Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

New types of organometallic oxo and related complexes containing molybdenum and tungsten Phillips, Everett C. 1989

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

Item Metadata

Download

Media
831-UBC_1989_A1 P44.pdf [ 12.38MB ]
Metadata
JSON: 831-1.0060263.json
JSON-LD: 831-1.0060263-ld.json
RDF/XML (Pretty): 831-1.0060263-rdf.xml
RDF/JSON: 831-1.0060263-rdf.json
Turtle: 831-1.0060263-turtle.txt
N-Triples: 831-1.0060263-rdf-ntriples.txt
Original Record: 831-1.0060263-source.json
Full Text
831-1.0060263-fulltext.txt
Citation
831-1.0060263.ris

Full Text

NEW TYPES OF ORGANOMETALLIC OXO AND RELATED COMPLEXES CONTAINING MOLYBDENUM AND TUNGSTEN by EVERETT C. PHILLIPS  B. Sc., University of British Columbia  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE STUDIES (Department of Chemistry)  We accept this thesis as conforming to the required standard  THE UNIVERSITY OF BRITISH COLUMBIA September 1989 ©Everett Charles Phillips, 1989  In  presenting  degree at the  this  thesis  in  University of  partial  fulfilment  of  of  department  this thesis for or  by  his  or  scholarly purposes may be her  representatives.  permission.  The University of British Columbia Vancouver, Canada  for  an advanced  Library shall make it  agree that permission for extensive  It  publication of this thesis for financial gain shall not  DE-6 (2/88)  requirements  British Columbia, I agree that the  freely available for reference and study. I further copying  the  is  granted  by the  understood  that  head of copying  my or  be allowed without my written  ii  ABSTRACT Treatment of solutions of the 16-electron dialkyl compounds Cp'M(NO)R2 (Cp* = Cp (r? -C H ) or Cp* (r -C Me ); M = Mo or W; R = CH SiMe , CH CMe , 5  5  5  5  ?  5  5  2  3  2  3  CH CMe2Ph, or Me) with either dioxygen or aqueous hydrogen peroxide in Et 0 at 2  2  ambient temperatures and pressures produces novel 16-electron dioxo alkyl complexes Cp'M(0) R which can be isolated in moderate yields (45 - 60 %). The results of labelling 2  studies with 0 are consistent with the first steps of the reactions involving 0 proceeding 1 8  2  2  via the coordination of the diatomic molecule to the metal centers in the organometallic reactants. The Cp'Mo(NO)(CH SiMe ) complexes react with adventitious 0 or H 0 to 2  3  2  2  2  form [Cp'Mo(NO)(CH SiMe )] -(/i-0) complexes, the Cp*-analogue of which has been 2  3  2  structurally characterized by X-ray crystallography. The unique members in the class of cyclopentadienylnitrosyl dialkyl complexes that exhibit limited reactivity with dioxygen are the nitrosyl bis(benzyl) complexes Cp'M(NO)(CH Ph')2 (M = Mo or W; Ph" = Q H 5 or C H -2,4,6-Me ). Investigations into 2  6  2  3  the molecular structure of these complexes has revealed that they are formally coordinatively saturated, 18-electron complexes possessing the unusual Cp'M(NO)(r; 2  CH Ph')(r7 -CH Ph') structure both in the solid-state and in low temperature solutions. 1  2  2  Variable temperature ^Ti and C NMR spectroscopic studies of these complexes in 1 3  CD C1 solutions have established that the bis(benzyl) complexes undergo a facile fluxional 2  2  process, the two benzyl ligands interchanging their modes of attachment to the molybdenum or tungsten metal center. The fluxional process is more facile for the tungsten bis(benzyl) complexes than the molybdenum species (AG* being between 8.4 -12.0 kcal/mol). Interestingly, only the tungsten species of this group of bis(benzyl) complexes Cp'M(NO)(CH Ph') react with dioxygen to afford the corresponding 2  2  Cp'W(0) (CH Ph) compounds. 2  2  Reactivity studies have revealed that the preferred sites of reactivity in the cyclopentadienyl dioxo alkyl complexes Cp'M(0) R are their M = 0 linkages. Thus, 2  iii  reactions of the dioxo alkyl complexes with 30% K^C^aq) in Et20 solution converts them cleanly to novel organometallic peroxo alkyl complexes Cp'M(0)(»? -C>2)R, several of which 2  have been isolated and characterized. These complexes are monomelic 16-electron entities having a piano-stool molecular structure, a fact which has been confirmed by a single crystal X-ray crystallographic analysis of C£W(0)(f? -C^)(CH2SiMe3). Reaction of 2  Cp*W(0)(r/ -0 )(CH SiMe ) with tetracyanoethylene (TCNE) results in the formation of 2  2  2  3  a 2:1 charge-transfer complex [Cp*W(0)(r? -02)(CH2SiMe3)]2- *-[(CN) C=C(CN)2] whose 2  r  2  solid-state molecular structure has been determined by X-ray analysis. In solution however, the charge-transfer species between Cp*W(0)(r? -02)(CH2SiMe3) and TCNE 2  exhibits Benesi-Hildebrand behavior and is, therefore, formulated as a 1:1 adduct. Treatment of representative Cp'W^O^R compounds with HC1 (or PCI5 or IV^SiCl) in Et20 produces the corresponding Cp'W(0)Cl2R compounds in high yields. These dichloro oxo complexes are very useful synthetic precursors in their own right, undergoing metathesis with alkyllithium or alkyl Grignard reagents to produce the oxo trialkyl compounds Cp'W(0)R R'3_ in which R and R' represent selected alkyl groups. Sterically n  n  crowded members of this class of compounds are not isolable under ambient conditions since they spontaneously convert to the corresponding oxo alkylidene species Cp'W(0)( = alkylidene)R, presumably via intramolecular a-H abstraction. The isolable oxo trialkyl complexes can be induced to undergo the same conversions simply by gentle warming of their solutions. Treatment of representative Cp'M(0) R complexes with /j-tolyl isocyanate in 2  toluene solution at reflux results in their conversion to the corresponding Cp*M(=NQH4-/?-Me)[-N(QH4-/?-Me)C(0)N(QH4-p-Me)]R compounds. These imido urea complexes are thought to form as a result of trapping of the bis(imido) complexes Cp*M(=NQH4-/>-Me)2R, analogues to the dioxo alkyl complexes Cp*M(0) R, with an 2  equivalent of the p-tolyl isocyanate reagent.  iv Table of Contents  Abstract  ii  List of Figures  ix  List of Tables  xiii  List of Schemes  xv  List of Abbreviations  xvi  Acknowledgements Chapter 1  xviii  An Introduction to Organometallic Oxo Compounds  1  A.  Applications of Organometallic Oxo Compounds  2  B.  A Brief Historical Development  5  C.  Structure and Bonding in Organometallic Oxo Compounds  8  D.  Characterization of Metal Oxo Complexes  10  E.  Preparation of Organometallic Oxo Compounds  11  F.  Scope of the Work Presented in this Thesis  15  G.  References and Notes  17  Chapter 2  New Types of Organometallic Oxo- and Dioxo-Alkyl Complexes of Molybdenum and Tungsten  23  Introduction  24  Experimental Section  26  Results and Discussion  41  A. Preparation of the Cyclopentadienylnitrosyl Dialkyl Starting Materials, Cp'M(NO)R , (Cp'= Cp and Cp*; 2  M = Mo and W; R = CH , CH SiMe , CH CMe , 3  CH CMe Ph, and CH Ph) 2  2  2  2  3  2  3  41  V  B. Spectroscopic and Some Physical Properties of the Cyclopentadienylnitrosyl Dialkyl Complexes, Cp'M(NO)R , 2  (Cp' = Cp and Cp*; M = Mo and W; R = CH , CH SiMe , 3  2  3  CH CMe , and CH CMe Ph) 2  3  2  46  2  C. Spectroscopic and Physical Properties and Molecular Structures of the [Cp'Mo(NO)(CH SiMe )] 0i-O) 2  3  2  (Cp' = Cp or Cp*) Complexes  50  D. Syntheses of a Series of Novel 16-electron Cyclopentadienyl Dioxo Alkyl Complexes, Cp'M(0) R (Cp' = Cp or Cp*, 2  R = CH SiMe , CH CMe , CH CMe Ph, CH Ph, or Me) 2  3  2  3  2  2  2  of Molybdenum and Tungsten  56  E. Some Physical Properties of the Cp'M(0) R Complexes 2  (Cp = Cp or Cp*, M = Mo or W, R = CH SiMe , CH CMe , 1  2  3  2  3  CH CMe Ph, CH Ph, or Me) 2  2  60  2  F. Possible Mechanisms for the Reactions of the Cp'M(NO)R  2  Complexes with Molecular Oxygen  65  Summary  73  References and Notes  75  Chapter 3  Synthesis, Characterization, Physical Properties and Unusual Solid-State Molecular Structures of a Series of Nitrosyl Bis(Benzyl) Complexes of Molybdenum and Tungsten: Cp'M(NO)(CH Ph') , [Cp'= Cp 2  2  (r, -C H ) or Cp* (r? -C Me ), Ph' = C H , or C H -2,4,6-Me )] 5  5  5  5  5  5  6  5  6  2  3  78  Introduction  79  Experimental Section  80  vi Results and Discussion  86  A. Synthesis and Some Physical Properties of the Nitrosyl Bis(benzyl) Complexes: Cp1A(^0)(CH^h%  (Cp' = Cp or Cp*;  M = Mo or W; Ph' = C H or C H -2,4,6-Me ) 6  5  6  2  86  3  B. Typical Benzyl Ligand Bonding Modes to Transition-metals Reported to Date in the Literature  87  C. Solid-State Molecular Structures of the Bis(benzyl) Complexes la, lb, 2,3 and 4  90  D. Comparison of the Intramolecular Dimensions of the Bis(benzyl) Complexes la, lb, 2,3, and 4, with Those Found in Related Transition-Metal-Benzyl Complexes  101  E. Spectroscopic Properties of the Nitrosyl Bis(benzyl) Complexes  106  F. Variable Temperature NMR Studies of the Bis(benzyl) Complexes la, lb, 2,3, and 4  110  G. Determination of the Free Energy of Activation for the Fluxional Process  121  H. A Formulation of the Bonding in the (rj -CH_Ph') Ligand of the 2  Bis(benzyl) Complexes la, lb, 2,3 and 4  124  Summary  125  References and Notes  126  Chapter 4  New Types of Cyclopentadienyl Oxo Trialkyl Complexes, Cp'W(0)R , 3  and Oxo Alkylidene Complexes, Cp'W(0)(=R')R, of Tungsten  129  Introduction  130  Experimental Section  131  vii  Results and Discussion  138  A. Synthesis, Spectroscopic and Some Physical Properties of the Cp'W(0)(Cl) R Complexes (Cp'= Cp or 2  (Cp = Cp or Cp*, R = C H or CH SiMe )  138  1  3  2  3  B. Characteristic Reactivity of the Cp'W(0)(Cl) R Complexes 2  (Cp* = Cp or Cp*, R = C H or CH SiMe ) 3  2  147  3  C. Future Studies  154  Summary  155  References and Notes  156  Chapter 5  Novel Organotransition Metal Peroxo Alkyl Complexes of Molybdenum and Tungsten: Cp'M(0)(r7 -0_)R  158  2  Introduction  159  Experimental Section  160  Results and Discussion  166  A. Synthesis and Some Physical Properties of the Cp'M(0)(r -0 )R Complexes (Cp' = Cp or Cp*, 2  ?  2  M = Mo or W, R = Me or CH SiMe ) 2  166  3  B. Reactions of Cp*W(0)(r? -0 )(CH SiMe ) with Ph P and 2  2  (CN) C=C(CN) 2  2  3  3  170  2  C. Spectrophotometry Determination of the Formation Constant Of the Charge Transfer Complex Between Cp*W(0)(r? -0 )(CH SiMe ) and TCNE 2  2  D. Future Studies  2  3  180 185  Summary  187  References and Notes  189  viii Chapter 6  Reactions of the Dioxo Alkyl Complexes, Cp'M(0) R, 2  with />-Tolyl Isocyanate  191  Introduction  192  Experimental Section  193  Results and Discussion  200  A. Synthesis and Some Physical Properties of the Cp*M(=NC H -/>-Me)[-N(C H -^-Me)C(0)N(C H ^-Me)]R 6  4  6  4  6  4  Complexes [M = W, R = CH SiMe (3a), C H (3b); 2  3  3  M = Mo, R = C H (3c)].  200  3  B. Solid-State Molecular Structure of Cp*W(=NC H -/»-Me)[-N(C H -^-Me)C(0)N(C H - -Me)]6  4  CH SiMe (3a) 2  3  C. Future Studies References and Notes Spectral Index  6  4  6  4  jP  205 209 210 211  ix  List of Figures  Figure 2.1  (a) 300 MHz H and (b) 75 MHz ^ ^ H } NMR spectra of J  Cp*Mo(NO)(CH SiMe ) in Q D 2  3  2  49  6  Figure 2.2  A view of the molecular structure of [Cp*Mo(NO)(CH SiMe )] -0* -O)  Figure 2.3  (a) 300 MHz *H NMR and (b) 75 MHz ^C^H} NMR spectra  2  3  2  51  of [Cp*Mo(NO)(CH SiMe )] -(>-0) in Q D solution at 2  3  2  6  ambient temperatures Figure 2.4  54  80 MHz *H NMR spectra at various times during the reaction between Cp*W(NO)(CH SiMe ) and 0 in Q D solution in the 2  3  2  2  6  presence of trace amounts of H 0 ; (a) time = 0 min, 2  (b) time = 1 h, (c) time = 1.5 h, (d) time = 2 h. Figure 2.5  70  Spectral region 1510-520 cm" of the IR spectrum of 1  (a) analytically pure Cp*W(0) (CH SiMe ) as a Nujol mull and 2  2  3  (b) the solid residue from the final reaction mixtures of the 1 8  0 labelling experiment as a Nujol mull  72  2  Figure 3.1.  A view of the molecular structure of la  92  Figure 3.2.  A stereo-view of the molecular structure of lb  96  Figure 3.3.  A view of the molecular structure of 2  97  Figure 3.4.  A view of the molecular structure of 3  99  Figure 3.5.  A view of the molecular structure of 4, showing the planarity of the rj -CH Ph ring with respect to the central tungsten atom 2  2  Figure 3.6.  100  (a) 300 MHz 4 l and (b) 75 MHz C ^H} NMR spectra of B  CpMo(NO)(CH C6H -2,4,6-Me )2, (lb), in CD C1 solutions 2  2  at ambient temperatures  3  2  2  112  Figure 3.7.  (a) 300 MHz H and (b) 75 MHz C { H} NMR spectra of X  B  X  Cp*W(NO)(CH2C6H ) , (4) in CD C1 solutions at ambient 5 2  2  2  temperatures Figure 3.8  NMR spectra of CpMo(NO)(CH C H2-2,4,6-Me3)2, (lb), in CD C1 2  6  2  2  solution at 208 K: (a) 300 MHz *H NMR spectrum, (b) 75 MHz C CH} NMR spectrum, (c) 75 MHz gated decoupled B  U  Figure 3.9  C NMR spectrum, (d) 75 MHz (APT) C NMR spectrum B  80 MHz H NMR spectra of CpMo(NO)(CH C H5)2, (la) in CD C1 J  2  6  2  solutions at 183 K. Proton-proton decoupling experiments: (a) methylene-proton region of undecoupled spectrum. (b) irradiation at -1.08 ppm, (c) irradiation at 1.62 ppm, (d) irradiation at 2.67 ppm, and (e) irradiation at 3.19 ppm Figure 3.10  (a) 20 MHz C ^H} NMR spectrum of CpMo(NO)(CH C H )2, B  2  6  5  (la), in CD C1 solution at 193 K ; (b) 100 MHz C B  2  2  CP/MAS NMR spectrum of CpMo(NO)(CH C H ) , (la), at 2  6  5  2  ambient temperature Figure 3.11  80 MHz *H NMR spectra of la in CD C1 solution 2  2  at various temperatures Figure 4.1  (a) 300 MHz H and (b) 75 MHz gated decoupled C NMR X  B  spectra of Cp*W(0)(Cl)2(CH SiMe ) in Q D 2  3  6  solution at ambient temperatures Figure 4.2  300 MHz *H NMR spectrum of Cp*W(0)Me (CH SiMe3) in Q D 2  2  solution at ambient temperatures Figure 4.3  300 MHz H NMR spectrum of Cp*W(0)Me(CH Ph)2 in QDg J  2  solution at ambient temperatures  6  2  xi Figure 5.1  NMR spectra of CpW(0)(r/ -0 )(CH SiMe ) in C D 2  2  2  3  6  6  solution at -25 °C, (a) 300 MHz H NMR spectrum J  and (b) 75 MHz C{ H} NMR spectrum 13  Figure 5.2  169  1  A view of the solid-state molecular structure of CpW(C% -0 )(CH SiMe )  171  2  2  Figure 5.3  2  3  An ORTEP plot of the molecular structure of [(IJ -C Me )W(0)(r -0 )(CH SiMe )] - n -[(CN) C=C(CN) ] 5  Figure 5.4  177  2  5  5  ?  2  2  3  2  2  2  Partial views of the molecular structure of [Cp* W(0)(f? -0 )(CH SiMe )] - u -[(CN) C=C(CN) ], 2  2  2  3  2  2  2  (A) a view illustrating the planarity of the bridging TCNE molecule between the two »? -0 peroxo groups and 2  2  (B) a view illustrating the skewed orientation of the . ri - 0 unit with respect to the C=C double bond of 2  2  the TCNE molecule Figure 5.5  179  UV-vis absorption spectra of the charge-transfer complex between Cp*W(0)(rj -0 )(CH SiMe ) and TCNE 2  2  2  3  in methylene chloride solutions; (A) ~ 0.0140 M Cp*W(0)(77 -0 )(CH SiMe ) at various concentrations 2  2  2  3  of TCNE (see Experimental Section); (B)_ 0.0155 M TCNE at various concentrations of Cp*W(0)(»7 -0 )(CH SiMe ) 2  2  2  3  (see Experimental Section). For comparison, the absorption spectra of solutions of 0.0143 M Cp*W(0)(r? -0 )(CH SiMe ) and 0.0155 M TCNE 2  2  2  3  alone in methylene chloride solution are indicated by (  ) and (••••) respectively  182  xii  Figure 5.6  Spectrophotometric detennination of the formation constant of the charge-transfer complex between Cp*W(C% -0 )(CH SiMe ) and TCNE in methylene 2  2  2  3  chloride at 25 °C according to the Benesi-Hildebrand method (eq. 5), a plot of [Cp*W(0)(r -0 )(CH SiMe )]/Absorbance versus [TCNE]" 2  ?  Figure 6.1  1  2  2  3  (a) 300 MHz H and (b) 75 MHz C{ H} NMR J  D  spectra of  1  W^NQH^-MeH-NtQH^-Me^O^-  tQjH^-MeWCHiSiMes), 3a Figure 6.2  184  202  Views of the solid-state molecular structure of Cp* W(=NC H4-/>-Me)[-N(QH -/>-Me)C(0)N6  4  (QH -/>-Me)](CH SiMe ), 3a: (a) side-view of molecule. 4  2  3  (b) view of molecule down the Cp^cen^d) - W axis (Cp* atoms omitted for clarity)  207  xin  List of Tables  Table 2-1  Selected Bond Lengths (A) in [Cp*Mo(NO)(CH SiMe )] -fc-O)  52  Table 2-H  Selected Bond Angles (°) in [Cp*Mo(NO)(CH SiMe )] -0i - O)  52  Table 2-D3  Analytical and IR Data for the Dioxo Alkyl Complexes  61  Table 2-IV  Mass Spectral and H and C{ H} NMR Data for the  2  3  2  X  B  3  2  2  1  Dioxo Alkyl Complexes Table 2-V  Assignment of the i /  w  =  62  Values to the Cp*W(0) (CH SiMe ) 2  Q  2  3  Isotopomers Table 3-1  71  Pertinent Crystallographic Data for the Complexes Cp'MCNOXCH^h'^ (Cp* = Cp or Cp*; M= Mo or W; Ph" = Q H 5 , or C6H -Me ), la, lb, 2, 3, and 4 2  Table 3-II  91  3  Bond Lengths (A) in the Complexes Cp'M(NO)(CH Ph) 2  2  (Cp'= Cp or Cp*; M= Mo or W) (la, 2, 3, and 4), and (^Mo(NO)(CH QH -2,4,6-Me ) , (lb) 2  Table 3-III  2  3  94  2  Bond Angles (deg) in the Complexes Cp'M(NO)(CH Ph) 2  2  (Cp'= Cp or Cp*; M= Mo or W) (la, 2, 3, and 4), and CpMo(NO)(CH C H -2,4,6-Me ) , (lb) 2  Table 3-IV  6  2  3  2  A Comparison of the Intramolecular Dimensions in Structurally Characterized Metal-Benzyl Complexes  Table 3-V  107  Variable Temperature *H and C{ H} NMR Data for the B  1  Nitrosyl Bis(Benzyl) Complexes Table 4-1  102  Analytical, IR and Mass Spectral Data for the Nitrosyl Bis(Benzyl) Complexes  Table 3-VI  95  108  Analytical and IR Data for the New Oxo Complexes Isolated in this Work  139  Table 4-H  Mass Spectral and *H and ^C^H} NMR Data for the Oxo Complexes  Table 5-1  Analytical and IR Data for the Peroxo Alkyl Complexes  Table 5-H  Mass Spectral and *H and C{ H} NMR Data B  1  for the Oxo Complexes Table 5-IH  Selected Bond Lengths (A) and Bond Angles (deg) in CpW(0)(r -0 )(CH SiMe ) 2  ?  Table 5-TV  2  2  3  Selected Bond Lengths (A) and Bond Angles (deg) in [Cp*W(0)(r/ -0 )(CH SiMe )] - M-[(CN) C=C(CN) ] 2  2  Table 5-V  2  3  2  2  Spectrophotometry Determination of the Formation Constant of the Charge-Transfer Complex between Cp*W(0)(rj -0 )(CH SiMe ) and TCNE 2  2  Table 6-1  2  3  Selected Bond Lengths (A) and Bond Angles (deg) in Cp*W(=NC H -/;-Me)[-N(C H -/7-Me)C(0)N6  4  6  (C H -/>-Me)](CH SiMe ) 6  4  2  3  4  2  XV  List of Schemes  Scheme 2-1  68  Scheme 5-1  173  Scheme 5-2  186  Scheme 5-3  186  Scheme 6-1  204  List of Abbreviations  Anal.  -  Analysis  'Bu  -  (CH ) C, tertiary-butyl  calcd  -  calculated  C D  -  benzene-rf  6  6  CD C1 2  3  3  6  dichloromethane-d  2  2  cm"  -  wavenumbers  Cp  -  » -C H  Cp*  -  r? -C Me  Cp  -  Cp or Cp*  CP-MAS  -  cross-polarization magic angle spinning  1  1  1 3  C  5  5  5  5  5  5  carbon-13  ^C^H}  -  proton-decoupled carbon-13  deg  -  degrees  Et  -  CH CH , ethyl  -  (CH CH2) 0, diethyl ether  Et 0 2  eV HOMO  3  2  3  2  electron volts -  highest occupied molecular orbital  *H  proton  IR  infrared  / •  coupling constant (in the NMR spectrum)  kcal  kilocalories  LUMO  -  lowest unoccupied molecular orbital  Me  -  C H , methyl 3  mmol  millimole  MO  molecular orbital  xvii  mol  mole  m/z  mass-to-charge ratio in the mass spectrum  NMR  nuclear magnetic resonance  P  +  Ph PMe PPh  3  3  TCNE  -  molecular ion (in the mass spectrum)  -  0 % phenyl  -  P(CH ) , trimethylphosphine  -  P(C6H ) , triphenylphosphine  -  tetracyanoethylene  3  3  5  3  THF  tetrahydrofuran  UV-vis  ultraviolet-visible  xviii Acknowledgements  I would like to express my deepest gratitude to Professor Peter Legzdins, who, over the course of the last few years, has been an outstanding friend and advocate, both academically and personally. Without his spirited enthusiasm and guidance, much of this work would not have been possible. I wish to thank the faculty and staff of the Chemistry Department for their assistance and useful discussions. Specifically, I thank those members of the Department who have made the summer months so enjoyable out on the softball field. Furthermore, discussions with Peter Borda, Steve Rak, Marietta Austria, Bev Gray and Lani Collins have been particularly helpful. I would also like to thank Vivien Yee and Drs. F. W. B. Einstein, R. H. Jones, S. J. Rettig, J. Trotter for the X-ray crystallographic analyses which are presented in this work. I also wish to thank the past and present Legzdin-ites for their help and friendship. Special thanks goes to Luis Slnchez, Allen Hunter, George Richter-Addo, Neil Dryden, and Teen Chin, with whom I spent many fruitful hours, both in and out of the lab. Finally, I must thank Nancy, my wife and closest friend, for her love and support.  xix  To my parents, Dave and Halga Phillips. Thanks for your love and continuous encouragements.  1  Chapter 1  An Introduction to Organometallic Oxo Compounds  2 The purpose of the following introduction is to provide some background to the research presented in this thesis. It is beyond the scope of this introduction to provide a thorough survey of the topic of organometallic oxo complexes. Instead, a synopsis of the field will be presented, highlighting a few important points of interest and developments up to the present. Furthermore, a brief discussion of some of the types of organometallic oxo complexes known, their structures and methods of preparation will help the reader appreciate some of the chemistry of high oxidation state organometallic oxo compounds. It should be noted that an excellent comprehensive review of the subject up to 1988 written by Bottomley and Sutin has recently appeared. Furthermore, for a detailed review of the 1  organometallic oxo complexes of rhenium, one should refer to recent articles by Herrmann.  2  A.  Applications of Organometallic Oxo Compounds. Since the 1940's and 1950's, high oxidation state metal oxide catalysts have been  used in many of the commercially important industrial processes that produce large quantities of organic chemicals via the oxidations of petroleum hydrocarbon feedstocks. The majority of the industrial processes of the past have involved vapor-phase oxidations over heterogeneous catalysts, an example being the ammoxidation of propylene to acrylonitrile by bismuth molybdate, eq. 1. [Bi 0 /Mo0 ] «-CH =CHCN 2  CH =CHCH 2  3  + NH  3  + 3/2 0  2  3  3  2  + 3^0  (1)  However, in the past few decades, homogeneous transition metal-oxide species have been used increasingly as selective liquid-phase oxidants in the synthesis of natural products and other fine organic chemicals.  3  Metal catalyzed oxidations of organic chemicals are often categorized into three broad classes, and commercial examples of each are provided below. The first class of  3  reactions involves free radical autoxidation reactions, such as the process for the production of terephthalic acid by the oxidation of /^-xylene, eq. 2. [Co(OAc) / Br"] • H0 C-QH -C0 H + 2H 0 H 0 2  CH3-C6H4-CH3 + 3 0  2  2  4  2  (2)  2  2  The second class of reactions involves nucleophilic attack of the metal species on coordinated organic substrates, such as the Wacker process which employs palladium complexes to catalyze the oxidation of ethylene to acetaldehyde, eq. 3. 4  [PdCl / CuCl ] 2  2CH = CH 2  2  + 0  2  •  2  2 CH CHO  (3)  3  H 0 2  An adaptation of this process in non-aqueous media involves the formation of vinyl acetate from the oxidation of ethylene in acetic acid, eq. 4. This process was reported by Moiseev and co-workers, and it is also being used in industry. 5  6  [PdCl / CuCl ] • CH =CHOAc + H 0 (4) HOAc 2  CH =CH 2  2  + l/2 0  2  + HO Ac  2  2  2  The third class of oxidations consists of reactions between organic substrates and hydroperoxides. Examples of these processes are the oxidation of propylene with alkylhydroperoxide in the presence of a Mo-oxide catalyst to produce propylene oxide, eq. 5, [Mo] CH CH = C H 3  2  + ROOH  •  2 CH CH - C H + ROH 3  2  and the hydroxylation of cyclohexene with H2O2 in the presence of OSO4, eq. 6.  +H 0 2  C^Q  H  2  +  (5)  4  Oxidations involving heterogeneous catalysts are often performed at high temperatures but with modest selectivity, while those involving homogeneous catalysts operate at much lower temperatures and provide increased selectivity. The problem of the cUrninishing world supply of petroleum, has resulted in the conservation of petroleum feedstocks becoming an important goal for the chemical industry. Therefore, in order for a catalytic oxidation process to be commercially viable, it must be efficient, and the catalysts used must be highly selective. While the currently employed metal-oxide catalysts are useful, the mechanism of the oxidation processes and the role of the metal-oxide species in these processes are still poorly understood. The interest in high oxidation state organometallic oxo complexes, therefore, results in part from the anticipation that a detailed study of their characteristic chemistry may provide the desired insight into how the metal oxides function as catalysts for these organic reactions and as oxidizing agents. It has been argued on theoretical grounds that the role 7  of the oxo ligands during these catalytic and stoichiometric processes is to stabilize various metallacyclic intermediates by an increase in the metal-oxygen bond order, e.g. during olefin metathesis  (  0 = M = CH + C H = C H 2  2  2  +  )  CH  0  •  ^CH XH  2  2  In any event, these studies may provide new developments towards the preparation of more active and/or more selective catalysts than those known at present. In addition, metal oxo species are thought to be intermediates in a number of biological oxidation processes (e.g., catalysis involving cytochrome P-450 and other related enzymes). The same mechanistic 8  concepts may be applied to homogeneous, heterogeneous and enzymatic catalytic processes, and, therefore, studies of organometallic metal oxo compounds may provide model systems for these oxidation processes.  5  B.  