The rearrangement of the trimethylsilylmethyl ligandafter insertion may add some driving force toward insertion into this ligand over others, butevidently not enough to overcome the rate of CO insertion into the neopentyl ligand.4.4 SummaryThe four alkyl and aryl chloride complexes synthesized in Chapter 3 have been used asprecursors for the preparation of a variety of mixed alkyl and aryl complexes of the typeCp*W(NO)(R)R1. The thermal chemistry of some of these complexes has been investigatedbriefly to determine if chemistry different from the symmetric bisalkyls could be induced. Onecompound, Cp*W(NO)(CH2Me3)Ph,has some interesting thermal chemistry, as it losesneopentane upon heating to form a putative ben.zyne intermediate, which then activates aromaticsolvents. It should be stressed here that not all of the mixed species have been investigated, and127in all probability, further interesting thermal chemistry of the mixed alkyl and aryl compounds hasyet to be uncovered.Isolation of the mixed compounds provide us with the opportunity to compare directly thereactivities of different tungsten-carbon a-bonds at the same metal center with a variety ofreagents. So far, càrbonylation reactions under ambient conditions have been performed on thesemixed species, and it has been found that each complex inserts one molecule of CO into one ofthe tungsten-carbon a-bonds. Conclusive identification of the carbonyation products leads to aranking of the migratory aptitudes of the five ligands used in the study, i.e. CH2Me3>CH2SIMe3> o-tolyl > Ph > Me. This trend appears to be sterically controlled, the most stericallydemanding ligands inserting CO preferentially in each case.Future work could possibly include the examination of the reactivity of this series of mixedcomplexes with other substrates. For example, the compounds could be reacted with hydrogento determine which ligand-tungsten link preferentially undergoes hydrogenation, or with oxygenor water to determine which dioxo alkyl or dioxo aryl complex is produced.1284.5 References and Notes(1) The Chemistry of the Metal-Carbon Bond; Hartley, F. R., Ed.; Wiley: New York, 1987;Vol. 4, Part 2.(2) Collman, J. P.; Hegedus, L. S.; Norton, J. R.; Finke, R. G. Principles and Applications ofOrganotransition Metal Chemistry; University Science Books: Mill Valley, CA, 1987;Chapters 3 and 14.(3) Legzdins, P.; Veitheer, J. E. Acc. Chem. Res., 1993, 26, 41.(4) Brunet, N.; Debad, J. D.; Legzdins, P.; Trotter, J.; Veitheer, J. E.; Yee, V. C.Organometallics 1993, 12, 4572.(5) Dryden, N. H.; Legzdins, P.; Lundmark, P. J. Organometallics 1993, 12, 2085.(6) Dryden, N. H.; Legzdins, P.; Rettig, S. J.; Veitheer, J. E. Organometallics 1992, 11,2583.(7) Legzdins, P.; Rettig, S. J.; Sanchez, L. Organometallics 1988, 7, 2394 and referencestherein.(8) Jordan, R. F.; Borkowsky, S. L.; Hinch, G. D. Organometallics 1991, 10, 1268.(9) Lappert, M. F.; Jeffery, J.; Luong-Thi, N. T.; Webb, M.; Atwood, J. L.; Hunter, W. E. J.Chem. Soc., Dalton Trans. 1981, 1594.(10) (a) Herrmann, W. A.; Felixberger, J. K.; Anwander, R.; Herdtweck, E.; Kiprof P.; Riede,J. Organometallics 1990, 9, 1434. (b) Teuben, J. H.; Luinstra, G. A. Organometallics1992, 1], 1793. (c) Bergman, R. G.; Foo, T. Organometallics 1992, 11, 1801.(11) Legzdins, P.; Rettig, S. J.; Veitheer, J. E. J Am. Chem. Soc. 1992, 114, 6922.(12) Brouwer, R. B.; Legzdins, P.; Rettig, S. J.; Ross, K. J. Organometallics, in press.(13) For a recent review, see: Buchwald, S. L.; Nielsen, R. B. Chem. Rev. 1988, 88, 1047.129(14) (a) Reference 13 and references therein. (b) Cámpora, J.; Buchwald, S. L.Organometaiics 1993, 12, 4182. (c) Arnold, J.; Wilkinson, G.; Hussain, B.; Hursthouse,M. B. Organometaiics 1989, 8, 415, and references therein.(15) (a) Reference 13 and references therein. (b) Chamberlain, L. R.; Kerschner, J. L.; Rothwell,A. P.; Rothwell, I. P.; Hufihian, J. C. J Am. Chem. Soc. 1987, 109, 6471. (c) Miller, F.D.; Sanner, R. D. Organometallics 1988, 7, 818.(16) (a) Durfee, L. D.; Rothwell, I. P. Chem. Rev. 1988, 88, 1059. (b) Petersen, J. L.; Egan, J.W., Jr. Organometallics 1987, 6, 2007. (c) Marks, T. I.; Mintz, E. A.; Sonnenberger, D.C. J Am. Chem. Soc. 1984, 106, 3484. (d) Marks, T. J.; Fagan, P. J.; Manriquez, J. M. JAm. Chem. Soc. 1978, 100, 7112. (e) Lappert, M. F.; Raston, C. L.; Engeihart, L. M.;White, A. H. I Chem. Soc., Chem. Commun. 1985, 521. (f) Andersen, R. A.; Simpson, S.J. I Am. Chem. Soc. 1981, 103, 4063.(17) 1HNMR data for Cp*W(NO)(CH2Me)(OC{CH}SiM:ö 0.16 (s, 9H, SiMe3), 1.44(s, 9H, CMe3), 1.68 (s, 15H, C,Me5), 4.25 and 5.81 (s, 1H each, OC(CH2)SiMe3,neopentyl methylene signals not observed.(18) Reference 2, Chapter 6.(19) Reviews of CO insertions: (a) Wojcicki, A. Adv. Organomet. Chem. 1973, 11, 87. (b)Kuhlman, E. J.; Alexander, 3. J. Coord. Chem Rev. 1980, 33, 195. (c) Alexander, 3. J. TheChemistry of the Metal-Carbon Bond; Hartley, F. R., Ed.; Wiley: New York, 1987; Vol. 2,Chapter 5.(20) (a) Cotton, 3. D.; Crisp, G. T.; Latif L. Inorg. Chim. Acta. 1981, 47, 171. (b) Cotton, J.D.; Crisp, G. T.; Daly, V. A. Inorg. Chim. Ada. 1981, 47, 165. (c) Cotton, J. D.; Tracey,L. B. Organometallics 1991, 10, 3156. (d) Casey, C. P.; Scheck, D. M. I Am. Chem.Soc. 1980, 102, 2723. (e) Pruett, R. L.; Fiato, R. A.; Cawse, J. N. I Organomet. Chem.1979, 172, 405. (f) Green, M.; Craig, P. 3. 1 Chem. Soc. (A) 1968, 1978. (g) Green, M.;Westlake, D. 3. 1 Chem. Soc. (A) 1971, 367.(21) Erker, G. Ace. Chem. Res. 1984, 17, 103.(22) Dormond, A.; Dachour, A. J Organomet. Chem. 1980, 193, 321.(23) Bergman, R. G.; Foo, T. Organometallics 1992, 11, 1811.130131CHAPTER 5Mechanism of Formation and Reactivity of Cp*W(NO)(H)(r)2PPh6H4).5.1 Introduction 1315.2 Experimental Procedures 1325.3 Results and Discussion 1415.4 Summary and Future Work 1555.5 References and Notes 1575.1 IntroductionIntramolecular activation of C-H bonds in transition-metal complexes is very prevalent inorganometallic chemistry.’ Both aromatic and aliphatic activations are known and, in fact, thefirst unambiguous example of C-H activation by a transition-metal complex was theintramolecular activation of a coordinated ligand.2 Various uses for this type of reaction havebeen developed, including uses in organic and asymmetric syntheses.3 The reversibility of manyof these metalations has led to their utilization in catalysis and in the modeling of catalytictransformations.4Although many examples of intramolecular activation exist, more work isneeded to decipher the mechanism of these reactions, especially since different mechanismsappear to be favored for different systems.1The orthometalated-phosphine complex prepared in Chapter 2,Cp*W(NO)(H)(rl2PPh6H4)(2.9), is an example of a transition-metal compound formed viaintramolecular C-H activation. The complex is presumably formed via alkane elimination from anintermediate phosphine-trapped alkyl hydride complex followed by orthometalation of one of the132phenyl groups on the phosphine ligand. Other mechanisms can be proposed to account for theproduct, however, and the different possibilities will be discussed in this Chapter. In addition, alabeling study that provides evidence for one particular mechanism is also described.An interesting property of this 18-valence-electron metalated complex is that underappropriate conditions, it undergoes reversible metalation. Thus, when exposed to Lewis bases,the hydride ligand and the W-C bond of the metalated phosphine aryl group undergo reductiveelimination to allow for coordination of the base to the metal center. Addition ofPPh3,acetonitrile, and acetone to complex 2.9 is presented along with some accompanying kinetic datafor the reaction with acetone.Complex 2.9 is also shown to activate benzene. The product of this reaction is discussed,and a comparison of the rate ofC6D versus C6H activation is presented as evidence for aproposed mechanism of aryl C-H bond activation. Finally, the reaction of the hydride complexwith chloroform is discussed.5.2 Experimental Procedures5.2.1 MethodsThe synthetic methodologies employed throughout this thesis are described in detail inSection 2.2.1. Isolated yields, and the spectroscopic and physical properties of all complexes arecollected in Tables 5.1 - ReagentsAcetone and acetonitrile were distilled twice from CaH2. CHCI3 (BDH),bis-(trimethylsilyl)acetylene, and TMS (Aldrich) were vacuum transferred fromP205. PPh3(Aldrich) was recrystallized from hexanes. PPh3-d15,hexamethylbenzene, Superhydride(LIB(C2H5)3),Red-Al (Na[CH3OCH2CH)1J,Aldrich), KH, NaBH4 (BDH), and133acetylene (Matheson) were used as received. Other reagents were prepared as described inprevious Chapters.5.2.3 Synthesis5.2.3.1 Preparation of Cp*W(NO)(D)(r2P{C6DS}4)(2.9-d15)Cp*W(NO)(CHSIMe)(0.25 g, 0.48 mmol) and PPh3-d15 (0.13 g, 0.48 mmol) weredissolved in hexanes (30 mL) and transferred to a 100 mL thick-walled bomb. H2 (18 psig) wasadded, and the solution was left for 5 hours. The vessel was then vented, and the yellowprecipitate was collected on a flit. This powder was then crystallized from Et20to obtain pure2.9-d15 (0.21 g, 70% yield). Preparation of Cp*W(NO)(PPh3)2(5.1)Cp*W(NO)(H)(rlPPh6H4)(0.30 g, 0.49 mmol) and PPh3 (0.13 g, 0.50 mmol) wereplaced in a small, thick-walled bomb. THF (15 mL) was added, and the vessel was then sealedand heated at 45 °C for 6 hours. During this time, the solution changed color from a brightyellow to a deep orange-red. The solvent was removed under vacuum, and the residue waswashed with a small amount ofEt20 (15 mL) to leave a bright orange powder (0.37 g,85% yield). Some of this product was crystallized fromCH21/hexanes to obtain samples of5.1 for elemental analysis. Preparation of Cp*W(NO)(PPh3)NCMe (5.2)Cp*W(NO)(H)(rlPPh6H4)was dissolved in NCMe in a small bomb. The vessel washeated at 50 °C for 2 hours, during which time its color changed from yellow to orange-brown.The solvent was removed under vacuum, and the residue was used directly for spectroscopiccharacterization. Repeated attempts to crystallize 5.2 from a variety of solvents led only to134decomposition of the product. Nevertheless, 5.2 was characterized by NIVIR and JRspectroscopies and by mass spectrometry. Preparation of Cp*W(NO)(PPh3)(q2OCMez) (5.3)Cp*W(NO)(H)(1l.PPh6H)(0.37 g, 0.61 mmol) was dissolved in acetone (20 mL) andthe solution was left unstirred at room temperature for 3 days, during which time the productcrystallized as yellow-orange blocks. The mixture was placed in a freezer overnight at -30 °C tocomplete the crystallization. The supernatant was then removed, and the product was driedunder vacuum (0.28 g, 68% yield). Reactions of 2.9 with Alkenes and AlkynesIn a typical experiment, Cp*W(NO)(H)(lPPh6H4)(0.050 g, 0.080 mmol) was weighedinto a vial in a glovebox. Approximately two equivalents of solid or liquid reagent were thenweighed into the same vial. The contents of the vial were dissolved in a minimum amount ofEt20, and the solution was transferred to a small bomb. The solution was heated at 45 °C for6-8 hours, after which time the solvent and volatiles were removed under vacuum. The residuewas dissolved in C6D,and a 1H NIVIR spectrum of this solution was then recorded. In this way,2,3 -dimethylbutene, 1 -pentyne, bis-(trimethylsilyl)acetylene, phenylacetylene, anddiphenylacetylene were all reacted with complex 2.9. Butadiene and acetylene were also used asreagents, but in an excess, since these reagents were vacuum transferred into the reaction vessel.None of these reactions formed one major product as judged by 1H NMR spectroscopy, andmost led to the formation of greater than ten products as judged by the number of resonances inthe Cp* region of their spectra.1355.2.3.6 Preparation of Cp*W(NO)(H)(Ph)PPh3(5.4)Cp*W(NO)(H)(rlPPhH4)(0.10 g, 0.16 mmol) was dissolved in benzene (15 mL) in asmall bomb, and the solution was heated at 50 °C overnight. The orange solution was thenconcentrated by removing approximately one-half of the solvent under vacuum. Hexanes(10 mL) were added, and the solution was then placed in a freezer maintained at -30 °C to inducecrystallization. Yellow crystals of complex 5.4 that formed were isolated by removing thesupernatant liquid via cannulation, and the crystals were dried under vacuum (0.10 g, 89% yield). Preparation of Cp*W(NO)(D)(C6D5)PPh3(5.4-d6)This deuterated analogue of 5.4 was prepared similarly to the non-deuterated compound,with the exception that C6D was used instead ofC6H. The product was also isolated as yellowcrystals. Preparation of Cp*W(NO)(C1)(12_PPh6H4)(5.5)Cp*W(NO)(H)(rlPPh6H4)(0.28 g, 0.46 mmol) was dissolved in CHC13 (25 mL), andthe resulting solution was stirred for 4 hours, during which time a flocculent yellow precipitateformed. Et20 (20 mL) was