Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Development of tungsten nitrosyl alkylidene complexes for activation of hydrocarbons Tran, Elizabeth 2002

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

Item Metadata

Download

Media
831-ubc_2002-732533.pdf [ 9.66MB ]
Metadata
JSON: 831-1.0061332.json
JSON-LD: 831-1.0061332-ld.json
RDF/XML (Pretty): 831-1.0061332-rdf.xml
RDF/JSON: 831-1.0061332-rdf.json
Turtle: 831-1.0061332-turtle.txt
N-Triples: 831-1.0061332-rdf-ntriples.txt
Original Record: 831-1.0061332-source.json
Full Text
831-1.0061332-fulltext.txt
Citation
831-1.0061332.ris

Full Text

D E V E L O P M E N T O F T U N G S T E N N I T R O S Y L A L K Y L I D E N E C O M P L E X E S F O R A C T I V A T I O N O F H Y D R O C A R B O N S by Elizabeth T r a n B.Sc. (Hons), University of Ottawa, 1994 A THESIS S U B M I T T E D I N P A R T I A L F U L F I L L M E N T O F T H E R E Q U I R E M E N T S F O R T H E D E G R E E OF D O C T O R OF P H I L O S O P H Y in T H E F A C U L T Y OF G R A D U A T E S T U D I E S (Department of Chemistry) We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH C O L U M B I A December 2001 © Elizabeth Tran, 2001 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of Cji P tm /_T ~f r KJ The University of British Columbia ^ Vancouver, Canada Date bee //j ZOO/ DE-6 (2/88) Abstract A series of tungsten alkyls of stoichiometry Cp 'W(NO)(R)(R ' ) [Cp' = Cp, Cp*; R = C H 2 C M e 3 , C H 2 C M e 2 P h , C H 2 S i M e 3 ; R ' = C H 2 C M e 3 , C H 2 C M e 2 E t , C H 2 C M e 2 P h , C H 2 S i M e 3 , C H 3 , Ph, C l , C H 2 P h , O C H 3 , N M e 2 ] is shown by N M R and, in some cases, IR spectroscopies to adopt an a-agostic structure when secondary interactions such as lone pair or 7i-electron donation are not possible. This was confirmed by X-ray diffraction studies of several of these compounds and by a neutron diffraction study of C p * W ( N O ) ( C H 2 C M e 3 ) 2 . In each, the agostic interaction is best described as a closed, 3-center, 2-electron bond and serves to give the electron-deficient tungsten center an 18-electron configuration. The complexes in which R and R ' are neopentyl-like display fluxional behavior in solution, exhibiting averaged N M R spectra as a result of a competition between R and R ' for formation of an a-agostic bond with the metal. This competition is manifested in terms of unusual N M R chemical shifts and coupling constants for the c t -CH 2 groups. These data establish that the strength of the C - H — W interaction is dependent on the nature of C p ' , R, and R ' . For complexes with R = R ' , it varies in the order Cp* > Cp and R = R ' = C H 2 C M e 3 ~ C H 2 C M e 2 P h > C H 2 S i M e 3 . For complexes with R ± R ' , two trends are observed: First, among all the alkyl ligands studied, the relative a-agostic donor ability is C H 2 C M e 3 > C H 2 C M e 2 E t > C H 2 C M e 2 P h > C H 2 S i M e 3 » C H 3 ( C H 2 P h is nonagostic). Second, when R = C H 2 C M e 3 and R ' = C H 2 C M e 3 , C H 2 C M e 2 E t , C H 2 C M e 2 P h , C H 2 S i M e 3 , C H 3 , Ph, C l , or C H 2 P h , the a-agostic donor ability of R decreases as R ' is changed from Ph to C H 3 to C l to C H 2 S i M e 3 to C H 2 C M e 2 P h to C H 2 C M e 2 E t to C H 2 C M e 3 to C H 2 P h . The neopentyl complexes, C p ' W ( N O ) ( C H 2 C M e 3 ) 2 [Cp' = Cp, Cp*] and C p * W ( N O ) ( C H 2 C M e 3 ) ( C H 2 P h ) , undergo facile a-abstraction of neopentane at 60-70 °C to give transiently the 16-electron alkylidene complexes, [Cp 'W(NO)(=CHCMe 3 ) ] and [Cp*W(NO)(=CHPh)], respectively. Both [Cp 'W(NO)(=CHCMe 3 ) ] and [Cp*W(NO)(=CHPh)] can be trapped with tertiary phosphines in T H F solvent to yield the corresponding phosphine adducts as the anti rotamer. In the case of [Cp*W(NO)(=CHCMe 3 ) ] , the trapping can also be effected in neat pyridine or M e 2 N E t . Cp*W(NO)(=CHCMe 3 ) (PMe 3 ) undergoes associative addition reactions with M e 2 N H , M e O H , and P h C 0 2 H to give C p * W ( N O ) ( C H 2 C M e 3 ) ( E R ) [ER = NMe2, O M e , 02CPh] and free PMe3. The amido complex is also formed smoothly when [Cp*W(NO)(=CHCMe 3 ) j is generated in T H F in the presence of M e 2 N H . When [Cp*W(NO)(=CHCMe3)] is generated in alkene solvents, it undergoes [2+2] cycloaddition and/or C - H activation reactions depending on the alkene structure. Cycloaddition is observed for acyclic alkenes which lack allylic C - H bonds and for cyclic alkenes with strained C=C bonds. In the case where the acyclic alkenes contain accessible allylic hydrogens, complicated C - H activation product mixtures are formed. Alkanes and arenes also react readily with [Cp*W(NO)(=CHCMe 3 ) ] at 70 °C by C - H activation. Reactions with alkanes proceed with moderate steric selectivity and produce mixed bis(alkyl), metallacyclobutane, alkene, or allyl complexes depending on the alkane structure. Cp*W(NO)(CH 2CMe3)(CH2SiMe 3) is produced in high yield in the reaction with M e 4 S i . The major product of the reaction with 1,1,2,2-tetramethylcyclopropane is a metallacycle formed by intermolecular C - H activation followed by y-cyclometalation. Neohexane and cyclohexane, which are capable of P-elimination, react to give alkene complexes which can be trapped by PMe3 or M e 2 N H . In contrast, alkanes such as pentane and methylcyclohexane react to give Cp*W(NO)(7 3 -al lyl)(H) complexes in both the presence and absence of PMe3. Reactions with arenes are similarly sensitive to steric effects and involve competitive aryl vs benzyl C - H activation. The benzyl-activated products are unstable at 70 °C, decomposing to give benzylidene complexes that also readily cleave aryl and benzyl C - H bonds. Mechanistic studies, including isotopic labeling, isotope effect, and kinetic studies, strongly implicate that the thermal decompositions of C p ' W ( N O ) ( C H 2 C M e 3 ) 2 proceed by a reversible, rate-determining a-abstraction process to form the neopentane complex, [Cp 'W(NO)(=CHCMe 3)(77 2 -H-CH 2 CMe 3 ) ] , in which the metal can coordinate interchangeably to the neopentane a- and y - C - H bonds. Concurrently, the coordinated neopentane dissociates irreversibly from the metal center to yield [Cp 'W(NO)(=CHCMe 3 ) ] , which can rapidly form adducts with dative ligands or coordinate another hydrocarbon and undergo addition reactions. iv The thermal decompositions of the mixed bis(alkyl) complexes, C p * W ( N O ) ( C H 2 C M e 3 ) ( R ) , where R = C H 2 C M e 2 E t , C H 2 C M e 2 P h , C H 2 S i M e 3 , C H 3 , and Ph, are also reported. These studies show that the pathway by which these compounds decompose is governed primarily by the strengths of the M - C bond broken and formed, and that the presence of a-agostic interactions in the ground state has little, i f any, effect on the mechanism of decomposition. V Table of Contents Abstract i i Table of Contents v List of Figures x i i i List of Tables xv i List of Abbreviations xix Acknowledgments xx i i Dedication x x i i i CHAPTER 1: Background, Scope, and Format 1 1.1 Background 1 1.2 Thesis Outline 2 1.3 Thesis Format 3 1.4 References and Notes 4 CHAPTER 2: Spectroscopic and Structural Studies of a-Agostic Bonding in Cp'W(NO)(R)(R') Complexes 6 2.1 Introduction 6 2.2 Experimental Section 9 2.2.1 Methods 9 2.2.2 Reagents 10 2.2.3 Preparation of E t M e 2 C C H 2 B r 11 2.2.4 Preparation of P h M e 2 C C D 2 B r 11 2.2.5 1 3 C N M R D a t a o f C p W ( N O ) ( C H 2 C M e 3 ) 2 ( 2 . 1 ) 12 2.2.6 l 3 C N M R D a t a o f C p W ( N O ) ( C H 2 C M e 2 P h ) 2 ( 2 . 2 ) 12 vi 2.2.7 M S and 2 H N M R Data of C p W ( N O ) ( C D 2 C M e 2 P h ) 2 (2.2-d4) 13 2.2.8 1 3 C N M R D a t a o f C p W ( N O ) ( C H 2 S i M e 3 ) 2 ( 2 . 3 ) 13 2.2.9 Improved Preparation of C p * W ( N O ) ( C H 2 C M e 3 ) 2 (2.4) 13 2.2.10 Preparation of C p * W ( N O ) ( C H 2 C M e 3 ) ( C D 2 C M e 3 ) (2.4-rf2) 14 2.2.11 1 3 C N M R D a t a o f C p * W ( N O ) ( C H 2 C M e 2 P l i ) 2 ( 2 . 5 ) 14 2.2.12 1 3 C D a t a o f C p * W ( N O ) ( C H 2 S i M e 3 ) 2 ( 2 . 6 ) 14 2.2.13 Preparation of C p * W ( N O ) ( C H 2 C M e 3 ) ( C H 2 C M e 2 E t ) (2.7) 14 2.2.14 1 3 C N M R D a t a o f C p * W ( N O ) ( C H 2 C M e 3 ) ( C H 2 C M e 2 P h ) ( 2 . 8 ) 15 2.2.15 l 3 C and ' H N O E Difference N M R Data of C p * W ( N O ) ( C H 2 C M e 3 ) ( C H 2 S i M e 2 ) ( 2 . 9 ) 15 2.2.16 1 3 C N M R Data of C p * W ( N O ) ( C H 2 C M e 3 ) ( C H 3 ) (2.10) 15 2.2.17 1 3 C N M R Data of Cp*W(NO)(CH 2 CMe 3 ) (Ph) (2.11) 16 2.2.18 1 3 C N M R Data of C p * W ( N O ) ( C H 2 C M e 2 P h ) ( C H 2 S i M e 3 ) (2.12) 16 2.2.19 1 3 C N M R Data of C p * W ( N O ) ( C H 2 C M e 2 P h ) ( C H 3 ) (2.13) 16 2.2.20 1 3 C N M R Data of C p * W ( N O ) ( C H 2 C M e 3 ) ( C l ) (2.14) 16 2.2.21 Preparation of C p * W ( N O ) ( C H 2 C M e 3 ) ( C H 2 P h ) (2.15) 16 2.2.22 Neutron Structure of C p * W ( N O ) ( C H 2 C M e 3 ) 2 (2.4) [See also addendum] . . . . 17 2.3 Results and Discussion 18 2.3.1 Synthesis 18 2.3.2 Spectroscopic Characterization 20 2.3.2.1 IR Studies 20 2.3.2.2 N M R Studies 21 2.3.3 X - R a y Structural Studies 33 2.3.4 Molecular Orbital Description of the Agostic Bonds 40 2.4 Epilogue 41 2.5 Addendum: Neutron Structure of C p * W ( N O ) ( C H 2 C M e 3 ) 2 (2.4) and v i i Reinterpretation of Spectroscopic and Structural Trends 4 4 2.6 References and Notes 48 CHAPTER 3: Thermolysis of Cp'W(NO)(CH2CMe 3 )2 and Cp*W(NO)(CH2CMe3)(CH2Ph) Complexes: Generation and Trapping Reactions of [Cp'W(NO)(=CHCMe3)] and [Cp*W(NO)(=CHPh)] Intermediates 5 4 3.1 Introduction 54 3.2 Experimental Section 57 3.2.1 Methods 57 3.2.2 Reagents 58 3.2.3 Preparation of Cp 'W(NO)(=CHCMe 3 ) (L) Complexes [Cp' = Cp, L = P M e 3 (3.1), PEt 3 (3.2), P (OMe) 3 (3.3); C p ' = Cp*, L = P M e 3 (3.4), P M e 2 P h (3.5), PEt 3 (3.6),P(OMe) 3 (3.7)] 58 3.2.3.1 Complex 3.1 58 3.2.3.2 Complex 3.2 5 9 3.2.3.3 Complex 3.3 5 9 3.2.3.4 Complex 3.4 5 9 3.2.3.5 Complex 3.5 60 3.2.3.6 Complex 3.6 60 3.2.3.7 Complex 3.7 60 3.2.4 Preparation of Cp*W(NO)(=CHCMe 3 ) (NEtMe 2 ) (3.8) 61 3.2.5 Preparation of Cp*W(NO)(=CHCMe 3 ) (Py) (3.9) 61 3.2.6 Preparation of Cp*W(NO)(=CHPh)(PMe 3 ) (3.10) 62 3.2.7 Attempted Trapping of [Cp*W(NO)(=CHCMe 3 ) ] with T H F , E t 2 0 , E t 3 N M e C N , and M e 3 C C N 62 3.2.8 Attempted Preparation o fCp*W(NO)(=CHCMe 3 ) (PMe 3 ) (3.4) in CH2CI2 .... 62 3.2.9 Preparation of C p ' W ( N O ) ( C H 2 C M e 3 ) ( N M e 2 ) [Cp' = Cp* (2.17); C p ' = Cp v i i i (3.12)] 63 3.2.9.1 Complex 2.17 63 3.2.9.2 Complex 3.12 63 3.2.10 Preparation of CpW(NO)(CHDCMe3)(NMe2)(3.12-</) 64 3.2.11 Reactions of Cp*W(NO)(=CHCMe 3 ) (PMe 3 ) (3.4) with M e O H , M e 2 N H , and P h C 0 2 H 64 3.2.11.1 Preparation of C p * W ( N O ) ( C H 2 C M e 3 ) ( O M e ) (2.16) 64 3.2.11.2 Preparation of C p * W ( N O ) ( C H 2 C M e 3 ) ( N M e 2 ) (2.17) 64 3.2.11.3 Preparation of C p * W ( N O ) ( C H 2 C M e 3 ) ( 0 2 C P h ) (3.13) 65 3.2.12 Preparation of C p * W ( N O ) [ C H ( C M e 3 ) C H 2 ( C M e 3 ) C H ] (3.14) 65 3.2.13 Preparation of C p * W ( N O ) [ C H ( C M e 3 ) C H ( C 3 H 6 ) C H ] (3.15) 66 3.2.14 Preparation of Cp*W(NO)[CH(CMe 3 )CH(C 3 H 6 )C(E t ) ] (3.16) 66 3.2.15 Preparation of C p * W ( N O ) [ C H ( C M e 3 ) C H ( C 5 H 6 ) C H ] (3.17) 67 3.2.16 Preparation of C p * W ( N O ) ( C H 2 C M e 3 ) ( 7 3 - C H 2 C ( E t ) C H 2 ) (3.18), Cp*W(NO)(H)(77 3 -CH 2 C(Et)CH 2 ) (3.19) and Cp*W(NO)(7 4-2-methylbutadiene) (3.20) 68 3.2.16.1 Complex 3.18 68 3.2.16.2 Complexes 3.19 and 3.20 69 3.2.17 Preparation of C p * W ( N O ) ( 7 3 - C i 0 H 1 8 ) (3.21) 69 3.3 Results and Discussion 71 3.3.1 Trapping of [Cp 'W(NO)(=CHCMe 3 ) ] and [Cp*W(NO)(=CHPh)] with Lewis Bases 71 3.3.2 Reactions of [Cp 'W(NO)(=CHCMe 3 ) ] and Cp*W(NO)(=CHCMe 3 ) (PMe 3 ) with Heteroatom-Hydrogen Bonds 78 3.3.3 Reaction of [Cp*W(NO)(=CHCMe 3 ) ] with Cycl ic and Acycl ic Alkenes . . . . 80 3.3.3.1 [2+2] Cycloaddition 80 ix 3.3.3.2 C - H Bond Activation 88 3.4 Epilogue 9 6 3.5 References and Notes ^ CHAPTER 4: Selective Activation of Aliphatic and Aromatic C-H Bonds by [Cp*W(NO)(=CHCMe3)] and Related Complexes 105 4.1 Introduction 105 4.2 Experimental Section 107 4.2.1 Methods 107 4.2.2 Reagents 107 4.2.3 Preparation of Cp*W(NO)(CH2CMe3)(CH2SiMe3) (2.9) 107 4.2.4 Preparation of C p * W ( N O ) [ C H 2 C M e C M e 2 C H ] (4.1) 108 4.2.5 Preparation of Cp*W(NO)(772-alkene)(PMe3) Complexes [alkene = cyclopentene (4.2), cyclohexene (4.3), neohexene (4.4)] 108 4.2.5.1 Complex 4.2 1 0 9 4.2.5.2 Complex 4.3 1 0 9 4.2.5.3 Complexes 4.4 1 1 0 4.2.6 N M R Study of the Thermolysis of 2.4 in Cyclohexane in the Presence of P M e 3 : Effect of Varying P M e 3 Concentration 110 4.2.7 N M R Study of the Thermolysis of 2.4 in Neohexane in the Presence of PMe3 111 4.2.8 Thermolysis of C p * W ( N O ) ( C H 2 C M e 3 ) ( C H 2 C M e 2 E t ) (2.7) in Neohexane in the Presence of P M e 3 1 1 1 4.2.9 Thermolysis of C p * W ( N O ) ( C H 2 C M e 3 ) ( C H 2 C M e 2 E t ) (2.7) in T H F in the Presence of PMe3 112 4.2.10 Preparation of Cp*W(NO)(cyclohexyl)(NMe 2 ) (4.7) 113 4.2.11 Preparation of Cp*W(NO)(7 3 -al lyl)(H) Complexes [// 3-allyl = 7 3 - C 6 H 9 X (4.8), ^ - C y H n (4.9), ^ - C 8 H 1 3 (4.10), ; / 3 - C 5 H 9 (4.11)] 113 4.2.11.1 Complexes 4.8 114 4.2.11.2 Complex 4.9endo 1 1 4 4.2.11.3 Complexes 4.10 115 4.2.11.4 Complex 4.1 l e n d 0 115 4.2.12 Reactions of 2.4 with Arenes 116 4.2.12.1 Reaction of 2.4 with a,a,a-Trifluorotoluene 116 4.2.12.2 Reaction of 2.4 with Anisole 1 1 7 4.2.12.3 Reaction of 2.4 with Toluene 1 1 8 4.2.12.4 Reaction of 2.4 with o-Xylene I 1 8 4.2.12.5 Reaction of 2.4 with m-Xylene 1 2 ° 4.2.12.6 Reaction of 2.4 with ^"-Xylene I 2 1 4.2.12.7 Reaction of 2.4 with Mesitylene 1 2 2 4.2.13 Thermolysis of C p W ( N O ) ( C H 2 C M e 3 ) 2 (2.1) in Neat Tetramethylsilane, C 6 D 6 , and /^-Xylene 122 4.2.14 Thermolysis of C p W ( N O ) ( C H 2 C M e 3 ) 2 (2.1) in Cyclohexane in the Presence of Added P M e 3 123 4.3 Results and Discussion 123 4.3.1 Reactions of 2.4 with Tetramethylsilane and Alkanes 123 4.3.2 Reactions of 2.4 with Arenes 139 4.3.3 Thermal Reactions of C p W ( N O ) ( C H 2 C M e 3 ) 2 (2.1) with Aliphatic and Aromatic C - H Bonds: Qualitative Observations 147 4.4 Conclusions I 4 8 4.5 References and Notes 150 CHAPTER 5: Kinetic and Mechanistic Studies of the Generation and C - H x i Activation Reactions of the [Cp'W(NO)(=CHCMe3)] Fragments 157 5.1 Introduction 157 5.2 Experimental Section 160 5.2.1 Methods 160 5.2.2 Reagents 160 5.2.3 N M R and Mass Spectral Data for C p * W ( N O ) ( C H 2 C M e 3 ) ( C D 2 C M e 2 P h ) (2.8-di) 161 5.2.4 Preparation of (CH3)2Si(CD3)2 161 5.2.5 Preparation of C p * W ( N O ) ( C H 2 C M e 3 ) ( C 6 D 5 ) (2.1 l-d5) 162 5.2.6 Kinetic Measurements 162 5.2.7 *H and 2 { ' H } N M R Analyses of the Thermolysis of (a) 2.1-</4 and 2.4-</4 in T H F in the Presence of P M e 3 and (b) 2.4-</4 in C 6 D 6 163 5.2.8 Thermolysis of 2.4 in T H F in the Presence of Excess P M e 3 and CeD6 164 5.2.9 G C - M S Analysis of Neopentane Generated from the Thermolysis of (a) 2.1-d4 and 2.4-</4 in T H F in the Presence of Excess P M e 3 and (b) 2.4-</4 in Neat P M e 3 . . . 164 5.2.10 Thermolysis of 2.4 in S i ( C D 3 ) 4 164 5.2.11 Thermolysis of 2.4 in C 6 D 6 165 5.2.12 Thermolysis of 2.4 in ( C H 3 ) 2 S i ( C D 3 ) 2 165 5.2.13 Thermolysis of 2.4 in 1:1 C 6 H 6 / C 6 D 6 166 5.2.14 Thermolysis of 2.4 in l , 3 , 5 - C 6 H 3 D 3 - c i 3 166 5.2.15 Thermolysis of C p * W ( N O ) ( C H 2 C M e 3 ) ( C H 2 C M e 2 P h ) (2.8) in T H F in the Presence of Added P M e 3 167 5.2.16 Thermolysis of C p * W ( N O ) ( C H 2 C M e 3 ) ( C D 2 C M e 2 P h ) (2.8-</2) in T H F in the Presence of Added P M e 3 168 5.2.17 Thermolysis of C p * W ( N O ) ( C H D C M e 3 ) [ C D 2 S i ( C D 3 ) 3 ] (2.9-</,2) in S i ( C H 3 ) 4 169 5.2.18 Thermolysis of C p * W ( N O ) ( C H 2 C M e 3 ) ( C H 3 ) (2.10) 169 x i i 52.19 Thermolysis of Cp*W(NO)(CH2CMe3)(C6D5) (2.1 l-rf5) in C 6 D 6 170 5.3 Results and Discussion 170 5.3.1 Isotopic Labeling Studies 170 5.3.2 Isotope Effect Studies 175 5.3.3 Kinetic Studies 178 5.3.4 Thermolyses of Other Cp 'W(NO)(R)(R ' ) Complexes: Qualitative Observations 184 5.4 Conclusions 189 5.5 References and Notes 190 A P P E N D I X 193 X l l l List of Figures Figure 2.1. Nujol-mull IR spectrum of (A) C p W ( N O ) ( C H 2 C M e 2 P h ) 2 (2.2) and (B) C p W ( N O ) ( C D 2 C M e 2 P h ) 2 (2.2-d4) 21 Figure 2.2. 500-MHz *H N M R spectrum of the methylene protons of 2.4 in C 6 D 6 at 20 °C 25 Figure 2.3. Plot of 1JQH versus 8 C a as a function of C p ' and R for complexes 2.1-2.6 ( C p = A ; C p * = B ) .' 26 Figure 2.4. 500-MHz ! H N M R spectrum of 2.9 in C 6 D 6 at 20 °C 27 Figure 2.5. Schematic representation of 2.9 along the C p * - W vector. The Cp* ligand (situated above) is not shown for reasons of clarity 28 Figure 2.6. Plot of the neopentyl ligand's ' JCH versus 8 C a as a function of R ' for complexes 2.7-2.11 and 2.14-2.15 29 Figure 2.7. O R T E P diagrams of (A) C p * W ( N O ) ( C D 2 C M e 3 ) 2 (2A-d4), (B) C p * W ( N O ) ( C H 2 C M e 3 ) ( C H 2 S i M e 3 ) (2.9), (C) C p * W ( N O ) ( C H 2 C M e 3 ) ( 2 , 5 - C 6 H 3 - M e 2 ) (2.11'), and (D) C p * W ( N O ) ( C H 2 C M e 3 ) ( N M e 2 ) (2.17) 3 6 Figure 2.8. Qualitative molecular orbital diagram for the double a-agostic interaction in Cp 'W(NO)(R ' ) (R ' ) complexes 41 Figure 2.9. Two views of the neutron structure of C p * W ( N O ) ( C H 2 C M e 3 ) 2 (2.4) along with selected bond distances and angles for 2.4 (neutron) and 2A-d4 (X-ray) 45 Figure 3.1. O R T E P diagrams and selected bond distances and angles for complexes (A) CpW(NO)(=CHCMe 3 ) (PMe 3 ) (3.1) and (B) Cp*W(NO)(=CHCMe 3 ) (PMe 3 ) (3.4)... 77 Figure 3.2. 500-MHz ' H and 75-MHz 1 3 C { ' H } N M R spectra of 3.16 in C D C 1 3 85 Figure 3.3. O R T E P diagrams and bonding parameters for the tungstacyclic rings of (A) C p * W ( N O ) [ C H ( C M e 3 ) C H 2 C H ( C M e 3 ) ] (3.14), (B) xiv Cp*W(NO)[CH(CMe3)CH(C 3 H6)CH] (3.15), and (C) Cp*W(NO)[CH (CMe3 )CH(C 5 H6)CH] (3.17) 87 Figure 3.4. O R T E P diagram and selected bond distances and angles of complex C p * W ( ^ 0 ) [ 7 4 - O T 2 C ( M e ) O T C H 2 C ( M e ) 2 C H ( M e ) ] (3.21a) 93 Figure 4.1. O R T E P diagrams and selected bond distances and angles for complexes (A) Cp*W(NO)(cyclohexene)(PMe 3) (4.3) and (B) Cp*W(NO)(neohexene)(PMe 3) (4.4syn) 128 Figure 4.2. O R T E P diagram along with selected bond distances and angles of complex Cp*W(NO)(cyclohexyl)(NMe 2 ) (4.7) 133 Figure 4.3. O R T E P drawings and selected bond distances and angles of (A) Cp*W(NO)(77 3-C 7H 1 1)(H) (4.9e„do) and (B) Cp*W(NO)(773-C8H13)(H)(4.11exo) 137 Figure 4.4. Variable-temperature 500-MHz *H N M R spectra of 2.11' in toluene-Jg ( • ) 1 4 3 Figure 5.1. 500-MHz ' H N M R spectra of the methylene regions of (A) Cp*W(NO)(CH 2 CMe 3 )(C6H 5 ) (2.11) and C p * W ( N O ) ( C H D C M e 3 ) ( C 6 D 5 ) (2.11-</6) and (B) Cp*W(NO)(CH 2 CMe 3 ) (2 ,4 ,6 -C 6 H 2 D3) (2.11-</3-H) and Cp*W(NO)(CHDCMe3)(3,5-C6H 3 D 2 ) (2.11-</3-D) 177 Figure 5.2. Representative first-order kinetic plots for the decomposition of 2.1 (•) and 2.4 (•) in T H F in the presence of excess P M e 3 at 81 °C 180 Figure 5.3. Eyring plots for the thermal decompositions of 2.1 (•) and 2.4 (•) in cyclohexane in the presence of excess PMe3 182 Figure 5.4. O R T E P diagram and selected bond distances and angles of complex Cp*W(NO)(CH 2 CMe 2 -o -C6H4 ) (PMe 3 ) (5.1) 186 Figure A l . Plot of the kinetic data for the products of the thermolysis of 2.4 in neohexane in the presence of excess P M e 3 : Cp*W(NO)(=CHCMe 3 ) (PMe 3 ) (3.4) = • ; Cp*W(NO)(syH-neohexene) (PMe 3 ) (4.4syn) = • ; Cp*W(NO)(arcri-neohexene)(PMe 3) XV ( 4 . 4 a n t i ) = ± 193 Figure A 2 . ! H - ' H E X S Y spectrum of Cp*W(NO)(CH 2CMe3)(C 6H3-2,5-Me2) (2.11') in toluene-^ at -50 °C. 194 Figure A 3 . ' H N M R spectra of the methylene regions of (A) Cp*W(NO)(CH 2CMe3)[CH 2Si(CH 3)(CD3)2] (2.9-</6-H) and Cp*W(NO)(CHDCMe 3 ) [CD 2 Si (CH3 ) 2 (CD 3 ) ] (2.9-</3-D) [obtained from thermolysis of 2.4 in (CH3)2Si(CD3)2, k^lko = 1.35(5)], and (B) m- and /(-isomers of C p * W ( N O ) ( C H 2 C M e 3 ) ( C 6 H 4 M e ) and C p * W ( N O ) ( C H D C M e 3 ) ( C 6 D 4 M e ) [obtained from thermolysis of 2.4 in 1:1 toluene/toluene-dg, kn/ko = 1.04(5), unpublished] 195 xv i List of Tables Table 2.1. Selected IR Data for Complexes 2.2, 2.2-d4, 2.5, and 2.11 21 Table 2.2. Pertinent *H and 1 3 C N M R Data for Complexes 2.1-2.17 in C 6 D 6 Unless Otherwise Noted 23 Table 2.3. Selected Bond Distances and Angles for Complexes 2.3, 2A-d4, 2.9, 2.11', and 2.17 37 Table 2.4. Bond Distances and Angles Involving the M - C H 2 C M e 3 Fragments of Complexes C p * W ( N O ) ( C D 2 C M e 3 ) 2 (2A-d4), C p W ( = N A d ) ( C H 2 C M e 3 ) 2 , a and C p N b ( = N A r ) ( C H 2 C M e 3 ) 2 6 (The C p ' ligands are not drawn below for clarity) 38 Table 3.1. Selected *H, 1 3 C , and 3 1 P N M R Data for Cp 'W(NO)(=CHCMe 3 ) (L ) and Cp*W(NO)(=CHPh)(PMe 3 ) Complexes" 75 Table 3.2. Selected Bond Distances and Angles for 3.14, 3.15, and 3.17 88 Table 4.1. Selected *H and 1 3 C N M R Data for the 7 3 - A l l y l Moieties of Complexes 4.8-4.11 1 3 6 Table 4.2. Relative Kinetic Selectivities of [Cp*W(NO)(=CHCMe 3 ) ] and [Cp*W(NO)(=CHAr)] for Activation of A r y l and Benzyl C - H Bonds 1 4 5 Table 4.3. Kinetic Selectivities of [Cp*W(NO)(=CHCMe 3 ) ] for Activation of A r y l C - H Bonds in Monosubstituted Benzenes ^ Table 5.1. Isotopic Composition of Neopentane Derived from Thermal 173 Decompositions of 2A-d4 and 2A-d4 Table 5.2. Rate Constants for the Decompositions of 2.1 and 2.4 as a Function of Phosphine and Phosphine Concentration 179 Table 5.3. Rate Constants for the Thermal Decompositions of 2.1 and 2.4 in the Presence of Excess P M e 3 f l 181 xv i i Table 5.4. Activation Parameters for the Thermal Decompositions of 2.1 and 2.4 in T H F and Cyclohexane in the Presence of Excess P M e 3 a 182 Table 5.5. Rate Constants for Thermolyses of 2.1-</4 and 2.4-</4 in the Presence of Excess P M e 3 . 183 Table A l . Selected IR Spectral Data for Complexes 2.1-2.1 \a 193 Table A2. tert-Butyl Ions in the Mass Spectra of Neopentanes 195 Table A3. Crystallographic Data for Complexes 2.4-</4, 2.4, 2.9, 2.11', 2.17, and 3.1 .. 196 Table A4. Crystallographic Data for Complexes 3.4, 3.14, 3.15, 3.17, and 3.21a 197 Table A5. Crystallographic Data for Complexes 4.3, 4.4syn, 4.7, 4.9e„do, 4.11eXo, and 5.1 -. 198 Table A6. Fractional coordinates and equivalent isotropic displacement parameters for Cp*W(NO)(CD2CMe3)2 (2.4-</4) 199 Table A7. Fractional coordinates and equivalent isotropic displacement parameters for C p * W ( N O ) ( C H 2 C M e 3 ) 2 (2.4) 200 Table A8. Fractional coordinates and equivalent isotropic displacement parameters for C p * W ( N O ) ( C H 2 C M e 3 ) ( C H 2 S i M e 3 ) (2.9) 201 Table A9. Fractional coordinates and equivalent isotropic displacement parameters for C p * W ( N O ) ( C H 2 C M e 3 ) ( C 6 H 3 - 2 , 5 - M e 2 ) (2.11') 202 Table A10. Fractional coordinates and equivalent isotropic displacement parameters for C p * W ( N O ) ( C H 2 C M e 3 ) ( N M e 2 ) (2.17) 203 Table A l l . Fractional coordinates and equivalent isotropic displacement parameters for CpW(NO)(=CHCMe 3 ) (PMe 3 ) (3.1) 204 Table A12. Fractional coordinates and equivalent isotropic displacement parameters for Cp*W(NO)(=CHCMe 3 ) (PMe 3 ) (3.4) 205 Table A13. Fractional coordinates and equivalent isotropic displacement parameters for C p * W ( N O ) [ C H ( C M e 3 ) C H 2 ( C M e 3 ) C H ] (3.14) 206 xvi i i Table A14. Fractional coordinates and equivalent isotropic displacement parameters for C p * W ( N O ) ( C i 0 H i 8 ) (3.15) 207 Table A15. Fractional coordinates and equivalent isotropic displacement parameters for C p * W ( N O ) ( C i 2 H 1 8 ) (3.17) 208 Table A16. Fractional coordinates and equivalent isotropic displacement parameters for Cp*W(NO)(77 3 -C 1 0 H 1 8 ) (3.21a) 209 Table A17. Fractional coordinates and equivalent isotropic displacement parameters for Cp*W(NO)(cyclohexene)(PMe 3) (4.2) 210 Table A18. Fractional coordinates and equivalent isotropic displacement parameters for Cp*W(NO)(neohexene)(PMe 3) (4.4syn) 211 Table A19. Fractional coordinates and equivalent isotropic displacement parameters for Cp*W(NO)(cyclohexyl)(NMe 2 ) (4.7) 212 Table A20. Fractional coordinates and equivalent isotropic displacement parameters for Cp*W(NO)(77 3-C 7H 1 ,)(H) (4.9endo) 2 1 3 Table A21. Fractional coordinates and equivalent isotropic displacement parameters for Cp*W(NO)( /7 3 -C 8 H 1 3 ) (H) (4.11M„) 2 1 4 Table A22. Fractional coordinates and equivalent isotropic displacement parameters for Cp*W(NO)(CH2CMe 2-o-C6H 4)(PMe 3) (5.1) 2 1 5 Lis t of Abbreviations A angstrom, 10" 1 0 m A d adamantyl anal. analysis A P T attached proton test 1 3 C { ' H } N M R A r argon or aryl atm atmosphere(s) bp boiling point br broad B u butyl °C degree Celsius 1 3 C carbon-13 1 3 C { ' H } 13 proton-decoupled C cal calorie calcd calculated C 6 D 6 benzene-^ C D C I 3 chloroform-cfi CD2CI2 dichloromethane-c?2 C H 2 C M e 3 neopentyl C H 2 C M e 2 P h neophyl CI chemical ionization c m - 1 wavenumbers C O S Y Correlated 2D N M R Spectroscopy Cp if'-C5H5, cyclopentadienyl Cp* rf-CsMes, permethylated Cp C p ' Cp or Cp* c P r cyclopropyl 6 N M R chemical shift in ppm d doublet or day(s) D M F dimethylformamide A heat (thermolysis) EI electron impact Et ethyl E t 2 0 diethyl ether X X eu entropy unit (cal/K-mol) g gram(s) G C gas chromatography A G * free energy of activation h Planck's constant *H proton 2 H deuterium AH* enthalpy of activation H M Q C Heteronuclear Multiple Quantum Coherence H M B C Heteronuclear Multiple Bond Coherence H z Hertz IR infrared spectroscopy or spectrum J coupling constant V A B n-bond coupling constant between atoms A and B K degree Kelv in or equilibrium constant kcal kilocalorie kJ kilojoule(s) k0bS observed rate constant L U M O lowest unoccupied molecular orbital m multiplet m meta M molar or mega m/z mass-to-charge ratio M e methyl M e C N acetonitrile mg milligram(s) m L millilitre(s) min minute(s) mmol millimole mol mole M S mass spectrometry v infrared stretching frequency N M R nuclear magnetic resonance N O E nuclear Overhauser effect o ortho O R T E P Oak Ridge Thermal Ell ipsoid Program p para 3 1 P phosphorus-31 3 1 P { 1H} 1 H-decoupled phosphorus Ph phenyl Pr propyl ppm parts per mil l ion q quartet AS* entropy of activation s singlet t triplet T H F tetrahydrofuran U V - v i s ultraviolet-visible 1 8 3 W tungsten-183 X X l l Acknowledgments I would like to take this opportunity to thank those people without whose help this thesis would not have been possible. I first would like to thank Dave Burkey, Sean Lumb, Steve Sayers, Miguel Romero, J im Nieman, and the late Prof. Larry Weiler for numerous helpful discussions, and especially Dave Burkey for showing me how to be a better technical writer. M y sincerest thanks also go to the following members of my guidance committee: Prof. Mike Fryzuk for constructive criticisms of Chapters 2, 3, and 5, Prof. Derek Gates for proof-reading the Abstract and Chapter 1, and Prof. Chris Orvig for assuming the role of acting supervisor and for ensuring that this thesis be defended. M y appreciation must also go to Ms . Liane Darge, Mrs . Marietta Austria, M r . Peter Borda, M r . Marshall Lapawa, M s . Lina Madilao, and Dr. N ick Burlinson for technical assistance and/or training. I also would like to thank the following collaborators: Prof. Bob Bau of U S C and Dr. Sax A . Mason of Institut Laue-Langevin in Grenoble for the neutron structure reported in Chapter 2; Dr. Tom Koetzle of Brookhaven National Lab, Prof. Bob Bau and Dr. Manfred Bortz of U S C , and Drs. Art Schultz and Rebecca Mi l l e r of Argonne National Lab for an attempted but unsuccessful neutron diffraction study; and Drs. Victor Young, Jr. and Maren Pink of the University of Minnesota, Prof. Fred Einstein and Dr. Ray Batchelor of Simon Fraser University, and the late Dr. Steve Rettig of U B C for the X-ray structures reported in Chapters 2-5 . To Profs. Maurice Brookhart of the University of North Carolina and Pete Wolczanski of Cornell, I would like express my gratitude for a helpful discussion concerning the work in Chapters 2 and 5, respectively. M y deep appreciation also goes to Prof. E d Piers and the late Prof. Larry Weiler for advice, and to Prof. Tony Durst of the University of Ottawa and Dr. Prabat Arya of N R C for introducing me to chemical research and for continued guidance. Finally, I must acknowledge the constant love and support shown to me by my parents and family and the unwavering encouragement shown to me by my friends. I am especially indebted to my brothers, John and Chuck, for computer assistance and conversation, and my friends, Hoa, Judy, Ell ie , Gunjeet, and Kamla, for making me feel blessed. XX111 To my loving parents 1 CHAPTER 1 Background, Scope, and Format 1.1 Background . 1.2 Thesis Scope .2 1.3 Thesis Format 3 1.4 References and Notes ,4 1.1 Background A n ongoing goal of research in the Legzdins group has been the development of organotransition-metal nitrosyl complexes as specific and selective reagents for organic and organometallic transformations of practical significance.1 In particular, major efforts have been directed toward the synthesis, characterization, and reactivity studies of the molybdenum and tungsten complexes Cp 'M(NO)R 2 (Cp' = Cp, Cp* and R = alkyl, aryl).2 These complexes are formally d 4 , 16-electron species. They are monomeric and possess a three-legged piano-stool structure in which the M - N O linkage is essentially linear. In this configuration, the metal d x y and d x z orbitals are significantly stabilized by backbonding to the nitrosyl %* orbitals, while the d x 2 - y 2 orbital (= L U M O ) remains high in energy and available for bonding to additional ligands.3 In general, the molybdenum and tungsten Cp'M(NO)R2 complexes exhibit reactivity patterns that are very similar. Although most are relatively air-stable and thermally stable as solids, their solution chemistry is quite diverse. For example, in addition to forming 1:1 metal-centered adducts with small Lewis bases and isonitrosyl adducts with Lewis acids,4 they both react readily with oxygen 4' 5 to form the bis(oxo) complexes, Cp'M(=0)2R, and with sulfur,4 isocyanides,4 N O , 4 C O , 6 and heterocumulenes to form products arising from insertion into the 7 8 9 M - C o-bonds. More interestingly, however, are their reactions with water and hydrogen, ' 2 which vary depending on the identity of the metal. For example, while exposure of Cp*Mo(NO)(CH 2 SiMe 3 ) 2 to H 2 leads to the formation of [Cp'Mo(NO)(CH 2 SiMe 3 )](u-N)[Cp'Mo(0)(CH 2 SiMe 3 ) ] , 8 the same reaction with the tungsten analogue produces the hydride complex, Cp*W(NO)(CH 2 SiMe 3 )(H), as a short-lived species.9 The objective of this thesis is to investigate further the chemistry of the tungsten bis(alkyl) complexes. In particular, research has been focused on the structures and thermal reactivity of these compounds. 1.2 Thesis Outline This thesis is divided into five chapters, each of which is self-contained, and an appendix section. Chapter 2 reports a systematic spectroscopic and structural investigation of a series of Cp'W(NO)(R)(R') complexes, where R = alkyl, and R ' = alkyl, aryl, chloride, alkoxide, or amide. In contrast to previous reports,2 this study shows that, in the absence of secondary interactions such as lone pair or Tt-electron donation, these complexes are in fact electronically saturated species due to the presence of an a-agostic C - H interaction. These interactions exist in both the solid and solution states and give rise to a number of unusual spectroscopic and structural features that have allowed the strength of the a-C-H—W interaction to be established as a function of Cp' , R, and R' . Thermolysis of the neopentyl complexes, Cp 'W(NO)(CH 2 CMe 3 ) 2 [Cp' = Cp, Cp*] and Cp*W(NO)(CH 2 CMe 3 )(CH 2 Ph), leads to extrusion of neopentane and formation of [Cp'W(NO)(=CHCMe 3)] and [Cp*W(NO)(=CHPh)], respectively, as reactive intermediates. The reactivity of these alkylidene intermediates with a range of trapping agents forms the basis of Chapter 3. The reactions of [Cp'W(NO)(=CHCMe 3)] and [Cp*W(NO)(=CHPh)] with L , where L = tertiary phosphine, pyridine, or tertiary amine, afford Cp'W(NO)(=CHCMe 3 )(L) and Cp*W(NO)(=CHPh)(L) complexes. Both [Cp*W(NO)(=CHCMe 3)] and its P M e 3 adduct undergo addition reactions with R E H substrates to form complexes of the type, Cp*W(NO)(CH 2 CMe 3 )(ER) [ER = N R 2 , OR, OC(0)R]. Alkenes also react readily with [Cp*W(NO)(=CHCMe 3)]. These reactions, however, lead to complicated product mixtures resulting from [2+2] cycloaddition and/or C - H activation depending on the alkene structure. 3 The products for both types of reactions are discussed, and mechanisms are proposed for their formation. Chapter 4 is an extension of Chapter 3 and is concerned with the alkane and arene C - H activation reactions of [Cp*W(NO)(=CHCMe 3)], and to a lesser extent the Cp analogue. The reactions of [Cp*W(NO)(=CHCMe3)] with alkanes proceed with moderate steric selectivity and are typically followed by rapid rearrangement transformations that result in overall dehydrogenation of the alkane. The reactions with arenes are similarly sensitive to steric effects and involve competitive aryl vs benzyl C - H activation. The benzyl-activated products are thermally unstable at the temperature at which they are formed, decomposing to give neopentane and benzylidene complexes that also readily cleave aryl and benzyl C - H bonds. Finally, Chapter 5 reports a detailed mechanistic investigation of the chemistry described in Chapters 3 and 4. Specifically, the kinetics of the thermal decompositions of Cp'W(NO)(CH2CMe3)2 in the presence of phosphines are presented, along with the results of an isotope effect and an isotope labeling study, which not only provide further support for the intermediacy of [Cp'W(NO)(=CHCMe3)] but which also provide strong evidence for the involvement of a - C - H complexes in the formation and C - H activation reactions of these coordinatively unsaturated alkylidene species. The thermal decompositions of the mixed bis(alkyl) complexes, Cp*W(NO)(CH 2 CMe 3 )(R), where R = C H 2 C M e 2 E t , C H 2 C M e 2 P h , CH 2SiMe3, C H 3 , and Ph, are also presented. These studies show that the pathway by which these compounds decompose is governed primarily by the strengths of the M - C bond broken and formed, and that the presence of oc-agostic interactions in the ground state has little, i f any, effect on the mechanism of decomposition. 1.3 Thesis Format This thesis is formatted in a manner similar to that described in the Ph. D. thesis of former group member, J. E. Veltheer,1 0 with Chapters 2 through 5 having five major sections. For a given Chapter X , the sections are: X . l Introduction, X.2 Experimental Section, X.3 Results and 4 Discussion, X.4 Epilogue or Conclusions, and X.5 References and Notes. Subsections of these categories are numbered using legal outlining procedures, e.g. X . l . l , X.l.1.1, X.1.2, X.l.2.1, etc. In each chapter, all new compounds prepared are catalogued numerically. For example, those in Chapter 2 are referred to as 2.1,2.2 , 2.3, and so on. Schemes, figures, and equations are similarly sequenced. The standard experimental methodologies outlined in Section 2.2.1 apply to the entire thesis. A n Appendix is also included which contains tables of crystallographic data, etc. 1.4 References and Notes (1) For a recent comprehensive discussion of transition-metal nitrosyl chemistry, see: Richter-Addo, G. G.; Legzdins, P. Metal Nitrosyls; Oxford University Press: New York, 1992. (2) Legzdins, P.; Veltheer, J. E. Acc. Chem. Res. 1993, 26, 41, and references therein. (3) (a) Legzdins, P.; Rettig, S. J.; Sanchez, L. ; Bursten, B. E.; Gatter, M . G. J. Am. Chem. Soc. 1985,107, 1411. (b) Bursten, B . E.; Cayton, R. H . Organometallics 1987, 6, 2004. (4) Legzdins, P.; Sanchez, L. Organometallics 1988, 7, 2394. (5) Legzdins, P.; Phillips, E. C ; Sanchez, L. Organometallics 1989, 8, 940. (6) Debad, J. D.; Legzdins, P.; Batchelor, R. J.; Einstein, F. W. B. Organometallics 1993, 12, 2094. (7) (a) Legzdins, P.; Rettig, S. J.; Ross, K . J.; Veltheer, J. E. J. Am. Chem. Soc. 1991,113, 4361. (b) Legzdins, P.; Lundmark, P. J.; Phillips, E. C ; Rettig, S. J.; Veltheer, J. E. Organometallics 1992, 11, 2991. (8) Debad, J. D.; Legzdins, P.; Reina, R.; Young, M . A . ; Batchelor, R. J.; Einstein, F. W. B . Organometallics 1994, 13, 4315. (a) Legzdins, P.; Martin, J. T.; Einstein, F. W. B. ; Jones, R. H . Organometallics 1987, 6, 1826. (b) Debad, J. D.; Legzdins, P.; Batchelor, R. J.; Einstein, F. W. B . Organometallics 1992,13, 6. J. E. Veltheer, Ph. D. Thesis, 1993, University of British Columbia. 6 C H A P T E R 2 Spectroscopic and Structural Studies of a-Agostic Bonding in Cp'W(NO)(R)(R') Complexes 2.1 Introduction 6 2.2 Experimental Section 9 2.3 Results and Discussion 18 2.4 Epilogue 41 2.5 Addendum: Neutron Structure of Cp*W(NO)(CH 2 CMe 3 )2 (2.4) and Reinterpretation of Spectroscopic and Structural Data 44 2.6 References and Notes 48 2.1 Introduction The interaction between a ligand C - H bond and an electronically unsaturated metal center is called an agostic interaction.1 First observed in the late 1960s by Ibers,2 Shaw,3 and Trofimenko,4 such an interaction has since become increasingly widespread, existing not only in transition-metal complexes but also in complexes of the lanthanide and actinide series. In all cases the C - H moiety acts as a weak two-electron donor to the metal, forming a three-center, two-electron bond whose strength and geometry depend upon the energy of the vacant metal orbital and the steric and conformational effects of the molecule. In general, the energy of an agostic interaction is 10-20 kcal mol" 1, 5 ' 6 while its geometry varies from a linear or open arrangement (I) to a canted side-on or closed structure (II) in which the H atom is closer to the metal. The bonding, on the other hand, is best described in terms of two synergistic components: (1) a rj-type interaction that involves C - H (rj)-to-metal direct donation, and (2) a 71-type interaction that involves metal-to-C-H (a*) back donation. 7 C - H - M H-. / ' .M (I) (II) It has been proposed that the driving force for the formation of an agostic C - H - - M interaction is to reduce the electron deficiency of the metal and to stabilize what would otherwise be an unsaturated species.7 While this interaction is often benign, its strength can be such as to have a marked effect on the chemical properties of the coordinated C - H bond and the reactivity of the molecule. For example, it is now well established that agostic C - H binding can significantly weaken the C - H bond and increase its Brensted acidity,8 thereby rendering it susceptible to a wide range of C - H activation reactions. 8 d ' 9 ' 1 0 , 1 1 In other instances, the coordinated C - H bond never becomes broken; instead, it plays an important role in lowering the activation barrier and in influencing the stereochemical outcome of chemical transformations such as Ziegler-Natta-type polymerization of alkenes 1 2 ' 1 3 and alkynes.1 4 In general, agostic C - H - - M linkages may be detected in the solid state by X-ray and neutron diffraction studies and in solution by IR and N M R spectroscopic methods. Keys to the structural identification of these links are short M — H and M—C distances, a lengthening of the coordinated C - H bond, and (in certain cases) a geometrical distortion of the ligand that contains this group. Spectroscopically, the anomalous chemical shifts of the agostic H and C atoms and the reduced values of the 1 JQH coupling constant evident in the N M R spectra and the V C H stretching frequency of the coordinated C - H bond in the IR spectra are most diagnostic of C - H binding. The latter two parameters, in particular, have also proven important in determining the degree of agosticity of a C - H - - M interaction, especially when comparisons are made for members of a related series of compounds. Typically, these 1JQH and V C H values fall in the ranges 50-100 Hz and 2300-2700 cm"1, respectively, compared to 120-130 Hz and 2900-3100 cm"1 for the corresponding unperturbed C - H groups.1 8 This chapter describes the results of a spectroscopic and X-ray structural study of a series of d 2 tungsten alkyl complexes of the type Cp'W(NO)(R)(R'), where Cp ' represents a cyclopentadienyl or pentamethylcyclopentadienyl ring, R is an alkyl ligand that contains a-H atoms, and R ' is either an alkyl group, an aryl moiety, or a potential multi-electron donor ligand. The principal objectives of this work are to determine the solution and solid-state structures of these compounds and to rationalize some of their chemical behavior which typically involves the cleavage of an oc-C-H bond. As depicted in eqs 2.1-2.3, this may proceed by deprotonation15 as reported recently by Legzdins and Sayers, or it may involve H -atom abstraction16 or exchange as described in the ensuing chapters of this thesis. B © M ® Cp*W(NO)(CH 2 SiMe 3 ) 2 »• [Cp*W(NO)(=CHSiMe 3 )(CH 2 SiMe 3 )]M (2.1) - B H L Cp 'W(NO)(CH 2 CMe 3 ) 2 • Cp'W(NO)(=CHCMe 3 )L + C M e 4 (2.2) A A1203 Cp*W(NO)(CHDCMe 3 ) (C 6 D 5 ) >• Cp*W(NO)(CH 2 CMe 3 ) (C 6 D 5 ) (2.3) Hexane In this chapter it wil l be demonstrated that Cp'W(NO)(R)(R') complexes, which lack K or lone-pair donation, are not 16-electron complexes as had previously been inferred,17 but are in fact more electronically rich due to the presence of a i* least one a-agostic C - H interaction. In particular, it wi l l be established that these interactions not only exist in both the solution and solid states but also vary in strength and multiplicity depending on the nature of Cp ' , R, and R' . For example, when R and R ' are both bulky alkyl ligands containing a-H atoms, the molecule assumes a structure in which two a - C - H units from different alkyl ligands form three-center, two-electron bonds with tungsten. Examples of multiple agostic C - H bonds in a mononuclear complex are rare, with the majority occurring in transition-metal complexes with very low. electron count 1 8 ' 1 9 or 9 9 0 91 9 9 complexes of the rare earth elements. ' ' Indeed, it was not until very recently that three 16-electron transition metal complexes containing multiple agostic bonds were discovered and reported. Interestingly, these complexes, which include the bis(methyl) complex, ('Pr2-2,6-C 6 H 3 N = ) 2 M o M e 2 , 2 3 and the bis(neopentyl) compounds, CpW(=CAd)(CH 2 CMe 3 ) 2 2 4 (Ad = 1-adamantyl) and CpNb(=NAr)(CH 2 CMe 3 ) 2 (Ar = C 6 H 3 -2,6- 'Pr 2 ) , 2 5 also possess cc-agostic structures. The first complex has been proposed to contain four a-agostic interactions based on neutron-diffraction data, while the second and third complexes have been shown to exhibit two such interactions. A comparison of the structure of these latter complexes with that of the Cp'W(NO)(R)(R') bis(alkyl) systems wil l therefore also be presented. 2.2 Experimental Section 2.2.1 Methods The experimental methodologies described in this chapter apply to the entire thesis. A l l reactions and subsequent manipulations of air- and/or water-sensitive compounds were performed under anaerobic and anhydrous conditions under an atmosphere of pre-purified dinitrogen or argon. Purification of these gases was achieved by sequential passage over MnO on vermiculite and activated 4-A molecular sieves. Conventional glovebox and Schlenk techniques were utilized throughout. The specific glovebox used in this work was an Innovative Technology dual-station glovebox. IR spectra were recorded at room temperature as solutions in NaCl cells or as Nujol mulls between KC1 plates on an ATI Mattson Genesis Series FT-IR spectrophotometer using WinFIRST 2.0 software. N M R spectra were obtained at room temperature unless otherwise noted on either a Varian X L (300 M H z , 'H), Bruker W H (400 M H z , 'H), or Bruker A M X (500 M H z , 'H) spectrometer. Where necessary, selective homonuclear decoupling, ] H N O E difference, 1 3 C(APT) , gated l 3 C { ' H } , ' H - ' H C O S Y , ' H - 1 3 C H M Q C , and ' H - 1 3 C H M B C experiments were carried out to correlate and assign ' H and 1 3 C signals. ' H and 1 3 C chemical shifts are recorded in ppm, relative to the residual proton or natural-abundance carbon signal(s) 10 of the solvent employed. 2 H{'H} N M R signals are referenced to C 6 H 5 D (8 7.15). A l l coupling constants are reported in Hz. Mrs. M . T. Austria and Ms. L. K . Darge of the U B C N M R laboratory assisted in obtaining some N M R data. Low-resolution mass spectra (EI, 70 eV, probe temperature 150 °C) were recorded by Mr. M . Lapawa and Ms. L. Madilao of the U B C mass spectrometry facility using a Kratos MS-50 spectrometer. Elemental analyses were performed by Mr. P. Borda of this Department. Reported yields are not optimized unless specified. 2.2.2 Reagents Hexanes, pentane, THF, and E t 2 0 were distilled from Na/benzophenone ketyl under dinitrogen. Deuterated solvents were purchased from Cambridge Isotope Laboratories (all >99.9 atom % D) and then degassed and vacuum-transferred from Na/benzophenone ketyl (C6D6), Na (toluene-Jg), or activated 4-A molecular sieves (CD 2C1 2). Celite and neutral alumina I were oven-dried (>130 °C) and cooled under vacuum prior to use. K H (Alfa, 50% in oil) was washed with hexanes and dried in vacuo prior to use. Dialkylmagnesium reagents R2Mg-(dioxane) (R = Me, C H 2 C M e 3 , C H 2 S i M e 3 , CH 2 CMe 2 Ph) were prepared by modified literature procedures26 using 2.2 equiv of dioxane; the titre of each was determined by back-titration with 0.10 N aqueous HC1 using phenolphthalein as indicator. With the exception of Cp*W(NO)(CH 2 CMe 3 ) 2 (2.4), the known complexes 2.1-2.6 2 7 ' 2 8 and 2.8-2.14 2 9 ' 3 0 were prepared according to published procedures. The complexes Cp*W(NO)(CH 2 CMe 3 )(OMe) (2.16), Cp*W(NO)(CH 2 CMe 3 )(NMe 2 ) (2.17), and Cp*W(NO)(CH 2 CMe 3 )(C 6 H 3 -2,5-Me 2 ) (2.11') were prepared as described in Sections 3.2.2 and 4.2.2. For brevity, the complete *H N M R data of the known complexes are excluded since they do not differ significantly from those reported in the literature and since they are not necessary for the analysis. The a-proton and cc-carbon signals that are listed in Table 2.2 are also omitted to avoid redundancy. 11 2.2.3.Preparation of E t M e 2 C C H 2 B r To a 1-L, three-necked round-bottomed flask containing L1AIH4 (6.30 g, 166 mmol) and connected to a condenser and a 100-mL addition funnel was added THF (400 mL) by cannula. The stirred grey suspension was cooled to -78 °C and was then treated dropwise with a 50-mL THF solution of 2,2-dimethylbutyric acid (18.9 g, 163 mmol) via addition funnel. Once the addition was complete, the resulting mixture was stirred at -78 °C for 45 min and at room temperature for ca. 1.5 d, whereupon no noticeable change occurred. In air, the reaction mixture was cooled to 0 °C and quenched successively with 3N aqueous NaOH (25 mL) and water (200 mL). The white precipitate formed was removed by filtration, and the filtrate was extracted with E t 2 0 (200 mL). The aqueous layer was re-extracted with E t 2 0 (3 x 150 mL). The combined organic layers were reduced in volume on a rotary evaporator, washed with water (2 x 30 mL), dried over MgS04, filtered, and concentrated in vacuo. The crude mixture was distilled in air at 140-150 °C to obtain the desired 2,2-dimethylbutanol as a viscous, pale yellow liquid (16.0 g, 98 % yield). To a portion of this alcohol (13.0 g, 0.127 mmol) was added dry D M F (75 mL). The resulting solution was cooled to 0 °C, and then the sequential slow addition of « - B u 3 P (50.0 mL, 200 mmol) by cannula, and Br 2 (7.0 mL, 200 mmol) by disposable syringe were effected. After being stirred for 3.5 h in the dark, the bright orange solution was distilled in air at 140-205 °C. The pale yellow distillate was extracted with water. The bottom organic layer was re-extracted with water ( 2 x 1 0 mL), dried over M g S 0 4 , and filtered. The filtrate was next distilled from P 2 0 5 under A r (1 atm) at -150-200 °C to obtain the desired alkyl bromide as a colorless liquid (6.50 g, 31% yield). Conversion of this alkyl bromide to the corresponding dialkyl M g reagent was carried out in a manner analogous to that described for the preparation of (Me 3CCH 2) 2Mg-(dioxane). 