A Brief Historical Development The first examples of organometallic oxo complexes reported in the literature were  the vanadium-containing species, CpV(0)X (Cp = f? -C5H . X = CI or Br). These 5  2  5  complexes were obtained by Fischer, Vigoureux and Kuzel in the late 1950's from the 9  reaction of CpV(CO)4 with HX (X= CI or Br) in the presence of dioxygen, eq. 7. CpV(CO)  4  + HX + 0  *-CpV(0)X  2  2  10  + 4CO + H 0 2  (7)  All the early organometallic oxo complexes were classified as organometallic by virtue of their containing the cyclopentadienyl ligand, the exception being the vanadium oxo aryl complex (r? -QH5)V(0)Cl , which contains a metal-carbon sigma bond. During the 1  11  2  1960's, reports were primarily concerned with the preparation of cyclopentadienyl metal oxo compounds of the Group 4, 5 12  1 3 3  and 6 " metals. In most cases, the oxo 14a  e  complexes were obtained by exposure of a variety of cyclopentadienyl metal carbonyl compounds to excess dioxygen under various experimental conditions. With respect to the chemistry presented in this thesis, the most notable complexes reported during this period were the compounds CpMo(0) X, (X= CI and Br), " the first examples of 14c  e  2  cyclopentadienyl dioxo complexes. It was not until the mid-1970's that several examples of organometallic oxo alkyl and oxo aryl complexes were reported, the early examples being of the Group 5 metals, i.e. V(0)(CH SiMe ) , ' 13b  2  3  13c  3  [ra(CH CMe ) 0x-O)] , 2  3  3  n  13d  and M(0)X Me.2L, (M= Ta, Nb; 2  X= CI, Br; L= OPMe , OPPh^. * The first examples of oxo alkyl complexes of rhenium 13  3  appeared, i.e. Re(0)R.4 (R= Me and CH SiMe ) and cis-Re(0) Me . 2  3  2  15  3  Furthermore, the  series of organometallic dioxo complexes was expanded slightly to include such members as, W(0) (Cl)Me.2L, ' g Mo(0) (r -l,3,5-C H Me ), 14f  14  1  2  2  ?  6  2  3  14h  and Mo(0) (bpy)(Br)R, (R= 2  Me, Et, CH CMe , n-C H , i-C H , CMe ). ' J It is surprising that none of the early 14i  2  3  3  7  3  7  14  3  examples of oxo alkyl complexes contained the cyclopentadienyl ligand since it is known to  6  stabilize high oxidation states by reducing the effective charge on the central metal through donation of JT -electron density from the ring into the empty d-orbitals on the metal. Another development that occurred in the mid-1970's was the appearance of the first examples of a class of cyclopentadienyl metal oxo acetylene complexes, i.e. CpM(0)(X)(r/ 2  C R ), [M= Mo or W; X= Ph , C l 16a  2  1 6 b  2  or SQFj ^ * ; R= Ph or CF ]. 16  16  1  3  As the number of isolated organometallic oxo complexes increased, researchers realized that a study of their chemistry could provide an understanding of the processes involved in the selective oxidations of organic chemicals by metal-oxide catalysts. The field, therefore, burst to life in the early 1980's, and still remains as one of the fastest growing fields in inorganic chemistry. This interest is manifested by the appearance of approximately 150 papers since 1980. To date, the metals involved in organometallic oxo complexes found in the literature are primarily restricted to the oxophilic elements in Groups 4 to 6 and the Group 7 element Re as shown below. 1  18 2  IS  3  4  S  6  7  8  8  10  11  14  15  16  17  12  Ti V Cr Zr Nb Mo Ru Hf Ta W Re Os  Sm  u There are however, a few examples of organometallic oxo complexes containing the metals U, Sm Cr, Ru, and Os, and for this reason they appear in italics above. The complexes [Cp* Sm] -0i-O) and [Cp U] -(/i-0) were isolated recently. Group 8 monomelic and 17  2  18  2  3  2  dimeric metal oxo alkyl or aryl complexes are extremely rare, the known examples being Os(0)(CH SiMe ) , Os(0) (r, -l,3,5-C H Me ) , and [Ru(CH SiMe ) ] k-0) . > 19  2  3  4  1  2  20  D  2  3  2  21  2  3  3  2  2  22  7 Since the research presented in this thesis began, several new types of oxo complexes have appeared, namely oxo hydride, oxo carbonyl, and oxo olefin complexes, as depicted below. There are, however, still only a few examples of each known.  Rare examples of organometallic oxo hydride complexes are [Cp*MH] -(/i-0) (M= Zr and 2  Hf) , Cp^TaCCOH, and Re(0)H(rj -RC=CR) . 23  24  2  2  25  Carbonyl and other *-acid ligands  such as olefins may bind strongly to metal oxide species if the metal is in a d configuration 2  as in W(0)Cl (PMePh ) (CO). This compound is the only monomelic oxo carbonyl 26  2  2  2  complex known, although there are many examples of metal oxo clusters having carbonyl ligands. Furthermore, the only known examples of oxo olefin compounds are 1  W(0)Cl (PMePh ) (r7 ^ C j H ^ 2  2  and Mo(0)(r; -S CN Pr) (» -C_(CN) )  2 6  2  2  n  2  2  2  7  However, oxo  27  4  acetylene complexes are much more common than oxo olefin compounds. In fact, 1  examples of bis(acetylene) oxo compounds of rhenium (HI) have been prepared recently, e.g. [Re(0)X(r -RC_CR) ] , (n= 0, X= CI, Br, I, H or Et; n= 1, X= py, PPh ). > 2  n+  ?  28  2  25  3  Of considerable interest are oxo alkylidene and oxo alkylidyne complexes of which many examples have been reported containing primarily the metals T a , M o and W . 29  30  30,31  Interestingly, the first example of a rhenium oxo alkylidene species, namely, Re(0) (=CHCMe )(CH CMe ), was reported last year. These types of complexes are 32  2  3  2  3  often invoked as intermediates in olefin and alkyne metathesis reactions, and some isolated compounds are active catalysts, an example being W(0)(Cl) (=CHCMe )(PEt3) (n= 1 or 2  2), which has been isolated by Schrock and co-workers  3 3  3  n  Some new types of  cyclopentadienyl oxo alkylidene complexes of tungsten are presented in Chapter 4 of this thesis.  8  Most recently, a novel class of organometallic peroxo complexes has been reported. For example, the complexes Cp'M(0)(r? -0 )R (Cp'= Cp or Cp*; M= Mo or W; R = 2  2  CH SiMe3 or Me), 2  34  Cp* Ta(»? -0 )R (R= Me, Et, Pr, Ph, or C H P h ) and Cp* Nb(r? 2  2  n  35  2  2  2  0 )X (Cp* = C5H4SiMe3, X= CI, or Me).  36  2  2  These complexes are of interest as possible  oxygen atom transfer agents for various organic transformations, e.g. epoxidation reactions of olefins. The Mo and W systems listed above are described in Chapter 5 of this thesis. C.  Structure and Bonding in Organometallic Oxo Compounds Nearly all the elements of the periodic table are known to form oxide compounds.  The term "organometallic oxo compound" applies to any complex that contains both the formal 0 " ligand and a metal-carbon linkage. The 0 " group can bind to metals in a 2  2  number of different ways, as depicted below:  M  M  M=0  1  M  wr  M  2  M  3  M  X  M  M  4  The organometallic oxo complexes that will be encountered in this thesis fall within the first two basic structural types above; (1) monomeric complexes containing terminal, multiply bonded metal oxygen linkages (M[=0] , n = 1, 2, or 3), or (2) dimetallic species n  that contain one or two oxygen atoms bridging between the two metal centers ([MQ*0) M], n = 1 or 2). In a few instances, complexes that contain a mixture of the two n  structural types 1 and 2 above will be encountered. The reader should consult reviews for discussions of trinuclear oxo complexes containing triply bridging [M3(ji-0)] linkages and tetranuclear oxo complexes containing quadruply bridging [M4Q1-O)] 4, respectively.  1  linkages, types 3 and  9  All transition metal-oxygen linkages contain some degree of 0 - » M x character in addition to the a bond. This results from the transition metal having some d-orbitals which are empty and, therefore, available to interact with the lone pair orbitals on the oxygen atom. In multiple metal-oxygen linkages of type 1 above, two r -interactions between the metal and the oxygen atom are possible which would give rise to a metaloxygen triple bond ( M H O ) . However, due to the lack of a suitable reference for a single M-O bond, it is often difficult to ascertain the actual metal-oxygen bond order extant in many oxo complexes, an order which must lie somewhere between 1 and 3. At present, evidence suggests that a bond of order near two is present in most oxo species and, therefore, an isolated metal-oxygen linkage is commonly depicted as M=0. It should be noted that in several oxo complexes of rhenium, e.g. Re(0)I(MeCsCMe)2,  28b  a Re_0  triple bond is invoked. Virtually all the dinuclear mono oxo-bridged organometallic complexes, [M(/i-0)M], have essentially linear M-O-M bridges. Indeed, a few complexes are known to contain an M-O-M angle of 180°, but most complexes contain M-O-M angles that fall in the range 168-175°. The linear M-O-M linkage may be considered to result from the situation where two M-O a-bonds of the M-O-M link are formed using two sp-hybrid orbitals on the oxygen atom, and the two pairs of *-electrons left in pure p orbitals on the oxygen atom interact with the empty d * -orbitals on the metal center. Large deviations from 180° are found in some complexes containing the M-O-M unit, for example [Cp2Zr(SPh)] -0*-O)  37a  2  and [Cp*V(O)Cl] -0i-O) contain Zr-O-Zr and V-O-V angles of 165.8° and 142.2°, 37b  2  respectively, thus, illustrating that a * -interaction between the O and the M does not necessarily require a linear M-O-M linkage. Often the M-O bond order in the M-O-M link is suggested to lie somewhere between the two limits, M-O-M and M=0=M. Introducing another bridging oxygen atom between the two metal centers in complexes of the type [M(/i-0)M] i.e. [Mfc-O^M], causes the M-O bond lengths to 38  increase slightly and the M-O-M angles to decrease substantially. For example, the  10 compound [Cp*Mo(0) ] (/j-0)  39  2  2  contains the intramolecular dimensions M o - O ^ ^ ^ =  1.864 A and Mo-O-Mo = 177.9°, while [Cp*Mo(0)] (>-0) 2  contain Mo-0^  ridge  38a  2  and [CpMoCO^Oi-O^  3815  ^ bond lengths between 1.946 A and 1.940 A, and average Mo-O-Mo  bond angles between 84° and 85°. Indeed, all complexes of the type [M(/i-0) M], contain 2  average M-O-M angles that are within the range of 83-93°.  383-11  Due to the existence of a  metal-metal bond in the complexes of the type [M(/i-0) M], it is difficult to describe the 2  bonding involved in the M-O-M linkages. The M-O-M angle is likely small in order for the two metal atoms to be within bonding distance from each other. D.  Characterization of Metal Oxo Complexes The most frequently used technique for the characterization of organometallic oxo  compounds is IR spectroscopy. All monomeric complexes containing a single metal-oxygen double bond (M = 0) exhibit in their IR spectra as Nujol mulls or KBr pellets a single strong band in the spectral region 980-850 cm" . There does not appear to be a direct 1  correlation to the observed frequency of the M - O vibration with a periodic change in the metal center. However, the observed ^  M  =  Q  is dependent on both the metal and the ligand  environment surrounding the metal center. IR spectra of complexes of the types M(=0) and [M(=0) ](/i-0) exhibit two 2  strong ^  2  ' with nearly equal intensity in the spectral region 960-880 cm . These are s  M = 0  due to asymmetric O=M=O ( v  -1  M = Q  asym) and symmetric 0=M=0 ( J >  vibrations of the cis-dioxo complex, the ^ higher in wavelength than the v  M = Q  m = 0  sym)  o &sym typically occurring some 20-40 cm"  1  M =  sym. Furthermore, IR spectra of complexes with cis-  dioxo ligands typically exhibit a third band in the region 390-370 cm" which is attributable 1  to a symmetrical bending motion of the 0 = M = 0 unit, i.e. 3 Q _  M = 0  -  IR spectra of oxo complexes of the type [M(/x-0)_M], (n= 1 or 2), often display a single broad band in the spectral region 820-700 cm" , which may or may not contain a 1  lower intensity shoulder at a lower wavelength.  11  Metal peroxo complexes containing a M^-C^) unit typically exhibit in their IR spectra one strong band in the region of 880-850 cm" and two medium intensity bands in 1  the region of 560-515 cm" . These bands are attributable to 1  i/  M 0  2  I/Q.Q  and J>MO2 asym and  sym bands, respectively. X-ray crystallography is also commonly used to characterize organometallic oxo  complexes. Recently, a statistical study of complexes containing metals of Groups 4 to 8 of the type M( = 0) , (n = 1,2 or 3) has shown that typical metal-oxygen bond lengths vary n  between 1.59 A and 1.76 A .  40  Furthermore, typical M-O bond lengths in complexes of the  type [M(/i-0)_M], (n= 1 or 2), usually fall within the range 1.80 - 2.10 E.  A.138,39,41  Preparation of Organometallic Oxo Compounds The preparation of organometallic oxo and oxo alkyl complexes is usually  accomplished by one of four basic methods: (1) oxidation of an organometallic compound, (2) hydrolysis of an alkyl complex, (3) labile ligand exchange reactions of oxo complexes, and (4) alkylation of an oxo halide or an oxo alkoxide complex with alkyl-Grignard, -lithium, -zinc or -aluminum reagents. Examples of each of these types of syntheses will be considered in turn below. (1) Oxidation of an Organometallic Compound The most useful method employed to prepare organometallic oxo complexes is by exposure of solutions of an organometallic compound to an oxidizing agent such as dioxygen or a nitrogen oxide (NO, N 0, RNO, and R NO). Reactions of this type 2  3  employing dioxygen are the most numerous probably due to the number of cases in which the isolation of organometallic oxo complexes probably resulted from inadvertent admission of air into a system containing an organometallic reaction. In any event, the most useful routes to cyclopentadienyl-metal oxo compounds are reactions of cyclopentadienyl-metal carbonyl complexes with dioxygen or hydrogen peroxide. The first  12 organometallic oxo complexes were prepared in this fashion (eq. 7), and other examples of this method are given below in equations 8-11. Cp NbCl(CO)  +  2  0  Cp Nb(0)Cl  2  (S)  36  2  hu/CHCU  [CpMo(CO) ] 3  +  2  0  +>  2  [CpMo(O) ] 0i-O)  +  2 2  [Cp*W(CO) ] 3  +  2  0  [CpMo(0) ] 2  +  2  CpMo(0) Cl 2  •[Cp*W(CO) ][CpW(0) ]  2  3  2  (9)  14d  (10)  42  O./hi. Cp*Re(CO)  CpRe(0)  3  or H 0 2  (ll)  3  43  2  This method has been generally restricted to the preparation of cyclopentadienyl metal oxo complexes with no alkyl ligands. There are relatively few cases where oxidations of metal alkyl complexes with 0 or H 0 have afforded oxo alkyl complexes; one such case, eq. 12, 2  2  2  is the subject of Chapter 2 of this thesis. CpM(NO)(R)  + 0  2  • CpM(0) R  2  + "RNO"  2  (12)  Oxidation reactions of organometallic complexes with nitrogen oxides are very useful for the preparation of oxo species. Examples of such oxidations are provided in eqs. 13-16. Cp* Ti  + N 0  2  Cp* VCl 2  2  2  •[Cp*Ti] 0x-O)0i-T :» -C Me CH )  (13)  • Cp*V(0)Cl + N 0 + "Cp*"  (14)  V(0)(CH SiMe )  (15)™°  1  2  + 2 NO  V(CH SiMe ) + 2  '  3  4  5  7  7  2  NO  5  +  2  3  3  NO  CpW(0)(f? -C2H )Me 2  2  2  2  2  CpW(CO) [CH = CHC(=0)Me]  4  44  45  *• +  CpW(0)(r -C H )(C(0)Me) 2  ?  