added, and the precipitate was collected on a fit and washed withmore Et20 (20 mL). The precipitate was dried under vacuum (0.26 g, 88% yield). Analyticallypure product was obtained by crystallization of the precipitate fromCH1/hexanes. Reactions of Cp*W(NO)(CI)(q2PPh6B4)with Hydride SourcesComplex 5.5 was reacted with a number of hydride sources in an attempt to metathesize thechloride for a hydride ligand. These reactions, all of which were unsuccessful, are brieflydescribed here.Cp*W(NO)(Cl)(rl2PPh6H4)(0.10 g, 0.16 mmol) and NaBH4 (0.015 g, 0.40 mmol) wereweighed into a large vial in a glovebox. THF was added, and the solution was then stirred for136three days at room temperature. An IR spectrum of the solution was then recorded whichappeared identical to the starting material.In a similar way, Cp*W(NO)(Cl)(fl2PPh6H4)was reacted with KH and also withSuperhydride in THF. JR spectroscopic monitoring of the solutions indicated that no reactionshad occurred after three days.An excess of RedAl was also reacted with the chloride complex. After allowing the reactionto stir for two days at room temperature, the solvent was removed, and a 1H NIVIR spectrum wasrecorded. Although a reaction occurred, the spectrum revealed that none of the expectedhydride product had been produced.5.2.4 Kinetic Monitoring of the Formation of Cp*W(NO)(PPh3)(i12_OCMe (5.3)All kinetic samples used for N4R spectroscopic monitoring of this reaction were preparedby the procedure described here. A sample ofCp*W(NO)(H)(1l2PPh6H4)(2.9) was weighedinto a small vial in a glovebox. The appropriate amount of acetone-d6was weighed into anothervial, and enough dioxane-d8to make the combined solution up to 0.500 mL was weighed into athird vial. The contents of the three vials were then mixed, whereupon the solid dissolved. Thefinal solutions were then transferred into an NMR tube bearing a Teflon stopcock. A smallamount of TMS (O.005 mL) was added, and the tube was then sealed. In this way, samplescontaining acetone concentrations of 10, 15, 21, 25 and 30 times greater than that of theorganometaffic reagent were prepared.The samples were then transferred to a pre-warmed NMR probe at 45, 55, 65, or 75 °C. Aninitial 1H NMR spectrum was recorded, and additional spectra were recorded at timed intervalsthereafter. Each measurement consisted of two single-scan spectra recorded one minute apart.The average of these two spectra was then used for integration measurements. The run at 75 °Cwas too fast to allow this sort of monitoring, and so each measurement consisted of only onescan. The integrations of the resonances due to the starting material and product Cp* ligands137were used to monitor the extent of the reaction after they were divided by the integration valuefor the TMS resonance.5.2.5 Kinetic Comparison ofC6D Versus C611 Activation by Complex 2.9The reaction of 2.9 with C6H and C6D to give the activated products 5.4 and 5.4-d6,respectively, were performed side-by-side and monitored identically. Each sample was preparedas follows: 0.200 g of 2.9 and 0.025 g of hexamethylbenzene (used as an integration standard)were dissolved in 8.790 g ofC6H (or 9.500 g C6D)and transferred to a stoppered test-tube.The tubes were placed in the cold well of a drybox that had been pre-warmed to55 °C by placing a constant-temperature warm bath on the exterior of the cold well.Approximately 0.5 mL aliquots of each solution were removed at timed intervals and placed inNMR tubes. The samples in C6H were taken to dryness under vacuum, and the residue wasredissolved in C6D. 1H NMR spectra for each sample were then recorded, each spectrumconsisting of four averaged scans (scans separated by one minute intervals). The integrations forthe peaks due to the Cp* resonances of the product and starting material were used to monitorthe extent of the reaction after being divided by the integration value for the hexamethylbenzenepeak.1385.2.6 Characterization Data for Complexes 2.9-d15,5.1 - 5.5Table 5.1. Numbering Scheme, Color, Yield, and Elemental Analysis Data for Complexes2.9-d15,5.1 - 5.5.complex compd color anal. found (calcd)no. (yield, %) C H NCp*W(NO)(D) 2.9-tii yellow - - -(i12-P(C6D5)4Cp*W(NO)(PPh3 5.1 orange (85) 60.57(63.28) 5.09(5.19) 1.39(1.63)Cp*W(NO)(PPh)NCMe 5.2 orange - - -Cp*W(NO)cPPh)(1l2OCMe 5.3 yellow (68) 55.30(55.61) 5.41(5.42) 1.98(2.09)Cp*W(NO)(H)(C6H5)PPh3 5.4 yellow (89) 59.45(59.23) 5.35(5.26) 2.02(2.03)Cp*W(NO)(D)(CD)PPh 5.4-d6 yellow - - -Cp*W(NO)(Cl)(r2PPh24 5.5 yellow (88) 5 1.91(52.07) 4.43(4.53) 2.11(2.17)Table 5.2. Selected Mass Spectral and Infrared Data for Complexes 2.9-d15,5.1 - 5.5.compd MS tempb IR (Nujol mull) (cm-i)HO. (°C) VNO other bands2.9-d15 626 [Pj 150 15495.1 873 [P] 120 15245.2 652 [Pj 120 1532 VCN17485.3 669 [Pi 120 1513, 15255.4 690 [p++HIC - 1572 VWH182O5.4-d6 695 [P - - 15665.5 645 [p+J 100 1593a m/z values are for the highest intensity peak of the calculated isotqpic cluster, i.e. 184W.b Probe temperatures. C FAB-MS (matrix: 3-nitrobenzyl alcohol). a DCI (NH3 carrier).139Table 5.3. NMR Data for Complexes 2.9-d15,5.1 - 5.5.compd 1 and 2H NMR ‘3C{’H} NMRno. o2.9-d15 ‘H (C6D):1.79 (s, 15H, C5Me)2H (C6H):7.8 (br, 6D, ArD)7.0 (br, 9D, ArD)3.95 (br, 1D, WD)5.la,c 7.20 (br, 12H, ArH) 139.85 (d, 1jc= 42 Hz, C10)7.13 (br, 18H, ArH) 133.88, 133.63, 133.49, 138.74,1.45 (s, 15H, C5Me) 128.55128.18, 127,43, 127.31 (C1)100.35 (CMe)10.29 (C5Me)52d 7.5 (m, 12H, ArH) 141.02, 140.46, 134.12, 134.056.9 (m, 18H, ArH) 133.98, 128.85, 128.74, 127.02 (C1)1.55 (s, 15H, C5Me) 100.27 (C5Me)10.57 (C5Me)53b,c 7.85 (br, 6H, o-ArH) 135, 130.7, 128.5 (br, C1)7.05 (br, 9H, m/p-ArH) 107.64 (C5Me)2.15 (s, 3H, Me) 70.33 = 51.1 Hz, C0)2.03 (s, 3H, Me) 33.30, 30.00 (Me)1.67 (s, 15H, C5Me) 10.63 (C5.A4’e)54C 7.63 (m, 8H, A.rH) 166.64 (d, ‘Jp 22 Hz, C10)6.90 (m, 12H, ArH) 142.79, 137.09, 136.511.61 (s, 15H, C5Me) 134.28, 134.15, 129.211.20 (d, 2PH = 94 Hz, 1H, WH) 128.33, 123.87, 123.84 (C1)106.87 (C5Me)10.23 (CA’fe5)5.4-d6 ‘H (C6D):7.62 (m, 6H, AsH)6.90 (m, 9H, ArH)1.62 (s, 15H, C5Me)2H (C6H):7.6 (br, 2D, ArD)6.8 (br, 3D, ArD)1.19 (d, PD = 15 Hz, 1D, WD)55a,c 7.8 (in, 2H, ArH) 137.30, 136.98, 134.39, 134.237.6 (m, 2H, ArH) 132.94, 132.82, 132.61, 132.567.5 (in, 3H, AiB) 130.76, 130.73, 130.66, 130.647.4 (m, 2H, AsH) 129.81, 128.79, 128.64, 128.617.35 (in, 2H, ArH) 125.37, 125.27 (C1)7.17 (in, 3H, ArH) 111.25 (CMe)1.70 (s, 15H, C5Me) 10.11 (C5Me)140a In CDC13. b In CD21. C NMR (ö, ppm): 5.1 = 45.90, Jw = 295 Hz; 5.3 = 28.24, Jp 210 Hz. 5.422.83, Jp = 118 Hz; 5.5 = -38.13, ‘w = 130 Hz. d In C6D.14L5.3 Results and Discussion5.3.1 Mechanism of Formation of Cp*W(NO)(H)(q2PPh6H4)(2.9)The synthesis of the orthometalated complex, 2.9, is first presented in Chapter 2 during adiscussion of the hydrogenation chemistry of Cp*W(NO)(CHSiMe3)This symmetric dialkylspecies reacts with molecular hydrogen in the presence of one equivalent of triphenyiphosphineto afford the metalated-phosphine complex in good yield. NMR spectroscopic monitoring of thishydrogenation reaction in C6D reveals that the conversion is quantitative in the presence of oneequivalent ofPPh3, and that two equivalents of TMS are liberated during the reaction. The firststep of the reaction is likely the formation of the intermediate alkyl hydride complex,Scheme 5.1N”IB0 RH2-TMSN”lR0 RR= CH2SiMe30NHO/TMS142Cp*W(NO)(CH2SiMe3)H,which is the same initial intermediate that is proposed to form in all ofthe hydrogenation reactions of the dialkyl species. The next step of the mechanism, shown inScheme 5.1, is most likely trapping of this hydride intermediate by PPh3 to form the 18-valence-electron complex Cp*W(NO)(CH2SiMe3)(H)PPh This intermediate is obviously unstableunder the conditions employed, since it is not observed during the monitoring of the reaction byNMR spectroscopy.The next step of the proposed mechanism needs to account for the reductive elimination ofthe second equivalent of TMS and the orthometalation of the phosphine ligand. It would seemintuitive that these reactions occur simultaneously to give the observed product, but there aresubtle differences in how this step could occur. Alkane elimination could be complete beforeorthometalation takes place. In this case, the hydride ligand in the product would originateentirely from the metalated aryl group of the phosphine ligand. Another possibility is theintermediacy of a 20-electron species formed via oxidative addition of the aryl C-H bond beforealkane elimination. This latter possibility can be confidently discarded since elimination ofdihydrogen instead of TMS from this intermediate would produce an alkyl-containing productwhich is not observed.If a concerted mechanism is adopted for the formation of 2.9 in which alkane is eliminatedand the phosphine undergoes metalation in one step, then the question arises as to whichhydrogen is lost with the alkyl group. Is it the one present on the intermediate hydride or is it theone arising from the metalated phosphine group? In other words, is the mechanism a truereductive alkane elimination followed by a fast oxidative addition of the aryl C-H bond, or isthere a four-centered transition state involved between the ortho C-H bond and the W-H bond?If the latter is true, then the hydride ligand in theproduct would originate solely from the orintermediate hydride, and the aryl-H would beeliminated with the alkane.143Some evidence for which pathway is involved has been obtained by performing a simplelabeling study. This has been accomplished by repeating the synthesis of 2.9 using deuteriumlabeled PPh3 such that the fate of the activated aryl deuteride could be determined. The synthesishas been performed identically to the preparation of the non-deuterated hydride. The product,2.9-d15,has been recrystallized from Et20 and characterized by 1H NMR, 2H NMR, and IRspectroscopies, and by mass spectrometry. All of the data are consistent with the formulation ofthe product as Cp*W(NO)(D)(rj2P C6D)4.The mass spectrum indicates a P of m/z 626which is fifteen mass units above the non-deuterated hydride parent ion peak. The 1H NMRspectrum of the product exhibits no signals assignable to the aryl or hydride groups, but onlycontains a single peak that is assignable to the Cp* ligand. The 2H NMR spectrum of thecomplex exhibits peaks that are consistent with those not observed in the 1H NMR spectrum; thearyl deuterons of the phosphine ligand are represented by two broad peaks at 7.0 and 7.8 ppm,and a peak at 4 ppm is assignable to the deuteride resonance. The IR spectrum contains anitrosyl-stretching band that is similar in energy to the non-deuterated analogue, butunfortunately does not have any bands attributable to a vxri. The analogousv.1band is observed in the IR spectrum of the non-deuterated hydride compound (1929cm-i). Using a simple Hooke’s Law calculation the analogous absorption for the deuteratedcomplex should appear around 1378 cm1. Unfortunately, this band cannot be definitivelyassigned because of the many bands in this area.The existence of a flilly-deuterated product at the metal-hydrogen link helps to narrow downthe mechanistic possibilities. Obviously, the proton lost with the alkane arises exclusively fromthe intermediate hydride and not from the phosphine ligand. Consistently, the hydride in themetalated product originates on the phosphine ligand. This evidence rules out the concerted,four-centered transition state mechanism and also any mechanism involving 20-electronintermediates since no scrambling of the protons occurs. The proposed mechanism for theformation of 2.9, based on this evidence, is shown in Scheme 5.1. It is not known if thereductive elimination occurs concurrently with orthometalation or just prior to it, but in either144case the latter step must be fast compared to the production of the alkyl hydride intermediate byhydrogenation, since no intermediates are observable.5.3.2 Reactions of Cp*W(NO)(H)(1l2PPh6H4)(2.9) with Lewis BasesDuring the preparations of the metalated hydride product, 2.9, it had been noticed that ifmore than one equivalent ofPPh3 is used, the product appears orange instead ofyellow and animpurity is clearly observable by NMR spectroscopy. This impurity exhibits only Cp* and arylresonances, and so it had been proposed that this impurity could be the bis-(phosphine) complex,Cp*W(NO)(PPh3)2.Two routes to this bis-(phosphine) product can easily be envisioned. One involves the directelimination of alkane from the phosphine-trapped alkyl hydride intermediate,Cp*W(NO)(R)(H)(PPh3),and concomitant trapping with another equivalent of phosphine. Theother route involves reaction of the metalated hydride product with excess phosphine. In thismechanism, the metalated phosphine is required to undergo