2.2.4 Preparation of P h M e 2 C C D 2 B r To a THF (350 mL) mixture of K H (6.85 g, 171 mmol, 3 equiv) at -10 °C was added dropwise P h C H 2 C 0 2 M e (8.0 mL, 56.8 mmol) by cannula. The resulting clear, colorless solution over white solids was stirred at ambient temperatures for 2 h, after which time it was 12 re-cooled to -10 °C. Freshly distilled M e l (28.0 mL, 450 mmol) was added slowly via dropping funnel over a period of 1 h, causing the reaction solution to become pale yellow. This mixture was stirred at room temperature for another 2 d, whereupon the solution became clear and colorless. Working in air, the reaction mixture was filtered through a plug of Celite (5 x 10 cm). The filtrate was concentrated in vacuo, and then distilled under reduced pressure to obtain pure PhCMe2C02Me as a clear, colorless liquid (4.22 g, 80 % yield). Reduction of this ester (4.22 g, 23.7 mmol) with L i A l D 4 (0.498 g, 11.9 mmol) in E t 2 0 (200 mL) based on the procedure for the preparation of Me3CCD20H led to the isolation of PhCMe2CD 2 OH as a slightly viscous colorless liquid (3.60 g, 99% yield) after simple distillation under reduced pressure. This alcohol (3.60 g, 23.7 mmol) was then converted to the bromo derivative by treatment with n-Bu 3P (8.85 mL, 35.6 mmol, 1.5 equiv) and excess B r 2 (1.71 mL, 33.2 mmol, 1.4 equiv) in D M F as described above. The bromide product was purified by flash chromatography on silica gel (230-400 mesh, 3 x 1 0 cm) using pure hexanes (i?f = 0.55) as the eluting solution and isolated as a colorless liquid (2.46 g, 48% yield). The bromide was then converted into the desired magnesium reagent in the usual manner. 2.2.5 1 3 C N M R Data of C p W ( N O ) ( C H 2 C M e 3 ) 2 (2.1) Gated 1 3 C{ 'H} N M R (75 M H z , C 6 D 6 ) 5 34.3 (q, [JCU = 124, CMe 3 ) , 39.4 (s, CMe 3 ) , 101.2 (d, ' J C H = 178, C 5 H 5 ) . 2.2.6 1 3 C N M R Data of CpW(NO)(CH 2 CMe 2 Ph) 2 (2.2) Gated 1 3 C{ 'H} N M R (75 M H z , C 6 D 6 ) 8 32.7 (q, ' J C H = 125, CMeMe), 33.5 (q, ' j C H = 125, CMeMe), 46.1 (s, CMe 2 ) , 101.4 (d, [JCH = 178, C 5 H 5 ) , 125.6 (d, ' J C H = 159, Ar CH), 126.0 (d, lJCH = 156, Ar CH), 128.3 (d, %H = 158, Ar CH), 152.8 (s, C i p s o ) . 13 2.2.7 MS and 2 H NMR Data of CpW(NO)(CD2CMe2Ph)2 (2.2-d4) M S (LREI, m/z) 549 [ M + , 1 8 4 W ] . 2 H{'H} (77 M H z , C 6 H 6 ) 5-1.05 (br s, 2D, W C D 2 ) , 3.55 (brs, 2D, W C D 2 ) . 2.2.8 1 3 C NMR Data of CpW(NO)(CH2SiMe3)2 (2.3) Gated 1 3 C{ 'H} N M R (75 M H z , C 6 D 6 ) 5 2.8 (q, lJCH = 118, SiMe 3 ) , 101.5 (d, [JCH = 178, C 5 H 5 ) . 2.2.9 Improved Preparation of Cp*W(NO)(CH2CMe3)2 (2.4) To a Schlenk flask containing a mixture of Cp*W(NO)Cl 2 (909 mg, 2.16 mmol) and (Me 3CCH 2) 2Mg-(dioxane) (277 mg, 2.16 mmol) at -196 °C was added THF (-40 mL) by vacuum transfer. The resulting brown mixture was thawed and stirred at -10 °C for 20 min, during which time it became a dark purple solution. THF was then removed in vacuo, and the flask was returned to the glovebox, charged with additional (Me 3CCH 2) 2Mg-(dioxane) (277 mg, 2.16 mmol), and reconnected to a high vacuum line. E t 2 0 was then added by vacuum transfer at -196 °C. The resulting mixture was thawed and stirred at room temperature for 30 min, during which time a red solution containing a pale yellow precipitate formed. Next, the solvent was removed in vacuo and the residue was extracted with 5:1 hexanes/Et20 (30 mL). The extracts were filtered through alumina 1(3 x 2 cm) supported on a medium-porosity frit. The alumina column was washed with additional 5:1 hexanes/Et20 until the filtrate was colorless. Diminishing the volume of the combined filtrates under reduced pressure, followed by cooling to -30 °C for 2 d afforded 2.4 (684 mg, 65% yield in 3 crops) as dark wine-red needles. Anal. Calcd. for C 2 0 H 3 7 N O W : C, 48.89; H , 7.59; N , 2.85. Found: C, 48.63; H , 7.73; N , 2.77. M S (LREI, m/z) 491 [ M + , 1 8 4 W ] . ' H N M R (500 M H z , C 6 D 6 ) 8 1.32 (s, 18H, CMe 3 ) , 14 1.54 (s, 15H, C 5 Me 5 ) . Gated 1 3 C{ 'H} N M R (75 M H z , C 6 D 6 ) 8 9.9 (q, LJCH = 128, C5Me5), 34.7 (q, 1JCB = 124, CMe 3 ) , 39.2 (s, CMe 3 ) , 110.2 (s, C 5 Me 5 ) . 2.2.10 Preparation of Cp*W(NO)(CH2CMe3)(CD2CMe3) (2.4-rf2) This complex was prepared in the same manner as described for the synthesis of 2.4 but using (Me 3CCD 2)Mg-(dioxane) for the second alkylation step. The only exception was that the hexanes/Et20 extracts were filtered through a pad of Celite rather than alumina I. M S (LREI, m/z) 493 [ M + , 1 8 4 W ] . IR (cm"1, Nujol) no reduced v C A - H observed. Gated 1 3 C{ 'H} N M R spectra obtained at -80 and +20 °C were identical to that of 2.4 at +20 °C. 2.2.11 1 3C NMR Data of Cp*W(NO)(CH2CMe2Ph)2 (2.5) Gated 1 3 C{ 'H} N M R (75 M H z , C 6 D 6 ) 8 9.4 (q, LJCH = 127, C5Me5), 32.4 (q, LJCH = 125, CMeMe), 33.8 (q, ' / C H = 125, CMeMe), 45.9 (s, CMe 2 ) , 109.7 (s, C 5 Me 5 ) , 125.5 (d, ' JCH = 159, Ar CH), 126.4 (d, ' / C H = 157, Ar CH), 128.4 (d, ' j C H = 158, Ar CH), 153.2 (s, C i p s o ) . 2.2.12 1 3C Data of Cp*W(NO)(CH2SiMe3)2 (2.6) Gated 1 3 C{ 'H} N M R (75 M H z , C 6 D 6 ) 5 2.9 (q, [JCLI =117, SiMe 3 ) , 9.9 (q, LJCn = 127, C 5 Me 5 ) , 110.2 (s, C 5 Me 5 ) . 2.2.13 Preparation of Cp*W(NO)(CH2CMe3)(CH2CMe2Et) (2.7) To a mixture of Cp*W(NO)(CH 2 CMe 3 )Cl (375 mg, 0.823 mmol) and (EtMe 2CCH 2) 2Mg-(dioxane) (128 mg, 0.823 mmol) at-196 °C was added E t 2 0 (10 mL) by vacuum transfer. The resulting contents were thawed and stirred at ambient temperature for 10 min, during which time the mixture changed from purple to deep red. The solvent was then removed in vacuo. The residue was triturated with pentane ( 3 ^ 5 mL) and filtered through 15 Celite. The dark red filtrate was reduced in volume at room temperature and then stored at -30 °C overnight to obtain 2.7 as shiny rust-colored crystals (279 mg, 67% yield). Anal. Calcd. for C 2 1 H 3 9 N O W : C, 49.91; H , 7.78; N , 2.77. Found: C, 49.63; H , 7.73; N , 2.77. IR (cm"1) v N O 1576 (s). MS (LREI, m/z) 505 [ M + , l 8 4 W ] . ' H N M R (500 M H z , C 6 D 6 ) 5 0.97 (t, V H H = 7.5, 3H, CH 2 Me), 1.23 (s, 3H, CMeMe), 1.38 (s, 9H, CMe 3 ) , 1.42 (s, 3H, CMeMe), 1.52 (s, 15H, C 5 Me 5 ) , 1.63 (m, 2H, Gr7 2Me). 1 3 C{ 'H} N M R (75 M H z , C 6 D 6 ) 5 9.78 (q, '7CH = 125, CH 2 Me), 9.83 (q, lJCH = 127, C 5 Me 5 ) , 30.9 (q, lJCH ~ 125, CMeMe), 31.1 (q, ' J C H = 124, CMeMe), 34.4 (q, 'J C H = 124, CMe 3 ) , 39.2 (s, CMe 3 ) , 39.5 (t, ' / C H = 124, CH 2 Me), 41.2 (s, CMe 2 Et), 109.6 (s, C 5 Me 5 ) . 2.2.14 1 3 C N M R Data of Cp*W(NO)(CH 2CMe 3)(CH 2CMe 2Ph) (2.8) Gated ' ^ { ' H } N M R (75 M H z , C 6 D 6 ) 8 9.7 (q, 'J C H = 127, C 5 Me 5 ) , 32.6 (q, lJCH = 125, CMeMe), 34.0 (q, [JCH = 124, CMeMe), 34.2 (q, lJCn = 124, CMe 3 ) , 39.6 (s, CMe 3 ) , 45.4 (s, CMe 2 ) , 109.7 (s, C 5 Me 5 ) , 125.5 (d, ' y C H = 159, Ar CH), 126.3 (d, lJCH = 157, Ar CH), 128.4 (d, 'J C H = 158, Ar CH), 153.6 (s, C i p s o ) . 2.2.15 1 3 C and *H NOE Difference N M R Data of Cp*W(NO)(CH 2CMe 3)(CH 2SiMe 2) (2.9) Gated 1 3 C { ] H } N M R (75 M H z , C 6 D 6 ) 8 2.8 (q, 'JCH = 117, SiMe 3 ) , 9.8 (q, lJCH = 127, C 5 Me 5 ) , 34.2 (q, ]Jm = 124, CMe 3 ) , 39.8 (s, CMe 3 ) , 109.8 (s, C 5 Me 5 ) . ' H NOEDS (400 M H z , C 6 D 6 ) irrad. at -2.12 ppm, NOEs at 8 -1.29 and 3.28; irrad. at -1.29 ppm, NOEs at 8 -2.12 and 1.07; irrad. at 1.07 ppm, N O E at 8-1.29; irrad. at 3.28 ppm, N O E at 8 -2.12. 2.2.16 1 3 C N M R Data of Cp*W(NO)(CH 2CMe 3)(CH 3) (2.10) Gated 1 3 C{ 'H} N M R (75 M H z , C 6 D 6 ) 8 9.6 (q, lJm = 127, C 5 Me 5 ) , 33.9 (q, 'JCH = 124, CMe 3 ) , 40.3 (s, CMe 3 ) , 109.5 (s, C 5 Me 5 ) . 16 2.2.17 U C N M R Data of Cp*W(NO)(CH 2CMe 3)(Ph) (2.11) Gated 1 3 C{ 'H} N M R (75 M H z , CD 2 C1 2 ) 5 10.2 (q, V C H = 127, C5Me5), 33.5 (q, V C H = 124, CMe 3 ) , 41.4 (s, CMe 3 ) , 111.4 (s, C 5 Me 5 ) , 127.2 (d, V C H = 158, A r CH), 128.3 (d, V C H = 159, A r C H ) , 136.9 (d, V C H = 155, Ar CH), 181.0 (s, W C i p s o ) . 2.2.18 1 3 C N M R Data of Cp*W(NO)(CH2CMe2Ph)(CH2SiMe3) (2.12) Gated 1 3 C{ 'H} (75 M H z , C 6 D 6 ) 5 2.8 (q, V C H = 118, SiMe 3 ) , 9.6 (q, V C H = 127, C5Me5), 32.8 (q, V C H = 125, CMeMe), 33.0 (q, V C H = 125, CMeMe), 46.1 (s, CMeMe), 110.0 (s, C 5 Me 5 ) , 125.7 (d, V C H = 159, Ar CH), 126.3 (d, V C H = 156 , Ar CH), 128.3 (d, V C H = 156, A r C H ) , 153.3 (s, C i p s o ) . 2.2.19 1 3 C N M R Data of Cp*W(NO)(CH 2CMe 2Ph)(CH 3) (2.13) Gated 1 3 C{ 'H} N M R (75 M H z , C 6 D 6 ) 5 9.4 (q, V C H = 127, C5Me5), 31.7 (q, V C H = 121, CMeMe), 33.1 (q, V C H = 121, CMeMe), 46.7 (s, CMe 2 ) , 109.6 (s, C 5 Me 5 ) , 125.5 (d, V C H = 159, Ar CH), 126.4 (d, V C H = 158, Ar CH), 128.2 (d, V C H = 157, Ar CH), 152.7 (s, C i p s o ) . 2.2.20 1 3 C N M R Data of Cp*W(NO)(CH 2CMe 3)(Cl) (2.14) Gated ' ^ { ' H } N M R (75 M H z , C 6 D 6 ) 8 9.8 (q, V C H = 127, C 5 Me 5 ) , 33.7 (q, V C H = 124, CMe 3 ) , 39.1 (s, CMe 3 ) , 112.6 (s, C 5 Me 5 ) . 2.2.21 Preparation of Cp*W(NO)(CH 2CMe 3)(CH 2Ph) (2.15) To a mixture of Cp*W(NO)(CH 2 CMe 3 )Cl (162 mg, 0.356 mmol) and (PhCH2)2Mg-(dioxane) (60 mg, 0.356 mmol) in a Schlenk rube at -196 °C was added E t 2 0 (20 mL) by vacuum transfer. The resulting contents were thawed and stirred at ambient 17 temperature for ca. 15 min, during which time the mixture changed from purple to orange. The solvent was then removed in vacuo. The residue was redissolved in E t 2 0 and filtered through a small plug of alumina I. The alumina column was washed with an additional 2 mL of E t 2 0 . The combined dark orange filtrates were reduced in volume at room temperature, an operation that resulted in the formation of a small amount of orange crystals. Toluene (1 mL) was then added, and the resulting homogeneous solution was stored at -30 °C overnight to induce the precipitation of 2.15 as orange crystals (136 mg, 75% yield). Anal. Calcd. for C21H39NOW: C, 51.67; H , 6.51; N , 2.74. Found: C, 51.79; H , 6.59; N , 2.65. IR (cm - 1) v N O 1539 (s). M S (LREI, mlz) 511 [ M + , 1 8 4 W ] . ' H N M R (300 M H z , C 6 D 6 ) 5 1.15 (s, 9H, CMe 3 ) , 1.55 (s, 15H, C 5 Me 5 ) , 6.88 (m, 3H, H„, H m ) , 7.28 (pseudo t, V H H = 6.9, 1H, Hp). Gated 1 3 C{ 'H} N M R (75 M H z , CD 2 C1 2 ) At 20 °C: 5 10.4 (q, ' j C H = 127, C5Me5), 33.9 (q, ' y C H = 123, CMe3), 38.8 (s, CMe 3 ) , 109.5 (s, C 5 Me 5 ) , 127.0 (d, 1JCH = 158, Cp), 129.3 (d, 'JCH = 158, C0 or C m ) , 131.6 (d, LJCH = 158, C 0 or Cm), 133.2 (C i p s o ) . At -80 °C: 5 10.2 (q, % H = 127, C5Me5), 33.2 (q, ' J C H = 122, CMe 3 ) , 36.4 (s, CMe 3 ) , 107.8 (s, C 5 Me 5 ) , 117.2 (s, Ci p S o) , 129.2 (d, ' J C H = 159, C G or C m ) , 129.7 (d, LJCH = 160, Cp), 133.1 (d, LJCH = 159, C 0 or C m ) . 2.2.22 Neutron Structure of C p * W ( N O ) ( C H 2 C M e 3 ) 2 (2.4) [See also Addendum] Single crystals of 2.4 were grown from Et20/pentane solution at -30 °C. Neutron diffraction data were collected at the Institut Laue-Langevin in Grenoble, France on instrument D19 using 4° x 64° area detector. Under an inert atmosphere, a red-brown crystal with approximate dimensions 4.8 x l .8 x 0.48 mm was mounted inside a thin-walled quartz tube, jammed between two plugs of quartz wool to prevent slippage. It was then placed inside a Displex cryostat and slowly cooled. At -100 K , the crystal appeared to undergo a reversible phase transition, and consequently the temperature was raised to 120 K for the subsequent data collection. Data were collected with neutrons of wavelengths 1.5365(2) A and integrated in three dimensions: a total of 4644 reflections emerged to give a final data set of 2940 unique reflections (R m e r g e = 4.6%) which were used in the subsequent structure analysis. 18 2.3 Results and Discussion 2.3.1 Synthesis A list of the Cp'W(NO)(R)(R') complexes selected for this investigation, including the synthetic routes to the bis(alkyl) species, is outlined in Scheme 2.1. These complexes are divided into four groups based on the nature of Cp' , R, and R ' . With the exception of 2.4, Scheme 2.1 R 2Mg-(diox) Cp 'W(NO)Cl 2 • Cp'W(NO)(R)(R') THF 0.5 R 2Mg-(diox) \ / 0.5 R' 2Mg-(diox) Cp*W(NO)(R)(Cl) THF ' E t 2 0 A Cp' = Cp B Cp' = Cp* 2.1 R = R' = C H 2 C M e 3 2.4 R = R' = C H 2 C M e 3 2.2 R = R* = C H 2 S i M e 3 2.5 R = R '= C H 2 S i M e 3 2.3 R = R' = C H 2 C M e 2 P h 2.6 R = R' = C H 2 C M e 2 P h C Cp' = Cp* 2.7 R = C H 2 C M e 3 R' = C H 2 C M e 2 E t 2.8 R = C H 2 C M e 3 R' = C H 2 C M e 2 P h 2.9 R = C H 2 C M e 3 R' = C H 2 S i M e 3 2.10 R = C H 2 C M e 3 R' = C H 3 2.11 R = C H 2 C M e 3 R' = Ph 2.12 R = C H 2 C M e 2 P h R' = C H 2 S i M e 3 2.13 R = C H 2 C M e 2 P h R* = C H 3 D Cp* = Cp* 2.14 R = C H 2 C M e 3 R' = C l 2.15 R = C H 2 C M e 3 R' = C H 2 P h 2.16 R = C H 2 C M e 3 R '= OMe 2.17 R = C H 2 C M e 3 R' = N M e 2 19 2.16, and 2.17, the known complexes 2.1-2.6 ' and 2.8-2.14 ' were prepared according to literature procedures, while the new complexes 2.7 and 2.15 were prepared in 67% and 75% isolated yields by analogous methods. The synthesis of 2.4 as reported previously2 8 involves metathesis of Cp*W(NO)Cl2 with 1 equiv of (Me3CCH2)2Mg-(dioxane) in THF. This method, unfortunately, affords the desired wine-red product in only 27—31 % yield at best. A more efficient one-pot synthesis was thus developed for the work described here. This new approach, which reproducibly provides 2.4 in yields as high as 65-73%, is analogous to that for the synthesis of the mixed bis(alkyl) species and involves sequential use of THF and Et20 as solvents. The role of Et20 is unclear at the moment.31 In contrast to the formation of complexes 2.1-2.15, the preparation of Cp*W(NO)(CH 2 CMe3)(OMe) (2.16) and Cp*WfNO)(CH 2 CMe3)(NMe 2 ) (2.17) by metathesis routes have met with little success. The reactions of Cp*W(NO)(CH 2 CMe 3 )Cl with NaOMe or LiNMe2 in either THF or Et^O, for example, proceeded with only partial conversion and led to an intractable mixture of products containing <20% of 2.16 and 2.17, respectively. Complexes 2.16 and 2.17 were thus prepared as shown in eq 2.4. This alternative method, which is described in full detail in Chapter 3, involves the use of the alkylidene complex, Cp*W(NO)(=CHMe 3)(PMe3), as the starting material and offers the following three attractive features: (1) it is quantitative (>97% N M R yield), (2) it is unaffected by the presence of excess alcohol or amine, and (3) it produces volatile PMe 3 as the only by-product. MeOH Cp*W(NO)(=CHCMe 3)(PMe 3) 1 *• -PMe 3 M e 2 N H >• THF Cp*W(NO)(CH 2CMe 3)(OMe) 2.16 Cp*W(NO)(CH 2 CMe 3 )(NMe 2 ) 2.17 (2.4) 20 2.3.2 Spectroscopic Characterization With the exception of complexes 2.7 and 2.15-2.17, the spectral data of complexes 2.1-2.6 and 2.8-2.14 have previously been reported. 2 7 - 3 0 These reported data, however, are incomplete and have not been fully interpreted. For example, the anomalous spectral features of these complexes have been neither identified nor examined by previous investigators. 2 7 - 3 0 The objective of this section is to discuss the spectral properties of all complexes listed in Scheme 2.1 and, in particular, to address and interpret the unusual spectral features that these complexes exhibit including those that had earlier been overlooked. These data not only show that in most cases these compounds contain at least one a-agostic C - H bond, but also that the strength of each of these bonds is dependent on the nature of the alkyl ligands and the effective electron density at the metal. In this section, a discussion of the IR data wil l be presented first, followed by a discussion of the *H and 1 3 C N M R data of complexes of groups A and B , and of groups C and D, respectively. 2.3.2.1 IR Studies The first indication that several complexes of the type Cp'W(NO)(R)(R') contain an a-agostic structure comes from the Nujol-mull IR spectra of complexes 2.2, 2.5, and 2.11, which show intense V N O and weak, broad V C H stretching vibrations at 1547-1558 cm"1 and 2478-2715 cm"1, respectively (Table 2.1). The extraordinarily low V C H is consistent with the presence of a C - H bond whose strength is diminished due to a strong interaction with an electron-deficient metal center.1 For complex 2.2, the V C H absorption appears as a broad band with two maxima that, as would be expected for an a-agostic complex, is shifted to 1992-2142 cm"1 upon substitution of the four a-H atoms by deuterium atoms (Figure 2.1). It is also ca. 220 and 170 cm"1 higher in energy than that of the Cp* complexes 2.5 and 2.11, respectively, a feature that can be attributed to there being less electron density at this metal center for M (d^-to-C-H (o*) backbonding.32 Since the extent of backbonding is the chief factor affecting the C - H bond order, and hence the V C H value, it can be concluded that the latter compounds have stronger agostic interactions. 21 Table 2.1. Selected IR Data for Complexes 2.2, 2.2-</4, 2.5, and 2.11* Cmpd Nujol* No. R R' v N 0 v C H 2.2 CH 2CMe 2Ph CH 2CMe 2Ph 1558 (s) 2686,2715 (br w) 2.2-rf4 CD 2CMe 2Ph CD 2CMe 2Ph 1556 (s) 2148,2038, 1992 (w) 2.5 CH 2CMe 2Ph CH 2CMe 2Ph 1547 (s) 2478 (br w) 2.11 CH 2 CMe 3 Ph 1547 (s) 2529 (brw) "v in cm"1. bs = strong; w = weak; br = broad. 3000 2500 2000 1500 1000 Wavenumbers Figure 2.1. Nujol-mull IR spectrum of (A) CpW(NO)(CH 2 CMe 2 Ph) 2 (2.2) and (B) CpW(NO)(CD 2 CMe 2 Ph) 2 (2.2-d4). 2.3.2.2 N M R Studies The room-temperature ' H and 1 3 C N M R spectra of complexes 2.1-2.17 in CeD6, and of 2.11 and 2.15 in CD 2 C1 2 have been examined. In general, these data are in agreement with 22 previously reported observations and assignments except for those pertaining to the a-moieties of the alkyl ligands. Several spectral parameters of the oc-H and a-C atoms are unusual, and provide further evidence for the presence of a-agostic interactions in these complexes. As summarized in Table 2.2, these include the ' H and 1 3 C chemical shifts, the ' F I - ' H chemical shift differences, the ' H - H , ' H - ^ C , and ' H - 1 8 3 W coupling constants. The N M R properties of complexes in groups A and B differ from those of group C, which differ in turn from those of group D. Each of these cases wil l therefore be discussed separately. As previously reported, the l H and l 3 C N M R spectra of complexes in groups A and B (i.e. 2.1-2.6) are all similar and are consistent with the compounds possessing a three-legged piano-stool molecular structure in which a mirror plane renders the two alkyl groups chemically equivalent.2 7'2 8 Noteworthy, however, are the Jwo sets of resonances integrating for 2H each due to the diastereotopic methylene protons. In contrast to literature reports, these appear as an A A ' X X ' Z splitting pattern (where Z = 1 8 3 W , / = lA, 14% abundance) rather than as a pattern for an A 2 B 2 2 9 or A 2 X 2 2 3 spin system (see Figure 2.2 for the ' H N M R spectrum of the methylene regions of 2.4). Thus, for each complex one of the methylene resonances is a slightly distorted doublet that appears at 5 2 to 4 and shows a 2JHH coupling constant of 11-12 Hz; while the other resonance is a prominent pseudodoublet that is highly shielded (5 -0.5 to -1.5) and exhibits additional splittings due to coupling of these protons to tungsten (VHW ~ 8-13 H z ) 3 3 as well as to each other (VHH ~ 1-4 Hz) . 3 2 In general, a four-bond H - H coupling across a metal center, a large chemical shift difference for the diastereotopic methylene resonances (2.79 < A8HH < 5.12 for 2.1-2.6), and a high-field shift for one of these resonances are spectral manifestations of strong agostic interactions.1 A negative ' H chemical shift, in particular, is often characteristic of a hydrogen atom that is hydridic in nature due to coordination to a metal. For comparison, the W-H chemical shifts of Cp*W(NO)(r] 3-allyl)(H) 3 4 a and CpW(NO)(H)(I)(PR 3) 3 1 b have also been found to occur ca. 5 -0.5 to -2 , while V H H values of ca. 1-3 Hz have been reported for the double oc-C-H agostic complex, CpW(=CAd)(CH 2 CMe 3 ) 2 , 2 4 the single a - C - H agostic complexes Cp*Ta(=CHPh)(CH=PPh 3)Cl 3 5 and [77 5 -C 5 (CMe 3 ) (CH 2 CMe 3 ) 2 (CH 2 CMe 2 CH 2 ) 2 ]TaCl 2 , 3 6 and the p - S i - H agostic complex Cp 2 Zr(H)(N'BuSiMe 2 H). 3 7 CO CN T3 CD o 13 Q SJO u a 7 u X u "o, e o U 1-1 cS ea Q o ro T3 C 03 a CD C ' £ Cn CM — S3 H B U CO U o U CO CO < CO i ; a X X CO X to < to ^ 3 X to T3 Cc, Ov co 7 u X o u u o OS CN N O O 2 -a -a o oo T3 J3 o tN X U PL, CN 5 u tN X u T3 - d o SO ro ON SO r~ CN SO ro SO CN ro CN o oo SO OS cn ? 3 CN X u X u CO r i o -a 0 0 oo 7 S-u X u u U <N CO OS (N CN OS OS •a •a ON Os CN OS OS T 3 •a CN O o CO CN •a o r o CN •a CO 7 X i u X u X i u tN X u cn ei o <N sc u SO CN OS SO OS OS <N cn W tu S u tN u u CN Os os •a •a Os o CN SO OS •a T 3 CO T3 ro ro o" T 3 •a CN OS O CN O © ro SO CN c>i 2 U X u u X u ro O •a •a o <N •a •a cn ; j cn ro CN 7 <o co" oo so" ro ro CN 7 ro" ? CN ro ro (dd, (dd, dd, so ro cn q ro ro CN — ' • a -a s—^ CO o 0 0 <N CN ro o CN SO CN T3 T3 C N £ 2 T 3 T 3 O N C M o\ co i n C N MO o oo ON VN (dd, 1' oo C N c f CN p o T f ON X> T 3 ON O C N C N ON C N ON O O C N C N T 3 X J •St-O ON c o C N r--ON T3 X I C N T 3 • O >0 ON T3 •a </1 v "*cr o NO o — i N O CN — ' NO c o O N N O N O o O N C N ON © o CM T 3 -a c o r -CM 7 — T 3 T 3 00 o T T CM o O N T T O O O NO NO CM CO T3 CM CM " ° co' ^ , . , m . —- —' — <N T3 X J § 2 •cr T 3 CM I ON O 00 C N T 3 T3 N O m , — • o CO CM CN 1 — ' 1 c f T 3 T l " »*o C N NO" 7 2 T3 CM o CM cr x< o I T ? >-C N C O C N CO CM 1 CO 3 T f i n oo 2 o T3 N O C N CO XI P-t u CJ 2 o o X u u X u 2 u o o u X u u x u u x u u u 2 u o O N C M O r i co C M r i 25 2.85 2.80 2.75 2.70 2.65 2.60 PPT) -1.35 -1.40 -1.45 -1.50 -1.55 Figure 2.2. 500-MHz ' H N M R spectrum of the methylene protons of 2.4 in C 6 D 6 at 20 °C. In the gated 1 3 C{ 'H} N M R spectra of 2.1-2.6, the a-carbons of each complex resonate as a doublet of doublets with a normal and a reduced 1 3 C - ' H coupling constant (1JCH) of 117-124 Hz and 99-104 Hz, respectively. In all instances, the latter is associated with the upfield methylene protons as judged from the 1 3 C satellites in the ' H N M R spectra. As illustrated in Figure 2.3, it is also inversely proportional to the chemical shift of the a-carbon resonances (8c) and is consistently smaller for Cp* complexes and for complexes containing the neopentyl and neophyl ligands. As noted in the introduction, a low 1JCH value is indicative of C - H - M interaction, and can be attributed to C - H bond weakening. A 1 3 C chemical shift to higher frequency as 1 JCH decreases, on the other hand, is suggestive of a geometrical distortion and an increase in sp2 character at this carbon. That the latter is likely the case is based on Schrock's studies of the a-agostic interactions in a series of neopentyl and neopentylidene complexes of the type T a ( C H 2 C M e 3 ) x C l y (x = 1-3, y = 2-4) 3 8 and Ta(=CHCMe3)2(PMe3)2(R) (R = alkyl). 3 9 These studies show that as the ' JCH of the neopentyl and neopentylidene ligands decrease, the 1 3 C shifts of the a-carbon of these ligands move to higher and lower frequencies, respectively, consistent with the a-carbon of the neopentyl ligand becoming more sp -hybridized and the a-carbon of the neopentylidene ligand becoming more sp-hybridized. 26 Figure 2.3. Plot of VCH versus S C a as a function of Cp ' and R for complexes 2.1-2.6 (Cp = A ; C p * = • ) . The above data thus not only indicate that the upfield methylene hydrogens in 2.1-2.6 are agostic but also establish that the degree of agostic bonding in these complexes varies with Cp ' and R. Assuming that a reduction in ' / C H value is representative of the strength of the agostic interaction, the order for decreasing degree of agostic bonding in 2.1-2.6 is as follows: Cp ' = Cp* > Cp and R = C H 2 C M e 3 * C H 2 C M e 2 P h > C H 2 S i M e 3 . That a stronger a-agostic interaction is observed for more electron-rich metal centers and for sterically more bulky alkyl ligands is not unprecedented. Indeed, similar steric and electronic effects have been observed by Schrock and co-workers in their spectroscopic and structural studies of a-agostic alkylidene complexes.4 0 In contrast to complexes of groups A and B, the complexes of group C are chiral and, as such, they show not only separate N M R signals for each of the a-carbons and a-methylene protons but also different splitting patterns for the latter depending upon the nature of R and R' . For complexes 2.7, 2.8, 2.9, and 2.12 (i.e. R * R' = C H 2 C M e 3 , C H 2 C M e 2 E t , C H 2 C M e 2 P h , or CH 2 SiMe 3 ) , the a-protons of each alkyl ligand resonate as a doublet and a doublet of doublets at ca. 8 1.1 to 4.3 and 8 -0.5 to -2.8, respectively. Again, as observed for the 27 symmetrical bis(alkyl) species, the splittings of the upfield resonances are due to geminal coupling as well as four-bond ' F I - ' H and two-bond ' H - 1 8 3 W couplings. This is clearly illustrated by the ' H N M R spectrum of 2.9 shown in Figure 2.4. Cp* C M e 3 C«2CM<:3 V C H 2 S i M e 3 Ctf 2 SiMe 3 C t f 2 C M e 3 Figure 2.4. 500-MHz ' H N M R spectrum of 2.9 in C 6 D 6 at 20 °C. In order to establish the stereochemical relationship of the a-methylene protons, an N O E difference experiment on complex 2.9 has been performed. The results of this experiment show that the protons that give rise to the upfield doublets of doublets are on average in close proximity to each other. Thus, i f it can be assumed on the basis of steric grounds that the (3-moiety of each alkyl ligand is directed away from the Cp* ligand, then the upfield protons must be situated trans to the N O ligand as shown in Figure 2.5: 28 O N H M e 3 C \ / H H H (dd, 5-2.12) (dd,5-1.29) N O E SiMe, Figure 2.5. Schematic representation of 2.9 along the Cp*-W vector. The Cp* ligand (situated above) is not shown for reasons of clarity. For complexes 2.10 and 2.13 in which R = C H 2 C M e 3 or C H 2 C M e 2 P h and R ' = C H 3 , the methyl-ligand proton signals are observed at 8 0.5 as a doublet, while the methylene proton signals are observed at ca. 8 4.2 and 8 -3.0 as a doublet and a doublet of quartets. Of significance here is that the chemical shift difference for the two methylene resonances (i.e. A8HH) is even greater than those of the bis(alkyl) species described above, and that the chemical shift for the methyl resonance appears to be normal, being roughly mid-way between the methylene signals. In the case of the phenyl complex 2.11 (i.e. R ' = Ph), a pair of diastereotopic methylene doublets with a similarly large chemical shift difference is again observed. A comparison of the ' H N M R data of complexes 2.7-2.13 reveals that the magnitude of the chemical shift difference for each pair of methylene resonances (A8 H H) is dependent upon two variables. First, it depends on the identity of the alkyl ligand, decreasing in the following order: C H 2 C M e 3 > CFf 2 CMe 2 Et > C H 2 C M e 2 P h > C H 2 S i M e 3 . Second, it decreases as the neighboring alkyl group becomes more neopentyl-like. For example, for complexes 2.7-2.11 the ASHH value of the neopentyl ligand decreases as R ' is varied from C H 3 and Ph to C H 2 S i M e 3 to C H 2 C M e 2 P h to CH 2 CMe 2 Et . Similarly, the A 8 H H value of the neophyl ligand in complexes 2.12 and 2.13 decreases as R ' is varied from C H 3 to C H 2 S i M e 3 . The observation of A8HH values which are not only large but also ligand-dependent strongly indicates that these trends are neither by accident nor a consequence of diamagnetic anisotropy. A more 29 reasonable explanation is that a large A8HH value is caused by coordination of one of the methylene C - H bonds to the metal center and that the magnitude of this spectral parameter is primarily affected by the strength of this coordination. In the gated 1 3 C{ 'H} N M R spectra of complexes 2.7-2.13, each oc-methylene carbon resonance appears at low field (8 55-122) as a doublet of doublets with normal and reduced 'JCH values similar to those of the symmetrical bis(alkyl) species. By contrast, the a-methyl signal of complexes 2.10 and 2.13 each appears as a high-field quartet (8 29) with a normal ' JCH value of 123 Hz. Comparison of the chemical shift (8c) and the reduced ' JCH values of the oc-methylene groups in 2.7-2.13 reveals that they too are ligand dependent and that they follow essentially the same trends as determined from the ' H N M R data. Although the differences are small, it is nevertheless noticeable that 8c is shifted to higher field while 1 JCH decreases as the alkyl ligand is varied from C H 2 S i M e 3 to C H 2 C M e 2 P h to C H 2 C M e 2 E t and CH 2 CMe3. The same changes are also observed on changing the neighboring alkyl group. For example, comparison of the ' JCH values of the neopentyl ligand in complexes 2.7-2.11 shows that it decreases as R ' is varied from C H 2 C M e 2 P h to C H 2 S i M e 3 to C H 3 to Ph (Figure 2.6). 115 A C H 2 P h N I X o - 5 105 95 85 80 C H , C M e 3 • CH 2 CMe 2 Et C H 2 C M e 2 P h • Cl *CH 2 S iMe 3 • Me A Ph] 100 120 5C a (ppm) Figure 2.6. Plot of the neopentyl ligand's ' JCH versus 8 C a as a function of R ' for complexes 2.7-2.11 and 2.14-2.15. 30 The observation of a normal 8H, a high-field 8c, and a normal VCH for the methyl ligand in complexes 2.10 and 2.13 suggests that this ligand is weakly agostic at best, i f at all. It thus can be concluded that each of these compounds, along with the phenyl complex 2.11, possesses essentially one agostic interaction and that this interaction involves a neopentyl or a neophyl cc-C-H bond. In the case of complexes 2.7-2.9 and 2.12-2.13, the observation of large A8HH and low VCH values for both alkyl ligands can be rationalized by considering that an a - C - H bond in each ligand is involved in agostic bonding, though not necessarily simultaneously (see later). Furthermore, the fact that one alkyl ligand has a larger A 8 H H and a lower 1 JCH than the other strongly suggests that these agostic interactions are unequal. Thus, on the basis of the ' H and 1 3 C N M R spectral trends described above, it can be concluded that the degree of agosticity of a given alkyl ligand is optimized when the neighboring alkyl ligand is small and/or lacks a-hydrogens and that the neopentyl ligand is the best agostic donor among the alkyl ligands investigated in this study. In contrast to complexes of groups A - C , the complexes of group D all contain a potential 7i-donating ligand and exhibit N M R spectra consistent with the presence of an a-agostic interaction for some but not for other complexes. For the chloro complex 2.14, both its ' H and 1 3 C N M R spectra display methylene signals with chemical shifts and coupling constants (A8 H H = 3.68 ppm, VCH = 93, 131 Hz) indicative of a highly distorted a-agostic neopentyl ligand. The presence of lone-pair electrons on the chlorine atom thus has no effect on the ability of the neopentyl to form a three-center, two-electron bond with tungsten. A -similar lack of effects by C l ligands has been reported for the titanium methyl complex, (dmpe)(Cl)3Ti(CH 3), 4 1 which has been assigned an a-agostic structure by neutron diffraction, and for the a-agostic niobium alkyl species, Tp*Nb(Cl)(R)(PhC=CMe) (R = C H 3 , C H 2 C H 3 , C H 2 S i M e 3 ) . 4 2 For the benzyl complex 2.15, the situation is not as straightforward. For example, although the neopentyl methylene protons give rise to two distinct doublets at 8 -2.37 and 2.19 (VHH = 13.2 Hz, A8HH = 4.56 ppm) and although the upfield signal also exhibits coupling to tungsten (V H w =11.1 Hz), the corresponding carbon, however, resonates as a triplet with a 31 chemical shift (5 89.6 in CD2CI2 or 82.6 in C^De) and coupling constant CJQH = 112 Hz) that are more consistent with the average for an agostic and a nonagostic neopentyl methylene moiety. For comparison, the corresponding 8c and 1JQH values for complexes 2.1, 2.4, 2.7-2.11, and 2.14 in C 6 D 6 are in the ranges 8 91-125 and 87-103 Hz, respectively. The N M R signals for the benzyl ligand are similarly unusual. For example, the chemical shifts (8 2.22 and 2.98, A8HH = 0.76 ppm) and coupling constants ( 2 JHH = 8.4 Hz) of the methylene proton signals of this ligand are intermediate for an 77' -benzyl (8 0 - 1 , V H H ~ 10 Hz) and an if-benzyl (8 2 - 3 , 2JHH ~ 5-7 Hz) ligands. 2 6 ' 4 3 The same trend is also observed for the ipso and methylene carbon resonances, which occur at 8 133.2 and 5 51.6 (t, LJCH = 134 Hz). For comparison, the 1 3 C shifts for the ipso carbon of rj1-benzyl and if -benzyl groups in structurally related complexes are in the regions 8 150-160 and 8 110-120, respectively.2 6'3 9 Taken together, these data can be rationalized in terms of the following equilibrium in which the neopentyl and benzyl ligands of 2.15 are rapidly exchanging their modes of attachment to the tungsten center at room temperature. W = Cp*W(NO) In support of this proposal, the N M R spectra of 2.15 are temperature-dependent. Cooling a sample to - 8 0 °C causes the gated 1 3 C{ 'H} N M R spectrum to change as follows. First, the signal due to the neopentyl methylene carbon is shifted upfield by ca. 30 ppm while its 1JQH coupling constant is increased to 117 Hz, a value that is characteristic of a nonagostic neopentyl ligand. Second, the signals for both the benzyl methylene (8 44.9) and benzyl ipso (8 117.2) carbons are similarly shifted to lower frequencies. Importantly, the chemical shift of the latter is in the region associated with static if -benzyl ligands, while the ' JCH coupling constant (143 Hz) for the former is consistent with considerable sp2 character at this carbon. 2 6 ' 3 7 On the basis of these spectral observations it can be concluded that (1) the low-temperature limiting structure for 2.15 is that in which the benzyl ligand is coordinated to 32 tungsten in an r/2-fashion and the neopentyl ligand is nonagostic, and (2) at room temperature this r]2-benzyl-metal interaction is fluxional, thereby allowing an a - C - H bond of the neopentyl ligand to compete for coordination. Finally, in contrast to the N M R spectra of complexes 2.1-2.15, the N M R spectra of the neopentyl complexes Cp*W(NO)(CH 2 CMe 3 )(OMe) (2.16) and Cp*W(NO)(CH 2 CMe 3 )(NMe 2 ) (2.17) display normal chemical shifts and coupling constants for the cc-methylene protons and carbons. The ' JCH values of 118 Hz for 2.16 and 120 Hz for 2.17, in particular, are indicative of nonagostic structures for these compounds. In the ' H N M R spectrum of 2.17, the methyl substituents of the N M e 2 ligand give rise to two sharp singlets at 8 2.59 and 3.71, which remain unchanged even when heated to 100 °C. This suggests that the amido ligand is functioning as a ft-donor. The lack of agostic interactions in this complex and in 2.16 therefore can be attributed to the inability of an a - C - H bond on the neopentyl group to compete with the lone-pairs of electrons on N and O for metal coordination. The results presented thus far clearly show that in the absence of efficient 7C-donation, alkyl complexes, Cp'W(NO)(R)(R'), possess an a-agostic structure in solution. When R is a sterically bulky alkyl ligand and R' is C H 3 , Ph, or C l , the metal center approaches a closed-shell configuration by forming a strong three-center, two-electron bond with one a - C - H bond. When R and R ' are both bulky alkyl moieties, the metal center is coordinated to two a - C - H groups, one from each alkyl ligand. The nature of the agostic interaction(s) in these latter complexes is not clear-cut, however. The fact that a reduced 1 JCH is observed for both alkyl ligands can be interpreted in terms of either a rapid site exchange of an a -H atom of one alkyl ligand and an a-H atom of the other for successive interaction with the metal center, or a simultaneous interaction of both a-H atoms. In order to distinguish between these two possibilities, a variable-temperature N M R study of the partially deuterated complex, Cp*W(NO)(CH 2 CMe 3 ) (CD 2 CMe 3 ) (2A-d2), was performed. Since the zero-point energy of a C - H bond is higher than that of a C - D bond and since this energy difference is greatest when the two bonds are nonagostic, there is a thermodynamic preference for the former to coordinate to the metal. Thus, i f two a - C - H bonds from different alkyl ligands are indeed 33 competing with each other for the metal, then cooling a solution of 2.4-rf2 should cause the equilibrium to shift in favor of the a - C - H agostic form (eq 2.4), and should lead to an asymmetric structure with a reduction in 1 JQH-O O N N C \ W / C — C W / C ( 2 - 4 ) H D / W = Cp'W(NO) ^ H T Both the ' H and the gated 1 3 C{ 'H} N M R spectra of 2.4-di have been investigated at several temperatures from -80 to +20 °C. As in the case of 2.4, the spectra of 2.4-</2 are consistent with there being a mirror plane in the molecule, even down to -80 °C. In addition, they show that the 1 JCH for the oc-carbons are invariant not only to partial deuterium substitution but also to temperature. These findings thus provide strong support that 2.4-rf2 and related bis(alkyl) complexes possess a time-averaged structure (see later) in which two oc-C - H bonds are coordinated to the metal center. 2.3.3 X-Ray Structural Studies The first Cp'W(NO)(R)(R') complex to have been structurally characterized by X-ray diffraction is the group A complex, CpW(NO)(CH 2 SiMe 3 ) 2 (2.3).31 Although the oc-H atoms were not located, the structure exhibits two features that are characteristic of oc-agostic species. First, the W - C a distances (2.103(9) and 2.108(9) A) are significantly shorter than simple tungsten-carbon single bonds, which are typically 2.18-2.25 A . Second, the W - C a - S i angles (125.5(5)° and 127.1(5)°) are more obtuse than those observed in nonagostic systems. In order to confirm that these unusual features are structural consequences of oc-agostic bonding and to compare the conformational preference of the alkyl ligands in this complex to those in complexes of groups B - D , X-ray crystallographic analyses of the following four compounds have been performed: Cp*W(NO)(CD 2 CMe 3 ) 2 (2A-d4), Cp*W(NO)(CH 2 CMe 3 ) (CH 2 SiMe 3 ) 34 (2.9), Cp*W(NO)(CH 2CMe 2)(C 6H3-2,5-Me 2) (2.11'), and Cp*W(NO)(CH 2 CMe 3 )(NMe 2 ) (2.17). 2.4-d4 and 2.17 are examples of complexes from groups B and D, respectively, while 2.9 and 2.11' are representative of group C. The last is a structural analogue of 2.11 whose synthesis and spectroscopic characterization are discussed in Chapter 4. The above four compounds were chosen because (1) X-ray quality crystals of each could easily be obtained; (2) they each contain a neopentyl ligand, which according to N M R studies, should be significantly distorted and should provide a more clear-cut situation than the (trimethylsilyl)methyl ligands in 2.3; and (3) by sharing the neopentyl ligand they not only allow for direct comparison between complexes, but also for comparison with several related agostic neopentyl complexes reported in the literature. Shown in Figures 2.7 are the ORTEP representations of the solid-state molecular structures of 2.4-d4, 2.9, 2.11', and 2.17. The relevant bond distances and angles, along with those of 2.3, are listed in Table 2.3. Comparison of the structures of 2.4-d4, 2.9, 2.11', and 2.17, with 2.3 shows that they all share a three-legged piano-stool geometry about tungsten and that the intramolecular dimensions involving the C p ' - W and W - N O portions in each molecule are unremarkable. In all cases, the Cp ' ligands are coordinated in an rf fashion while the W - N O linkages are essentially linear with W - N distances (ca. 1.76 A) and W - N - 0 angles (ca. 166.9 -169.5°) in the range found for other Cp'W(NO)-containing systems.2 7'3 9 This indicates that the N O ligands are functioning as formal three-electron donors and that there is considerable W—> N O backbonding. A n examination of the W-alkyl fragments reveals that (1) they each adopt a conformation almost identical to that in 2.3; that is, for all complexes the alkyl ligands are obtuse and oriented with their (3-moieties pointing in the general direction of the N O ligand but away from the plane formed by the Cp'(centroid), the metal, and the N O ligand (Figure 2.7a); and (2) the degree of distortion in each case varies depending upon the alkyl group and the effective electron density at the metal. For complex 2.4-d4, one of the ter/-butyl groups is disordered over two positions in a 0.54:0.46 ratio. The quaternary carbon could be refined anisotropically, but the methyls could 35 not. Comparison of the bonding parameters of the neopentyl ligands in 2.4-d4 reveals that they are inequivalent. Thus, while the W(l)-C(2) distance of 2.169(10) A and W( l ) -C( l ) -C(2) angle of 126.9(6)° are within the range observed for the alkyl ligands of 2.3, the corresponding distance (2.101(10) A) and angle (134.0(7)°) for the neighboring neopentyl group indicate an even greater distortion. For comparison, the M - C a - C p angles in the bis(neopentyl) complexes CpW(=CAd)(CH 2 CMe 3 ) 2 2 4 and CpNb(=NAr)(CH 2 CMe 3 ) 2 , 2 5 in which the presence of two a-agostic interactions has been confirmed by X-ray diffraction, are 134.8(4)° and 132.5-131.2°, respectively (Table 2.4). In order to gain conclusive evidence that 2.4-d4 also has an a-agostic structure, the positions of the a-D atoms were located during the structure refinement. This was achieved by placing the D atoms in idealized positions and then allowing the orientations of the resulting C - D bonds to refine freely. Summarized in Table 2.4 are the bonding parameters of the W - C D 2 C M e 3 fragments in 2.4-d4. Noteworthy is that while the W - C ( l ) - D ( l b ) and W-C(6)-D(6a) angles are normal, the W - C ( l ) - D ( l a ) and W-C(6)-D(6b) angles of 82.1° and 94.1°, respectively, are significantly contracted from tetrahedral and are consistent with these D's being drawn toward the metal. That this is indeed the case is confirmed by the short W--D(la) and W--D(6b) distances of 2.25 and 2.45 A and the relatively lengthened C( l ) -D( la ) and C(6)-D(6b) distances of 1.11 and 1.04 A. For comparison, the M - C a - H bond angles in the alkylidyne and imido complexes mentioned above are 93° and 86.5-89.4°, while the corresponding C - H - - M distances are 2.42 and 2.32-2.41 A, respectively (Table 2.4). In 2.4-d4, the D atom nearest to tungsten is associated with the more acute W - C a - D angle and the more obtuse W - C a - C p angle. Taken together, these observations not only strongly suggest that this complex contains two a-agostic interactions in the solid state, one from each neopentyl ligand, but also show that one of the interactions is stronger than the other and that each of these interactions is indeed the primary cause for the significant angular deformation at the a-carbons. 36 Figure 2.7. ORTEP diagrams of (A) Cp*W(N0)(CD 2 CMe 3 ) 2 (2.4-</4), (B) Cp*W(N0)(CH 2 CMe 3 ) (CH 2 SiMe 3 ) (2.9), (C) Cp*W(N0)(CH 2 CMe 3 ) (2 ,5-C 6 H 3 -Me 2 ) (2.11'), and (D) Cp*W(N0)(CH 2 CMe 3 ) (NMe 2 ) (2.17). 0 0 co i U tN Q O O * CX o co CD X ca I o U o CD 60 cd U CN" X u I > 60 < cl cd cn CD CD cl cd Q Cl o _cd "o 1-a "CD X I o c CD S-l cd CO C cd 60 u CD X H CD 5p a 3 z x Z & u d cd O (N u z-- > / ca ro CN CN 0 0 ^ 3 z -U / \ ? ca C N , 0 2 -u / \ « 2 Q o U 5 z x' z ex CD u U z I  ex U i r i U Q U O z * ex U CD CD c 3 Q C o ; _ T f x z V x z I o I as i 3S I x Z 5 3 SS I 2 ^ Q l as i as Q o NO Q is o - ~ o o —I o ca — —< Q Q I I U CJ CO _CD "Bo c <! T3 Cl O CQ cj i CJ I X CJ I u CJ I ON NO u OQ -H(a) -H(b) i-H(b; OO S o 1 00 O 1 ? CJ 1 Nb Nb H( a; T f o NO ON 00 o ON' O CO x* x) •H(2: •H(2: •H(2: •H(2: i CO CN CO CN CO <N o o CJ 1 i 1 H(a z z H(a as i o i Q l as i o i —. CN CN _r c-j T f S 2 2 Q N p CJ N£> CJ T f TT 00 o o xT x* C(l)-D( a X ( j o CJ 1 7 D(la NO o f X SD C J II 3 c ca e ca ca II 3 cr oo o o o c Cu) cx o CJ c o 39 In contrast to 2.4-d4, the alkyl groups of 2.9 initially appeared equivalent due to a crystallographically imposed mirror plane. The structure thus had to be refined using high-resolution data. By restraining the W - C a bonds to having the same bond length within a 0.010 A esd and all atoms in the disordered groups to having identical anisotropic displacement parameters, the S i - C and C - C bond lengths were successfully refined independently and a chemically reasonable structure was obtained (Figure 2.7b). The important feature to note about this structure is that the neopentyl ligand ( W - C a - C p = 133.3(11)°) is more severely distorted than the (trimethylsilyl)methyl ligand ( W - C a - S i p = 119.1(7)°), a fact that is consistent with the N M R data which suggest that the former has a stronger agostic interaction. In the case of complex 2.11', the asymmetric unit contains two molecules which are virtually mirror images of each other, one of which is shown in Figure 2.7c. Like the bis(alkyl) complexes 2.4-d4 and 2.9, the neopentyl ligand of 2.11' also exhibits a markedly short W - C a distance (2.084(11) A) and a large W - C a - C p angle (133.9(8)°). That the latter is in the range observed for 2.4-d4 and 2.9 implies a similar degree of agostic bonding for all three systems. For complex 2.17, on the other hand, hydrogen bonding between the Cp* ligand of one molecule and the N O ligand of another occurs. This is evident from the C(6)-H(6b)--0(1) distance of 2.555 A and C(6)-H(6b)--0(1) angle of 167.8°. Examination of the amido ligand reveals that the N atom adopts a planar geometry with bonding parameters consistent with it functioning as a rc-donor. For example, the W - N M e 2 distance of 1.931(4) A is indicative of considerable multiple bond character in this linkage, while the N(l ) -W(l) -N(2)-C(16) and N(l ) -W(l) -N(2)-C(17) torsion angles of 177.5(4)° and -5.9(5)°, respectively, are of proper magnitude for maximum 7i-bonding. The neopentyl ligand displays a normal W - C a distance of 2.186(4) A and a W - C a - C p angle (122.9(3)°) that is substantially less than that in 2.4-d4, 2.9, and 2.11'. Thus, although the W - C a - C p angle deviates from the ideal 109° for sp3-hybridized centers, it is more likely that this deviation is a consequence of the intrinsic steric 40 repulsion between the tert-butyl group and the rest of the molecule rather than oc-agostic bonding. 2.3.4 Molecular Orbital Description of the Agostic Bonds As noted in Chapter 1, Cp'W(NO)(R)(R') bis(alkyl) complexes are formally 16-electron species which possess a low-lying unoccupied molecular orbital (LUMO) that is approximately d x 2_ y 2 in character.44 Although the orbital is situated such that three of its four lobes are suitably oriented for bonding to additional ligands, the spectroscopic and structural data presented above, however, show that only the lobe that is straddled by the alkyl ligands is involved in oc-agostic bonding. For complexes in which there is only one OC -C-H- -W interaction, donation of an electron pair from a C - H bond gives the metal a configuration approaching 18 electrons. For complexes in which two such interactions are present, the bonding situation is more complicated, for the metal now has an electron count of 20 electrons. In general, transition-metal complexes with greater than an 18-electron count are oxidatively unstable since antibonding orbitals are used to accommodate the excess electron density.45 Two representative examples of such complexes are cobaltocene and nickelocene, which are a 19-and 20-electron complexes, respectively 4 6 In the case of the Cp'W(NO)(R)(R') bis(alkyl) complexes, however, the formation of a double oc-agostic interaction need not result in the antibonding orbitals being occupied. This is possible by considering the molecular orbital model shown in Figure 2.8. In this model, which is analogous to that proposed previously by Schrock and co-workers for the alkylidyne bis(neopentyl) system mentioned above,2 3 mixing of two oc-C-H o orbitals with the metal-based L U M O leads to formation of C - H - W bonding, nonbonding, and antibonding molecular orbitals. Since there are only two pairs of C - H a electrons, only the bonding and nonbonding molecular orbitals wil l be occupied. 