2  2  (16)  46  13 Furthermore, Bottomley and co-workers have had much success in preparing organometallic metal oxo clusters by similar routes.  1  (2) Hydrolysis of a Transition-Metal Alkyl Complex or a Metal Cyclopentadienyl Complex Another method for the preparation of organometallic oxo complexes involves hydrolysis of a transition-metal alkyl complex or a metal cyclopentadienyl complex. Examples of such reactions are common for complexes of the Group 4 and 6 metals, eqs. 17 - 21. CpTiCl  [CpTiCl ] fc-0)  + H Q  3  2  2  2  [Cp*Zr(N )]_>-N ) + H Q  [Cp*Zr(H)] 0*-O) + 3 N  Cp HfMe  [Cp HfMe] 0i-O)  2  2  2  2  2  + H Q  2  2  2  2  2  (17)  1 2 a  (18)  2 3  (19)  4 7  4 1 b  W(CCMe )(CH CMe ) + H Q  [W(O)(CH CMe ) ] 0i-O)  (20)  Cp MoCl  Cp Mo(0)  (2D  3  2  2  2  3  3  2  + OH"  2  3  3  2  2  48  ( 3 ) Labile Ligand Exchange Reactions A third, and relatively new, preparative route to novel organometallic oxo complexes has recently been found. This involves the reaction of organometallic oxo complexes that contain labile ligands (such as phosphines) with olefins, acetylenes and carbon monoxide. Few such reactions exist, some examples being given in eqs. 22 - 24. However, as these types of complexes become more common, this synthetic route will likely provide interesting new types of organometallic oxo complexes. Re(0)I (EPh ) 3  3  3  + 2MeC = CMe  •Re(0)I(*; -C Me ) 2  2  2  2  (22)  28c  E= Por As W(0)Cl (PMePh ) 2  2  3  + CO  -^W(0)Cl (PMePh ) (CO) 2  2  2  (13)  26  14 W(0)Cl (PMePh ) + CH =CHR 2  2  3  2  <  •  W(0)Cl (PMePh ) (r -CH =CHR )  (24)  2  2  2  2  ?  2  26  2  (4) Alkylation of an Oxo Compound The fourth preparative route to organometallic oxo compounds involves chemical modification of an inorganic or organometallic metal-oxyhalide or -oxyalkoxide complex by alkylation. Metal-oxyhalides or -oxyalkoxides are typically reacted with alkyl- or aryllithium, -Grignard, -zinc, -mercury or -aluminum reagents to form the new metal carbon linkage(s). Some of the first examples of this method are given in equations 25-27. V(0)C1  + Ph Hg  3  • (r, -C H )V(0)Cl + PhHgCl  (25)  1  2  6  5  11  2  + L 2 W(0)C1  4  Re(0)Cl  + Me Mg  2 W(0)Cl (Me)L + MgCl (L = OEt^ dppe, or tdpo)  2  3  + 4 MeLi  4  ^Re(0)Me  4  (26) & 14  2  + 4 LiCl  49  (27)  50  Recently, there have appeared a few examples where one of the M = 0 bonds of an organometallic oxo compound of the type M(=0) , n= 2, 3, or 4, is also alkylated to n  afford new oxo alkyl complexes, eqs. 28 and 29. Cp*Re(0)  3  + A1R (or ZnR ) 3  Os(0) + 2 (Me SiCH ) Mg 4  3  • Cp*Re(0)R  2  2  2  •Os(0)(CH SiMe ) + Mg0 2  3  4  (28)  2  2b38f  (29)  19  2  Furthermore, a few research groups have recently exploited the reaction of the M = 0 bond with chlorinating agents (such as Me SiCH Cl, PC1 , Cl , HC1 or GeCl ) to 3  2  5  2  2  obtain new organometallic metal oxo halides. Examples of these conversions are given in eqs. 30 - 32, the chemistry of reaction 30 being considered in detail in Chapter 4 of this thesis.  15 Cp'W(0) R + XC1 —  *• Cp*W(0)(Cl) R + XO  2  (30)  2  (X= Me SiCH -, PC1 ,H) 3  2  4  HC1 or Me SiCl 3  [W(0) (OCMe )] ( n n -Et C CH CH C Et4) 5  2  3  5  2  4  5  2  2  5  [ W ( 0 ) C l 3 ] ( r , , ^ - E t ^ C ^ C H ^ E g (31),51 5  2  Cp*Re(0)  + GeCl »l,4-dioxane  3  *- Cp*Re(0)Cl  2  (SI) ' 2  2  52  The product oxo halide complexes may be alkylated as discussed above, and as a result, many new oxo alkyl compounds can be prepared in a more efficient two step procedure.  F.  Scope of the Work Presented in this Thesis. Chapter 2 describes the successful attempts to prepare a new series of  cyclopentadienyl dioxo alkyl complexes, Cp'M(0) R, of both molybdenum and tungsten 2  with a variety of alkyl ligands. These Cp'M(0) R complexes are prepared by reactions of 2  cyclopentadienylnitrosyl dialkyl complexes, Cp'M(NO)R , and dioxygen. The results of the 2  investigations into the mechanism by which these dioxo complexes are formed from their nitrosyl dialkyl starting materials are also discussed. Furthermore, the Cp'Mo(NO)(CH SiMe ) starting materials prepared in this work were found to be 2  3  2  thermally- and moisture-sensitive and to convert to new types of [Cp'Mo(NO)(CH SiMe )] (/i-0) complexes which have been isolated and characterized. 2  3  2  Chapter 3 deals with the preparation, physical properties and molecular structures of the unique members in the class of dialkyl nitrosyl complexes, namely a series of cyclopentadienylnitrosyl bis(benzyl) complexes of molybdenum and tungsten, Cp'M(NO)(CH Ph) . These complexes exhibit limited reactivity towards dioxygen due to 2  2  their possessing an interesting 18-electron, coordinatively saturated molecular structure.  16 Aspects of this chemistry and details of their characteristic temperature-dependent dynamic processes in solution are presented. Chapter 4 describes some characteristic reactivity of representative examples of the cyclopentadienyl dioxo alkyl complexes of tungsten, Cp'W(0) R, with electrophiles. Most 2  importantly, Chapter 4 outlines a direct synthetic route to a number of interesting oxo trialkyl, Cp'VJ(0)(R') (R), and oxo alkylidene, Cp'W(0)(=CHR)(R'), complexes of 2  tungsten. Chapter 5 is also concerned with some reactivity of the cyclopentadienyl dioxo alkyl complexes. The preparation, characterization and solid-state molecular structure of a series of novel cyclopentadienyl-metal oxo peroxo complexes of both molybdenum and tungsten, Cp'M(0)(r? -0 )R, are described. Furthermore, reactions of representative 2  2  examples of these complexes with PPh and tetracyanoethylene are described. The latter 3  reactions afford novel charge-transfer complexes [Cp'W(0)(r? -02)R]2-^[(CN) C=C(CN) ] 2  2  2  (R = CH , CH SiMe ), and a discussion of the preparation, characterization, X-ray crystal 3  2  3  structure and a UV-vis absorption study of these complexes constitutes the last part of this Chapter. In Chapter 6 reactions of representative members of the cyclopentadienyl dioxo alkyl complexes, Cp'M(0) R, with />-tolylisocyanate (p-MeC H N = C = 0) are described. 2  6  4  Rather than oxo imido complexes of the type Cp'M(0)(=NR')R, other complexes resulting from the incorporation of three />-MeC H N = C=0 molecules have been isolated, and the 6  4  physical properties and molecular structures of the corresponding Cp'M( = NC H - p-Me)(N(C H -p-Me)C(0)-N(C H - p-Me)]R complexes are delineated. 6  4  j  6  4  6  4  j  17  References and Notes:  Bottomley, F.; Sutin, L. Adv. in Organomet. Chem. 1988,28,339. (a) Herrmann, W. A. Angew. Chem. Int. Ed. Engl 1988,27,1297, and references therein, (b) Herrmann, W. A. /. Organomet. Chem. 1986,300, 111. Kochi, J. K.; Sheldon, R. A. Metal-Catalyzed Oxidations of Organic Compounds. Academic Press, Inc.; New York: 1981. Jira, R.; Sedlmeier, J . ; Smidt, J. Justus Liebigs Ann. Chem. 1966,693, 99 and references therein. Moiseev, 1.1.; Vargaftik, M. N.; Syrian, Y. Proc. Acad. Set USSR, Phys. Chem. Sect. 1960,133, 801. (a) Miller, S. A. in "Ethylene and Its Industrial Derivatives" (Miller, S. A., ed.), pp 946-956. Benn, London, 1969. (b) Krekeler, H . ; Kronig, W. Proc. 7  t h  World Pet.  Congr. 1967,5,41. Rappe, A. K.; Goddard, W. A. /. Am. Chem. Soc. 1980,102,5114.  Cytochrome P-450: Structure, Mechanism, and Biochemistry; Ortiz de Montella R., Ed.; Plenum: New York, 1986. (a) Fischer, E. O.; Vigoureux, S. Chem. Ber. 1958, 91,1342. (b) Fischer, E. O.; Vigoureux, S.; Kuzel, P. Chem. Ber. 1960,93,701. As we shall see later, this type of reaction has become the most common method for the preparation of cyclopentadienyl oxo complexes. Carrick, W. L . ; Reichle, W. T.; Pennella, F.; Smith, J. J. /. Am. Chem. Soc. 1960, 82,3887. For examples of Group 4 metal oxo complexes (Ti, Zr, and Hf) see : (a) Corrandini, P.; Allegra, G. /. Am. Chem. Soc. 1959,81,5510. (b) Samuel, E. Soc.  18  Bull Chim. Fr. 1966, 3548. (c) Reid, A. F.; Shannon, J. S.; Swan, J. M . ; Wailes, P. C. AustJ. Chem. 1965,18,173. (d) Brainina, E. M . ; Freidlina, R. Kh.; Nesmeyanov, A. N. Dokl Akad. Nauk. SSSR 1964,154, 143. For examples of Group 5 metal oxo complexes (V, Nb) see : (a) Treichel, P. M . ; Werber, G. P. /. Am. Chem. Soc. 1968,90,1753. (b) Mowat, W.; Shortland, A . ; Wilkinson, G . ; Yagupsky, G. /. Chem. Soc. Chem. Commun. 1970,1369. (c) Hill, N. J . ; Mowat, W.; Shortland, A . ; Wilkinson, G; Yagupsky, G . ; Yagupsky, M. /.  Chem. Soc. Dalton Trans. 1972, 533. (d) Schrock, R. R. /. Am. Chem. Soc. 1976, 9 5399. (e) Santini-Scampucci, C.; Riess, J. G. /. Chem. Soc., Dalton Trans. 1974, 1433. For examples of Group 6 metal oxo complexes (Cr, Mo and W) see : (a) Fischer, E. O.; Ulm, K.; Fritz, H. P. Chem. Ber. 1960, 93, 2167. (b) Bottomley, F.; Paez, D. E . ; White, P. S. /. Am. Chem. Soc. 1981,103, 5581. (c) Cousins, M . ; Green, M. L. H. /. Chem. Soc. 1963, 889. (d) Cousins, M . ; Green, M. L. H. /. Chem. Soc. 1964,1567. (e) Cousins, M . ; Green, M. L. H. /. Chem. Soc A 1969,16. (f) Santini-Scampucci, C.; Riess, J. G. /. Organomet. Chem. 1974, 73, C13. (g) Santini-Scampucci, C.; Riess, J. G. /. Chem. Soc, Dalton Trans. 1976, 195. (h) Heyn, B. ;Hoffman, R. Z Chem. 1976,16,195. (i) Hughes, L . ; Grate, J. H . ; Moorehead, E. L . ; Schrauzer, G. N. /. Am. Chem. Soc 1978,100,4760. 0) Hughes, L . ; Schrauzer, G. N.; Strampach, N. Z. Naturforsch, 1982,37b, 380. (a) Mertis, K.; Williamson, D. H . ; Wilkinson, G. /. Chem. Soc, Dalton Trans. 1975, 607. (b) Mertis, K.; Wilkinson, G. /. Chem. Soc, Dalton Trans. 1976,1488. (c) Green, J. C.; Lloyd, D. R.; Galyer, L . ; Mertis, K.; Wilkinson, G. /. Chem. Soc, Dalton Trans. 1978,1403. (d) Galyer, L . ; Mertis, K.; Wilkinson, G. /. • Organomet. Chem. 1975,85, 031. (a) Bokiy, N. G . ; Gatilov, Yu. V . ; Struchkov, Yu. T.; Ustynyuk, N. A- /.  19  Organomet. Chem. 1973,54, 213. (b) Davidson, J. L . ; Green, M . ; Sharp, D. W. A.; Stone, F. G. A . ; Welch, A. J. /. Chem. Soc., Chem. Commun. 1974, 706. (c) Braterman, P. S.; Davidson, J. L . ; Sharp, D. W. A. /. Chem. Soc, Dalton Trans. 1976, 241. (d) Howard, J. A. K . ; Stanfield, R. F. D.; Woodward, P. /. Chem. Soc, Dalton Trans. 1976, 246. (17)  Atwood, J. L . ; Bloom, I.; Evans, W. J . ; Grate, J. W.; Hunter, W. E. /. Am. Chem. Soc. 1985,107,405.  (18)  Organometallic Chemistry ofthef-Elements, Fischer, R. D.; Marks, T. J. Eds.: Reidel; Dordrecht, 1979, pp. 1-35.  (19)  Alves, A. S.; Anderson, R. A . ; Moore, D. S.; Wilkinson, G. Polyhedron 1982,1, 83.  (20)  Behling, T.; Edwards, P. G . ; Hursthouse, M. B.; Motevalli, M . ; Stravropoulos, P.; Wilkinson, G. /. Chem. Soc, Dalton Trans. 1987, 169.  (21)  Hursthouse, M. B.; Motevalli, M . ; Tooze, R. P.; Wilkinson, G. /. Chem. Soc, Dalton Trans. 1986, 2711.  (22)  It should be noted that there are many well characterized inorganic compounds with terminal, multiply bonded oxo ligands for the entire group 7 and 8 metals see: Cotton, F. A.; Wilkinson, G. Advanced Inorganic Chemistry; 4th Ed.; Wiley Interscience: Toronto, 1980.  (23)  Bercaw, J. E . ; Hillhouse, G. L. /. Am. Chem. Soc. 1984,106,5472.  (24)  Bercaw, J. E . ; Burger,B. J . ; Gibson, V. C.; van Asselt, A. /. Am. Chem. Soc. 1986, 108, 5347.  (25)  Critchlow, S. C.; Erikson, T. K. G . ; Mayer, J. M . ; Spaltenstein, E. /. Am. Chem. Soc. 1989, 111, 617.  (26)  Cooper, C.; Geib, S. J . ; Mayer, J. M . ; Rheingold, A. L . ; Su, F.-M. /. Am. Chem. Soc. 1986,108,3545.  