an r12-to-’q’ transformation. To testif this latter route could produce a product identical to the impurity, a sample of pure complex2.9 has been reacted with PPh3. When effected in THF with one equivalent ofPPh3, a productcan be isolated in good yield. In fact, if the reaction is monitored by NIvIR spectroscopy, it isseen to be quantitative. This product has been conclusively identified as the bis-(phosphine)complex, Cp*W(NO)(PPh3)2(5.1) (equation 5.1), by NMP. and IR spectroscopies, massspectrometry, and elemental analysis.PPh3THF N 5.10” PPh3B PPh32.9 5.1145Complexes of the type Cp’W(NO)L2are well known. All of the precursors for the work inthis Thesis, if fact, are synthesized from one such complex. Cp*W(NO)(CO)2is reacted withPC15 to synthesize Cp*W(NO)(Cl)2which is used to make all of the alkyl complexes whosechemistry is discussed here. A series of compounds containing two-electron donors in the Cpsystem have been prepared in these laboratories by the reduction of the dihalides in the presenceof the donor ligand.5 Similar methodology could presumably be applied to the synthesis of theanalogous complexes containing a Cp* ligand, although this has not yet been attempted. Thisreduction method is feasible for the synthesis of complexes with two identical ligands, but couldnot be used to make mixed-ligand complexes of the type Cp*W(NO)(L)LI. The observedreactivity of the hydride complex, 2.9, presents us with the opportunity to synthesize these typesof species, since two-electron donors (L) capable of causing the metalated-phosphine ligand toadopt an re-coordination mode will produce the mixed complexes Cp*W(NO)(PPh3)L.Heating a solution of the metalated-phosphine compound 2.9 in acetonitrile produces themixed-ligand complex, Cp*W(NO)(PPh3)NCMe (5.2), which is isolated by removing the excesssolvent under vacuum. The product resists crystallization, and therefore prevents satisfactoryelemental analysis data from being obtained. However, the product has been spectroscopicallycharacterized. Mass spectrometry confirms the formulation of the product, and the 1H and 13CNMR data are consistent as well. The reaction has also been performed in deuterated acetonitrileto allow for monitoring of the reaction, which is observed to be quantitative by 1H NMRspectroscopy.Ph2 NCMe NpPh 5.2“ IN Me2.9 5.2146The IR spectrum of complex 5.2 is quite informative as to the nature of the bonding of thenitrile ligand to the tungsten center. A band is observed at 1748 cm1 and is the only feature inthe region that can be assigned to the vCN. This stretching frequency is quite low compared toother known N-bound nitrile complexes that usually display this band in the region 2200 -2300 cm-’. The observed value of 1748 cm-1 is, however, consistent with the nitrile being boundin ani2-fashion (equation 5.2). This type of nitrile bonding is not very common. Only a fewexamples have been observed,6with a handful having been structurally characterized.6a Forexample,Cp2Mo(r-NCMe) was the first structurally characterized example of a dihapto nitrilecomplex.6aThe vCN for this complex occurs at 1750 cm-’.CpIr(PPh3)-NC{p-ClCH4})exhibits a similar JR band at 1758 cm4.6 A recent tungstenexample isWC12(PMe3ri-NC e); however, this and other similar complexes reported by thesame authors do not exhibit any observable vCN bands.7Having established that complex 2.9 can be used to synthesize mixed-ligand complexes uponaddition of suitable Lewis bases, the same reaction has been attempted with acetone to determinewhich type of bonding is preferred by this ligand. Mass spectrometry and elemental analysisconfirm the formulation of the product isolated from this reaction as the expected acetoneadduct, Cp*W(NO)(PPh3)(rl2OCMe(5.3). The bonding mode of the acetone to the tungstenhas been determined from the NMR spectroscopic data. The 13C NMR spectrum of the productcontains, along with the expected aryl and Cp* resonances, two signals assignable to the acetonemethyl groups, and a signal at 70.33 ppm. This latter resonance is assigned to the carbonylcarbon of the acetone ligand. It is this signal that confirms the bonding of the acetone in adihapto fashion, because it exhibits tungsten satellites (wc = 51 Hz) which indicate that thiscarbon is bound directly to the tungsten center. The position of the carbonyl carbon resonance inthe 13C NMR spectrum is also helpful in confirming thei-2-bonding of the carbonyl group inketone and aldehyde complexes. Whereasrj1-bound ketones and aldehydes exhibit carbonylresonances in the normal range for the free ligand (204 ppm for acetone), the147ri2-.bound ligands display the same resonance at much higher field, i.e. 45 - 110 ppm (70 ppm forcomplex 5.3).8acetoneoN1’NPPh3 5.35.3 MeIn an attempt to determine if other ligands that are capable ofi2-coordination to thetungsten center could persuade the metalated phosphine to adopt an ri’-coordination mode,complex 2.9 has been reacted with a variety of unsaturated Lewis bases. It was hoped that in thisway a it-bound olefin or acetylene complex could be produced since we have not yet been able tocoordinate such molecules to our systems. The reactions between complex 2.9 and a variety ofalkenes and alkynes have been monitored by 1H NMR spectroscopy (see Section Noneof the final reaction solutions contain one major product, and therefore the reactions have notbeen attempted on a preparative scale. The failure of some of these reactions to produce onemajor product may be due in part to the ability of complex 2.9 to activate aromatic C-H bonds,as discussed later. Kinetic Study of Acetone Addition to Cp*W(NO)(H)(12PPh6H4)(2.9)A limited kinetic study of the addition of acetone to complex 2.9 has been performedprimarily to determine the order of the reaction with respect to each of the reactants, and theresults used to gain insite into the mechanism of the transformation.The reaction 5.3 has been monitored by 1H NMR spectroscopy. Acetone-d6has beenemployed instead of its non-deuterated analogue to avoid very large peaks in the spectra thatwould interfere with the integrations of the starting material and product peaks. The solventchosen for the study is dioxane-d8. Although the hydride complex reacts slowly with this solvent2.9148at the temperatures used for the study, it reacts preferentially with acetone when dissolved in amixture of the two solvents. Chlorinated solvents such as methylene chloride and chloroformcould not be used because they react with the hydride starting material, as discussed later.The experimental details of the kinetic runs are outlined in Section 5.2.4. All kinetic datahave been tabulated in the Appendix along with examples of all pertinent calculations.The reaction has been monitored at one concentration of the organometallic reactant withvarying amounts (10 to 40 times) of excess acetone-d6,and the rates of the reactions weremonitored at 55 °C in the presence of a small amount of TMS used as an integration standard.The rates of disappearance of the hydride starting material and the appearance of the productwere measured. Both were monitored to determine the reliability of the measurements and togive an approximate value of the experimental errors involved, since the rates of productappearance and hydride disappearance0-02-0.4-0.6-0.8• -12-1.6-1.8-2acxo sa 4cxo 45OFigure 5.1. Sample plot of the data obtained during the reaction of compound 2.9 with acetoneat 55 °C in the presence of a ten-fold excess of acetone.0 500 1OD 15(X) D)time (sec)149should be identical because no intermediates are observed. A sample plot of the data obtainedfor the experiment using a 10-fold excess acetone is shown in Figure 5.1. The integration of theCp* resonance of the hydride starting material has been monitored with time. The values forln[(At-A,)/(A0- )],where A is the ratio of the hydride peak integration to the integration ofthe standard TMS peak, are plotted versus time. Each run, conducted with a different excessconcentration of acetone, produced a different rate, with the runs containing a higherconcentration of acetone generally being faster. The data for each different concentration ofacetone produced a linear plot, indicating the reaction is first order in organometallic reagent atthe one concentration used for this study.A plot of reaction rates versus acetone concentration is shown in Figure 5.2, with a best-fitline drawn through the points. As is obvious from the figure, the points are quite scattered andthe resulting line does not pass through the origin, and its resulting slope is not very steep. Both11 I109I IIFigure 5.2. Plot ofk0bs versus acetone concentration for the addition of acetone to complex2.9. The data have been obtained by monitoring the disappearance of starting material.0.5 1 1.5 2acetone concentration (M)150of these features would be expected for data that implied a first order dependence of the rate onthe acetone concentration. Thus it appears that the reaction is not strictly first order in acetone,although it is obvious from the plot that there is a marked dependence of the rate of the reactionon the acetone concentration, with the rate increasing with increasing concentration.The dependence of the reaction rates on temperature has also been determined. The reactionof the hydride complex with a 25-fold excess of acetone has been monitored at 45, 55, 65, and75 °C, and kobs for each temperature has been calculated from the plot ofln[I-AI/IA0-A ]versus time for both the appearance of the product and disappearance of thehydride starting material. These rates and their associated temperatures have been used toconstruct an Eyring plot (Figure 5.3). From the slope and intercept values of the fitted line for.-10 —-10.5-11.5-12-1Z5.2.85 aisFigure 5.3. Plot of ln(k,bs/T) versus lIT for the addition of acetone to complex 2.9. Data wereobtained by monitoring the disappearance of the starting material.this graph, the following activation parameters have been calculated: Ht = 82 ± 10 kJ/mol and= -45 ± 14 J/molK. The value of MIt is not very informative about the mechanism of2.9 2.96 3 ac aiI1Tx1000K)151reaction, especially an associative one. The value of the entropy of activation, however, can bevery useful in determining what kind of mechanism is operative.9 A negative value, such as theone observed here, indicates an associative mechanism. The measured value is not verylarge, however, and its significance must not be overstated, since other effects such as solventreorganization can affect the value somewhat, especially for systems with polar solvents like theone utilized here.Although the kinetic analysis presented here is not at all comprehensive, some conclusionscan be drawn about the mechanism of the reaction. Firstly, for the one concentration measured,the reaction is first order in organometallic reagent. Secondly, there is a dependence of the rateof reaction on the acetone concentration, although as judged by the plot shown in Figure 5.2, thisdependence is not first order. The entropy of activation indicates an associative reaction, whichsupports the latter dependence. It can thus be concluded that the acetone is involved somewhatin forcing the phosphine ligand to undergo demetalation. In other words, the phosphine does notundergo spontaneous elimination of its tungsten-aryl bond to form a 16-valence-electron species,which is then trapped by the base. Demetalation only occurs when a Lewis base is present thatcan coordinate to the metal center.5.3.3 Reaction of Cp*W(NO)(H)(r2PPh6H4)(2.9) with Benzene.During the characterization of the metalated-phosphine complex 2.9, it had been observedthat the compound slowly decomposes in benzene-d6,and does so in a very clean manner toproduce a single product. This reaction has been repeated on a preparative scale using C6H anda higher temperature to increase the rate of the reaction. The product has been isolated andcharacterized as the phenyl hydride species, Cp*W(NO)(H)(Ph)PPh3(5.4). The IR spectrum ofthis compound shows a v of 1820 cm1, and the 1H NMR spectrum contains a doublet at 1.20ppm assignable to the hydride resonance. The coupling of the hydride ligand to the phosphorusatom is very strong(2Jp = 94 Hz), and it is thus assumed that the hydride is cis to the phosphineligand.152‘Ph2 0N-NPPh3Qil5.4The fact that this system activates benzene is not surprising, since complex 2.9 is formed byC-H activation of a phenyl group on the phosphine ligand. The driving force for the reaction inequation 5.4 is undoubtedly the decrease in strain associated with the metalated-phosphine link,since the overall number and type of bonds to tungsten has not changed upon completion of thereaction. Unfortunately, the system does not activate