41 0 \ ^ 0 H H Energy \ C H H + Figure 2.8. Qualitative molecular orbital diagram for the double oc-agostic interaction in Cp'W(NO)(R')(R') complexes. 2.4 Epilogue In summary, a series of 17 alkyl complexes of the type Cp'W(NO)(R)(R') have been systematically studied by IR and N M R spectroscopies. The X-ray structures of several of these species have also been determined. The picture that emerges from these studies is that these compounds adopt an oc-agostic structure when secondary interactions such as lone pair or 7i-electron donation are not possible. Thus, with the exception of the alkoxo and amido complexes, 2.16 and 2.17, in which the metal center attains an 18-electron configuration via lone-pair donation from the O and N atoms, the remaining complexes exhibit cc-methylene agostic interactions of varying strengths and multiplicities. In solution the oc-agostic interactions give rise to a number of unusual spectroscopic features for each oc-methylene moiety. Most noteworthy are (1) a negative ' H shift for one of the methylene protons, indicating the presence of an oc-agostic interaction between this hydrogen and the tungsten, (2) reduced VCH and 'JCH values, consistent with a weakening of 42 the oc-C-H bond, and (3) a 1 3 C shift to higher frequency as 'JCH decreases, indicating an increase in sp2 character and a geometrical distortion at this carbon. In the solid state, these oc-agostic interactions lead to shortened W - C a bonds and markedly obtuse W-C a -C(Si)p angles. That these abnormal structural features are indeed consequences of oc-agostic bonding has been confirmed by the X-ray structure of complex 2.4-rf4, in which all four D atoms were located. Although the uncertainties associated with the metrical parameters of these atoms are large, it nevertheless is clear that one C - D bond on each methylene group is normal while the other is agostic. Two types of oc-agostic complexes have been established. In one type, which includes complexes in which R = CH2CMe3 and R ' = Ph, C H 3 , C l , or CH 2 Ph , one oc-agostic interaction involving the neopentyl ligand is observed. The order for the increasing degree of oc-agostic bonding among these complexes based on the magnitude of the 'JCH value is R ' = PhCH.2 < C l < CH3 < Ph. That the benzyl complex contains the weakest oc-agostic interaction is consistent with the fact that in this complex the neopentyl oc-C-H bond has to compete with the 7T,-system of the benzyl ligand for the vacant metal orbital. In the case of the phenyl complex, the interaction between the oc-C-H bond and the metal is strongest because the phenyl ligand can only act as a o-donor. The second type of oc-agostic complexes is significantly more striking, for two oc-agostic interactions from different alkyl ligands are present. On the basis of the N M R data, the strength of these interactions varies as follows. First, among all the alkyl ligands studied, the order of decreasing oc-agostic donor ability is C H 2 C M e 3 > C H 2 C M e 2 E t > C H 2 C M e 2 P h > C H 2 S i M e 3 . Second, for a series of Cp* complexes in which R = C H 2 C M e 3 and R ' = C H 2 C M e 3 , C H 2 C M e 2 E t , C H 2 C M e 2 P h , or C H 2 S i M e 3 , the degree of oc-agosticity of R decreases as R ' is changed from C H 2 S i M e 3 to C H 2 C M e 2 P h to C H 2 C M e 2 E t to C H 2 C M e 3 . Third, for a series of Cp and Cp* complexes in which R = R ' , the agostic interactions in the Cp* series are consistently stronger than those in the Cp analogues, a fact that can be ascribed to there being a greater degree of metal-to-C-H (o*) backbonding in the former complexes. A similar 43 phenomenon has been well established for related rf-U.2 compounds, whose existence can only be detected when the metal has d electrons for back-donation.47 Although the trends above clearly show that in solution the a-agostic interactions in Cp'W(NO)(R)(R') complexes are unequal when R ^ R' and that the degree of agosticity of a given alkyl group changes as R ' is varied, this is not entirely true in the solid state as the neopentyl ligands in complexes Cp*W(NO)(CH2CMe3)(CH2SiMe3) (2.9) and Cp*W(NO)(CH 2CMe 3)(C 6H3-2,5-Me2) (2.11') are equally distorted. Furthermore, although the N M R data of complexes in which R = R' indicate a plane of symmetry in solution and therefore imply that the agostic interactions are equal, the solid-state structure of 2.4-</4 shows it to contain an asymmetric structure in which one neopentyl group is not only markedly more distorted than the other but is also equally distorted compared to that in 2.9 and 2.11'. Taken together, these results suggest that (1) in the solid state, regardless of whether R = R ' or R * R' , the double a-agostic interactions in Cp'W(NO)(R)(R') bis(alkyl) complexes are unequal and do not vary as a function of the neighboring alkyl group, and (2) in solution, they exist as time-averaged structures due equilibration of the type shown in eq 2.5. O O N N R I R' R w / R ' ( 2 ' 5 ) C ^ V ^ C X X C ' / ' N > C / H W = Cp'W(NO) X H n The observation that only the (trimethylsilyl)methyl and neopentyl-like ligands are agostic suggests that steric factors play an important role in the occurrence of agostic interactions. This is not surprising since in order to minimize steric crowding at the metal, the W - C a - C p would have to widen, thereby placing an a - C - H bond close the metal. Among the ligands studied, the neopentyl is the most agostic. The reason for this selectivity is not clear. It appears not to be solely steric in origin, since i f this were the case, then the C H 2 C M e 2 E t and C H 2 C M e 2 P h (neophyl) ligands, which are larger than the neopentyl ligand, would be expected 44 to have stronger oc-agostic interactions. Furthermore, although it could be argued that the neophyl ligand has a weaker agostic bond because of the electronic effects of the phenyl substituent, this is also unlikely since the strengths of the agostic bond of this ligand and of the neopentyl ligand are identical when the neighboring alkyl group of both complexes is neophyl or neopentyl, respectively, or when it is a methyl or a trimethylsilylmethyl group. Finally, the reason why two oc-agostic interactions are formed, when one is sufficient for electronic saturation, is also uncertain. Given the fact these interactions only occur in complexes in which both alkyl ligands are bulky, steric factors again appear to play an active role in making these interactions feasible. 2.5 Addendum: Neutron Structure of Cp*W(NO)(CH 2CMe 3) 2 (2.4) and Reinterpretation of Spectroscopic and Structural Data As noted above, the position of hydrogens as determined by X-ray diffraction is subject to appreciable error. This is due to the fact that peaks due to hydrogen atoms are typically near the noise level of the electron-density maps. Thus, in order to unambiguously establish the multiplicity of the oc-agostic interaction in 2.4, a neutron diffraction study of this complex was undertaken. This was achieved at 120 K and yielded the results shown in Figure 2.9. As was found in the X-ray analysis of 2A-d4, one of the fert-butyl groups in 2.4 exhibits packing disorder — the three methyl groups at C(2) thus had to be refined as two sets of half-methyls, approximately equally populated and related to each other by a 60° rotation about the C(l)-C(2) bond. This disorder, fortunately, has no effect on the atomic positions of the neopentyl methylene groups. As illustrated in Figure 2.9, the neutron structure of 2.4 and the X-ray structure of 2.4-d4 are very similar, differing significantly only in the extent by which the C(1)-H(1A) and C(6)-H(6B) bonds distort. Thus, as in 2A-d4 the neopentyl ligands in 2.4 are also distorted from regular tetrahedral geometry at C( l ) and C(6). The W( l ) -C( l ) -C(2 ) angle and the 45 2.4-rf4 2.4 Cf; 010 W ( l ) - N ( l ) 1.752(9) 1.785(3) W ( l ) - C ( l ) 2.101(10) 2.110(3) W(l ) -C(6) 2.169(8) 2.175(3) C( l ) -C(2) 1.528(13) 1.538(3) C(6)-C(7) 1.482(13) 1.542(3) N( l ) -0 (1) 1.244(9) 1.228(3) W ( 1 ) - H ( 1 A ) 2.25(10) 2.233(6) W ( 1 ) - H ( 1 B ) N / A 2.693(6) W ( 1 ) - H ( 6 A ) N / A 2.745(6) W ( 1 ) - H ( 6 B ) 2.45(9) 2.582(6) C(1)-H(1A) 1.110 1.153(6) C(1)-H(1B) 0.839 1.070(8) C(6)-H(6A) 0.896 1.094(6) © H!B HOB ;gj . • ® \ ' £ f 01 C(6)-H(6B) 1.042 1.083(7) W ( l ) - C ( l ) - C ( 2 ) 134.0(7) 133.4(2) W(l ) -C(6) -C(7) 126.9(6) 125.4(2) W(1)-C(1)-H(1A) 82.1 80.6(3) W(1)-C(1)-H(1B) 101.4 111.5(3) W(1)-C(6)-H(6A) 102.2 109.7(3) W(1)-C(6)-H(6B) 94.1 99.3(3) H(1A)-C(1)-H(1B) 108.4 103.9(5) H(6A)-C(6)-H(6B) 104.2 103.6(5) W ( l ) - N ( l ) - 0 ( 1 ) 168.7(7) 169.1(2) N ( l ) - W ( l ) - C ( l ) 98.7(4) 99.4(2) N ( l ) - W ( l ) - C ( 6 ) 98.8(4) 99.42(14) C ( l ) - W ( l ) - C ( 6 ) 106.4(4) 106.48(14) Figure 2.9. Two views of the neutron structure of Cp*W(NO)(CH 2 CMe 3 ) 2 selected bond distances and angles for 2.4 (neutron) and 2.4-rf4 (X-ray). (2.4) alon; >g with 46 W ( l ) - C ( l ) distance are 133.4(2)° and 2.110(3) A, respectively, and are essentially identical to those found in 2A-d4. In contrast, the W(l)-C(6)-C(7) angle of 125.4(2)° and the W(l)-C(6) distance of 2.175(3) A are now only slightly outside the ranges observed in Cp*W(NO)(CH 2 CMe 3 )(NMe 2 ) (see Table 2.3 above) and (dmpe)W(=CCMe3)(=CHCMe3)(CH2CMe3),4 8 which do not contain an oc-agostic neopentyl ligand. With regard to the four methylene hydrogens, the bonding parameters associated with H(1B) and H(6A) are unexceptional, as expected.49 H(1A) and H(6B), on the other hand, are bent toward the metal center such that the W(1)-C(1)-H(1A) and W(1)-C(6)-H(6B) angles are significantly smaller and larger, respectively, than in 2A-d4. The W(1)-C(6)-H(6B) angle is 99.3(3)° and can be compared to analogous angles of 104.2(6)° and 101.1(5)° for the anagostic C H 2 C M e 3 group in Cp* 2 Th(CH 2 CMe 3 ) 2 , 2 2 and 95.7(8)° and 94.0(13)° in Cp*La[CH(SiMe 3 ) 2 ] 2 , 5 0 which does not exhibit oc-agostic C - H interactions with the metal center. Consequently, the W(1)-H(6B) distance is 2.582(6) A, which is too long for there to be a genuine agostic interaction.51 Consistent with this contention, the C(6)-H(6B) bond does not lie in the plane containing the metal L U M O 4 4 [N(1)-W(1)-C(6)-H(6B) = -132.6(4)°] and the C(6)-H(6B) distance is in the range expected for normal aliphatic C - H bonds. In contrast to H(6B), the W(1)-C(1)-H(1A) angle is 80.6(3)° and the W(1)-H(1A) distance is 2.233(6) A, which is only -0.34 A longer than the average dihydrogen W - H distance (1.89(1) A) in the neutron structure of W(CO) 3 (P'Pr 3 ) 2 ( 77 2-H 2). 4 7 A n agostic interaction must be responsible for this short distance since both C( l ) and H(1A) lie in the plane containing the metal L U M O [N(1)-W(1)-C(1)-H(1A) = 87.7(4)°] and since C(1)-H(1A) = 1.153(6) A, which is among the longest C - H bonds ever to be definitively characterized.1"'52 Thus, in contrast to the X-ray structure of 2A-d4 where the presence of two oc -C-H-W interactions, one from each neopentyl ligand, is strongly suggested, the neutron structure of 2.4 establishes that only one such interaction, in fact, exists. 47 On the basis of the close similarity between the neutron structure of 2.4 and the X-ray structures of CpW(NO)(CH 2 SiMe 3 )2 (2.3), Cp*W(NO)(CH 2 CMe 3 ) (CH 2 SiMe 3 ) (2.9), and Cp*W(NO)(CH 2 CMe 3 )(C 6 H 3 -2,5-Me 2 ) (2.11'), it can therefore be concluded that 2.3, 2.9, and 2.11' each also contain only one oagostic C - H interaction. The observation that the neopentyl ligands in 2.9 and 2.11' are equally distorted compared to the agostic neopentyl ligand in 2.4 strongly suggests that in the solid state the agostic interactions in these three complexes are comparable in strength. The oc-C-H-"W interaction in 2.4 (and by analogy in 2.3, 2.9, and 2.11') is best described as a closed, three-center, two-electron bond and serves to give the electron-deficient tungsten center an effective closed-shell configuration. This finding implies that in solution the fluxional process illustrated in eq 2.6 is in fact responsible for the averaging of the alkyl ligands in 2.4 and in other Cp'W(NO)(R)(R') bis(alkyl) complexes and that this exchange process is fast on the N M R time scale even at low temperatures. It also implies that, for a series of Cp* complexes in which R = C H 2 C M e 3 and R ' = C H 2 C M e 3 , C H 2 C M e 2 E t , C H 2 C M e 2 P h , or C H 2 S i M e 3 , the observed 'JCH values for R and R ' vary because of the change in position of this equilibrium depending on the nature of R ' . That is, they change because R is a better oc-agostic donor than R ' and not because it binds more strongly than R ' . Consistent with this new proposal, when First, among all the alkyl ligands studied, the order of decreasing oc-agostic donor ability is C H 2 C M e 3 > C H 2 C M e 2 E t > C H 2 C M e 2 P h > C H 2 S i M e 3 . Second, for a series of Cp* complexes in which R = C H 2 C M e 3 and R ' = C H 2 C M e 3 , C H 2 C M e 2 E t , C H 2 C M e 2 P h , or C H 2 S i M e 3 , the degree of oc-agosticity of R decreases as R ' is changed from C H 2 S i M e 3 to C H 2 C M e 2 P h to C H 2 C M e 2 E t to C H 2 C M e 3 . Third, for a series of Cp and Cp* complexes in which R = R' , the agostic interactions in the Cp* series are consistently stronger than those in the Cp analogues, a fact that can be ascribed to there being a greater degree of metal-to-C-H (a*) backbonding in the former complexes. 48 O N O N R \ R' W = Cp'W(NO) R R' (2.6) A A ' 2.6 References and Notes (1) For reviews, see: (a) Brookhart, M . ; Green, M . L . H . J. Organomet. Chem. 1983, 250, 395. (b) Brookhart, M . ; Green, M . L . Ff.; Wong, L . - L . Prog. Inorg. Chem. 1988, 36, 1. (c) Crabtree, R. H . ; Hamilton, D. G. Adv. Organomet. Chem. 1988, 28, 299. (d) Crabtree, R. H . Angew. Chem. Int. Ed. Engl. 1993, 32, 789. (2) LaPlaca, S. J.; Ibers, J. A . Inorg. Chem. 1965, 4, 778. (3) Bailey, N . A . ; Jenkins, J. M . ; Mason, R.; Shaw, B. L. J. Chem. Soc. Chem. Comm. 1965, 237. (4) Trofimenko, S. J. Am. Chem. Soc. 1967, 89, 6288. (5) (a) Cotton, F. A . ; Stanislowski, A . G. J. Am. Chem. Soc. 1974, 96, 5074. (b) Burger, B . J.; Thompson, M . E.; Cotter, W. D.; Bercaw, J. E. J. Am. Chem. Soc. 1990, 112, (6) (a) Kawamura-Kuribayashi, H. ; Koga, N . ; Morokuma, K . J. Am. Chem. Soc. 1992, 114, 2359. (b) Lohrenz, J. C. W.; Woo, T. K . ; Ziegler, T. J. Am. Chem. Soc. 1995, 117, 12793. (c) Margl, P.; Lohrenz, J. C. W.; Ziegler, T.; Blochl, P. E. J. Am. Chem. Soc. 1996, 775,4434. 1566. (7) Eisenstein, O.; Jean, Y . J. Am. Chem. Soc. 1985, 707, 1177. 49 For examples, see: (a) Lamanna, W.; Brookhart, M . J. Am. Chem. Soc. 1981,103, 989. (b) Lamanna, W.; Brookhart, M . ; Humphrey, M . B. J. Am. Chem. Soc. 1982,104, 2117. (c) Crabtree, R. H . ; Lavin, M . J. Chem. Soc, Chem. Commun. 1985, 794. (d) Kanamori, K . ; Broderick, W. E.; Jordan, R. F.; Willett, R. D.; Legg, J. I. J. Am. Chem. Soc. 1986,108, 7122. (e) Vigalok, A . ; Uzan, O.; Shimon, L. J. W.; Ben-David, Y . ; Martin, J. M . L . ; Milstein, D. J. Am. Chem. Soc. 1998,120, 12539. For reviews, see: (a) ref. 1. (b) Hall, C ; Perutz, R. N . Chem. Rev. 1996, 96, 3125. (a) Turner, H . W.; Schrock, R. R. J. Am. Chem. Soc. 1982,104, 2331. (b) Turner, H . W.; Schrock, R. R.; Fellman, J. D.; Holmes, S. J. / . Am. Chem. Soc. 1983,105, 4942. Blomberg, M . R. A . ; Brandemark, U . ; Siegbahn, P. E. M . J. Am. Chem. Soc. 1983, 105, 5557. Grubbs, R. H ; Coates, G. W. Acc. Chem. Res. 1996, 29, 85. Piers, W. E.; Bercaw, J. E. J. Am. Chem. Soc. 1990, 112, 9406. Katz, T. J.; Sivavec, T. M . J. Am. Chem. Soc. 1985,107, 737. (a) Legzdins, P.; Sayers, S. F. Organometallics 1996,15, 3907. (b) Legzdins, P.; Sayers, S. F. Chem. Eur. J. 1997, 3, 1579. Tran, E.; Legzdins, P. J. Am. Chem. Soc. 1997,119, 5071. Legzdins, P.; Veltheer, J. E. Acc. Chem. Res. 1993, 26, 41, and references therein. Mena, M . ; Pellinghelli, M . A . ; Royo, P.; Serrano, R.; Tiripicchio, A . J. Chem. Soc, Chem. Commun. 1996, 1118. (a) Huang, D.; Streib, W. E.; Eisenstein, O.; Caulton, K . G. Angew. Chem. Int. Ed. Engl. 1997, 36, 2004. (b) Cooper, A . C ; Streib, W. E.; Eisenstein, O.; Caulton, K . G. J. Am. Chem. Soc. 1997,119, 9069. (c) Ujaque, G.; Cooper, A . C ; Maseras, F.; 50 Eisenstein, O.; Caulton, K . G. J, Am. Chem. Soc. 1998, 720, 361. (d) Huang, D.; Streib, W. E.; Bollinger, J. C ; Caulton, K . G.; Winter, R. F.; Scheiring, T. J. Am. Chem. Soc. 1999, 727, 8087. (20) van der Heijden, H. ; Schaverien, C. J.; Orpen, A . G. Organometallics 1989, 5, 255. (21) Jeske, G.; Lauke, H . ; Mauermann, J.; Swepston, P. N . ; Schumann, H . ; Marks, T. J. J. Am. Chem. Soc. 1985, 707, 8091. (22) Bruno, J. W.; Smith, G. M . ; Marks, T. J.; Fair, C. K . ; Schultz, A . J.; Williams, J. M . J. Am. Chem. Soc. 1986, 705, 40. (23) Poole, A . D.; Williams, D. N . ; Kenwright, A . M . ; Gibson, V . C ; Clegg, W.; Hockless, D. C. R.; O'Neil , P. A . Organometallics 1993, 72, 2549. (24) Warren, T. H . ; Schrock, R. R.; Davis, W. M . J. Organomet. Chem. 1998, 569, 125. (25) Cole, J. M . ; Gibson, V . C ; Howard, J. A . K . ; Mclntyre, G. J.; Walker, G. L. P. Chem. Commun. 1998, 1829. (26) Dryden, N . H . ; Legzdins, P.; Trotter, J.; Yee, V . C. Organometallics 1991, 70, 2857. (27) Legzdins, P.; Rettig, S. J.; Sanchez, L . Organometallics 1988, 7, 2394. (28) Debad, J. D.; Legzdins, P.; Rettig, S. J.; Veltheer, J. E. Organometallics 1993,12, 2714. (29) Dryden, N . H. ; Legzdins, P.; Rettig, S. J.; Veltheer, J. E. Organometallics 1992, 77, 2583. (30) Debad, J. D.; Legzdins, P.; Batchelor, R. J.; Einstein, F. W. B . Organometallics 1993, 72,2094. (31) Although their identities have not yet been unambiguously established, the side-products based on their brown and forest green colors are presumably the bridging 51 nitrido complexes, Cp*W(NO)(Cl)(p.-N)[Cp*W(=0)Cl] / y a and Cp*W(NO)(CH 2 CMe 3 )( | l-N)[Cp*W(=0)Cl], 2 9 b formed by reduction of the starting material and the mono(alkyl) intermediate, Cp*W(NO)Cl2 and Cp*W(NO)(CH 2 CMe 3 )Cl , respectively. See: (a) Legzdins, P.; Ross, K . J.; Sayers, S. F.; Rettig, S. J. Organometallics 1997,16, 190. (b) Debad, J. D.; Legzdins, P.; Reina, R.; Young, M . A . ; Batchelor, R. J.; Einstein, F. W. B. Organometallics 1994, 13, 4315. Wood, C. D.; McLain, S. J.; Schrock, R. R. J. Am. Chem. Soc. 1978,101, 3210. In principle, the coupling between the upfield protons to tungsten and to each other could also be considered as one-bond and three-bond couplings, respectively, owing to the agostic interactions. However, such consideration would be misleading because true one-bond tungsten-hydrogen couplings in Cp'W(NO)-containing complexes are all >100 H z , 3 3 while three-bond ' H - ' H couplings across a transition-metal center are typically >7 H z . 3 3 a Furthermore, that the long-range ' H - ' H coupling could be due to W (the letter, not tungsten) coupling can also be ruled out, since such couplings are usually >3 Hz for cyclic systems and <0.5 Hz for acyclic ones. 3 3 b (a) Wenzel, T. T.; Bergman, R. G. J. Am. Chem. Soc. 1986,108, 4856. (b) For a comprehensive review, see: Barfield, M . ; Chakrabarti, B. Chem. Rev. 1969, 69, 757. (a) See chapters 3 and 4 of this thesis, (b) Legzdins, P.; Martin, J. T.; Oxley, J. C. Organometallics 1985, 4, 1263. Messerle, L . W.; Jennische, P.; Schrock, R. R.; Stucky, G. J. Am. Chem. Soc. 1980, 102, 6744. van der Heijden, H. ; Gal, A . W.; Pasman, P.; Orpen, A . G. Organometallics 1985, 4, 1847. Procopio, L . J.; Carroll, P. J.; Berry, D. H . J. Am. Chem. Soc. 1994,116, 177. 52 Fellmann, J. D.; Schrock, R. R.; Traficante, D. D. Organometallics 1982, 7, 481. Fellmann, J. D.; Schrock, R. R.; Rupprecht, G. A . J. Am. Chem. Soc. 1981,103, 5752. Schultz, A . J.; Brown, R. K . ; Williams, J. M.; Schrock, R. R. J. Am. Chem. Soc. 1981, 103, 169, and references therein. Dawoodi, Z.; Green, M . L. H. ; Mtetwa, V . S. B.; Prout, K . ; Schultz, A . J.; Williams, J. M . ; Koetzle, T. F. J. Chem. Soc. Dalton Trans. 1986, 1629. Etienne, M . ; Mathieu, R.; Donnadieu, B. J. Am. Chem. Soc. 1997, 779, 3218. Legzdins, P.; Jones, R. H . ; Phillips, E. C ; Yee, V . C.; Trotter, J.; Einstein, F. W. B. Organometallics 1991, 10, 1002. (a) Legzdins, P.; Rettig, S. J.; Sanchez, L. ; Bursten, B. E. ; Gatter, M . G. J. Am. Chem. Soc. 1985, 707, 1411. (b) Bursten, B. E.; Cayton, R. H . Organometallics 1987, 6, 2004. . Lauer, J. W.; Hoffmann, R. J. Am. Chem. Soc. 1976, 98, 1729. Green, M . L . H . In Organometallic Compounds., 3 r d ed; Methuen and Co. Ltd.: London, 1968, Vo l . 2, pp 90-164. Kubas, G. J. Acc. Chem. Res. 1988, 27, 120. Churchill, M . R.; Youngs, W. J. Inorg. Chem. 1979,18, 2454. (a) Kuchitsu, K . J. Chem. Phys. 1968, 49, 4456. (b) Bartell, M . S.; Kuchitsu, K . ; De Neui, R. J. J. Chem. Phys. 1965, 35, 1211. (a) Klooster, W. T.; Brammer, L. ; Schaverien, C. J.; Budzelaar, P. H . M . J. Am. Chem. Soc. 1999, 727, 1381. (b) Schaverien, C. J.; Budzelaar, P. H . M . , private communication. 53 For comparison, the agostic C H - W distance in W(CO)3(PCy3)2 is about 2.27 A 5 1 A and the agostic C H - M o distance in [Et 2 B(pz) 2 ][rf-CH 2 C(Ph)CH 2 ](CO) 2 Mo appears to be 2.27(8) A and possibly as short as about 2.15 A 5 1 B : (a) Wasserman, H . J.; Kubas, G. J.; Ryan, R. R. J. Am. Chem. Soc. 1986,108 , 2294. (b) Cotton, F. A . ; LaCour, T.; Stanislowski, A . G. J. Am. Chem. Soc. 1974, 96, ISA. Brown, R. K . ; Williams, J. M . ; Schultz, A . J.; Stucky, G. D.; Ittel, S. D.; Harlow, R. L. J. Am. Chem. Soc. 1980, 102, 981. 54 C H A P T E R 3 Thermolysis of Cp'W(NO)(CH2CMe3)2 and Cp*W(NO)(CH2CMe3)(CH2Ph) Complexes: Generation and Trapping Reactions of [Cp'W(NO)(=CHCMe3)] and [Cp*W(NO)(=CHPh)] Intermediates 3.1 Introduction 54 3.2 Experimental Section 57 3.3 Results and Discussion 71 3.4 Epilogue 96 3.5 References and Notes 100 3.1 Introduction The first transition-metal complexes to contain a direct metal-carbon double bond were prepared in the mid-1960s.1 Since then, thousands have been reported and, in general, tend to fall into one of three distinct classes based on their mode of reactivity. The first two classes are complexes called Fischer-type carbenes2 and Schrock-type alkylidenes,3 named after their discoverers. Fischer-type carbene complexes are mostly generated by alkylation of a metal-bound carbonyl ligand (eq 3.1), are electrophilic, and have found widespread utility in the cyclopropanation of alkenes.4 In addition, the metal in these complexes is usually in a low oxidation state, while the carbene carbon is often stabilized by a heteroatom. 1. M e L i / M ( C O ) 6 (CO) 5 M=C (3.1) 2. M e 3 0 + Me 55 Schrock-type alkylidene complexes, on the other hand, are nucleophilic and are prepared almost exclusively by a-hydrogen abstraction routes (eq 3.2).5 They are also key intermediates in numerous important stoichiometric and catalytic transformations. These include formation of metallacycles via 2+2 cycloaddition reactions with unsaturated organic substrates,3 alkene metathesis,6 ring-opening metathesis polymerization,7 acyclic diene8 and alkyne9 polymerizations, ring-closing metathesis,10 and carbonyl olefinations.11 C H 2 R L n M \ - R ' H R' A L n M = C H \ (3.2) R The difference in the chemical reactivity between Fischer carbene and Schrock alkylidene complexes can be attributed to a difference in the bonding within their metal-carbon double bonds.1 2 In Fischer carbene complexes, this bonding is a composite of electron donation from a doubly occupied orbital on the carbene ligand and a 71-back-donation from a filled metal d orbital. The M=C bond in alkylidene complexes, in contrast, is best considered in terms of an ethylene-like covalent interaction (i.e. one in which the electrons are nearly equally distributed between the metal fragment and the alkylidene moiety). Finally, the third class of M=C complexes is amphiphilic, reacting as both electrophiles and nucleophiles. Examples include (CO)2(Ph3P)2Ru(=CF2),13 which reacts with HC1 and CH3NH2 to give (CO) 2(Ph 3P)2Ru(CF 2H)Cl and (CO)2(Ph3P)2Ru(CNCH3), respectively, and Cp(CO)2Re(=CHCH 2CH 2CMe 3), 1 4 which reacts with HC1 and P M e 3 to produce Cp(CO)2Re(CH 2CH2CH 2CMe 3)Cl and zwitterionic Cp(CO) 2 Re[CH(PMe 3 )CH 2 CH 2 CMe 3 ] . The investigation into M=C complexes by the Legzdins group began in the early 1990s with the discovery that the molybdenum bis(alkyl) complex, CpMo(NO)(CH 2 CMe 3 ) 2 , decomposes at room temperature to give a bimetallic species with a unique bridging 7]':7]2-NO ligand (Scheme 3.1).1 5 The reaction presumably proceeds via a-elimination of neopentane and formation of the putative complex CpMo(NO)(=CHCMe 3). Although this 16-electron 56 intermediate exists only transiently, it can be trapped by coordination to various Lewis bases. It can also add heteroatom-hydrogen bonds to its Mo=C linkage in manner consistent with it being nucleophilic at carbon.1 6 Scheme 3.1 E = NR, O, S L CpMo(NO)(=CHCMe 3)(L) Subsequent to the discovery of the molybdenum neopentylidene system, anionic alkylidene complexes of the type [Cp*M(NO)(=CHSiMe 3)(CH 2SiMe3)]~M' + [M = Mo, W; M ' = Na, Li] have been prepared by deprotonation of an alkyl ligand of Cp*M(NO)(CH 2 SiMe 3 )2. 1 7 While these 18-electron compounds exhibit interesting structures, their chemistry, unfortunately, has been limited to only protonation reactions that lead to the starting material. As a continuation of the development of unsaturated alkylidene complexes containing the Cp 'M(NO) fragment, the thermal decompositions of a series of tungsten bis(alkyl) 57 complexes of the type Cp'W(NO)(CH 2 CMe 3 )(R), where Cp' = Cp or Cp* and R = C H 2 C M e 3 , C H 2 C M e 2 E t , C H 2 C M e 2 P h , C H 2 S i M e 3 , C H 3 , Ph, or CH 2 Ph, have been investigated. O f these compounds, the neopentyl complexes Cp 'W(N0) (CH 2 CMe 3 ) 2 and Cp*W(N0)(CH 2 CMe 3 ) (CH 2 Ph) have been found to undergo a-abstraction to generate neopentane and the transient alkylidene complexes [Cp'W(NO)(=CHCMe 3)] and [Cp*W(NO)(=CHPh)], respectively. The decompositions of the remaining complexes, in contrast, are complicated by alternative and/or competing reactions, each of which wil l be discussed in subsequent chapters of this thesis. In the present chapter several aspects of the chemistry of the above unsaturated tungsten(II) alkylidene complexes, which have allowed the similarities and differences between the molybdenum and tungsten systems and the Cp and Cp* systems to be established, are described. These include (1) the trapping reactions of these species with Lewis bases, (2) the reactions of the neopentylidene species and their PMe 3 adducts with substrates containing heteroatom-hydrogen bonds, and (3) the reactions of the Cp* neopentylidene intermediate with cyclic and acyclic alkenes, which in an unprecedented manner lead not only to novel 2+2 cycloaddition complexes but also to a wide range of structurally diverse C - H bond-activation products. The ability of the neopentylidene species [Cp*W(NO)(=CHCMe 3)] to activate alkane C - H bonds has recently been reported,18 full details of this chemistry are described in Chapters 4 and 5. 3.2 Experimental Section 3.2.1 Methods The synthetic methodologies employed throughout this thesis are described in detail in section 2.2.1. G C - M S analyses were performed by Ms. L. Madilao of the U B C mass spectrometry facility using a Carlo Erba G C and Kratos MS-80 unit. Unless otherwise noted all thermolyses were carried out in thick-walled glass vessels or J. Young N M R tubes immersed in an oil bath at 60-70 °C for ca. 1.5-2 d. 58 3.2.2 Reagents The complexes CpW(NO)(CH 2 CMe 3)2 (2.1), Cp*W(NO)(CH 2 CMe 3 ) 2 (2.4), and Cp*W(NO)(CH 2 CMe 3 )(CH 2 Ph) (2.15) were prepared as described in Section 2.2.1. PMe 3 , PMe 2 Ph, PEt 3 , P(OMe) 3 , EtNMe 2 , E t 3 N, neohexene, 2-methyl-1-butene, 2-methyl-2-butene, cyclopentene (Aldrich), ethylcyclopentene (Wiley Organics), and norbornadiene (Eastman Kodak) were dried over sodium and vacuum-transferred prior to use. Acetonitrile and tert-butylnitrile (Aldrich) were distilled from CaH 2 . Pyridine (BDH) and methanol (Aldrich) were dried over activated 4-A molecular sieves. M e 2 N H (99.9%, Aldrich) was used as received. Benzoic acid (Fisher) was recrystallized from hexanes, and then dried under vacuum at room temperature. 3.2.3 Preparation of Cp'W(NO)(=CHCMe3)(L) Complexes [Cp' = Cp, L = PMe 3 (3.1), PEt 3 (3.2), P(OMe) 3 (3.3); Cp' = Cp*, L = PMe 3 (3.4), PMe 2Ph (3.5), PEt 3 (3.6), P(OMe)3 (3.7)] Complexes 3.1-3.7 were prepared in a similar manner using the appropriate Lewis base. A n exception is that the thermolyses for the preparation of the Cp complexes were carried out at ca. 65 °C for 1 d, while those for the Cp* series were effected at ca. 75 °C for 2 d. The preparation of 3.4 is described as a representative example. To a Pyrex bomb containing 2.4 (172 mg, 0.350 mmol) were added sequentially THF (2 mL) and P M e 3 (>10 equiv) by vacuum transfer. The resulting wine-red mixture was heated at 75 °C for 2 d, during which time it became a yellow-brown solution. Next, the organic volatiles were removed in vacuo, and the residue was extracted with 4:1 hexanes/Et20 (10 mL). The extracts were filtered through Celite ( l x l cm), which was washed with additional 4:1 hexanes/Et20 until the filtrate was colorless. Diminishing the volume of the combined filtrates in vacuo, followed by cooling to -30 °C for 2 d afforded 3.4 (147 mg, 85 % yield) as yellow plates. 3.2.3.1 Complex 3.1 was isolated as dark yellow blocks (93% yield) from E t 2 0 at -30 °C. 59 Anal. Calcd. for C i 3 H 2 4 N O P W : C, 36.72; H , 5.69; N , 3.29. Found: C, 36.89; H , 5.84; N , 3.19. IR(cm"')VNO 1520(s). M S 425 [ M + , 1 8 4 W ] . ' H N M R (300 M H z , C 6 D 6 ) 5 1.09 (d, V H P = 9.6, 9H, PMe 3 ) , 1.43 (s, 9H, CMe 3 ) , 5.26 (d, JHP = 1.2, 5H, C 5 H 5 ) . 1 3 C{ 'H} N M R (75 M H z , C 6 D 6 ) 5 21.0 (d, lJCP = 34, PMe 3 ) , 32.1 (CMe 3), 52.7 (CMe 3 ) , 96.4 (C 5 H 5 ) . 3.2.3.2 Complex 3.2 was isolated as pale yellow clusters (57% yield) from 1:5 hexanes/Et20 a t - 3 0 ° C . Anal. Calcd. for C i 6 H 3 0 N O P W : C, 41.13; H , 6.47; N , 3.00. Found: C, 41.28; H , 6.41; N , 3.01. jR(cm"')VNO 1514(s). M S 467 [ M + , 1 8 4 W ] . ' H N M R (300 M H z , CDC1 3) 5 1.02 (m, 9H, CH2G1Y3), 1.12 (s, 9H, CMe 3 ) , 1.6-1.9 (m, 6H, Ci7 2 CH 3 ) , 5.62 (d, JHP = 0.9, 5H, C 5 H 5 ) . ' ^ { ' H } N M R (75 M H z , CDC1 3) 5 7.9 (CH 2 CH 3 ) , 21.6 (d, XJC? = 31, C H 2 C H 3 ) , 31.6 (CMe 3), 52.4 (CMe 3 ) , 96.5 (C 5 H 5 ) . 3.2.3.3 Complex 3.3. The formation of this complex in THF was not clean. A black precipitate also formed, which was removed by filtration through alumina I(1><1 cm). The filtrate was then concentrated in vacuo to obtain relatively pure 3.3 as judged by ' H N M R spectroscopy. ' H N M R (300 M H z , CDC1 3) 5 1.14 (s, 9H, CMe 3 ) , 3.57 (d, V H P = 12.0, 9H, OCH 3 ) , 5.70 (d, JHp = 1.2, 5H, C 5 H 5 ) . 1 3 C{ 'H} N M R (75 M H z , CDC1 3) 5 30.9 (CMe 3), 52.0 (CMe 3 ), 52.3 (OCH3), 96.8 (C5H5). 3.2.3.4 Complex 3.4. Anal. Calcd. for C , 8 H 3 4 N O P W : C, 43.65; H , 6.92; N , 2.83. Found: C, 43.62; H , 7.00; N , 2.78. IR (cm"1) v N 0 1513 (s). M S 495 [ M + , 1 8 4 W]. ' H N M R (300 M H z , C 6 D 6 ) 5 1.06 (d, 2 J H P = 9.0, 9H, PMe 3 ) , 1.43 (s, 9H, CMe 3 ) , 1.87 (s, 15H, C 5 Me 5 ) . 1 3 C{ 'H} N M R (75 M H z , C 6 D 6 ) 8 11.0 (C5Me5), 19.7 (d, ]JCP = 32, PMe 3 ) , 32.0 (CMe 3), 52.2 (CMe 3 ), 106.3 (C 5 Me 5 ) . 60 3.2.3.5 Complex 3.5 is highly soluble in hydrocarbon solvents, but can be isolated as a tan solid by storing a concentrated pentane/Et20 mixture at -30 °C for 1-2 months. The purity of this material, however, was ca. 90% at best. IR (cm"1) 1509 (s). M S 557 [ M + , 1 8 4 W ] . ' H N M R (300 M H z , CD 2 C1 2 ) 5 1.15 (s, 9H, CMe 3 ) , 1.82 (s, 15H, C 5 Me 5 ) , 1.72 (s, 3H, PMe 2 ) , 1.75 (s, 3H, PMe 2 ) , 7.4 (m, 5H, Ph). I 3 C { ' H } N M R (75 M H z , CD 2 C1 2 ) 8 10.6 (C5Me5), 19.9 (PMeMe), 20.4 (PMeMe), 31.5 (CMe 3), 32.6 (CMe 3 ), 107.0 (C 5 Me 5 ) , 128.8 (d, JCP = 10, C o / m ) , 130.2 (C p ), 130.9 (d, Jc? =11, C o / m ) , 136.9 (CipSo). 3.2.3.6 Complex 3.6 was isolated as pale yellow blocks (74% yield) from a 3:1 mixture of hexanes/Et20 at -30 °C. Anal. Calcd. for C 2 ,H 4 oNOPW: C, 46.94; H , 7.50; N , 2.61. Found: C, 47.34; H , 7.50; N,2.81. IR(cm"')VNO 1527 (s). M S 537 [ M + , 1 8 4 W ] . ' H N M R (500 M H z , CDC1 3) 5 0.76 (pent, VHH = 7.6, V H P = 9.0, 9H, C H 2 C / / 3 ) , 1.43 (m, 6H, Gtf 2 CH 3 ) , 1.87 (s, 15H, C 5 Me 5 ) . 1 3 C{ 'H} N M R (75 M H z , CDC1 3) 5 7.6 ( C H 2 C H 3 ) , 10.9 (C5Me5), 19.0 (d, V C P = 29, C H 2 C H 3 ) , 31.7 (CMe 3), 52.2 (CMe 3 ), 106.3 (C 5 Me 5 ) . 3.2.3.7 Complex 3.7 was isolated as pale yellow plates (51% yield) from a 3:1 mixture of hexanes/Et20 at -30 °C. Anal. Calcd. for C i 8 H 3 4 N 0 4 P W : C, 39.79; H , 6.31; N , 2.58. Found: C, 39.67; H , 6.33; N , 2.53. IR (cm"1) v N 0 1527 (s). M S 543 [ M + , 1 8 4 W ] . l H N M R (300 M H z , CDC1 3) 5 1.12 (s, 9H, CMe 3 ) , 2.02 (s, 15H, C 5 Me 5 ) , 3.50 (d, V H P = 11.7, 9H, P(OMe) 3). 1 3 C{ 'H} N M R (75 M H z , CDC1 3) 8 10.6 (C5Me5), 31.1 (CMe 3), 51.8 (OMe), 52.6 (CMe 3 ) , 107.3 (C 5 Me 5 ) . 61 3.2.4 Preparation of Cp*W(NO)(=CHCMe3)(NEtMe2) (3.8) A wine-red solution of 2.4 (62 mg, 0.13 mmol) in EtNMe 2 (2 mL) was heated in the dark at 70 °C for 2 d, during which time it became brown-yellow. In the presence of room light, the organic volatiles were removed in vacuo to obtain a brown oil which was redissolved in hexanes (2 mL) and filtered through Celite ( l x l cm). The filtrate was concentrated and then cooled at -30 °C for several days to induce the deposition of 3.8 as dark yellow crystals (22 mg, 35% yield). Anal. Calcd. for C i 9 H 3 6 N 2 O W : C, 46.35; H , 7.37; N , 5.69. Found: C, 46.08; H , 7.44; N , 5.66. IR(cm"')VNO 1512 (s). MS 492 [ M + , 1 8 4 W ] . ' H N M R (300 M H z , C 6 D 6 ) 5 0.75 (t, VHH = 7.4, 3H, CU2CH3), 1.55 (s, 9H, CMe 3 ) , 1.80 (s, 15H, C 5 Me 5 ) , 2.00 (m, 2H, Gtf 2 CH 3 ) , 2.30 (s, 3H, NMe), 2.40 (s, 3H, NMe). 1 3 C{ 'H} N M R (75 M H z , C 6 D 6 ) 8 10.8 (C 5 Me 5 ) , 11.3 (CH 2 CH 3 ) , 33.0 (CMe 3 ) , 33.9 (CMe 3), 55.8 (NMe), 56.9 (NMe), 60.7 (CH 2 CH 3 ) , 106.8 (C 5 Me 5 ) . 3.2.5 Preparation of Cp*W(NO)(=CHCMe3)(Py) (3.9) In a manner similar to the preparation of 3.8, a solution of 2.4 (56 mg, 0.11 mmol) in pyridine (3 mL) was heated in the dark at 70 °C for 2 d, during which time it became orange-brown. In the presence of light, the organic volatiles were removed in vacuo to obtain a brown oil which was extracted into E t 2 0 (1 mL) and then cooled to -30 °C. Orange-brown crystals of 3.9 formed after 1 d (15 mg, 27% yield) and were isolated by removing the mother liquor with a pipet. Anal. Calcd. for C 2 0 H 3 0 N 2 O W : C, 48.20; H , 6.07; N , 5.62. Found: C, 48.13; H , 5.97; N , 5.47. I R t c m - V N O 1551 (s). M S 498 [ M + , 1 8 4 W ] . ' H N M R (300 M H z , C 6 D 6 ) 5 1.63 (s, 9H, CMe 3 ) , 1.79 (s, 15H, C 5 Me 5 ) , 6.08 (pseudo t, V C H = 6.4, 2H, H m ) , 6.48 (t, VHH = 8.5,1H, Hp), 8.29 (d, V H H = 6.5, 2H, H 0 ) . I 3 C{ 'H} N M R (75 M H z , C 6 D 6 ) 5 10.4 (C 5 Me 5 ) , 33.9 (CMe 3), 49.0 (CMe 3 ), 107.3 (C 5 Me 5 ) , 124.6 (C o / m ) , 136.3 (C p), 156.1 (Co/m). 62 3.2.6 Preparation of Cp*W(NO)(=CHPh)(PMe3) (3.10) Complex 3.10 was prepared from Cp*W(NO)(CH 2 CMe 3 )(CH 2 Ph) in a manner analogous to that described for the preparation of Cp*W(NO)(=CHCMe3)(PMe3). It was crystallized from 5:1 E t 2 0 /THF at -30 °C and isolated as brown-orange rods (65% yield, 2 crops). Anal. Calcd. for C20H30NOPW: C, 46.62; H , 5.87; N , 2.72. Found: C, 46.42; H , 5.69; N , 2.95. M S 515 [ M + , 1 8 4 W ] . ' H N M R (400 M H z , CD 2 C1 2 ) 5 1.34 (d, 2JHP = 8.7, 9H, PMe 3 ) , 2.12 (s, 15H, C 5 Me 5 ) , 7.03 (t, V H H = 7.3, IH, H p ) , 7.22 (t, V H H = 7.6, 2H, H m ) , 7.56 (d, V H H = 7.6, 2H, H 0 ) . 1 3 C{ 'H} N M R (75 M H z , CD 2 C1 2 ) 5 11.1 (C5Me5), 17.6 (d, lJCP = 34, PMe 3 ) , 108.4 (C 5 Me 5 ) , 125.5 (C p), 128.4 (C m ) , 128.8 (C 0), 154.1 (d, V C P = 4, C i p s o ) . 3.2.7 Attempted Trapping of [Cp*W(NO)(=CHCMe3)] with THF, E t 2 0 , E t 3 N MeCN, and M e 3 C C N Each of these reactions was carried out in a manner similar to the preparation of 3.8 and 3.9 using the appropriate trapping agent. In all cases the ' H N M R spectrum of the crude reaction mixture revealed either decomposition or a plethora of intractable non-alkylidene products. 3.2.8 Attempted Preparation of Cp*W(NO)(=CHCMe3)(PMe3) (3.4) in CH 2 C1 2 In a manner similar to the preparation of 3.4 in THF, a dichloromethane solution (2 mL) of 2.4 (48 mg, 0.098 mmol) and excess PMe 3 was heated at 70 °C for 2 d, during which time a pale green-yellow precipitate formed and the reaction solution became orange. The organic volatiles were then removed in vacuo to obtain a residue, which upon addition of E t 2 0 gave a pale yellow solid and an orange solution. The solid (presumably a complex mixture of phosphonium chlorides based on F A B mass spectrometry) was removed by filtration. 63 Concentration of the filtrate followed by cooling to -30 °C for 2 d provided Cp*W(NO)(PMe 3 )Cl 2 (3.11) (12 mg, 25% yield) as brown-orange needles. M S 420 [ M + - P M e 3 , 1 8 4 W ] . ' H N M R (300 M H z , C 6 D 6 ) 8 1.63 (d, 2JHp = 10.5, 9H, PMe 3 ) , 2.03 (s, 15H, C 5 Me 5 ) . l 3 C { ' H } N M R (75 M H z , C 6 D 6 ) 5 10.5 (C5Me5), 13.4 (d, lJCP = 34,PMe 3 ) , 113.2(C 5Me 5). 3 1 P{ 'H} N M R (121 M H z , C 6 D 6 ) 8 12.0 ( ' j P W = 251). 3.2.9 Preparation of C p ' W ( N O ) ( C H 2 C M e 3 ) ( N M e 2 ) [Cp' = C p * (2.17); C p ' = Cp (3.12)] 3.2.9.1 Complex 2.17. To a thick-walled Pyrex bomb containing 2.4 (47 mg, 0.096 mmol) was added THF (10 mL) and M e 2 N H (3-5 equiv from a calibrated gas bulb) by vacuum transfer. The resulting mixture was thawed and heated at 70 °C for 2 d, during which time its color darkened. The organic volatiles were then removed in vacuo to obtain a residue, which was extracted into 2:1 hexanes/Et20 and filtered through Celite (1 x 1 cm). Concentration of the filtrate followed by cooling to -30 °C for 2 d provided 2.17 as orange-yellow needles (43 mg, 91% yield). Anal. Calcd. for C i 7 H 3 2 N 2 O W : C, 43.97; H , 6.95; N , 6.03. Found: C, 43.89; H , 7.09; N , 5.91. IR (cm"1) v N 0 1566 (s). M S 464 [ M + , 1 8 4 W ] . *H N M R (500 M H z , C 6 D 6 ) 8 0.94 (AB q, 2JHH = 13.5, 2H, W C H 2 ) , 1.37 (s, 9H, CMe 3 ) , 1.65 (s, 15H, C 5 Me 5 ) , 2.59 (s, 3H, NMe), 3.71 (s, 3H, NMe). 1 3 C{ 'H} N M R (125 M H z , C 6 D 6 ) 8 9.52 (C 5 Me 5 ) , 34.5 (CMe 3), 42.2 (CMe 3 ), 51.3 (NMe), 57.7 (WCH 2 ) , 59.8 (NMe), 109.8 (C 5 Me 5 ) . 3.2.9.2 Complex 3.12 (yellow-orange blocks, 93% yield) was prepared from 2.1 in a manner similar to that described for 2.17 except that the thermolysis was effected at 65 °C for 1 d. Anal. Calcd. for C 1 2 H 2 2 N 2 O W : C, 36.56; H , 5.63; N , 7,11. Found: C, 36.85; H , 5.77; N , 7.17. IR (cm 4) v N O 1602 (s). M S 394 [ M + , 1 8 4 W ] . ' H N M R (300 M H z , C 6 D 6 ) 8 1.32 (s, 9H, CMe 3 ) , 1.50 (AB q, V H H = 12.7, 2H, W C H 2 ) , 2.62 (NMe), 3.68 (NMe), 5.14 (C 5 H 5 ) . 64 1 3 C{ 'H} N M R (75 M H z , C 6 D 6 ) 8 34.3 (CAfe3), 37.4 (CMe 3 ), 47.5 (WCH 2 ) , 56.0 (NMe), 59.5 (NMe), 102.5 (C 5 H 5 ) . 3.2.10 Preparation of CpW(NO)(CHDCMe 3)(NMe 2) (3.12-rf) This compound was prepared from CpW(NO)(CD 2 CMe 3 ) 2 and M e 2 N H in a manner analogous to that described for 3.12. The procedure afforded 3.12-rf in >95% yield as judged by ' H N M R spectroscopy. ' H N M R (300 M H z , C 6 D 6 ) 8 1.32 (s, 9H, CMe 3 ) , 1.50 (br s, 1H, W C H D ) , 2.62 (NMe), 3.68 (NMe), 5.14 (C 5 H 5 ) . 1 3 C{ 'H} N M R (75 M H z , C 6 D 6 ) 8 34.3 (CMe 3), 37.4 (CMe 3 ) , 46.9 (t, 'JCD = 18.5, WCHD) , 56.0 (NMe), 59.6 (NMe), 102.5 (C 5 H 5 ) . 3.2.11 Reactions of Cp*W(NO)(=CHCMe3)(PMe3) (3.4) with MeOH, Me 2 NH, and P h C 0 2 H 3.2.11.1 Preparation of Cp*W(NO)(CH 2CMe 3)(OMe) (2.16). To a sample of 3.4 (10 mg, 0.020 mmol) in an N M R tube was added methanol by vacuum transfer. The resulting solution was left at room temperature for 6 h, whereupon the once yellow solution appeared red. The organic volatiles were then removed in vacuo, and an N M R sample in CeD6 was prepared. The ' H N M R spectrum of the crude reaction mixture revealed the quantitative conversion of 3.4 into 2.16. ' H N M R (500 M H z , C 6 D 6 ) 8 1.33 (AB q, 2 J H H = 14.3, 2H, W C H 2 ) , 1.41 (s, 9H, CMe 3 ) , 1.57 (s, 15H, C 5 Me 5 ) , 4.75 (s, 3H, OMe). 1 3 C{ 'H} N M R (75 M H z , C 6 D 6 ) 8 9.1 (C 5 Me 5 ) , 32.8 (WCH 2 ) , 34.0 (CMe 3), 36.8 (CMe 3 ), 70.0 (OMe), 111.6 (C 5 Me 5 ) . 3.2.11.2 Preparation of Cp*W(NO)(CH 2CMe 3)(NMe 2) (2.17). A solution of 3.4 (12 mg, 0.024 mmol) in THF (0.5 mL) was treated with excess M e 2 N H by vacuum transfer. The resulting solution was heated at 70 °C for 1 d, during which time it slowly turned yellow-65 orange. The organic volatiles were then removed in vacuo. The ' H N M R spectrum of the crude reaction mixture in C6D 6 revealed the quantitative conversion of 3.4 into 2.17. 3.2.11.3 Preparation of Cp*W(NO)(CH 2CMe 3)(0 2CPh) (3.13). To an N M R tube containing 3.4 (17 mg, 0.034 mmol) and benzoic acid (4 mg, 1 equiv) was added CeD6 (0.5 mL) by vacuum transfer. Upon thawing, the color of the mixture immediately changed from yellow to orange. The ' H N M R spectrum indicated clean conversion to 3.13, whose identity was confirmed by comparison of its N M R spectra to those of an authentic sample.1 9 3.2.12 Preparation of Cp*W(NO)[CH(CMe 3)CH 2(CMe 3)CH] (3.14) To a Pyrex bomb containing 2.4 (909 mg, 2.16 mmol) was added neohexene (3 mL) by vacuum transfer. The mixture was heated at 70 °C for 2 d, during which time it became orange-yellow. The organic volatiles were then removed in vacuo to obtain a yellow solid, which was redissolved in 4:1 hexanes/Et20 (1 mL) and cooled to -30 °C. Yellow needles of 3.14 (903 mg, 97% yield in 2 crops), which were suitable for an X-ray diffraction study, formed after 2 d and were isolated by removing the mother liquor with a pipet. Anal. Calcd. for C 2 1 H 3 1 N O W : C, 50.10; H , 7.41; N , 2.78. Found: C, 49.87; H , 7.40; N , 2.76. IR (cm"1) v N 0 1518 (s). M S 503 [ M + , 1 8 4 W ] . ' H N M R (300 M H z , CDC1 3) 8 -2.29 (td, 2JHH = 4.5, V H H = 10.2, IH, Hp syn to Cp*), -1.01 (pseudo q, 2JHH = 4.5, V H H = 6.0, IH, Hp anti to Cp*), 1.09 (s, 18H, CMe 3 ) , 1.91 (s, 15H, C 5 Me 5 ) , 3.92 (dd, V H H = 6.0, 10.2, 2H, WCH) . 1 3 C{ 'H} N M R (75 M H z , CDC1 3) 8 -4.8 (CH 2 ), 10.3 (C 5 Me 5 ) , 32.2 (CMe 3), 43.2 (s, CMe 3 ) , 108.8 (C 5 Me 5 ) , 124.2 ('yCw = 124, WCH) . NOEDS (400 M H z , CDC1 3) irrad. at 1.91 ppm (Cp*), N O E at 8 3.92 (H«); irrad. at 3.92 ppm (H a ) , NOEs at 8 -2.29 (Hp), 1.22 (CMe 3 ), 1.65 (Cp*). 66 3.2.13 Preparation of Cp*W(NO)[CH(CMe 3)CH(C 3H 6)CH] (3.15) A solution of 2.4 (68 mg, 0.13 mmol) in cyclopentene (4 mL) was heated at 70 °C for 2 d, during which time it became orange-yellow. The organic volatiles were then collected by vacuum transfer and analyzed by G C - M S , which showed, in addition to neopentane (m/z 57) and the solvent (m/z 70), the presence of 3-cyclopentylcyclopentene (m/z 136) and 1,1'-bicyclopentyl (m/z 138). The identity of the latter two compounds was confirmed by comparison of their mass spectral fragmentation patterns with those of authentic samples.20 Their yields, unfortunately, could not be estimated due to partial peak overlap in the G C trace. Extraction of the crude reaction mixture into E t 2 0 , followed by filtration through Celite ( l x l cm) and cooling to -30 °C for several days provided 3.15 as orange crystals (30 mg, 45% yield). Anal. Calcd. for C20H33NOW: C, 49.29; H , 6.83; N , 2.87. Found: C, 49.59; H , 7.03; N , 2.89. M S 485 [ M + , 1 8 4 W ] . ' H N M R (500 M H z , CDC1 3) 5-0.37 (septet, V H H = 5.3, 10.2, IH, Hp), 0.38 (m, IH, CH 2 ) , 1.00 (s, 9H, CMe 3 ) , 1.75 (m, IH , C H 2 ) , 2.00 (s, 15H, C 5 Me 5 ) , 2.04 (m, IH , CH 2 ) , 2.32 (m, IH, CH 2 ) , 2.46 (m, IH , CH 2 ) , 2.61 (m, IH , CH 2 ) , 3.29 (d, 3 J H H = 5.1, IH , Gr7CMe 3), 7.68 (m, IH, WCH). 1 3 C{ 'H} N M R (75 M H z , CDC1 3) 5 10.4 (C 5 M? 5 ) , 15.5 (p CH), 32.1 (CMe 3), 33.4 (CH 2 ) , 37.2 (CH 2 ) , 40.6 (CH 2 ) , 41.2 (CMe 3 ) , 109.5 (C 5 Me 5 ) , 115.3 (CHCMe 3 ) , 141.8 (WCH). 3.2.14 Preparation of Cp*W(NO)[CH(CMe 3)CH(C 3H 6)C(Et)] (3.16) A solution of 2.4 (33 mg, 0.049 mmol) in ethylcyclopentene (4 mL) was heated at 75 °C for 2 d, during which time it became orange-yellow. The organic volatiles were then collected by vacuum transfer and analyzed by G C - M S , which showed in addition to neopentane and the solvent (m/z 96) the presence of a complex mixture of unidentifiable products with parent peaks at m/z 98, 134, 136, 164, and 166. Examination of the crude reaction mixture by *H N M R spectroscopy indicated the presence of 3.16 and a trace amount of decomposition products. This mixture was then triturated with pentane ( 3 x 5 mL) and 67 filtered through alumina 1 (1x1 cm). The alumina column was washed with additional 5:1 pentane/Et20 until the filtrate was colorless. Diminishing the volume of the combined filtrates under reduced pressure followed by cooling to -30 °C for several days afforded 3.16 as orange-yellow crystals (6 mg, 17% yield). MS515 [ M + , 1 8 4 W ] . ' H N M R (500 M H z , CDC1 3) 8-0.62 (quint, V H H = 5.4, 10.6, 1H, Hp), 0.47 (m, 1H, C a H 2 ) , 1.01 (s, 9H, CMe 3 ) , 1.11 (t, V H H = 7.2, 3H, C H 2 C / / 3 ) , 1.76 (m, VHH = 5.4, 11.7, 1H, C b H 2 ) , 1.98 (s, 15H, C 5 Me 5 ) , 2.13 (m, 1H, C C H 2 ) , 2.27 (m, 1H, C C H 2 ) , 2.35 (m, 2H, C a H 2 , C b H 2 ) , 2.48 (m, 1H, C d H 2 ) , 3.08 (m, 1H, C d H 2 ) , 3.12 (d, V H H = 5.4, 1H, WCH). 1 3 C{ 'H} N M R (75 M H z , CDC1 3) 8 10.3 (C5Me5), 17.0 (CH 2 CH 3 ) , 27.4 ((3 CH), 32.1 (CMe 3), 33.4 (C a H 2 ) , 37.9 (C b H 2 ) , 40.9 (CMe 3), 42.9 (C C H 2 ) , 46.9 (C d H 2 ) , 109.2 (C 5 Me 5 ) , 113.2 (WCH), 173.2 (WC). NOEDS (400 M H z , CDC1 3): irrad. at 5 1.01, NOEs at -0.62 and 3.12 ppm. 3.2.15 Preparation of Cp*W(NO)[CH(CMe 3 )CH(C 5 H 6 )CH] (3.17) A solution of 2.4 (24 mg, 0.049 mmol) in norbomadiene (~2 mL) was heated at 70 °C for 2 d, during which time it became pale orange. The organic volatiles were then removed in vacuo, leaving an insoluble white material (presumably products of the ring-opening polymerization of norbomadiene) and an orange crystalline solid. The latter was extracted into E t 2 0 (3 mL) and filtered through Celite ( l x l cm). Concentration of the filtrate followed by cooling to -30 °C for several days provided 3.17 as orange blocks. These blocks, which were suitable for an X-ray crystallographic analysis, were obtained in 32% yield (8 mg). M S 511 [ M + , 1 8 4 W ] . ' H N M R (500 M H z , C 6 D 6 ) 8-0.12 (m, JHH = 2.3,6.6, 1H, Hp), 0.02 (d, J H H = 8.9, 1H, CH 2 ) , 1.25 (m, J H H = 8.8, 2.6, 1H, CH 2 ) , 2.38 (s, 1H, CH), 2.78 (s, 1H, CH), 3.09 (d, VHH = 6.2, 1H, Gr7CMe 3), 5.94 (dd, JHH = 3.0, 5.2, 1H, =CH), 6.18 (dd, JHH = 3.1, 5.5, 1H, =CH), 6.94 (dd, 7HH = 2.9, 7.2, 1H, WCH). 1 3 C{ 'H} N M R (125 M H z , C 6 D 6 ) 8 10.4 (C 5 Me 5 ) , 18.0 ((3 CH), 32.9 (CAfe3), 34.1 (CMe 3 ), 43.6 (CH), 45.3 (CH 2 ) , 47.5 (CH), 108.9 (C 5 Me 5 ) , 122.1 (CHCMe 3 ) , 131.6 (WCH), 135.8 (=CH), 142.3 (=CH). 68 3.2.16 Preparation of Cp*W(NO)(CH2CMe3)(7r1-CH2C(Et)CH2) (3.18), Cp*W(NO)(H)(n 3-CH 2C(Et)CH 2) (3.19) and Cp*W(NO)(rj4-2-methylbutadiene) (3.20). In a manner similar to the preparation of 3.14, a solution of 2.4 (58 mg, 0.118 mmol) in 2-methyl-l-butene (2 mL) was heated at 70 °C for 2 d, during which time it became yellow-brown. The organic volatiles were then collected by vacuum transfer and were examined by G C - M S , which showed the presence of neopentane (mlz 57), the solvent (mlz 70), and a trace amount of an unidentifiable material whose mass is equivalent to two C5H10 units. Analysis of the crude reaction mixture by ' H N M R spectroscopy indicated that 3.18, 3.19, and 3.20 (in ~1.2:1:0.4 ratio) were formed in a combined yield of 60-75%; the remaining products could not be ascertained due to inadequate signal-to-noise of their signals and to spectral overlap. Next, the N M R mixture was taken to dryness, extracted into pentane, and filtered through alumina I (1 x 1 cm). The alumina column was washed with additional 4:1 pentane/Et20 until the filtrate was colorless. Concentration of the combined filtrates followed by cooling to -30 °C for several days afforded 3.18 (13 mg, 22% yield) as sticky white rosettes. The pale yellow mother liquor was then reduced in volume and stored at -30 °C for 2 d to induce the deposition of 3.18, 3.19, and 3.20 as yellow crystals in a ratio of -1:2:1. 