20  (a) Ricard, L, ; Weiss, R. Inorg. Nucl Chem. Lett. 1974,10, 217. (b) Corbin, J. L . ; McDonald, J. W.; Newton, W. E . ; Ricard, L . ; Weiss, R. Inorg. Chem. 1980,19, 1997. (a) Calabrese, J. C.; Mayer, J. M . ; Tulip, T. H . ; Valencia E. /. Am. Chem. Soc. 1987,109,157. (b) Mayer, J. M . ; Thorn, D. L . ; Tulip, T. H. /. Am. Chem. Soc. 1985,107, 7454. (c) Mayer, J. M . ; Tulip, T. H. /. Am. Chem. Soc 1984,106,3878. (a) Fellmann, J. D.; Messerle, L. W.; Rupprecht, G. A . ; Schrock, R. R. /. Am. Chem. Soc. 1980,102,6236. (b) McLain, S. J . ; Schrock, R. R.; Wood, C. D. /. Am. Chem. Soc. 1979,101,3210. (c) Schultz, A. J.; Williams, J. M . ; Fellmann, J. D.; Rupprecht, G. A . ; Schrock, R. R. /. Am. Chem. Soc. 1979,101,1593. (d) Churchill, M. R.; Youngs, W. J. Inorg. Chem. 1979,18,1930. (e) Churchill, M. R;; Hollander, F. J. Inorg. Chem. 1978,17,1957. Schrock, R. R. Acc. Chem. Res. 1986,19,342, and references therein. (a) Freudenberger, J. H . ; Schrock, R. R. Organometallics 1985,4,1937 and references therein, (b) Clark, D. N.; Pedersen, S. F.; Rocklage, S. M . ; Sancho, J.; Schrock, R. R.; Wengrovius, J. H. Organometallics 1982,1,1645 and references therein, (c) Kress, J. R. M . ; Le Ny, J.-P.; Osborn, J. A . ; Wesolek, M. G. /. Chem. Soc, Chem. Commun. 1981,1039. (d) Kress, J. R. M . ; Osborn, J. A . ; Russell, M. J. M . ; Wesolek, M. G. /. Chem. Soc, Chem. Commun. 1980,431. Cai, S.; Hoffman, D. M . ; Wierda, D. A. /. Chem. Soc, Chem. Commun. 1988,1489. (a) Schrock, R. R.; Wengrovius, J. H . ; Churchill, M. R.; Missert, J. R.; Youngs, W. J. /. Am. Chem. Soc. 1980,102,4515. (b) Wengrovius, J. H . ; Schrock, R. R. Organometallics 1982,1,148. (a) Legzdins, P.; Phillips, E. C.; Sanchez, L. Organometallics 1989,8, 930. (b) Legzdins, P.; Phillips, E. C.; Rettig, S. J . ; Sanchez, L . ; Trotter, J . ; Yee, V. C. Organometallics 1988, 7,1877. (c) Faller, J. W.; Ma, Y. Organometallics 1988, 7, 558.  21  Bercaw, J. E . ; Henling, L. M . ; Trimmer, M. S.; van Asselt, A. /. Am. Chem. Soc. 1988,110, 8254. Antinolo, A . ; Martinez de Darduya, J . ; Otero, A . ; Royo, P.; Manotti Lanfredi, A. M . ; Tiripicchio, A. /. Chem. Soc, Dalton Trans. 1988,2685. (a) Peterson, J. L. /. Organomet. Chem. 1979,166,179. (b) Herberhold, M . ; Kremnitz, W.; Kuhnlein, M . ; Brunn, K.; Ziegler, M.L. Z. Naturforsh. 1987,42b, 1520. For examples of complexes of the type [M(^-0) M] (M = Ti, Mo, Re) see : (a) 2  Arzoumanian, H . ; Baldy, A . ; Petrignani, J.-F.; Pierrot, M. /. Organomet. Chem. 1985,294,321. (b) Couldwell, C.; Prout, K. Acta Crystallogr. 1978, B34, 933. (c) Bottomley, F.; Egharevba, G. O.; Lin, I. J. B.; White, P. S. Organometallics 1985, 4,550. (d) Huggins, J. M . ; Lebioda, L . ; Whitt, D. R. /. Organomet. Chem. 1986, 312, C15. (e) Felixberger, J. K.; Herdtweck, E . ; Herrmann, W. A . ; Kuchler, J. G.; Wagner, W. Angew. Chem. Int. Ed. Engl 1988,27,394. (f) Guggolz, E . ; Herrmann, W. A . ; Kiisthardt, U . ; Serrano, R.; Zahn, T.; Ziegler, M. L. Angew. Chem. Int. Ed Engl. 1984,23, 515. (g) Guggolz, E . ; Herrmann, W. A . ; Nuber, B.; Kiisthardt, U . ; Serrano, R.; Ziegler, M. L. /. Organomet. Chem. 1985,287,329. (h) Fidel, M . ; Herdtweck, E . ; Herrmann, W. A . ; Kulpe, J . ; Kiisthardt, U . ; Okuda, J. Polyhedron 1987,6,1165. Faller, J. W.; Ma, Y. /. Organomet. Chem. 1988,340,59. Mayer, J. M. Inorg. Chem. 1988,27,3899. see also references 41 (a) and (b). (a) Herberhold, M . ; Kremnitz, W.; Razavi, A . ; Schollhorn, H . ; Thewalt, U. Angew. Chem. Int. Ed Engl 1985,24, 601. (b) Feinstein-Jaffe, I.; Gibson, D . ; Lippard, S. J . ; Schrock, R. R.; Spool, A. /. Am. Chem. Soc. 1984,106,6305. Alt, H. G . ; Hayen, H. I. /. Chem. Soc, Chem. Commun. 1987,1795. (a) Klahn-Oliva, A. H . ; Sutton, D. Organometallics 1984,3,1313. (b) Bock, H . ; Herrmann, W. A . ; Serrano, R. Angew. Chem. Int. Ed Engl 1984,23,383. (c)  22  Fidel, M . ; Herrmann, W. A . ; Voss, E. /. Organomet. Chem. 1985,297, C5. (44)  Bottomley, F.; Egharevba, G. O.; Lin, I. J. B.; White, P. S. Organometallics 1985, 4, 550.  (45)  Bottomley, F.; Darkwa, J . ; Sutin, L . ; White, P. S. Organometallics 1986,5, 2165.  (46)  Alt, H. G . ; Hayen, H. I. Angew. Chem. Int. Ed. Engl 1985,24,497.  (47)  Alt, H. G . ; Baker, E. C.; Fronczek, F. R.; Sharp, P. R.; Rausch, M. D.; Raymond, K. N. Inorg. Chem. 1976,15, 2284.  (48)  (a) Green, M. L. H . ; Lynch, A. H . ; Swanwick, M. G. /. Chem. Soc, Dalton Trans. 1972,1445. (b) Chiang, M. Y . ; Silavwe, N. D.; Tyler, D. R. Inorg. Chem. 1985,24, 4219.  (49)  Muetterties, E. L . ; Band, E. /. Am. Chem. Soc. 1980,102, 6572.  (50)  Edwards, P. G . ; Hursthouse, M. B.; Malik, M. A . ; Wilkinson, G. /. Chem. Soc, Dalton Trans. 1980,2467.  (51)  Dewan, J. C.; MacLaughlin, S. A . ; Murray, R. C.; Schrock, R. R. Organometallics 1985,4, 796.  (52)  Floel, M . ; Herdtweck, E . ; Herrmann, W. A . ; Kulpe, J . ; Kiisthardt, U . ; Voss, E. /. Organomet. Chem. 1986,314,151.  23  Chapter 2  New Types of Organometallic Oxo- and Dioxo-Alkyl Complexes of Molybdenum and Tungsten  1  24  Introduction As discussed in the previous chapter, the study of high oxidation state organometallic oxo compounds has been one of the fastest growing areas of research in inorganic chemistry since the early 1980's. The renewed interest in these oxo compounds results in part from the anticipation that a detailed understanding of their characteristic chemistry may provide some insight into how metal oxides function as catalysts for organic reactions and as oxidizing agents for organic compounds.  2  My interest in organometallic oxo complexes stems from some of the original investigations of the chemical properties of CpW(NO)(CH2SiMe3)2. During these studies, 3  Luis Sanchez, a post-doctoral research fellow working in our laboratories, encountered the first examples of three new types of cyclopentadienyl oxo alkyl complexes, namely 1-3 depicted below.  4  Oy^VcHjSiMe, ^/\ ^n 3nvie 2  f 2 CH SiMe  O ^ f ^I C^CHSiMe HSiMe,  3  3  V 2  H  -  2  SiMe  H  -  A*  3  s  SiMe  3  3  The initial discoveries of complexes 1-3 were somewhat serendipitous, to say the least. It was found that during the synthesis of the complex CpW(NO)(CH SiMe3)2, eq. 1, 2  admission of air into the reaction mixture caused the formation of small quantities of complex 1. 1. 2Me3SiCH MgCl/Et20 2  CpW(NO)I  2  2. H 0 2  —^2 CpW(NO)(CH SiMe )2 + CIMgl 2  3  (1)  25  The trialkyl-oxo compound 2 was isolated as a by-product in very low yield (=! 3%) from reaction 1. The oxo alkylidene compound 3 resulted from the thermal decomposition of the parent dialkyl nitrosyl compound in the solid state. Nevertheless, I was sufficiently 3  intrigued by these classes of compounds to embark on a study of their chemistry. The work described in this chapter, therefore, began as a direct continuation of the above studies. The first objective was to prepare a variety of dioxo alkyl complexes of tungsten with various alkyl groups, and then, possibly extend the chemistry to include the molybdenum analogues, i.e. eq. 2.  C_ or H2O2 Cp'M(NO)R  •  2  Cp'M(0) R 2  (2)  [M= MoorW]  In order to accomplish this, many new cyclopentadienyldialkyl nitrosyl starting materials, Cp'M(NO)R2 (Cp'= Cp or Cp*), had to be prepared for the first time (especially in the case for M= Mo). As a result, this chapter begins with a description of the method for the preparation of the molybdenum and tungsten Cp'M(NO)R2 complexes. Investigations into the chemical properties of the Cp'Mo(NO)(CH2SiMe3)2 compounds has resulted in the isolation of new types of cyclopentadienylnitrosyl oxo bridged dimolybdenum species, namely [Cp'Mo(NO)(CH SiMe3)]20i-O). Hence, a discussion of the preparation, 2  characterization, and the solid-state molecular structures of the new Cp'M(NO)R2 and [Cp'Mo(NO)(CH SiMe3)]20i-O) complexes constitutes thefirstpart of this chapter. 2  Following this are presented the full details for the preparation and characterization of a series of dioxo alkyl complexes, Cp'M(0) R, for both molybdenum and tungsten with a 2  variety of alkyl ligands. These complexes are successfully prepared from reactions of the Cp'M(NO)R2 complexes employing either C_ or H C_ as the oxidizing agent. 2  The second objective of this study was to determine the mechanism by which the dioxo alkyl complexes are formed from the dialkyl nitrosyl complexes, Cp'M(NO)R2, and  26 air, and possibly to isolate or identify the by-products of the reactions. Results of these investigations are also presented in this chapter. Furthermore, it was hoped that a reliable synthetic route could be found for the preparation of complexes of the type 2 and 3, considered above. Indeed, such a route has been found, but the details concerning this work are presented in Chapter 4.  Experimental Section  All reactions and subsequent manipulations involving organometallic reagents were performed, unless mentioned otherwise, under anhydrous and anaerobic conditions using conventional Schlenk-tube techniques or in a Vacuum Atmospheres Corp. Dri-Lab 5,6  Model HE-43-2 drybox, under an atmosphere of dinitrogen or argon. All solvents were purchased from either BDH Chemicals or Aldrich and were purified according to documented procedures; pentane (Omnisolv.) and hexanes (ACS grade) were distilled 7  from CaH , toluene (ACS grade) was distilled from Na, diethyl ether and tetrahydrofuran 2  (both anhydrous ACS grade) were distilled from Na/benzophenone ketyl, and methylene chloride (spectroscopic grade) was distilled from P 0 . The solvents were freshly distilled 2  5  and deaerated with a dinitrogen atmosphere for approximately 10 min prior to use. All other chemicals were of reagent grade or comparable purity, and were either purchased from commercial suppliers or prepared by published procedures. Standard analytical and spectroscopic techniques were used to ascertain their purity. The samples of dioxygen, 0 , and 0 , used during this work were purchased from Medigas and 16  18  2  2  Matheson in 99.5% purity and 98% isotopic purity, respectively, and were not purified further before use. The hydrogen peroxide reagent was obtained as a 30% by weight aqueous solution from BDH Chemicals. The methanol and f-butyl alcohol were distilled from CaH and stored over 3A molecular sieves. The /?-methylphenol was purified by 2  27 vacuum distillation. The alkyl-Grignard reagents PhCH MgCl, Me PhCCH_MgCl and 2  2  Me CCH MgCl were prepared by the literature procedure, and Me SiCH MgCl was 8  3  2  3  2  purchased from Aldrich as a 1.0 M solution in Et_0. Methyl-lithium, CH Li, was also 3  purchased from Aldrich, but as a 1.4 M solution in Et_0. The CpM(NO)I complexes (M = Mo and W ) and Cp*W(NO)I were 9  10  1 1  12  2  2  prepared by literature methods. The complex Cp*Mo(NO)I was prepared with I in a 2  2  similar manner as the perhydro cyclopentadienyl analogue, and the Cp'M(NO)C_ 13  complexes (Cp'= Cp and Cp*; M= Mo, or W), were synthesized by treating the appropriate Cp'M(CO) (NO) complex with an equimolar amount of PC1 in EtjO. The 14  2  5  purity of the above reagents was checked by elemental analysis before use. Infrared spectra were recorded on a Nicolet 5DX FT-IR instrument internally calibrated with a He/Ne laser. All *H NMR spectra were obtained on a Varian Associates XL-300 or Bruker WP-80 spectrometer with reference to the residual proton signal of the deuterated solvent employed (usually benzene-^). The C NMR spectra were recorded B  at 75 MHz on a Varian Associates XL-300 spectrometer with reference to the C signal of B  the solvent employed. All *H and C chemical shifts are reported in parts per million 1 3  downfield from Me Si. Low-resolution mass spectra were recorded at 70 eV on an Atlas 4  CH4B or a Kratos MS50 spectrometer using the direct-insertion method by the staff of the Mass Spectrometry Laboratory of this Department headed by Dr. G.K. Eigendorf. Elemental analyses were performed by Mr. P. Borda of this Department. The melting point determinations were performed on a Gallenkamp Melting Point Apparatus, and the reported temperatures are uncorrected.  Preparation of the Cyclopentadienylnitrosyl Dialkyl Starting Materials, Cp'M(NO)R2 [Cp' = Cp or Cp*, M = Mo or W, R = CH , CH SiMe , CH CMe , 3  2  3  2  3  CH CMe Ph or CH Ph], All these syntheses involved treating the appropriate 2  2  2  Cp'M(NO)X precursors (Cp' = Cp or Cp*, M = Mo or W, X = CI or I) with the requisite 9  2  amount of either an alkyl-Grignard or an alkyl-lithium reagent at low temperatures. The  28 Cp'W(NO)R complexes (Cp'= Cp, R = CH SiMe , CH CMe , CH CMe Ph or CH Ph; 2  2  3  2  3  2  2  3  2  Cp' = Cp*, R = CH SiMe ) were prepared by their published procedures, and their L5  2  3  purity was checked by elemental analysis and H NMR spectroscopy before use. The new 1  dialkyl nitrosyl complexes used in this work were prepared as outlined below. (a) Preparation of the Cp'Mo(NO)(CH SiMe3) Complexes, [Cp"= Cp or Cp*]. 2  2  The method for the synthesis of CpMo(NO)(CH SiMe ) from CpMo(NO)I and 2  3  2  2  Me SiCH MgCl has been reported previously. Attempts to repeat this synthesis proved 3  3  2  difficult in this work. However, optimum yields of CpMo(NO)(CH SiMe ) were obtained 2  3  2  when this complex was prepared from CpMo(NO)Cl in a manner similar to that outlined 2  below for the preparation of its pentamethylcyclopentadienyl analogue (method I). The full experimental details for the synthesis of Cp*Mo(NO)(CH SiMe ) from 2  3  2  Cp Mo(NO)Cl (method I) and Cp Mo(NO)I (method II) are described below as 2  2  representative examples to illustrate the difficulties that were encountered in the preparation of these complexes. Method I. Reaction of Cp*Mo(NO)Cl with Me SiCH MgCl. 2  3  2  To a rapidly  stirred, brown suspension of Cp*Mo(NO)Cl (2.10 g, 6.32 mmol) in E^O (150 mL), which 2  had been cooled to -20 °C using a saturated CaCl (aq.)