non-aromatic C-H bonds. This is also asexpected, because the preference of the system for aryl, rather than alkyl, bonds to tungsten hasalready been observed during the production of 2.9, where alkane elimination followed by arylactivation occurs readily. Once the aryl hydride complex Cp*W(NO)(H)(Ph)PPh3(5.4) isformed, it does not exchange with solvent. Thus, the 1H NMR spectrum of a sample of 5.4 doesnot change upon heating. The hydride peak does not diminish, as would be expected if thecomplex eliminated C6H and activated C6D,and no peaks attributable to liberated C6H areobserved. Kinetic Study of Benzene Addition to Cp*W(NO)(R)(rI2PPh6H4)(2.9)A comparison of the rates ofC6FL and C6D activation by a transition-metal complex canlead to the observation of a kinetic isotope effect for the reaction. This information is useful indetermining the mechanism of C-H activation reactions of this type, and therefore the experimenthas been performed for this system. Two solutions of complex 2.9, one in C6H and the other inC6D,have been heated at 55 °C and the disappearance of the starting material and appearanceof the product have been monitored by 1H NMR spectroscopy. The recorded data and pertinentcalculations are contained in the Appendix.2.9153A plot of the data acquired during the monitoring of the activation ofC6H and C6D bycomplex 2.9 is shown in Figure 5.4. The graph is a plot of ln[(A-A)/(A0- ]versus time forboth runs, where A is the integration value for the Cp* resonance of the hydride starting materialdivided by the standard hexamethylbenzene peak area. The slope of the line drawn with theC6H activation data is greater than that for the C6D data, indicating that the reaction withC6H is faster than that with C6D for this system. The kinetic isotope effect is calculated to bekH/kD = 1.9 ± 0.5, and although small, this effect is significant.0.50— -0.5-1.E -2-2.5-3-3.525DFigure 5.4. Plot of the data obtained during the monitoring of the reactions between complex2.9 and C6H, (.) and C6D (x).The product aryl hydride complex, when synthesized with C6D,exhibits no hydride signalin its 1H NMR spectrum, indicating that the original hydride ligand is transferred quantitatively tothe aryl group of the phosphine ligand. This observation rules out a four-centered transition statebetween the incoming aryl C-H and the metalated W-C bonds. Since there is an observed kineticisotope effect for the benzene activation reaction, the transition state must involve some degree0 5XO iOXD 15tXX 2cXXXtime (sec)154of C-H bond breaking. An associative mechanism is thus proposed in which benzeneC-H activation follows elimination of the metalated-phosphine link and the hydride ligand. Thetransient formation of an‘q2-benzene complex cannot be dismissed, and in fact would seem likelyfollowing the results of the Lewis base addition reactions discussed above. If such anintermediate is formed, however, it is formed prior to the rate determining step, since if thisprocess was rate-determining, a kinetic isotope effect would not be observed for the C-Hactivation. The value for kHIkD of 1.9 is comparable to other reported isotope effects for C-Hactivation reactions in other systems, although measurements of this type are rare. For example,the intramolecular activation of a phenyl group in Ir(PPh3)C1 exhibits an isotope effect of 1 4,1Owhile the activation of the alkyl C-H bond in (PEt2Pt(CHCMe exhibits an isotope effect ofabout 311 A small value for kH/kD has been attributed to a transition state in which the C-Hbond is mostly intact, a situation referred to as an early transition state. 125.3.4 Reaction of Cp*W(NO)(H)(l2PPh6H4)with ChloroformOne of the standard properties of transition-metal hydride complexes is that they react withchlorinated solvents in such a way as to replace the hydride with a chloride ligand.13 Thehydride complex, 2.9, is no different. When this complex is dissolved in chloroform, a yellowprecipitate slowly forms that, upon isolation, has been characterized asCp*W(NO)(Cl)(r2PPh6H4)(5.5). The NMR data for this compound are nearly identical tothose for the hydride from which it is derived, except for the absence of the hydride signal in theNMR spectrum. The IR spectrum of the chloride complex displays a nitrosyl-stretchingCHCI3N Ph2 5.52.9 5.5155frequency of 1593 cm1, 48 cm higher than that of the hydride. This is expected since thegreater electronegativity of the chloride ligand results in less backbonding from the metal to thenitrosyl ligand than occurs for the hydride compound.Complex 5.5 has been prepared as a possible precursor for the synthesis of the deuteridecompound, Cp*W(NO)(D)(2PPh6H4).Had this latter compound been preparable, a kineticcomparison of the rates of benzene activation would have been made between it and the hydrideanalogue to complement the results of the C6H versus C6D activation study. The deuteridecomplex, however, could not be prepared. Many hydride sources have been reacted withcomplex 5.5 in an attempt to metathesize the chloride for a hydride ligand. No evidence for theformation of the desired hydride was observed in any of the attempted reactions. Hence, it wasconcluded that the analogous deuteride sources would not afford the desired deuteride complex.5.4 Summary and Future WorkIt has been shown in this Chapter that the metalated complex,Cp*W(NO)(H)(r,2PPh6H),undergoes reversible metalation of the W-aryl and W-H bondswhen exposed to appropriate reagents. Concomitant Lewis-base coordination at the metal centercauses this reaction to occur, and this method has been exploited to produce mixed-ligandcomplexes of the type Cp*W(NO)(PPh3)L. This methodology therefore provides a route tonovel mixed-ligand species, and allows for their study in the future. When the Lewis base isacetonitrile or acetone, these ligands bind to the metal center in anri2-fash on. The characteristicreactivity of these dihapto complexes is worth future study, since there has been very littlereported on the reactivity of such dihapto ketones and nitriles. Molecular structure determinationof an ‘q2-nitrile complex of this type should also be considered, as very few of these types ofcompounds have been structurally characterized.The addition of acetone to the hydride complex has been shown to proceed via anassociative mechanism, implying the need for an incoming ligand to force the reverse-metalationof the phosphine ligand. The Cp*W(NO)(PPh3)fragment has the ability to backbond to156it-acceptor ligands, and seems to prefer them since the acetone addition reaction is observed tobe much faster than the analogous reaction with PPh3.The hydride complex 2.9 has also been shown to activate benzene. Although a pseudo-firstorder study to determine the order of the reaction has not been possible, a comparison of therates of activation of benzene and deuterated benzene has been performed. A kinetic isotopeeffect was observed, indicating some C-H bond activation in the rate-determining step of thereaction. A mechanism involving C-H activation following demetalation of the phosphine arylgroup is proposed based on the observed isotope effect and the distribution of the activatedhydride in the product. Unfortunately, the deuterium analogue of the hydride complex 2.9 couldnot be synthesized, and so the kinetic isotope effect for the reductive elimination portion of thereaction could not be measured.1573.5 References and Notes(1) (a) Ryabov, A. D. Chem. Rev. 1990, 90, 403. (b) Crabtree, R. H. Chem. Rev. 1985, 85,245.(2) Kleiman, J. P.; Dubeck, M. J Am. Chem. Soc. 1963, 85, 1544.(3) (a) Ryabov, A. D. Synthesis 1985, 233. (b) Sokolov, V. I. Pure Appi. Chem. 1983, 55,1837.(4) (a) Lewis, L. N. .1 Am. Chem. Soc. 1986, 108, 743. (b) Stolzenberg, A. M.; Muetterties,E. L. Organometaiics 1985, 4, 1739. (c) footnote la and references therein.(5) Hunter, A. D.; Legzdins, P. Organometallics 1986, 5, 1001.(6) (a) Wright, T. C.; Wilkinson, G.; Motevalli, M.; Hursthouse, M. B. .1 Chem. Soc., DaltonTrans. 1986, 2017. (b) Harman, W. D.; Sabat, M.; Barrera, J. .1 Am. Chem. Soc. 1991,113, 8178. (c) Chetcuti, P. A.; Knobler, C. B.; Hawthorne, M. F. Organometallics 1988,7, 650.(7) Barrera, J.; Sabat, M. Harman, W. D. Organometallics 1993, 12, 4381.(8) Huang, Y.-H.; Gladysz, J. A. J Chem. Ed 1988, 65, 298 and references cited therein.(9) Atwood, J. D. Inorganic and Organometallic Reaction Mechanisms; Brooks/ColePublishing Co.: Monerey, CA, 1985; Chapter 1.(10) Parshall, G. W. Ace. Chem. Res. 1970, 3, 139.(11) Moore, S. S.; DiCosimo, R.; Sowinske, S. F.; Whitesides, G. M.; I Am. Chem. Soc. 1981,103, 948.(12) (a) Jones, W. D. Activation and Functionalization of Alkanes; Hill, C. L., Ed.; WileyInterscience: 1989; Chapter IV. (b) Janowicz, A. H.; Bergman, R. G. I Am. Chem. Soc.1983, 105, 3929.(13) Crabtree, R. H. Comp. Coord. Chem. 1987, 2, 689158CHAPTER 6Reactivity of the Alkenyl Complex, Cp*W(NO)(CH2SiMe3)(CPh=C.6.1 Introduction 1586.2 Experimental Procedures 1606.3 Results and Discussion 1676.4 Summary 1846.5 References and Notes 1866.1 IntroductionMany types of a- and it-bound hydrocarbyl ligands are available for study on transition-metal complexes. Studies in our laboratories, however, have almost entirely been limited toinvestigating the chemistry of alkyl and aryl ligands in Cp’M(NO)(R)2systems. The onlyexceptions have been the recent isolation of the alkylidene species CpMo(NO)(=CHCMe3)L1and the diene complexes Cp’M(NO)(r4-diene .2Although the syntheses of the diene compoundsappear generally applicable, the route to the alkylidene species remains specific to the CpMosystem. Some attempts have been made to synthesize compounds containing hydrocarbyl ligandssuch as olefins, acetylenes, 3-hydrogen-containing alkyl groups, and other alkylidenes on theCpM(NO) fragments, but without success. We can, however, produce complex 2.8,Cp*W(NO)(CH2SiMe3)(CPh=CH which is the first alkenyl species that we have been able toisolate. This alkenyl ligand is the only a-bound ligand with f3-hydrogens, other than aryl groups,that we have observed on the Cp’M(NO) fragments. The stability of this alkenyl complex hasbeen attributed to an interaction between the alkenyl double bond and the electronically159unsaturated metal center (see Chapter 2). It is not surprising, however, that the complex isthermally sensitive, and readily undergoes f3-hydride elimination upon heating.The synthesis and isolation of the alkyl alkenyl compound,Cp*W(NO)(CH2SIMe3)(CPh=C (2.8), is presented in Chapter 2. The complex is formedupon hydrogenation of the bis-(trimethylsilylmethyl) species in the presence of phenylacetylene,and is isolated in very low yield because of competing reactions that occur during hydrogenation.This abysmal yield of complex 2.8 hindered studies on its reactivity and so an improved route tothis compound was desired. This Chapter begins with a discussion of the synthetic methodologyinvolved in synthesizing this alkenyl species in high yield, which in turn allows for further studyof its characteristic chemistry.This Chapter presents some investigations into the thermal chemistry of this alkenylcompound, 2.8. Its chemistry is dominated by f3-hydrogen elimination from the alkenyl ligand,which is accompanied by TMS liberation to form a highly reactive intermediate. Thisintermediate is so reactive that it is capable of activating both aromatic and aliphatic C-H bonds.The products of ben.zene, hexane, pentane, and Et20 activation are presented, and mechanismsaccounting for their structures are discussed.Reaction of the alkenyl complex with PMe3 results in the formation of a 1:1 adduct, but notthe expected metal-centered one. The Lewis base attacks the alpha carbon of the alkenyl ligandto form a metallacyclopropane compound. Characterization data for this product are presentedalong with its crystallographically-established molecular structure.When complex 2.8 is produced via the hydrogenation of Cp*W(NO)(CH2SiMe3)in thepresence of phenylacetylene (Section, an organic compound, 1,4-diphenyl-1,3-butadiene,can also be isolated along with the organometallic species. The thermal chemistry of complex 2.8that has been uncovered during the C-H activation studies is helpful in explaining the appearanceof this organic product, and a mechanism involving alkyne coupling has been proposed toaccount for its formation.1606.2 Experimental Procedures6.2.1 MethodsThe synthetic methodologies employed throughout this Thesis are described in detail inSection 2.2.1. Isolated yields, physical properties, and spectroscopic data for all complexes arelisted in Tables 6.1- Reagents(CPhCH)Mg.X(dioxane) was prepared by a procedure analogous to that used tosynthesize the dialkyl- and diarylmagnesium reagents used in this Thesis.3 