3.2.16.1 Complex 3.18. Anal. Calcd. for C 2 0 H 3 7 N O W : C, 49.09; H , 7.21; N , 2.86. Found: C, 48.96; H , 7.14; N , 2.86. IR (cm"1) v N 0 1565 (s). M S 489 [ M + , 1 8 4 W]. ' H N M R (500 M H z , CDCI3) 5 0.75 (d, J H H = 2.7, 1H, C a H 2 ) , 0.96 (s, 9H, CMe 3 ) , 1.03 (d, V H H = 12.8, 1H, W C H 2 ) , 1.23 (t, V H H = 7.7, 3H, CH 2 Gr7 3 ), 1.60 (d, V H H = 12.8, 1H, W C H 2 ) , 1.81 (s, 15H, C 5 Me 5 ) , 1.89 (br s, 1H, C b H 2 ) , 2.22 (d, JHH = 2.7, 1H, C a H 2 ) , 2.25 (m, 2H, C# 2 CH 3 ) , 3.64 (d, JHH = 3.9, 1H, C b H 2 ) . 1 3 C{ 'H} N M R (75 M H z , CDCI3) 8 10.2 (C5Me5), 17.6 ( C H 2 C H 3 ) , 27.6 (WCH 2 ) , 29.5 (CH2CH3), 34.4 (CMe 3), 37.2 (CMe 3 ), 42.5 (C a H 2 ) , 74.2 (C b H 2 ) , 107.2 (C 5 Me 5 ) , 135.6 (4° C). NOEDS (400 M H z , C 6 D 6 ) irrad. at 8 0.75, NOEs at 1.81, 2.22 ppm; irrad. at 8 1.23, NOEs at 2.22, 2.25 ppm; irrad. at 8 3.64, NOEs at 0.96, 1.60, 1.89 ppm. 69 3.2.16.2 Complexes 3.19 and 3.20. IR (cm"1) v N O 1590 (s), 1565 (s, br); v ™ 1888 (w). M S 3.19: 416 [ M + - H 2 , 1 8 4 W ] . ' H N M R (500 M H z , C 6 D 6 ) 8 3.19: -0.88 (s, 'y H w =118, IH, WH), 0.45 (s, IH , C a H 2 ) , 0.75 (s, IH, C b H 2 ) , 1.09 (t, V H H = 7.5, 3H, CH 2 Gr7 3 ) , 1.78 (s, 15H, C 5 Me 5 ) , 2.20 (m, IH, C# 2 CH 3 ) , 2.50 (m, IH, C/7 2 CH 3 ) , 2.96 (br s, IH, C a H 2 ) , 4.15 (br s, IH, C b H 2 ) . 3.20: 0.95 (dd, 2 J H H = 4 . 0 , 3 J H H = 0.9, IH, C C H 2 ) , 1.1 (obscured by Gr7 2 CH 3 of 3.19, IH , C d H) , 1.75 (s, 15H, C 5 Me 5 ) , 2.04 (s, 3H, CH 3 ) , 2.3 (obscured by C r Y 2 C H 3 of 3.19, IH, C e H 2 ) , 3.14 (dd, 2 J H H = 4.0, V H H = 14.0, IH, C e H 2 ) , 3.29 (dd, 2 J H H = 4 . 0 , V H H = 0.7, IH, C C H 2 ) . 1 3 C{ 'H} N M R (75 M H z , C 6 D 6 ) 5 3.19: 10.7 (C 5 Me 5 ) , 14.9 (CH 2 CH 3 ) , 27.2 (WCH 2 ) , 31.0 (CH 2 CH 3 ) , 35.0 (CMe 3), 37.6 (CMe 3 ), 46.2 (C a H 2 ) , 53.0 (C b H 2 ) , 104.7 (C 5 Me 5 ) , 124.6 (4° C). 3.20: 9.9 (C5Me5), 18.8 (CH 3 ) , 50.7 (C C H 2 ) , 55.5 (C e H 2 ) , 80.5 (C d H), 106.5 (C 5 Me 5 ) , 115.8 (4° C). 3.2.17 Preparation of Cp*W(NO)(773-C10H,g) (3.21) A solution of 2.4 (73 mg, 0.149 mmol) in 2-methyl-2-butene (2 mL) was heated at 70 °C for 2 d, during which time it became yellow-orange. The organic volatiles were then collected by vacuum transfer and were analyzed by G C - M S , which showed peaks due to neopentane and the solvent. Examination of the crude reaction mixture by *H N M R spectroscopy revealed the presence of 3.21a-c in a relative yield of about 2:1:0.8 along with unidentified products. Next, the N M R mixture was taken to dryness, extracted into pentane, and filtered through Celite ( l x l cm). Concentration of the filtrate followed by cooling to -30 °C for 2 d provided 3.21 as pale yellow crystals (30 mg, 41% yield). X-ray quality crystals of 3.21a were obtained by recrystallizing this batch of crystals from a mixture of 5:1 pentane/toluene at -30 °C. 70 Anal. Calcd. for C 2 0 H 3 3NOW: C, 49.29; H , 6.83; N , 2.87. Found: C, 48.98; H , 6.77; N,2.95. IR (cm 1 ) VNO 1585, 1621, 1651 (s). MS 487 [ M + , 1 8 4 W ] . ' H N M R (500 M H z , C 6 D 6 ) 8 3.21a: 0.41 (d, 2JHH = 2.7, IH, C a H 2 anti), 0.93 (s, 3H, C g H 3 ) , 1.22 (s, 3H, C h H 3 ) , 1.50 (obscured, IH, C H ) , 1.59 (s, 15H, C 5 Me 5 ) , 1.85 (dd, JHH = 4.2, 9.5, IH , C e H 2 ) , 1.94 (d, 3 J H H = 7.5, 3H, CjH 3), 1.94 (d, 2JHH = 3.1, IH, C a H 2 syn), 2.38 (s, 3H, C C H 3 ) , 2.56 (t, 2J HH = 10.3, IH, C e H 2 ) , 2.66 (dd, 3 J H H = 4.9, 10.1, IH, C d H). 3.21b: 0.38 (d, 2J HH = 2.7, IH, C a H 2 anti), 0.87 (s, 3H, C H 3 ) , 1.19 (s, 3H, CH 3 ) , 1.67 (s, 15H, C 5 Me 5 ) , 1.55 (obscured, 3H, CH 3 ) , 1.61 (obscured, IH, C H 2 ) , 2.01 (obscured, IH, CH 2 ) , 2.08 (d, 2JHH = 3.0, IH, C a H 2 syn), 2.25 (s, 3H, C C H 3 ) , 2.51 (q, V H H = 7.9, IH, QH) , 2.67 (obscured, IH, C d H). 3.21c: 0.49 (d, 2 J H H = 2.7, IH, C a H 2 anti), 0.77 (d, VHH = 7.1, 3H, CH 3 ) , 1.52 (s, 15H, C 5 Me 5 ) , 1.89 (obscured, IH, CH), 1.97 (obscured, IH, CH 2 ) , 2.05 (m, IH, C d H), 2.22 (s, 3H, CH 3 ) , 2.36 (d, 2JHH = 2.8, IH, C a H 2 syn). 1 3 C{ 'H} N M R (75 M H z , C 6 D 6 ) 8 3.21a: 10.2 (C 5 Me 5 ) , 18.2 (C C H 3 ) , 24.3 (CjH 3), 28.9 (C g H 3 ) , 31.7 (C h H 3 ) , 41.1 (C e H 2 ) , 42.2 (QH), 44.4 (C a H 2 ) , 60.1 (C f), 90.6 (C d H), 106.2 (C 5 Me 5 ) , 128.8 (C b). 3.21b: 42.6 (C a H 2 ) , 61.9 (C f), 96.1 (C d H), 123.94 (C b). 3.21c: 48.1 (C a H 2 ) , 76.6 (C d H), 123.92 (C b). Remaining signals for 3.21b and 3.21c: 9.8, 10.7 (C 5 Me 5 ) , 14.3, 18.0, 18.4, 19.5, 26.0, 30.3 (CH 3 ) , 37.3, 35.7 (CH), 32.4, 43.3 (CH 2 ), 106.9 (C 5 Me 5 ) . NOEDS (400 M H z , C 6 D 6 ) 3.21a: irrad. at 0.41 ppm (C a H 2 ) , NOEs at 8 1.92 (C a H 2 ) , 2.66 (C d H); irrad. at 0.93 ppm (C g H 3 ) , NOEs at 1.22, 1.85, 2.65; irrad. at 1.22 ppm (C h H 3 ) , NOEs at 8 0.93, 1.92 (CjH 3), 2.55; irrad. at 1.67 ppm, NOEs at 8 0.41, 0.93; irrad. 2.38 ppm (C C H 3 ) , NOEs at 8 1.22, 1.92 (C a H 2 ) ; irrad. at 2.65 ppm, NOEs at 8 0.41, 0.93, 1.85. 71 3.3 Results and Discussion 3.3.1 Trapping of [Cp'W(NO)(=CHCMe 3 )] and [Cp*W(NO)(=CHPh)] with Lewis Bases In analogy to the behavior of the molybdenum bis(neopentyl) complex, CpMo(NO)(CH2CMe3)2, studied earlier,15 the thermolysis of the tungsten bis(alkyl) analogues Cp 'W(NO)(CH 2 CMe 3 )2 [Cp' = Cp (2.1), Cp* (2.4)] and Cp*W(NO)(CH 2 CMe 3 )(CH 2 Ph) (2.15) occurs exclusively by an a-abstraction process that yields neopentane and the transient neopentylidene and benzylidene intermediates [Cp'W(NO)(=CHCMe 3)] and [Cp*W(NO)(==CHPh)]. Three features of the generation and subsequent trapping of these tungsten alkylidene intermediates contrast sharply with the molybdenum reactions, however. First, unlike CpMo(NO)(CH 2 CMe 3 ) 2 , which decomposes in several hours at room temperature, the tungsten bis(alkyl) precursors are thermally robust, decomposing in solution only under forcing conditions: ca. 15 h at 60 °C for 2.1 and 2 d at 70 °C for 2.4 and 2.15. Second, thermolysis of solutions of 2.1 or 2.4 in the absence of a trapping agent does not produce a bridging alkylidene complex as observed for the molybdenum system but rather leads to decomposition. Third, while the molybdenum reactions may be conducted in a range of solvents including hexanes, E t 2 0 , and dichloromethane, the reactions of the tungsten complexes are highly solvent sensitive and depend on the nature of the trapping ligand. In general, the tungsten alkylidene complexes (only the neopentylidene systems were studied in detail) are readily trapped by Lewis bases (L) such as phosphines, phosphites, and amines, as shown in Scheme 3.2, but not with acetonitrile, ter/-butylnitrile, E t 2 0 , THF, or Me 2 S. The observation of the tungsten ligand-trapped products, however, depends on the steric bulk and donor strength of L and on the type of Cp' ligand. For complex 2.1, adduct formation leading to complexes 3.1-3.3 is observed with PMe 3 , PEt 3 , and P(OMe) 3 , while an intractable mixture of products is observed upon reaction with pyridine. Complex 2.4, in contrast, reacts with not only PMe 3 , PMe 2 Ph, PEt 3 , and P(OMe) 3 , but also with dimethylethylamine and pyridine, yielding complexes 3.4-3.9 in high yields, a behavior that is 72 Scheme 3.2 Cp'W(NO)' C H 2 C M e 3 \ C H 2 C M e 3 - C M e 4 [Cp*W(NO)(=CHCMe3)] Cp'W(NO)' H ^ U c M e 3 2.1 Cp' = Cp 2.4 Cp' = Cp* Cp' = Cp Cp' = Cp* 3.1 L = P M e 3 3.2 L = PEt 3 3.3 L = P(OMe) 3 3.4 L = P M e 3 3.5 L = PMe 2 Ph 3.6 L = PEt 3 3.7 L = P(OMe) 3 3.8 L = NMe 2 Et 3.9 L = Py C H 2 C M e 3 7 Q o C Cp*W(NO) 2.15 / \ C H 2 P h PMe, -CMe, [Cp*W(NO)(=CHPh)] Cp*W(NO)' H ^ P h PMe, 3.10 more analogous to the CpMo bis(neopentyl) system.15 However, unlike the molybdenum system, which is relatively insensitive to the steric bulk of the trapping ligands, decomposition is observed when the tungsten complexes 2.1 and 2.4 are independently thermolyzed in the presence of sterically encumbered PPh 3 or when 2.4 is reacted with bulky triethylamine. The fact that the [CpMo(NO)(=CHCMe 3)] and [Cp*W(NO)(=CHCMe 3)] intermediates can be trapped by amines but not the CpW analogue, even though the electronic properties of this species are more comparable with those of the molybdenum system, strongly suggests that it is not the trapping of the unsaturated neopentylidene species that depends on the nature of L and Cp' . A more reasonable explanation is that it is the effect of L and Cp ' on the stability of the ligand-trapped products that determines whether these complexes wil l be observed and 73 isolated. On the basis of the above observations, this stability decreases as the size and donating ability of L increases and decreases, respectively. It also decreases when Cp' is changed from Cp* to Cp. Thus, unlike the CpMo system, which forms isolable pyridine and PPh 3 adducts, analogous CpW products are not observed presumably not because they are not formed but because they decompose at the temperature required for 2.1 to extrude neopentane. As noted above, the observation of the tungsten ligand-trapped alkylidene complexes is also markedly dependent on the choice of solvent. This is due to the fact that the unsaturated alkylidene intermediates are highly reactive not only toward dative ligands but also toward most solvents. For reactions involving amine ligands, the amine must also function as the solvent in order to avoid competitive side-product formation and decomposition. For reactions with phosphine and phosphite ligands, on the other hand, THF but not E t 2 0 or acetonitrile may be used as solvent without deleterious effects. Use of sterically hindered 2,2,4,4-tetramethylpentane for the reaction of 2.4 with PMe3 is also equally effective. Changing this solvent to cyclohexane or other less encumbered aliphatic hydrocarbon solvents, however, yields a mixture of products in which the desired neopentylidene adduct 3.4 is formed in moderate yields at best, due to competitive reaction of the [Cp*W(NO)(=CHCMe 3)] intermediate with the C - H bonds of the solvent (see Chapter 4). Finally, the reaction of 2.4 with P M e 3 in dichloromethane does not lead to alkylidene products; rather, a complex mixture of compounds, in which Cp*W(NO)(Cl) 2(PMe 3) (3.11) is the major (and only identified) product, is observed (eq 3.3). The identity of 3.11 was established by N M R and mass spectroscopies and by comparison to an authentic sample prepared from reaction of Cp*W(NO)Cl 2 with PMe 3 . The mechanism by which 3.11 forms remains unclear, since the reaction also generates a large amount of insoluble phosphonium chlorides due to rapid reaction between P M e 3 and dichloromethane at 70 °C. A 2.4 + P M e 3 - Cp*W(NO)(PMe 3 )Cl 2 (3.3) C H 2 C 1 2 3.11 74 A l l alkylidene complexes outlined in Scheme 3.2 are 18-electron species. They are yellow to brown-orange materials that are readily soluble in aromatic and ethereal solvents and that, although they may be handled briefly in air as solids, are moisture sensitive in solution. With the exception of 3.3, the phosphine and phosphite adducts can be crystallized from hexanes/Et20 and are stable as solids for months at room temperature and to substitution by weaker a-donors. The amine adducts, in contrast, are unstable in the solid state, decomposing over several days at -30 °C in the absence of light. In the presence of excess PMe3, the pyridine adduct 3.10 slowly and quantitatively converts to the P M e 3 adduct 3.4. In CeD 6 solutions, the Cp and Cp* phosphine adducts show little degradation after being heated for 1 d at 100 and 120 °C, respectively, whereas the amine adducts are sensitive to concentration effects. For example, in dilute CeD6 solutions, complex 3.9 may be heated to 70 °C for 16 h or stored at 25 °C for ca. 1 week in the dark with only trace amounts of decomposition and no reaction with solvent. In concentrated solutions, it decomposes over a period of days at room temperature to intractable products, presumably via bimolecular pathways. For comparison, the CpMo pyridine adduct can be crystallized and isolated, but at room temperature it decomposes slowly in solution to give largely the asymmetric dimer [CpMo(NO)(CHCMe3)]2 (Scheme 3.1). The CpMo phosphine analogues, in contrast, show no sign of decomposition even upon heating in C^D^ at 180 °C for 24 h. Each of the complexes 3.1-3.10 was generated and isolated as a single isomer, which on the basis of ' H N M R and X-ray data is the anti rotamer in which the alkylidene substituent points away from the Cp ' ligand (see below). The syn rotamer was sometimes also observed in trace amounts (-5-10%) when solution or solid samples of the anti isomers were exposed to room light for more than several days. In the absence of light, no syn products could be detected, even at elevated temperatures. Thus, in contrast to many other alkylidene complexes including the CpMo(NO)(=CHCMe 3)(L) system, which can undergo syn and anti rotamer interconversion thermally,21 the syn and anti rotamers of the tungsten alkylidene complexes reported here can only be interconverted photochemically. 75 anti syn A l l alkylidene complexes in Scheme 3.2 were fully characterized by ' H and 1 3 C N M R spectroscopies and in the case of complexes 3.1 and 3.4 also by X-ray diffraction. The ' H and 1 3 C N M R spectra of complexes 3.1-3.10 are similar to those of their molybdenum counterparts (Table 3.1). For the phosphine and phosphite adduct complexes, the alkylidene H a and C a resonances appear as doublets due to coupling to phosphorus. For complexes containing a mixture of syn and anti rotamers, two sets of H a resonances, which differ in chemical shift by up to 1.4 ppm, are observed. The higher-field resonances are assigned to the anti isomers since the a-hydrogen atoms in these complexes are closer the Cp ' ligand and are Table 3.1. Selected ' H , 1 3 C , and 3 1 P N M R Data for Cp'W(NO)(=CHCMe 3 )(L) and Cp*W(NO)(=CHPh)(PMe 3) Complexes0 Complex H a C a 31p 3 r b -'HP 5 '-^CH 8C 'jpw CpW(NO)(=CHCMe3)(PMe3) (3.1 )c 12.00 (13.27) 3.0(5.4) 282.6 114 10 -5.1 (-1.5) 455 CpW(NO)(=CHCMe3)(PEt3) (3.2) 12.10 3.0 285.5 117 10 31.4 435 CpW(NO)(=CHCMe3)[P(OMe)3] (3.3) 12.80 3.0 Cp*W(NO)(=CHCMe3)(PMe3) (3.4)c 11.25 (12.81) 3.6 (4.8) 282.8 111 9 -6.5 (-2.9) 449 Cp*W(NO)(=CHCMe3)(PMe2Ph) (3.5/ 11.27(12.64) 3.3 (4.0) 285.4 113 9 9.8 (7.6) 446 Cp*W(NO)(=CHCMe3)(PEt3) (3.6) 11.25 (12.47) 3.6 (4.5) 285.5 112 9 25.0 430 Cp*W(NO)(=CHCMe3)[P(OMe)3] (3.7) 11.91 (13.30) 3.0 (3.9) 296.3 116 13 24.5 (19.0) 675 Cp*W(NO)(=CHCMe3)(NMe2Et) (3.8)c 10.10 257.9 116 Cp*W(NO)(=CHCMe3)(Py) (3.9)c 10.63 (12.32) 264.4 116 Cp*W(NO)(=CHPh)(PMe3) (3.10)rf 11.79 3.8 261.2 119 9 -6.5 443 "And rotamer in CDC13 unless otherwise noted. Vn t i rotamer (syn rotamer). cIn C 6 D 6 . dln CD 2C1 2. 76 therefore more shielded. Other notable features of the ' H and 1 3 C N M R data of 3.1-3.10 are the VCH coupling constants and the chemical shifts of the resonances of the alkylidene a - C H moieties. For all complexes, the [JCH values are low (111-119 Hz) 2 2 and fall within the range characteristic of 18-electron alkylidene complexes in which steric interaction between the substituent on the ct-carbon atom and the metal fragment renders the M = C a - C p angles large.2 3 That this is indeed the case has been confirmed by an X-ray structural determination of complexes 3.1 and 3.4 (see later). Both the chemical shifts of the alkylidene H a and C a resonances of 3.1-3.10 move progressively upfield as L varies in the series L = phosphite, phosphine, pyridine, amine. The same also happens to the H a resonance on replacing Cp by Cp*. The fact that an upfield shift is observed on replacing the phosphite or Cp ligand with the more electron-donating (and therefore more shielding) phosphine or Cp* group is undoubtedly electronic in origin, but why the pyridine and amine ligands, which are sterically unremarkable and are poorer donors, give rise to the furthest upfield resonances is unclear at the moment. Finally, it should also be noted that the ' H N M R spectra of the pyridine and benzylidene complexes 3.9 and 3.10 each display only three sets of resonances for the pyridine and benzylidene ring protons, thereby implying that these rings are rotating rapidly on the N M R time scale at 25 °C. The solid-state molecular structures of the neopentylidene complexes 3.1 and 3.4, as determined by X-ray diffraction, are shown in Figure 3.1, along with selected bond distances and angles. Noteworthy is that the N O ligand in each complex is essentially linear, while the orientation of the neopentylidene ligand, with the substituents parallel to the N O moiety ( N - W - C a - C p torsion angle ~ 0°) and the tert-butyl group pointing away from the Cp ' ligand (i.e. the anti orientation), is as expected for a conformation involving maximum W=C it-bonding and minimum steric interactions.24 The W - C a distances of 1.93-1.96 A are typical of 77 N(l ) -0 (1 ) 1.24(1) A W ( l ) - N ( l ) 1.77(1) A W(l) -C(6) 1.93(2) A W ( l ) - P ( l ) 2.418(4) A W ( l ) - N ( l ) - 0 ( 1 ) 1 7 5 . 3 ( 1 0 ) ° P ( l ) - W ( l ) - N ( l ) 9 1 . 3 ( 4 ) ° P ( l ) - W ( l ) - C ( 6 ) 9 3 . 4 ( 4 ) ° N ( l ) - W ( l ) - C ( 6 ) 1 0 1 . 2 ( 5 ) ° W( l ) -C(6) -C(7) 1 4 0 ( 1 ) ° B C(9) C(10) C(18) N(l ) -0(1) 1.22(2) A W ( l ) - N ( l ) 1.774(14) A W(l)-C(14) 1.961(9) A W ( l ) - P ( l ) 2.555(4) A W ( l ) - N ( l ) - 0 ( 1 ) 1 6 3 . 3 ( 1 4 ) ° P ( l ) - W ( l ) - N ( l ) 9 1 . 0 ( 5 ) ° P ( l ) -W( l ) -C(14) 9 3 . 8 ( 6 ) ° N ( l ) - W ( l ) - C ( 1 4 ) 9 8 . 7 ( 7 ) ° W( l ) -C(6) -C(7) 1 4 0 . 7 ( 1 1 ) ° F i g u r e 3.1. ORTEP diagrams and selected bond distances and angles for complexes (A) CpW(NO)(=CHCMe 3)(PMe 3) (3.1) and (B) Cp*W(NO)(=CHCMe 3)(PMe 3) (3.4). 78 W - C double bonds, while the W - C a - C p angles of 140° are obtuse for sp2-hybridized centers, confirming the results of the N M R studies. For comparison, the corresponding bond distances and angles in isoelectronic CpMo(NO)(=CHCMe 3)(PMePh 2) and [CpRe(NO)(=CHPh)(PPh 3)][PF 6] 2 , a complexes are 1.950(3) and 1.949(6) A , and 140.8(3) and 136.2(5)°, respectively. 3.3.2 Reactions of [Cp'W(NO)(=CHCMe3)] and Cp*W(NO)(=CHCMe3)(PMe3) with Heteroatom-Hydrogen Bonds In analogy with the molybdenum chemistry, the thermolysis of the tungsten bis(neopentyls) 2.1 and 2.4 in the presence of amines with accessible N - H bonds results in the quantitative formation of amido-alkyl complexes Cp'W(NO)(CH 2 CMe 3 ) (NR 2 ) - For example, thermolysis of a THF solution of 2.1 at 60 °C for 1 d or of 2.4 at 70 °C for 2 d in the presence of excess dimethylamine leads exclusively to the corresponding neopentyl amido complexes 3.12 and 2.17 as shown in Scheme 3.3. Heating the oc-deuterated complex CpW(NO)(CD 2 CMe 3 ) 2 (2.I-CI4) under similar conditions yields the monodeuterated species 3.12-</, for which the ' H N M R spectrum shows a broad singlet at 5 1.50 and the 1 3 C { ' H } N M R spectrum a triplet at 5 46.9 (lJcu = 18.5 Hz) due to the neopentyl a-proton and a-carbon atom, respectively. This result thus demonstrates that 2.1 and 2.4 react with amine N - H bonds via the neopentylidene [CpW(NO)(=CHCMe 3)] and [Cp*W(NO)(=CHCMe 3)] intermediates. Scheme 3.3 C X 2 C M e 3 A M e 2 N H C X H C M e , cp'WfNoy *• [Cp'W(NO)(=CXCMe3)] Cp 'WfNoy \ THF \ C X 2 C M e 3 + N M e 2 CMe4-c?o-3 Cp' = Cp X = H (3.12) X = D (3.12-d) Cp' = Cp* X = H (2.17) 2.1 Cp' =Cp 2.4 Cp' = Cp* 79 Stronger protic acids such as alcohols2 5 and carboxylic acids 1 9 do not react with 2.1 or 2.4 via the intermediacy of alkylidene complexes. Although O - H bond cleavage leading to formation of neopentyl alkoxo and carboxylato complexes is observed, these reactions instead proceed rapidly by protonolysis of a neopentyl ligand in the starting material. Formal addition of O - H and even N - H bonds across a W=C linkage, however, can be observed when amines, alcohols, and carboxylic acids are reacted with the ligand-trapped complexes Cp'W(NO)(=CHCMe3)(L). For example, as shown in eq 3.4-3.6 when complex 3.4 is treated with dimethylamine, methanol, or benzoic acid, loss of PMe 3 occurs with quantitative formation of the amido complex 2.17, the alkoxo complex 2.16, and the known carboxylate complex 3.13,19 respectively. For comparison, of the known CpMo(NO)(=CHCMe3)(L) systems, only the pyridine adduct shows any degree of substitutional lability toward substrates that function as both a Lewis base and a Bransted acid. 1 6 Qualitatively, the rate of the tungsten reactions varies with the acidity of the organic compounds, with benzoic acid reacting immediately upon contact at ca. 10 °C, methanol in several hours at room temperature, and dimethylamine in 1-2 days at 70 °C. Substrates with moderate acidity but lack a donor function are not sufficiently reactive toward 3.4. For example, phenylacetylene and methyl acetate, which have significantly lower pKa values than dimethylamine, do not react.26 Me 2 NH Cp'W(NO)(CH 2 CMe 3 ) (NMe 2 ) + P M e 3 (3.4) / THF A 3.12 Cp' = Cp 2.17 Cp' = Cp* C H C M e 3 / MeOH Cp'W(NO) * Cp*W(NO)(CH 2 CMe 3 )(OMe) + P M e 3 (3.5) \ PMe 3 2.16 3.1 Cp 1 =Cp U p U p N *- Cp*W(NO)(CH 2 CMe 3 ) (0 2 CPh) + P M e 3 (3.6) C6D6 3.13 Since the coordinatively unsaturated [Cp*W(NO)(=CHCMe 3)] complex is highly reactive toward the C - H bonds of hydrocarbon solvents even in the presence of excess 80 P M e 3 , 1 8 the fact that the PMe 3 adduct 3.4 is stable upon thermolysis in benzene for several days at 70-80 °C strongly suggests that loss of PMe 3 prior to heteroatom-hydrogen bond addition is unlikely. More probable is that the reaction proceeds via an associative mechanism in which a net addition of the heteroatom-hydrogen bond across the W=C bond in 3.4 occurs prior to and/or concurrently with phosphine dissociation. Since 3.4 is a coordinatively saturated, 18-electron complex, this addition reaction could occur in one of three ways: (a) by initial protonolysis of the oc-carbon atom, followed by coordination of the heteroatom to tungsten and dissociation of phosphine; (b) by an S^- l ike displacement of the phosphine as 5 3 the heteroatom-hydrogen bond is added across the alkylidene functionality; or (c) by an r\ -r\ ring slippage or rf—tf N O isomerization to open a coordination site, followed by heteroatom coordination, phosphine loss, and heteroatom-hydrogen bond addition. While the above results do not allow these possibilities to be distinguished, initial protonolysis is not unreasonable for the reaction with benzoic acid, for it would explain why this substrate, which is a poorer donor than dimethylamine and methanol, reacts instantaneously upon contact. 3.3.3 Reactions of [Cp*W(NO)(=CHCMe3)] with Cyclic and Acyclic Alkenes In addition to its reactions with Lewis bases and heteroatom-hydrogen bonds, the transient [Cp*W(NO)(=CHCMe 3)] complex reacts readily with cyclic and acyclic alkenes, leading to [2+2] cycloaddition and/or C - H activation products depending upon the structure of the alkene. 3.3.3.1 [2+2] Cycloaddition As shown in Scheme 3.4, heating 2.4 in neohexene at 70 °C for 2 d affords the tungstacyclobutane complex 3.14 as the sole product as judged by ' H N M R spectroscopy. Complex 3.14 is an orange-yellow solid which can be crystallized in 97% yield from hexanes/Et20 at -30 °C and which, in contrast to analogous acyclic bis(alkyl) complexes, is remarkably stable. For example, it is unaffected by air and moisture in the solid and solution 81 states and is inert to protic acids such as methanol. It is also stable to thermolysis (70 °C, 12 h) in toluene in the presence of excess PMe3. When 2.4 is heated in cyclopentene, ethylcyclopentene, or norbornadiene, the corresponding metallacycles 3.15, 3.16, and 3.17 are isolated in 45%, 17%, and 32% crystallized yields, respectively. These reactions, however, are accompanied by decomposition, formation of complex mixtures of organometallic side products, and formation of the following organic byproducts, which unfortunately could not be quantified. The reaction with norbornadiene, for example, is complicated by precipitation (particularly in concentrated solutions) of a white, amorphous, rubbery polymer that is presumably polynorbornadiene. This polymer is insoluble in all standard organic solvents, making its removal from the reaction vessel impossible2 7 and the isolation of 3.17 generally difficult. In the reaction with ethylcyclopentene, no polymeric products could be detected. Instead, G C - M S analysis of the volatiles reveals the presence of two major organic products whose identity remains to be determined but whose masses (m/z 166 and 164) suggest that they are formed respectively by coupling of a neopentylidene ligand to an ethylcyclopentene molecule and by dehydrogenation of this coupled product. The mechanism of the coupling reaction is unknown, but appears not to involve initial [2+2] cycloaddition of ethylcyclopentene to the [Cp*W(NO)(=CHCMe3)] intermediate (see later) since the product of this metathesis process would be expected to have a mass greater than 166. Finally, in the cyclopentene reaction, neither polymers nor coupled products analogous to those generated in the ethylcyclopentene reaction could be detected. Instead, products consistent with the dimerization of cyclopentene are observed. Thus, G C - M S analysis of the volatiles reveals the presence of 3-cyclopentylcyclopentene (m/z 136) and l,l'-bicyclopentyl (m/z 138) as the major organic species. These were identified by comparison of their mass spectral fragmentation patterns 20 with those of authentic samples reported in the literature. 82 Scheme 3.4 A plausible mechanism for the formation of 3-cyclopentylcyclopentene is shown in Scheme 3.5. The initial step is addition of a methylene C - H bond of cyclopentene (most likely at the allylic position based on results described below) across the W=C bond of [Cp*W(NO)(=CHCMe3)]. This leads to the neopentyl cyclopentenyl intermediate A , which undergoes (3-H abstraction to form the 16-electron alkene complex B. A catalytic cycle then 83 Scheme 3.5 84 ensues, which involves formation of the metallacyclopentane complex E by coordination and subsequent oxidative coupling of a second molecule of cyclopentene. Complex E then undergoes (3-H abstraction, followed by allylic C - H activation of a third molecule of cyclopentene to give G. Alkene insertion into the W - H bond of G then yields intermediate H. From H, 3-cyclopentylcyclopentene is liberated and B regenerated upon (3-H abstraction. The formation of cyclopentene trimers is not observed presumably because coupling between a monomer and a dimer is sterically unfavorable. Like 3.14, complexes 3.15-3.17 are orange, sparingly to moderately soluble in aliphatic hydrocarbon solvents, and stable indefinitely at 25 °C under dinitrogen. They also can be handled for long periods in air and, in the case of 3.15, heated in C6D6 for 12 h at 70 °C, without noticeable decomposition. Spectroscopically, the metallacyclobutane ligands of 3.14-3.17 are all similar; the ' H and 1 3 C N M R spectra of.3.16 in CDC1 3 are shown in Figure 3.2 as representative examples. Like the vast majority of early transition-metal metallacyclobutane complexes,2 8 the Hp and Cp resonances of 3.14-3.17 are found upfield in the range 8 -2.29 to -0.12 and 8 -4.8 to 27.4, respectively, while downfield shifts are observed for the H a and C a resonances at 8 3.09-7.68 and 8 113.2-141.8. The presence of a plane of symmetry in complex 3.14 is indicated by the observation of identical tert-butyl groups and a single W C H signal in the ' H and 1 3 C N M R spectra. That the tert-butyl groups point away from the Cp* ligand has been confirmed by an N O E difference experiment, which shows a strong enhancement of the H a resonance upon irradiation of the Cp* signal, and vice versa. In the ' H N M R spectrum of 3.14, the H a resonance appears as a doublet of doublets with 3JHH = 6.0 and 10.2 Hz due to coupling to the trans- and cis-$ protons, respectively. In 3.15, 3.16, and 3.17,the VHH coupling constants between the Gr7CMe 3 and P protons are 5.1-7.2 Hz, suggesting that they are trans to one another. Similar cis- and trans-coupling constants have been reported previously for other metallacyclobutane systems.2 8 b 85 ~i——1 r~ ppm 7 J ~ i • 1 1 r 6 5 4 2 1 I I I I I I I I I I I I I I I I M I 1 I M I I I I I I I I I I I I I I I I I I I I I I I I I I 1 I I M I I I I I 1 I I I I I I I I I I I I I I I I I I I I I M I I I I I I I I I I ' U 1 " 1 ' M I 1 " 1 1 200 180 . 160 140 120 100 SO 60 40 20 0 PPM Figure 3.2. 500-MHz l H and 75-MHz 1 3 C{ 'H} N M R spectra of 3.16 in CDC1 3 86 Complexes 3.14-3.17 are the first Cp'M(NO)-containing metallacyclobutanes to be synthesized, a class of compounds which to date cannot be accessed by traditional metathesis routes.29 In order to confirm the stereochemical assignments and to gain detailed structural information of these compounds, the solid-state molecular structures of 3.14, 3.15, and 3.17 were established by X-ray diffraction studies. Shown in Figure 3.2 are the ORTEP representations of these structures; selected bond distances and angles are listed in Table 3.2. As expected from the N M R experiments, the tert-butyl groups in 3.14 are cis to one another, while the tert-butyl groups in 3.15 and 3.17 are trans to the (3-substituents. For all three complexes, the N O ligand is essentially linear while the tungstacyclic ring is nearly planar and symmetric. Noteworthy is that the majority of the metrical parameters of the tungstacyclic rings in 3.14, 3.15, and 3.17 are virtually identical (Figure 3.2). The W - C a - C p - C a torsion angles are 10.3(4)°, -0.3(4)°, and -7.5(3)°, respectively, while the W - C „ (2.071-2.098 A) and W--Cp (2.37 A) distances and C a - W - C a angles (84-85°) are in the range often observed for d° tungstacyclobutane complexes. 2 8 3 ' 3 0 The most unusual aspects of the tungstacyclic rings are the Ccx-Cp bond distances. In 3.14, these distances (1.63 A) lie at the long end of the range found for transition-metal metallacyclobutane complexes (1.54-1.63 A) . In 3.15 and 3.17, the C a - C p distances belonging to the alkene fragment are 1.702(6) and 1.671(7) A , respectively, and are the longest reported to date. 60 W 88 Table 3.2. Selected Bond Distances and Angles for 3.14, 3.15, and 3.17 3.14 3.15 3.17 Bond Distances (A) N ( l ) - 0 ( 1 ) 1.230(5) N ( l ) - 0 ( 1 ) 1.239(5) N ( l ) - 0 ( 1 ) 1.227(3) W ( l ) - N ( l ) 1.767(4) W ( l ) - N ( l ) 1.763(4) W ( l ) - N ( l ) 1.771(3) W(l ) -C(12) 2.091(5) W( l ) -C(6 ) 2.098(4) W ( l ) - C ( l l ) 2.088(3) W(l ) -C(14) 2.080(5) W ( l ) - C ( l ) 2.071(5) W(l ) -C(17) 2.075(3) W(l ) -C(13) 2.375(6) W( l ) -C(5 ) 2.372(4) W( l ) -C(12) 2.373(3) C(12)-C(13) 1.638(7) C(5)-C(6) 1.571(6) C ( l l ) - C ( 1 2 ) 1.581(5) C(13)-C(14) 1.631(7) C( l ) -C(5) 1.702(6) C(12)-C(17) 1.671(5) Bond Angles (°) 0(1)-N(1)-W(1) 168.3(4) 0(1)-N(1)-W(1) 169.0(3) 0(1)-N(1)-W(1) 168.7(3) C(12)-W(l) -C(14) 84.3(2)C(1 ) -W( l ) -C(6 ) 85.0(2) C ( l l ) - W ( l ) - C ( 1 7 ) 84.13(12) W(l ) -C(12)-C(13) 78.1(3)W( 1)-C(1)-C(5) 77.2(2) W ( l ) - C ( l l ) - C ( 1 2 ) 79.3(2) W(l ) -C(14)-C(13) 78.6(3)W( 1)-C(6)-C(5) 79.1(2) W ( l ) - C(17)-C(12) 77.8(2) C(12)-C(13)-C(14) 117.8(4) C( l ) -C(5) -C(6) 118.7(4) C( l l ) -C(12 ) -C(17 ) 118.1 (3) W ( l ) - C ( 1 2 ) - C ( l l ) 133.5(3) W ( l ) - C(6)-C(7) 131.5(3) W ( l ) - C ( l l ) - C ( 1 9 ) 132.0(2) 3.3.3.2 C - H Bond Activation In contrast to the alkene reactions described above, the thermolysis of 2.4 in acyclic alkenes possessing allylic C - H bonds leads primarily to C - H bond activation products. For example, thermolysis of 2.4 in neat 2-methyl-l-butene at 70 °C for 2 d results in formation of a 6:5:4 mixture of the 773-allyl neopentyl complex 3.18, T7 3-allyl hydride complex 3.19, and rf-trans-diene complex 3.20 along with several intractable Cp*-containing products, which do not display N M R resonances characteristic of metallacyclobutane complexes (eq 3.7). G C - M S analysis of the volatile products shows neopentane and a barely detectable compound whose mass equals the sum of a neopentylidene ligand and a 2-methyl-1 -butene molecule. 89 Monitoring of the reaction by ' H N M R spectroscopy reveals that 3.18-3.20 are formed in a final combined yield of 60-70% and in a constant ratio over the course of the reaction. Complexes 3.18-3.20 can be heated in 2-methyl-l-butene, 2-methyl-2-butene, or benzene at 70-80 °C for 3-4 d without decomposition, rearrangement, or reaction with the solvent. These observations thus demonstrate that these complexes are thermally robust and are produced by independent pathways. (3.7) Et 3.18 Et 3.19 3.20 jf, -CMe 4 ' Complex 3.18 is a sticky off-white solid whose formation is consistent with initial allylic C - H activation by [Cp*W(NO)(=CHCMe 3)], followed by an rf- to 773-allyl conversion as shown in Scheme 3.6a. Although highly soluble in hydrocarbon solvents, 3.18 can be isolated in 20% yield by chromatography on alumina I, followed by painstaking crystallization from pentane/Et20 at -30 °C. It can also be fully characterized by spectroscopic and analytical techniques. In addition to resonances attributable to the Cp* and neopentyl ligands, the ' H and 1 3 C N M R spectra of 3.18 display characteristic patterns indicative of the presence of an rf-allyl moiety.3 1 Due to the chirality at the metal center, two sets of diastereotopic syn and anti proton resonances are observed in the ' H N M R spectrum; these appear at 8 2.22 and 3.64 and 8 0.75 and 1.89, respectively. In the l 3 C { ' H } N M R spectrum, the terminal carbons resonate at 8 42.5 and 74.2, while the signal for the central carbon appears at 8 135.6. A n endo conformation was assigned based on the results of an ' H N O E difference experiment, which shows a strong enhancement of the Cp* signal upon irradiation of the signal for the anti proton at 8 0.75. This assignment is consistent with the observation that the anti protons, being closer to the Cp* ligand, are more shielded and therefore resonate at higher field than the syn protons. 90 In contrast to 3.18, the 7]3-allyl hydride and t]4-trans-diens complexes 3.19 and 3.20 could not be isolated free of 3.18 and each other. They were therefore identified on the basis of IR and ' H / 1 3 C N M R spectroscopies, aided by H M Q C and C O S Y experiments. Spectroscopically, 3.20 is very similar to the previously reported CpMo(NO)(?ra«5-r/ 4-2-methylbutadiene)32 complex. Of note are the upfield shifts of the diene protons at 8 0.95, 1.1, 2.3, and 3.29, which are indicative of not only rj4-bonding but also significant 7t-backbonding from the tungsten metal. For complex 3.19, its IR and N M R characteristics are similar to those of the Cp*W(NO)(r; 3-allyl)(H) complexes described in Chapter 4. Key to its structural assignment as a hydride complex is a weak W - H stretch at 1888 cm"' in the Nujol-mull IR spectrum and a singlet integrating for one proton at 8 - 0 . 8 8 ('JWH = 1 1 8 Hz) in the ' H N M R spectrum (CeDe). The formation of complex 3.20 presumably occurs by initial C - H bond cleavage at the non-allylic methyl and/or (though less likely) the allylic methylene carbon, followed by p-abstraction and coordination of the pendant C=C bond (Scheme 3.6b). The formation of complex 3.19, on the other hand, can be rationalized by the reaction sequence shown in Scheme 3.6c. According to this multistep mechanism, which also yields an organic byproduct whose mass equals the sum of a neopentylidene ligand and a 2-methyl-1 -butene molecule, allylic C - H activation at the methylene position, followed by if- to n 3-allyl conversion first takes place to yield an r/ 3-allyl neopentyl intermediate E. In order to relieve allylic strain due to substitution at the terminal carbon, E rearranges to alkene adduct F by a reductive C - C coupling process similar to that proposed recently by Ipaktschi and coworkers for the • 5 1 9 9 9Qb conversion of CpW(NO)(n -allyl)(rj -alkynyl) complexes to r\ -alkene-n -allene complexes. F then reacts with another equivalent of 2-methyl-1-butene to yield 3.19 via a reaction pathway analogous to that proposed above for the formation of 3-cyclopentylcyclopentene. An exception is that formation of the n'-allyl-hydride-n^alkene complex H is followed by if-to r/ 3-allyl conversion with concomitant displacement of the alkene ligand rather than by insertion of the alkene into the W - H bond. This proposed mechanism assumes that (1) allylic C - H activation of the alkene ligand in F is slow compared to coordination of a second molecule of 2-methyl-1-butene and (2) for steric reasons, allylic C - H activation of the 2-91 Scheme 3.6 methyl- 1-butene ligand in G is faster than allylic C - H activation of the trisubstituted alkene as well as alkene-alkene coupling. The thermolysis of 2.4 in neat 2-methyl-2-butene at 70 °C for 2 d, in contrast, is cleaner than the 2-methyl- 1-butene reaction. It affords neopentane as the only detectable organic byproduct and shows little evidence, i f any, in the ' H N M R spectrum of the crude reaction mixture for formation of allyl neopentyl, allyl hydride, or diene complexes as described above. It produces instead a mixture of three major organometallic products in 92 70-80% yield in a ratio of-10:5:4, the most abundant of which is Cp*W(NO)[r/ 4-CH 2 C(Me)CHCH 2 C(Me) 2 CH(Me)] (3.21a) in which two molecules of 2-methyl-2-butene have coupled (eq 3.8). others (3.8) Neither complex 3.21a nor the second (3.21b) and third (3.21c) products could be separated from one another. A l l three compounds were therefore characterized as a mixture, obtained by crystallization from a concentrated pentane solution of the crude products at -30 °C. The identity of 3.21a was established on the basis of ' H and 1 3 C N M R spectroscopies, aided by C O S Y , H M Q C , H M B C , and N O E difference experiments, and confirmed by X-ray diffraction. In its ' H N M R spectrum, 3.21a gives rise to 11 signals due to the 11 different types of protons. Most prominent are the signals at 8 1.50 due to the W - C H proton and at 8 0.41, 1.98, and 2.66 due to the C H 2 and C H protons of the n 3 -CH 2 C(Me)CH moiety. In the 1 3 C N M R spectrum, the corresponding carbon resonances are found at 8 42.2, 44.4, and 90.6. Shown in Figure 3.4 is an ORTEP diagram of the solid-state molecular structure of 3.21a along with selected bond distances and angles. As expected from the N M R experiments, the structure contains a [CH(CH 3 )C(CH 3 ) 2 CH 2 CH(CH 3 )CH 2 ] ligand, bonded by a W-alkyl and a W-r/ 3 -a l lyl interaction. Noteworthy are the following two features. First, the W - C ( l 1) distance (2.307(7) A) is at the long end of the range for W - C single bonds. Second, 3 2 the allyl moiety is coordinated in an endo, syn fashion and shows no r\ —•rj,?] distortion typically observed in (773-allyl)(nitrosyl)tungsten and -molybdenum systems such as CpM(NO)(r/3-allyl)(halide) [M = M o , 3 3 W 3 4 ] , CpW(NO)(77 3-allyl)(n'-alkynyl), 2 9 b and 93 TpMo(NO)(r/ 3-allyl)(CO). 3 5 That the latter is the case is evidenced by the fact that the C(14)-C(15) (1.396(9) A) and C(15)-C(16) (1.420(9) A) distances are similar and in the range intermediate of double and single bonds. A reasonable explanation for this is that 3.21a is sufficiently electron-rich to allow the metal to back-donate to both allylic C - C bonds. N(l ) -0 (1 ) 1.225(6) A W ( l ) - N ( l ) 1.765(5) A W ( l ) - C ( l l ) 2.307(7) A W(l) -C(16) 2.239(6) A W(l) -C(15) 2.372(6) A W(l ) -C(14) 2.382(6) A C(15)-C(16) 1.420(9) A C(14)-C(15) 1.396(9) A 0(1)-N(1)-W(1) 170 .7 (5 ) ° C ( 1 2 ) - C ( l l ) - W ( l ) 114 .8 (4 ) ° C(14)-C(15)-C(16) 113 .8 (6 ) ° C(15)-C(14)-C(13) 127 .5 (6 ) ° Figure 3.4. ORTEP diagram and selected bond distances and angles of complex Cp*W(NO)[r7 4 -CT 2 C(Me)CTCH 2 C(Me) 2 CH(Me)] (3.21a). For complexes 3.21b and 3.21c, complete assignment of their ' H and l 3 C N M R resonances, unfortunately, proves not possible due to severe spectral overlap. Nevertheless, both compounds appear to be isomers of 3.21a on the basis of the following observations. First, their N M R and IR spectroscopic data are, for the most part, similar to those of 3.21a (see the Experimental Section). For example, that each complex also contains an r / 3 -CH 2 C(Me)CH moiety is clearly indicated by the presence of C H 2 and C H signals in the *H N M R spectrum at 8 0.38, 2.08, and 2.67 for 3.21b and 8 0.49, 2.36, and 2.05 for 3.21c and by the presence of the corresponding carbon resonances in the 1 3 C N M R spectrum at 8 42.6 and 96.1 for 3.21b and 8 48.1 and 76.6 for 3.21c. Second, the solubility properties of all three complexes are very 94 similar, as evidenced by the fact that they cannot be separated from one another by column chromatography and/or fractional crystallization. Third, elemental analysis and mass spectral data of a 1.9:1:1 mixture of 3.21a, 3.21b, and 3.21c are consistent with the formula shown in eq3.8. Complexes 3.21a-c are yellow, air-stable crystalline materials that can be heated in C6D6 at 70 °C without further transformation. Structurally, 3.21a-c are related to the zirconocene and the CpMo(NO) 774(o~,T}3)-allyl complexes shown in Scheme 3.7, prepared by reaction of the cis- and ?ra«5-Cp2Zr(r]4-diene) complexes with alkenes3 6 and of the CpMo(NO)(T]4-2,5-dimethyl-2,4-hexadiene) complex with alkynes.3 7 Taking into account the reactivity of these diene compounds, the formation of 3.21a is proposed to proceed via the reaction sequence shown in Scheme 3.8. The initial step is allylic C - H activation leading to the rf -allyl neopentyl complexes A and/or B , which subsequently cyclometallate by a 5-elimination process to give the metallacyclopentene complex C . Rapid insertion of a second equivalent of 2-methyl-2-butene into the sterically less hindered W - C bond of C would then afford 3.21a. Scheme 3.7 CpMofNO) R'C— CR' CpMo(NO) R" 95 Scheme 3.8 3.21a The proposal that A and/or B is unstable with respect to cyclometallation and formation of a five-membered tungstacycle, which can undergo further transformation in the presence of a Lewis base, is not unprecedented. Indeed, previous studies have shown that the bis(alkyl) complex, CpW(NO)(CH2SiMe3)(CH2CMe2Ph), also cyclometallates readily under thermolytic conditions in the presence of PMe3 to give the orthometallated complex CpW(NO)(CH 2 CMe 2 -o-C 6 H 4 ) (PMe 3 ) . 3 8 That this reaction proceeds via a metallacyclopentane intermediate wi l l be demonstrated in Chapter 5 for the thermal decomposition of the Cp* neopentyl neophyl derivative. Finally, since signals for complex 3.20 are not observed in the ' H N M R spectrum of the crude reaction mixture and since 3.20 is thermally stable in the presence of 2-methyl-2-butene, isomerization of C to 3.20 followed by 2-methyl-2-butene insertion in a manner analogous to the CpMo system can be ruled out. 96 In contrast to 2.4, the thermolysis of 2.1 in neat alkenes such as cyclopentene gives complex mixtures of products along with decomposition and formation of a brown-black precipitate. The CpMo(NO)(CH 2 CMe 3 )2 complex, on the other hand, shows little or no reactivity toward alkenes.1 5 b 3.4 Epilogue In summary, Cp 'W(NO)(CH 2 CMe 3 )2 and Cp*W(NO)(CH 2 CMe 3 )(CH 2 Ph) complexes are thermally unstable with respect to a-abstraction of neopentane. This leads to the transient 16-electron alkylidene complexes [Cp'W(NO)(=CHCMe 3)] and [Cp*W(NO)(=CHPh)], respectively, which differ from the [CpMo(NO)(=CHCMe 3)] system in a number of ways. 1 5 Notable differences and trends that emerge upon comparison of the trapping reactions of the neopentylidene systems with Lewis bases (L) are as follows. First, the rate of a-abstraction decreases significantly when the metal is changed from Mo to W and when Cp is replaced by Cp*, though to a lesser extent. Second, only one ' H N M R resonance is observed for the a-protons of the tungsten alkylidene complexes, indicating the presence of only the anti rotamer. Third, the tungsten complexes do not undergo dimerization in the absence of L, as observed for the CpMo system. Fourth, the trapping of [Cp'W(NO)(=CHCMe 3)] is markedly sensitive to the nature of L , favoring sterically undemanding and strong o-donors such as phosphines, phosphites, and, in the case of the Cp* system, amines. It is also strongly dependent on the choice of solvent. For example, when L = phosphine or phosphite, THF or 2,2,4,4-tetramethylpentane may be used as solvents with little or no deleterious effects. When L = amine, clean adduct formation is observed only when the amine also functions as the solvent. Fifth, the stability of the tungsten ligand-trapped complexes is sensitive to the nature of both L and Cp ' , decreasing as the size and electron-donating ability of L increases and decreases, respectively, and as Cp ' is changed from Cp* to Cp. Finally, the syn-anti isomerization of the tungsten adducts can be effected photochemically but not thermally, as has been reported for the molybdenum counterparts. 97 The thermal reactions of the Cp'W(NO)(CH 2CMe3)2 complexes with a variety of heteroatom-hydrogen bonds has also been investigated. The reaction with dimethylamine demonstrates that [Cp'W(NO)(=CHCMe 3)] species are initially generated, which then effect the cleavage of the amine N - H bond to give cleanly the neopentyl amides 3.12 and 2.17. With more acidic substrates such as methanol and benzoic acid, protonolysis of the bis(neopentyl) precursors occurs before generation of the [Cp'W(NO)(=CHCMe 3)] intermediates can take place. Complexes 3.12 and 2.17 alternatively can be prepared by thermolysis of the PMe 3 -trapped alkylidene complexes Cp'W(NO)(=CHCMe 3)(PMe 3) in the presence of excess dimethylamine. When Cp*W(NO)(=CHCMe 3)(PMe 3) is treated with methanol or benzoic acid, formation of the corresponding neopentyl alkoxide complex 2.16 and neopentyl carboxylate complex 3.13 is observed. Qualitatively, the rates of these reactions increase as the acidity of the organic substrates increases. The reaction of Cp'W(NO)(=CHCMe 3 )(PMe 3 ) complexes with heteroatom-hydrogen bonds most likely involves an associative mechanism rather than a mechanism analogous to that previously proposed for CpMo(NO)(=CHCMe 3)(Py), i.e. initial loss of PMe 3 , followed by N - H or O - H addition across the W=C bond. This is based on the fact that all Cp'W(NO)(=CHCMe 3 )(L) complexes, including the amine adducts, are stable to thermolysis (70 °C, 1-4 d) in neat hydrocarbon solvents. If dissociation of L to form coordinatively unsaturated [Cp'W(NO)(=CHCMe 3)] were facile, then C - H bond activation of these solvents should be observed, as demonstrated recently in a communication1 8 of the reaction of [Cp*W(NO)(=CHCMe 3)] with alkanes (full details of which are described in Chapters 4 and 5). The thermal reactivity of the coordinatively unsaturated [Cp'W(NO)(=CHCMe 3)] complexes with alkenes has also been examined. This work demonstrates that, unlike the CpMo system which shows essentially no reactivity toward alkenes, both the CpW and Cp*W complexes are reactive toward these substrates; only the Cp*W complex yields tractable products, however. [Cp*W(NO)(=CHCMe 3)] reacts with cyclic and acyclic alkenes to yield [2+2] cycloaddition and/or C - H activation products depending on the alkene structure. 