/Dry Ice bath, was added 12 mL of a 2  1.0 M E^O solution of Me SiCH MgCl (12 mmol) in a dropwise fashion from an addition 3  2  funnel. Immediately the reaction mixture turned black in color, then slowly became purple as a white precipitate deposited. After being stirred for an additional hour at -20 °C, the reaction mixture was taken to dryness in vacuo (note : no deaerated H Q was added). ?  Under rigorously anaerobic conditions, the remaining purple and white reaction residue was extracted with pentane (2 x 30 mL). The volume of the purple pentane extracts j >  N 0  1639 (w), 1601 (s) cm ] was slowly reduced in vacuo until the first signs of crystallization -1  were evident. Cooling of this mixture to -20 °C overnight induced the crystallization of 0.62 g (25% yield) of purple Cp*Mo(NO)(CH SiMe ) as an extremely air- and water-sensitive 2  3  2  29 solid. The crystals were collected by filtration and dried at 5 x 10'"* mm and 20 °C for 2 h and stored under a N atmosphere at -20 °C. There was no evidence apparent for the 2  formation of [Cp*Mo(NO)I] in this reaction, the major product obtained in method II 2  detailed below. Cp*Mo(NO)(CH SiMe ) ; purple crystals. Anal. Calcd for C H ONSi Mo: C, 2  3  2  lg  37  2  49.63; H, 8.56; N, 3.21. Found: C, 49.47; H, 8.71; N, 3.06. IR (Nujol mull): ! / cm" ; ^ i _  1243 (m) cm" ; IR (hexanes): « /  1  S  X  1  M e  N O  N O  1595 (s)  1601 (s) cm ; ^sj.Mg 1244 (m) cm .  H NMR (C D ) S 2.20 (d, 2H, ^ . H * = 10.8 Hz, 2 C# H SiMe ), 1.49 (s, 15H, 6  6  A  C (C/7 ) ), 0.37 (s, 18H, 2 Si(C# ) ), -1.17 (d, 2H, / H A . X = 2  5  3  5  3  3  x  1 0  H  -  3  8H z  « CH # SiMe ). 2  A  x  3  ^C^H} NMR (C D ) 6 110.52 (s, C (CH ) ), 66.54 (s, C H H ) , 9.97 (s, C (CH ) ), 2.59 6  6  5  3  5  A  X  5  3  5  (s, Si(CH ) ). Low-resolution mass spectrum (probe temperature 100 °C) m/z 437 ( P , +  3  98  3  Mo). Melting point: 72-73 °C. Method II. Reaction of Cp'MofNCOL with Me SiCH_MgCl. To a stirred, 3  purple suspension of Cp*Mo(NO)I (3.00g, 5.82 mmol) in Et_0 (200 mL) at room 2  temperature were added dropwise from an addition funnel 12 mL of a 1.0 M E12O solution of Me SiCH MgCl (12 mmol). The reaction mixture immediately became dark green, and 3  2  then slowly turned purple while a greenish-white precipitate formed. After 1 h, 0.3 mL of deaerated H 0 was added to the mixture. Instantly the green precipitate disappeared, and 2  a white precipitate formed, the supernatant solution remaining purple in color. The solvent was removed under reduced pressure until approximately 10-15 mL remained, at which point the solution was chromatographed on an alumina column (Fisher neutral, 80200 mesh, activity 3, 6 x 3 cm made up in hexanes) and eluted with Et_0. A purple band developed slowly away from a red-brown band that remained at the top of the column. As the purple band eluted down the column it turned brown in color suggesting that some decomposition was occurring. The first fraction of the eluate collected consisted of some of the remaining purple band and its accompanying brown band. The second fraction eluted  30 with CH C1 gave a deep red eluate. Fractional crystallization of these two fractions from 2  2  1:1 Et20:hexanes and CH Cl :hexanes solutions, respectively, at -20 °C led to the isolation 2  2  of 0.05g of Cp*Mo(NO)(CH SiMe ) (5% yield), and 0.08g of 2  3  [Cp*Mo(NO)(CH SiMe )] 0*-O) 2  3  2  2  10% yield) from the first fraction and 0.62g of  [Cp*Mo(NO)I] (27% yield based on Mo) from the second fraction. These complexes had 2  very similar solubilities in the solvents used and were difficult to separate analytically pure from each other by fractional crystallization. [Cp*Mo(NO)I] ; orange-red crystals. Anal. Calcd for C_ H N_O I Mo : C, 2  0  30  2  2  30.95 ; H, 3.90 ; N, 3.61. Found : C, 31.02 ; H, 3.94 ; N, 3.58. IR (Nujol mull) * 1578 (s) cm" ; IR ( C H C l ) r / 1  2  2  1611 (s) cm' . H NMR (CD C1) 1  N O  X  3  S  N  2  0  1599,  2.05 (s, C (Ctf ) ). 5  3  5  Low-resolution mass spectrum (probe temperature 120 °C) m/z 776 (P ). +  The analytical and spectroscopic data for Cp*Mo(NO)(CH SiMe ) are provided in 2  3  2  Method I above, while the data for [Cp*Mo(NO)(CH SiMe )] (/j-0) are presented below 2  3  2  in the section where it was isolated as the major product from the decomposition of Cp*Mo(NO)(CH SiMe ) . 2  3  2  (b) Preparation of the Cp'M(NO)(CH_Ph) Complexes, [Cp'= Cp or Cp*, M= Mo 2  or W]. All of these syntheses were performed in an analogous manner by the treatment of the appropriate Cp'M(NO)X2 precursors (Cp'= Cp or Cp*, M= Mo or W, X= CI or I) with the requisite amount of PhCH MgCl in E^O/THF solvent mixtures at 0 °C. The complete 2  details of the syntheses and the analytical and spectroscopic data for all of these nitrosyl bis(benzyl) complexes are presented in greater detail in Chapter 3 of this thesis. (c) Generation of the Cp'M(NO)(CH ) Complexes [Cp'= Cp or Cp*; M= Mo or 3  2  W]. These four complexes were generated in situ in an analogous manner by treating the appropriate Cp'M(NO)X precursors (Cp' = Cp or Cp*, M = Mo or W, X = CI or I) with 2  31 the requisite amount of MeLi at low temperatures in diethyl ether solutions. For details, see the section below which describes the preparation of the Cp'M(0) Me complexes. 2  Decomposition of the Cp'Mo(NO)(CH SiMe3) Complexes, (Cp'= Cp or Cp*). 2  2  (a) Isolation of [CpMo(NO)(CH SiMe3)] (/i-0). This complex was obtained in 2  2  low yield from the thermal decomposition of CpMo(NO)(CH SiMe )2 in an EtjO/hexanes 2  3  solution. Gradually over a period of a few months, a purple solution of CpMo(NO)(CH SiMe ) , stored under a dinitrogen atmosphere at -20 °C, changed to a red 2  3  2  color. Chromatography of the resulting solution through a Florisil column (6x3 cm) with Et^O as the eluant, caused two bands to develop. The second red band was collected separately from the first purple band (containing undecomposed CpMo(NO)(CH SiMe ) ), and the red eluate was concentrated in vacuo until the first signs 2  3  2  of crystallization appeared. Cooling the resulting mixture to -20 °C overnight resulted in the formation of a red-black crystalline solid, [CpMo(NO)(CH SiMe )] 0j-O). On 2  3  2  occasion, IR and H-NMR spectroscopic characterization of this solid sample showed there J  to be traces of CpMo(NO)(CH SiMe ) as an impurity. In these cases, fractional 2  3  2  recrystallization of the solid material from hexanes was sufficient to separate the two complexes. Anal. Calcd for C H N S i 0 M o : C, 37.76 ; H, 5.63 ; N, 4.89. Found: C, 38.03 ; 18  32  2  2  H, 5.51; N, 5.00. IR (Nujol mull): « / ^Mo-O-Mo  41 NMR  8  1  2  1  (C D )S C  2  1603 (s) cm' ;  i/ .  1  N O  (s) cm . IR (C H ): y 6  C  3  s i  1599 (s, br) o n , * 1  6  N O  5  5  2  B  C  H a  {41} NMR  _ 0  . 812 (s) cm' . 1  M c  2  5  2  3  I. 48 (d, I H , /  M o  Isomer A 5.32 (s, 5H, C H ), 1.80 (s, 2H, Ctf SiMe ), 0.34 (s, 9H,  Si(C// ) ). Isomer B 5.42 (s, 5H, C H ), 2.14 (d, IH, / 3  1258 (sh), 1244 (s) cm' ; 1  M e  -Hx  = 1 0  H a  -Hx  = 1 0 > 6  ^ C# H SiMe ), A  x  3  - H ' CH tt SiMe ), 0.28 (s, 9H, Si(C# ) ). Ratio of A:B = 3:1. 6  2  A  x  3  3  3  Isomer A 104.6 (s, C H ), 39.3 (s, CH SiMe ), 2,23 (s, Si(CH ) );  (C D )8 C  5  3  5  C  5  2  3  3  3  Isomer B 104.5 (s, C H ), 39.2 (s, CH H SiMe ), 1.36 (s, Si(CH ) ). Low-resolution mass 5  5  A  x  3  3  spectrum (probe temperature 150 °C) m/z 558 (P -Me, ^Mo). +  3  32 (b) Isolation of [Cp*Mo(NO)(CH SiMe3)] (p-0). To a rapidly stirred, brown 2  2  suspension of Cp*Mo(NO)Cl (1.18g, 3.56 mmol) in Et20 (150 mL) at room temperature 2  was added dropwise from an addition funnel 8.0 mL of a 1.0 M EtjO solution of Me SiCH MgCl (8 mmol). The initial brown color of the reaction mixture darkened 3  2  immediately, turning black then purple. As more of the Me SiCH MgCl reagent was 3  2  added over 30 min, the reaction mixture became bright purple (i/  1593 cm ) and a white -1  NO  precipitate formed. After lh, approximately 0.2 mL of deaerated H 0 was added to the 2  reaction mixture to destroy the excess Grignard reagent. The solvent volume was reduced to approximately 15 mL by slow evaporation of the EtjO under reduced pressure. The solution was then chromatographed on either an Alumina column (Fisher neutral, 80-200 mesh, activity 3, 6 x 3 cm) or a Florisil column (6x3 cm) made up in Et20 and using EtjO as the eluant. The purple band was slowly eluted down the column, whereupon it changed to a red-brown color. This red-brown band (^  1584 cm ) was collected, and the solvent -1  NO  was removed from the eluate in vacuo until the first signs of crystallization became evident. Cooling the mixture to -20 °C overnight resulted in the formation of 0.65 g of large red-black crystals of [Cp*Mo(NO)(CH SiMe )] fc-0) (51% yield). 2  3  2  Anal. Calcd for C2gH N Si 0 Mo : C, 47.16; H, 7.35; N, 3.93. Found: C, 47.07; 52  2  2  H, 7.64; N, 3.85. IR (Nujol mull): *  2  1560 (s) cm" ;  = 773 (s) cm' . IR (E^O): v  1  2  5  Me  Mo  Q  J  6  (s, 30H, C (Ctf ) ), 0.84 (d, 2H, / _ H b = 3  1  si  1584 (s, br) cm" . H NMR (C D )5 Isomer A 1.59  NO  5  * _ = 1240 (s) cm" ; v _ _  1  N O  1  M o  3  1 2  Ha  - ^ 3  2  6  Ctf H SiMe ), 0.75 (d, 2H, / 2  A  B  3  H a  .  j ^ - 12.3 Hz, 2 CH # SiMe ), 0.45 (s, 18H, 2 Si(C// ) ); Isomer B 1.71 (s, 30H, A  B  3  3  3  C (C// ) ), 0.417 (s, 18H, 2 Si(C# ) ). Ratio of A:B = 9:1. 5  3  5  3  3  B  C ^H} NMR (C D )5 6  6  Isomer A 112.6 (s, C (CH ) ), 36.5 (s, CH H_SiMe ), 9.96 (s, C (CH ) ), 2.83 (s, 5  3  5  A  3  5  3  5  Si(CH ) ); Isomer B 113.5 (s, C (CH ) ), 36.2 (s, CH H_SiMe ), 10.26 (s, C (CH ) ), 3  3  5  3  5  A  3  5  3  5  3.43 (s, Si(CH ) ). Low-resolution mass spectrum (probe temperature 120 °C) m/z 713 3  (P -2, Mo). +  98  3  33 Preparations of the Cp'M(0) R Complexes (Cp' = Cp or Cp*, M = Mo or W, R = 2  CKjSiMej, CH CMe3, or CH CMe Ph) by Employing Dioxygen. All of these syntheses 2  2  2  were carried out in a similar manner and involved treating the appropriate Cp'M(NO)R  2  precursors (Cp' = Cp or Cp*, M = Mo or W, R = CH SiMe , CH CMe , or CH_CMe Ph) 2  3  2  3  2  with an excess of dioxygen. The experimental procedures for the preparation of the Cp'M(0) R complexes vary only in the reaction solvent used; the method for the synthesis 2  of CpW(0) (CH SiMe ) is presented below as a representative example. 2  2  3  Analytically pure CpW(NO)(CH SiMe ) (0.906 g, 2.00 mmol) was dissolved in 2  3  2  hexanes (20 mL) at room temperature, and most of the dinitrogen atmosphere above the resulting dark purple solution was removed in vacuo. An excess of molecular dioxygen was then introduced into the reaction vessel until a pressure of 10 psig was attained. The reaction vessel was sealed, and the reaction mixture was stirred, whereupon a color change from purple to red occurred within 1 h. After 6 h, the final reaction mixture consisted of a fine brown precipitate and a red-brown supernatant solution. The volume of the final mixture was reduced to approximately 10 mL under reduced pressure, and the brown precipitate was collected by cannula-filtration, washed with cold (0 °C) hexanes (15 mL), and dried at 5 x 10" mm. Recrystallization of this solid from Et_0 at -20 °C afforded 0.41 g 3  (56% yield) of CpW(0) (CH SiMe ) as an analytically pure, white crystalline solid. 2  2  3  The analytical, mass spectral, IR and H and C NMR data for 1  B  CpW(0) (CH SiMe ) and all the other new dioxo alkyl complexes synthesized during this 2  2  3  work are presented in Tables 2-III and 2-IV. Using a method similar to that described above, white CpW(0) (CH CMe ) (0.146 2  2  3  g, 46% yield), CpW(0) (CH CMe Ph) (0.115 g, 29%? yield), CpMo(0) (CH SiMe ) (0.32 2  2  2  2  2  3  g, 67% yield), and Cp*W(0) (CH SiMe ) (0.46 g, 65% yield) were prepared from their 2  2  3  appropriate dialkyl nitrosyl starting materials in hexanes or Et_0 solutions. Interestingly, if Cp*W(NO)(CH SiMe ) (1.0 g, 1.9 mmol) was reacted with dioxygen in Et_0 (15 mL) in 2  3  2  the presence of 0.2 mL of H 0,0.71 g of Cp*W(0) (CH SiMe ) (85% yield) was obtained. 2  2  2  3  34 Preparation of Cp*Mo(0) (CH SiMe3). 2  Freshly prepared and analytically pure  2  purple Cp*Mo(NO)(CH SiMe ) (1.0 g, 1.9 mmol) in Etp (10-20 mL) was reacted with an 2  3  2  atmosphere of 0 at room temperature, in a manner similar to that described above. 