All other reagentswere prepared as described in previous Chapters.6.2.3 Improved Synthesis of Cp*W(NO)(CH2SiMe3)(CPh=C (2.8)Cp*W(NO)(Cl) (4.08 g, 9.70 mmol) and(CH2SiMe3Mg.X(dioxane) (0.850 g,4.86 mmol) were mixed in a small Erlenmeyer flask in a glovebox. The flask was cooled to-196 °C, and THF (20 mL) was added dropwise so that it froze on top of the reagents. Thecontents of the flask were allowed to warm slowly while being swirled. When all of the reagentswere dissolved and the solution was deep blue, a solution of(CPh=CH2)MgX(dioxane)(0.777 g, 4.85 mmol) in THF (5 mL) was added dropwise. The mixture was then allowed towarm to room temperature while being constantly stirred. The solvent was then removed fromthe red solution under vacuum, and the residue was left under dynamic vacuum for a furtherhour. After this time, the residue was extracted with Et20 (3 x 10 mL) and the combinedextracts were filtered through Florisil (1 x 3 cm) supported on a flit. The Florisil was washedwith more Et20 until the washings were colorless. The combined filtrates were thenconcentrated to approximately 20 mL, hexanes (15 mL) were added, and the resulting solution161was placed in a freezer maintained at -30 °C. Deep red crystals of complex 2.8 were depositedand were isolated by removing the supernatant solution via pipette (1.81 g, 69% yield).6.2.4 Preparation of Cp*W(NO)(Ph)(CPh=CH2)(6.1)Cp*W(NO)(CHSiMe3)(CPh=C (0.310 g, 0.575 mmol) was dissolved in C6H (10 mL)and placed in a small thick-walled bomb. The vessel was heated at 45 °C for two days, duringwhich time the solution turned from red to deep orange-brown. The solvent was then removedunder vacuum, and the brown residue was extracted withEt20/hexanes (10 mL of a 1:1mixture). The extract was filtered through Celite and placed in a freezer maintained at -30 °C.Brown-black crystals of 6.1 (0.25 g, 81% yield) were isolated by removing the supernatantsolution with a pipette.6.2.5 Thermolysis of Cp*W(NO)(CII2SiMe3)(CPh=CH in C6DCp*W(NO)(CHSiMe3)(CPh=CH (0.030 g) was dissolved in C6D (0.5mL) in an NMRtube. The tube was sealed and then warmed to 45 °C for two days. During this time the 1HNTvIR spectrum of the solution was periodically recorded. When the starting material had allbeen consumed, a final spectrum was recorded. Only one product was formed, as judged by asingle new Cp* resonance. A large peak at 0.0 ppm also indicated that TMS had been liberated.6.2.6 Measurements of the Rates ofC6D and C6H Activation by Complex 2.8The rates ofC6D and C6H activation by 2.8 were measured at 50 °C. In the experiment,0.150 g samples of complex 2.8 were dissolved in 7.00 mL ofC6H and 6.94 mL ofC6D,andmeasurements of the rates of disappearance of starting material were effected by the procedureoutlined in Section 5.2.5, with the exception that instead of monitoring the Cp* signals, thedisappearance of the SiMe3 signal of the starting material was monitored. All data and theirgraphical representation are contained in the Appendix.1626.2.7 Thermolysis Reactions of Complex 2.8Complex 2.8 was thermolyzed in the presence of a variety of reagents. All of these reactionswere performed by the procedure outlined below.For each reaction, Cp*W(NO)(CHSiMe3)(CPhCH (0.030 g, 0.056 mmol) andapproximately 1.5 equivalents of each reagent were weighed into a small vial in a glovebox. Thecontents of the vial were then dissolved in hexanes (10 mL) and transferred to a small, thick-walled bomb. The solution was heated at 45 °C for two days, after which time the solution wasplaced under vacuum to remove the solvent and any volatiles. The residue that remained wasdissolved in C6D,and a 1H NIvIR spectrum was recorded. In this way, PPh3 phenylacetylene,2,3-dimethyl-2-butene, 1-pentyne, benzophenone, benzophenoneimine, and benzaldehyde wereall used as reagents.In the case of phenylacetylene, one major product was formed, as judged by 1H NMRspectroscopy. In all other cases, decomposition occurred, as evidenced by a large number ofpeaks in the Cp* region of the 1H NMP. spectra. In most of these reactions, signals attributableto the product resulting from solvent activation could be observed in the 1H NMR spectra of thereaction mixtures.In some cases no solvent was required, since complex 2.8 dissolved directly in the neatreagent, and the solutions were treated as described above. In this way acetone, cyclohexanone,acetonitrile, THF, hexanes, Et20, and pentane were all tested as reagents. In the case ofhexanes, pentane and Et20, only one product was formed, as judged by the 1H NMR spectra ofthe reaction mixtures. In all other cases, evidence for decomposition to a number of productswas observed.6.2.8 Preparation of the Metallacyclic Complexes 6.2 - 6.4All three of these complexes were prepared by the procedure described below.163In a typical experiment, Cp*W(NO)(CH2SiMe3)(CPh=C (0.30 g, 0.56 mmol) wasdissolved in the appropriate solvent (Et20(10 mL) or hexanes or pentane (50 mL)) in a thick-walled bomb. The solution was heated at 45 °C for two days, during which time it turned fromred to orange-brown. The solvent was then removed under vacuum, and the residue wasdissolved in a minimum amount ofEt20. The solution was filtered through a small column ofFlorisil (2 x 2 cm) supported on a fit, and the column was washed with Et20 until the washingswere colorless. The solvent was removed from the combined filtrates under vacuum to leave anoily yellow-brown residue which was dissolved in a minimum amount of hexanes. The resultingsolution was then placed in a freezer (-30 °C) to induce crystallization. Each of the compoundscrystallized as yellow-brown rosettes. Complex 6.3 was ffirther recrystallized from hexanes,while complexes 6.2 and 6.4 were recrystallized fromEt20/hexanes. All three compoundscrystallized as yellow-orange needles, and were isolated by removing the supernatant solution viapipette (6.2, 57%; 6.3, 23%; 6.4, 32% yield).6.2.9 Synthesis ofCp*W(NO)(CHSiMe)(q_C{PMe}PhCH(6.5)Cp*W(NO)(CHSiMe3)(CPhCH (0.250 g, 0.464 mmol) was dissolved in hexanes(40 mL) in a thick-walled bomb. The solution was cooled to -196 °C, and PMe3 (excess) wasvacuum transferred into the bomb. The resulting solution was then heated at 45 °C overnight,during which time it turned from red to bright yellow and small, needle-like crystals weredeposited. The solvent and excess PMe3 were removed under vacuum, and the yellow residuewas then dissolved in CH21 (5 mL). Hexanes (10 mL) was added, and the solution was placedin a freezer (-30 °C) to induce crystallization. Complex 6.5 was isolated as yellow crystals (0.18g, 61% yield) by removing the supernatant via cannulation.1646.2.10 Characterization Data for Complexes 6.1 - 6.5Table 6.1. Numbering Scheme, Color, Yield and Elemental Analysis Data forComplexes 6.1- 6.5.compd color anal, found (calcd)complex no. (yield, %) C H NCp*W(NO)(Ph)(CPh,CH2) 6.1 brown (81) 54.46(54.76) 5.14(5.36) 2.65(2.42)Cp*W(NO)(l4 6.2 orange (57) 52.18(52.78) 6.48(6.36) 2.60(2.68)CHPhCH2WPr)Cp*W(NO)(l4 6.3 orange (23) 53.90(53.46) 6.6 1(6.57) 2.54(2.6 1)CHPhCH2”Bu)Cp*W(NO)(l4 6.4 orange (32) 50.3(50.29) 5.91(5.95) 2.58(2.67)CHPhCH2(OEt)C)Cp*W(NO)(CHSiMe3 6.5 yellow (61) 48.94(48.78) 7.07(6.88) 2.44(2.28)(i2-C{PMe3}PhCH2)cmpd MS tempb IRn°’,fl/Za (°C) (Nujol, cm1)6.1 529 [Pj 120 1572, 156215536.2 523 [Pj 100 VNO 15716.3 537 [Pj 150 VNO 15726.4 525 [P] 120 VNO 15606.5 615 [Pj 80 VNO 1484972, 953a m/z values are for the highest intensity peak of the calculated isotopic cluster (184f).b Probe temperatures.Table 6.2. Selected Mass Spectral and Infrared Data for Complexes 6.1 - 6.5.165Table 6.3. 1H NI’ER Data for Complexes 6.1 - 6.5.compd 1H NMR (C6D) ‘3C{1H} NMR (C6D)no. 6 66.1 7.99 (d, 3J= 7.8 Hz, 211, o-ArH) 225.09 (WCPh=)7.90 (d, 3HH = 7.8 Hz, 2H, o-ArH) 179.00, 143.33 140.807.27 (m, 4H, ArH) 130.28, 129.33, 128.877.18 (m, 2H, AiR) 127.53, 126.41 (C1)4.28 (d, 2HH = 6.9 Hz, WH = 6.3 Hz, 110.23 (C5Me)1H, =CH2) 78.95 (=CH2)3.71 (d, 6.9 Hz, 1H, WH 9.56 (C5Me)11.4 Hz, CH2)1.54 (s, 15H, C5Me)6.2 6.97 (br, 2H, ArH) 121.30, 120.99 (C1)6.64 (t, J= 7 Hz, 1H, ArH) 106.50 (C5Me)2.98 (m, 111, CHCH2H) 71.30 (WCHPh, Jew = 15 Hz)2.51 (m, 111, WCHCI]) 5269 (WCH2CH)1.83 (m, 1H, WCH) 44.24 (CH3C)1.67 (dt, J= 9.4, 2.8 Hz, 1H, WCH2) 36.06 (WCH2,‘ew = 88 Hz)1.4- 1.6 (m, 5H) 35.27 (CHCH)1.58 (s, 15H, C5Me) 22.37 (CHC)1.10 (dd, J= 11.0, 9.4 Hz, 1H, WCH2) 15.07 (CH3)0.98 (t, J= 7.0 Hz, 3H, CH3) 10.07 (C5Me)6.3 7.00 (br, 211, ArH) 122.00, 121.13 (C1)6.68 (t, J = 7 Hz, IH, ArH) 10647 (C5Me)3.02 (m, 111) 7128 (WCHPh)2.54 (m, 1H) 53.11 (WCHCH)1.4 - 1.8 (m) 41.91 (CH)1.59 (s, 1511, C5Me) 36.25 (WCH2)1.15 (m, 2H) 34.08 (CH3C1.02 (m. 111) 31.72 (CHCI-I)0.97 (t, J= 7.6 Hz, 3H, CH3) 23.90 (CH3C)14.56 (CH3)10.08 (C5Me)6.4a 6.95 (br, 2H, AiR) 136, 131, 123 (br, C1)6.65 (t, J= 7 Hz, 111, ArH) 121.25, 120.83 (C1)3.99 (m, 1H, WCH) 107.72 (C5Me)3.72 (m, 1H, OCH2CH3) 87.46 (WCHPh)3.52 (m, 111, OCHCH 68 (br)343 (m, 1H, CHCH2H) 64.33 (OCH2CH2.15 (m, 1H, CHCHH) 58.20 (WCHCH)2.02(m, 1H, WCH2) 33.15 (WCH,Jwc 109 Hz,)1.54 (s, 1511, C5Me) 31.36 (CHCH2)1.35 (m, 211, WCH2CH) 16.02 (CH3)1.30 (t, J= 8.5 Hz, 3H, OCH2CH3 10.43 (C5Me)7.40 (m, 2H, o-ArH)7.11 (t, 3HH = 7.5 Hz, 2H, m-ArH)6.99 (m, 111, p-ArH)1.56 (d, 2J = 12.4 1{z, 9H, PMe3)1.50 (s, 1511, C5Me)1.40 (dd, J’= 7.8, 25Hz, 1H, CH2)0.98 (dd, J 7.8, 10.8 Hz, 111, CH2)0.88 (d, 2HH = 13.5 Hz, 1H,CH2SiMe3)0.10 (s, 9H, SiMe3)-1.24 (d, 2HH = 13.5 Hz, 1H,CHSiMe3)141.88 (C0)131.19, 127.46, 125.35 (C1)106.00 (C5Me)29.41 (d, Jp = 53 Hz, WC{PMe3})27.63 (CH2SiMe Jcw = Hz)11.76 (d, Jp = 56Hz, PMe3)9.97 (C5Me)4.60 (SiMe3)-10.04 (d,‘ci 4 Hz, WCH2C{PMe3},Jcw81H2)6.5a,C166a CDC13. b C7D8. C 31P{1H} 34.66.1676.3 Results and Discussion6.3.1 Improved Synthesis of Cp*W(NO)(CH2SiMe3)(CPh=C (2.8)To assist in the study of the reactivity ofCp*W(NO)(CH2SiMe3)(CPh=CH (2.8), a higher-yielding synthetic route to this species was required. It was decided to apply the samemethodology used to synthesize the mixed alkyl compounds presented in Chapter 4 to thesynthesis of this alkyl alkenyl complex. Thus, the appropriate dialkenylmagnesium reagent,(CPh=CH2)Mg.X(dioxane), has been prepared from the appropriate Grignard reagent,(CPh=CH)MgBr, which has, in turn, been synthesized from cL-bromostyrene and magnesium.The dialkenylmagnesium reagent can then be reacted with Cp*W(NO)(CH2SiMe3)Clthat isgenerated in situ to produce the desired product in good yield (equation 6.1). In fact, theGrignard reagent itself can be used in the preparation of the alkyl alkenyl complex, although thedialkenylmagnesium reagent is much easier to use and to store and results in better isolated yieldsof the product. This synthetic methodology allows for the preparation of complex 2.8 in gramquantities, and thus enables us to explore its characteristic chemistry.1) 1/2 (CHSiMe3Mg oxan Hw 2)1/2 (CII2CPh)zMrdioxane w J—H/\ /—{ 6.1Cl TIIF No ci 0 R©2.8 R = CH2SiMe31686.3.2 Thermal Chemistry of Complex Activation of Aromatic SolventsDuring the characterization of complex 2.8, it was noted that the complex decomposes whendissolved in toluene-d8or benzene-d6. The decomposition takes approximately two months togo to completion at room temperature, but it is very clean, as only one organometallic productbeing formed. The 1H NMR spectrum of the final reaction solution indicates that TMS is lostduring the reaction, and that the organometallic product contains a Cp* ligand and a phenylgroup. The reaction has been repeated on a preparative scale in benzene at a higher temperatureto increase the reaction rate.When a solution of complex 2.8 in C6H is heated at 45 °C, a reaction occurs and a newcompound can be isolated in good yield from the final mixture. This complex,Cp*W(NO)(Ph)(CPh=CH2)(6.1), is a product of solvent activation (equation 6.2). Thecharacteristic properties of the compound, both physical and spectroscopic, are very similar tothe alkyl alkenyl complex from which it originates. The spectroscopic properties of the alkenylligands in these two compounds are very similar, implying that they both exhibit a similar type ofinteraction between this ligand and the metal center. Thus, the‘3C{’H} NMR spectrum of 6.1displays peaks at 225 and 79 ppm assignable to the alpha and beta alkenyl carbons, respectively,and the 1H NMR spectrum contains two mutually-coupled alkenyl proton resonances at about4 ppm(2J= 6.9 Hz) that also display coupling to the tungsten nucleus (J = 6.3 and11.4 Hz). The observation of such coupling is used in Chapter 2 as a criterion for inferring an/W%./LH-SIMe46.2N N0 Rç 0 Phç2.8 R = CH2S1Me3 6.1169interaction between the alkenyl double bond and the metal center in complex 2.8.The above reaction has been repeated in C6D so that it could be monitored by 1H NMRspectroscopy. The spectrum