98 Reactions with neohexene, cyclopentene, ethylcyclopentene, and norbornadiene give the tungstacyclobutane complexes 3.14-3.17, respectively (Scheme 3.4). The formation of 3.14 is quantitative and indicative of a regio- and stereospecific [2+2] cycloaddition reaction. The formation of 3.15-3.17, in contrast, is accompanied by side product formation and decomposition. In the norbornadiene reaction, an insoluble white rubber, presumably polynorbornadiene, is also produced. The reaction with cyclopentene, which contains unhindered allylic C - H bonds, yields 3-cyclopentylcyclopentene and l,r-bicyclopentyl as the major organic products. The formation of 3-cyclopentylcyclopentene can be rationalized by the multistep mechanism shown in Scheme 3.5. According to this mechanism, the key initial step is addition of a methylene C - H bond of cyclopentene (most likely at the allylic position) across the W=C bond of [Cp*W(NO)(=CHCMe 3)]. This leads to a mixed bis(alkyl) complex which can undergo further transformation in the presence of cyclopentene to give the observed product. The reaction with ethylcyclopentene, in contrast, yields two organic compounds whose masses strongly suggest that they are not formed by metathesis pathways but rather by coupling of a neopentylidene ligand to an ethylcyclopentene molecule and by a dehydrogenation of this coupled product. The coupling of an alkylidene ligand to an alkene molecule via a non-metathesis route is extremely rare but not unprecedented. A n example is the reactions of the neopentylidene complexes CpM(=CHCMe3)Cl2 (M = Ta, Nb) with terminal alkenes.39 In this case, Schrock and coworkers showed that these reactions lead to new alkene products, formed by selective insertion of the neopentylidene ligand into an olefinic C - H bond of the alkene substrate. Although the identity of the organic products in the reaction of [Cp*W(NO)(=CHCMe3)] with ethylcyclopentene is unknown, it is unlikely that they are formed by a similar insertion process. This is especially so given the fact that cyclopentene, which is structurally more similar to ethylcyclopentene, does not give rise to such products, whereas acyclic 2-methyl-l-butene does. A common feature of ethylcyclopentene and 2-methyl- 1-butene is the presence of an ethyl group at the double bond. Since the coupling of a neopentylidene ligand to a 2-methyl-1 -butene molecule can be rationalized by a mechanism involving initial allylic C - H bond activation at the methylene position (Scheme 3.6c), an analogous proposed mechanism such as that shown below may be invoked for the ethylcyclopentene reaction. 99 W=CHCMe 3 + W w Unlike neohexene and the cycloalkenes above, the reactions of [Cp*W(NO)(=CHCMe3)] with acyclic alkenes containing allylic hydrogens are dominated by C - H bond activation pathways. In fact, no tungstacyclobutane complexes are detected from the reaction of this species with 2-methyl-1-butene. Instead, the major organometallic products are allyl neopentyl complex 3.18, allyl hydride complex 3.19, and diene complex 3.20 (eq 3.7). A small amount of an organic product whose mass equals the sum of a neopentylidene ligand and a molecule of 2-methyl-1-butene is also generated. The reaction of [Cp*W(NO)(=CHCMe3)] with 2-methyl-2-butene similarly proceeds primarily by allylic C - H activation. However, unlike the 2-methyl-1-butene reaction, this leads to formation of the six-membered metallacycles 3.21a-c as the major products (eq 3.8), formed by an unusual coupling of two molecules of 2-methyl-2-butene. The formation of complexes 3.18-3.21 is significant, for it represents the first clear-cut examples of alkene C - H bond activation by a transition-metal alkylidene complex. 4 0 Further, it shows that activation of allylic C - H bonds is preferred, as would be expected on the basis of bond dissociation energies,41 and that allylic C - H activation is a facile process for unstrained alkenes. Thus, whether an alkene wil l undergo [2+2] cycloaddition and/or C - H activation reactions with [Cp*W(NO)(=CHCMe 3)] depends on the degree of strain of the alkene double bond and the presence and accessibility-of allylic hydrogens. 100 3.5 References and Notes (1) Fischer, E. O.; Maasbol, A . Angew. Chem. Int. Ed. Engl. 1964, 3, 580. (2) For reviews, see: (a) Fischer, E. O. Adv. Organomet. Chem. 1976, 14, 1. (b) Gallop, M . A . ; Roper, W. R. Adv. Organomet. Chem. 1986, 25, 121. (c) Dotz, K . H . Angew. Chem., Int. Ed. Engl. 1984, 23, 587. (d) Wulff, W. D.; Tang, P. C.; Chan, K . S.; McCallum, J. S.; Yang, D. C ; Gilbertson, S. R. Tetrahedron 1985, 41, 5813. (3) For reviews, see: (a) Schrock, R. R. In Reactions of Coordinated Ligands; Braterman, P. S.; Ed.; Plenum: New York, 1986; Vo l . 1, pp 221-283. (b) Feldman, J.; Schrock, R. R. Prog. Inorg. Chem. 1991, 39, 1. (c) Beckhaus, R. Angew. Chem. Int. Ed. Engl. 1997, 36, 686. (4) (a) Dotz, K . H . Pure Appl. Chem. 1983,55 , 1689. (b) Hegedus, L. S.; McGuire, M . A . ; Schultze, L . M . ; Yijun, C ; Anderson, O. P. J. Am. Chem. Soc. 1984,106, 2680. (c) Brookhart, M . ; Studabaker, W. B . Chem. Rev. 1987, 87, 411. (5) Nugent, W. A . ; Mayer, J. M . Metal-Ligand Multiple Bonds; John Wiley & Sons: New York, 1988; Chapter 3. (6) For leading references on the metathesis reaction, see: (a) Grubbs, R. H . In Comprehensive Organometallic Chemistry; Wilkinson, G., Ed.; Pergamon Press, Ltd.: New York, 1982; Vo l . 8, pp 499-551. (b) Leconte, M . ; Basset, J. M . ; Quinard, F.; Larroche, C. In Reactions of Coordinated Ligands; Braterman, P. S., Ed.; Plenum: New York, 1986; Vo l . 1, pp 371-420. (c) Schrock, R. R. Pure Appl. Chem. 1994, 66, 1447. (d) Schuster, M . ; Blechert, S. Angew. Chem. Int. Ed. Engl. 1997, 36, 2036. (e) Grubbs, R. FL; Chang, S. Tetrahedron 1998,54 , 4413. (7) (a) Grubbs, R. H . ; Tumas, W. Science 1989, 243, 907. (b) Bazan, G. C ; Khosravi, E.; Schrock, R. R.; Feast, W. J.; Gibson, V . C ; O'Regan, M . B. ; Thomas, J. K . ; Davis, W. M . J. Am. Chem. Soc. 1990,112, 8378. (c) Schrock, R. R. Acc. Chem. Res. 1990, 23, 101 158. (d) Bazan, G. C ; Oskam, J. FL; Cho, H.-N. ; Park, L. Y . ; Schrock, R. R. J. Am. Chem. Soc. 1991,773, 6899. (e) Conticello, V . P.; Gin, D. L. ; Grubbs, R. H . J. Am. Chem. Soc. 1992, 114, 9708. (a) Wagener, K . B.; Nel, J. G.; Konzelman, J.; Boncella, J. M . Macromolecules 1990, 23, 5155. (b) Wagener, K . B. ; Smith, D. W. Jr. Macromolecules 1991, 24, 6073. (c) Wagener, K . B. ; Boncella, J. M . ; Nel, J. G. Macromolecules 1991, 24, 2649. (a) Schlund, R.; Schrock, R. R.; Crowe, W. E. J. Am. Chem. Soc. 1989,111 , 8004. (b) Bazan, G. C ; Khosravi, E.; Schrock, R. R.; Feast, W. J.; Gibson, V . C ; O'Regan, M . B.; Thomas, J. K . ; Davis, W. M . J. Am. Chem. Soc. 1990,112, 8378. (c) Park, L. Y . ; Schrock, R. R.; Stieglitz, S. G.; Crowe, W. E. Macromolecules 1991, 24, 3489. (d) Masuda, T.; Mishima, K . ; Fujimori, J.; Nishida, M . ; Muramatsu, H . ; Higashimura, T. Macromolecules 1992, 25, 1401. (e) Makio, H . ; Masuda, T.; Higashimura, T. Polymer 1993, 34, 2218. (f) Fox, H . H . ; Wolf, M . O.; O'Dell , R.; Lin, B . L . ; Schrock, R. R.; Wrighton, M . S. J. Am. Chem. Soc. 1994,116, 2827. (g) Schrock, R. R.; Luo, S.; Lee, J. C ; Zanetti, N . C ; Davis, W. M . J. Am. Chem. Soc. 1996, 118, 3883. (a) Fu, G. C ; Grubbs, R. H . / . Am. Chem. Soc. 1992,114, 5426. (b) Fu, G. C ; Grubbs, R. H . J. Am. Chem. Soc. 1992,114, 7324. (c) Fu, G. C ; Grubbs, R. H . J. Am. Chem. Soc. 1993,115, 3800. (d) Fujimura, O.; Fu, G. C ; Grubbs, R. H . J. Org. Chem. 1994, 59, 4029. (e) Kim, S. H . ; Bowden, N . ; Grubbs, R. H . J. Am. Chem. Soc. 1994,116, 10801. (f) Grubbs, R. H . ; Miller, S. J.; Fu, G. C. Acc. Chem. Res. 1995, 28, 446. (g) Armstrong, S. K . J. Chem. Soc, Perkin Trans. 1998,1, 371. (a) Brown-Wens ley, K . A . ; Buchwald, S. L. ; Cannizzo, L. ; Clawson, L. ; Ho, S.; Meinhardt, D.; Stille, J. R.; Strauss, D.; Grubbs, R. H . Pure Appl. Chem. 1983, 55, 1733. (b) Petasis, N . A . ; Bzowej, E. I. J. Am. Chem. Soc. 1990,772 , 6392. (c) Bazan, G. C ; Schrock, R. R.; O'Regan, M . B. Organometallics 1991,10, 1062. (d) Fujimura, O.; Fu, G. C ; Rothemund, R. W. K . ; Grubbs, R. H . J. Am. Chem. Soc. 1995, 777, 102 2355. (e) Petasis, N . A . ; Lu, S.-P.; Bzowej, E. I.; Fu, D.-K. ; Staszewski, J. P.; Akritopoulou-Zanze, I.; Patane, M . A . ; Hu, Y . - H . Pure Appl. Chem. 1996, 67, 667. (12) (a) Carter, E. A. ; Goddard, W. A . J. Am. Chem. Soc. 1986,108, 2180. (b) Carter, E. A . ; Goddard, W. A . J. Am. Chem. Soc. 1986, 705; 4746. (13) (a) Clark, G. R.; Hoskins, S. V . ; Jones, T. C ; Roper, W. R. J. Chem. Soc, Chem. Commun. 1983, 719. (b) Brothers, P. J.; Roper, W. R. Chem. Rev. 1988, 88, 1293. (14) Casey, C. P.; Vosejpka, P. C ; Askham, F. R. J. Am. Chem. Soc. 1990, 772, 3713. (15) (a) Legzdins, P.; Rettig, S. J.; Veltheer, J. E. J. Am. Chem. Soc. 1992, 774, 6922. (b) Legzdins, P.; Rettig, S. J.; Veltheer, J. E.; Batchelor, R. J.; Einstein, F. W. B. Organometallics 1993, 72, 3575. (16) Legzdins, P.; Rettig, S. J.; Veltheer, J. E.; Young, M . A . ; Batchelor, R. J.; Einstein, F. W. B . Organometallics 1995, 72, 407. (17) (a) Legzdins, P.; Sayers, S. F. Organometallics 1996, 75, 3907. (b) Legzdins, P.; Sayers, S. F. Chem. Eur. J. 1997, 3, 1579. (18) Tran, E.; Legzdins, P. J. Am. Chem. Soc. 1997, 779, 5071. (19) Sayers, S. F. Ph.D. Dissertation, The University of British Columbia, 1997. (20) (a) Eight Peak Index of Mass Spectra; 2 n d Ed.; Mass Spectrometry Data Center: Reading, U K ; 1974; Vo l . 2. (b) The NIST/EPA/NIH Online Mass Spectra Database; U . S. Department of Commerce, 1995. (21) For examples, see: (a) Kie l , W. A. ; Lin, G. -Y. ; Constable, A . G.; McCormick, F. B.; Strouse, C. E.; Eisenstein, O.; Gladysz, J. A . J. Am. Chem. Soc 1982,104, 4865. (b) Toreki, R.; Schrock, R. R.; Davis, W. M . J. Am. Chem. Soc. 1992,114, 3367. (c) Johnson, L . K . ; Frey, M . ; Ulibarri, T. A . ; Virgi l , S. C ; Grubbs, R. H . ; Ziller, J. W. J. Am. Chem. Soc. 1993, 775, 8167. 103 (22) For comparison, the 'JCH values of idealized sp 2 -CH bonds are typically 140-150 Hz. (23) (a) Schrock, R. R. Acc. Chem. Res. 1979,12, 98. (b) Pedersen, S. F.; Schrock, R. R. J. Am. Chem. Soc. 1982,104, 7483. (c) Brookhart, M . ; Green, M . L. H . ; Wong, L . - L . Prog. Inorg. Chem. 1988, 36, 1. (24) Gibson, V . C. J. Chem. Soc, Dalton Trans. 1994, 1607. (25) Tran, E.; Legzdins, P. Unpublished results. (26) The approximate pKa values of phenylacetylene, methyl acetate, and dimethylamine are 29, 25, and 35, respectively. See: Bordwell, F. G. Acc. Chem. Res. 1988, 21, 456. (27) This polymer is also unaffected by strong bases (e.g. N a O H / H 2 0 and KOH/EtOH). It, however, disintegrates upon contact with concentrated sulfuric acid. (28) For examples, see: (a) Schrock, R. R.; DePue, R.; Feldman, J.; Schaverien, C. J.; Dewan, J. C.; Liu, A . H . J. Am. Chem. Soc. 1988, 110, 1423. (b) Grubbs, R. H. ; Gilliom, L. R. J. Am. Chem. Soc. 1986,108, 733. (c) van Doom, J. A . ; van der Heijden, H . ; Orpen, A . G. Organometallics 1995,14 , 1278. (29) For example, treatment of Cp*W(NO)X 2 [X = C l , I] with 1,3-dimagnesiopropane or 1,3-dilithiopropane simply results formation of the monohalide dimer, [Cp*W(NO)X] 2 , in high yield. Tran, E. Unpublished results. (30) Wallace, K . C ; Liu , A . H. ; Dewan, J. C ; Schrock, R. R. J. Am. Chem. Soc. 1988,110, 4964. (31) (a) Collman, J. P.; Hegedus, L. S.; Norton, J. R.; Finke, G. R. Principles and Applications of Organotransition Metal Chemistry, University Science Books: M i l l Valley, C A , 1987; p. 176. (b) Ipaktschi, J.; Mirzaei, F.; Demuth-Eberle, G. J.; Beck, J.; Serafin, M . Organometallics 1997,16, 3965. (c) Chapter 4 of this thesis. 104 (32) (a) Hunter, A . D.; Legzdins, P.; Nurse, C. R.; Einstein, F. W. B. ; Will is , A . C. J. Am. Chem. Soc. 1985, 707, 1791. (b) Christensen, N . J.; Hunter, A . D.; Legzdins, P. Organometallics 1989, 8, 930. (33) Faller, J. W.; Nguyen, J. T.; Ellis, W.; Mazzieri, M . R. Organometallics 1993,12, 1434. (34) Greenhough, T. J.; Legzdins, P.; Martin, D. T.; Trotter, J. Inorg. Chem. 1979,11, 3268. (35) (a) Ward, Y . D. , Villanueva, L . A . ; Allred, G. D.; Payne, S. C.; Semones, M . A . ; Liebeskind, L . S. Organometallics 1995, 14, 4132. (b) Villanueva, L . A . ; Ward, Y . D.; Lachicotte, R.; Liebeskind, L. S. Organometallics 1996,15, 4190. (36) (a) Erker, G.; Engel, K . ; Dorf, U . ; Atwood, J. L . ; Hunter, W. E. Angew. Chem. Int. Ed. Engl. 1982, 21, 914. (b) Nakamura, A . J. Organomet. Chem. 1990, 400, 35 and references therein. (37) Christensen, N . J.; Legzdins, P.; Einstein, F. W. B.; Jones, R. H . Organometallics 1991, 10, 3070. (38) Brunet, N . ; Debad, J. D.; Legzdins, P.; Trotter, J.; Veltheer, J. E.; Yee, V . C. Organometallics 1993, 12, 4572. (39) McLain, S. J.; Wood, C. D.; Schrock, R. R. J. Am. Chem. Soc. 1976, 99, 3519. (40) Although alkene C - H activation by transition-metal alkylidene complexes is unprecedented, activation of arene and even alkane C - H bonds by these species has previously been reported, see: (a) van der Heijden, H. ; Hessen, B . J. Chem. Soc, Chem. Commun. 1995, 145. (b) Coles, M . P.; Gibson, V . C ; Clegg, W.; Elsegood, M . R. J.; Porrelli, P. A . Chem. Commun. 1996, 1963. (c) Cheon, J.; Rogers, D. M . ; Girolami, G. S. J. Am. Chem. Soc. 1997,119, 6814. (d) Reference 18. (41) McMil len, D. F.; Golden, D. M . Ann. Rev. Phys. Chem. 1982, 33, 493. 105 CHAPTER 4 Selective Activation of Aliphatic, Aromatic, and Benzylic C - H Bonds by [Cp*W(NO)(=CHCMe3)] and Related Complexes 4.1 Introduction 105 4.2 Experimental Section 107 4.3 Results and Discussion 123 4.4 Conclusions 148 4.5 References and Notes 150 4.1 Introduction The intermolecular activation of C - H bonds,1 especially those of alkanes, by soluble organometallic complexes has been the subject of much experimental2 and theoretical3 study in the last 20 years. Much of the impetus for this work has been provided by the search for reagents which can selectively convert alkanes, an abundant and cheap source of hydrocarbon feedstocks, into more valuable commodity chemicals. Although a large number of systems are now known which can effect the activation of C - H bonds intermolecularly, most are nonselective and involve the photochemical or thermal generation of a high-energy reactive intermediate.4 In general, organometallic complexes that can effect the activation of C - H bonds intermolecularly are of four main groups. First, there is the largest group which consists of low-valent late transition-metal complexes, and involves an oxidative-addition mechanism.5 Second is a group that consists of high-valent early transition-metal complexes,6 lanthanide complexes, and actinide complexes, and involves a o-bond metathesis mechanism. In the 106 third and fourth group, the activation occurs as a result of H-atom abstraction by metallaradicals9 and C - H bond addition across a M=X bond (X = O , 1 0 N R , 1 1 " 1 5 C H R 1 6 ) . Once rare, the reaction between a C - H bond and a metal-ligand multiple bond has flourished in recent years. However, while reactions of this type have become well established for transition-metal imido complexes, with specific examples now known for T i , 1 1 Z r , 1 2 V , 1 3 Ta, 1 4 and W , 1 5 there are at present few transition-metal oxo and alkylidene complexes that can effect the cleavage of C - H bonds intermolecularly.16 Fewer still are alkylidene complexes which can effect the cleavage of alkane C - H bonds. Indeed, prior to the preliminary report160 of the work described in this chapter, alkylidene complexes were only known to activate aliphatic C - H bonds intramolecularly.17'18'19'20 In the preceding chapter, it was shown that thermolysis of Cp*W(NO)(CH2CMe3)2 leads to loss of neopentane and formation of the 16-electron neopentylidene complex, [Cp*W(NO)(=CHCMe3)], as a highly reactive intermediate, which (a) can be trapped by a range of Lewis bases to yield the corresponding base adducts as the anti rotamer, (b) undergo addition reactions with the N - H and O - H bonds of amines and alcohols, respectively, and (c) react with alkenes to give novel [2+2] cycloaddition and/or C - H activation products depending on the alkene structure. The objective of this chapter is to report the full details of the reactivity of [Cp*W(NO)(=CHCMe 3)] with the C - H bonds of Me 4 Si , alkanes, and arenes. As will soon be seen, these C - H activation reactions occur with some selectivity and lead to an array of C - H activation products depending on the alkane and arene structure. The reactivity of the neopentylidene complex, [CpW(NO)(=CHCMe 3)], and benzylidene complexes, [Cp*W(NO)(=CHAr)], with representative aliphatic and aromatic C - H bonds will also be reported and compared to that of the Cp* neopentylidene system. 107 4.2 Experimental Section 4.2.1 Methods The synthetic methodologies employed throughout this thesis are described in detail in Section 2.2.1. G C - M S analyses were performed by Ms. L . Madilao of the U B C mass spectrometry facility using a Carlo Erba G C and Kratos MS-80 unit. 3 1 P and 1 9 F chemical shifts were recorded relative to P(OMe) 3 (5 141.0 in C 6 D 6 ) and C F 3 C 0 2 H (6 0.00) standards. A l l thermolyses were carried out in thick-walled glass vessels or J. Young N M R tubes immersed in an oil bath at 60-75 °C for ca. 1.5-2 d unless otherwise noted. 4.2.2 Reagents The organometallic precursors CpW(NO)(CH 2CMe 3)2 (2.1), Cp*W(NO)(CH 2 CMe 3 ) 2 (2.4), Cp*W(NO)(CH 2 CMe 3 ) (CH 2 CMe 2 Et) (2.7), and Cp*W(NO)(CH 2 CMe 3 )(CH 2 Ph) (2.15) were prepared as described in Section 2.2.2. M e 2 N H (Aldrich, 99.9%) was used as received. 1,1,2,2-Tetramethylcyclopropane (Wiley Organics), tetramethylsilane, cyclopentane, neohexane, methylcyclopentane, methylcyclohexane, ethylcyclohexane, anisole, toluene, and mesitylene (Aldrich) were vacuum-transferred from sodium. PMe 3 , pentane, cyclohexane, o-xylene, m-xylene, and/>xylene (Aldrich) were vacuum-distilled from sodium benzophenone ketyl. ot,a,a-Trifluorotoluene (Aldrich) was dried over activated 4 A molecular sieves. 4.2.3 Preparation of Cp*W(NO) (CH 2 CMe 3 ) (CH 2 S iMe 3 ) (2.9) In the glovebox, a Pyrex vessel was charged with 2.4 (48 mg, 0.098 mmol) and tetramethylsilane (1 mL). The resulting wine-red solution was heated at ca. 70 °C for 2 d, during which time its color darkened. The organic volatiles were then removed in vacuo to obtain a red-black solid, which was extracted into pentane and filtered through Celite ( l x l cm). Concentration of the filtrate followed by cooling to -30 °C for 2 d provided 2.9 (45 mg, 108 90% yield) as maroon crystals. The identity of 2.9 was confirmed by comparison of its ' H and 1 3 C N M R spectra to those of an authentic sample.21 4.2.4 Preparation of C p * W ( N O ) [ C H 2 C M e C M e 2 C H ] (4.1) To Pyrex vessel containing 2.4 (35 mg, 0.071 mmol) was added 1,1,2,2-tetramethylcyclopropane (2 mL) by vacuum transfer. The resulting wine-red mixture was thawed and heated at ca. 70 °C for 2 d, during which time it became orange-yellow. The organic volatiles were then removed in vacuo to obtain an oily residue, which was redissolved in hexanes and filtered through alumina 1 (1x1 cm). The alumina column was washed with E t 2 0 until the filtrate was colorless. Concentration of the combined filtrates followed by cooling to -30 °C for several days afforded 4.1 (10 mg, 32% yield) as bright yellow crystals. Anal. Calcd. for C , 7 H 2 7 N O W : C, 45.85; H , 6.11; N , 3.15. Found: C, 45.55; H , 6.02; N , 3.00. IR (cm"1) v N 0 1588 (s). MS 445 [ M + , 1 8 4 W ] . ' H N M R (400 M H z , C 6 D 6 ) 8 0.72 (dd, V H H = 4.3, V H H = 1-2, 1H, CHsynH to Cp*), 1.16 (br s, 1H, CH), 1.63 (s, 3H, CMe2), 1.68 (s, 15H, C 5 Me 5 ) , 2.17 (s, 3H, CMz 2 ) , 2.31 (s, 3H, CH 2 CMe) , 3.25 (d, V H H = 4.3, 1H, Cf7 a n t i H to Cp*). 1 3 C{ 'H} N M R (125 M H z , C 6 D 6 ) 5 10.4 (C 5 Me 5 ) , 18.8 (CH 2 CMe), 27.5 (1 CMe 2 ) , 30.1 (1 CMe 2 ) , 55.8 ( V C W = 37, W C H 2 ) , 80.9 (CMe 2 ), 83.0 (WCH), 104.3 (C 5 Me 5 ) , 106.0 (CH 2 CMe). NOEDS (400 M H z , C 6 D 6 ) irrad. at 0.72 ppm, NOEs at 8 1.16, 1.68, 3.25; irrad. at 1.16 ppm, NOEs at 8 0.72, 1.68; irrad. at 1.63 ppm, NOEs at 8 1.16, 2.17; irrad. at 1.68 ppm, NOEs at 8 0.72, 1.16; irrad. at 2.17 ppm, N O E at 8 1.63; irrad. at 2.31 ppm, N O E at 8 3.25; irrad. at 3.25 ppm, NOEs at 8 0.72, 2.31. 4.2.5 Preparation of Cp*W(NO)(rj2-alkene)(PMe3) Complexes [alkene = cyclopentene (4.2), cyclohexene (4.3), neohexene (4.4)] The reactions leading to complexes 4.2-4.4 were carried out in a similar manner. In each case, the neopentylidene complex Cp*W(NO)(=CHCMe 3)(PMe 3) (3.4) (-10-30% yields) was detected as a minor product by *H and 3 1 P{ 'H} N M R analyses of the crude reaction 109 mixture. The preparation of 4.2 is described as a representative example. To a Pyrex vessel containing 2.4 (56 mg, 0.11 mmol) was added cyclopentane (9 mL) and PMe 3 (5-10 equiv) by vacuum transfer. The resulting wine-red mixture was thawed and heated at ca. 70 °C for 2 d, during which time it became a dark yellow solution. Next, the organic volatiles were removed in vacuo to obtain a dark yellow oil, which was extracted into 5:1 hexanes/Et20 (2 mL) and filtered through alumina 1(2 x 1 cm). The alumina column was washed with additional 5:1 hexanes/Et20 until the filtrate was colorless. Concentration of the combined filtrates followed by cooling to -30 °C for 2 d afforded 4.2 (25 mg, 44% yield) as pale yellow plates. 4.2.5.1 Complex 4.2. Anal. Calcd. for C 1 8 H 3 2 N O P W : C, 43.83; H , 6.54; N , 2.84. Found: C, 43.86; H , 6.61; N , 2.76. IR (cm"1) v N 0 1522 (s). MS 493 [ M + , 1 8 4 W]. ' H N M R (400 MHz, C 6 D 6 ) 8 1.06 (m, 1H, a=CH), 1.19 (d, 2Jm = 8.4, 9H, PMe 3 ) , 1.58 (m, l H , y C H 2 ) , 1.66 (s, 15H, C 5 Me 5 ) , 1.72 (m, 1H, y C H 2 ) , 1.99 (m, 1H, a ' =CH), 2.34 (dd, 2 J m = 6.6, V H H = 12.4, 1H, p CH 2 ) , 2.64 (m, 1H, (3' CH 2 ) , 3.00 (sept, 1H, (3' CH 2 ) , 3.26 (sept, 1H, (3 CH 2 ) . ' ^ { ' H } N M R (75 M H z , C 6 D 6 ) 5 10.5 (C5M25), 16.8 (d, [JCP = 29, PMe 3 ) , 23.2 (yCH 2 ) , 34.9 (p' CH 2 ) , 35.4 (d, 3 J C P = 4, p CH 2 ) , 44.2 ( ' j C w = 40, a ' =CH), 46.1 (d, 2JC? = 13, a =CH), 103.0 (C 5 Me 5 ) . 3 I P { ' H } N M R (121 M H z , C 6 D 6 ) 8-13.2 ('jpW = 364). 4.2.5.2 Complex 4.3 was prepared by heating 2.4 (81 mg, 0.16 mmol) in a cyclohexane solution (8 mL) of PMe 3 (5-15 equiv). It was crystallized from a 3:1 mixture of pentane/Et20 at -30 °C and isolated as pale yellow needles (32 mg, 51%) yield). Anal. Calcd. for C19H34NOPW: C, 44.98; H , 6.76; N , 2.76. Found: C, 44.20; H , 6.41; N , 2.75. IR (cm 1) VNO 1518 (s). H R M S Calcd. for C i 9 H 3 4 N O P W : 507.2881 [NT1", 1 8 4 W ] . Found: 507.1888. ' H N M R (500 MHz, C 6 D 6 ) 8 0.8 (m, 1H, a ' =CH), 1.22 (d, 2JHP = 9H, PMe 3 ) , 1.5 (m, 1H, a=CH), 1.6 (m, 2H, f CH 2 ) , 1.67 (s, 15H, C 5 M e 5 ) , 1.8 (m, 2H, y C H 2 ) , 2.4 (m, 1H, p ' CH 2 ) , 2.8 (m, 2H, p CH 2 ) , 3.1 (m, 1H, p ' CH 2 ) . 1 3 C { ' H } N M R (125 M H z , C 6 D 6 ) 8 10.3 (C 5 Me 5 ) , 17.4 (d, 2JCP = 29, PMe 3 ) , 24.2 (yCH 2 ) , 25.3 (y' C H 2 ) , 28.3 (p CH 2 ) , 30.8 (d, 3 J C P = 4, p ' CH 2 ) , 36.9 (lJcv/ = 39, a CH), 40.0 (d, 2JCP = 12, a ' CH), 103.0 (C 5 Me 5 ) . 3 1 P{ 'H} N M R (202 M H z , C 6 D 6 ) 8-13.2 (s, lJPW = 313). 110 4.2.5.3 Complexes 4.4 were prepared as a 5.7:1 mixture of 4.4 s y n and 4.4 a n ti by heating 2.4 (66 mg, 0.13 mmol) in a neohexane solution (8 mL) of PMe 3 (0.38 mmol, 3 equiv). The two rotamers were isolated together as yellow blocks (38 mg, 56% yield) from a 4:1 mixture of hexanes/Et20 at -30 °C. Anal. Calcd. for C 1 9 H 3 6NOPW: C, 44.80; H , 7.13; N , 2.75. Found: C, 44.95; H , 7.09; N.2.68. IRCcm'VNO 1513(s). MS 509 [ M + , 1 8 4 W ] . l H N M R (400 M H z , C 6 D 6 ) 5 4.4 s y n: 0.38 (m, IH , =C/ / s y n H), 1.00 (s, 9H, CMe 3 ) , 1.33 (d, 2 J H P = 8.4, 9H, PMe 3 ) , 1.73 (s, 15H, C 5 Me 5 ) , 2.3 (m, 2H, =C/ / a n t i H, =C//CMe 3 ) . 4.4 a n t i: -0.30 (m, IH , =C/ / s y n H), 1.16 (d, 2 J H p = 8.6, 9H, PMe 3 ) , 1.34 (s, 9H, CMe 3 ) , 1.43 (m, 2H, =Cr7 a n t jH, =Gr7CMe 3), 1.67 (s, 15H, C 5 Me 5 ) . 1 3 C{ 'H} N M R (75 M H z , C 6 D 6 ) 8 4.4 s y n: 10.9 (C 5 Me 5 ) , 18.3 (d, lJCP = 30, PMe 3 ) , 25.1 ('Jew = 33, =CH 2), 34.7 (CMe 3), 35.2 (d, 2JC? = 2, CMe 3 ) , 57.9 (d, lJCP = 12, =CHCMe 3 ) , 103.9 (C 5 Me 5 ) . 4.4 a n t i: 10.5 (C 5 Me 5 ) , 16.0 (d, % P = 32, PMe 3 ) , 27.2 (d, 2 J C p = 22, =CH 2), 32.3 (CMe 3), 33.1 (CMe 3 ) , 55.7 (=CHCMe 3), 103.0 (C 5 Me 5 ) . 3 1 P{ 'H} N M R (202 M H z , C 6 D 6 ) 8 4.4 s y n: -17.3 ( 'Jp w = 358). 4.4 a n t i: -9.6 ( ' j P W = 361). NOEDS (400 M H z , C 6 D 6 ) 4.4 s y n: irrad. at 0.38 ppm, NOEs at 8 1.00, 1.73,2.27; irrad. at 1.00 ppm, NOEs at 8 0.38, 1.33, 1.73,2.27; irrad. at 1.73 ppm, NOEs at 8 0.38, 1.00, 1.33. 4.4 a n t i: irrad. at-0.30 ppm, NOEs at 5 1.16, 1.43; irrad. at 1.16 ppm, NOEs at 8-0.30, 1.67; irrad. at 1.34 ppm, N O E at 8 1.43; irrad. at 1.67 ppm, NOEs at 8 1.16. 4.2.6 Thermolysis of 2.4 in Cyclohexane in the Presence of PMe 3 : Effect of Varying PMe 3 Concentration Into each of two J. Young N M R tubes was added 1.0 mL of a 0.020 M solution of 2.4 in cyclohexane. 6 quiv of PMe 3 were then added to one tube and 12 equiv to the other. The resulting solutions were heated side-by-side in 70 °C bath and the reaction monitored periodically for 2 d by 3 1 P{ 'H} N M R spectroscopy. In each case, formation of 3.4 and 4.3 as the sole products was observed; these were formed in a constant relative yield of 25% and 75% and 49% and 51%), respectively. I l l 4.2.7 Kinetics of the Thermolysis of 2.4 in Neohexane in the Presence of PMe3 Two J. Young N M R tubes, each containing 2.4 (16 mg, 0.033 mmol), neohexane (0.80 mL), PMe3 (3.5 equiv), and a capillary tube of P(o-tolyl)3 in CeD 6 , were heated to 70 °C in a constant-temperature bath and the reaction monitored periodically for 7 d by 3 1 P{ 'H} N M R spectroscopy. In each case, parallel formations of 3.4 and 4.4 s y n were observed at early conversion. As the reaction progressed, the intensity of the signal due to 4 .4 a n t i appeared and increased at the expense of the intensity of the resonance due 4.4 s y n . The rate constant (average of the two runs) for the formation of 3.4 and 4.4 s y n was 4.22 x 10"5 s"1, while the rate constant for the isomerization of 4.4 s y n to 4 .4 a n t i was 7.41 x 10"6 s"1. 4.2.8 Thermolysis of Cp*W(NO)(CH 2CMe 3)(CH 2CMe 2Et) (2.7) in Neohexane in the Presence of PMe 3 This reaction was effected in a manner analogous to that described for reaction between 2.4, neohexane, and PMe3. To a Pyrex bomb containing 2.7 (19 mg, 0.038 mmol) was added neohexane and excess PMe 3 by vacuum transfer. The resulting wine-red mixture was thawed and heated at ca. 70 °C for 2 d, during which time it became orange. The organic volatiles were then collected by vacuum transfer and were examined by G C - M S , which showed the presence of neopentane and neohexane but not neohexene, dimethylcyclopentane, or 2,2,6,6-tetramethyloctane. Analysis of the crude reaction mixture by 3 1 P{ 'H} N M R spectroscopy indicated that 4.4 s y n and 4.4 a n t i were formed as the major products. Several minor products were also observed, which gave rise to 3 1 P resonances at 8 35.6, 33.3, -5.7, -5.8, and -23.0 (relative to 85% H 3 P 0 4 at 8 0.00) in a ratio of about 1:1:0.5:0.5:0.8 based on relative 3 I P N M R peak heights. The products with resonances at 8 -5.6, -5.8, and -23.0 were assigned as Cp*W(NO)(=CHCMe 3)(PMe 3) (3.4), Cp*W(NO)(=CHCMe 2Et)(PMe 3) (4.5), and Cp*W(NO)(PMe 3 ) 2 (4.6), respectively, on the basis of ' H and 3 1 P N M R spectral data of the crude product mixtures (see below, Section 4.2.9). In the case of 4.6, its identity was further confirmed by comparison of its l H and 3 1 P{ 'H} N M R spectra to those an of authentic sample prepared by Na/Hg reduction of Cp*W(NO)Cl 2 in THF in the presence of P M e 3 . 2 2 112 4.2.9 Thermolysis of Cp*W(NO)(CH 2CMe 3)(CH 2CMe2Et) (2.7) in THF in the Presence of PMe 3 This reaction was carried out in a manner analogous to that described above using THF as solvent. Analysis of the organic volatiles by GC-MS again showed peaks for neopentane and neohexane, but none for neohexene, dimethylcyclopentane, and 2,2,6,6-tetramethyloctane. Examination of the crude reaction mixture by L H and 3 1 P{ 'H} N M R spectroscopies again indicated a complex mixture of products. In this case, complexes 3.4, 4.5, and 4.6 were observed as the major (and only identified) products in a ratio of about 1:1:1 based on relative 3 1 P N M R peak heights. The minor products gave rise to 3 1 P signals in the 10-18 ppm region. Signals consistent with the presence of 4.4 were not observed. The N M R mixture was then taken to dryness, extracted into 3:1 hexanes/Et20, and filtered through alumina I (1 x 1 cm). The alumina column was washed with additional 3:1 hexanes/Et20 until the filtrate was colorless. Next, the combined filtrates were reduced in volume and stored at -30 °C for several days to induce the deposition of 3.4, 4.5, and 4.6 as brown-orange crystals in a ratio of -1:1:3. Cp*W(NO)(=CHCMe2Et)(PMe3) (4.5). MS 509 [ M + , 1 8 4 W ] . ' H N M R (500 M H z , C 6 D 6 ) 8 1.06 (d, 2 J H P = 9.0, 9H, PMe 3 ) , 1.40 (s, 3H, CMeMe), 1.46 (s, 3H, CMeMe), 1.86 (s, 15H, C 5 M e 5 ) , 11.27 (d, VHp = 3.3, W=CH), CH2CH3 unobserved due to peak overlap. 1 3 C{ 'H} N M R (125 M H z , C 6 D 6 ) 8 9.7 ( C C H 2 C H 3 ) , 10.71 (C 5 Me 5 ) , 19.4 (d, ' j C P = 32, PMe 3 ) , 28.7 (CMeMe), 28.8 (CMeMe), 36.6 (CH 2 CH 3 ) , 54.0 (CMeMe), 106.1 (C 5 Me 5 ) . 3 1 P{ 'H} (121 M H z , C 6 D 6 , referenced to external H 3 P 0 4 ) 8 -7.6 ( 'y P W = 445). Cp*W(NO)(PMe 3) 2 (4.6). MS 501 [ M + , 1 8 4 W ] . ' H N M R (500 M H z , C 6 D 6 ) 8 (d, 2JHP = 7.2, 18H,PMe 3 ) , 1.92 (s, 15H,C 5 Me 5 ) . 1 3 C{ 'H} N M R (125 M H z , C 6 D 6 ) 8 11.9 (C 5 Me 5 ) , 24.0 (m, 2, 6, PMe 3 ) , 98.8 (C 5 Me 5 ) . 3 1 P{ 'H} (121 M H z , C 6 D 6 , referenced to external H 3 P 0 4 ) 8-22.7 ( ' / P W = 453). 113 4.2.10 Preparation of Cp*W(NO)(cyclohexyl)(NMe2) (4.7) In a manner similar to the preparation of alkene complex 4.3, a cyclohexane solution (10 mL) of 2.4 (60 mg, 0.12 mmol) and M e 2 N H (5-10 equiv from a calibrated gas bulb) was heated at ca. 70 °C for 2 d, during which time its color changed from wine-red to dark yellow. The organic volatiles were then removed in vacuo to obtain a yellow-brown oil , which was triturated with pentane ( 3 x 5 mL), extracted into 3:1 pentane/Et20, and filtered through alumina 1(1 x 1 cm). The alumina column was washed with additional 3:1 pentane/Et20 until the filtrate was colorless. Concentration of the combined filtrates followed by cooling to -30 °C for several days provided 4.7 (31 mg, 54 % yield) as large, pale yellow needles. The needles were spectroscopically pure, but a satisfactory elemental analysis could not be obtained. IR (CH 2 C1 2 , cm"1) v N O 1529 (s). MS 476 [ M + , 1 8 4 W ] . ' H N M R (500 M H z , C 6 D 6 ) 5 1.17 (m, IH, WCH) , 1.45 (m, 3H, C a /7H, Cb#H, CCHH), 1.66 (s, 15H, C 5 M e 5 ) , 1.72 (m, IH, C d H 2 ) , 1.82 (m, IH , C a H 2 ) , 1.98 (m, 2H, CbJTH, CJM), 2.27 (m, IH, C e H 2 ) , 2.63 (s, 3H, NMe a ) , 2.87 (br d, JHH = 14.0, IH, C e H 2 ) , 3.72 (s, 3H, NMe b ) . 1 3 C{ 'H} N M R (125 M H z , C 6 D 6 ) 8 9.4 (C5Me5), 27.8 (C a H 2 ) , 32.9 (C b H 2 ) , 33.0 (C C H 2 ) , 37.2 (C d H 2 ) , 40.0 (C e H 2 ) , 51.1 (NMe a), 52.1 (WCH, 'Jew = 94 Hz), 60.6 (NMe b), 109.9 (C 5 Me 5 ) . 4.2.11 Preparation of Cp*W(NO)(7j3-allyl)(H) Complexes [ff-allyl = 773-C6H9 (4.8), rf-C 7 H„ (4.9), jf-Cslll3 (4.10), rf-C5K9 (4.11)]. The reactions leading to complexes 4.8-4.11 were all carried out in a similar manner. The preparation of 4.8 is described as a representative example. To a Pyrex vessel containing 2.4 (46 mg, 0.094 mmol) was added methylcyclopentane (2 mL) by vacuum transfer. The resulting wine-red mixture was thawed and heated at ca. 70 °C for 2 d, during which time it became a brown-yellow solution. The organic volatiles were then removed in vacuo to obtain a brown-yellow oil, which was extracted into 4:1 hexanes/Et20 (0.5 mL) and filtered through alumina I (2 x 1 cm). The alumina column was washed with additional 4:1 hexanes/Et20 (3 114 mL) followed by a minimum amount of Et20 until the filtrate was colorless. The combined filtrates were concentrated at room temperature and then stored at -30 °C overnight to obtain 4.8e ndo and 4.8 e„do* as clusters of pale yellow needles (28 mg, 70% combined yield). Complete N M R characterization of 4.8e„do* could not be achieved due to inadequate signal-to-noise of its signals and to spectral overlap. 4.2.11.1 Complexes 4.8. Anal. Calcd. for C i 6 H 2 5 N O W : C, 44.56; H , 5.84; N , 3.25. Found: C, 44.68; H , 5.93; N , 3.29. IR (cm"1) V W H 2161, 1911; v N O 1565, 1548 (s). MS 429 [ M + - H 2 , 1 8 4 W ] . ' H N M R (400 MHz , C 6 D 6 ) 5 4.8e n do: -1-05 (s, lJHW = 119, WH), 0.68 (s, 1H, allyl C / / a n t i H ) , 1.69 (m, 1H, C a H 2 ) , 1.75 (s, 15H, C 5 Me 5 ) , 1.95 (m, 1H, C a H 2 ) , 2.44 (s, 1H, allyl CH), 2.51 (dd, 2 J H H = 8.2, V H H = 14.5, 1H, C b H 2 ) , 2.83 (m, 2H, Cbfflf, CCHH), 2.95 (m, 1H, C C H 2 ) , 3.06 (m, 1H, allyl CHsynH). 4.8endo*: -0.69 (s, '7Hw = 121, 1H, WH), 0.96 (br s, 1H, allyl CH 2 ) , 1.7 (m, 1H, allyl CH), 1.73 (s, 15H, C 5 Me 5 ) , 4.21 (br s, 1H, allyl CH 2 ) , other resonances were masked by those of major isomer. 1 3 C{ 'H} (75 M H z , CeDe) 8 4.8e mio: 10-7 (C5Me5), 23.8 (C a H 2 ) , 35.6 (C b H 2 ) , 35.9 (C C H 2 ) , 38.3 (allyl CH 2 ) , 80.3 (allyl CH), 104.7 (C 5 Me 5 ) , 125.3 (allyl 4° C). 4.8endo*: 10.4 (C 5 Me 5 ) , 23.2 (CH 2 ) , 33.9 (CH 2 ) , 36.2 (CH 2 ) , 46.2 (allyl CH 2 ) , 67.7 (allyl CH), 104.5 (C 5 Me 5 ) , 126.8 (allyl 4 °C). NOEDS (400 M H z , C 6 D 6 ) 4.8e n do: irrad. at -1.05 ppm, NOEs at 5 1.95, 2.44, 2.83; irrad. at 0.68 ppm, NOEs at 5 2.44, 3.05; irrad. at 1.75 ppm, NOEs at 8 -1.05, 0.68, 2.44; irrad. at 3.05 ppm, N O E at 8 0.68. 4.8 e n do*: irrad. at 0.96 ppm, N O E at 8 4.21; irrad. at 4.21, NOEs at 8 -0.69, 0.96. 4.2.11.2 Complex 4 .9 e n d o was prepared by heating 2.4 in methylcyclohexane. It was isolated as clear pale yellow rods (57% yield) from 5:1 pentane/Et20 at -30 °C. Anal. Calcd. for C i 7 H 2 7 N O W : C, 45.86; H , 6.11; N , 3.15. Found: C, 45.99; H , 6.19; N , 2.97. IR (cm"1) v W H 1905 (w); v N 0 1564 (s). MS 443 [ M + - H 2 , 1 8 4 W ] . ' H N M R (500 M H z , C 6 D 6 ) 8-0.73 (s, ' J H W = 123, 1H, WH), 0.31 (m, 1H, allyl C / / a n t i H ) , 1.39 (m, 1H, C b H 2 ) , 1.54 (m, 1H, C a H 2 ) , 1.71 (m, 1H, C b H 2 ) , 1.76 (s, 15H, C 5 Me 5 ) , 1.84 (m, 1H, C a H 2 ) , 2.28 (br s, 1H, allyl CH), 2.60 (m, 1H, allyl CHsynH), 2.68 (m, 2H, Q M I , CdHH), 2.91 (q, V H H = 7.9, C C H 2 ) , 115 2.97 (m, IH, C d H 2 ) . 1 3 C{ 'H} (75MHZ, C 6 D 6 ) 5 10.7 (C5Me5\ 22.1 (C a H 2 ) , 22.5 (C b H 2 ) , 28.8 (C C H 2 ) , 29.6 (C d H 2 ) , 41.3 (allyl CH 2 ) , 80.0 (allyl CH), 104.9 (C 5 Me 5 ) , 121.1 (allyl 4° C). 4.2.11.3 Complexes 4.10 were prepared as a mixture of 4.10exo and 4.10endo by heating 2.4 in ethylcyclohexane. They were isolated as brown-yellow blocks (37% combined yield) from a minimum amount of pentane and 2 drops of E t 2 0 at -30 °C. Complete spectroscopic characterization of 4.10e„do could not be achieved due to spectral overlap and to inadequate signal-to-noise of its N M R signals. Anal. Calcd. for C 1 8 H 2 9 N O W : C, 47.07; H , 6.36; N , 3.05. Found: C, 46.86; H , 6.42; N , 2.96. IR (cm - 1) 4.10exo v W H 1935 (m); v N 0 1575 (br s). 4.10endo: v W H 2172 (m). MS 457 [ M + - H 2 , l 8 4 W ] . ' H N M R (400 MHz , C 6 D 6 ) 8 4.10e x o: -0.92 (s, ' j H w = 125, IH , WH), 1.42 (m, IH, C b H 2 ) , 1.58 (m, 3H, CaHH, CCHR), 1.71 (s, 15H, C 5 Me 5 ) , 1.79 (m, IH, C b H 2 ) , 1.90 (m, IH, C d H 2 ) , 2.00 (m, IH, C C H 2 ) , 2.15 (m, 3H, C d / f f l , allyl Gr7 s y n H, CeHH), 2.44 (dd, IH, 2JHH = 3.1, V H H = 13.3, allyl C/Y a n t i H), 2.51 (m, IH, C e H 2 ) , 2.66 (dd, 3 7 H H = 7.8, V H H = 13.3, IH, allyl CH). 4.10e„do: -0.71 (s, ' j W H = 126, IH, WH), 0.60 (m, IH, allyl CHantiH), 2.87 (dd, 2 JHH = 3.3, V H H = 7.2, IH, allyl CHmK), 2.89 (m, IH, CH 2 ) , 4.69 (dd, 3 J H H = 7.2, V H H = 11-8, IH, allyl CH), other signals are obscured. 1 3 C{'H}(75 M H z , C 6 D 6 ) 8 4.10e x„: 10.7 (C 5 Me 5 ) , 27.1 (C a H 2 ) , 30.6 (C b H 2 ) , 32.6 (C C H 2 ) , 35.1 (C d H 2 ) , 37.2 ('yCw = 25, allyl C H 2 ) , 41.0 (C e H 2 ) , 93.2 (allyl CH), 103.8 (C 5 Me 5 ) , 109.0 (allyl 4 °C). 4.10endo: 10.5 (C5Me5), 27.0 (CH 2 ) , 30.5 (CH 2 ) , 32.5 (CH 2 ) , 33.0 (CH 2 ) , 40.9 (allyl CH 2 ) , 42.8 (CH 2 ) , 100.9 (allyl CH), 104.8 (C 5 Me 5 ) . NOEDS (400 M H z , C 6 D 6 ) 4.10exo: irrad. at-0.92 ppm, NOEs at 8 2.01, 2.15, 2.51; irrad. at 2.51 ppm, NOEs at 8-0.92, 2.15; irrad. at 2.44 ppm, N O E at 8 2.15; irrad. at 2.66 ppm, N O E at82.15. 4.2.11.4 Complex 4.11endo- This compound was prepared by thermolysis of 2.4 in pentane. Crystallization of the filtered crude reaction mixture typically gave 4.11c ndo in 2-5% yield as bright yellow crystals. 116 M S 417 [ M + - H 2 , 1 8 4 W ] . ' H N M R (500 M H z , C 6 D 6 ) 8-1.40 (s, V H W = 120, 1H, WH), 0.16 (dd, V H H = 1-3, V H H = 10.1, 1H, allyl C b H 2 ) , 1.09 (t, V H H = 7.0, 3H, C H 2 C f / 3 ) , 1.74 (s, 15H, C 5 Me 5 ) , 1.82 (m, 1H, a allyl C C H), 2.34 (m, 1H, C a / f 2 C H 3 ) , 2.48 (m, 1H, C a / / 2 C H 3 ) , 2.79 (m, 1H, allyl C b H 2 ) , 4.59 (m, 1H, p allyl C d H). nC{lB} N M R (125 M H z , C 6 D 6 ) 8 10.6 (C5Me5), 17.2 (CH 3 ) , 27.8 (allyl C a H 2 ) , 38.9 (allyl C b H 2 ) , 84.4 (a allyl C C H), 101.4 (p allyl C d H) , 104.5 (C 5 Me 5 ) . 4.2.12 Reactions of 2.4 with Arenes The reactions of 2.4 with aromatic solvents were all effected in a similar manner. In all cases, the crude products were subjected to *H N M R analysis (at either 400 or 500 MHz) prior to workup. The reaction with a,a,oc-trifluorotoluene is described as a representative example. 4.2.12.1 Reaction with oc,a,oc-Trifluorotoluene. A solution of 2.4 (19 mg, 0.39 mmol) in a,a,a-trifluorotoluene (2 mL) in a Pyrex vessel was heated at 70-75 °C for 2 d, during which time it became brown-red. The organic volatiles were then removed in vacuo, and an N M R sample in C6D 6 was prepared. The ' H N M R spectrum revealed the quantitative conversion of 2.4 into o-, m-, andp- Cp*W(NO)(CH 2 CMe 3 ) (C 6 H 4 CF 3 ) (4.13) in a ratio of-1:66:33, respectively (by integration of the methylene regions). 4.130 was identified by the doublets in the ' H N M R spectrum at 8 -2.68 (VH H = 11.2) and 5.05 (V H H = 11.6) due to the presence of a C H 2 C M e 3 group attached to a [Cp*W(NO)(aryl)] fragment. Next, the N M R mixture was taken to dryness, extracted into pentane, and then quickly filtered through a plug of alumina I (1 x 1 cm). Concentration of the filtrate followed by cooling to -30 °C for 2 d afforded 4.13m and 4.13p as dark red crystals (38% combined yield). Anal. Calcd. for C 2 2 H 3 0 F 3 N O W : C, 46.74; H , 5.35; N , 2.48. Found: C, 46.43; H , 5.29; N , 2.48. IR (cm - 1) v N 0 1575. MS 565 [ M + , 1 8 4 W ] . ' H N M R (300 M H z , CDC1 3) 8 4.13m: -2.58 (d, V H H = 10.9, 1H, W C H 2 ) , 1.12 (s, 9H, CMe 3 ) , 1.84 (s, 15H, C 5 Me 5 ) , 5.11 (d, V H H = 10.8, 1H, W C H 2 ) , 7.31 (t, V H H = 7.3, 1H, ArH), 7.38 (d, V H H = 7.3, 1H, ArH), 7.65 (s, 1H, ArH), 7.77 (d, V H H = 7.2, 1H, ArH). 4.13„: -2.61 (d, V H H = 10.4, 1H, W C H 2 ) , 1.12 (s, 117 9H, CMe 3 ) , 1.842 (s, 15H, C 5 Me 5 ) , 5.13 (d, 2 J H H = 10.4, IH, W C H 2 ) , 7.41 (d, 3 J H H = 7.4, 2H, ArH), 7.59 (d, V H H = 7.8, 2H, ArH). 1 3 C{ 'H} N M R (75 M H z , CDC1 3) 5 10.2 (C 5 Me 5 ) , 33.3 (CMe 3), 42.2 (CMe 3 ) , 111.5 (C 5 Me 5 ) , 123.6 (m, Ar CH), 123.8 (m, Ar CH), 124.7 (q, lJCF = 270, C F 3 of 4.12m), 127.7 (WCH 2 ) , 129.3 (q, V C F = 30, C C F 3 of 4.12m), 132.3 (d, 3JCF = 14, Ar CH), 133.0 (m, A r CH), 137.4 (Ar CH), 141.2 (Ar CH), 178.7 ( W C i p s o of 4.12m), 184.4 ( W C i p S 0 of 4.12p). C F 3 and C C F 3 resonances of 4.13p were unobserved. 1 9 F N M R (188 M H z , CDC1 3) 8 4.13m: 13.6 (s). 4.13p: 13.3 (s). 4.2.12.2 Reaction with Anisole cleanly produced a mixture of o-, m-, and p-Cp*W(NO)(CH 2 CMe 3 ) (C 6 H 4 OCH 3 ) (4.12) in a ratio of approximately 88:7:5, respectively. After removal of CeD6 from the N M R sample, the residue was redissolved in hexanes (1 mL), and then stored at -30 °C overnight to obtain o-, m-, andp-4.12 (12 mg, 60% combined yield) as wine-colored crystals. Because of the low relative yields of 4.12m and 4.12^, spectral data other than *H N M R were not obtained. Anal. Calcd. for C 2 2 H 3 3 N 0 2 W : C, 50.10; H , 6.31; N , 2.66. Found: C, 49.87; H , 6.61; N , 2.65. IR (cm - 1) v N 0 1564 (s). MS 527 [ M + , 1 8 4 W ] . ' H N M R (400 M H z , CDC1 3) 8 4.12c: -2.90 (d, 2 J H H = 12.0, IH, W C H 2 ) , 1.05 (s, 9H, CMe 3 ) , 1.83 (s, 15H, C 5 Me 5 ) , 3.89 (s, 3H, OCH 3 ) , 4.91 (d, V H H = 12.0, IH, W C H 2 ) , 6.78 (d, V H H = 8.0, IH, ArH), 6.88 (td, V H H = 7.2, V H H = 0.8, IH , ArH), 7.19 (td, 3 J H H = 8.0, V H H = 1.8, IH, ArH), 7.97 (dd, V H H = 7.1, V H H = 2.0, IH, ArH). 4.12m: -1.92 (d, 2 J H H = 12.0, W C H 2 ) , 1.06/1.07 (s, 9H, CMe 3 ) , 1.85/1.86 (s, 15H, C 5 Me 5 ) , 3.77 (s, 3H, OCH 3 ) , 4.15 (d, 2JHH = 12.0, IH, W C H 2 ) , 6.68 (dd, V H H = 8.4, V H H = 2.4, IH, ArH), 7.03 (d, V H H = 7.2, IH, ArH), other signals were obscured by signals of 4.120. 4.12p: -2.21 (d, V H H = 12.0, IH, W C H 2 ) , 1.06/1.07 (s, 9H, CMe 3 ) , 1.85/1.86 (s, 15H, C 5 Me 5 ) , 3.75 (s, 3H, OCH 3 ) , 4.66 (d, 2 J H H = 12.0, IH, W C H 2 ) , 6.76 (d, V H H = 8-7, 2H, ArH), 7.56 (d, 3 7 H H = 8.6, 2H, ArH). 1 3 C{ 'H} N M R (75 M H z , CDC1 3) 8 4.120: 10.5 (C 5 Me 5 ) , 33.7 (CMe 3),41.3 (CMe 3 ), 54.6 (OCH 3 ) , 108.9 (C 5 Me 5 ) , 111.1 (WCH 2 ) , 121.2 (Ar CH), 128.9 (Ar CH), 130.5 (Ar CH), 140.9 (Ar CH), 159.8 (C i p s o OCH 3 ) , 168.2 (WC i p s o ) . 118 4.2.12.3. Reaction with Toluene produced a mixture of ca. 82% m- and p-Cp*W(NO)(CH 2 CMe 3 ) (C 6 H 4 Me) (4.14) in a 1.4:1 ratio and ca. 18% Cp*W(NO)(CH 2 Ph)(C 6 H 4 Me). The latter species were identified by the doublets in the ' H N M R spectrum at 8 2.38 ( V H H = 6.1) and 3.35 ( V H H = 6.1) due to the presence of a W - C H 2 P h group, and by mass spectral analysis, which showed a molecular ion peak at mlz 531. After removal of C^D^ from the N M R sample, the residue was chromatographed on alumina 1 ( 1 x 1 cm) by eluting with hexanes until a purple band became visible, and then with 4:1 hexanes/Et20 to elute it. Next, the eluate was taken to dryness, redissolved in a minimum amount of pentane, and then stored at -30 °C for 1 d to obtain 4.14m and 4.14 p as mauve microcrystals in 56% combined yield. Anal. Calcd. for C 2 0 H 3 0 N O W : C, 48.20; H , 6.07; N , 5.62. Found: C, 48.13; H , 5.97; N , 5.47. IR (cm - 1) v N O 1549 (s). MS 511 [ M + , 1 8 4 W ] . ' H N M R (500 M H z , dioxane-d8) 8 4.14m: -2.13 (d, V H H =11.5, 1H, W C H 2 ) , 1.05 (s, 9H, CMe 3 ) , 1.82 (s, 15H, C 5 Me 5 ) , 2.27 (s, 3H, ArMe), 4.45 (d, V H H = 11.5, 1H, W C H 2 ) , 6.95 (d, V H H = 7.5, 1H, ArH), 7.09 (t, V H H = 7.5, 1H, ArH), 7.21 (d, V H H = 7.3, 1H, ArH), 7.39 (s, 1H, ArH). 4.14 p: -2.05 (d, V H a H x =11-8, IH, W C H 2 ) , 1.07 (s, 9H, CMe 3 ) , 1.83 (s, 15H, C 5 Me 5 ) , 2.22 (s, 3H, ArMe), 4.29 (d, V H a H x = II . 8, 1H, W C H 2 ) , 7.03 (d, V H H = 7.5, 2H, ArH), 7.44 (d, V H H = 7.8, 2H, ArH). 1 3 C{ 'H} N M R (125 M H z , dioxane-Jg) 8 4.14m: 10.02 (C 5 Me 5 ) , 21.4 (ArMe), 33.86 (CMe 3), 41.3 (CMe 3 ) , 111.45 (C 5 Me 5 ) , 121.0 (WCH 2 ) , 127.7 (ArCH) , 128.8 ( A r C H ) , 137.2 (Ci p s o Me), 137.3 (Ar CH), 182.9 (WCipS 0). 4.14 p: 10.00 (C5Me5), 21.5 (ArMe), 33.89 (CMe 3), 41.0 (CMe 3 ), 111.43 (C 5 Me 5 ) , 118.6 (WCH 2 ) , 128.6 (Ar CH), 133.6 (Ar CH), 137.6 (C i p s o Me) , 180.2 (WC i p s o ) . 4.2.12.4 Reaction with o-Xylene produced a mixture of Cp*W(NO)(CH 2 CMe 3 )(C 6 H 3 -3,4-Me 2 ) (4.15), Cp*W(NO)(CH 2 CMe 3 ) (CH 2 C 6 H 4 -2-Me) (4.16), and Cp*W(NO)(CH 2 C 6 H 4 -2-Me)(C 6 H 3 -3,4-Me 2 ) (4.17) in ca. 66%, 19%, and 13% yields, respectively. A trace amount of a fourth product, tentatively assigned as Cp*W(NO)(CH 2 C 6 H 4 -2-Me) 2 , was identified by a doublet in the ' H N M R spectrum at 8 3.43 ( V H H = 6.0). After removal of C 6 D 6 from the N M R sample, the residue was dissolved in a minimum amount of E t 2 0 and stored at -30 °C overnight to obtain 4.16 as dark orange crystals (20% yield). Next, the mother liquor was 119 chromatographed on alumina 1(2 x 1 cm) by developing with hexanes until a purple band became visible, and then with 3:1 hexanes/Et20 to elute it. Concentration of the eluate in vacuo, followed by cooling to -30 °C for several days afforded 4.15 as mauve needles (39% yield). Cp*W(NO)(CH2CMe3)(C6H3-3,4-Me2) (4.15): Anal. Calcd. for C23H35NOW: C, 52.58; H , 6.72; N , 2.67. Found: C, 52.79; H , 6.78; N , 2.90. IR (cm - 1) v N 0 1559. M S 525 [ M + , 1 8 4 W ] . ' H N M R (300 MHz, C 6 D 6 ) 5 -1.76 (d, V H H = 11.7, 1H, W C H 2 ) , 1.28 (s, 9H, CMe 3 ) , 1.59 (s, 15H, C 5 Me 5 ) , 1.99 (s, 3H, ArMe), 2.08 (s, 3H, ArMe), 4.26 (d, V H H = 11.7, 1H, W C H 2 ) , 7.04 (d, V H H = 7.5, 1H, ArH), 7.47 (d, V H H = 7.2, 1H, ArH), 7.82 (s, 1H, ArH). 1 3 C { ' H } N M R (75 M H z , C 6 D 6 ) 8 10.1 (C 5 Me 5 ) , 19.7 (ArMe), 20.0 (ArMe), 34.0 (CMe3), 40.7 (CMe 3 ), 110.6 (C 5 Me 5 ) , 117.4 ( V C W = 90, W C H 2 ) , 128.9 ( A r C H ) , 134.0 ( A r C H ) , 135.7 (C i p s o Me) , 136.3 (C i p s o Me), 139.4 (ArCH) , 181.0 (WC i p s o ) . Cp*W(NO)(CH 2CMe 3)(CH 2C 6H 4-2-Me) (4.16): Anal. Calcd. for C23H35NOW: C, 52.58; H , 6.72; N , 2.67. Found: C, 52.72; H , 6.54; N , 2.55. IR (cm - 1) v N O 1543. MS 525 [ M + , 1 8 4 W ] . ' H N M R (400 M H z , CDC1 3) 5 -2.75 (d, V H H = 13.5, 1H, GF/ 2 CMe 3 ) , 0.72 (s, 9H, CMe 3 ) , 1.60 (d, V H H = 13.5, 1H, Gtf 2 CMe 3 ) , 1.91 (s, 15H, C 5 Me 5 ) , 2.13 (d, V H H = 7.6, 1H, Gf/ 2 Ar), 2.26 (s, 3H, ArMe), 3.01 (d, V H H = 7.6.1H, C/f 2 Ar), 6.00 (d, V H H = 7.4, 1H, ArH), 6.62 (t, V H H = 7.4, 1H, ArH), 7.00 (d, V H H = 7.5, 1H, ArH), 7.58 (t, V H H = 7.5, 1H, ArH). 1 3 C{ 'H} N M R (75 M H z , CDC1 3) 8 10.6 (C5Me5), 20.1 (ArMe), 33.8 (CMe 3), 37.3 (CMe 3 ), 44.8 (CH 2 Ar) , 70.0 (br s, C H 2 C M e 3 ) , 108.5 (C 5 Me 5 ) , 121.2 ( C H 2 C i p s o ) , 125.4 (Ar CH), 128.0 (Ar CH), 129.6 (Ar CH), 133.4 (Ar CH), 145.5 (C i p s 0 Me). Cp*W(NO)(CH2C6H4-2-Me)(C6H3-3,4-Me2) (4.17): ' H N M R (500 M H z , C 6 D 6 ) 8 1.61 (s, 15H, C 5 Me 5 ) , 2.10 (s, 3H, ArMe), 2.19 (s, 3H, ArMe), 2.31 (s, 3H, ArMe), 3.53 (d, V H H = 6.0, 1H, W C H 2 ) , 6.08 (d, V H H = 7.3, 1H, ArH), 6.33 (d, V H H = 7.5, 1H, ArH), 6.62 (t, V H H = 7.5, 1H, ArH), 6.87 (d, V H H = 7.4, 1H, ArH). Others resonances were obscured by peak overlap. 120 4.2.12.5 Reaction with m-Xylene produced a mixture of Cp*W(NO)(CH 2 CMe 3 ) (C 6 H 3 -3 ,5-Me 2 ) (4.18) and Cp*W(NO)(CH 2C 6H4-3-Me)(C 6H 4-3,5-Me 2) (4.19) in ca. 50% and 41% yields, respectively. A third product, tentatively assigned as Cp*W(NO)(CH 2C6H4-3-Me) 2 based on the presence of a benzylic methylene doublet in the ' H N M R spectrum at 8 0.65 (2-/HH = 6.0), was detected in 9% yield. After removal of C 6 D 6 from the N M R sample, the residue was dissolved in a minimum amount of hexanes, and then stored at -30 °C overnight to obtain 4.19 as bright yellow crystals (33% yield). Next, the mother liquor was chromatographed on alumina I (1.5 x 0.7 cm) by developing with hexanes until a purple band became visible, and then with 3:1 hexanes/Et20 to elute it. Concentration of the eluate in vacuo, followed by cooling to -30 °C for several days afforded 4.18 as mauve rosettes (33% yield). Cp*W(NO)(CH 2 CMe 3 ) (C 6 H 3 -3 ,5 -Me 2 ) (4.18): Anal. Calcd. for C 2 3 H 3 5 N O W : C, 52.58; H , 6.72; N , 2.67. Found: C, 52.63; H , 6.87; N , 2.60. IR (cm - 1) v N 0 1550 (s). MS 525 [ M + , 1 8 4 W ] . ' H N M R (300 M H z , CD 2 C1 2 ) 8 -2.14 (d, 2Jm = 11.4, IH, W C H 2 ) , 1.09 (s, 9H, CMe 3 ) , 1.83 (s, 15H, C 5 Me 5 ) , 2.25 (s, 3H, ArMe), 4.55 (d, 2JHH = 11.4, IH, W C H 2 ) , 6.80 (s, IH, ArH), 7.07 (s, 2H, ArH). 1 3 C{ 'H} N M R (75 M H z , CDC1 3) 8 10.1 (C 5 Me 5 ) , 21.4 (ArMe), 33.6 (CMe 3), 41.1 (CMe 3 ), 111.2 (C 5 Me 5 ) , 122.3 (WCH 2 ) , 129.1 (Ar CH), 134.2 (2 Ar CH), 136.2 (2 QpsoMe), 181.7 (WC i p s o ) . Cp*W(NO)(CH 2C 6H4-3-Me)(C 6H4-3,5-Me2) (4.19): Anal. Calcd. for C 2 6 H 3 3 N O W : C, 55.82; H , 5.95; N , 2.50. Found: C, 54.31; H , 5.78; N , 2.30. IR (cm - 1) VNO 1546 (s). H R M S Calcd. for C 2 6 H 3 3 N O W : 559.2072 [ M + , 1 8 4 W ] . Found: 559.2080. *H N M R (300 M H z , CDC1 3) 8 1.80 (s, 15H, C 5 Me 5 ) , 1.92 (s, 3H, ArMe), 2.07 (s, 6H, 2 ArMe), 2.30 (d, 2 J H H = 6.0, IH, W C H 2 ) , 3.23 (d, 2JHH = 6.0, IH, W C H 2 ) , 6.35 (br hump, IH, ArH), 6.43 (s, IH, ArH), 6.59 (s, IH, ArH), 6.70 (s, 2H, ArH), 6.73 (t, V H H = 7.5, IH, ArH), 7.09 (br d, V H H = 6.3, IH, ArH). 