2  Extraction of the final reaction mixtures with E^O, and subsequent crystallization resulted in the co-crystallization of two colored products. The two types of crystals were physically separated from each other. The major product was pale-yellow Cp*Mo(0) (CH SiMe ) 2  2  3  (0.35 g, 53% yield), its analytical and spectroscopic data being presented in Tables 4-1 and 4-II. The minor product was yellow-brown crystalline [Cp*Mo(0) ] (/i-0) (0.06 g, 12% 2  2  yield). [ 4 1 NMR spectra of crude reaction mixtures from these reactions in C D 6  6  established the relative yields of Cp*Mo(0) (CH SiMe ) to [Cp*Mo(O) ] 0u-O) to be 2  2  3  2 2  _2:1.] [Cp*Mo(O) ] 0i-O); brown crystals. Anal. Calcd for C^H^C^Mo^ C, 44.45; H, 2 2  5.60. Found: C, 44.31; H, 5.67. IR (Nujol mull): ^  910, 878 (s) cm" ; 1  M  o  =  0  f  M o  _ -Mo  7  6  2  0  (br) cm" . H NMR (C D ) 6 1.78 (s, C(C// ) ). Low-resolution mass spectrum (probe 1  J  6  6  3  5  temperature 120 °C) m/z 526 (P -0). +  If the Cp*Mo(NO)(CH SiMe ) starting material was allowed to remain as a solid 2  3  2  or in solution for a long period of time, i.e. 1-2 days, the subsequent reaction with dioxygen gave yellow-brown [Cp*Mo(0) ] u-0) in higher isolated yields (-20-30%). 2  2v  Preparation of Cp'W(0) (CH Ph) (Cp' = Cp or Cp*). 2  2  These complexes were  prepared in a manner similar to that described above. Reaction of CpW(NO)(CH Ph) 2  2  (0.09 g, 0.19 mmol) with excess molecular dioxygen in toluene (10 mL) gave 0.04 g of CpW(0) (CH Ph) (55% yield) as a white crystalline solid after extraction of the final 2  2  reaction mixtures with EtjO and crystallization at -20 °C. In a similar manner, reaction of Cp*W(NO)(CH Ph) (0.36 g, 0.68 mmol) with excess 0 in E^O afforded 0.078 g of 2  2  2  brownish white crystals of Cp*W(0) (CH Ph) (26% yield). 2  2  35  Preparations of the Cp'M(0) R Complexes by Employing Hydrogen Peroxide, 2  H 0 . In general, these conversions were more rapid than those involving 0 , usually 2  2  2  being complete in less than 1 h and producing comparable isolated yields of the CpM(0) R 2  products. However, the stoichiometry of the oxidizing agent had to be carefully controlled in order to avoid further conversion of the desired CpM(0) R compounds to their 2  CpM(f? -0 )(0)R derivatives (see Chapter 5). Again, the preparation of 2  2  CpW(0) (CH SiMe ) is described below as a representative example. 2  2  3  To a stirred, purple solution of CpW(NO)(CH SiMe ) (1.20 g, 2.65 mmol) in E _ 0 2  3  2  (50 mL) was added a 30% by weight aqueous solution of H 0 (0.22 mL, 2.8 mmol H 0 ) 2  2  2  2  by microsyringe. The initial purple color of the reaction mixture faded over the course of 1 h to a pale yellow. An IR spectrum of the final yellow solution was devoid of absorptions due to the nitrosyl reactant. Volatiles were removed from the final reaction mixture under reduced pressure to obtain a sticky yellow solid which was dried at 20 °C and 5 x 10" mm 3  for 2 h. Recrystallization of the resulting pale yellow solid from 1:1 E^O/hexanes at -20 °C afforded 0.50 g (51% yield) of CpW(0) (CH SiMe ) as a white microcrystalline solid. 2  2  3  When the pentamethylcyclopentadienyl analogue, Cp*W(0) (CH SiMe ), was 2  2  3  prepared in a manner similar to that described in the preceding paragraph for the cyclopentadienyl species, the oxo peroxo complex, Cp W(0)(r? -O )(CH SiMe ), was also z  2  3  obtained (see Chapter 5) as a by-product in ca. 10% yield. The two organometallic oxo complexes could be conveniently separated by chromatography on Florisil with E^O as eluant (vide infra). When the analogous reaction of Cp*Mo(NO)(CH SiMe ) (0.16 g, 0.37 mmol) with 2  3  2  a 30% by weight aqueous solution of H 0 (0.03 mL, 0.38 mmol H 0 ) was performed, 2  2  2  2  0.02 g of yellow-brown crystals of [Cp*Mo(0) ] (/i-0) (20% yield based on Mo) were 2 2  isolated, and no Cp*Mo(0) (CH SiMe ) was obtained. 2  2  3  36 Preparations of the Cp'M(0) Me Complexes. All these syntheses involved first 2  treating the appropriate Cp'M(NO)X2 precursors (Cp* = Cp or Cp*, M = Mo or W, X = 9  CI or I) with the requisite amount of MeLi at low temperatures to generate Cp'M(NO)Me  2  in situ, and then exposing these dialkyl complexes to H 0 . The representative 2  2  experimental procedure employed to synthesize Cp*Mo(0) Me was as follows. 2  A stirred suspension of Cp*Mo(NO)Cl (1.2 g, 3.61 mmol) in Et_0 (250 mL) was 2  cooled to -78 °C using a Dry Ice/acetone bath. Dropwise addition of a 1.4 M solution of MeLi in Et_0 (5.5 mL, 7.7 mmol, diluted with 10 mL E^O) to the cooled mixture resulted in a gradual color change to orange-red over the course of 1 h as the organometallic reactant was consumed. The cold bath was then removed, and the reaction mixture was permitted to warm slowly to approximately -35 °C using a saturated CaCl (aq.)/Dry Ice 2  bath. Gradually an orange white precipitate formed and the resulting solution was red in color. Then a 30% by weight aqueous solution of H 0 (0.2 mL, 0.5 mmol) was injected 2  2  into the reaction mixture by syringe. Immediately the solution changed color, becoming brown. After being stirred for 1 h at 20 °C, the final orange solution was first concentrated in vacuo to ca. 10 mL and then chromatographed on a Florisil column (2x4 cm) using E^O as the eluant. The pale yellow band that developed was eluted from the column and collected. The eluate was slowly concentrated under reduced pressure until white crystals began to deposit. The mixture was then maintained at -20 °C overnight to complete the crystallization. The white crystals were collected by filtration the next day to obtain 0.53 g (52% yield) of analytically pure Cp*Mo(0) Me. 2  The congeneric Cp*W(0) Me compound was prepared in a similar manner in 50% 2  isolated yield from a solution of Cp*W(NO)Me (generated from Cp*W(NO)Cl and 2  2  MeLi). The yields of the analogous CpM(0) Me (M = Mo or W) complexes were only in 2  the range of 10-15%, however, since appreciable amounts of the CpM(77 -C_)(0)Me 2  compounds were also formed as by-products. Use of dioxygen in place of hydrogen peroxide did not afford higher yields of the desired dioxo methyl products.  37 Preparation of Cp*W(0) Cl from the Reaction of Cp*W(NO)Cl with 0 . An 2  2  2  atmosphere of molecular oxygen was passed over a stirred, olive-green solution of Cp*W(NO)Cl (1.0 g, 2.6 mmol) in CH C1 (20 mL) at room temperature for 24 h, after 2  2  2  which time all the volatiles were removed in vacuo. The resulting brown solid residue was extracted with Et^O (3 x 10 mL). The yellow-brown extracts were concentrated and cooled to -20 °C overnight to induce the precipitation of 0.21 g of yellow-brown crystals of Cp*W(0) Cl (21% yield). 2  Reaction of [Cp*Mo(NO)(CH SiMe3)] 0i-O) with O 2  2  r  Dark red-black  [Cp*Mo(NO)(CH SiMe )] (/i-0) in hexanes (25 mL) at room temperature was reacted 2  3  2  with an atmosphere of 0 in a manner similar to that described above for the preparation 2  of the CpM(0) R complexes. Extraction of the final reaction mixtures with E^O, and 2  subsequent crystallization gave yellow-brown crystals of [Cp*Mo(O) ] 0i-O) which were 2 2  characterized by their known analytical and spectroscopic data, see above. Thermolysis of CpW(0) (CH SiMe ). A clear, colorless solution of 2  2  3  CpW(0) (CH SiMe ) (0.33 g, 0.90 mmol) in dioxane (80 mL) was heated in an oil bath to 2  2  3  125 °C for 3 days, after which time the dioxane was removed in vacuo. A bluish-white residue remained and was extracted with E^O. A sample of the extracts was subjected to thin layer chromatography on Silica gel 60 (Merck) with E^O, which showed that two compounds were present. Flash chromatography was then used to separate the two compounds. The first major compound separated was the starting material CpW(0) (CH SiMe ), whose identity was established by comparing its rf-value to 2  2  3  authentic CpW(0) (CH SiMe ) and by H NMR and mass spectroscopy. The second X  2  2  3  minor compound isolated was CpW(0) (CH ) which was characterized by its IR (Nujol 2  mull), H NMR and mass spectra. 1  3  38 Reaction of CpW(NO) (CH SiMe3) with PhNO. 2  A stirred, purple solution of  2  CpW(NO)(CH SiMe ) (0.53 g, 1.17 mmol) in E^O (30 mL) was cooled to -78 °C using a 2  3  2  Dry Ice/acetone bath. A solution containing PhNO (0.125 g, 1.17 mmol) in E^O (60 mL), which had also been cooled to -78 °C, was added to the reaction mixture by cannulation. After 0.5 h with no apparent reaction having occurred, the Dry Ice/acetone bath was removed and replaced with a saturated CaCl (aq)/Dry Ice bath (approximate temperature 2  of -40 °C). The solution was stirred for an additional 0.5 h, whereupon it slowly changed to light red. The solution was concentrated to approximately 10 mL in volume, and hexanes (20 mL) were added. Solvent was removed in vacuo until the first signs of crystallization occurred, and the resulting mixture was cooled to -20 °C overnight. This induced the formation of 0.110 g of yellow-orange crystals of CpW(0) (CH SiMe ) (26% yield) which 2  2  3  were identified by comparison of their IR and *H NMR spectra with those of an authentic sample. Reactions of the Cp'W(NO)(CH SiMe ) Complexes (Cp' = Cp or Cp*) with 2  3  2  Alcohols. All of these reactions were performed in a similar manner. A microsyringe was used to add the dry and deaerated alcohol (MeOH, BuOH, or />-MePhOH) to deuterated l  solvent solutions of the appropriate nitrosyl dialkyl complex, Cp'W(NO)(CH SiMe ) . 2  3  2  Similar reactions involving the Cp'W(NO)(CH Ph) complexes are outlined in the next 2  2  chapter. (a) Reaction of CpW(NO)(CH SiMe3) with MeOH. To a purple solution of 2  2  CpW(NO)(CH SiMe ) (0.05 g, 0.11 mmol) in C D (1 mL) was added MeOH (30/*L, 0.22 2  3  2  6  6  mmol) by using a microsyringe. The resulting mixture was cannulated into an NMR tube, and the tube was then flame-sealed under a slight vacuum. The ensuing reaction was monitored by *H and C NMR spectroscopy. New *H resonances attributable to B  CpW(NO)(OMe)(CH SiMe ) appeared at 6 5.31 (s, 5H, C # ), 1.28 (s, 3H, OC// ), 0.33 2  3  5  5  3  (s, 2H, C/f Si), 0.26 (s, 9H, Si(C// ) ), and these resonances slowly increased in intensity as 2  3  3  39 the H signals due to the CpW(NO)(CH SiMe ) starting material decreased in intensity. 1  2  3  2  Also, a proton resonance at 0.00 ppm assigned to liberated Si(CH ) grew in as the 3  4  reaction proceeded. After 4 h, the H resonances of CpW(NO)(CH SiMe ) appeared to J  2  3  2  be one-half their original intensity. After 11 h, *H signals at 6.7 (m, 4H) and 2.68 (pentet, 2H) showed that CpH (C H ) was being liberated. After 3 d, a H NMR spectrum of the 1  5  6  sample showed only signals attributable to CpH, TMS and excess MeOH. (b) Reaction ofCp*W(NO)(CH SiMe ) with MeOH. A C D solution of 2  3  2  6  6  Cp*W(NO)(CH SiMe ) (0.060 g, 0.113 mmol) and MeOH (8.0uL, 0.059 mmol) was 2  3  2  flame-sealed in an NMR tube under a slight vacuum. An initial H NMR spectrum of the 1  reaction mixture within 10 min after preparation of the sample showed that a reaction had already taken place. The *H signals of the Cp*W(NO)(CH SiMe ) starting material [ 1.53 2  (d, 1H, 0 / H S i ) , 1.52 (s, 15H, C H ), A  B  5  3  2  0.34 (s, 9H, Si(Ctf ) ), and -1.47 (d, 1H, / 2  3  5  3  H a H  b  =  10.9 Hz, CH //_Si) ] decreased in intensity over the course of 1.5 h while H signals which 1  A  could be assigned to Cp*W(NO)(OCH )(CH SiMe ) (see below) appeared and grew in a 3  2  3  quantitative manner. After 2 weeks the *H and C NMR spectra were unchanged, and B  there was no evidence of decomposition as in the reaction of CpW(NO)(CH SiMe ) with 2  3  2  MeOH described in (a) above. Furthermore, there was no evidence for further reaction of Cp*W(NO)(OCH )(CH SiMe ) with MeOH to afford Cp*W(NO)(OCH ) . 3  2  3  3  2  NMR data for Cp*W(NO)(OCH )(CH SiMe ): 300 MHz *H NMR (C D ) 6 4.68 3  (br, OCH ), 1.57 (s, 5H, C H ), 3  5  5  Ci/ H SiMe ), 0.84 (d, 1H, /  0.88 (d, 1H, /  B  3  3  2  2  A  2  H a H  b  = 11.9 Hz, J = 2  H a H b  =  HW  6  1  L  9  2 /  HW  =  4  2  6  ^  8.3 Hz, CH //_SiMe ), 0.31 (s, 9H, A  3  Si(C# ) ). 75 MHz ^C^H} NMR (C D ) * 111.84 (s, C H ), 50.2 (s, OCH ), 35.7 (s, 3  3  6  6  5  5  3  1 / ^ = H2.0 Hz, C H ^ , 9.47 (s, C (CH ) ), 2.12 (s, Si(CH ) ). (The C NMR spectrum B  5  3  5  3  3  also contains a singlet resonance at 0.00 ppm which can be assigned to liberated Si(CH ) .) 3  4  40  (c) Reaction of CpW(NO)(CH SiMe3) with tBuOH. A Q D solution of 2  2  6  CpW(NO)(CH SiMe ) (0.028 g, 0.062 mmol) and BuOH (20 nL, 0.21 mmol) was flamel  2  3  2  sealed in an NMR tube under a slight vacuum. An initial H NMR spectrum displayed 1  proton signals due only to the two starting materials. The NMR tube was then placed into an oil bath heated to 65 °C. After two weeks, a H NMR spectrum was devoid of signals 1  due to the CpW(NO)(CH SiMe ) starting material. H and C NMR spectroscopy 1  2  3  B  2  confirmed the quantitative formation of the complex CpW(NO)(OCMe )(CH SiMe ). 3  2  3  The H and C NMR spectra remained unchanged even after 12 months. 1  B  NMR data for CpW(NO)(OCMe )(CH SiMe ): 300 MHz *H NMR S 5.40 (s, 5H, 3  C5//5),  1.81 (d, IH, C t f H S i M e , J 1  a  b  3  HaHb  2  3  = 11.1 Hz), 1.37 (s, 9H, OC(Ctf ) ), 0.44 (d, IH, 3  CH // SiMe , ^ H a H t ^ 11-1 Hz), 0.283 (s, 9H, Si(CH ) ). a  b  3  3  3  3  75 MHz ^C^H} NMR 6 104.1  (s, C5H5), 84.2 (s, OC(CH ) ), 32.16 (s, OC(CH ) ), 25.15 (s, C H S i M e , J w = 100 Hz), 1  3  3  3  3  2  3  C  2.81 (s, Si(CH ) ). 