of the final product displays signals due to the aryl group on thealkenyl ligand and the Cp* ligand, but lacks signals for either of the alkenyl protons and the arylgroup attached to the metal center. This implies that the aryl ligand originates as solvent and alsothat both alkenyl protons are replaced by deuterons during the reaction.This behavior can be explained by the proposed mechanism shown in Scheme 6.1 whichoutlines the steps involved when the alkyl alkenyl complex, 2.8, is thermolyzed in C6D solution.The alkenyl ligand first undergoes 13-hydrogen elimination and TMS is then lost from the metalcenter to form anrj2-alkyne intermediate. This intermediate activates a C-D bond of the solventto form a species with a phenyl ligand and a new alkenyl ligand in which one of the methylenealkenyl positions is deuterated (complex 6.1-d in the Scheme). This step is obviously notcompletely stereoselective, and affords deuteration at either position on the alkenyl ligand, sinceboth of the alkenyl positions are eventually deuterated. Elimination ofC6D5HfromScheme 6.1R = CH2S1Me3M = Cp*W(NO)=:: [/2] 6.146BD-C6D511170this complex and activation of another molecule of solvent then produces the frilly deuteratedproduct (6.1-d7). The aryl group is then further exchanged with solvent in an equilibriumprocess. The Scheme displays the situation that quickly leads to full deuteration of the alkenylligand, although in practice, many benzene elimination-activation steps are presumably requiredbefore all alkenyl methylene protons are replaced by deuterons. The driving force for thereaction shown in Scheme 6.1 is presumably the greater bond strength of the metal-aryl bondcompared to that of the original metal-alkyl bond, but the smaller steric bulk of the phenyl overthe trimethylsilyl ligand may also be a contributing factor.To help confirm that the benzene exchange reaction shown in Scheme 6.1 is taking place, aC6D solution of a non-deuterated sample of 6.1, the phenyl alkenyl complex, has been heatedand the reaction monitored by 1H NMR spectroscopy. Eventually, the spectrum of this reactionappears identical to that obtained from thermolysis of 2.8 in C6D,thus implying that the sameproduct is formed. The alkenyl proton signals of complex 6.1 disappear along with the signalsdue to the aryl ligand, and a sharp peak in the aryl region indicates that benzene is liberated.Since no intermediates that contain only one alkenyl proton are observed during thethermolysis of complex 2.8 in C6D,it can be concluded that this benzene-exchange reaction ismuch faster than the elimination of alkane from the original reactant. Consistently, thedisappearance of the alkenyl methylene and the initial alkyl ligand signals occurs at approximatelythe same rate. Therefore, once the phenyl alkenyl complex is formed, it undergoes rapid solventexchange so that no intermediates containing a mono-deuterated alkenyl ligand are observed.This slow elimination of TMS as the rate-determining step followed by rapid aromaticsolvent activation would result in no kinetic isotope effect for benzene activation. The rates ofboth C6D and C6H activation by complex 2.8 have been measured using the same procedureemployed for the measurement of a kinetic isotope effect for benzene activation by complex 2.9(Section The measured rates for the reactions between complex 2.8 and C6H andC6D are 1.1 ± 0.2 x 10 s and 1.2 ± 0.3 x io s, respectively. The two rates are not171significantly different, and thus it must be concluded that, as expected, C-H bond activation is notinvolved in the rate-determining step.Only a few reports on the thermal chemistry of transition-metal alkenyl complexes arepresent in the literature.4 Most of these studies are of zirconocene alkenyl species, presumablybecause these compounds are easily obtained via hydrozirconation of acetylenes with readily-obtained zirconocene hydrides.4ab5One aspect of these systems that make them attractive forstudy is that the compounds involved are electronically and coordinatively unsaturated, and arethus inherently more reactive than 18-valence-electron alkenyl complexes. It is not surprising,therefore, that some similarities in reactivity exist between the zirconocene alkenyl systems andthe thermal chemistry exhibited by the 16-valence-electron complexes,Cp*W(NO)(CH2SiMe3)(CPh=CH (2.8) and Cp*W(NO)(Ph)(CPh=CH (6.1).Elimination of a f3-hydrogen from an alkenyl ligand is common in zirconocene systems.Bercaw and co-workers have proposed a 13-elimination step to account for ligand-coupledproducts that are observed upon thermolysis ofCp*2Zr(CH=CFllvIe).4Erker et al. haveproposed a similar process to account for acetylene-coupled products formed during theattempted synthesis of someCp2Zr(Cl)(CR=CHR’) species.4b Intermolecular C-H activation byunsaturated alkyne intermediates formed via these 3-hydrogen elimination processes is notobserved in these systems, rather the alkyne complex couples with free acetylenes or olefins toform metallacyclic products. However, one example of a side-bound acetylene complex capableof activating C-H bonds has been reported for a zirconocene complex. The thermolysis ofCp2Zr(THF)(i-Me3SiCCSiMe)results in the loss of THF and activation of a C-H bond of aCp ligand in another equivalent of starting material. The product of this reaction is a binuclearalkenyl species6 (equation 6.4). Other examples of C-H bond activation by transition-metalalkyne complexes have been observed in iron, nickel, and platinum systems.7172SiMe3—2THF 6.3SiMe3 Me3Si6.3.2.2 Activation of Aliphatic Solvents by Complex 2.8In an attempt to trap the alkyne intermediate that is proposed to form upon thermolysis ofCp*W(NO)(CHSiMe)(CPh=C (2.8), hexanes solutions of this compound have beenthermolyzed in the presence of a number of Lewis bases and unsaturated reagents. It was hopedthat the Lewis bases would directly stabilize the alkyne intermediate such that a complex of thetype Cp*W(NO)(CPhCH)L could be isolated. It was also hoped that the unsaturated reagents,such as alkynes and ketones, would trap this intermediate species by coupling with its alkyneligand to form an isolable metallacycle. The details of these reactions are presented in theExperimental Section, and can be briefly summarized here by saying that none of the reactionsproduce any major products except for the reaction with PMe3,which is discussed later. Thethermolysis of complex 2.8 without any added reagent forms one major product, however, andthis same product can be observed spectroscopically in many of these thermolysis reactions.The thermolysis of complex 2.8 in hexanes has been repeated on a preparative scale withoutadded reagents, and a product resulting from solvent activation, complex 6.3, has been isolatedfrom the reaction solution (equation 6.4). Thus it appears that if the species formed upon TMSliberation from complex 2.8 does not react with added reagents, then it reacts with the solvent.The elemental analysis and mass spectral data for complex 6.3 imply that this compound is acombination of the proposed alkyne intermediate and a molecule of hexane. The 1H NMR173spectroscopic data for the compound are very complicated, and so a structure for the complexcould not be proposed from these data alone. The thermolysis of compound 2.8 has also beenperformed using pentane and Et20 as solvents, and solvent-activation products are againobtained (complexes 6.2 and 6.4 respectively). Since their spectroscopic data are very muchalike, all three compounds presumably have a similar structure.CH3HR’W IL 6.4/ I —S1Me40NR02.8 R = CH2S1Me3 6.2 R’ = CH236.3 R’ =CB6.4 R’ = OCB2CH3An X-ray crystallographic analysis of the pentane activated product, 6.2, has beenperformed, and the solid-state molecular structure is displayed in Figure 6.1. The complexcontains a metallacycle that is stabilized by an3-benzyl interaction. It is clear from the structurewhich parts’ of the ligand originated as solvent and which originated as the initial alkenyl ligand.The pentane molecule has been activated at the terminal position and is now bound to the metalat that point. The beta position has also been activated, and is bound to the beta carbon of thealkenyl ligand to form the metallacycle. Thus, a double C-H activation of the pentane hasoccurred, and C-C bond formation between the pentane and the alkenyl ligand has formed themetallacyclic product. The bond distances around the metallacycle portion of the structure are allwithin the range of single bonds. The W-C(6) bond length of 2.226(4) A is comparable to othertungsten-carbon single bonds (for example, see Figure 6.4), while theW-C(19) length of 2.301(4) A is slightly longer, but within the range of a single bond. The C-Cbonds that make up the cyclic portion of the ligand are all between 1.509-1.534 A, and compare174Figure 6.1. View of the solid-state molecular structure of Cp*W(NO)_(114-CHPhCH2CHflPr)(6.2), including selected bond lengths and angles with esd’s inparentheses.C(15) C( 14)C(5) C(4)C( 11)___C(1) C(2) C(3) C(13)C( 12)C(23)C(24)Bond Lengths (A)W - C(6) 2.226(4)W.-C(19) 2.301(4)W-C(21) 2.381(4)W-C(22) 2.371(4)Bond Angles (deg)W-N-O 174.3(3)C(6)-W-C(19) 71.8(1)N-W-C(6) 88.0(1)N-W-C(22) 93.3(1)C(6)wC(22)C(8)C(7) C(20)C(10)C(26)C(9)C(25)175well to the single C-C bonds in the dangling propyl group (1.5 18-1.538 A). Therefore, there isno unsaturation in this portion of the ligand.The phenyl group is bound to the tungsten center in what can be described as an ri3-benzylinteraction, with the alpha, ipso, and ortho carbons all within bonding distance of the metalcenter. The ipso and ortho carbon-tungsten distances (2.38 1(4) and 2.371(4) A, respectively)are slightly longer than the alpha carbon-tungsten length (2.301(4) A), which is a typicalobservation in complexes containing this type of interaction.8Once the molecular structure of complex 6.2 had been determined, its NIvIR data could thenbe assigned with the help of1H- COSY and1H-3CHETCOR spectra. The NIvIR data forcomplex 6.4 have been assigned in an analogous manner, and similar data for complex 6.3 havebeen assigned by analogy to the pentane-activation product. Many of the 13C resonances for thearyl groups in these compounds either are not observed in their 13C { 1H} NMR spectra or appearas very broad peaks. The signals for the ortho protons are also not observed in the 1H NMRspectra of all three compounds, and the other aryl protons are represented by a very broad peakassigned to the meta protons and a broad triplet assigned to the para proton.It is proposed that the aryl groups in all three compounds are stereochemically nonrigid andundergo a r13 —* r1 —> r3 process in solution, with a fast rotation at the ii stage. Thismechanism of exchange has been observed before in another3-benzyl compound,CpM(CO)(i13-CH26H5),and its derivatives.9 The 13C{1H} NMR spectra for all threecomplexes, 6.2 - 6.4, contain only two sharp peaks in the aryl region that are assignable to thepara and meta carbons. Broad peaks in this area are observable in the spectra of complex 6.4,and are presumably the other aryl carbon resonances. The same resonances in the spectra ofcomplexes 6.2 and 6.3 are likely too broad to be observed. The ortho proton signal is also notobserved in the 1H NMR spectra for any of the three complexes, and is presumably so broad asnot to be discernible from the baseline. Variable temperature NMR spectroscopy could possiblyrender the ortho signals observable, and this is planned for fhture work.