1 3 C{ 'H} N M R (75 M H z , CDC1 3) 8 10.6 (C 5 Me 5 ) , 20.7 (2 ArMe), 33.6 (ArMe), 46.9 (WCH 2 ) , 108.8 (C 5 Me 5 ) , 113.2 (CH 2C i p So), 125.9 (Ar CH), 129.1 ( A r C H ) , 132.5 (Ar CH), 132.7 (Ar CH), 134.0 (Ar CH), 135.0 (C i p s o Me), 136.3 (v br s, Ar CH), 139.6 (C i p S 0 Me), 171.9 (WC i p s o ) . 121 4.2.12.6 Reaction with /^-Xylene produced a mixture of Cp*W(NO)(CH 2CMe 3)(C 6H4-2,5-Me 2 ) (2.11') and Cp*W(NO)(CH 2 C 6 H 4 -4-Me)(C 6 H 4 -2,5-Me 2 ) (4.21), and Cp*W(NO)(CH 2 C 6 H 4 -4-Me) 2 (4.22) in ca. 44%, 33%, and 13% yields, respectively, the last of which was identified by comparison of its ' H N M R data to those of an authentic sample prepared by a metathesis route.24 After removal of CeD 6 from the N M R sample, the residue was chromatographed on alumina 1 ( 1 x 1 cm) by developing with hexanes until a purple band became visible, and then with 3:1 hexanes/Et20 to elute it. Concentration of the eluate in vacuo, followed by cooling to -30 °C for several days afforded 2.11' as mauve rods (33% yield). A second orange-yellow band was eluted with 2:1 hexanes/Et20. This fraction was then stored at -30 °C overnight to induce the deposition of 4.21 as yellow-orange blocks (23% yield). Cp*W(NO)(CH 2CMe 3)(C 6H 4-2,5-Me 2) (2.11'): Anal. Calcd. for C 2 3 H 3 5NOW: C, 52.58; H , 6.72; N , 2.67. Found: C, 52.43; H , 6.59; N , 2.52. IR (cm - 1) VNO 1538 (s). MS 525 [ M + , 1 8 4 W ] . ' H N M R (500 M H z , to\-ds, 213K) 5 major rotamer: -0.98 (d, V H H = 11.5, 1H, W C H 2 ) , 1.34 (s, 9H, CMe 3 ) , 1.47 (s, 15H, C 5 Me 5 ) , 2.20 (s, 3H, ArMe), 2.89 (s, 3H, ArMe), 4.08 (d, 2 J H H = 11.5, 1H, W C H 2 ) , 6.54 (s, 1H, ArH), 6.85 (d, V H H = 7.6, 1H, ArH), 7.17 (d, V H H = 7.9, 1H, ArH). minor rotamer: -3.28 (d, V H H = 10.4, 1H, W C H 2 ) , 1.33 (s, 9H, CMe 3 ) , 1.48 (s, 15H, C 5 Me 5 ) , 1.73 (s, 3H, ArMe), 2.31 (s, 3H, ArMe), 5.02 (d, V H H = 10.4, 1H, W C H 2 ) , 8.34 (s, 1H, ArH), 7.04 (d, V H H = 7.4, 1H, ArH), 7.21 (d, V H H = 7.7, 1H, ArH). 1 3 C { ' H } N M R (125 M H z , CD 2 C1 2 ) 8 major rotamer: 9.8 (C 5 Me 5 ) , 21.0 (ArMe), 25.5 (ArMe), 33.1 (CMe 3), 40.0 (CMe 3), 110.8 (C 5 Me 5 ) , 111.2 (WCH 2 ) , 127.7 ( A r C H ) , 128.1 (Ar CH), 129.9 (Ar CH), 130.9 (C i p s 0 Me), 146.0 (C i p s o Me), 185.6 (WC i p s 0 ) . minor rotamer: 10.0 (C 5 Me 5 ) , 20.6 (ArMe), 27.7 (ArMe), 32.7 (CAfe3), 42.2 (CMe 3 ), 109.5 (C 5 Me 5 ) , 127.8 (Ar CH), 129.5 ( A r C H ) , 133.2 (C i p s o Me), 134.7 (WCH 2 ) , 136.3 (C i p s o Me) , 140.5 ( A r C H ) , 180.8 ( W C i p S o ) . Cp*W(NO)(CH 2C 6H 4-4-Me)(C 6H 4-2,5-Me 2) (4.21): Anal. Calcd. for C 2 6 H 3 3 N O W : C, 55.82; H , 5.95; N , 2.50. Found: C, 55.54; H , 6.06; N , 2.68. IR (cm - 1) VNO 1548 (s). MS 559 [ M + , 1 8 4 W ] . ' H N M R (400 M H z , C 6 D 6 ) 8 1.57 (s, 15H, C 5 Me 5 ) , 1.58 (s, 3H, ArMe), 1.99 122 (s, 3H, ArMe), 2.38 (d, 2 J H H = 6.1, IH, W C H 2 ) , 2.93 (s, 3H, ArMe), 3.35 (d, 2JHH = 6.1, IH, W C H 2 ) , 5.33 (br s, IH , ArH), 6.35 (d, V H H = 7.8, 2H, ArH), 6.66 (d, 3 J H H = 7.5, IH, ArH), 6.94 (d, V H H = 7.8, 2H, ArH), 7.04 (d, V H H = 7.5, IH, ArH). 1 3 C { ' H } N M R (75 M H z , CDC1 3) 5 10.5 (C 5 Me 5 ) , 20.5 (ArMe), 21.4 (ArMe), 28.2 (ArMe), 49.1 (br s, 2 W C H 2 ) , 108.9 (C 5 Me 5 ) , 111.7 ( W C H 2 C i p s o ) , 124.6 (Ar CH), 127.5 (Ar CH), 129.8 (2 Ar CH), 130.1 (C i p s o Me) , 135.2 (2 Ar CH), 142.5 (C i p s o Me), 146.9 (C i p S 0 Me), 180.8 (WC i p s o ) . 4.2.12.7 Reaction with Mesitylene produced Cp*W(NO)(CH 2C 6H 3-3,5-Me2)2 (4.23) as the only detectable organometallic product in addition to decomposition, which gave rise to broad featureless peaks in the Cp* region. After removal of CeD 6 from the N M R sample, the oily brown-yellow residue was redissolved in 4:1 hexanes/Et20 and filtered through a plug of alumina 1 ( 1 x 1 cm). The filtrate was reduced in volume and then stored at -30 °C to induce the deposition of 4.23 as long orange-yellow needles (43% yield). Anal. Calcd. for C 2 4 H 3 7 N O W : C, 57.25; H , 6.35; N , 2.38. Found: C, 57.14; H , 6.42; N , 2.42. IR (cm - 1) v N 0 1547 (s). MS 587 [ M + , 1 8 4 W ] . ' H N M R (300 M H z , CDC1 3) 5 0.81 (d, 2 JHH = 8.7, 2H, W C H 2 ) , 2.11 (s, 15H, C 5 Me 5 ) , 2.46 (d, 2JHH = 8.7, 2H, W C H 2 ) , 2.50 (s, 12H, ArMe), 6.68 (s, 4H, ArH), 7.24 (s, 2H, ArH). 1 3 C{ 'H} N M R (75 M H z , CDC1 3) 5 10.3 (C5Me5), 21.2 (4 ArMe), 41.3 ( ' j Cw = 66, 2 W C H 2 ) , 107.7 (C 5 Me 5 ) , 128.7 (2 A r CH), 129.1 (4 Ar CH), 132.9 (4 C i p s o M e ) , 137.7 (br, 2 C H 2 C i p s o ) . 4.2.13 Thermolysis of CpW(NO)(CH 2CMe 3) 2 in Neat Me 4Si, C6D6, and />-Xylene Each of these reactions was carried out in a manner similar to the reaction of 2.4 with tetramethylsilane except the reaction was carried out at 60 °C for ca. 5-15 h. In all cases the ' H N M R spectrum of the crude reaction mixture revealed primarily decomposition. 123 4.2.14 Thermolysis of CpW(NO)(CH 2CMe 3) 2 in Cyclohexane in the Presence of PMe 3 In a manner analogous to the Cp* reaction described above, a 0.01 M solution of 2.1 in cyclohexane in the presence of excess PMe 3 was heated at 60 °C overnight, during which time it changed color from wine-red to yellow. The organic volatiles were then removed in vacuo, and an N M R sample in CeD 6 was prepared. The ' H N M R spectrum of the crude reaction mixture revealed the clean conversion of 2.1 into an ca. 3:2 mixture of CpW(NO)(=CHCMe 3)(PMe 3) (3.1) and CpW(NO)(cyclohexene)(PMe3), respectively. The identity of 3.1 was confirmed by comparison of its *H N M R data to those of an authentic sample, while that of CpW(NO)(cyclohexene)(PMe3) was established on the basis of mass spectral and ' H and 1 3 C N M R analyses of the crude reaction mixture. CpW(NO)(cyclohexene)(PMe3): MS 437 [ M + , 1 8 4 W ] . ' H (400 M H z , C 6 D 6 ) 8 1.13 (d, 2 JHP = 8.8, 9H, PMe 3 ) , 1.6-1.8 (m, 5H), 1.9 (m, 1H), 2.25 (m, 1H), 2.43 (m, 1H), 2.95 (m, 1H), 3.10 (m, 1H), 4.87 (s, 5H, C 5 H 5 ) . 1 3 C{ 'H} (100 M H z , C 6 D 6 ) 8 18.2 (d, lJCP = 30.8, PMe 3 ) , 24.6 (CH 2 ) , 24.8 (CH 2 ) , 30.6 (d, 2JCP = 3.9, =CH), 30.7 (CH 2 ) , 32.6 (CH 2 ) , 33.9 (d, 2JCP = 10.6, =CH), 95.4 (C 5 H 5 ) . 4.3 Results and Discussion 4.3.1 Thermal Reactions of Cp*W(NO)(CH 2CMe 3) 2 (2.4) with Tetramethylsilane and Alkanes As noted in the Introduction, thermolysis of the bis(neopentyl) complex Cp*W(NO)(CH 2 Me 3 ) 2 (2.4) leads to the formation of [Cp*W(NO)(=CHCMe 3)J as a highly reactive intermediate. In the following section, it is shown that [Cp*W(NO)(=CHCMe 3)] reacts rapidly with the C - H bonds of tetramethylsilane and alkanes and that the 124 thermodynamic product of C - H activation often involves subsequent rearrangement of the initially formed complex. As shown in Scheme 4.1, thermolysis of 2.4 in tetramethylsilane at 70 °C for ca. 2 d leads to formation of the known bis(alkyl) complex Cp*W(NO)(CH 2CMe3)(CH 2SiMe3) (2.9) in 90% isolated yield. Thermolysis in 1,1,2,2-tetramethylcyclopropane, in contrast, gives a complex mixture containing Cp*W(NO)[CH 2 C(Me)C(Me) 2 CH] (4.1) as the major and only identified product. The formation of 4.1 most likely proceeds via the intermediacy of Cp*W(NO)(CH2CMe3)(2,2,3,3-tetramethylcyclopropyl), which then rapidly rearranges to 4.1 by y-hydrogen abstraction. The driving force presumably is derived from the formation of a Scheme 4.1 M e 3 S i O C M e 3 2.9 Me 4 Si 4.1 2.4 125 strong W - c P r bond in the first step and the release of severe steric congestion between the cyclopropyl and neopentyl groups in the second. The alternative possibility involving initial formation of Cp*W(NO)(CH2CMe3)[(l,2,2-trimethylcyclopropyl)methyl] (not shown) appears highly unlikely, since it would involve activation a strong cyclopropyl C - H bond to form a highly strained product, and since y-hydrogen abstraction has so far not been observed with other Cp*W(NO)(R)(R') complexes in which R and R ' are primary alkyl ligands (and neopentyl-like). Complex 4.1 was isolated as bright yellow crystals in 32% yield. It is stable to air and moisture and can be stored indefinitely at 25 °C under nitrogen. In the 1 3 C { ' H } N M R spectrum (C 6 D 6 ) , the W - C H 2 and W - C H carbons resonate at 5 55.8 ( ' j C w = 37 Hz) and 83.0, respectively. In the ' H N M R spectrum, the corresponding methine proton appears as a broad singlet at 8 1.16, while the diastereotopic methylene protons appear as a doublet and a doublet of doublets at 5 3.25 ( 2 J H H = 4.3 Hz) and 5 0.72 ( 2 7 H H = 4.3 Hz, 4 J H H = 1.2 Hz). That the cyclopropyl group points away from the Cp* ligand has been confirmed by an N O E difference experiment, which shows a strong enhancement of the methine resonance upon irradiation of the Cp* methyl protons, and vice versa. In contrast to reactions with tetramethylsilane and 1,1,2,2-tetramethylcyclopropane, only decomposition is observed upon thermolysis of 2.4 in neat cyclopentane, cyclohexane, or neohexane. More tractable chemistry, however, is observed in the presence of a trapping agent such as PMe3. As shown in Scheme 4.2, thermolysis of 2.4 in cyclopentane, cyclohexane, or neohexane in the presence of excess PMe 3 at 70 °C for ca. 2 d cleanly produces the 7]2-alkene complexes 4.2-4.4, respectively, and Cp*W(NO)(=CHCMe 3)(PMe 3) (3.4), in ratios that depend on the amount of PMe3 used. From the crude reaction mixtures, it can be observed by ' H and 3 1 P{ 1 H} N M R spectroscopies that 4.2 and 4.3 exist as a single isomer, whereas 4.4 exists as a 5:1 mixture of syn and anti rotamers,23 respectively, which have proven to be inseparable by physical means. Complex 3.4 was identified by comparison of its N M R spectral data to those of an authentic sample prepared by thermolysis of a THF or 2,2,4,4-tetramethylpentane solution of 2.4 in the presence of excess PMe3 (see Chapter 3). Complexes 126 4.2-4.4, on the other hand, were isolated as air-stable yellow crystalline materials by chromatography on alumina I, followed by crystallization from hexanes/Et20 at -30 °C. They were characterized by standard spectroscopic and analytical methods, aided by H M Q C , H M B C , and/or N O E difference experiments, and in the case of 4.3 and 4.4 s y n also by X-ray diffraction. I Scheme 4.2 O N \ / ^;PMe3 + M e 3 P P M e 3 N C M e 3 A W 4.2, n = l I i I 4 -3 ,n = 2 O 3.4 N M e 3 C o C M e 3 2.4 PMe 3 V O N ^ P M e 3 = ^ O N ^ P M e 3 + 3.4 4'4 Sy n 4.4an(i Spectroscopically, complexes 4.2-4.4 are very similar to the rhenium alkene complexes, [CpRe(NO)(CH 2=CHR)(PPh 3)] +X", 2 4 synthesized by Gladysz and coworkers. In the ' H N M R spectra (C^D^), the olefinic proton resonances appear at about 5 -0.3 to 1.0 and 8 1.4 to 2.3. In all cases, the upfield signals correspond to the protons lying syn to the Cp* ligand and cis to NO. In the 1 3 C{ 'H} N M R spectra (C 6 D 6 ) , the corresponding carbon resonances appear as a doublet and a singlet at 8 40.0-46.1 ( V C P = 12-13 Hz) and 8 36.9-44.2 ('•/cw - 39-40 Hz), respectively. These upfield shifts are indicative of substantial 2 5 2 6 metallacyclopropane character and can be ascribed to strong Tt-backbonding by tungsten. ' Key to the stereochemical assignment of 4.4 s y n in solution is the observation of a strong N O E 127 enhancement of the t e r t - b u t y l resonance upon irradiation of the Cp* and PMe3 signals. In the case of 4 .4 a n t i , irradiation of the PMe3 signal leads to an N O E enhancement of the =CH 2 signal at 8 -0.30 and irradiation of either the Cp* or =CH 2 signal at 8 -0.30 leads to an enhancement of the signal for the =C/ /CMe 3 proton at 8 1.43.27 Shown in Figure 4.1 are the ORTEP diagrams of the solid-state molecular structures of complexes 4.3 and 4.4 s y n along with selected bond distances and angles. As in the case of the [CpRe(NO)(CH 2=CHR)(PPh 3)] +X" complexes, the nitrosyl ligand in each of 4.3 and 4.4 s y n is essentially linear, while the alkene ligand is oriented such that its C=C bond axis is perpendicular to the W - N - 0 vector, as would be expected for maximum 7i-bonding.28 The alkene ligand is coordinated in a manner such that the two W - C a distances are unequal. In each complex, the bond cis to N O is shorter than that cis to PMe3. This asymmetry appears to be due to steric crowding by the alkene substituents, since the opposite trend is observed for the [CpRe(NO)(CH 2=CHR)(PPh 3)] +X~ complexes, where the =CHR group is cis to NO, and since in 4.3 and 4.4 s y n the W - C a distance also increases as the size of the substituent on carbon increases. Finally, as expected from the N M R experiments, there is a substantial lengthening of the carbon double bonds in 4.3 and 4.4 s y n . In both compounds, the C(14)-C(15) bond distances are essentially identical at 1.447 A . 2 9 128 B C(18) C(12) C(19) W ( l ) - N ( l ) 2.444(2) A N( l ) -0 (1 ) 1.768(6) A W ( l ) -P ( l ) 1.251(8) A W(l) -C(14) 2.246(9) A W(l) -C(15) 2.193(9) A C(14)-C(15) 1.447(12) A N ( l ) - W ( l ) - C ( 1 4 ) 105.8(3)° N ( l ) - W ( l ) - C ( 1 5 ) 96.3(3)° N ( l ) - W ( l ) - P ( l ) 87.1(2)° W ( l ) - N ( l ) - 0 ( 1 ) 172.6(6)° W ( l ) - N ( l ) 2.441(2) A N( l ) -0 (1 ) 1.770(5) A W ( l ) - P ( l ) 1.227(6) A W ( l ) -C(14) 2.173(6) A W ( l ) -C(15) 2.298(6) A C(14)-C(15) 1.446(8) A N ( l ) -W ( l ) -C(14 ) 93.0(3)° N ( l ) - W ( l ) - C ( 1 5 ) 94.6(2)° N ( l ) - W ( l ) - P ( l ) 82.5(2)° W ( l ) - N ( l ) - 0 ( 1 ) 170.7(5)° Figure 4.1. ORTEP diagrams and selected bond distances and angles for complexes (A) Cp*W(NO)(cyclohexene)(PMe3) (4.3) and (B) Cp*W(NO)(neohexene)(PMe3) (4.4 s y n). 129 In order to gain insight into the reactions outlined in Scheme 4.2, the thermolysis of 2.4 in cyclohexane in the presence of various of amounts of PMe 3 was monitored b y 3 1 P { ' H } N M R spectroscopy. This showed that the ratio of the neopentylidene complex 3.4 to the cyclohexene complex 4.3 is independent of the reaction time and that this ratio increases as the initial concentration of PMe3 increases. These results thus indicate that 3.4 and 4.3 are formed as a result of a competition between PMe 3 and cyclohexane for reaction with [Cp*W(NO)(=CHCMe 3)] and that these complexes do not interconvert. The thermolysis of 2.4 in neohexane in the presence of excess PMe 3 was also monitored by 3 1 P{ 'H} N M R spectroscopy. In this case, parallel formation of 3.4 and 4.4syn was observed at early conversion. As the reaction progressed, 4.4a n ti was generated at the expense of 4.4Syn. At 70 °C, the isomerization of 4.4Syn to 4.4anti (^ obs ~7.41 x 10"6 s"')is independent of P M e 3 concentration and occurs at a rate that is approximately one order of magnitude lower than that for formation of 4.4syn (& o b s ~ 4.22 x 10"5 s"1). In addition, it may be effected in neat C^D^ or in CeD^ in the presence of a large excess of PMe 3 without solvent activation or formation of the known bis(phosphine) complex Cp*W(NO)(PMe3)2.2 These results therefore suggest that 4.4syn is the kinetic product of the double C - H activation of neohexane and that, in contrast to the [CpRe(NO)(CH 2=CHR)(PPh 3)] +X" complexes, which upon thermolysis undergo exchange of alkene enantioface by way of a o - C - H complex, 3 0 the isomerization of 4.4syn to 4.4an ti occurs by simple rotation and without phosphine or alkene ligand dissociation. To determine whether the formation of 4.4 occurs with concomitant reversible tert-butyl C - H activation, the bis(alkyl) complex Cp*W(NO)(CH 2 CMe 3 ) (CH 2 CMe 2 Et) (2.7) was prepared and its thermal decomposition investigated. If 2.7, the expected product of tert-butyl C - H activation, was formed reversibly, then heating a solution of 2.7 and P M e 3 should cleanly produce complexes 3.4 and 4.4 when the solvent is neohexane, and complex 3.4 when the solvent is THF. This was not the case, however, when the thermolysis of 2.7 in the presence of excess P M e 3 was performed. As shown in eq 4.1, the thermolysis of 2.7 in neohexane is complex, producing not only 4.4 and 3.4 as major products but also 130 (4.1) THF *• 3.4 + 4.5 + 4.6 + others (4.2) P M e 3 75 °C Cp*W(NO)(=CH 2CMe 2Et)(PMe 3) (4.5) and Cp*W(NO)(PMe 3 ) 2 (4.6), along with several unidentified minor products which display 3 'P N M R signals in the 8 33-36 region. The thermolysis of 2.7 in THF is similarly complex, affording an approximately 1:1:1 mixture of 3.4, 4.5, and 4.6 based on relative 3 1 P N M R peak heights, as well as several minor unidentified products which display 3 1 P N M R signals in the 8 10-18 region (eq 4.2). In this case, 4.4 could not be observed by 3 l P { ' H } N M R spectroscopy, most likely because the concentration of free neohexane was too low to compete with PMe 3 for reaction with [Cp*W(NO)(=CHMe 3)]. Although complexes 4.4, 3.4, 4.5 and 4.6 could not be isolated pure, their identities could be established by ' H N M R , 1 3 C N M R , and/or 3 1 P N M R analyses of semi-purified product mixtures. Spectroscopically, 4.5 is very similar to 3.4; it was thus identified by the characteristic signals of its alkylidene a-proton and PMe 3 phosphorus at 8 11.27 ( V H P = 3.3 Hz) and 8 -7.6 ('./pw = 445 Hz), respectively. In the case of 4.6, its identity was confirmed by 131 Scheme 4.3 2.4 comparison of its ' H , i 3 C , and 3 1 P N M R spectral data to those of an authentic sample3 5 prepared by reducing Cp*W(NO)Cl2 with sodium amalgam in the presence of excess PMe3. Together these results strongly suggest that in the activation of neohexane by [Cp*W(NO)(=CHCMe 3)] attack at the tert-butyl position occurs rarely, i f at all. The activation of neohexane is therefore regiospecific and involves initial attack at the CH2C//3 position. The observation that 4.2, 4.3, and 4.4 s y n are formed stereospecifically can be rationalized in terms of a mechanism in which the initially formed complex eliminates neopentane by a direct (3-hydrogen abstraction reaction. As illustrated in Scheme 4.3, a 132 bicyclic transition state with little conformational freedom is involved. In this transition state, in order to minimize unfavorable steric interactions with the neopentyl tert-butyl group, rotation about the neohexyl C a - C p bond occurs so as to place the tert-butyl syn with respect to the Cp* ligand. Subsequent (3-hydrogen abstraction leading to 4.4 s y n therefore occurs stereospecifically for steric reasons. In the formation of 4.2 and 4.3, on the other hand, such a rotation is not possible due to the geometric constraints imposed by the cycloalkyl rings. Consequently, (3-hydrogen abstraction leads to formation of alkene complexes in which the =CHR hydrogen points toward the Cp* ligand. The trapping of the above double C-H-activated intermediates is not limited to phosphine ligands. As shown in eq 4.3, thermolysis of 2.4 in cyclohexane in the presence of excess Me2NH also occurs cleanly, yielding Cp*W(NO)(cyclohexyl)(NMe2) (4.7) and Cp*W(NO)(CH 2 CMe 3 ) (NMe 2 ) (2.17) (see Chapter 3), in ratios that depend upon the amount of amine added. As noted in Chapter 3, complex 2.17 can also be generated upon thermolysis of 2.4 or Cp*W(NO)(=CHCMe 3)(PMe 3) (3.4) in THF in the presence of excess M e 2 N H . Since the conversion of 2.4 to 2.17 involves initial formation of [Cp*W(NO)(=CHCMe 3)] as the reactive intermediate rather than o-bond metathesis, the formation of 4.7 therefore is in the same manner as that proposed above for reaction in the presence of PMe 3 . A n exception is that amine coordination is followed by conversion of the cyclohexene moiety into a cyclohexyl group by oc-hydrogen migration. (4.3) 70 °C 2.4 4.7 2.17 Complex 4.7 was isolated as a yellow crystalline solid by chromatography on alumina I, followed by crystallization from hexanes/Et20. It was then characterized by ' H and 1 3 C N M R spectroscopies and by an X-ray crystallographic study. Spectroscopically, 4.7 is very 133 similar to 2.17. In addition to peaks due to the Cp* and N M e 2 ligands, the 1 3 C { 'H} N M R spectrum of 4.7 in CeD6 contains signals attributable to the cyclohexyl methine carbon at 5 52.1 ('Jew = 94 Hz) and methylene carbons at 8 27.8, 32.9, 33.0, 37.2, and 40.0. In the ' H N M R spectrum, the cyclohexyl methine proton resonates as a multiplet at 8 1.17, while the amido methyl groups resonate as two sharp singlets at 8 2.63 and 3.72 even at 70 °C. This suggests that the amido ligand is functioning as a strong 7t-donor toward the tungsten center. Shown in Figure 4.2 is an ORTEP diagram of the solid-state molecular structure of 4.7 along with selected bond distances and angles. Like the structure of 2.17,the amido ligand of 4.7 adopts a planar geometry with bonding parameters consistent with it acting as a K-donor. For example, the W - N M e 2 distance of 1.924(8) A is indicative of considerable multiple bond character and the N(l)-W-N(2)-C(7) and N(l)-W-N(2)-C(6) torsion angles of-8(4)° and 177.1(6)°, respectively, are of proper magnitude for maximum 7>bonding. W - N ( l ) 1.754(7) A W-N(2) 1.924(8) A N(l) -0(1) 1.232(9) A W-C(21) 2.148(9) A N(2)-C(6) 1.477(11) A N(2)-C(7) 1.453(14) A W - N ( l ) - 0 ( 1 ) 1 6 9 . 2 ( 1 0 ) ° N ( l ) - W - N ( 2 ) 9 7 . 3 ( 4 ) ° W-N(2)-C(6) 1 2 3 . 5 ( 7 ) ° W-N(2)-C(7) 1 2 8 . 8 ( 6 ) ° Figure 4.2. ORTEP diagram along with selected bond distances and angles of complex Cp*W(NO)(cyclohexyl)(NMe 2) (4.7). Finally, in contrast to the formation of alkene complexes from cyclopentane, cyclohexane, and neohexane, the thermolysis of 2.4 in methylcyclopentane, 134 methylcyclohexane, ethylcyclohexane, or pentane (70 °C, 2 d) results in the formation of the rj 3-allyl hydride complexes 4.8—4.11 in low to moderate yields in both the presence and absence of PMe3 (Scheme 4.5). Reactions in the absence of PMe3 proceed with concomitant decomposition, while reactions in the presence of PMe3 proceed with some decomposition (especially in the case of pentane) as well as give 3.4 and PMe3-trapped alkene complexes as minor products. Scheme 4.5 (* = minor isomer detectable by ' H and 1 3 C N M R spectroscopies) 4.11endo 135 Complexes 4.8-4.11 were isolated as yellow crystalline materials by crystallization from hexanes or pentane/Et20. They are thermally quite stable; for example, 4.8 and 4.10 can be heated in toluene-Jg for several hours at 70-90 °C without detectable decomposition or reaction with solvent. The formulation of 4.8-4.11 is supported by N M R , IR, and mass spectroscopies. The highest peak in the electron-impact mass spectra corresponds to the molecular ion minus H 2 . The W - H stretch in the Nujol-mull IR spectra appears as a weak absorbance at 1900-2200 cm"1. In the ' H N M R spectra, the hydride resonance appears as a singlet at 8 -0.7 to -1.4 with ' j H w = 119-123 Hz. The ' H and 1 3 C N M R spectra of 4.9 and 4.11 indicate the presence of one isomer. In the ' H and 1 3 C N M R spectra of 4.8 and 4.10, two isomers were observed in ratios of >85:15. The major and minor isomers of 4.8 were assigned endo structures in which the hydride ligand is cis and trans, respectively, to the substituted end of the allyl ligand based on ' H N O E difference measurements (see the Experimental Section), ' H and 1 3 C N M R chemical shifts for the allyl fragment, and comparison of these N M R data to those of 4.9, which has been characterized by X-ray diffraction (see below). The isomers of 4.10, in contrast, differ from one another in the orientation of the allyl fragment relative to the Cp* ligand. The assignment of the major isomer as an exo complex in which the hydride ligand is cis to the substituted end of the allyl was based on results of an N O E difference experiment and an X-ray diffraction study. In the case of the minor isomer of 4.10, its assignment as an endo complex in which the hydride ligand is cis to the substituted end of the allyl was based on the fact that its allyl proton and carbon resonances appear at chemical shifts that are very similar to those of 4.8endo and 4.9endo, and quite different from those of 4.10eXo- As shown in Table 4.1, the chemical shifts for the allyl protons and carbons are particularly diagnostic of the orientation of the allyl ligand with respect to the Cp* ligand. In general the chemical shift difference for the (X-CH2 resonances of the endo isomer (A8 ~ 2.3-3.3 ppm) is much larger than that of the exo isomer (AS ~ 0.3 ppm). In addition, the resonance for the central proton of the endo isomer appears at much lower field than that of the exo isomer. 1 3 6 Table 4.1. Selected ' H and 1 3 C N M R Data for the rf -Al ly l Moieties of Complexes 4.8-4.11" compd 'H NMR (8) l 3 C NMR (8) a-CH 2 (anti) a-CH 2 (syn) a-CH (anti) (3-CH oc-CH2 a-CH p-CH A Q b ^ • ° e n d u 0 .68 3 .06 2 . 4 4 3 8 . 3 8 0 . 3 4 . 8 e n d o 0 . 9 6 4 .21 1.7 4 6 . 2 6 7 . 7 4 9 / 0.31 2 . 6 0 2 . 2 8 4 1 . 3 8 0 . 0 4.11 e n t i 0 0 . 1 6 2 . 7 9 1.82 4 . 5 9 3 8 . 9 8 4 . 4 101 .4 4.10 c n d o c 0 . 6 0 2 . 8 7 4 . 6 9 4 0 . 9 100 .9 4.10exo* 2 . 4 4 2 .15 2 . 6 6 3 7 . 2 1 0 9 . 0 "In CeDg. 4 Major isomer. c Minor isomer. dOn\y one isomer was detected in the N M R spectra of purified samples. As mentioned above, the solid-state molecular structures of 4 . 9 e n d o and 4.10 e x o have been determined by X-ray diffraction. Shown in Figure 4 . 3 are the ORTEP representations of these structures along with selected bond distances and angles. Structurally, 4 . 9 e n d o and 4.10 e x o are very similar to the o,7]3(endo) complex, Cp*W(NO)[CH 2C(Me)CHCH 2C(Me)2CH(Me)], reported in Chapter 3 . Noteworthy are the following features. First, in contrast to the allyl ligands in nitrosyl complexes such as CpM(NO)(r/3-allyl)(halide) (M = M o , 3 1 W 3 2 ) , CpW(NO)(773-allyl)(V-alkynyl),33 and TpMo(NO)(773-allyl)(CO),34 which exhibit if^atf distortion, the allyl C - C distances in 4 . 9 e n d o and 4.10 e x o ( 1 . 3 7 - 1 . 3 9 A) are all within 0 . 0 2 A of one another and in the range intermediate of double and single bonds. Second, in each complex the substituted allyl terminus is cis to the hydride ligand, which was located but not refined, and is farther from the metal center than the unsubstituted terminus, presumably in order to minimize steric repulsion between the allyl substituents and the [Cp*W(NO)(H)] fragment. The observation that 4.8-4.11 are formed in low to moderate yields in both the presence and absence of excess PMe 3 and that, in all but the case of 4.8ei,do*, the hydride and the substituted end of the allyl ligand are adjacent to one another can be explained in terms of 137 B C(7) W(l)-C(7) 2.234(11) A W ( l ) - C ( l ) 2.369(9) A W(l ) -C(2) 2.397(11) A C(l ) -C(2) 1.366(14) A C(l ) -C(7) 1.39(2) A W ( l ) - N ( l ) 1.787(9) A N( l ) -0 (1) 1.216(11) A C(2)-C(l)-C(7.) 1 1 8 . 7 ( 1 0 ) ° C(2)-C(l ) -C(6) 1 2 0 . 5 ( 9 ) ° C( l ) -C(2)-C(3) 1 2 3 . 8 ( 1 0 ) ° C(7)-C(l ) -C(6) 1 1 9 . 7 ( 1 0 ) ° C(2)-C(3)-C(4) 1 1 1 . 5 ( 1 0 ) ° 0(1)-N(1)-W(1) 172 .8 (9 ) ° W ( l ) - C ( l l ) 2.267(8) A W(l)-C(12) 2.275(7) A W(l)-C(13) 2.476(7) A C( l l ) -C(12 ) 1.390(11) A C(12)-C(13) 1.380(10) A W ( l ) - N ( l ) 1.768(6) A N( l ) -0(1) 1.227(7) A C(l l ) -C(12)-C(13) 125 .7 (7 ) ° C(12)-C(13)-C(18) 122 .7 (7 ) ° 0(1)-N(1)-W(1) 1 7 2 . 8 ( 9 ) ° F i g u r e 4.3. ORTEP drawings and selected bond distances and angles of (A) Cp*W(NO)(77 C 7 H,,)(H) (4.9 e n d 0) and (B) Cp*W(NO)(r73-C8H13)(H)(4.10exo). 138 the reaction sequence outlined in Scheme 4.6. According to this proposed reaction sequence (using pentane as a representative example), both methyl and methylene C - H bonds are attacked by [Cp*W(NO)(=CHCMe 3)]. However, based on the relative yields of 4.8-4.11 and on the assumption that these complexes are primarily formed as a result of initial methyl C - H activation, [Cp*W(NO)(=CHCMe 3)] presumably reacts with the 2° C - H bonds of methylcyclopentane, methylcyclohexane, and ethylcyclohexane reluctantly, whereas with pentane it favors the 1° C - H bonds over the 2° C - H bonds weakly, i f at all. This leads to Scheme 4.6 139 formation of bis(alkyl) complexes A and A ' , which rearrange to the alkene intermediates B , B ' , and B " in a manner analogous that described above for the activation of neohexane, cyclopentane, and cyclohexane. Allylic C - H activation then ensues depending on the structure and orientation of the alkene ligand. For intermediate B , allylic C - H activation is apparently fast and irreversible, yielding the observed product 4 .11 e n d 0 even in the presence of PMe3. In the case of intermediate B " , the orientation of its alkene ligand does not permit such a transformation; adduct formation therefore occurs when the reaction is carried out in the presence of PMe 3 . Intermediate B ' , on the other hand, is structurally similar to the alkene intermediates in the cyclopentane and cyclohexane reactions with one exception. Subsequent allylic C - H activation in B ' would lead to formation of a 1,3-disubstituted allyl ligand in an endo and not an exo conformation. As noted above, thermolysis of 2.4 in cyclopentane or cyclohexane in the presence of PMe3 leads to the clean formation of alkene complexes 4.3 and 4.4, respectively, whereas use of pentane as solvent results in concomitant decomposition. A possible explanation for these observations therefore is that allylic C - H activation in [Cp*W(NO)(alkene)] complexes is stereospecific,35 occurring only when it leads to formation of an endo allyl ligand, and that 1,3-disubstituted allyl ligands are unstable, decomposing at the temperature at which they form. 4.3.2 Thermal Reactions of 2.4 with Arenes In addition to its reactions with alkanes, the unsaturated neopentylidene complex [Cp*W(NO)(=CHCMe 3)] is also highly reactive toward aromatic solvents. As shown in Scheme 4.6, thermolysis of 2.4 in neat a,a,a-trifluorotoluene or anisole at 70-75 °C for -1.5 d leads to formation of products resulting from aryl C - H bond addition across the W=C linkage. In a,a,a-trifluorotoluene, the products are o-, m-, and/>Cp*W(NO)(CH2CMe3)(C6H5CF3) (4.12), formed in a ratio of ~1:66:33 by integration of the methylene regions of the ' H N M R spectrum (CeD 6). Similarly, the reaction with anisole yields the corresponding o-, m-, andp-Cp*W(NO)(CH 2CMe3)(C6H 5OCH3) complexes (4.13) in a ratio of-88:7:5. 140 Scheme 4.6 4.12 o:m:p = 88 : 7 : 5 2.4 4.13 o:m:p = 1 : 66 : 33 The thermolysis of 2.4 in methylated benzenes at 70-75 °C yields both aryl and benzyl C - H activation products as shown in Scheme 4.7. The thermolysis in toluene produces a mixture of ca. 82% m- and / ? -Cp*W(NOXCH 2 CMe 3 XC 6 H 4 Me) (4.14) in a ratio of 1.4:1 and ca. 18% Cp*W(NO)(CH 2 Ph)(C 6 H 4 Me). The formation of Cp*W(NO)(CH 2 Ph)(C 6 H 4 Me) occurs with Cp*W(NO)(CH 2CMe3)(CH 2Ph) as an observable intermediate. As noted in the Introduction, Cp*W(NO)(CH 2CMe3)(CH 2Ph) is unstable with respect to cc-abstraction of neopentane. This leads to formation of the benzylidene complex, [Cp*W(NO)(=CHPh)], which apparently is also capable of cleaving C - H bonds intermolecularly. Because of the low yields of Cp*W(NO)(CH 2Ph)(CeH 4Me), isolation was not attempted; assignment of their structures therefore is only tentative and rests primarily on the fact that the W - C / / 2 P h resonances exhibit chemical shifts and coupling constants very similar to those of other benzyl-containing complexes such as 4.17, 4.19, 4.21, and 4.22 (see below). 3 6 Complexes 4.14, on the other hand, were purified as a mixture by crystallization and characterized fully by spectroscopic and analytical methods (see below). Scheme 4.7 4.14m 4.14p (-48 : 34 : 18) 4.15 4.16 4.17 + isomer (-66 : 19 : 13 : 1) 4.18 4.19 4.20 (-50 : 41 : 9) 2.11' 4.21 4.22 (-44 : 43 : 13) + decomposition 4.23 142 Heating 2.4 in o-xylene results in the formation of complexes 4.15, 4.16, and 4.17 in ca. 66%, 19%), and 13% yields, respectively, and a trace amount of a product tentatively assigned as the bis(benzyl), Cp*W(NO)(CH 2 C 6 H 4 -2-Me)2 (by ] H N M R ) . Complete conversion of 4.16 to 4.17 and Cp*W(NO)(CH 2C 6H4-2-Me)2 was not possible, as decomposition of 4.16 to unknown products became competitive at longer reaction times. The reaction with m-xylene gives complexes 4.18, 4.19, and 4.20 in respective yields of ca. 50%>, 41%, and 9%, according to *H N M R analysis, while the reaction withp-xylene produces 2.11', 4.21, and 4.22 in ca. 44%, 33%, and 13% yields, respectively. The intermediates Cp*W(NO)(CH 2 CMe 3 )(CH 2 C 6 H4-3-Me) and Cp*W(NO)(CH 2 CMe3)(CH 2 C 6 H4-4-Me) were not observed, presumably because they decomposed too rapidly to allow build up during the reaction. The reaction with mesitylene, in contrast, proceeds with concomitant decomposition and yields the benzyl-activated complex, Cp*W(NO)(CH 2C6H3-3,5-Me 2) 2 (4.23), as the only identifiable product in 43% isolated yield. In order to confirm the identity of several of the products produced in the reactions of 2.4 with o-, m-, and /^-xylenes, the ] H N M R data of 4.22 were compared to those of an authentic sample, while complexes 4.15, 4.16, 4.18, 4.19, 4.20, 4.21, and 2.11' were purified and characterized fully by spectroscopic and analytical techniques. With the exception of 4.19 and 2.11', the remaining isolated products exhibit straightforward N M R spectra,37 which indicate the presence of a methylene agostic3 8 interaction ('Jca-H ^ 90 Hz) in 4.15 and 4.18, and an if -benzyl interaction in 4.16 ( V H H = 6 Hz, 8Cj p s o =111) and 4.21 ( 2JHH = 7 Hz, 8Cj p s o = 121).3 9 Broad features, on the other hand, complicate the N M R spectra of 4.19 and 2.11'. Variable-temperature N M R studies of 4.19 suggest that it undergoes two separate dynamic 2 3 2 * processes in solution. The room-temperature process involves an 77 —>7] —>t] interconversion of the m-methylbenzyl ligand. This is supported by the observation of (1) a structureless hump at 8 6.35 that can be ascribed to the aromatic proton that is at the 6-position, (2) a severely broad signal at 8 136.3 due to the carbon to which this proton is attached, and (3) a sharp upfield signal at 8 113.2 due to W C H 2 C i p s o . 4 0 At temperatures below -50 °C, the r)2-benzyl interaction is frozen out, and a second fluxional process that involves hindered rotation about 1 143 223 K 300 K JUL u ppm Q Figure 4.4. Variable-temperature 500-MHz ' H N M R spectra of 2.11' in toluene-rfg (•)• the 3,5-xylyl ligand is observed. For complex 2.11', hindered rotation about the 2,5-xylyl ligand is evidenced at room temperature. As the sample is cooled to -30 °C, two sets of well-resolved resonances, consistent with the presence of two conformational isomers of 2.11', appear in the ' H N M R spectrum (Figure 4.4). The two isomers exist in a 3:1 ratio that is invariant to temperatures between -20 and -80 °C. A ' H - ' H E X S Y experiment (mixing time = 75 ms) in toluene-^ at -50 °C reveals that they are not static structures but are in fact interconverting on the N M R time scale (see the Appendix). 144 The above benzyl-containing products are in general relatively stable. There is no evidence of isomerization during the reaction, and arene exchange is not observed. The neopentyl aryl complexes behave similarly. An exception is that these latter complexes are less stable than their benzyl analogues, often decomposing in trace amounts to unknown products under the conditions of the experiment. Under these conditions no appreciable change in the product ratios is observed. The selectivity with which the reactive intermediates [Cp*W(NO)(=CHCMe 3)] and [Cp*W(NO)(=CHAr)] react with different types of aryl and benzyl C - H bonds therefore can be calculated from these ratios by correcting for the number of each type of bond (see Tables 4.4 and 4.5). With regard to reactions with methylated arenes, Table 4.2 shows that [Cp*W(NO)(=CHAr)] is significantly more discriminating than [Cp*W(NO)(=CHCMe 3)] with respect to aryl vs benzyl C - H activation and that, in all but the case of mesitylene, both species favor the former over the latter, despite the latter being much weaker bonds. Comparison of the reactions of [Cp*W(NO)(=CHCMe 3)] indicates a strong preference for activation of the meta and para C - H bonds relative to the bonds at the ortho and benzylic position. Attack at the ortho position, in fact, is significant only when the ortho C - H bond is not too sterically hindered and the meta and para C - H bonds are absent. Thus, in the absence of severe steric hindrance there is a strong preference for aryl activation over benzyl activation, presumably because of initial formation of 7]2-arene complexes prior to C - H activation and because of formation of a much stronger metal-aryl bond. The observation that [Cp*W(NO)(=CHAr)] is more reactive toward aryl C - H bonds than [Cp*W(NO)(=CHCMe 3)] can be attributed to the presence of the oc-aryl group, which mediates the alkylidene's nucleophilicity. It can also be attributed to the fact that the oc-aryl group is less sterically demanding than the tert-butyl group in [Cp*W(NO)(=CHCMe 3)], thereby making it is easier for arenes to form n2-complexes prior to C - H activation. 145 Table 4.2. Relative Kinetic Selectivities of [Cp*W(NO)(=CHCMe 3)] and [Cp*W(NO)(=CHAr)] for Activation of Aryl and Benzyl C - H Bonds 2.4 -CMe 4 Me 3 C NO ArMe W CH 2 CMe 3 \ -Ar K + ArMe W \ CH 2 CMe 3 CH 2 Ar L -CMe, W CH 2 Ar \ Ar N + W \ CH 2 Ar CH 2 Ar O NO relative product yields (%)" ArMe K L N o Xaryl:Zbenzyl [Cp*W(NO)(=CHCMe3)]c [Cp*W(NO)(=CHAr)]c toluene 82 0 18 0 82:18 2.73:1 >99:1 o-xylene* 66 19 14 1 66:34 2.91:1 19.5:1 m-xylene* 50 0 41 9 50:50 1.50:1 6.83:1 /7-xylene6 44 0 43 13 44:56 1.18:1 4.96:1 mesitylene <20 0 0 >80 -20:80 -1:1.33 -0:100 "Determined by repeated integration of the 'H N M R spectra recorded at either 400 or 500 M H z ; errors are to ca. 3-5%. 'Product distributions obtained from side-by-side thermolyses at 75.0 °C. "Statistically corrected. Table 4.3. Kinetic Selectivities of [Cp*W(NO)(=CHCMe 3)] for Activation of Aryl C - H Bonds in Monosubstituted Benzenes relative product yields" X o:m:p CF 3 ~1 66 33 0:1:1 OCH 3 88 7 5 12.5:1:1.4 C H 3 ~0C 66 33 0:1:1.4 "Determined by repeated integration of the ' H N M R spectra recorded at either 400 or 500 M H z ; errors are to ca. 3-5%. 'Statistically corrected. c Not detected, but possibly present at 5% of total product. In general, transition-metal complexes whose aryl C - H activation reactivity is electronically controlled exhibit high selectivity for para vs meta activation. For example, Mayer and co-workers have shown that the (HBpz3)Re(0)(Cl)(I) complex [HBpz 3 = hydridotris(l-pyrazolyl)borate] reacts with toluene to give the meta- and para-activated products in a 1:10.5 ratio and with more electron-rich substrates such as anisole, only the product of para activation is observed.41 Similarly, Graham and Sweet have reported that the [CpRe(NO)(CO)][BF 4] complex activates toluene to give a 1:31 mixture of meta:para products, but with a,a,a-trifluorotoluene the reverse trend is observed as would be expected since C F 3 is an electron-withdrawing group.4 2 With regard to reactions of [Cp*W(NO)(=CHCMe 3)] with monosubstituted benzenes, Table 4.3 shows that para C - H 147 bonds react only slightly faster than meta C - H bonds when the benzene substituent is electron-donating. When the benzene substituent is electron-withdrawing, no significant difference in rates between meta and para activation is observed. This suggests that electronic effects play a very minor role (if at all) in controlling the selectivity of the aryl C - H reaction of the neopentylidene intermediate. Consistent with this proposal, both a,a,a-trifluorotoluene and toluene, which differ markedly in electronic properties, show nearly identical reactivity toward ortho C - H bonds. That both give essentially no ortho C - H activation strongly suggests that these reactions are governed primarily by steric effects. However, unlike a,a,ot-trifluorotoluene and toluene, anisole reacts to give mostly ortho C-H-activated products, despite having meta/para selectivity essentially identical with that observed for toluene. Clearly, steric effects are not operative in this case. A more reasonable explanation is that anisole reacts by way of a site-directed43 reaction in which prior coordination of the methoxy group to the metal center of [Cp*W(NO)(=CHCMe 3)] directs attack at the ortho position 4 4 4.3.3 Thermal Reactions of CpW(NO)(CH 2CMe 3) 2 (2.1) with Aliphatic and Aromatic C - H Bonds: Qualitative Observations In order to study the effect of the Cp' ligand on the C - H activation chemistry described above, the thermolysis of the Cp complex 2.1 in cyclohexane in the presence of excess PMe 3 and in neat p-xylene, C^D^, and tetramethylsilane was examined. In contrast to 2.4, thermolysis of 2.1 in neat ^ -xylene or tetramethylsilane to 60 °C yields a complex mixture of intractable products. In the case of CeD6, only decomposition is observed, as judged by ' H N M R monitoring of the reaction. When 2.1 is heated in cyclohexane in the presence of excess PMe 3 , on the other hand, the reaction proceeds in the same manner as that described above (4.4) 2.1 4.24 3.1 148 for 2.4 with one exception. The expected C - H activation product, CpW(NO)(cyclohexene)(PMe3) (4.24), is generated as a minor product unless the amount (and concentration) of PMe 3 added is very low compared to that used in the Cp* experiments. Taking together, these observations can be interpreted as follows. The reactivity of the Cp intermediate [CpW(NO)(=CHCMe3)] is analogous to that of the Cp* system, except (a) it does not yield stable products with substrates such as p-xylene, C^De, and tetramethylsilane, and (b) its selectivity for PMe3 coordination vs cyclohexane activation is significantly greater than that observed with its Cp* counterpart, presumably for electronic as well as steric reasons. 4.4 Conclusions In summary, the highly reactive neopentylidene intermediate [Cp*W(NO)(=CHCMe3)] has been shown to undergo rapid addition reactions with the C - H bonds of Me4Si, alkanes, and arenes. The reaction with Me4Si produces the bis(alkyl) complex 2.9 in high yield. The reactions with alkanes, in contrast, lead to different types of rearranged products depending on the alkane structure. In addition, they demonstrate that, in the absence of severe steric hindrance, the C - H bonds that are preferentially attacked are generally those having the highest bond dissociation energies: i.e., cyclopropyl > 1° alkyl ~ 2° alkyl > cycloalkyl » tert-butyl. This is based on the following observations. The major product in the reaction with 1,1,2,2-tetramethylcyclopropane is the metallacycle complex 4.1, formed by initial addition of a cyclopropyl C - H bond across the W=C bond followed by y-cyclometalation. Neohexane, which contains three different types of C - H bonds, is activated regiospecifically at the CH2C / /3 position. In this case, the initially formed complex then undergoes (3-hydrogen abstraction in a stereospecific manner, yielding an rj2-alkene complex having the tert-butyl group syn to the Cp* ligand. In the presence of added PMe3, this alkene complex can be trapped and isolated as 4.4 s y n . Cyclopentane and cyclohexane react similarly in the presence of excess PMe3 to give the corresponding cycloalkene complexes 4.2 and 4.3 as a single isomer. The reactions with methylcyclopentane, methylcyclohexane, ethylcyclohexane, and pentane, in contrast, give complex mixtures containing the r/ 3-allyl hydride complexes 4.8-4.11, respectively, in both the presence and absence of PMe3. The formation of 4.8-4.11 149 is proposed to proceed via the reaction sequence outlined in Scheme 4.6. This proposed mechanism assumes that (1) P-hydrogen abstraction occurs with the same stereospecificity as that observed for the neohexane reaction, (2) subsequent allylic C - H activation is also stereospecific, occurring only when it leads to formation of an endo and acyclic allyl ligand, and (3) 1,3-disubstituted allyl ligands are unstable, decomposing at the temperature at which they form. Although rare, the conversion of an alkane molecule into a ^-coordinated ligand by multiple H-atom loss (i.e., dehydrogenation) is precedented. However, in contrast to the formation of 4.2-4.11, where the neopentylidene ligand accepts two hydrogens from the alkane solvent, most systems in the literature require an added alkene as hydrogen acceptor.45 In the reactions of [Cp*W(NO)(=CHCMe 3)] with arenes, meta- and para-C-H bonds are preferentially attacked, followed by ortho and benzyl C - H bonds. Exceptions are the reaction with anisole, where ortho attack predominates, presumably because of prior coordination of the oxygen atom to metal center, and the reaction with mesitylene, where benzylic activation is strongly favored, presumably because all the aryl hydrogens in this molecule are in sterically hindered environments. Like the Cp*W(NO)(CH2CMe3)(CH2Ph) complex reported in Chapter 3, the benzyl-activated products are unstable with respect to o> abstraction of neopentane. This leads to formation of [Cp*W(NO)(=CHAr)] complexes which favor aryl C - H bonds significantly more than [Cp*W(NO)(=CHCMe 3)]. Finally, the above study also shows that the Cp neopentylidene complex, [CpW(NO)(=CHCMe 3)], is also capable of cleaving C - H bonds intermolecularly. However, unlike the Cp* system, it is significantly more reactive toward P M e 3 than the C - H bonds of cyclohexane and its C - H activation products are in general not stable at the temperature at which they form. 150 4.5 References and Notes (1) The term "activation" of a rj-bond has two meanings in organometallic chemistry: (a) an increase in reactivity of a bond due to some action, and (b) the cleavage of a bond. In this thesis, all " C - H activation reactions" are processes in which the C - H bond is broken. (2) For reviews, see: (a) Shilov, A . E. Activation of Saturated Hydrocarbons by Transition Metal Complexes; D. Reidel: Boston, 1984. (b) Crabtree, R. H . Chem. Rev. 1985, 85, 245. (c) Activation and Functionalization of Alkanes; H i l l , C. H . , Ed.; Wiley, 1989. (d) Selective Hydrocarbon Activation, Principles and Progress; Davies, J. A . , Watson, P. L. , Liebman, J. F., Greenberg, A . , Eds.; V C H : N Y , 1990. (e) Arndtsen, B . A . ; Bergman, R. G.; Mobley, T. A . ; Peterson, T. H . Acc. Chem. Res. 1995, 28, 154. (f) Bengali, A . A . ; Arndtsen, B. A. ; Burger, P. M . ; Schultz, R. H . ; Weiller, B. H . ; Kyle, K . R.; Moore, C. B. ; Bergman, R. G. Pure Appl. Chem. 1995, 67, 281. (g) Shilov, A . E.; Shul'pin, G. B. Chem. Rev. 1997, 97, 2879. (h) Stahl, S. S.; Labinger, J. A . ; Bercaw, J. E. Angew. Chem., Int. Ed. Engl. 1998, 37, 2180. (3) (a) Saillard, J.-Y.; Hoffmann, R. J. Am. Chem. Soc. 1984,106, 2006. (b) Cundari, T. R. J. Am. Chem. Soc. 1994, 116, 340. (c) Siegbahn, P. E. M . ; Svensson, M . J. Am. Chem. Soc. 1994,116, 10124. (d) Siegbahn, P. E. M . J. Am. Chem. Soc. 1996,118, 1487. (e) Niu, S.; Hall, M . B. J. Am. Chem. Soc. 1999,121, 3992. (4) For an example of an exception, see: Arndsten, B . A . ; Bergman, R. G. Science 1995, 270, 1970. (5) (a) Wenzel, T. T.; Bergman, R. G. J. Am. Chem. Soc. 1986, 4856. (b) Desrosiers, P. J.; Shinomoto, R. S.; Flood, T. C. J. Am. Chem. Soc. 1986, 108, 1346. (b) Hackett, M . ; Whitesides, G. M . J. Am. Chem. Soc. 1988,110, 1449. (6) (a) Jordan, R. F.; Guram, A . S. Organometallics 1990, 9, 2116. (b) Guram, A . S.; Jordan, R. F.; Taylor, D. F. J. Am. Chem. Soc. 1991,113, 1833. (c) Lecuyer, C ; 151 Quignard, F.; Choplin, A . ; Olivier, D.; Bassett, J. M . Angew. Chem., Int. Ed. Engl. 1991, 30, 1660. (d) Quignard, F.; Choplin, A . ; Bassett, J. M . J. Chem. Soc., Chem. Commun. 1991, 1589. (a) Watson, P. L . J. Am. Chem. Soc. 1983,105, 6491. (b) Watson, P. L . ; Parshall, G. W. Acc. Chem. Res. 1985, 18, 51. (c) Thompson, M . E.; Bercaw, J. E. Pure Appl. Chem. 1984, 56, 1. (d) Thompson, M . E.; Baxter, S. M . ; Bulls, A . R.; Burger, B. I ; Nolan, M . C ; Santarsiero, B. D.; Schaefer, W. R.; Bercaw, J. E. J. Am. Chem. Soc. 1987,109, 203. (e) Bunel, E.; Burger, B . J.; Bercaw, J. E. J. Am. Chem. Soc. 1988, 110, 976. (f) Evans, W. J.; Chamberlain, L . R.; Ulibarri, T. A . ; Ziller, J. W. J. Am. Chem. Soc. 1988, 110, 6423. (a) Bruno, J. W.; Marks, T. J.; Day, V . W. J. Am. Chem. Soc. 1982,104, 7357. (b) Fendrick, C. M . ; Marks, T. J. J. Am. Chem. Soc. 1986, 108, 425. (a) Groves, J. T.; Nemo, T. E. J. Am. Chem. Soc. 1983, 105, 6243. (b) Traylor, T. G. J. Chem. Soc, Chem. Commun. 