• 176A proposed mechanism for the formation of compounds 6.2 - 6.4 is offered in Scheme 6.2,with the specific example of pentane activation being shown. The initial steps involve theelimination of TMS to form the alkyne intermediate, which then activates the C-H bond of thesubstrate to give an alkyl alkenyl complex. These steps are analogous to the mechanism forarene C-H activation discussed above. The alkyl ligand of the alkyl alkenyl intermediate thusformed is unstable toward f3-hydrogen elimination, and an alkene hydride complex is thusproduced. It is worth noting here that complex 2.8, the starting material in this reaction, doesScheme 6.2R = CH2S1Me3M = Cp*W(NO)_©—SIMe4 M/HIMJ’$)6.2177not possess f3-hydrogens on its alkyl ligand and is thus stable toward this type of decomposition.It is also noteworthy that we have not been able to synthesize 16-valence-electron species of thetype Cp’M(NO)(R)R’ with ligands that contain f3-hydrogens, presumably because theydecompose via this elimination pathway.The next step in the proposed mechanism shown in Scheme 6.2 is hydride migration onto thea-carbon of the alkenyl ligand. Alternatively, this step can be thought of as reductive eliminationto form the bis-olefin intermediate shown. Olefin coupling then affords the observed products.Theoretically, many isomers are possible from the coupling of the two asymmetric olefins, butexperimentally one major isomer is formed. Some small peaks are observable in the NMRspectra for the alkane activation products, but if these are due to another isomer, they indicatethat the other isomer is present in less than 10% that of the major isomer. As an example of thedifferent isomers that could possibly form, the alkene ligand originating from the alkane substratecould couple in an opposite manner to form a metallacycle that is substituted at the alphaposition. However, this isomer would be sterically disfavored. The stereoselectivity at the olefincontaining the phenyl ring could be explained by invoking anr3-benzyl-like interaction betweenphenyl group and the metal center that presumably ‘holds’ the alkene in position while couplingoccurs, and stabilizes the product once it has formed. It is also probable that the couplingreaction is reversible, such that isomers that are sterically unfavored or do not contain the phenylring on the alpha carbon uncouple to re-form the bis-olefin intermediate. Once the most stablemetallacycle is formed, it would be ‘locked’ in position and stabilized by the metal-benzylinteraction.Each of the steps in the mechanism proposed in Scheme 6.2 has precedents in otherorganotransition-metal systems. Reductive elimination, olefin coupling, and 3-hydrogenelimination are all very well-known reactions. 10 The activation of aliphatic C-H bonds bytransition-metal complexes has also been observed many times, but has been less studied than theother reactions mentioned.11 The reaction between complex 2.8 and hexane or pentaneconstitutes the functionalization of a saturated hydrocarbon, which is a highly sought-after178process in organometallic chemistry. Although this reaction is not catalytic, it remains interestingbecause it represents a double C-H bond activation of a saturated hydrocarbon with concomitantC-C bond formation to produce the metallacycle. Attempts to remove these newly formedorganic ligands from the metal center or to fi.irther fhnctionalize them have not been made as yet,but are planned for future work.The unique feature of the C-H bond activation reactions presented here is the selectivenature of the activation process. A major goal in the functionalizing of aliphatic hydrocarbons isselectivity: to activate the hydrocarbon using a complex that differentiates between primary,secondary, and tertiary C-H bonds such that a mixture of isomers is not produced due toactivation at different positions on the organic substrate. The problems involved are exemplifiedby a study by Janowicz and Bergman, who observed all possible isomers due to activation ofboth primary and secondary C-H bonds in pentane upon photolysis of Cp’Ir(L)H2complexes.’2Fortunately, upon warming the isomer mixture to 110 °C, all of the secondary productsisomerized to the primary product, the neopentyl hydride compound. The system studied in thisThesis is unique in that only one major isomer is formed. This may well be due to the nature ofthe alkene hydride intermediate that is proposed to form during the reactions (Scheme 6.2). Atthis stage of the mechanism, reversible olefin insertion/hydride elimination could conceivablyisomerize the organic group to the more sterically favored c-olefin, even if the alkane is initiallyactivated at an internal position of the hydrocarbon. This isomerization is well known in olefinhydrometalation chemistry.13 Once the isomerization to the terminal olefin has occurred,coupling with the other olefin produces the metallacyclic product. It is interesting thatthermolysis of the alkenyl complex, 2.8, in cyclohexane does not lead to a species analogous tothe C-H activation products. This may reflect an unfavorable steric environment rather than theinability of the system to activate methylene groups. Obviously, more experiments are needed todetermine the generality of the activation reactions.Not only is the mechanism of formation of the metallacyclic products 6.2- 6.4 noteworthy,but the products themselves are intriguing. We have been unable to synthesize 16-valence-179electron metallacycles on the Cp’M(NO) fragment, even though many analogous dialkyl anddiaryl species are known. Although many metallacyclic magnesium reagents can besynthesized,14none that we have tried have been successfiul in generatingCp’M(NO)(t2-hydrocarbyl) species. The only such metallacycles that we have prepared containa Lewis base coordinated to the metal center,15 the added stabilization provided by the baseevidently being required to isolate these species. The base may cause a change in the orbitalarrangements on the metal such that adjacent metal-carbon bonds can be accommodated. Thecomplexes prepared here resemble the base-stabilized metallacycles previously isolated, as ther3-benzylic interaction donates extra electron density to the metal and presumably stabilizes thecomplexes to allow isolation. Reaction of Complex 2.8 with PMe3Complex 2.8 has been thermolyzed in the presence of an excess ofPMe3, and a product hasbeen isolated. Rather than the expected product formed by trapping of the proposed alkyneintermediate, the phosphine attacks the starting material to form a 1:1 adduct before f3-hydrogenelimination takes place. The product, Cp*W(NO)(CH2SiMe3)(1{PMe}PhCH(6.5), is notthe metal-centered phosphine adduct as might be expected, but rather a metallacyclopropanespecies formed via PMe3 attack at the alpha carbon of the alkenyl ligand (equation 6.5).A,PMe3 0N”IHN”I7e36.5o R0 0 R6.52.8 R = CH2S1Me3The spectroscopic properties of complex 6.5 support this formulation. Thus, thecompound’s 1H NMR spectrum exhibits signals assignable to the methylene protons of the180trimethylsilylmethyl ligand that are not coupled to phosphorus, while the signals due to thecyclopropane methylene protons do exhibit such coupling. The 13C NMR spectrum of complex6.5 indicates that the two metallacyclopropane carbons are strongly coupled to phosphorus, andtheir signals are shifted far upfield compared to their positions in the alkenyl starting material.The JR spectrum of the complex exhibits a very low nitrosyl-stretching frequency of 1484 cm-’,which reflects increased back-bonding to the nitrosyl from the formally anionic metal center. Thechemical shift of the phosphorus nucleus in the compound’s 31P NI4R spectrum (34 ppm) isconsistent with the phosphonium nature of the phosphorus atom. Also, the phosphorus signalappears much farther downfield than the analogous signals for metal-centered PMe3 adductsdescribed in this Thesis (for example: -4.3 ppm for complex 3.7 and -21.9 ppm for complex 3.9).Nucleophilic attack on the alpha position of alkenyl ligands has been observed in othersystems, although the reaction is not a common one.’6 In all cases, the alkenyl ligand containselectron withdrawing groups that enhance the electrophilic character of the alpha carbon of theligand. Selected examples are shown in Figure 6.2, all of which are derived from attack of aphosphine on a parent alkenyl complex.C’PMe3Me3/ COR oC’21)1u13Figure 6.2. Selected examples of metallacyclopropane compounds formed via phosphine attackon alkenyl ligands (references: A16, B16b, C16c)The solid-state molecular structure of complex 6.5 has been determined by an X-raydiffraction study (Figure 6.4). The most interesting feature of the structure is the geometry ofB181the metallacyclopropane ligand. The distance between the two metal-bound carbon atoms in thisligand, 1.463(7) A, can be compared with analogous bond lengths in the three compounds shownin Figure 6.3: 1.44(2) A for A, 1.448(13) A for B, and 1.488(4) A for C. These lengths, whichare longer than a C-C double bond (1.34 A for ethylene) but shorter than a C-C single bond(1.54 A), imply that the ligands display some ylide characteristics, which makes the C-C bonddistances short due to a contribution from theW_(12-(C=C—Pj) resonance structure (Figure6.3). This resonance form places a negative charge on the metal center, which explains the lownitrosyl-stretching frequency observed for complex 6.5. The substitutents on the alpha carbonare bent back away from the metal center in a way that resembles a metallacyclopentane ratherthan anri2-olefin complex, implying that the olefin resonance form is not the only contributor tothe overall description of the bonding.+Ph PMe3Figure 6.3. Resonance forms contributing to the bonding in complex 6.5. W—H H6.3.3 Insights into the Synthesis of Complex 2.8 via HydrogenationAs discussed above and in Chapter 2, the alkyl alkenyl complex, 2.8, may be synthesized byhydrogenation of Cp*W(NO)(CH2SiMe3)in the presence of phenylacetylene. Along with theorganometallic alkenyl product, an organic species is also isolated from the reaction mixture.This product, 1,4-diphenyl-2,3-butadiene, has been characterized by comparing its spectroscopicproperties to an authentic sample. The appearance of this diene could not be explained at thetime, but in light of the thermal chemistry of the alkenyl complex that is discussed above, anexplanation can now be proposed.H182Figure 6.4. View of the solid-state molecular structure ofCp*W(NO)(CH2SiMe3)(i2-C{PMe3}PhCH2)(6.5), including selected bond lengths and angles with esd’s inparentheses.Bond Lengths (A)W-C(20) 2.155(5)W-C(19) 2.243(4)W-C(6) 2.258(4)P-C(19) 1.768(4)Bond Angles (deg)W-N-O 170.9(3)C(20) - W - C(19) 38.8(3)W - C(20) - C(l9) 73.9(3)W - C(19) - C(20) 67.3(3)C(1lC( 12)0C(14)C(8)C(22) C(23)C(2 1)SiC(9)C(24)C(25)C(7)C(18) C( 16)C(17)183The mechanism to explain the appearance of the diene involves the initial formation of thealkenyl complex Cp*W(NO)(CH2S1Me3)(CPh=C (2.8) via hydrotungstation ofphenylacetylene, as presented in Chapter 2. This product then thermally decomposes to theproposed alkyne intermediate, Cp*W(NO)(CPhCH), as discussed above (Scheme 6.3). Sincethis hydrogenation reaction is pet-formed using an excess of phenylacetylene, it is proposed thatthis intermediate then reacts with the excess alkyne in a way that couples the two alkynes to forma metallacyclopentadiene ligand. Hydrogenolysis ofthe metallacycle carbon-tungsten bonds thenreleases the butadiene. This type of alkyne coupling is well known in organotransition-metalScheme 6.3_©‘i-SU4e4 M/HCCPhH2.8R = CHS1Me3M = Cp*W(NO)Hz [Jchemistry,17and there is no reason why the reaction should not occur on the unsaturatedCp*W(NO) fragment.Some support for this mechanism has been obtained. If a solution of the isolated alkenylcomplex, 2.8, is placed under a hydrogen atmosphere in the presence of phenylacetylene, areaction occurs and the same organic product can again be isolated. If complex 2.8 isthermolyzed in the presence of phenylacetylene without a hydrogen atmosphere, no products canbe isolated from the reaction mixture, but as judged by 1H NMR spectroscopy one major speciesis formed. All attempts to isolate this product lead to its decomposition. Thus, it seems that ifRH184this product is the metallacyclopentadiene complex shown in Scheme 6.3, it is not stable anddecomposes during attempted isolation.Although very little of the butadiene is isolated from these reactions, its formation isintriguing since it indicates that the metal center in the Cp*W(NO) fragment is capable of alkynecoupling, which has not been observed before in these systems. Further experiments, includingthe attempted coupling of other alkynes, are warranted.6.4 SummaryThis Chapter presents some of the thermal chemistry exhibited by the alkyl alkenyl complex,Cp*W(NO)(CH2SiMe3)(CPh=C.Upon heating, the alkenyl compound undergoes13-elimination of one of the alkenyl methylene hydrogens, and TMS is liberated from themolecule. The intermediate thus formed is very reactive, and has been shown to activate botharomatic and aliphatic C-H bonds. Activation of benzene in this manner results in the formationof the phenyl alkenyl species, Cp*W(NO)(Ph)(CPh=CH2),but the activation of saturatedhydrocarbons such as pentane and hexanes is much more complicated and forms metallacyclicproducts. The latter reaction can be viewed as double C-H bond activation of the substrate withC-C bond formation to form the metallacycle, and constitutes a functionalization of the saturatedorganic substrate.The alkyl alkenyl species has also been thermolyzed in the presence of a number of reagentsin an attempt to trap the intermediate formed upon TMS liberation. This intermediate isproposed to be an alkyne complex. When this species is generated in the presence of an alkyne, ametallacyclobutadiene complex is proposed to form which, upon reaction with hydrogen,liberates the butadiene. It thus appears that this metal fragment is capable of alkyne coupling.The presence ofPMe3 during the thermolysis of the alkyl alkenyl compound prevents theelimination of TMS by an alternate reaction with the starting material. The base attacks thea-carbon of the alkenyl ligand to form a metallacyclopropane with a phosphonium groupattached. This complex has been structurally characterized and compared to known similarspecies.Much work remains in the study of the chemistry of the alkyl alkenyl compound,Cp*W(NO)(CH2SiM3)(CPh=C.The thermal chemistry of this complex has only brieflybeen investigated, and the reactions presented in this Chapter are only a cursory look at thissystem that will undoubtedly continue to yield interesting results.1851866.5 References and Notes(1) Legzdins, P.; Rettig, S. J.; Veitheer, J. E.; Batchelor, R. I.; Einstein, F. W. B.Organometallics 1993, 12, 3575.