1984, 279. (c) Wayland, B . B. ; Del Rossi, K . J. J. Organomet. Chem. 1984, 276, C47. (d) Wayland, B . B. ; Ba, S.; Sherry, A . E. J. Am. Chem. Soc. 1991,113, 5305. (e) Zhang, X . - X . ; Wayland, B . B . J. Am. Chem. Soc. 1994,116, 7897. (f) Brown, S. N . ; Mayer, J. M . J. Am. Chem. Soc. 1994,116, 2219. (a) Cook, G. K . ; Mayer, J. M . J. Am. Chem. Soc. 1994,116, 1855. (b) Brown, S. N . ; Myers, A . W.; Fulton, J. R.; Mayers, J. M . Organometallics 1998,17, 3364. (a) Cummins, C. C ; Schaller, C. P.; Van Duyne, G. D.; Wolczanski, P. T.; Chan, E. A . -W. ; Hoffmann, R. J. Am. Chem. Soc. 1991, 113, 2985. (b) Bennett, J. L. ; Wolczanski, P. T. J. Am. Chem. Soc. 1994, 116, 2179. (c) Bennett, J. L . ; Wolczanski, P. T. J. Am. Chem. Soc. 1997,119, 10696. (a) Walsh, P. J.; Hollander, F. J.; Bergman, R. G. J. Am. Chem. Soc. 1988,110, 8730. (b) Cummins, C. C ; Baxter, S. M . ; Wolczanski, P. T. J. Am. Chem. Soc. 1988,110, 8731. (c) Walsh, P. J.; Hollander, F. J.; Bergman, R. G. Organometallics 1993,12, 152 3705. (d) Lee, S. Y . ; Bergman, R. G. J. Am. Chem. Soc. 1995,117, 5877. (e) Schaller, C. P.; Bonanno, J. B. ; Wolczanski, P. T. J. Am. Chem. Soc. 1994, 116, 4133. (f) Schaller, C. P.; Cummins, C. C ; Wolczanski, P. T. J. Am. Chem. Soc. 1996,118, 591. de With, J.; Horton, A . D. Angew. Chem., Int. Ed. Engl. 1993, 32, 903. Schaller, C. P.; Wolczanski, P. T. Inorg. Chem. 1993, 32, 131. Schafer, D. F.; Wolczanski, P. T. J. Am. Chem. Soc. 1998, 120, 4881. (a) van der Heijden, H. ; Hessen, B. J. Chem. Soc, Chem. Commun. 1995, 145. (b) Coles, M . P.; Gibson, V . C ; Clegg, W.; Elsegood, M . R. J.; Porrelli, P. A . Chem. Commun. 1996, 1963. (c) Tran, E.; Legzdins, P. J. Am. Chem. Soc. 1997, 119, 5071. (d) Cheon, J.; Rogers, D. M . ; Girolami, G. S. J. Am. Chem. Soc. 1997,119, 6814. (a) McDade, C ; Green, J. C ; Bercaw, J. E. Organometallics 1987,1, 1629. (b) van Doom, J. A . ; van der Heijden, H . ; Orpen, A . G. Organometallics 1994,13, 4271. Bulls, A . R.; Schaefer, W. P.; Serfas, M . ; Bercaw, J. E. Organometallics 1987, 6, 1219. (a) Chamberlain, L . ; Rothwell, I. P.; Huffman, J. C. J. Am. Chem. Soc. 1982, 104, 7338. (b) Chamberlain, L . R.; Rothwell, I. P.; Huffman, J. C. J. Am. Chem. Soc. 1986, 108, 1502. (c) Lockwood, M . A . ; Clark, J. R.; Parkin, B . C ; Rothwell, I. P. Chem. Commun. 1996, 1973. Schrock, R. R.; DePue, R. T.; Feldman, J.; Yap, K . B.; Yang, D. C ; Davis, W. M . ; Park, L . ; DiMare, M . ; Schofield, M . ; Anhaus, J.; Walborsky,E.; Evitt, E.; Kriiger, C ; Betz, P. Organometallics 1990, 9, 2262. (a) Debad, J. D.; Legzdins, P.; Batchelor, R. J.; Einstein, F. W. B . Organometallics 1993,12, 2094. (b) For complete spectroscopic and structural characterizations, see Chapter 2 of this thesis. Martin, J. T. Ph.D. Thesis, University of British Columbia, 1988. 153 The syn and anti notation simply represents the orientation of the neohexene tert-butyl group with respect to the Cp* ligand. Bodner, G. S.; Peng, T.-S.; Arif, A . M . ; Gladysz, J. A . Organometallics 1990, 9, 1191. For comparison, the olefinic carbons of free cyclopentane and cyclohexane resonate at about 8 130. See: Silverstein, R. M . ; Bassler, G. C ; Morrill , T. C. Spectrometric Identification of Organic Compounds; John Wiley & Sons, Inc.: New York, 1991; pp 217 and 238. The exceptionally strong 7i-donor ability of the related Cp*W(NO)(PPh3) fragment has recently been documented, see: Burkey, D. J.; Debad, J. D.; Legzdins, P. J. Am. Chem. Soc. 1997,119, 1139. Irradiation of the tert-butyl group was not feasible since its chemical shift is too close to that of =Cr7 a n t iH and =C/7CMe 3. (a) Schilling, B. E. R.; Hoffmann, R.; Faller, J. W. J. Am. Chem. Soc. 1979, 101, 592. (b) Kie l , W. A . ; Lin, G.-Y. ; Constable, A . G.; McCormick, F. B. ; Strouse, C. E.; Eisenstein, O.; Gladysz, J. A . J. Am. Chem. Soc. 1982,104, 4865. (c) Peng, T.-S.; Pu, J.; Gladysz, J. A . Organometallics 1994,13, 929. (d) Gibson, V . C. J. Chem. Soc, Dalton Trans. 1994, 1607. (e) Peng, T.-S.; Winter, C. H . ; Gladysz, J. A . Inorg. Chem. 1994, 33, 2534. For comparison, typical C - C single and C - C double bond distances are 1.54 A and 1.34 A, respectively. (a) Peng, T.-S.; Gladysz, J. A . J. Chem. Soc, Chem. Commun. 1990, 902. (b) Peng, T.-S.; Gladysz, J. A . J. Am. Chem. Soc. 1992,114, 4174. (c) Peng, T.-S.; Arif, A . M . ; Gladysz, J. A . Helv. Chim. Acta 1992, 75, 442. Faller, J. W.; Nguyen, J. T.; Ellis, W.; Mazzieri, M . R. Organometallics 1993,12, 1434. 154 Greenhough, T. J.; Legzdins, P.; Martin, D. T.; Trotter, J. Inorg. Chem. 1979,11, 3268. Ipatkschi, J.; Mirzaei, F.; Demuth-Eberle, G. J.; Beck, J.; Serafin, M . Organometallics 1997,16, 3965. (a) Ward, Y . D.; Villanueva, L . A . ; Allied, G. D.; Payne, S. C.; Semones, M . A . ; Liebeskind, L. S. Organometallics 1995,14, 4132. (b) Villanueva, L . A . ; Ward, Y . D.; Lachicotte, R.; Liebeskind, L. S. Organometallics 1996,15, 4190. Although less likely, it cannot be ruled out that allylic C-H activation is simply highly stereoselective. In general, the resonances for the W - C 7 / 2 A r protons of the benzyl aryl complexes Cp*W(NO)(CH 2Ar)(Ar) appear as two doublets at 5 2 and 3 ( 2 J H H ~ 8 Hz), whereas those of the bis(benzyl) Cp*W(NO)(CH 2 Ar) 2 appear at 5 0.8 and 3 ( 2 7 H H ~ 8 Hz). In general, the resonances for the W-Cr7 2 CMe3 protons of neopentyl aryl complexes appear as two doublets at 8 -2 and 4 (27HH ~12 Hz), whereas those of neopentyl benzyl complexes appear at 8 -2 and 2 ( 2JRH ~12 Hz). (a) Brookhart, M . ; Green, M . L . H . J. Organomet. Chem. 1983, 250, 395. (b) Brookhart, M . ; Green, M . L . H . ; Wong, L . - L . Prog. Inorg. Chem. 1988, 36, 1. (c) Crabtree, R. H. ; Hamilton, D. G. Adv. Organomet. Chem. 1988, 28, 299. For comparison, the corresponding V H H values of T]'-benzyl ligands in related complexes are ~10 Hz, while the ipso carbons are located at 8 150-160. For examples, see: (a) Legzdins, P.; Jones, R. H. ; Phillips, E. C ; Yee, V . C ; Trotter, J.; Einstein, F. W. B . Organometallics 1991,10, 986. (b) Dryden, N . H . ; Legzdins, P.; Trotter, J.; Yee, V . C. Organometallics 1991,10, 2857. Because the remaining ' H and 1 3 C resonances are sharp, it is unlikely that the broadness of the ortho C H group is due to a fluxional process involving either a y-155 agostic C - H interaction or a slow rotation about the C a - C p bond of the m-methylbenzyl ligand. Brown, S. N . ; Myers, A . W.; Fulton, J. R.; Mayer, J. M . Organometallics 1998,17, 3364. Sweet, J. R.; Graham, W. A . G. J. Am. Chem. Soc. 1983,105, 305. For a review of substrate-directable chemical reactions, see: Hoveyda, A . H . ; Fu, G. C ; Evans, D. A . Chem. Rev. 1993, 93, 1307. In further support of this proposal, thermolysis of 2.4 in 1-ethoxy-4-methylbenzene (see below) has been found to produce complexes 4.12* and 4.12** in ~90% and 10% yields, respectively. 4.12*: Anal. Calcd. for C 2 4 H 3 7 N O 2 W : C, 51.90; H , 6.72; N , 2.52. Found: C, 51.63; H , 6.78; N , 2.53. IR (cm - 1) VNO 1565 (s). MS 555 [ M + , 1 8 4 W ] . ' H N M R (400 M H z , C 6 D 6 ) 8 -2.63 (d, V H H = 11.6, 1H, W C H 2 ) , 1.03 (t, V H H = 6.8, 3H, OCH 2 Gtf 3 ) , 1.33 (s, 9H, CMe 3 ) , 1.65 (s, 15H, C 5 Me 5 ) , 2.23 (s, 3H, ArMe), 3.50 (q, V H H = 7.1, 2H, O C / / 2 C H 3 ) , 4.98 (d, V H H = 11.6, 1H, W C H 2 ) , 6.52 (d, V H H = 8.1, 1H, ArH), 6.96 (dd, V H H = 8.1, V H H = 1.9, 1H, ArH), 8.21 (br s, 1H, ArH). 1 3 C { ' H } (75 M H z , CDC1 3) 8 10.2 (C5Me5), 15.3 (OCH 2 CH 3 ) , 20.5 (ArMe), 33.4 (CMe 3), 41.6 (CMe 3 ), 62.9 (OCH 2 CH 3 ) , 109.6 (ArCH) , 111.0 (C 5 Me 5 ) , 129.8 (WCH 2 ) , 134.0 (CipS 0Me), 141.0 (ArCH) , 157.1 (C i p s o OEt) , 168.5 (WCi p s o ) . NOEDS (400 M H z , C 6 D 6 ) irrad. at 2.23 ppm, NOEs at 8 6.96, 8.15. 4.12**: MS 619 [ M + , 1 8 4 W ] . ' H N M R (300 M H z , C 6 D 6 ) 8 1.45 (t, V H H = 6.8, 3H, OCH 2 Cr7 3 ) , 4.10 (m, 2H, OGrY 2 CH 3 ) , 6.25 (d, V H H = 8.1, 1H, ArH), 6.49 (d, V H H = 8.1, ArH). Other signals were obscured. 156 (45) For examples, see: (a) Crabtree, R. H . ; Mellea, M . F.; Mihelcic, J. M . ; Quirk, J. M . J. Am. Chem. Soc 1982, 104, 107. (b) Burk, M . J.; Crabtree, R. H . ; McGrath, D. V . ; J. Chem. Soc, Chem. Comm. 1985, 1829. (c) Baudry, D.; Ephritikhine, M . ; Felkin, H. ; Holmes-Smith, R. J. Chem. Soc, Chem. Commun. 1983, 788. (d) McGhee, W. D.; Bergman, R. G. J. Am. Chem. Soc. 1988,110, 4246. 157 C H A P T E R 5 Kinetic and Mechanistic Studies of the Generation and C - H Activation Reactions of the [Cp'W(NO)(=CHCMe3)] Fragments 5.1 Introduction 157 5.2 Experimental Section 160 5.3 Results and Discussion 170 5.4 Conclusions 189 5.5 References and Notes 190 5.1 Introduction a-Hydrogen abstraction from an alkyl ligand is a fundamental process in organometallic chemistry and has played an important role in the development of transition-metal alkylidene1/ carbene2 complexes. As with the more ubiquitous (3-hydrogen abstraction reaction, an oc-hydrogen may be abstracted by an external reagent,3 the metal,4 or a neighboring ligand.5 Abstraction of an oc-hydrogen by the metal is often reversible, only takes place in coordinatively unsaturated <f complexes, where n > 2, and may or may not be followed by reductive elimination of H X , where X is a ligand originally attached to the metal. Direct transfer of an oc-hydrogen atom to a neighboring ligand on the other hand occurs with no formal change in the metal oxidation state, is rarely reversible, and has a rate that often depends on factors such as the degree of steric congestion at the metal center, the solvent, and the presence of added donor ligands. More complicated pathways involving initial Cp* ring-metalation or y - C - H activation followed by a-hydrogen abstraction have also been noted. For example, it has been shown that (1) the reaction of the bis(alkyl) Cp*W(NO)(CH 2 SiMe 3 ) 2 with L D A leads to formation of an "ate" alkylidene complex in which the ring-metallated species [(T? 5,?? 1-158 C 5Me4CH2)W(NO)(CH2SiMe3)2][Li 2(THF)3] is an observable intermediate, 6 and (2) thermolysis of (PMe3)2Ta(Me)(Mes)Br3 (Mes = C6H 3-2,4,6-Me 3) leads to loss o f methane and formation of a substituted benzylidene complex via abstraction of a y-hydrogen atom and subsequent rearrangement of an unobserved benzotantallacyclobutene compound. 7 In the preceding chapters, it was reported that the bis(alkyl) complexes Cp*W(NO)(CH 2 CMe3) 2 and its Cp analogue exhibit a-agostic structures in solution and solid states and that they decompose readily at elevated temperatures to give neopentane and r presumably the coordinatively unsaturated neopentylidene fragments, [Cp'W(NO)(=CHCMe3)], which can effect the C - H bond cleavage of alkanes, arenes, and certain alkenes. In the activation of alkanes and arenes, several remarkable selectivity features o f [Cp*W(NO)(=CHCMe 3)] have been noted (see Chapter 4). First, this intermediate can activate aliphatic C - H bonds even in the presence of excess PMe3. 8 Second, it reacts most rapidly with aromatic compounds, showing a strong preference for activation of the meta and para C - H bonds relative to the bonds at the ortho and benzylic positions. Attack at the ortho position, in fact, is significant only when the reaction is site-directed or when the ortho C - H bond is not too sterically hindered and the meta and para C - H bonds are absent. Third, in the activation of alkanes the C - H bonds that are preferentially attacked are generally those having the highest bond dissociation energies; the qualitative order of reactivity, for example, is cyclopropyl > 1° alkyl ~ 2° alkyl > cycloalkyl » tert-butyl. There are a number of cases now known in which the kinetic selectivity for different types of C - H bonds is similar to that of [Cp*W(NO)(=CHCMe 3)]. 9 For many of these systems, it has been found that strong M - C bonds are formed when strong C - H bonds are broken and that the relative ease of C - H activation is determined largely by the strengths of the product M - C bonds and by the binding affinities o f the hydrocarbon to the metal center prior to activation. 1 0 ' 1 1 For example, arenes are generally more reactive than alkanes because of prior coordination of the metal to the 7i-electrons to give n 2-arene complexes and because of formation of a much stronger M - C bond. 159 Molecular complexes in which an arene is weakly coordinated to an unsaturated metal center in an 7]2-fashion were first formulated as early as the mid-1960s.1 2 Since then, the area of arene as well as alkane coordination has grown steadily, and has led to the discovery that such adducts not only exist as independent entities 1 1 ' 1 3 but also as metastable intermediates in a number of C - H bond making and breaking reactions.14 For reactions involving C - H activation, the initial formation of alkane a-complexes and 772-arene complexes is apparently necessary for two reasons: (1) to allow the metal to coordinate interchangeably (and hence discriminate) different types of C - H bonds, and (2) to provide a lower energy pathway for the bond-breaking step.15 Recently, there has been considerable research effort devoted to understanding the mechanisms of intermolecular C - H activation by soluble transition-metal complexes. However, in few of these investigations has it been possible to obtain direct information about the reactive intermediates without examining also the reverse reactions. Much of the current understanding of C - H activation by way of oxidative addition, for example, has come from mechanistic studies of alkane reductive elimination reactions, where several lines of evidence based on isotope effect and isotope tracer experiments have implicated the existence of alkane a-complexes. With respect to the mechanism by which the Cp'W(NO)(CH 2CMe 3)2 complexes decompose and react, no intermediates have yet been directly observed, although several trapping reactions with phosphines and alkenes (see Chapter 3) have strongly implicated the intermediacy of [Cp'W(NO)(=CHCMe 3)] in which the tert-butyl group points away from Cp' ligand. In this chapter, the first part reports an in-depth study of the kinetics and mechanisms of the thermal decompositions and subsequent C - H activation reactions of the bis(neopentyl) complexes. Specifically, kinetic isotope effect and labeling studies are described which not only further support the intermediacy of [Cp'W(NO)(=CHCMe 3)], but which also strongly implicate the intermediacy of alkane adducts, [Cp'W(NO)(=CHCMe 3)(?7 2-R-H)], along the pathway for both the forward and reverse reactions. A mechanism involving a net transfer of an cc-hydrogen from one neopentyl group to the cc-carbon of the adjacent neopentyl group is proposed; whether 160 neopentane is lost by direct a-hydrogen abstraction or by a-hydride elimination followed by reductive elimination could not be distinguished. In the second part of this chapter, an investigation of the thermal stabilities of a series Cp'W(NO)(R)(R') complexes that are electronically and structurally analogous to the bis(neopentyl) compounds is described. Two major conclusions from this work are the following: (1) the pathway by which these compounds decompose is governed primarily by the strengths of the M - C bond formed and broken, and (2) the presence of a-agostic interactions in the ground state structures has little, i f any, effect on the rate and pathway of decomposition. 5.2 Experimental Section 5.2.1 Methods The synthetic methodologies employed throughout this thesis are described in detail in Section 2.2.1. Unless otherwise stated, all thermolyses were carried out in J. Young N M R tubes or thick-walled glass vessels immersed in an oil bath at 65-70 °C for ca. 2 d. 5.2.2 Reagents THF and cyclohexane used in kinetic experiments were distilled from Na/benzophenone ketyl and then stored in glass bombs under dinitrogen. Dimethylphenylphosphine, trimethylphosphine, triethylphosphine, «-butyl ether (Aldrich), C^De (Cambridge Isotopes), C 6 H 3 D 3 (98 %, Aldrich), and tetramethylsilane-dn were dried over sodium. Dimethyldichlorosilane (Aldrich) was freshly distilled. C D 3 I (99.5+ atom % D, Aldrich) was dried over 4-A molecular sieves. The organometallic complexes CpW(NO)(CH2CMe3)2 (2.1), Cp*W(NO)(CH 2 CMe 3 ) 2 (2.4), Cp*W(NO)(CH 2 CMe 3 ) (CH 2 CMe 2 Ph) (2.8), and Cp*W(NO)(CH 2 CMe 3 ) (CH 3 ) (2.10) were prepared as described in Section 2.2.1. The a-deuterated complexes CpW(NO)(CD 2 CMe 3 ) 2 (2.1-rf4), Cp*W(NO)(CD 2 CMe 3 ) 2 (2.4-rf4), and Cp*W(NO)(CH 2 CMe 3 ) (CD 2 CMe 2 Ph) (2.8-</2) were prepared in a similar manner using the dialkylmagnesium reagents R2Mg-(dioxane) (R = C D 2 C M e 3 , CD 2 CMe 2 Ph) ; an exception is that 161 the crude product mixtures were filtered through Celite instead of alumina I prior to crystallization. In all cases, l H N M R spectroscopy at 300 M H z revealed no detectable methylene proton signals, thus indicating at least 98% D incorporation at C a . 5.2.3 NMR and Mass Spectral Data for Cp*W(NO)(CH2CMe3)(CD2CMe2Ph) (2.8-</2) M S 555 [ M + , 1 8 4 W ] . ' H N M R (500 M H z , CDC1 3) 5-2.18 (d, 2 J H H = 12.5, 1H, W C H 2 ) , 1.27 (s, 9H, CMe 3 ) , 1.45 (s, 15H, C 5 Me 5 ) , 1.74 (s, 3H, CMeMe), 1.85 (s, 3H, CMeMe), 3.16 (d, V H H = 12.4, 1H, W C H 2 ) , 7.06 (t, V H H = 7.3, 1H, Hp), 7.23 (t, V H H = 7.4, 2H, H m ) , 7.50 (d, V H H = 7.4, 2H, H 0 ) . 2 H{ 'H} N M R (77 M H z , C 6 H 6 ) 8 -0.54 (s, ID, W C D 2 ) , 2.47 (s, ID, W C D 2 ) . 5.2.4 Preparation of (CH 3) 2Si(CD 3) 2 M g turnings (4.00 g, 167 mmol) and ra-butyl ether (350 mL) were added to a 500-mL three-necked flask equipped with a stir-bar and a reflux condenser. The flask was cooled to -78 °C and C D 3 I (6.0 mL, 94.4 mmol) was added dropwise by syringe. The resulting mixture was stirred at room temperature for 3.5 h, during which time its color changed to grey-black. The supernatant solution was filtered through Celite ( 2 x 2 cm) into a 500-mL two-necked flask and cooled to -40 °C. Freshly distilled M e 2 S i C l 2 (5.10 mL, 42.0 mmol) was added dropwise by syringe. Once the addition was complete, the flask was removed from the cold bath, wrapped with aluminum foil, and the contents were stirred (under a static atmosphere of argon) for another 16 h. A pale brown-yellow solution resulted which contained an off-white precipitate in suspension. At this point, the contents of the flask were quickly transferred to a one-necked flask equipped with a short-path distillation head. Under a flow of argon, colorless (CH 3 ) 2 Si (CD 3 ) 2 (~l-2 mL, b.p. 28-32 °C) was distilled into a receiving flask immersed in a dry ice/acetone bath (Care should be taken to avoid bumping!). To remove iodide byproducts, the distillate was transferred to a Pyrex bomb containing PPh 3 (~200 mg), degassed, and stored in the dark for ~5 d. Next, it was vacuum transferred into another bomb containing a Na mirror and allowed to dry for 4 d before use. The purity of (CH 3 ) 2 Si (CD 3 ) 2 was established by ' H N M R spectroscopy and by G C - M S mlz 94 [M + ] . 162 5.2.5 Preparation of Cp*W(NO)(CH2CMe3)(C6D5) (2.11-rf5) A wine-red solution of 2.4 (60 mg, 0.12 mmol) in C g D 6 (~1 mL) was heated at 70 °C for ~2 d, during which time its color changed to purple-red. ' H N M R analysis of the sample revealed the formation of free neopentane and the known Cp*W(NO)(CHDCMe3)(C6D5) (2.11-rf6)'6 (~95%) along with a small amount of unidentified products, which displayed signals in the range 8 1-2. The organic volatiles were then removed in vacuo to obtain an oily residue, which was chromatographed on alumina 1(1 x 1.5 cm) by eluting with hexanes. The first band to appear, a pink fraction, was collected by eluting with 4:1 hexanes/Et20. Concentration of this fraction, followed by cooling to -30 °C for 1 d provided 2.11-</5 as dark wine-red needles (18 mg, 30% yield) [Occasionally, a residual amount of l.W-d^ could still be detected by ' H N M R ] . The ' l i N M R spectrum of 2.11-rfs is identical to that of the protio analogue 2.11 (see Chapter 2), except it does not display aromatic resonances. 5.2.6 Kinetic Measurements A l l kinetic experiments were performed by UV-vis spectroscopy. Spectra were recorded on a HP 8452A Diode Array UV-vis spectrophotometer equipped with a deuterium lamp and a thermostated cell holder connected to a V W R Scientific constant-temperature bath (model 1160A) filled with ethylene glycol/water and maintained to ±0.05 °C. The spectrophotometer was interfaced to a 586 PC computer using HP 89531A operating software. Stock solutions of organometallic reagent were prepared in the appropriate solvent in the glovebox and stored at -30 °C between runs. A representative stock solution of 2.1, 2.1-di, 2.4, or 2.4-</4 contained 0.013 mmol of the complex dissolved in 25.0 mL of solvent in a volumetric flask. In a typical experiment, a UV-vis cell equipped with a Kontes Teflon stopcock was charged sequentially with 3.0 mL of the solution of the organometallic reagent and a measured amount of trapping reagent where required. Each run was monitored for >3 half-lives. In all but a few cases, each rate constant was determined at least three times for each system and each temperature studied. Spectra were recorded at regular time intervals after the solutions had been allowed to reach the desired temperatures (5-15 min). The absorbance intensities of 2.1, 2.4, and their deuterated 163 analogues at A, = 486 nm were monitored in all cases and then plotted versus time. Infinity values were determined by non-linear regression with least-squares fitting of the raw data using SigmaPlot 3.0; Jandsen Graphing Software, 1987-1992. Using Microsoft Excel, the observed rate constants (& 0bs) were then calculated from the slope of the plots of ln[(At-Aoo)/(At-Ao)] versus time, which displayed excellent linearity with correlation coefficients typically 0.995 or better. Activation parameters, AH* and AS*, were determined from Eyring plots of ln(& 0b s/T) versus 1/T, where AH* = - R x slope, AS* = R[intercept - \n(k^/h)] and R, &B, and h are the gas constant, Boltzmann's constant, and Planck's constant, respectively. Errors are 95% confidence limits of linearly regressed lines. 5.2.7 ! H and 2{1H} N M R Analyses of the Thermolysis of (a) 2A-d4 and 2A-d4 in THF in the Presence of PMe3 and (b) 2A-d4 in CeD 6 In a typical reaction, a J. Young N M R tube containing 10-20 mg of organometallic reagent was charged with ca. 0.5 mL of solvent and, where needed, an excess amount of PMe3 by vacuum transfer. The resulting mixture was heated in a constant-temperature bath at 65.0 °C in the case of 2.l-d4 and 75.0 °C in the case of 2A-d4 and monitored periodically by ' H (CeD^) and 2 H{ 'H} (CeUe) N M R spectroscopies. In all cases, the thermolyses were relatively clean and H/D exchange between the a- and y-positions of the neopentyl ligands of the starting materials during the course of the thermolyses was evidenced by the presence of low-intensity ' H N M R signals in the methylene regions. At >90% conversion, the ' H and 2 H N M R spectrum of the reactions of 2.1-d4 and 2A-d4 with PMe 3 showed peaks consistent with the presence of both Cp'W(NO)(=CDCMe 3 )(PMe 3 ) (major) and Cp'W(NO)[=CHCMe 2 (CD 3 )](PMe 3 ) (minor). The ' H and 2 H{ 'H} N M R spectra of the C 6 D 6 reaction of 2A-d4 similarly revealed the formation of Cp*W(NO)(CHDCMe 3 ) (C 6 D 5 ) along with a small amount of Cp*W(NO)[CH 2CMe2(CD 3)](C 6D5). 164 5.2.8 Thermolysis of 2.4 in THF in the Presence of Excess PMe 3 and C 6 D 6 In the glovebox, a small Pyrex vessel was charged with 2.4 (24 mg, 0.049 mmol), THF (2-3 mL), and CeD6 (21 uL, 0.24 mmol, 5 equiv). The wine-red solution was then degassed by three freeze-pump-thaw cycles. Next, PMe3 (>10 equiv) was added by vacuum transfer and the resulting solution was heated at 70 °C for ca. 16 h, after which time the organic volatiles were removed in vacuo. ' H N M R spectroscopy showed that Cp*W(NO)(=CHCMe 3)(PMe 3) (3.1) was the only product formed. A n identical result was also obtained when Si(CH 3 )4 was substituted for C^De-5.2.9 G C - M S Analysis of Neopentane Generated from the Thermolysis of (a) 2.1-d4 and 2A-d4 in THF in the Presence of Excess PMe 3 and (b) 2A-d4 in Neat PMe 3 These experiments were carried out in 0.4-0.5 mL of solvent and were stopped after 2-3 d at 65 °C in the case of 2.1-d4 and 75 °C in the case of 2A-d4. The organic volatiles were then removed in vacuo and analyzed by G C - M S , which showed the absence of dineopentyl. The mass spectra of the reaction of 2.\-d4 with PMe 3 in THF showed the presence of neopentane-fifo, -d\, -di, and -rf3 in respective yields of ca. 2%, 17%, 4%, and 77% (see Table 5.1). In the case of 2.4-^ 4, the corresponding isotopic distribution was ca. 0%, 18%>, 3%, and 79% in THF and ca. 0%, 10%, 0%, and 90% in neat PMe 3 . As expected, ' H N M R (C 6 D 6 ) analysis of the nonvolatiles of the reaction of 2A-d4 in neat PMe 3 revealed the presence of a small amount of hydrogen at the cc-position. 5.2.10 Thermolysis of 2.4 in Si(CD 3) 4 A sample of 2.4 (29 mg, 0.057 mmol) in tetramethylsilane-Ji2 (0.5 mL) in a J. Young N M R tube was heated in an oil bath at 70 °C for 2.5 d, during which time its color became deep purple red. A *H N M R spectrum of this solution showed the clean formation of Cp*W(NO)(CHDCMe 3 ) [CD 2 Si(CD 3 ) 3 ] (2.9-dn) and 1 equiv of neopentane. The organic volatiles were removed in vacuo. The remaining red-black solid was extracted with pentane and 165 filtered through Celite. Cooling this filtrate to -30 °C overnight afforded 2.9-dn (27 mg, 93% yield) as mauve crystals. Anal. Calcd. for C 1 9 H 2 5 D i 2 N O S i W : C, 43.94; H , 7.13; N , 2.70. Found: C, 44.26; H , 7.09; N , 2.74. IR (cm - 1) v N O 1554 (s), v C D 2204 (w). MS 519 [ M + , 1 8 4 W ] . *H N M R (300 MHz, C 6 D 6 ) 5 1.36(s, 9H, CMe 3 ) , 1.51 (s, 15H, C 5 Me 5 ) , 3.19 (br s, lH,WGr7DC). 2 H { ' H } N M R ( 7 7 M H z , C 6 H 6 ) 8-2.16 (s, ID, WCHDC) , -1.38 (s, ID, WCD 2 Si ) , 0.33 (s, 9D, Si(CD 3 ) 3 ) , 1.06 (s, ID, WCD 2 Si ) . 1 3 C{ 'H} N M R (75 M H z , C 6 D 6 ) 8 9.9 (C 5 Me 5 ) , 34.2 (CMe 3), 39.7 (CMe 3 ), 104.8 (t, 'JCD = 14.9, W C H D C ) , 109.8 (C 5 Me 5 ) . 1 3 C resonances attributable to C D 2 and Si(CD 3 ) 3 could not be observed, presumably due to C - D coupling, quadrupolar effect, and lack of nuclear Overhauser enhancement. 5.2.11 Thermolysis of 2.4 in C^D^ See the first part of Section 5.2.5. 5.2.12 Thermolysis of 2.4 in (CH3)2Si(CD3)2 A sample of 2.4 (14 mg, 0.025 mmol) in a J. Young N M R tube was dissolved in (CH 3 ) 2 Si (CD 3 ) 2 (~0.3 mL) and then heated in a constant-temperature bath at 70.0 °C for 1.5 d, after which time the volatiles were removed in vacuo. The residue was dissolved in CeD6 and two ' H N M R spectra (400 M H z , pulse delay of 10 seconds) were recorded. The ratio of Cp*W(NO)(CH 2 CMe 3 ) [CH 2 Si(CH 3 ) (CD 3 ) 2 ] (2.9-d6-H) to Cp*W(NO)(CHDCMe 3 ) [CD 2 Si(CD 3 ) (CH 3 ) 2 ] (2.9-</6-D) was determined to be 1.35(5):1 by integration of the methylene signals at 8 3.20 and 3.27. ' H N M R (500 M H z , C 6 D 6 ) 8-2.11 (dd, 2 J H H = 12.4, 4JHH = 1.4, 1H, WCH 2 C) , -1 .30 (dd, 2 J H H = 12.0, V H H = 1.4, WCH 2 Si ) , 0.38 [s, SiMe(CD 3 ) 2 and SiMe 2 (CD 3 )] , 1.07 (s, CMe 3 ) , 1.35 (d, 2JHH = 12.4, WCH 2 Si ) , 1.52 (s, C 5 Me 5 ) , 3.20 (s, W C H D C ) , 3.27 (d, V H H = 12.4, 1H, W C H 2 C ) . 166 5.2.13 Thermolysis of 2.4 in 1:1 C 6 H 6 / C 6 D 6 Two samples of 2.4 (16 mg, 0.033 mmol) in J. Young N M R tubes were dissolved in an equimolar mixture of C6H 6/C6D 6 (0.4 mL). The tubes were then heated side by side in a constant-temperature bath at 70.0 °C for ~1.5 d. After removal of the organic volatiles in vacuo, the residues were dissolved in CeDe and two ' H N M R spectra (400 M H z , pulse delay of 10 s) were recorded for each experiment. The average ratio of Cp*W(NO)(CH 2CMe 3)(C6H 5) (2.11) to Cp*W(NO)(CHDCMe 3 ) (C 6 D 5 ) (2.11-rf6) was determined to be 1.05(5): 1 by integration of the methylene signals at 8 4.37 and 4.50. ' H N M R (500 M H z , C 6 D 6 ) 8-2.05 (d, V H H =11.4, 1H, W C H 2 ) , 1.26 (s, CMe 3 ) , 1.53 (s, C 5 Me 5 ) , 4.37 (s, WCHD) , 4.50 (d, V H H = 11.4, 1H, W C H 2 ) , 7.08 (t, 3 J H H = 7.4, ArH), 7.19 (t, VHH = 7.4, ArH), 7.71 (d, V H H = 7.7, ArH). 5.2.14 Thermolysis of 2.4 in l,3,5-C 6H 3D 3-</ 3 Two samples of 2.4 (14 mg, 0.025 mmol) in J. Young N M R tubes were dissolved in l ,3,5-CeH 3 D 3 (~0.3 mL). Each sample was heated in a constant-temperature bath at 70.0 °C for ~1.5 d, after which time the organic volatiles were removed in vacuo. The residues were then dissolved in C 6 D 6 and two ' H N M R spectra (400 M H z , pulse delay of 10 seconds) were recorded for each experiment. The average ratio of Cp*W(NO)(CH 2 CMe 3 ) (C 6 H 2 D 3 ) (2.11-</3-H) to Cp*W(NO)(CHDCMe 3 ) (C 6 H 3 D 2 ) (2.11-</3-D) was determined to be 1.35(5):1 by integration of the methylene signals at 8 4.34 and 8 4.48. ' H N M R (400 M H z , C 6 D 6 ) 8-2.02 (d, V H H =11.4, 1H, W C H 2 ) , 1.25 (s, CMe 3 ) , 1.54 (s, C 5 Me 5 ) , 4.34 (s, WCHD) , 4.48 (d, V H H = 11.4, 1H, W C H 2 ) , 7.08 (s, ArH), 7.18 (s, ArH), 7.71 (s, ArH). 167 5.2.15 Thermolysis of Cp*W(NO)(CH 2CMe3)(CH 2CMe 2Ph) (2.8) in THF in the Presence of Added PMe 3 To a Pyrex reactor containing 2.8 (50 mg, 0.090 mmol) was added THF (2 mL) and excess PMe3 by vacuum transfer. The resulting mixture was thawed and heated at 75 °C overnight, during which time its color changed from brown-red to yellow-orange. The organic volatiles were removed in vacuo and the residue was analyzed by ' H N M R (CeD^) spectroscopy, which showed the presence of Cp*W(NO)(CH2CMe2-o-C6H4)(PMe3) (5.1) as the major species along with Cp*W(NO)(=CHCMe 2)(PMe 3) (3.4) and Cp*W(NO)(CH 2CMe3)(C 6H 4CMe 3) (5.2) (as a ~1.4:1 mixture of m- and/>isomers) in a combined yield of ~15%. Next, the crude product sample was redissolved in a minimum amount of E t 2 0 and stored at -30 °C for 2 d to obtain 5.1 (15 mg, 30% yield) as yellow-orange blocks. The mother liquor was taken to dryness and chromatographed on alumina I by eluting with hexanes. The first band to appear, a pink fraction, was collected by eluting with 4:1 hexanes/Et20. Concentration of this fraction, followed by cooling to -30 °C for 2 d afforded 5.2 (~2 mg) as mauve needles contaminated by a trace amount of 5.1. Cp*W(NO)(CH 2CMe 2-o-C 6H 4)(PMe 3) (5.1): Anal. Calcd. for C 2 3 H 3 6NOPW: C, 49.56; H , 6.51; N , 2.51. Found: C, 49.78; H , 6.54; N , 2.52. M S 557 [ M + , 1 8 4 W ] . ' H N M R (500 M H z , CDC1 3) 5 1.18 (s, 3H, CMeMe), 1.21 (d, V H P = 8.9, 9H, PMe 3 ) , 1.36 (s, 3H, CMeMe), 1.49 (d, 2JHH = 8.8, IH, W C H 2 ) , 1.65 (dd, 2JHH = 8.8, V H P = 2.3, IH , W C H 2 ) , 1.87 (s, 15H, C 5 M e 5 ) , 6.84 (m, IH, ArH), 6.94 (m, 2H, ArH), 7.11 (d, 3 J H H = 7.6, IH, ArH). 1 3 C{ 'H} N M R (75 M H z , CDC1 3) 5 10.5 (C 5 Me 5 ) , 14.1 (d, LJCP = 29, PMe 3 ) , 30.6 (CMeMe), 36.1 (d, V C P = 3, CMeMe), 51.5 (d, 3 J C P = 3, CMeMe), 57.4 ( 2 J C P = 7, W C H 2 ) , 107.0 (C 5 Me 5 ) , 122.7 (Ar CH), 123.0 (Ar CH), 124.2 (Ar CH), 138.4 (d, VCp = 13, Ar CH), 165.9 (d, V C P = 9, C i p s o ) , 173.2 (d, 2JCP = 21, WCipSo). 3 1 P{ 'H} N M R (202 M H z , C 6 D 6 ) 6-20.4 (^PW = 211). Cp*W(NO)(CH 2CMe 3)(C 6H 4CMe 3) (5.2): *H N M R (500 M H z , C 6 D 6 ) 6 5.2m: -1.82 (d, 2 J H H =11.5, IH, W C H 2 ) , 1.26 (s, 9H, CH 2 CMe 3 ) , 1.29 (s, 9H, C i p s o C M e 3 ) , 1.58 (s, 15H, C 5 Me 5 ) , 4.38 (d, 2 J H H = 11.6, IH, W C H 2 ) , 7.19 (d, 3 JHH = 7.2, IH , ArH), 7.22 (t, V H H = 7.4, IH, 168 ArH), 7.62 (d, V H H = 7.5, 1H, ArH), 7.84 (s, 1H, ArH). 5.2,: -1.92 (d, V H H = 11.5, 1H, W C H 2 ) , 1.22 (s, 9H, C i p s o C M e 3 ) 1.26 (s, 9H, C H 2 C M e 3 ) , 1.58 (s, 15H, C 5 M e 5 ) , 4.38 (d, V H H = 11.6, 1H, W C H 2 ) , 7.29 (d, V H H = 8.0, 2H, ArH), 7.78 (d, V H H = 7.9, 2H, ArH). 5.2.16 Thermolysis of C p * W ( N O ) ( C H 2 C M e 3 ) ( C D 2 C M e 2 P h ) (2.8-rf2) in T H F in the Presence of Added P M e 3 This reaction was carried out in the same manner as that described above for the protio complex 2.8. Analysis of the crude reaction mixture by ' H N M R spectroscopy indicated the presence of Cp*W(NO)(CD 2 CMe 2 -o -C 6 H 4 ) (5.1-</2), Cp*W(NO)(=CHCMe 3)(PMe 3) (3.4) (W=CH8 11.2), and Cp*W(NO)(CH 2 CMe 3 ) [C 6 H 4 CMe 2 (CD 2 H)] (5.2-d2). Complexes 5.1-d2 and 5.2-d2 were purified, respectively, by crystallization from E t 2 0 and by chromatography on alumina I with hexanes and hexanes/Et20 as described above. 5.1- d2: MS mlz 559 [ M + , 1 8 4 W ] . ' H N M R (500 M H z , C 6 D 6 ) 8 0.89 (d, 27Hp = 8.8, 9H, PMe 3 ) , 1.58 (s, 15H, C 5 Me 5 ) , 1.68 (s, 3H, CMeMe), 1.75 (s, 3H, CMeMe), 7.06 (t, V H H = 6.9, 1H, ArH), 7.10 (d, V H H = 6.9, 1H, ArH), 7.16 (t, V H H = 7.1, 1H, ArH), 7.23 (d, V H H = 7.5, 1H, ArH). 2 H{ 'H} N M R (77 M H z , C 6 H 6 ) 8 1.80 (br s, 2D, W C D 2 ) . 1 3 C { ' H } (CDC1 3) 8 10.5 (C 5 Me 5 ) , 14.1 (d, V C P = 29, PMe 3 ) , 30.6 (CMeMe), 36.1 (CMeMe), 51.3 (CMeMe), 107.0 (C 5 Me 5 ) , 122.6 (Ar CH), 123.0 (Ar CH), 124.2 (Ar CH), 138.4 (d, V C P = 13, A r CH), 165.9 ( C ^ ) , 173.2 (d,JCp = 2 1 , W C i p s o ) . 5.2- d2\ ' H N M R (500 M H z , C 6 D 6 ) 8 5.2m-d2: -1.82 (d, V H H = 11.5, 1H, W C H 2 ) , 1.26 (s, 9H, C H 2 C M e 3 ) , 1.29 [s, 7H, Ci P soCMe 2 (CD 2 H)] , 1.58 (s, 15H, C 5 M e 5 ) , 4.38 (d, VHH =11.6, 1H, W C H 2 ) , 7.19 (d, V H H = 7.2, 1H, ArH), 7.22 (t, V H H = 7.4, 1H, ArH), 7.62 (d, V H H = 7.5, 1H, ArH), 7.84 (s, 1H, ArH). 5 . 2 ^ : -1.82 (d, V H H =11.5, 1H, W C H 2 ) , 1.26 [s, 7H, C i p s 0 M e 2 ( C D 2 H ) ] , 1.29 (s, 9H, C H 2 C M e 3 ) , 1.58 (s, 15H, C 5 Me 5 ) , 4.38 (d, VHH = 11-6, 1H, W C H 2 ) , 7.29 (d, V H H = 7.8, 2H, ArH), 7.78 (d, V H H = 7.9, 2H, ArH). 2 H{ 'H} N M R (77 M H z , C 6 H 6 ) 8 1.26 [br s, CMe 2 (CD 2 H)] . 169 5.2.17 Thermolysis of Cp*W(NO)(CHDCMe 3)[CD 2Si(CD 3)3] (2.9-dn) in Si(CH 3) 4 A solution of 2.9-du in ~0.5 mL of tetramethylsilane was heated in a J. Young N M R tube at ~70 °C for 2 d. Neither reaction nor H/D exchange could be detected in the ' H and 2 H{ 'H} N M R spectra of a CeD6 sample of the crude reaction mixture. 5.2.18 Thermolysis of Cp*W(NO)(CH 2CMe 3)(CH 3) (2.10) To a CeD 6 solution (~0.5 mL) of freshly sublimed 2.10 in a J. Young N M R tube was added P M e 3 (~1.2 equiv) by vacuum transfer. A n immediate color change from dark red to yellow resulted. Inspection of the ' H N M R spectrum at room temperature revealed the presence of an equilibrium mixture containing Cp*W(NO)(CH 2CMe3)(CH 3)(PMe 3), 2.10, and free PMe 3 . Only one isomer of Cp*W(NO)(CH 2 CMe 3 )(CH 3 )(PMe 3 ) was detected at 400 M H z , namely that in which the P M e 3 is coordinated between the methyl and nitrosyl ligands. Cp*W(NO)(CH 2CMe 3)(CH 3)(PMe 3): ' H N M R (400 M H z , C 6 D 6 ) 8 0.00 (d, 3 7 H P = 16.7, 3H, W C H 3 ) , 0.61 (dd, 2JHH = 11.5, V H P = 5.9, IH, W C H 2 ) , 1.03 (d, 2 7 H p = 8.7, 9H, PMe 3 ) , I. 42 (d, 2JHH= 11.5, 1 H , W C H 2 ) , 1.52 (s, 9H, CMe 3 ) , 1.54 (s, 15H, C 5 M e 5 ) . 1 3 C { ' H } N M R (75 M H z , C 6 D 6 ) 5 9.8 (C 5 Me 5 ) , 11.6 (d, lJc? = 28, PMe 3 ) , 12.8 (d, 2 J C P = 21, lJCH = 128, W C H 3 ) , 35.4 (CMe 3), 38.4 (CMe 3 ) , 54.6 (d, 2JCP =11, ' J C H = 120, W C H 2 ) , 106.0 (C 5 Me 5 ) . 3 1 P{ 'H} N M R (202 M H z , C 6 D 6 ) 8 -17.4 ( ' j P W = 140). NOEDS (400 MHz) irrad. at 0.00 ppm, NOEs at 8 0.61 and 1.03; irrad. at 0.61 ppm, NOEs at 8 0.00 and 1.42; irrad. at 1.03 ppm, NOEs at 8 0.00 and 1.54. When THF-dg was used in place of CeD 6 as solvent, an immediate color change from dark red to yellow also resulted. Thermolysis of this mixture to 70 °C in the dark showed no noticeable reaction after 2 days. After an additional 3 d at ~85 °C, traces of 3.4 (W=CH 8 II . 25) was detected by ' H N M R spectroscopy (CeD6) along with similar amounts of unidentified decomposition products, which displayed broadened peaks in the Cp* region. No methylidene products were observed. 170 5.2.19 Thermolysis of Cp*W(NO)(CH2CMe3)(C6D5) (2.11-</5) in C 6 D 6 A C 6 D 6 solution (~0.4 mL) of 2.11-</5 in a J. Young N M R tube was prepared. The tube was then placed in an N M R probe at 70 °C. ' H N M R spectra were acquired at regular time intervals over the course of ~10 h. Although a small amount of decomposition was observed, neither arene exchange nor H/D scrambling was evident. 5.3 Results and Discussion 5.3.1 Isotopic Labeling Studies The most straightforward mechanism for the thermal decomposition and C - H activation reaction of Cp 'W(NO)(CH 2 CMe 3 )2 [Cp' = Cp (2.1), Cp* (2.4)] is that shown in Scheme 5.1. Because this reaction sequence involves both C - H bond scission and formation, valuable mechanistic information may be obtained by replacement of the migrating hydrogen with deuterium. In this context, two types of deuterium-labeling experiments have been carried out in order to probe this mechanistic possibility. Scheme 5.1 Me 3 C H CMe 3 171 First, thermolysis of 2.4 in neat Me4Si-iii2 at ~70 °C for 2.5 d results in the quantitative formation of neopentane and the mixed bis(alkyl) complex Cp*W(NO)(CHDCMe 3 ) [CD 2 Si(CD 3 ) 3 ] (2.9-d12) in which a deuterium is incorporated into the a-position of the neopentyl ligand (Scheme 5.2). Similarly, thermolysis of 2.4 in neat CeD6 at -70 °C for 2 d produces the phenyl neopentyl derivative Cp*W(NO)(CHDCMe 3 ) (C 6 D 5 ) (2.11-de) in which the a-carbon of the neopentyl ligand is monodeuterated. Minor resonances downfield of the Cp* region are also detected in the ' H N M R spectrum in this case and are assignable to products arising from slow decomposition of 2.11-^6 during the late stages of the reaction (see later). For both complexes 2.9-dn and 2.11-</6, analysis of the crude product mixture by N M R spectroscopy indicates that only one of the neopentyl methylene protons is replaced by deuterium. Particularly indicative of the structures illustrated in Scheme 5.2 is the ' H N M R resonance of the remaining proton, which appears as a singlet and is shifted upfield relative to the resonance observed in the unlabeled complex by ca. 0.07 ppm for 2.9-dn and 0.14 ppm for 2.11-</6- Thus, although no intermediates could be detected when the reaction with CeD6 was followed by ' H N M R spectroscopy, these results unambiguously demonstrate that [Cp*W(NO)(=CHCMe 3)] is a key intermediate in the thermal decomposition of 2.4, that (in agreement with the results of the trapping reactions with phosphines and certain alkenes) the Scheme 5.2 2.9-d 12 2.4 2.11-d6 172 neopentylidene moiety is anti with respect to the Cp* ligand, and that 2.9-du and 2 . 1 a r e formed as a result of a cis C - D addition across the tungsten-carbon double bond in this unobserved unsaturated intermediate. In the second type of deuterium-labeling study, the preparation and thermal decomposition of the oc-deuterated derivatives Cp 'W(NO)(CD 2 CMe 3 )2 [Cp' = Cp (2.l-d4), Cp* (2.4-d4)] were carried out. The preparation of 2.1-d4 and 2.4-d4 can be carried out in a manner analogous to that for 2.1 and 2.4 using (Me3CCD2)2Mg-(dioxane) as the alkylating agent. These deuterium-containing compounds, however, undergo H/D exchange with Et20/pentane and Et20/hexanes when filtered slowly through alumina I. Their purification therefore is best achieved by filtration of the crude product mixtures through Celite followed by crystallization. For both compounds, the methylene resonances were the only signals observed in the 2 H { 1 H } N M R spectrum. No corresponding methylene signals were observed in the ' H N M R spectrum, indicating an isotopic purity of at least 98%. To determine the fate of the oc-hydrogen abstracted and the source of the hydrogen consumed in converting one neopentyl group to neopentane, the thermal decompositions of 2.1-d4 and 2.4-d4 were performed in THF in the presence of excess P M e 3 for 2-3 d at 65 °C and 75 °C, respectively. As can be seen from Scheme 5.3 and Table 5.1, analysis of the crude reaction mixtures by lH and 2 H N M R spectroscopies and the organic volatiles by GC/mass spectrometry shows the formation of Cp'W(NO)(=CDCMe 3)(PMe 3) and neopentane-<i3 as the major products Scheme 5.3 T H F A R = H (2. W 4 ) ; Me (2.4-d4) + CMe4-d3 + CMe 4-cf 1 173 in these reactions. In each case, small quantities of Cp'W(NO)[=CHCMe 2 (CD 3 )](PMe 3 ) and neopentane-c/i are also observed, with the latter constituting 17-18% of the neopentane mixture. A reasonable explanation for these observations is that loss of neopentane from 2A-d4 and 2.4-d4 occurs reversibly during the course of the thermolysis and that this process involves an a-hydrogen abstraction/elimination pathway in which an a-deuterium (hydrogen) from one neopentyl group is transferred to the a-carbon atom of the other. Since neopentane-^ is not observed as a product, it can also be concluded that neither ring metalation nor y-hydrogen activation is a competitive side-reaction. Table 5.1. Isotopic Composition of Neopentane Derived from Thermal Decompositions of 2.1-d4 and 2A-d4 compd isotopic composition of neopentane (%)" reaction medium -do" -d\ -d2» -di 2.1-d4 THF, excess PMe 3 2 17 4 77 2.4-d4 THF, excess PMe 3 0 18 3 79 2.4-d4 neat PMe 3 0 10 0 90 "These are estimates derived directly from the GC-MS data (see the Appendix); they are not corrected for isotopic impurities present in the starting materials. 'These percentages are within experimental error. Further evidence that loss of neopentane occurs reversibly during the thermolysis of 2.1 and 2.4 comes from reaction of 2A-d4 with CeD 6 . When 2A-d4 is heated in neat CeD6 at 70 °C, H/D exchange between the a- and y-positions of the neopentyl ligands in 2A-d4 is again found to occur, as judged by ' H and 2 H{ 'H} N M R spectroscopies. Concurrently, neopentane and the benzene-activated isotopomer Cp*W(NO)(CD 2CMe 3)(C6D 5) form. A small quantity of Cp*W(NO)[CHDCMe 2 (CD 3 )](C6D 5 ) is also observed with deuterium incorporated into the same a-position as that described above for 2.11-^ 6. This is based on the observation that the W-CHD resonance appears exclusively as a low-intensity broad singlet at 8 4.37. This demonstrates that the thermal decomposition of 2A-d4 proceeds by an a-H abstraction route 174 wherein isotopic label scrambling between the a- and y-positions of the neopentyl ligands occurs prior to product formation. The H/D exchange described above in principle could occur inter- or intramolecularly as depicted in Scheme 5.4. Path A can be discounted since thermolysis of 2.4 (70 °C, 16 h) in THF in the presence of excess PMe3 (>10 equiv) and 5 equiv of Me4Si or CeD6 produces Cp*W(NO)(=CHCMe 3)(PMe3), with no detectable products resulting from intermolecular C-H(D) activation. This is not unexpected and can be attributed to the fact that the added hydrocarbon in each case is too low in concentration to compete with PMe3 for reaction with intermediate [Cp*W(NO)(=CHCMe3)]. The possibility of a solvent-caged mechanism (i.e. path B) can also be tested by a competition experiment. When 2.4-</4 is heated to 75 °C in neat PMe3, Cp*W(NO)(=CDCMe3)(PMe3) is observed with detectable proton incorporation in the op-position. In addition, the neopentane produced is approximately 90% d3 and 10% d\ as determined by G C / M S analysis (see Table 5.1). This strongly suggests that neopentane does not react as a solvent-caged species. More likely is that the H/D exchange occurs intramolecularly via path C. According to this path, the starting bis(neopentyl) does not decompose to give [Cp*W(NO)(=CDCMe3)] directly. Rather, the initial step involves reversible formation of the o-complex, [Cp*W(NOX=CDCMe 3 )(77 2 -D-CD 2 CMe 3 )]. The observation of neopentane-^, in addition to the expected neopentane-c/3, indicates that the metal in this complex also migrates readily and intramolecularly from complexation with an cc-C-D bond to complexation with a y-C - H bond with eventual formation of Cp*W(NO)(CHDCMe 3 ) [CH 2 CMe 2 (CD3)] . It also strongly suggests that this exchange process occurs competitively with loss of neopentane from [Cp*W(NO)(=CDCMe 3)(77 2-D-CD 2CMe 3)] and that the latter process is irreversible. This leads to formation of the [Cp*W(NO)(=CDCMe3)] intermediate, which then coordinates PMe 3 or another (and more concentrated) hydrocarbon. 175 Scheme 5.4 5.3.2 Isotope Effect Studies Additional evidence that an intermediate alkane 0-complex exists along the pathway by which 2.1 and 2.4 eliminate neopentane comes from studies of the isotope effects on the reverse reaction, i.e., the C - H activation reaction of [Cp*W(NO)(=CHCMe 3)]. The isotope effects on the C - H activation reaction of [Cp*W(NO)(=CHCMe 3)] were measured as shown in Scheme 5.5. Thermolysis of 2.4 in a mixture of 1:1 C 6 H 6 / C 6 D 6 at 70.0 °C for 1.5 d results in the formation of the benzene C - H and C - D activation products, 2.11 and 2.11-</6. By integration of the methylene proton signals at 8 4.50 and 4.37 (Figure 5.1 A) , the ratio of 2.11 to 2.ll-d6, was found to be 1.05(5): 1. Similarly, the thermolysis of 2.4 in 1,3,5-C 6H 3D 3 at 70.0 °C for 1.5 d produces Cp*W(NO)(CH 2 CMe 3 )(2,4,6-C 6 H 2 D 3 ) (2.11-rf3-H) and Cp*W(NO)(CHDCMe 3)(3,5-176 Scheme 5.5 1 : 1 C 6 H 6 / C 6 D 6 A Me 3 C o 2.11 + N Me 3 C g kH/kD= 1.05(5): 1 2.11-rf,; 1,3,5-C 6 H 3 D 3 1 N 1 Me 3 C o C M e 3 2.4 ( C H 3 ) 2 S i ( C D 3 ) 2 i N D + Me 3 C O D ^ - D 2.11-</3-H N Me 3 C o D D %/* D = 1.35(5): 1 N 2.11-rf3-D Me 3 C o Si(CH 3)(CD 3) 2 %/)tD= 1.35(5): 1 N I 2.9-</6-H Me 3 C $ Si(CH 3) 2(CD 3) 2.9-</6-D CeH 3 D 2 ) (2.11-^ 3-D). The ratio of the benzene products in this case (Figure 5. IB) is 1.35(5): 1, indicating a slight preferential activation of a C - H bond over a C - D bond. A n analogous result is obtained when (CH 3 ) 2 Si(CD 3 ) 2 is substituted for 1,3,5-C 6H 3D 3 as solvent. (CH 3 ) 2 Si(CD 3 ) 2 was prepared in a straightforward manner as shown in eq 5.1. Thermolysis of 2.4 in (CH 3 ) 2 Si (CD 3 ) 2 at 70.0 °C for 1.5 d leads to the formation of Cp*W(NO)(CH 2 CMe 3 ) [CH 2 Si(CH 3 ) (CD 3 ) 2 ] (2.9-</6-H) and Cp*W(NO)(CHDCMe 3 ) [CD 2 Si(CH 3 ) 2 (CD 3 ) ] (2.9-</3-D) in a 1.35(5):1 ratio by integration of the methylene proton resonances at 5 3.27 and 3.20, respectively. 177 xs Mg 1/2 (CH 3 ) 2 SiCl 2 C D 3 I *• C D 3 M g I ( C H 3 ) 2 S i ( C D 3 ) 2 (5.1) «-Bu 20 K-BU 20 •pr 1 1 1 1 1 1 j 11—1—1—1—r —1—1—1—r~1—|—1—1—1—1—1—1—r ppm A.A ppm - 2 . 0 B • 111111111111111111 1 1 1 1 1 1 1 1 ppm 4 . 4 ppm - 2 . 0 Figure 5.1. 500-MHz ' H N M R spectra of the methylene regions of (A) Cp*W(NO)(CH 2 CMe 3 ) (C 6 H 5 ) (2.11) and Cp*W(NO)(CHDCMe 3 ) (C 6 D 5 ) (2.11-rf6) and (B) Cp*W(NO)(CH 2CMe 3)(2,4,6-C 6H 2D3) (2.11-^ -H) and Cp*W(NO)(CHDCMe 3 )(3,5-C 6 H 3 D 2 ) (2.11-^ -D). Each of the experiments outlined in Scheme 5.5 was carried out under conditions that ensure that once C - H or C - D cleavage has occurred isomerization of the initially formed product is not followed. The observed product ratios therefore reflect the kinetic isotope effects for activation of benzene and tetramethylsilane by [Cp*W(NO)(=CHCMe 3)]. This demonstrates that a small but real isotope effect (ku/kD = 1.35(5)) is observed when [Cp*W(NO)(=CHCMe 3)] has to compete for C - H / C - D bonds in the same molecule. When the C - H / C - D bonds belong to separate molecules, the isotope effect (kHlkG = 1.05(5)) is negligible, i f present at all. These results require that direct C - H bond addition across the tungsten neopentylidene moiety is not occurring and that an intermediate o-complex is involved. The involvement of an intermediate 178 o-complex allows the C-H(D) bonds in the same molecule to coordinate interchangeably with the metal center prior to bond activation. Since the zero-point energy of a C - H bond is higher than that of a C - D and since this energy difference is greatest when the two bonds are terminal, there is a thermodynamic preference for the former to coordinate to the metal and undergo subsequent bond cleavage. Similar isotope effects have been reported by Jones et. al., Bergman et. ai, and Flood et. al. for the thermal activation of benzene by [Cp*Rh(PMe 3)], [Cp*Ir(PMe 3)(77 2-C 3H 6)], and [CnRh(PMe 3)][BAr 4] [Cn= l,4,7-trimethyl-l,4,7-triazacyclononane; A r = C 6 H 3 -3,5-(CF 3 ) 2 ] , respectively. Bergman and Flood proposed that their observed isotope effects could be due to initial formation of either a o-complex or an n2-arene complex. Jones and Feher, on the other hand, postulated the intermediacy of an r/2-arene complex for their reactions. As additional evidence, they found that when benzene is replaced by£>-di-/ert-butylbenzene, an r/2 complex can be directly observed by ' H N M R spectroscopy at low temperatures. In the activation of benzene by [Cp*W(NO)(=CHCMe 3)], the presence of an intermediate Tj2-arene complex would provide a lower energy pathway for the interconversion of one a-complex to another. Unfortunately, such a species can neither be proved nor disproved based on the observed kinetic isotope effects discussed above. The fact that the kH/ko values for reactions with 1,3,5-C6H3D3 and (CH 3) 2Si(CD 3)2 are identical may simply be coincidental. 