(2) (a) Debad, J. D.; Legzdins, P.; Young, M. A. I Am. Chem. Soc. 1993, 115, 2051. (b)Christensen, N. J.; Hunter, A. J.; Legzdins, P. Organometaiics 1989, 8, 930.(3) Dryden, N. H.; Legzdins, P.; Rettig, S. J.; Veitheer, I. E. Organometallics 1992, 11,2583.(4) (a) Doherty McGrady, N.; McDade, C.; Bercaw, J. E. in Organometallic Compounds.Synthesis, Structure, and Reactants, Shapiro, B. L., Ed.; Texan A+M University Press:Texas, 1983; p 46. (b) Erker, G.; Zwettler, R.; Kruger, C.; Hyla-Kryspin, I.; Gleiter, R.Organometallics 1990, 9, 524. (c) Bruce, G. C.; Knox, S. A. R.; Phillips, A. J. I Chem.Soc., Chem. Commun. 1990, 716. (d) Omori, H.; Suzuki, H.; Moro-oka, Y.Organometallics 1989, 89, 1576.(5) (a) Hyla-Kryspin, I.; Gleiter, R.; Kruger, C.; Zwettler, R.; Erker, G. Organometallics1990, 9, 517. (b) Schwartz, J.; Labinger, J. A. Angew. Chem. mt. Ed Engi. 1976, 15,333.(6) Rosenthal, U.; Ohff A.; Michalik, M.; Görls, H.; Burlakov, V. V.; Shur, V. B. Angew.Chem. mt. Ed EngL 1993, 32, 1193.(7) Shilov, A. E. in Activation and Functionalization ofAlkanes; Hill, C. L., Ed., WileyInterscience: New York; 1989, Chapter I.(8) (a) Carmona, E.; Mann, J. M.; Paneque, M,; Poveda, M. L.; Organometallics 1987, 6,1757. (b) Bleeke, J. R.; Burch, R. R.; Coulman, C. L.; Schardt, B. C. Inorg. Chem. 1981,20, 1316. (c) Burch, R. R.; Muetterties, E. L.; Day, V. W. Organometallics 1982, 1, 188.(9) Cotton, F. A.; Marks, T. J. I Am. Chem. Soc. 1969, 91, 1339.187(10) Coliman, J. P.; Hegedus, L. S.; Norton, J. R.; Finke, R. G. Principles and Applications ofOrganotransition Metal Chemistry; University Science Books: Mill Valley, CA, 1987.(11) For recent reviews, see: (a) Jones, W. D. in Activation and Functionalization ofAlkanes;Hill, C. L., Ed., Wiley-Interscience: New York; 1989, Chapter IV. (b) Crabtree, R. H. inThe Chemistry ofAlkanes and Cycloalkanes; Patai, S. and Rappopori, Z., Eds., WileyInterscience: New York; 1992, p 653.(12) Janowicz, A. H.; Bergman, R. G. J Am. Chem. Soc. 1983, 105, 3929.(13) Hart, D. W.; Schwartz, J. J Am. Chem. Soc. 1974, 96, 8115.(14) See, for example: (a) footnote 9, Chapter 9 and references therein. (b) Lappert, M. F.;Martin, T. R.; Mime, C. R. C.; Atwood, J. L.; Hunter, W. E.; Pentilla, R. E. J Organomet.Chem. 1980, 192, C35. (c) Diversi, P.; Ingrosso, G.; Lucherini, A. Inorg. Synth. 1985, 22,167. (d) Seetz, J. W. F. L.; Van de Heisteeg, J. J.; Schat, G.; Akkerman, 0. S.;Bickeihaupt, F. I Mol. Catal. 1985, 28, 71.(15) Brunet, N.; Debad, J. D.; Legzdins, P.; Trotter, J.; Veltheer, J. E.; Yee, V. C.Organometallics 1993, 12, 4572.(16) (a) Scordina, H.; Kergoat, R.; Kubicki, M. M.; Guerchais, J. E.; L’Haridon, P.Organometaiics 1983, 2, 1681. (b) Carmona, E.; Giiitiérrez-Puebla, E.; Monge, A.;Mann, J. M.; Paneque, M.; Poveda, M. L. Organometallics 1989, 8, 967. (c) Alt, H. G.;Thewait, U. I Organomet. Chem. 1984, 268, 235.(17) Footnote 10, Chapters 9 and 18, and references therein.188AppendixKinetic Data and Calculations for Chapters 4 and 5.A.1 Kinetic Data and Calculations of Rates and Activation Parameters for the ReactionBetween Acetone and Complex 2.9Table Al. Data for the reaction between complex 2.9 and acetone at different acetoneconcentrations, all performed at 55 °C. The numbers represent values of ln[j-AtI/IA0-A},where A is the integrated value of the hydride or product Cp* peak divided by the integrationvalue of the standard TMS peak.[acetone] 0.685 M [acetone] 1.027 Mtime hydride product time hydride product0 0 0 0 0 0240 -0.11666 -0.14324 300 -0.12389 -0.17502480 -0.21381 -0.19176 600 -0.2623 -029466720 -0.33466 -0.32198 900 -0.37212 -0.3464960 -0.43671 -0.42246 1200 -0.50608 -0.490411200 -0.5613 -0.51498 1500 -0.67368 -0.66411500 -0.71478 -0.67035 1800 -0.75814 -0.742131800 -0.81644 -0.77451 2100 -0.88428 -0.841362100 -0.98505 -0.93979 2400 -0.99687 -1.013172400 -1.10994 -1.04635 2700 -1.17692 -1.158792700 -1.24041 -1.17096 3000 -1.41554 -1.254883000 -1.36398 -1.2804 3300 -1.37184 -1.328473300 -1.51439 -1.3981 3600 -1.5858 -1.467893600 -1.58651 -1.52279 3900 -1.77004 -1.605473900 -1.82556 -1.71374 4200 -1.77464 -1.720954200 -1.83607 -1.76313 4500 -2.02814 -1.864664500 -2.0387 -1.92584189ln[IA-AtI/IAo-AIj values:A, At, and A0 represent values of the integration of the Cp* peak of the compound beingmonitored divided by the integration for the standard TMS peak. The infinity value was[acetone] = 1.437 M [acetone] = 1.711 Mtime hydride product time hydride product0 0 0240 -0.17368 -0.15743480 -0.31197 -0.27808720 -0.47016 -0.42133960 -0.64997 -0.587811200 -0.81166 -0.738961440 -0.93882 -0.86121680 -1.16874 -1.040311920 -1.27288 -1.173682160 -1.33425 -1.310972400 -1.52764 -1.464082640 -1.76595 -1.625752880 -1.96366 -1.782113120 -2.0178 -1.946023360 -2.20237 -2.086283600 -2.21191 -2.205133840 -2.58298 -2.353050 0 0300 -0.18376 -0.1625600 -0.33045 -0.25674900 -0.50621 -0.419361200 -0.66333 -0.576841500 -0.82929 -0.720461800 -0.98831 -0.877342100 -1.15106 -1.026122400 -1.26006 -1.142282700 -1.42883 -1.274473000 -1.73391 -1.516463300 -1.7683 1 -1.637473600 -2.08353 -1.84163900 -2.04054 -1.85561[acetone] = 2.054 Mtime hydride product[acetone] 2.738 Mtime hydride product0 0 0240 -0.19072 -0.15633480 -0.38023 -0.29529720 -0.59953 -0.47077960 -0.8055 -0.616811200 -1.04117 -0.776611440 -1.31425 -0.918181680 -1.4158 -0.992371920 -1.65593 -1.151172160 -2.23252 -1.359082400 -2.29347 -1.438162640 -2.39428 -1.642852880 -2.3935 -2.035943120 -2.54474 -2.156783360 -2.45582 -2.286433600 -2.93897 -2.742130 0 0240 -0.19883 -0.16296480 -0.42407 -0.37577720 -0.56494 -0.5 1678960 -0.80176 -0.723 171200 -1.06989 -0.934091440 -1.16637 -1.069861680 -1.47387 -1.285771920 -1.62584 -1.485892160 -2.16536 -1.748132400 -2.53021 -2.024322640 -3.02846 -2.154062880 -2.42896 -2.24675measured when all of the starting material had reacted as judged by the disappearance of the Cp*190resonance of this compound. The initial value, A0, was measured at time = 0, and A values weremeasured at timed intervals thereafter. The absolute values of the differences are used so thatstarting material and product data could be calculated using the same equation, and their valuescould be compared directly.191Table A2. Data for the reaction between complex 2.9 and acetone (25 times excess) at differenttemperatures. The numbers represent values of ln[I -A/A0j], where A is the integratedvalue of the hydride or product Cp* peak divided by the standard TMS peak area.45 deg. 65 deg.tune (sec) hydride product time (sec) hydride product0 0 0 0 0 0600 -0.11388 -0.11831 90 -0.10674 -0.103241200 -0.22678 -0.21882 180 -0.26708 -0.259911800 -0.36418 -0.36389 270 -0.45282 -0.436942400 -0.45526 -0.429 390 -0.66476 -0.62983000 -0.58634 -0.55314 480 -0.82427 -0.78653600 -0.7165 -0.66282 570 -0.92586 -0.883784200 -0.84444 -0.79483 660 -1.09401 -1.045664800 -0.98484 -0.9175 750 -1.34048 -1.219335400 -1.04483 -0.98348 840 -1.26943 -1.23684930 -1.44355 -1.433311020 -1.7995 -1.612851110 -1.99085 -1.778571200 -2.24645 -1.955731290 -2.31284 -2.098131380 -2.36932 -2.17351470 -2.17954 -2.308675deg.triall 75deg.trial2time (sec) hydride product time (sec) hydride product0 0 0 0 0 090 -0.38534 -0.33908 60 -0.24876 -0.18974180 -0.8558 -0.78444 180 -0.71334 -0.59852270 -1.40598 -1.31487 300 -1.13948 -1.03849360 -1.77947 -1.6472 420 -1.63846 -1.48606450 -2.24614 -2.05739 630 -2.3664 -2.27747540 -3.78956 -2.97676 820 -3.74085630 -4.3242192-9.5-10-10.5R.t-11-11.5-12-125Figure A.1. Plot of ln(kobs/T) versus lIT for the reaction of complex 2.9 with acetone. Datawere obtained by monitoring the disappearance of the hydride starting material.Plots of ln(k/T) versus l/T for both sets of data (starting material and product monitoring)are shown in Figures A. 1 and A.2. The slope of the graph produced using data from themonitoring of starting material is -9662.8 with an intercept of 18.0 10. The equation used tocalculate the activation parameters is:ln(k!T) ln(k’Ih) + zS/R - AH/RT,where k’ is the Boltzmann’s constant, and h is Planck’s constant. Thus,slope = -9662.8 = - z\H/RAH (9662.8)R = 80.34 kJ/mola2.9 2.95 3 ac5 3.1 a151ITi1000K193-9-9.5-10.-10.5..c— -11.5-12-12.5—13 •l-- I I285 3.( 3.15lIT xl000KFigure A.2. Plot of ln(kobs/T) versus lIT for the reaction of complex 2.9 with acetone. Datawere obtained by monitoring the appearance of the acetone-containing product.The value for the entropy of activation is found from the intercept value:intercept ln(k’/h) +18.010 = 23.8 + 5/8.314ASt= -48.14 JImolKSimilar calculations using the data obtained by monitoring the product appearance give thefollowing values:= 82.80 id/mo!AS -41.24 J/molK.Error analysis can be accomplished in many ways. The one chosen here is to use thedifference between the two values as one-half of the experimental error and average thecalculated values. This method gives the following values and errors: = 81.6 ± 10 kJ/mol2.9 2.95 3 3.1194and AS = -45 ± 14 J/moIK. This is by no means a rigorous error analysis, but should give arough estimate of the experimental error involved.A.2 Kinetic Data and Calculation of the Kinetic Isotope Effect for the Reaction BetweenBenzene and Complex 2.9Table A.3. Data for the monitoring of the reactions between 2.9 and C6H and 2.9 and C6D.The data given are values of ln[IA-AtI/lAo-Al], where A is the integrated value of the hydrideor product Cp* peak divided by the standard hexamethylbenzene peak area.Beuzene Benzene-d6time(sec) hydride product time(sec) hydride product0 0 0 0 0 01800 -0.28975 -0.24029 1800 -0.17106 -0.144083600 -0.5761 -0.50924 3600 -0.35869 -0.31725400 -0.83389 -0.82619 5400 -0.53472 -0.497047200 -1.104 -1.09657 7200 -0.62146 -0.645599000 -1.34618 -1.43835 9000 -0.75205 -0.8559410800 -1.71975 -1.77312 10800 -0.95671 -0.9771112600 -1.82965 -2.07431 12600 -1.14413 -1.1692514400 -2.1297 -2.58675 14400 -1.18734 -1.2958818000 -2.95585 -3.10305 18000 -1.48375 -1.6584521600 -3.42191 -3.59204 21600 -1.83254 -2.05188Calculation of kinetic isotope effect:The rates of the reactions between 2.9 and C6H and 2.9 and C6D are given by the slopesof the plots of ln[j .-A1j/j0-AI] versus time. An approximate estimate of the experimentalerror is obtained by using the difference between the rate obtained for the starting material andthe product.195For C6H:Hydride monitoring: kobs = 1.579 X 1O S.Product monitoring: kobs = 1.735 X S.The average, with associated error, is: kj05 = 1.7 ± 0.3 x 10 s1.For C6D:Hydride monitoring: kobs = 0.835 1 X S.Product monitoring: kobs = 0.9382 X io s-’.The average, with associated error, is: kD,obs = 0.9 ± 0.2 x 10 s.The kinetic isotope effect is thus:kHobIkDobs = 1.7 ± 0.3 x 10-4/0.9 ± 0.2 x io = 1.9 ± 0.5196A.3 Kinetic Data for the Reaction BetweenC6H/CDand Complex 2.8Table A.4. Data for the monitoring of the reactions between 2.8 and C6H and 2.8 and C6D.The data given are values of1n[(At-A)/(A0-A)],where A is the integrated value of the startingmaterial’s SiMe3 peak divided by the standard hexamethylbenzene peak area.time C6H C6])0 0 01800 -0.15956 -0.153083600 -0.37579 -0.336145400 -0.5276 -0.576237200 -0.71925 -0.678549000 -0.95153 -0.9572710800 -1.1298 -1.1709512600 -1.34601 -1.2667314400 -1.6914 -1.6584516200 -1.84037 -1.7340418000 -2.13664 -1.9260721600 -2.61225 -2.16629Rate Constants:From the data above, the observed, pseudo-first order rate constants for the activation ofC6D and C6H by complex 2.8 are 1.21 x io and 1.06 x 1O s, respectively. Since this isthe same method used to measure rate constants for the activation of benzene by complex 2.9,the same experimental errors are involved. Therefore, the same error is applied to thesenumbers, which gives values of:kp,05 1.1±0.2x 104skD,ob = 1.2 ± 0.3 x io s1.0.1-04C-09-14CC—24-22197time (sec)0 51X’O icoo 150w 2XJ00 250DFigure A.3. Plot of the data contained in Table A.4 (C6D (.); C6H (•)).


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