5.3.3 Kinetic Studies Kinetic studies were carried out on the thermal decomposition of CpW(NO)(CH 2 CMe 3 ) 2 (2.1) and Cp*W(NO)(CH 2 CMe 3 ) 2 (2.4) in THF and cyclohexane in the presence of greater than 5-fold excess of phosphine, where phosphine = PMe 3 , PMe 2 Ph, or PEt 3 . The rates of disappearance of 2.1 and 2.4 were measured by UV-vis spectroscopy by monitoring the decrease in absorbance intensity at 486 nm; measured rate constants are given in Tables 5.2 and 5.3. In all cases examined, the rates of decomposition were first-order in metal complex for >3 half-lives and showed no dependence on the concentration and nature of added phosphine. 179 Table 5.2. Rate Constants for the Decomposition of 2.1 and 2.4 as a Function of Phosphine and Phosphine Concentration entry compd" solvent temp (°C)* phosphine [phosphine] (M) Us (S-')C 1 2.4 cyclohexane 91.0 PMe3 0.0032 4.84(10) x 10"4 2 91.0 PMe3 0.016 4.72(7) x 10"4 3 91.0 PMe3 0.16 4.55(7) x 10~4 4 2.4 THF 81.5 PMe3 0.016 2.16(4) x 10"4 5 81.5 PMe2Ph 0.016 2.30(3) x 10"4 6 81.5 PEt3 0.016 2.13(2) x 10"4 7 2.1 THF 81.0 PMe3 0.016 5.72(5) x 10"4 8 81.0 PMe2Ph 0.016 5.75(4) x 10"4 9 81.0 PEt3 0.016 5.52(4) x if/ 4 "[organometallic] ~ 5.2 x 10"4 M . 6 Error - 1 °C. "Entries 1-3 were obtained from 2-3 runs and entries 4-9 were obtained from 3-4 runs; errors were calculated according to [E(Sj) 2 ] 1 / 2 /n, where Sj is the standard deviation of k\. Comparison of the rate data of entries 4 and 7 of Table 5.2 indicates that 2.4 decomposes slower than 2.1 by a factor of ca. 2.7 (Figure 5.2); the latter in turn decomposes at a rate ca. 10 times slower than the previously reported CpMo(NO)(CD2CMe3)217 system. 180 time (s) Figure 5.2. Representative first-order kinetic plots for the decomposition of 2.1 (•) and 2.4 (•) in THF in the presence of excess PMe3 at 81 °C. For both compounds 2.1 and 2.4, the decomposition is slightly faster in THF solvent than in cyclohexane (Table 5.3). By arbitrarily setting the k0bS value derived from reaction in cyclohexane at 81 °C to 1.0, the relative rates of decomposition of 2.1 and 2.4 in THF are 2.7 and 1.3, respectively. These results stand in contrast to the rate of decomposition of CpTa(CH2CMe 3 )2Cl 2 1 8 reported by Schrock and co-workers, which is slowed by donor solvents, and to the rate of decomposition of CpMo(NO)(CD 2 CMe 3 )2, 1 6 which is invariant to changes in solvent polarity and donor ability. 181 Table 5.3. Rate Constants for the Thermal Decompositions of 2.1 and 2.4 in the Presence of Excess P M e 3 a entry compd" temp (°C)6 solvent *obs ( s ' r solvent £obs (S"')C 1 2.4 71.5 THF 8.43(11) x 10"5 cyclohexane 5.23(3) x 10-5 2 81.5 2.16(4) x 10 4 1.67(2) x if/ 4 3 91.0 5.61(8) x 10"4 4.58(5) x 10"4 4 99.0 1.20(6) x 10"3 1.23(2) x 10"3 5 2.1 62.0 THF 8.74(12) x 10~5 cyclohexane 2.46(1) x 10"5 6 71.0 2.31(5) x 10"4 6.48(3) x 10"5 7 81.0 5.72(5) x 10-4 2.09(2) x 10"4 8 90.0 1.42(4) x 10"3 4.90(7) x 10"4 "[organometallic] ~ 5.2 x 10"4 M . 6 Error ~ 1 °C. "Obtained from 3-4 runs; errors were calculated according to [Z(s,)2] " 2 /n, where Sj is the standard deviation of k\. The rates of thermal decomposition of 2.1 between 60 and 90 °C and of 2.4 between 71 and 99 °C in THF and cyclohexane gave linear Eyring plots as shown in Figure 5.2. From these plots, the activation parameters listed in Table 5.4 were determined. The close to zero entropies of activation for both 2.1 and 2.4 in cyclohexane are comparable to those observed by Bercaw and co-workers for the cc-hydrogen transfer reactions involving Cp*M(CH2R)2 compounds [M = Ti (AS* = -2.85(71) eu), 1 9 H f (AS* = 1(3) eu)2 0], and suggest that there is little reorganization on approach to the transition state in this solvent. 182 Figure 5.3. Eyring plots for the thermal decompositions of 2.1 (•) and 2.4 (•) in cyclohexane in the presence of excess PMe3. Table 5.4. Activation Parameters for the Thermal Decompositions of 2.1 and 2.4 in THF and Cyclohexane in the Presence of Excess P M e 3 a Entry compd" solvent AH* (kcal/mol)A AS*(eu) fcc 1 2.4 THF 24.0(2.2) -7.9(6.1) 2 cyclohexane 28.3(3.5) 3.6(9.8) 3 2.1 THF 23.3(1.7) -7.9(4.9) 4 cyclohexane 25.1(2.1) -4.8(6.1) "[Organometallic] ~ 5.2 x 10"4 M . 'Errors are 95% confidence limits. c l eu = 1 cal/K-mol = 4.1 84 J/K-mol. A n examination of the kinetics of the thermal decomposition of CpW(NO)(CD 2 CMe 3 ) 2 (2.1-</4) and Cp*W(NO)(CD 2 CMe 3 ) 2 (2.4-rf4) reveals a decrease in rate compared to 2.1 and 2.4 (Table 5.5). At ca. 81 °C, the kH/kD values for decompositions of 2.1/2.1-</4 and 2.4/2.4-</4 in THF are 2.86 and 2.50, respectively, indicating that an cc-C-H(D) bond is being broken in the transition state of the rate-determining step. These isotope effects are smaller than values of 183 3-5, 5 a ' 2 1 which are generally observed for simple intramolecular oc-hydrogen abstraction reactions. They are, however, similar in magnitude to those reported for the thermal decompositions of Cp* 2 Ti(CH 3 ) 2 ( k H / k D = 2.92(10) at 110 °C) 1 8 and Ta(CH 2 CMe 3 ) 5 (kH/kD = 2.7 at 60 °C) . 2 2 A common feature of Cp* 2 Ti(CH 3 ) 2 and Ta(CH 2 CMe 3 )5 is that when deuterium is substituted for hydrogen in the oc-position of the alkyl groups, they undergo alkane elimination not only by oc-deuterium abstraction but also by ring metallation and y-hydrogen abstraction, respectively. The slightly reduced kH/kD values observed in these reactions therefore can be ascribed to the lack of kinetic contributions from these side reactions. In contrast to Cp* 2 Ti(CD 3 ) 2 and Ta(CD 2CMe 3)s, abstraction of a y-hydrogen or a hydrogen from a Cp or Cp* ring is not observed in the thermal decompositions of CpW(NO)(CD 2 CMe 3 ) 2 (2.l-d4) and Cp*W(NO)(CD 2 CMe 3 ) 2 (2A-d4) (see above). Instead, thermolyses of these tungsten complexes show that H/D exchange between the cc- and y-positions of the neopentyl ligands, leading to formation of Cp 'W(NO)(CDHCMe 3 ) [CH 2 CMe 2 (CD 3 ) ] , occurs competitively with loss of neopentane-^-, from the starting compounds. The origin of the normal but relatively small isotope effects observed in the decomposition of Cp 'W(NO)(CD 2 CMe 3 ) 2 therefore is due to the kinetic contribution from loss of neopentane from Cp 'W(NO)(CD 2 CMe 3 ) 2 and Cp 'W(NO)(CDHCMe 3 ) [CH 2 CMe 2 (CD 3 ) ] . Since the latter species can only undergo cc-C-H bond cleavage, the overall deuterium isotope effect is diluted by its decomposition. Table 5.5. Rate Constants for Thermolyses of 2.1-</4and 2.4-^4 in the Presence of Excess PMe 3 compd" solvent temp (°C)* Us (S"')C 2.1-d4 THF 81.5 2.00(4) x 10"4 2.86(6) 2A-d4 THF 81.5 8.64(24) x 10"5 2.50(8) 2A-d4 cyclohexane 91.0 1.92(3) x 10"4 2.39(5) "[Organometallic] ~ 5.2 x 10"4 M . 'Error ~ 1 °C. "Obtained from duplicate runs; errors were calculated according to [2(Si/k;)2] 1 / 2 (k H /k D ) , where s, is the standard deviation of kv 184 5.3.4 Thermolyses of Other Cp'W(NO)(R)(R') Complexes: Qualitative Observations In the preceding chapters it was shown that the benzyl neopentyl complexes, Cp*W(NO)(CH 2 CMe 3 ) (CH 2 Ar) (Ar = Ph, C 6 H 4 -o-Me, C 6 H 4 -m-Me, C 6H 4-/>Me) also decompose by a-hydrogen abstraction to give neopentane. Significantly, it was found that while the benzyl, w-methylbenzyl, and /j-methylbenzyl complexes decompose cleanly and readily at a rate that is comparable to that of 2.4 (i.e., 1.5-2 d at 70 °C), the o-methylbenzyl complex decomposes partially (<20%) and rather messily under similar conditions. Two features of these results are significant. First, although the neopentyl but not the benzyl ligand is involved in a-agostic bonding (see Chapter 2), a-hydrogen abstraction from a benzyl ligand is more facile than a-hydrogen abstraction from a neopentyl ligand. This selectivity presumably is driven by the breaking of a relatively weak W-neopentyl bond and the formation of a strong W=C(H)Ar bond. Second, even when the benzyl methylene hydrogens are hindered by o-methyl substitution, a-hydrogen abstraction from the neopentyl ligand does not take place. This result suggests that the presence of an a-agostic interaction in the ground state structure is not a prerequisite for the cleavage of an a - C - H bond. To gain additional insight into the structural features which determine the thermal reactivity of other Cp'W(NO)(R)(R') complexes, the thermolyses of the bis(alkyl) and alkyl aryl complexes Cp*W(NO)(CH 2 CMe 3 ) (CH 2 CMe 2 Et) (2.7) (described in detail in Chapter 4), Cp*W(NO)(CH 2 CMe 3 ) (CH 2 CMe 2 Ph) (2.8), Cp* W(NO)(CHDCMe 3 ) [CD 2 Si(CD 3 ) 3 ] (2.9-dn), Cp*W(NO)(CH 2 CMe 3 ) (CH 3 ) (2.10), and Cp*W(NO)(CH 2 CMe 3 ) (C 6 D 5 ) (2.11-rf5) were carried out. Qualitatively 2.7 and 2.8 decompose at a rate comparable to that observed for 2.4, whereas the sterically less bulky 2.9-</i2,2.10 , and 2.11-</5 are significantly more stable. As noted in Chapter 4, thermolysis of 2.7 in THF in the presence of excess PMe 3 at 70-75 °C for 1.5 d produces a mixture containing as the major (and only identified) products Cp*W(NO)(=CHCMe 3)(PMe 3) (3.4) and Cp*W(NO)(=CHCMe 2Et)(PMe 3) (4.7) in a ratio of ~1:1, and Cp*W(NO)(PMe 3 ) 2 , whose mechanism of formation is not yet known (eq 5.2). Under similar reaction conditions, thermolysis of the neophyl neopentyl complex 2.8 at 75 °C 185 (5.2) 2.7 + ^ W - ^ Me.P i PMe 3 J N O overnight in the presence of excess PMe 3 results in the formation of a yellow-orange mixture containing the ortho-metalated complex Cp*W(NO)(CH2CMe2-o-C6H4)(PMe3) (5.1) as the major product (eq 5.3). The minor products, formed in trace amounts and identified on the basis of their ' H N M R spectral data, are 3.4 and a 1.4:1 mixture of m- and p-Cp*W(NO)(CH2CMe3)(C6H4CMe3) (5.2). Complex 5.1 was isolated as yellow-orange blocks in 30% yield. Its structure was confirmed by X-ray diffraction. As shown in Figure 5.4, the ORTEP diagram of 5.1 clearly shows that the phenyl ring is orthometalated. The W(l)-C(18) and W ( l ) - C ( l 1) bond lengths are 2.244(5) and 2.236(4) A, respectively, and are slightly longer than those in previously characterized CpW(NO)(CH 2 CMe 2 -o-C 6 H 4 ) (NCMe) (2.232(8) and 2.201(8) A). 2 3 In addition, the PMe 3 is coordinated cis to the W-C(sp 2 ) bond, and not cis to the W-C(sp 3 ) bond as observed for the N C M e ligand in the Cp compound. (5.3) 5.2 (m:p= 1.4:1) 186 C(14) C(20) C(19) W ( l ) - N ( l ) 1.793(4) W ( l ) - P ( l ) 2.5089(11) W(l) -C(18) 2.244(5) W ( l ) - C ( l l ) 2.236(4) N( l ) -0 (1 ) 1.216(5) C(17)-C(18) 1.539(7) C ( l l ) - C ( 1 2 ) 1.498(7) W ( l ) - N ( l ) - 0 ( 1 ) 174.6(4) N ( l ) - W ( l ) - C ( l l ) 115.74(19) N( l ) -W( l ) -C(18) 84.47(19) N ( l ) - W ( l ) - P ( l ) 77.30(13) C(17)-C(18)-W(l) 112.8(3) Figure 5.4. ORTEP diagram and selected bond distances and angles of complex Cp*W(NO)(CH2CMe2-o-C6H4)(PMe3) (5.1). To determine whether 5.1 is formed by 8-hydrogen abstraction in unobserved intermediate, [Cp*W(NO)(=CHCMe 2Ph)], the a-deuterated complex Cp*W(NO)(CH 2 CMe 3 ) (CD 2 CMe 2 Ph) (2.8-d2) was prepared and its thermal reactivity examined. The thermolysis of 2.8-</2 was carried out in THF in the presence of excess PMe3 and afforded a mixture containing 3.4,5.1- d2, and 5.2-*/2 in yields similar to those observed in the reaction of 2.8 (eq 5.4). Complex 5.2-</2 was isolated in >90% purity by chromatography on alumina I and characterized by ' H and 2 H{ 'H} N M R spectroscopies. Complex 5.1-</2 was isolated by crystallization from E t 2 0 and characterized by mass spectrometry and by ' H , 2 H { ' H } , and 1 3 C { ' H } N M R spectroscopies. Key to the identification of 5.1-</2 is the absence of methylene resonances in its ' H and 1 3 C N M R spectra and the presence of a peak (m/z 559) corresponding to the molecular ion in its EI mass spectrum. This demonstrates that 5.1-</2 is formed as a result of metalation at the 5-position of the neophyl ligand, followed by coordination of PMe 3 , and that cc-hydrogen abstraction from C H 2 C M e 3 and 8-hydrogen abstraction from C H 2 C M e 2 P h are competitive processes in the thermal decomposition of 2.8. The observation that 8-hydrogen abstraction from a neophyl ligand is preferred over cc-hydrogen 187 5.2-</2 (m:p= 1.4:1) abstraction is not without precedent. The thermal decompositions of (dppe)Pt(CH2CMe2Ph)2 and CpV(CH2CMe2Ph)2(PMe3),25 for example, have been shown to cyclometalate exclusively through 8-H activation. In the case of 2.8, 5-H activation is preferred even though both alkyl ligands are involved in a-agostic bonding with the metal (see Chapter 2). This kinetic preference presumably is driven by the formation of a strong M-C(aryl) bond. In contrast to the above bis(alkyls), Cp*W(NO)(CHDCMe 3 ) [CD 2 Si(CD 3 ) 3 ] (2.9-dn) shows no evidence of decomposition or H/D scrambling analogous to that observed for 2.4-d4 after a sample was heated in neat Me4Si at -70 °C for 2 d. In the case of Cp*W(NO)(CH 2 CMe 3 ) (CH 3 ) (2.10), it reacts instantaneously with P M e 3 in either C 6 D 6 or THF at ambient temperature to give a yellow solution whose ' H N M R spectrum is consistent with the formation of Cp*W(NO)(CH 2 CMe 3 )(CH 3 )(PMe 3 ) as an equilibrium mixture (eq 5.5). Only one isomer of Cp*W(NO)(CH 2 CMe 3 )(CH 3 )(PMe 3 ) is formed, namely that in which the PMe 3 is cis to the methyl and nitrosyl ligands, according to N O E difference data. Like 2.9-dn, both 2.10 and its P M e 3 adduct are fairly stable thermally. They decompose partially in THF at -85 °C for 3 d to give traces of Cp*W(NO)(=CHCMe 3)(PMe 3) (3.4) and unidentified products. Although the methylidene complex Cp*W(NO)(=CH 2)(PMe 3) was not apparent in the ' H N M R spectrum of the crude reaction mixture, it is possible that [Cp*W(NO)(=CH 2)] was also generated but that this species was too reactive to be trapped by PMe 3 . Thus it appears that while a-abstraction of 188 3.4 (5.5) Me 3 C 3 + PMe 3 T H F ; W - N O f \ ^ P M e 3 A M e3C Me + others 2.10 alkane occurs readily in 2.4, 2.7, 2.8, and certain neopentyl benzyl derivatives, this process is at best sluggish for 2.9-du and 2.10 under similar experimental conditions. While it can be argued that loss of tetramethylsilane from 2.9-</i2 or methane from 2.10 is highly unfavorable because it would involve the breaking of fairly strong M - C D 2 S i ( C D 3 ) 3 and M - C H 3 bonds,2 6 the reason why these complexes do not undergo facile loss of neopentane as observed for Cp*W(NO)(CH 2 CMe 3 ) (CH 2 Ar) is at present unclear. Finally, in the above section it was shown that thermolysis of 2.4 in CeD^ at 70 °C results in the formation of Cp*W(NO)(CHDCMe 3 ) (C 6 D 5 ) (2.11-rf6). To determine whether benzene is activated reversibly, the thermolysis of isotopically labeled Cp*W(NO)(CH 2CMe 3)(Ph-c/5) (2.11-rf5) was investigated. This compound was prepared as shown in eq 5.6 by taking advantage of the acidic nature of the agostic C - D bond. The thermolysis of 2.11-</5 was carried out in neat CeD 6 and monitored periodically by ' H N M R spectroscopy at 70 °C. Throughout the course of the thermolysis (~10 h), slow decomposition was evidenced by the appearance minor peaks in the Cp* region. Arene exchange and H/D exchange between the neopentyl and phenyl ligands, however, were not observed. This result demonstrates that 2.11-rf5 does not decompose by an a-hydrogen abstraction pathway, presumably because this would require the breaking of the strong M-C(aryl) bond. To gain additional insight into the thermal decomposition of 2.11, the thermolysis of 2.4 in CeD6 was carried out at 70-75 °C for 3 d. The organic volatiles were then analyzed by G C / M S , which indicated the presence of neopentane-dq, -d\, and <5?2 in approximately 81%, 11%, and 7% yields, respectively. The formation of neopentane-di and -di strongly suggests that Cp*W(NO)(CHDCMe 3)(Ph-<i5) (2.11-</6) decomposes by two competitive pathways, namely ring metalation and (3-hydrogen abstraction. 189 (5.6) 2.11-rf6 2.11-</5 5.4 Conclusions The thermal decompositions of the bis(neopentyl) complexes Cp'W(NO)(CH2CMe3)2 have been shown to proceed by a reversible, rate-determining formation of a neopentane a-complex, [Cp'W(NO)(=CHCMe3)(7]2-H-CH2CMe3)], in which the metal can coordinate interchangeably to the a- and y - C - H bonds of the neopentane molecule. Concurrently, neopentane dissociates irreversibly from the metal center to give the 16-electron intermediate, [Cp'W(NO)(=CHCMe 3)], which can coordinate to another hydrocarbon or to a Lewis base such as PMe 3 . Comparison of the relative rates and products of the thermal decompositions of 2.7, 2.8, 2.9-dn, 2.10 and 2.11-rfs with those of 2.4 and the benzyl neopentyl complexes Cp*W(NO)(CH 2 CMe3)(CH 2 Ar) demonstrates that (1) 2.4 and the benzyl neopentyl complexes are unstable with respect to neopentane loss by a-hydrogen abstraction, (2) 2.7 and 2.8 decompose at a rate comparable to that of 2.4 but give several products as a result of competitive a-hydrogen abstraction from CH 2 CMe3 and C H 2 C M e 2 E t and competitive a-hydrogen abstraction from CH 2 CMe3 and 5-hydrogen abstraction from C H 2 C M e 2 P h , and (3) 2.9-dn, 2.10, and 2.11-</5 are thermally robust; thermolysis of 2.10 and 2.11-</5 at extended times demonstrates that a-hydrogen abstraction resulting in loss of neopentane is a very minor reaction at best. 190 5.5 References and Notes (1) (a) Schrock, R. R. In Reactions of Coordinated Ligands; Braterman, P. S.; Ed.; Plenum: New York, 1986; Vol . 1, pp 221-223. (b) Feldman, J.; Schrock, R. R. Prog. Inorg. Chem. 1991, 39, 1. (c) LaPointe, A . M . ; Schrock, R. R.; Davis, W. M . J. Am. Chem. Soc. 1995, 777,4802. (2) Luecke, H . F.; Arndtsen, B. A. ; Burger, P.; Bergman, R. G. J. Am. Chem. Soc. 1996, 775,2517. (3) For examples, see: (a) Schrock, R. R. J. Am. Chem. Soc. 1975, 97, 6577. (b) Schrock, R. R.; Sharp, P. R. J. Am. Chem. Soc. 1978, 700, 2389. (c) Schrock, R. R.; Messerle, L. W.; Wood, C. D.; Guggenberger, L. J. J. Am. Chem. Soc. 1978, 700, 3793. (d) Kiel , W. A . ; Lin , G. -Y. ; Constable, A . G.; McCormick, F. B. ; Strouse, C. E.; Eisenstein, O.; Gladysz, J. A . J. Am. Chem. Soc. 1982, 104, 4865. (e) Kie l , W. A . ; Lin, G. -Y. ; Bodner, G. S.; Gladysz, J. A . J. Am. Chem. Soc. 1983, 705, 4958. (4) For examples, see: (a) Cooper, N . J.; Green, M . L . H . J. Chem. Soc, Dalton Trans. 1979, 1121. (b) Fellmann, J. D.; Turner, H . W.; Schrock, R. R. J. Am. Chem. Soc 1980, 702, 6608. (c) Fellmann, J. D.; Schrock, R. R.; Traficante, D. D. Organometallics 1982, 1, 481. (d) Turner, H . W.; Schrock, R. R.; Fellmann, J. D.; Holmes, S. J. J. Am. Chem. Soc. 1983, 705, 4942. (e) Parkin, G.; van Asselt, A . ; Leahy, D. J.; Whinnery, L. ; Hua, N . G.; Quan, R. W.; Henling, L. M . ; Schaefer, W. P.; Santarsiero, B . D.; Bercaw, J. E. Inorg. Chem. 1992, 31, 82. (f) Schrock, R. R.; Shih, K . - Y . ; Dobbs, D. A . ; Davis, W. M . J. Am. Chem. Soc. 1995, 777, 6609. (g) Schrock, R. R.; Seidel, S. W.; Mosch-Zanetti, N . C ; Shih, K . - Y . ; O'Donoghue, M . B. ; Davis, W. M . ; Reiff, W. M . J. Am. Chem. Soc. 1997, 779, 11876. (h) Schrock, R. R ; Seidel, S. W.; Mosch-Zanetti, N . C ; Dobbs, D. A. ; Shih, K . - Y . ; Davis, W. M . Organometallics 1997, 16, 5195. (5) For examples, see: (a) Caulton, K . G.; Chisholm, M . H. ; Streib, W. E.; Ziling, X . J. Am. Chem. Soc. 1991,113, 6082. (b) References l a and lb. (6) Legzdins, P.; Sayers, S. F. Chem. Eur. J. 1997, 3, 1579. 191 Sharp, P. R.; Astruc, D.; Schrock, R. R. J. Organomet. Chem. 1979,752, All. Tran, E.; Legzdins, P. J. Am. Chem. Soc. 1997,119, 5071. Jones, W. D.; Hessell, E. T.-V. Am. Chem. Soc. 1993,115, 554. (a) McNamara, B . K . ; Yeston, J. S.; Bergman, R. G.; Moore, C.B.J. Am. Chem. Soc. 1999,121, 6437. Brown, C. E.; Ishikawa, Y . ; Hackett, P. A. ; Rayner, D. M . J. Am. Chem. Soc. 1990, 112, 2530. (a) Chatt, J.; Davidson, J. M . J. Chem. Soc. 1965, 843. (b) Hodges, R. J.; Garnett, J. L . J. Phys. Chem. 1968, 72, 1969. (a) Graham, M . A . ; Perutz, R. N . ; Poliakoff, M . ; Turner, J. J. J. Organomet. Chem. 1972, 34, C34. (b) Perutz, R. N . ; Turner, J. J. J. Am. Chem. Soc. 1975, 97, 4791. (c) Hall, C.; Perutz, R. Chem. Rev. 1996, 96, 3125. (d) Sun, X . - Z . ; Grills, D. C ; Nikiforov, S. M . ; Poliakoff, M . ; George, M . W. J. Am. Chem. Soc. 1997, 119, 7521. (e) Evans, D. R.; Drovetskaya, T.; Bau, R.; Reed, C. A . ; Boyd, P. D. W. J. Am. Chem. Soc. 1997,119, 3633. (f) Geftakis, S.; Ball, G. E. J. Am. Chem. Soc. 1998, 720, 9953. (a) Buchanan, J. M . ; Stryker, J. M . ; Bergman, R. G. J. Am. Chem. Soc. 1986, 705, 1537. (b) Jones, W. D. ; Feher, F. J. J. Am. Chem. Soc. 1986, 705, 4814. (c) Periana, R. A. ; Bergman, R. G. J. Am. Chem. Soc. 1986, 705, 7332. (d) McGhee, W. D.; Bergman, R. G. J. Am. Chem. Soc. 1988, 770, 4226.. (e) Bullock, R. M . ; Headford, C. E. L . ; Hennessy, K . M . ; Kegley, S. E.; Norton, J. R. J. Am. Chem. Soc. 1989, 777, 3897. (f) Parkin, G.; Bercaw, J. E. Organometallics 1989, 5, 1172. (g) Gould, G. L. ; Heinekey, D. M . J. Am. Chem. Soc. 1989, 777, 5502. (h) Wang, C; Ziller, J. W.; Flood, T. C. J. Am. Chem. Soc. 1995, 777, 1647. (i) Stahl, S. S.; Labinger, J. A . ; Bercaw, J. E. J. Am. Chem. Soc. 1996, 775, 5961. (j) Schafer, D. F.; Wolczanski, P. T. J. Am. Chem. Soc. 1998, 720, 4881. (k) Gross, C. L. ; Girolami, G. S. J. Am. Chem. Soc. 1998, 720, 6605. 192 (1) Mobley, T. A . ; Schade, C ; Bergman, R. G. Organometallics 1998,77, 3574. (m) Wick, D. D.; Reynolds, K . A . ; Jones, W. D. J. Am. Chem. Soc. 1999, 121, 3974. (15) Jones, W. D.; Feher, F. J. J. Am. Chem. Soc. 1984,106, 1650. (16) This reaction was reported previously by former group member, K . J. Ross. The product, Cp*W(NO)(CHDCMe 3 ) (C 6 D 5 ) , was fully characterized by ' H and 2 H N M R spectroscopies, but no minor products were suggested. Ross, K . J. Ph.D. Dissertation, University of British Columbia, 1994. (17) Legzdins, P.; Veltheer, J. E.; Young, M . A . Organometallics 1995, 14, 407. (18) Wood, C. D; McLain, S. J.; Schrock, R. R. J. Am. Chem. Soc. 1979,101, 3210. (19) McDade, C ; Green, J. C ; Bercaw, J. E. Organometallics 1982,7 , 1629. (20) Bulls, A . R.; Schaefer, W. P.; Serfas, M . ; Bercaw, J. E. Organometallics 1987, 6, 1219. (21) (a) Cheon, J.; Rogers, D. M . ; Girolami, G. S. J. Am. Chem. Soc. 1997, 779, 6804. (b) Ajou, J. A . N . ; Rice, G. L. ; Scott, S. L. J. Am. Chem. Soc. 1998, 720, 13436. (22) Schrock, R. R.; Fellmann, J. D. J. Am. Chem. Soc. 1978, 700, 3359. (23) Brunet, N . ; Debad, J. D.; Legzdins, P.; Trotter, J.; Veltheer, J. E.; Yee, V . C. Organometallics 1993,12, 4572. (24) Ankianiec, B . C ; Hardy, D. T.; Thomson, S. K. ; Watkins, W. N . ; Young, G. B . Organometallics 1992, 77, 2591. (25) Hessen, B. ; Buijink, J. K . F.; Meetsa, A . ; Teuben, J. H . ; Helgesson, G.; Hakansson, M . ; Jagner, S.; Spek, A . L. Organometallics 1993, 72, 2268. (26) Bruno, J. W.; Marks, T. J.; Morss, L . R. J. Am. Chem. Soc. 1983, 705, 6824. 193 APPENDIX Table A l . Selected IR Spectral Data for Complexes 2.1-2.11" cmpd R R' Nujol KBr vNo V c o H v N o V C a H 2.1 CH 2 CMe 3 CH 2 CMe 3 1558 (s) 1555 (s) 2.2 CH 2CMe 2Ph CH 2CMe 2Ph 1554 (s) 2690, 2719 (br) 1560 (s) 2690, 2719 (br) 2.3 CH 2 SiMe 3 CH 2SiMe 3 1596 (s) c 2.4 CH 2 CMe 3 CH 2 CMe 3 1551 (s) 1551 (s) 2.5 CH 2CMe 2Ph CH 2CMe 2Ph 1547 (s) 2478 (sh w) 1551 (s) 2478 (sh w) 2.6 CH 2 SiMe 3 CH 2SiMe 3 1549 (s) c 2.7 CH 2 CMe 3 CH 2CMe 2Et 1576 (s) 1554 (s) 2.8 CH 2 CMe 3 CH 2CMe 2Ph 1574 (s) 1557 (s) 2.9 CH 2 CMe 3 CH 2SiMe 3 1553 (s) 1552 (s) 2.10 CH 2 CMe 3 C H 3 1587 (s) 1551 (s) 2.11 CH 2 CMe 3 Ph 1547 (s) 2529 (br w) 1551 (s) "v in cm"'. *Not observed. "Not determined. 0.7 0.6 -\ 0 20000 40000 60000 Time (sec) Figure A l . Plot of the kinetic data for the products of the thermolysis of 2.4 in neohexane in the presence of excess PMe 3 : Cp*W(NO)(=CHCMe 3)(PMe 3) (3.4) = • ; Cp*W(NO)(5v«-neohexene)(PMe3) (4.4 s y n) = • ; Cp*W(NO)(anri-neohexene)(PMe3) _ ( 4 . 4 a n t i ) = A . 80000 194 Since &4.4Sy„ = & 3 .4 , fcuanti can be calculated according to the equation: £ m a X 4 . 4 s y n = [ l / (^ 4 . 4 s y n -^ 4 . 4 a n t i)] ln(A: 4 . 4 s y n /^ 4 .4 a n t i) , where / m a X 4 . 4 S y n is the time at which [4.4syn] reaches its maximum value. Figure A2 . ' H - ' H E X S Y spectrum of Cp*W(NO)(CH2CMe3)(C6H3-2,5-Me2) (2.11') in toluene-Jg at -50 °C. G C - M S Analysis of Neopentane-rfx: Neopentane does not give a molecular ion on electron impact; the base peak is [M - methyl]+ (i.e., C4H9, mlz 57). Assuming that (a) neopentane-^ is solely (CH 3)3C(CD 3), neopentane-fik is solely (CH3)3C(CD 2H), etc., (b) there are no isotope effects for methyl- and hydrogen-losses, and (c) the highest mass peak in the 1 3 C corrected data corresponds to the amount of that isotope, the isotopic distribution in the neopentanes produced from thermolyses of 2.1-*/4, 2.4-rf4, and 2.11-^6 can be calculated by subtracting the appropriate theoretical amount from all the lower mass peak totals; three such operations would give the amounts of neopentane-^, -d2, and -d\. The amount remaining in the mlz 57 column is due to neopentane-fifo; however, this must be normalized by multiplying by 0.75 since an mlz 57 fragment is generated 4 times out of 4 but neopentane-Ji, -d2, or -d3 gives mlz 58, 59, and 60 195 fragments, respectively, only 3 times out of 4. See, Schrock, R. R.; Fellmann, J. D. J. Am. Chem. Soc. 1978,100, 3359. Table A 2 . terr-Butyl Ions in the Mass Spectra of Neopentanes mlz" 57 58 59 60 2.1-d4/THF/PMe3 45.07 25.96 8.62 100.00 2.4-rf4/THF/PMe3 37.73 28.52 7.74 100.00 2.4-</4/PMe3 32.63 14.91 5.85 100.00 2.11-rf6/C6D6 100.00 -9.60 -6.58 0 Neopentane-^o* 100.00 0.13 Neopentane-(3?i* 35.07 100.00 Neopentane-^6 34.66 3.29 100.00 Neopentane-of3* 32.66 3.30 2.71 100.00 "All corrected for 1 3 C . 'Calculated spectra. Schrock, R. R.; Fellmann, J. D. J. Am. Chem. Soc. 1978,100, 3359. Figure A 3 . l H N M R spectra of the methylene regions of (A) Cp*W(NO)(CH 2CMe3)[CH2Si(CH3)(CD 3)2] (2.9-rf6-H) and Cp*W(NO)(CHDCMe3)[CD2Si(CH3)2(CD3)] (2.9-</3-D) [obtained from thermolysis of 2.4 in (CH 3)2Si(CD 3) 2, kH/kD = 1.35(5)], and (B) m- and /^-isomers of Cp*W(NO)(CH2CMe3)(C6Fl4Me) and Cp*W(NO)(CHDCMe 3 ) (C 6 D 4 Me) [obtained from thermolysis of 2.4 in 1:1 toluene/toluene-fig, kn/ku = 1.04(5), unpublished]. ON p. o SC c i 1 .3 2 g o .5 S £ < r- oo ^ . m ( N O — — 00 S O O s O c o r i O SC c j -3-S O Tl-o X JD s o _c t-l o J3 t C N I N so" 2 Q 55 o so O N O N ca o z sc u O .a ^ 2 C — w S =tt f-u-> so O N OO ° ) co" o o C N — OO ~ D u C J u < 3 m a oo O N GO o 2 0 0 r-m o T 3 O cd P-, O d r> O N •<» i rr r i 00 u X a o U ON r i a o S --H» s at r i 00 O SC C J o :z sc o C J fc o d d E o O e o o c o s C N eu ON O s — r o OO f N r o i n i n r o i n C N OO i n C N i n o C N 5 S O 0 0 _c c^ so o < Q o o C J ru cd — 00 OH _ § " '> cd cj - S Q —i o ON r o r o o vo — O c o d — i cd Q o IS 13 -4-» & U — « "*3 i r i T3 O I o CJ o Q SC o C J 'S. E w (U o s x o O C O d E .s I f iS *3 <u o o ir\ — -~ C N o f 00 E ea N 8 o H < D o o C J E S CJ P3 s o. E CJ o J3 cr c S3 _3 "S CJ s s 00 so T f o d ^ 7( -I -i ^ S o o o o z (-1 o o x o o X f ^ • 0 0 r^T (N >n m OO 0\ 0 0 ON O od 1 E Q iz> U • a o J 3 O o _ H ri O Z r> K O _o o J3 X • a 2 SO OO CO O l 0 0 r- r-'2,/c 0.50 4.65 4.16 09.9 in © PH —< <—1 '—1 — — PH < s E Q t o O T3 O -H ^- f^ l O Z o u CN CN O O —c CN ^ «n m. ^ oo oo 0 0 •3-PH H 0 0 E Q • a o J 3 O o cn - H - H i n i—I ci o O — m zS O o 1 o •s o X ! e o ^? in w 0 0 o in cn as cn VO 0 0 o cn VO V3 Q u u u CO oo 2 o 0 0 ~ H - H i n • a o X i O O PH o z "3-X CN g X "i = 2 O o O o — 0 0 Os ~I CN Os PH H P i s CO E O o U T3 C o I o U a H2 •c 'I w •S o "o u o. a cd 60 > N S O 5 P i c P i P i — o 2 M Z ~n V '-3 _ iS _3 — c H - . o c • a o o O as Vi PH o z so C J o io T f d 1 o O m 10 in CN T f T f « Q o u u oo 60 2 o c o oo mc o c © o o c o CN atter in r o o r--r o CU o o oo cj fc o X , s i n o CN CN T f r o i n OO _c o CN Os "o r o i n OO i n o c T f o r o oo o c O N oo O N i n » n CN mo CN T f r o T f i n O mo OO — O N a o o u o o Q < cd 2 O — ~ oo in — — S c o LH OH tTi -a 03 •TT o z cj o (N O CO 2 f CN O N G\ o <n T f r- CN \o oo T f oo i n CO CO r-—. O N T f ^ ^ 8 E t3 <D C3 o o o —'• Os c oo o •TT o cj 5 p, o z_ X Os u •c a . o T f T f d O 3 T f CN PH cn —• £ K °. so T f o C\ — — — § 5 Q O C J C J o o - X 2 • V T f CO CN O N i n *n CN CO CN — — T f s eS P . e PH m O O (N O ~ T f 0 0 o d O U u «s H-* Q o HS H-» u — C3 H T3 O 6 o U P, o Ol o o -2 " o d o s u u C J CH 3 CN CN i n T f °< •<! •< > N E 1 3 P So u o as s W O T f r o r~- — r-~ r- »n ^ O N r o P i c T3 O ^ = -1 -s 0 0 o d P i 2" P i H - • o c •a o o O 199 Table A6. Fractional coordinates and equivalent isotropic displacement parameters for Cp*W(NO)(CD 2 CMe 3 ) 2 (2.4-</4) x y z U(eq) W( l ) 7797(1) 1370(1) 6588(1) 27(1) 1 N( l ) 5765(10) 1327(4) 6170(7) 38(2) 1 0(1) 4356(8) 1233(4) 5706(7) 53(2) 1 C( l ) 8040(12) 2318(5) 6918(9) 42(3) 1 C(2) 6942(11) 2866(4) 6582(8) 38(3) 0.54(3) C(3) 5199(17) 2690(10) 6230(3) 62(8) 0.54(3) C(4) 7240(3) 3191(11) 5466 (18) 58(8) 0.54(3) C(5) 7180(3) 3318(11) 7654(19) 75 (9) 0.54(3) C(2') 6942(11) 2866(4) 6582(8) 38(3) 0.46(3) C(3') 5500(3) 2765(12) 7010(3) 60(10) 0.46(3) C(4') 6520(4) 2984(14) 5205(14) 66(10) 0.46(3) C(5') 7810(3) 3438(9) 7250(2) 54(9) 0.46(3) C(6) 8258(11) 931(5) 8365(8) 30(2) 1 C(7) 7105(10) 768(4) 9041(9) 41(3) 1 C(8) 6009(12) 260(5) 8408(10) 56(3) 1 C(9) 7968(12) 543(5) 10351(9) 54(3) 1 C(10) 6131(12) 1330(5) 9222(10) 48(3) 1 C ( l l ) 7958(12) 767(5) 4948(9) 39(3) 1 C(12) 8492(11) 1369(4) 4736(9) 28(2) 1 C(13) 9893(11) 1477(4) 5627(9) 32(2) 1 C(14) 10245(10) 950(5) 6397(8) 29(2) 1 C(15) 9056(11) 511(4) 5971(9) 31(2) 1 C(16) 6550(12) 434(6) 4144(11) 58(3) 1 C(17) 7716(13) 1747(6) 3628(10) 62(4) 1 C(18) 10957(12) 2018(5) 5698(11) 52(3) 1 C(19) 11682(11) 867(6) 7436(9) 49(3) 1 C(20) 9006(13) 138(5) 6435(11) 52(3) 1 Table A7. Fractional coordinates and equivalent isotropic displacement parameters for Cp*W(NO)(CH 2 CMe 3 )2 (2.4) X y z U(cq) U(1) 7802(3) 1371(1) 6593(2) 22(1) mn 5730(2) 1318(1) 6163(1) 28(1) c m 4337(3) 1225(1) 5703(3) 39(1) CCD 8651(3) 2325(1) 6930(2) 34(1) cm 6929(3) 2871(1) 6591(2) 33(1) C(3) 5143(10) 2680(5) 6266(11) 54(2) C(4) 7140(12) 3174(4) 5433(10) 51(2) C(5) 7195(15) 3325(4) 764BC9) 72(3) ca*) 5544(14) 2776(5) 6962(17) 5fi(3) « 4 ' ) 6722(32) 2963(12) 5232(21) 84(4) ccS") 7794(19) 3448(5) 7242(1?) 69(3) cm &302(3) 926(1) 2377(2) 26(1) tm 7083(2) 768(1) 9076(2) 2«i(1) « 8 ) 6010(3) 259(1) 8408(2) 43(1) tm 7960(3) 541(1) 10371(2) 34(1) C{10) 6110(3) 1328(1) 9223(2) 40(1) €(11) 7950(3) 764(1) 4936(2) 28(1) C<12) 8481(3) 1364(1) 4723(2) 26(1) €(13) 9924(2) 1478(1) 5634(2) 26(1) C(14>. 10274(2) 955(1) 6411(2) 26(1) C<1S) 9074(3) 509(1) 5970(2) 26(1) CC16) 6565(3) 440(1) 4126(2) 41(1) C{17) 7708(4) 1755(1) 3637(2) 40(1) C(18) 10975(3) 2021 CD 5706(3) 38(1) C(19> 11727(3) 878(1) 7445(2) 36(1) CC2Q) 9025(3) •135(1) 6431(3) 37(1) Table A8. Fractional coordinates and equivalent isotropic displacement parameters for Cp*W(NO)(CH 2CMe3)(CH 2SiMe3) (2.9) X y z U(eq) W( l ) 1576(1) 7500 4928(1) 28(1) 1 N( l ) 3324(4) 7500 5428(3) 34(1) 1 O( l ) 4412(5) 7500 5958(4) 58(1) 1 C( l ) 1687(19) 6652(5) 3795(7) 39(2) 0.50 Si(l) 3091(6) 5998(3) 3944(4) 45(1) 0.50 C(2) 4825(22) 6418(16) 3888(20) 76(4) 0.50 C(3) 2936(27) 5389(12) 2800(16) 73(3) 0.50 C(4) 3024(21) 5459(5) 5199(12) 62(2) 0.50 C ( r ) 1815(19) 6545(5) 4037(7) 39(2) 0.50 C(l") 3032(20) 6003(10) 3939(14) 45(1) 0.50 C(2') 4479(24) 6350(16) 3691(21) 76(4) 0.50 C(3') 2722(28) 5440(13) 3057(18) 73(3) 0.50 C(4) 3069(21) 5653(5) 5086(13) 62(2) 0.50 C ( l l ) 586(4) 7128(2) 6543(2) 36(1) 1 C(12) -394(4) 6897(2) 5737(3) 43(1) 1 C(13) -980(5) 7500 5230(5) 49(1) 1 C(14) 1376(6) 6679(3) 7333(3) 60(1) 1 C(15) -855(7) 6157(3) 5544(6) 74(2) 1 C(16) -2092(7) 7500 4353(8) 91(4) 1 202 Table A9 . Fractional coordinates and equivalent isotropic displacement parameters for Cp*W(NO)(CH2CMe3)(C6H3-2,5-Me2) (2.11') atom X y z Be, W(l) -0.29851(3) 0.11513(4) 0.20928(2) 1.509 W(2) 0.22121(3) 0.11527(4) 0.28286(2) 1.597 0(1) -0.2504(6) -0.0809(6) 0.2474(4) 3.7(2) 0(2) 0.2387(6) -0.0787(6) 0.2391(4) 3.6(2) N(l) -0.2610(6) 0.0018(7) 0.2349(5) 2.2(2) N(2) 0.2400(7) 0.0040(7) 0.2547(5) 3.3(3) C(l) -0.4252(6) 0.1339(8) 0.1147(5) 1.5(2) C(2) -0.4421(8) 0.0634(9) 0.1532(6) 2.6(3) C(3) -0.4499(7) 0.1041(9) 0.2176(6) 2.2(3) C(4) -0.4329(8) 0.2020(8) 0.2112(6) 2.3(3) C(5) -0.4207(7) 0.2198(8) 0.1484(5) 1.5(2) C(6) -0.4261(8) 0.1265(10) 0.0443(6) 3.5(3) C(7) -0.4623(9) -0.0347(9) 0.1353(7) 3.2(3) C(8) -0.4751(9) 0.0568(10) 0.2756(7) 3.8(4) C(9) -0.4405(8) 0.2728(8) 0.2654(6) 2.3(3) C(10) -0.4076(8) 0.3136(9) 0.1210(6) 2.8(3) C(ll) -0.2243(7) 0.1379(9) 0.1367(6) 2.0(3) C(12) -0.1989(8) 0.0739(9) 0.0923(6) 2.3(3) C(13) -0.1480(8) 0.1054(10) 0.0500(6) 2.8(3) C(14) -0.1226(8) 0.1969(11) 0.0479(6) 3.0(3 C(15) -0.1468(8) 0.2610(9) 0.0900(6) 2.6(3 C(16) -0.1978(7) 0.2296(9) 0.1326(5) 1.9(3 C(17) -0.2237(10) -0.0279(9) 0.0909(7) 3.6(4 C(18) -0.1202(8) 0.3624(10) 0.0922(6) 3.5(3 Table A10. Fractional coordinates and equivalent isotropic displacement parameters for Cp*W(NO)(CH 2 CMe 3 )(NMe 2 ) (2.17) X y z U ( e q ) SC W ( l ) 7668 (1) 9528 (1) 8843 (1) 18 (1) 1 0 ( 1 ) 4412 (5) 8430 (4) 9183 (2) 40 (1) 1 N ( l ) 5741 (5) 8789 (4) 8982 (2) 25 (1) 1 N ( 2 ) 8873 (6) 8009 (4) 8613 (2) 26 (1) 1 C ( l ) 9528 (6) 10124 (5) 9622 (2) 23 (1) 1 C ( 2 ) 9811 (6) 11026 (5) 9157 (2) 23 (1) 1 C ( 3 ) 8377 (7) 11755 (5) 9084 (2) 26 (1) 1 C ( 4 ) 7168 (6) 11317 (5) 9490 (2) 25 (1) 1 C ( 5 ) 7873 (6) 10266 (5) 9819 (2) 23 (1) 1 C ( 6 ) 10773 (8) 9236 (7) 9904 (3) 35 (2) 1 C ( 7 ) 11436 (7) 11300 (7) 8861 (4) 38 (1) 1 C ( 8 ) 8253 (10) 12924 (6) 8685 (3) 40 (2) 1 C ( 9 ) 5476 (9) 11832 (8) 9575 (4) 45 (2) 1 C ( 1 0 ) 7077 (9) 9571 (8) 10330 (3) 44 (2) 1 C ( l l ) 7085 (6) 10423 (5) 7980 (2) 23 (1) 1 C ( 1 2 ) 6090 (6) 9741 (5) 7489 (2) 23 (1) 1 C ( 1 3 ) 6139 (8) 10551 (8) 6914 (3) 37 (1) 1 C ( 1 4 ) 4295 (7) 9597 (7) 7678 (3) 32 (1) 1 C ( 1 5 ) 6787 (9) 8407 (7) 7339 (3) 36 (2) 1 C ( 1 6 ) 10590 (7) 8082 (7) 8421 (3) 38 (2) 1 C ( 1 7 ) 8338 (11) 6680 (7) 8626 (4) 40 (2) 1 204 Table A l l . Fractional coordinates and equivalent isotropic displacement parameters for CpW(NO)(=CHCMe 3)(PMe 3) (3.1) X y z U(eq) w o ) 0.56731(7) 0.35417(4) 0.27778(5) 3.99(1) P(i) 0.7964(4) 0.3512(3) 0.3905(3) 4.36(9) 0(1) 0.639(1) 0.1799(7) 0.1519(9) 6.0(3) N( l ) 0.611(1) 0.2494(9) 0.2078(9) 40(3) C( l ) 0.572(4) 0.481(2) 0.152(2) 88(8) C(2) 0.432(4) 0.452(1) 0.151(2) . 8.7(7) C(3) 0.385(3) 0.475(2) 0.246(3) 90(8) C(4) 0.491(3) 0.518(1) 0.307(2) 69(6) C(5) 0.607(3) 0.524(1) 0.252(3) 7.5(7) C(6) 0.469(2) 0.2983(9) 0.388(1) 47(4) C(7) 0.419(2) 0.199(1) 0.424(1) 5.9(5) C(8) 0.467(3) 0.194(1) 0.543(2) 11.5(9) C(9) 0.479(2) 0.115(1) 0.366(2) 71(6) C(10) 0.259(2) 0.198(1) 0.404(2) 10.0(7) C ( l l ) 0.804(2) 0.416(1) 0.516(1) 93(6) C(12) 0.945(2) 0.401(1) 0.334(2) 81(6) C(13) 0.861(2) 0.234(1) 0.433(2) 79(6) 205 Table A12. Fractional coordinates and equivalent isotropic displacement parameters for Cp*W(NO)(=CHCMe 3)(PMe 3) (3.4) X y z U(eq) W( l ) 2584(1) 7503(2) 8145(1) 32(1) N( l ) 1248(13) 6735(14) 8898(13) 85(6) O( l ) 95(15) 6260(12) 9063(16) 107(5) C( l ) 1642(9) 7463(13) 5468(9) 49(3) C(2) 2515(16) 8288(11) 5859(21) 46(4) C(3) 4155(13) 8009(11) 6336(18) 40(4) C(4) 4177(13) 6978(11) 6218(21) 39(4) C(5) 2644(17) 6655(12) 5704(21) 49(5) C(6) -104(11) 7399(26) 4811(11) 99(8) C(7) 2017(30) 9308(12) 5497(18) 110(12) C(8) 5570(21) 8632(18) 6594(16) 98(12) C(9) 5657(23) 6403(14) 6610(15) 79(9) C(10) 2138(32) 5614(14) 5492(25) 133(15) P(l) 1449(4) 8963(3) 9327(4) 45(1) C ( l l ) 2630(18) 10022(12) 9594(20) 74(5) C(12) -462(19) 9389(16) 8346(19) 86(6) C(13) 913(17) 8655(13) 11157(14) 68(5) C(14) 4388(11) 7389(17) 9777(10) 28(4) C(15) 4914(15) 7008(10) 11253(14) 50(3) C(16) 3621(18) 6467(11) 11840(16) 62(4) C(17) 6271(18) 6339(14) 11142(18) 79(5) C(18) 5533(16) 7784(12)., - 12495(13) 67(6) 206 Table A13. Fractional coordinates and equivalent isotropic displacement parameters for Cp*W(NO)[CH(CMe3)CH2(CMe3)CH] (3.14) X y z U(eq) WO) 0.201815(11) 0.148487(12) 0.182473(12) 1.283(6) 0(1) 0.2461(3) 0.3412(2) 0.2014(3) 3.27(11) N( l ) 0.0837(3) 0.2588(3) 0.1928(3) 2.02(10) C ( l ) 0.0788(3) 0.0797(4) 0.2345(4) 2.74(14) C(2) 0.0780(3) 0.1790(4) 0.2379(4) 2.92(14) C(3) 0.0832(3) 0.2117(4) 0.1618(4) 3.10(15) C(4) 0.0874(3) 0.1357(4) 0.1107(3) 2.15(12) C(5) 0.0793(4) 0.0537(3) 0.1564(4) 2.41(12) C(6) 0.0712(5) 0.0139(7) 0.2979(5) 6.8(3) C(7) 0.0668(4) 0.2380(7) 0.3064(5) 7.6(3) C(8) 0.0762(5) 0.1856(5) 0.1388(7) 6.9(2) C(9) 0.0890(4) 0.1396(6) 0.0265(5) 5.8(2) C(10) 0.3308(4) 0.0436(4) 0.1247(5) 5.2(2) C(ll) 0.2730(3) 0.1227(4) 0.0424(3) 2.17(12) C(12) 0.3142(4) 0.0854(3) 0.1007(3) 1.44(10) C(13) 0.2675(3) 0.0511(4) 0.1798(3) 2.36(13) C(14) 0.3207(3) 0.0707(3) 0.2590(3) 1.32(10) C(15) 0.3919(4) 0.0980(4) 0.3252(3) 1.93(12) C(16) 0.3734(4) 0.0470(4) 0.0232(3) 3.12(14) C(17) 0.2835(4) 0.2105(4) 0.0687(4) 3.29(15) C(18) 0.2686(4) 0.1458(4) -0.0294(4) 3.1(2) C(19) 0.3758(3) 0.1143(4) 0.3947(4) 2.83(14) C(20) 0.3701(4) 0.0158(4) 0.3428(4) 2.90(14). C(21) 0.3701(4) 0.1856(5) 0.3105(3) 2.70(14) 207 Table A14. Fractional coordinates and equivalent isotropic displacement parameters for Cp*W(NO)(Ci 0 H 18) (3.15) X y z U(eq) SOF W ( l ) 5 1 5 0 ( 1 ) 2 7 6 0 ( 1 ) 2 0 6 5 ( 1 ) 1 7 ( 1 ) l N ( l ) 6 7 8 8 ( 5 ) 3 6 1 8 ( 4 ) 1 8 5 4 ( 3 ) 2 1 ( 1 ) l 0 ( 1 ) 8 0 0 3 ( 4 ) 4 0 6 9 ( 5 ) 1 5 6 9 ( 3 ) 3 5 ( 1 ) l C ( l ) 3 2 1 0 ( 6 ) 4 9 7 7 ( 6 ) 2 4 6 7 ( 3 ) 2 4 ( 1 ) l C ( 2 ) 1 2 5 6 ( 6 ) 5 4 2 0 ( 6 ) 2 5 4 7 ( 4 ) 3 3 ( 1 ) l C ( 3 ) 7 9 9 ( 6 ) 5 4 7 4 ( 6 ) 3 5 8 1 ( 4 ) 3 3 ( 1 ) l C ( 4 ) 2 2 2 5 ( 6 ) 3 9 0 5 ( 6 ) 3 9 7 2 ( 4 ) 3 1 ( 1 ) i C ( 5 ) 3 9 1 4 ( 6 ) 3 9 0 4 ( 6 ) 3 4 9 7 ( 3 ) 2 0 ( 1 ) l C ( 6 ) 5 4 1 6 ( 5 ) 2 1 0 9 ( 5 ) 3 4 9 0 ( 3 ) 1 8 ( 1 ) l C ( 7 ) 6 9 3 5 ( 6 ) 1 8 0 1 ( 6 ) 4 1 1 8 ( 3 ) 2 4 ( 1 ) l C ( 8 ) 6 2 8 7 ( 7 ) 1 8 5 7 ( 7 ) 5 1 3 1 ( 3 ) 3 4 ( 1 ) l C ( 9 ) 7 8 3 8 ( 6 ) 3 0 8 6 ( 6 ) 3 9 1 4 ( 3 ) 2 9 ( 1 ) l C ( 1 0 ) 8 2 8 4 ( 7 ) 1 2 ( 6 ) 3 9 6 3 ( 4 ) 3 9 ( 1 ) l C ( l l ) 6 4 3 3 ( 6 ) 4 8 9 ( 6 ) 1 1 0 4 ( 3 ) 2 2 ( 1 ) l C ( 1 2 ) 5 3 0 1 ( 6 ) 1 8 9 1 ( 6 ) 5 6 5 ( 3 ) 2 3 ( 1 ) i C ( 1 3 ) 3 5 6 5 ( 6 ) 2 1 9 3 ( 6 ) 9 0 0 ( 3 ) 2 2 ( 1 ) l C ( 1 4 ) 3 6 2 2 ( 6 ) 9 6 8 ( 5 ) 1 6 6 0 ( 3 ) 2 0 ( 1 ) l C ( 1 5 ) 5 3 9 3 ( 6 ) - 9 1 ( 5 ) 1 7 7 5 ( 3 ) 2 2 ( 1 ) l C ( 1 6 ) 8 3 6 8 ( 6 ) - 2 4 6 ( 7 ) 9 4 3 ( 4 ) 3 4 ( 1 ) l C ( 1 7 ) 5 8 5 3 ( 7 ) 2 7 7 4 ( 6 ) - 2 7 6 ( 3 ) 3 0 ( 1 ) l C ( 1 8 ) 1 9 6 7 ( 6 ) 3 4 9 4 ( 6 ) 4 7 3 ( 4 ) 2 9 ( 1 ) 1 C ( 1 9 ) 2 1 0 0 ( 6 ) 6 9 8 ( 6 ) 2 1 8 0 ( 3 ) 2 8 ( 1 ) 1 C ( 2 0 ) 6 0 2 0 ( 6 ) - 1 6 6 3 ( 6 ) 2 4 3 6 ( 3 ) 2 8 ( 1 ) 1 Table A15. Fractional coordinates and equivalent isotropic displacement parameters for Cp*W(NO)(C 1 2 H 1 8 ) (3.17) X y z U ( e g ) SOF W ( l ) 7 3 3 6 ( 1 ) 2 1 8 5 ( 1 ) 8 8 3 4 ( 1 ) 21 (1) 1 0 ( 1 ) 6175 (3) 1602 (2) 6690 (2) 44 (1) 1 N ( l ) 6565 (3) 1919 (2) 7544 (2) 28 (1) 1 C ( l ) 7862 (4) 663 (2) 9325 (3) 37 (1) 1 C ( 2 ) 8 9 7 6 ( 4 ) 1047 (3) 9110 (3) 38 (1) 1 C ( 3 ) 9574 (3) 1725 (2) 9852 (3) 32 (1) 1 C ( 4 ) 8844 (3) 1739 (2) 10532 (2) 28 (1) 1 C ( 5 ) 7777 (4) 1093 (2) 10195 (3) 32 (1) 1 C ( 6 ) 6984 (8) - 1 0 1 (3) 8752 (4) 63 (2) 1 C ( 7 ) 9498 (6) 740 (4) 8299 (4) 60 (1) 1 C ( 8 ) 10814 (4) 2282 (4) 9930 (4) 49 (1) 1 C ( 9 ) 9255 (4) 2219 (3) 11532 (3) 36 (1) 1 C ( 1 0 ) 6843 (4) 849 (3) 10756 (3) 42 (1) 1 C ( l l ) 5684 (3) 2762 (2) 9114 (3) 25 (1) 1 C ( 1 2 ) 6 2 7 9 ( 3 ) 3628 (2) 8745 (3) 27 (1) 1 C ( 1 3 ) 6628 (4) 4425 (2) 9517 (3) 35 (1) 1 C ( 1 4 ) 6852 (5) 5261 (3) 8972 (3) 45 (1) 1 C ( 1 5 ) 8009 (5) 5163 (3) 8848 (3) 47 (1) 1 C ( 1 6 ) 8620 (4) 4261.(2) 9288 (3) 36 (1) 1 C {17 ) 7762 (3) 3522 (2) 8564 (2) 26 (1) 1 C ( 1 8 ) 8096 (5) 4 2 1 1 ( 3 ) 10173 (3) 43 (1) 1 C ( 1 9 ) 4145 (4) 2627 (3) 8610 (3) 33 (1) 1 C (20) 3767 (5) 1756 (3) 9042 (4) 53 (1) 1 C ( 2 1 ) 3695 (4) 2530 (4) 7476 (3) 43 (1) 1 C (22) 3390 (4) 3428 (3) 8871 (4) 46 (1) 1 Table A16. Fractional coordinates and equivalent isotropic displacement parameters for Cp*W(NO)(n 3 -C 1 0 H 1 8 ) (3.21a) x y z U ( e q ) SOF W ( l ) 1213 (1) 4078 (1) 2 9 3 0 ( 1 ) 25 (1) 1 0 ( 1 ) 2190 (7) 5741 (3) 4015 (4) 54 (2) 1 N ( l ) 1920 (6) 5040 (3) 3 5 7 9 ( 4 ) 35 (1) 1 C ( l ) - 1 0 4 4 (7) 4724 (4) 2 1 4 0 ( 4 ) 28 (1) 1 C ( 2 ) - 1 2 7 5 (7) 4489 (4) 3 0 3 8 ( 4 ) 29 (1) 1 C ( 3 ) - 1 2 4 8 (7) 3532 (4) 3096 (4) 29 (1) 1 C ( 4 ) - 1 0 4 0 (7) 3183 (4) 2220 (4) 30 (2) 1 C ( 5 ) - 9 5 5 (7) 3923 (4) 1624 (4) 28 (1) 1 C ( 6 ) - 1 1 3 6 (9) 5667 (4) 1753 (6) 43 (2) 1 C ( 7 ) - 1 6 0 3 (9) 5120 (5) 3775 (5) 42 (2) 1 C ( 8 ) - 1 6 2 3 (9) 2976 (4) 3883 (5) 40 (2) 1 C ( 9 ) - 1 0 8 8 ( 9 ) 2207 (5) 1 9 5 4 ( 5 ) 46 (2) 1 C ( 1 0 ) - 1 0 6 9 (9) 3884 (5) 5 8 4 ( 5 ) 43 (2) 1 C ( l l ) 2527 (8) 4501 (4) 1 7 9 6 ( 5 ) 41 (2) 1 C ( 1 2 ) 3577 (9) 3751 (5) 1 5 0 0 ( 5 ) 44 (2) 1 C ( 1 3 ) 4469 (9) 3302 (6) 2393 (5) 52 (2) 1 C ( 1 4 ) 3298 (7) 3069 (4) 2958 (5) 36 (2) 1 C ( 1 5 ) . 3299 (8) 3313 (4) 3875 (5) 34 (2) 1 C ( 1 6 ) 1879 (8) 3156 (4) 4149 (4) 36 (2) 1 C ( 1 7 ) 3351 (12) 5354 (6) 2009 (8) 71 (3) 1 C ( 1 8 ) 4657 (11) 4081 (6) 892 (7) 65 (3) 1 C ( 1 9 ) 2555 (9) 3039 (5) 9 0 5 ( 5 ) 51 (2) 1 C (20) 4547 (9) 3837 (5) 4 4 7 1 ( 5 ) 44 (2) 1 210 Table A17. Fractional coordinates and equivalent isotropic displacement parameters for Cp*W(NO)(cyclohexene)(PMe3) (4.2) X y z U(eq) W ( l ) 1305(1) 3312(1) 634(1) 22(1) P(l) 1754(1) 4776(3) -105(1) 34(1) N( l ) 1605(2) 1560(8) 507(4) 30(2) 0(1) 1840(2) 321(7) 518(4) 52(2) C( l ) 944(3) 3315(10) 1706(5) 30(2) C(2) 1402(3) 2782(9) 2051(5) 28(2) C(3) 1709(3) 4070(9) 2029(5) 22(2) C(4) 1422(3) 5430(8) 1675(5) 23(2) C(5) 954(3) 4960(9) 1459(5) 25(2) C(6) 521(4) 2404(14) 1735(7) 57(3) C(7) 1551(4) 1108(10) 2420(7) 53(3) C(8) 2216(3) 4030(12) 2422(6) 42(2) C(9) 1594(4) 7127(9) 1653(6) 38(2) C(10) 552(4) 6033(12) 1110(7) 47(3) C ( l l ) 2329(4) 5472(14) 521(7) 57(3) C(12) 1914(4) 3513(11) -873(6) 48(3) C(13) 1528(5) 6612(12) 682(7) 62(3) C(14) 752(3) 3687(11) -608(5) 4e(3) C(15) 619(3) 2463(11) 114(6) 38(2) C(16) 555(4) 687(12) -410(7) 54(3) C(17) 445(4) 556(14) 1357(6) 62(3) C(18) 791(4) 1521(11) 1653(6) 50(3) C(19) 736(3) 3269(11) 1523(6) 43(2) Table A18. Fractional coordinates and equivalent isotropic displacement parameters for Cp*W(NO)(neohexene)(PMe 3)(4.4 s y n) atom X y z W(l) 0.18313(2) 0.147497(14) 0.24211(2) 2.452(5) P(l) 0.3870(2) 0.19188(11) 0.38425(13) 3.63(4) 0(1) 0.2053(7) 0.3122(3) 0.1733(4) 6.8(2) N(l) 0.2076(6) 0.2432(3) 0.1996(4) 3.75(13) C(l) 0.0292(7) 0.0576(4) 0.3079(5) 3.42(15) C(2) 0.0029(7) 0.1376(4) 0.3322(5) 3.51(15) C(3) -0.0671(7) 0.1794(4) 0.2471(6) 3.9(2) C(4) -0.0810(7) 0.1260(4) 0.1676(5) 3.50(15) C(5) -0.0275(6) 0.0515(4) 0.2079(4) 3.19(14) C(6) 0.0770(8) -0.0099(5) 0.3787(5) 5.2(2) C(7) 0.0132(10) 0.1629(5) 0.4337(6) 6.3(3) C(8) -0.1239(9) 0.2633(5) 0.2383(8) 7.2(3) C(9) -0.1563(9) 0.1433(5) 0.0637(6) 6.6(2) C(10) -0.0486(8) -0.0251(4) 0.1516(6) 5.5(2) C(ll) 0.4646(10) 0.1301(5) 0.4907(6) 6.5(2) C(12) 0.5578(8) 0.2303(5) 0.3574(6) 5.4(2) C(13) 0.3285(10) 0.2789(5) 0.4380(7) 6.7(3) C(14) 0.2350(8) 0.0898(4) 0.1214(4) 3.9(2) C(15) 0.3818(7) 0.0925(4) 0.1961(5) 3.5(2) C(16) 0.4711(7) 0.0154(4) 0.2253(5) 4.1(2) C(17) 0.3830(9) -0.0475(5) 0.2626(6) 5.4(2) C(18) 0.5161(10) -0.0167(5) 0.1386(7) 6.9(3) C(19) 0.6242(9) 0.0281(5) 0.3075(7) 6.6(2) 212 Table A19. Fractional coordinates and equivalent isotropic displacement parameters for Cp*W(NO)(cyclohexyl)(NMe 2) (4.7) x y z U(eq) w 0.30124(2) 0.16682(4) 0.08101(2) 0.0425 0 0.3916(6) -0.1617(8) 0.1310(6) 0.087 N(l) 0.3568(5) -0.0263(8) 0.1036(5) 0.057 N(2) 0.3937(5) ' 0.2761(8) 0.0498(5) 0.049 C(l) 0.2168(6) 0.3906(10) 0.1164(6) 0.048 C(2) 0.3034(5) 0.3620(9) 0.1842(5) 0.047 C(3) 0.3019(6) 0.1985(10) 0.2135(5) 0.053 C(4) 0.2119(6) 0.1305(10) 0.1634(6) 0.053 C(5) 0.1600(6) 0.2438(11) 0.1051(6) 0.051 C(6) 0.3882(7) 0.4521(10) 0.0270(6) 0.062 C(7) 0.4810(7) 0.2108(12) 0.0484(7) 0.070 C(l l ) 0.1816(7) 0.5532(11) 0.0723(7) 0.066 C(12) 0.3816(7) 0.4849(13) 0.2255(6) 0.071 C(13) 0.3795(8) 0.1213(15) 0.2892(7) 0.086 C(14) 0.1779(8) -0.0392(12) 0.1787(8) 0.077 C(15) 0.0583(6) 0.2382(14) 0.0472(8) 0.075 C(21) 0.1982(7) 0.0864(15) -0.0386(6) 0.070 C(22) 0.1664(9) 0.2168(22) -0.1014(8) 0.105 C(23) 0.0886(11) 0.1550(28) -0.1863(10) 0.132 C(24) 0.1242(14) 0.0220(33) -0.2233(10) 0.130 C(25) 0.1593(15) -0.1211(25) -0.1588(12) 0.140 C(26) 0.2348(12) -0.0546(19) -0.0805(9) 0.109 213 Table A20. Fractional coordinates and equivalent isotropic displacement parameters for Cp*W(NO)(r ? 3 -C 7 H 1 1 )(H) (4.9 e n d o) X y z U(eq) W(l ) 3433(1) 6447(1) 2047(1) 36(1) C ( l ) 3341(13) 7560(7) 924(6) 37(2) C(2) 2630(14) 6721(8) 571(7) 38(2) C(3) 774(16) 6639(9) 96(8) 49(3) C(4) -442(15) 7436(9) 339(8) 58(3) C(5) 464(17) 8404(8) 278(9) 52(3) C(6) 2179(15) 8480(7) 914(8) 42(3) C(7) 5008(15) 7530(8) 1426(8) 53(3) C(8) 4147(17) 4761(9) 2080(8) 52(3) C(9) 5794(13) 5261(8) 2020(7) 42(3) C(10) 6333(11) 5814(9) 2795(7) 41(3) C ( l l ) 4911(15) 5742(9) 3271(7) 50(3) C(12) 3645(16) 5059(9) 2873(9) 55(3) C(13) 3313(21) 4036(11) 1463(11) 82(5) C(14) 6940(17) 5143(9) 1323(9) 62(4) C(15) 7923(18) 6414(8) 2938(11) 66(4) C(16) 4964(24) 6177(12) 4130(9) 84(5) C(17) 2035(20) 4691(10) 3228(12) 85(5) N( l ) 2300(12) 7312(7) 2608(6) 44(2) O( l ) 1620(12) 7847(8) 3072(6) 74(3) Table A21 Fractional coordinates and equivalent isotropic displacement parameters for Cp*W(NO)(n 3 -C 8 H 1 3 )(H) (4.11 e x o) x y z U(eq) W( l ) 0.64157(3) 0.17219(2) 0.36499(2) 2.012(6) 0(1) 0. 7957(7) -0.0135(4) 0.3192(4) 5.5(2) N( l ) 0.7258(7) 0.0622(5) 0.3323(4) 3.18(14) C( l ) 0.5883(7) 0.1911(5) 0.5156(4) 2.32(14) C(2) 0.7467(8) 0.1610(5) 0.5140(4) 2.45(14) C(3) 0.8301(7) 0.2353(5) 0.4724(5) 2.87(15) C(4) 0. 7204 (8) 0.3132(5) 0.4500(5) 2.56(15) C(s) 0.5711(7) 0.2831(5) 0.4770(5) 2.34(14) C(6) 0.4598(9) 0.1388(6) 0.5636(5) 37(2) C(7) 0.8188(11) 0.0698(6) 0.5556(6) 4.5(2) C(8) 1.0043(8) 0.2377(8) 0.4632(6) 5.0(2) C(9) 0.7587(11) 0.4104(6) 0.4129(6) 4.2(2) C(10) 0.4227(9) 0.3450(6) 0.4724(6) 4.2(2) C ( l l ) 0. 7270(9) 0.2429(7) 0.2416(5) 4.0(2) C(12) 0.5651(10) 0.2593(5) 0.2389(4) 33(2) C(13) 0.4491(9) 0.1890(5) 0.2304(5) 2.9(2) C(14) 0.2778(9) 0.2157(7) 0.2386(5) 40(2) C(15) 0.1744(9) 0.1336(7) 0.2640(6) 40(2) C(16) 0.1892(9) 0.0475(6) 0.2026(6) 43(2) C(17) 0.3569(9) 0.0161(6) 0.1978(5) 35(2) C(18) 0.4612(9) 0.0993(6) 0.1729(5) 33(2) 215 Table A22. Fractional coordinates and equivalent isotropic displacement parameters for Cp*W(NO)(CH 2 CMe 2 -o-C 6 H4)(PMe3) (5.1) X y z U(eq) W(l) 1757(1) 1247(1) 1918(1) 13(1) P( l ) 3741(1) 1248(1) 952(1) 18(1) 0(1) 1243(4) 2996(3) 951(2) 33(1) N(l ) 1502(4) 2312(3) 1362(2) 20(1) C ( l ) 145(6) 221(3) 2586(3) 22(1) C(2) -526(5) 721(3) 1940(3) 19(1) C(3) 75(6) 404(3) 1194(3) 20(1) C(4) 1109(6) -270(4) 1391(3) 24(1) C(5) 1166(6) -371(4) 2253(3) 24(1) C(6) -284(7) 221(4) 3455(3) 31(1) C(7) -1713(5) 1371(3) 2011(3) 29(1) C(8) -421(6) 666(4) 376(3) 35(1) C(9) 1811(7) -898(4) 802(3) 39(1) C(10) 1984(6) -1074(3) 2726(3) 35(1) C ( l l ) 3553(4) 1234(4) 2758(2) 14(1) C(12) 3677(5) 1975(3) 3309(3) 17(1) C(13) 4805(6) 2014(4) 3825(3) 26(1) C(14) 5771(5) 1313(5) 3839(3) 28(1) C(15) 5610(5) 538(4) 3336(3) 24(1) C(16) 4541(5) 516(3) 2790(3) 20(1) C(17) 2534(5) 2676(3) 3317 (3) 18(1) C(18) 1280(5) 2133 (3) 3004(3) 21(1) C(19) 2209(6) 3049(4) 4175(3) 31(1) C(20) 2872(6) 3530 (3) 2789(3) 27(1) C(21) 3161(6) 1439(4) -83(3) 33 (1) C(22) 5022(6) 330(4) 798(3) 31(1) C(23) 4823(6) 2258(4) 1128(3) 29(1) 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0061332/manifest

Comment

Related Items