3  3  (d) Reaction of CpW(NO)(CH SiMe3) with ^ - M e Q H ^ H . A QDs solution of 2  2  CpW(NO)(CH SiMe ) (0.05 g, 0.11 mmol) and/J-MeQHjOH (70uL, 0.221 mmol) was 2  3  2  flame-sealed in an NMR tube under a slight vacuum. Monitoring the reaction by *H NMR spectroscopy showed that the *H resonances due to the CpW(NO)(CH SiMe ) starting 2  3  2  material slowly decreased in intensity over a 24-hour period. New *H resonances that had grown in could be attributed to free SiMe4 [ 6 0.00 ppm] and CpW(NO)(CH SiMe )(OQH4-/?-Me) [ S 5.38 (s, 5H, C^s), 0.16 (s, 9H, CH Si(C// ) )]. 2  3  2  3  3  An additional set of H signals grew in at 5.56 (s, 5H), and 0.21 (s, 9H) which were not J  assigned.  41 Results and Discussion A.  Preparation of the Cyclopentadienylnitrosyl Dialkyl Starting Materials, Cp'M(NO)R2,  (Cp = Cp and Cp*; M= Mo and W; R= CH , CHjSiMej, CH CMe3, CH CMe Ph, and 3  2  2  2  CH Ph). 2  The first examples of these novel 16-electron cyclopentadienyldialkyl nitrosyl complexes were prepared by Dr. Luis Sanchez in our laboratories. The dialkyl complexes, CpM(NO)R (M= Mo, R= CH_SiMe ; M= W, R= CH SiMe , CH CMe , CH CMe Ph 2  3  2  3  and CH Ph), were prepared as outlined in equations 3 and 4 2  3  2  3  2  2  Treatment of the well  known nitrosyl diiodide complexes with two equivalents of a Grignard reagent, in which the alkyl group is bulky and contains no 0-hydrogens, gives intermediate iso-nitrosyl complexes which have been isolated as red crystalline solids by fractional crystallization, eq. 3.  2 CpM(NO)I  2  Et>0 =  + 4 RMgCl  •  [CpM(NO)R ] MgI Et 0 2  2  2  + 3 MgX^E^O) (X= CI or I)  2  (3)  A partial single-crystal X-ray crystallographic analysis of the prototypal [CpW(NO)(CH SiMe ) ] MgI Et_0 compound has revealed that the magnesium is 3  2  3  2  2  2  coordinated to the oxygens of the nitrosyl ligands. Water, a stronger Lewis base than the oxygen of the nitrosyl ligand, is added to this iso-nitrosyl adduct, eq. 4, to cleave the Lewis acid/base NO-Mg-ON interactions, and the reaction solutions turn from red to purple, the characteristic color of the dialkyl complexes. EUO [CpW(NO)(CH SiMe ) ] MgI Et 0 + 6 H 0 2  3  2  2  2  2  •  2  CpW(NO)R + [Mg(H 0) ] 2  2  6  2+  + 21" + Et_Q  (4)  42 The water also destroys any excess of the Grignard reagent present after the completion of the reaction. Reactions 3 and 4 need not be performed in a stepwise manner, i.e. isolation of the iso-nitrosyl complexes is not required, but may be effected in a sequential manner in situ to obtain the dialkyl nitrosyl complexes. The desired dialkyl nitrosyl complexes are obtained as dark red to purple crystalline solids in 50-60% isolated yields from the final reaction mixtures by fractional crystallization. In order to obtain a more comprehensive series of cyclopentadienyl dioxo alkyl complexes, Cp'M(0) R, the series of 16-electron dialkyl nitrosyl complexes, CpM(NO)R 2  2  (M= Mo, R= CH SiMe ; M= W, R= CH SiMe , CH CMe , CH CMe_Ph and CH Ph), 2  3  2  3  2  3  2  2  needed to be extended. Therefore, discussed below are the syntheses of the new molybdenum and tungsten dialkyl nitrosyl complexes prepared during this work, namely Cp*Mo(NO)(CH SiMe ) , Cp'M(NO)Me and Cp'M(NO)(CH Ph) [Cp'- Cp or Cp*; M = 2  3  2  2  2  2  Mo and W]. Underrigorouslyanaerobic and anhydrous conditions, purple Cp*Mo(NO)(CH SiMe ) can be prepared in modest yields (20-25%) by the treatment of 2  3  2  Cp*Mo(NO)Cl with slightly less than 2 equivalents of the Me SiCH MgCl Grignard 2  3  2  reagent, as outlined in eq. 5.  E t  Cp Mo(NO)Cl  2  + 2 Me SiCH MgCl 3  2  2°  = Cp*Mo(NO)(CH SiMe ) 2  3  2  +  • 2MgCl  (5)  2  The lower isolated yields of Cp*Mo(NO)(CH SiMe ) as compared to the congeneric 2  3  2  dialkyl complex of tungsten reflects to some extent the tedious work-up required to obtain a pure sample of this complex. For instance, water cannot be used to cleave the NO-MgON interactions of the iso-nitrosyl intermediate [Cp*Mo(NO)(CH SiMe ) ] MgI Et20 or 2  3  2  2  2  to destroy the excess of the Grignard reagent in these reactions, since Cp*Mo(NO)(CH SiMe ) is much more sensitive to H 0 than the corresponding dialkyl 2  3  2  2  43 complexes of tungsten. As a result, slightly less than stoichiometric amounts of the Me SiCH MgCl Grignard reagent are used. The chromatography step must be omitted 3  2  because Cp*Mo(NO)(CH SiMe )2 decomposes on alumina and Florisil columns. 2  3  Therefore, the solids obtained from the final reaction mixtures are extracted with hexanes/pentane and are fractionally recrystallized a number of times from pentane at low temperatures in order to obtain analytically pure purple solids of Cp*Mo(NO)(CH SiMe ) . 2  3  2  It is worth noting that Cp*Mo(NO)(CH SiMe ) cannot be prepared in reasonable 2  3  2  yields from the reaction of Cp*Mo(NO)I with 2 equivalents of the Me SiCH MgCl 2  3  2  Grignard reagent. [Recall that this is the method used in eqs 3 and 4 for the preparation of all the previously known dialkyl nitrosyl complexes of tungsten.] Instead, the major product isolated from the reaction in eq. 6, is the red, mono-iodide dimer, [Cp*Mo(NO)I] , which 16  2  has been characterized by elemental analysis and IR, H NMR and mass spectroscopy. J  The desired nitrosyl bis(trimethylsilylmethyl) complex, Cp*Mo(NO)(CH SiMe ) , 2  E  Cp Mo(NO)I + 2 Me SiCH MgCl 2  3  2  t  3  2  2°  — ^ C p Mo(NO)I] (27% yield) 2  + Cp*Mo(NO)(CH SiMe ) 2  3  2  (6)  (5% yield)  + [Cp*Mo(NO)(CH SiMe )] (>-0) (10% yield) 2  3  2  along with its decomposition product, [Cp*Mo(NO)(CH SiMe )] (/i-0) (see below), are 2  3  2  obtained from the final reaction mixtures as two minor products in low yields. The relatively low isolated yields of each product complex in eq. 6 reflects the difficulty of separating these three products from each other in analytically pure form by chromatography and fractional crystallization since they possess similar solubility properties.  44 Repeating the synthesis of the previously reported complex, CpMo(NO)(CH SiMe )2, using CpMo(NO)I as the precursor (eq. 7), affords lower yields 3  2  3  2  of the desired complex than what was reported (=* 10-20% versus 61%). It is very difficult to obtain a pure sample of CpMo(NO)(CH SiMe ) from reactions involving CpMo^O)^, 2  3  2  as illustrated by the fact that elemental analyses are inconsistent from EuO CpMo(NO)X2 + 2 Me SiCH MgCl 3  2  •CpMo(NO)(CH_SiMe ) [X-IorCl] + 2MgX 3  (7)  2  2  each reaction to the next. This may be due to the formation of considerable amounts of [CpMo(NO)I] which may co-crystallize with the desired CpMo(NO)(CH SiMe ) product. 2  2  3  2  In any event, optimum yields of CpMo(NO)(CH SiMe ) are obtained when the complex is 2  3  2  prepared from CpMo(NO)Cl as the reactant in eq. 7. 2  A recent electrochemical study of the reduction behavior of the dihalo complexes Cp'M(NO)X (Cp'= Cp or Cp*; M= Mo or W; X= CI, Br, or I) provides some insight 17  2  into the different reactivity exhibited by the two precursor complexes, Cp'Mo(NO)X2 (X= Cl and I), during their reactions with Grignard reagents. A mechanism involving initial electron transfer from the Grignard reagent to the organometallic complex has been demonstrated, this transfer resulting in the formation of the radical anion 17  [Cp'M(NO)X ] \ The electrochemical studies also showed that the stability of the radical _  2  anion species within a given series of Cp'M(NO)X complexes increases in the order 2  [CpM(NO)I ]"- < [CpM(NO)Br ]-- < [CpM(NO)Cl ]-\ The [Cp*Mo(NO)I ]"- species is 17  2  2  2  2  much less stable than the other [Cp'M(NO)I ] * species and undergoes rapid decomposition _  2  via T loss. If this I" loss occurs before the occurrence of alkyl ligand exchange with the Grignard reagent, then this would explain the relatively high yields of [Cp*Mo(NO)I]  2  formed in reaction 6. These results also suggest that the Cp'M(NO)Cl complexes should 2  be the most suitable reagents for metathesis with Grignard reagents. Indeed, using the  45 Cp'Mo(NO)Cl precursors in reactions 5 and 7 gives the desired Cp*Mo(NO)(CH SiMe ) 2  2  3  2  and CpMo(NO)(CH SiMe ) complexes in reasonable yields, presumably because the 2  3  2  decomposition of the initially formed [Cp'Mo(NO)Q ]** is slow enough that alkyl-for-halide 2  ligand exchange may occur. As a result, all the dialkyl nitrosyl Cp'M(NO)R complexes are 2  now prepared in our laboratories via the dichloro Cp'M(NO)Cl precursors. 2  Other new dialkyl nitrosyl complexes of molybdenum and tungsten have also been prepared during this work. The generation of the much desired Cp'M(NO)Me (Cp'= Cp 2  or Cp*; M= Mo or W) complexes has been accomplished at low temperatures by using a method analogous to that employed in reaction 5. The dichloro nitrosyl compounds are treated in Et^O solutions at low temperatures with exactly 2 equivalents of MeLi (eq. 8).  Cp'M(NO)Cl  + 2 MeLi  9  • "Cp'MXNOJMe," + 2LiCl  (8)  -78 °C Color changes similar to those observed during reactions 3 and 4 occur, but the nitrosyl dimethyl complexes are not, as yet, isolable. Possibly the two methyl groups do not sterically protect the metal center as do the alkyl groups in the complexes containing the CH SiMe , CH CMe , or CH CMe Ph ligands, and decomposition in solution occurs upon 2  3  2  3  2  2  warming of the reaction mixtures to room temperature. Addition of phosphines such as PMe or PPh to the final reaction mixtures of reactions 8 in order to stabilize the 3  3  coordinatively unsaturated Cp'M(NO)Me fragment and generate a more stable 182  electron species, namely Cp'M(NO)Me (L), proved unsuccessful. Nevertheless, color 2  changes occurred that were similar to those observed when the complexes Cp'W(NO)(CH SiMe ) , (Cp'= Cp or Cp*) are treated with PMe to form the isolable 2  3  2  3  Cp'W(NO)(CH SiMe ) (PMe ) species. ' In any event, the nitrosyl dimethyl complexes 3 15  2  3  2  3  of molybdenum or tungsten, Cp'M(NO)Me , may be generated at low temperatures in situ, 2  and, as we shall see later, reactions of these complexes with 0 or H 0 , do afford 2  2  2  46  cyclopentadienyl dioxo methyl complexes. The synthetic utility of these Cp'M(NO)Me2 complexes remains to be explored, however. Similarly, the complexes CpMo(NO)(CH Ph) , Cp*Mo(NO)(CH Ph) and 2  2  2  2  Cp*W(NO)(CH Ph) are preparable in a straightforward manner from reactions of the 2  2  appropriate Cp'M(NO)Cl complex and a benzyl-Grignard reagent. These nitrosyl 2  bis(benzyl) complexes, however, are unique members in the class of dialkyl nitrosyl complexes since they possess 18-electron metal centers and coordinatively saturated molecular structures. As a result, the discussion of their preparation, molecular structures and characteristic properties constitutes a separate story in this Thesis (see Chapter 3).  B.  Spectroscopic and Some Physical Properties of the Cyclopentadienylnitrosyl Dialkyl  Complexes, Cp'M(NO)R , (Cp'= Cp and Cp*; M= Mo and W; R= C H , CH SiMe3, 2  3  CH CMe3, and CH CMe Ph). 2  2  2  18  2  All of the dialkyl nitrosyl complexes, Cp'M(NO)R , are red to purple solids which 2  are very soluble in common organic solvents such as hexanes, Et 0, benzene, CH C1 and 2  2  2  THF. Species containing the Cp* and CH SiMe ligands are the most soluble members of 2  3  this class of compounds; for example, they are quite soluble in pentane. The dialkyl nitrosyl complexes of tungsten are air and thermally stable as solids, and they may be handled in air for short periods of time without noticeable decomposition (the exceptions being Cp'W(NO)Me2 which are generated in solution at low temperatures). The molybdenum analogues, specifically CpMo(NO)(CH SiMe ) and 2  3  2  Cp*Mo(NO)(CH SiMe ) , are also thermally stable as solids having melting points of 2  3  2  approximately 72°, but they are very air and moisture sensitive either as solids or in solutions. Therefore, they must be handled in a dinitrogen atmosphere at all times. Both CpMo(NO)(CH SiMe ) and Cp*Mo(NO)(CH SiMe ) react with adventitious water or 2  3  2  2  3  2  molecular oxygen to afford [CpMo(NO)(CH SiMe )] 0i-O) and 2  3  2  47  [C£*Mo(NO)(CH SiMe )] 0*-O), respectively, which are isolable in moderate yields as 2  3  2  analytically pure red-black crystalline solids. These conversions may be accelerated if purple hexanes or Et 0 solutions of CpMo(NO)(CH SiMe ) or Cp*Mo(NO)(CH SiMe ) 2  2  3  2  2  3  2  are filtered through alumina HI, Florisil, or even to some extent Celite. The formulations of these Cp'M(NO)R complexes as monomeric, 16-electron 2  entities are supported by their low-resolution mass spectral data. For instance, the parent ion at 437 m/z (^Mo) is observed in the 70 eV mass spectrum of Cp*Mo(NO)(CH SiMe ) . The IR, and *H and C NMR spectra of all of the dialkyl n  2  3  2  nitrosyl complexes are consistent with their possessing monomeric three-legged piano stool molecular structures consisting of a linear M-N-O linkage and two normal a-bonded alkyl groups, i.e.,  N  III  O  a fact that has been established for CpW(NO)(CH SiMe ) in the solid state by a single2  3  2  crystal X-ray crystallographic analysis. The methylene protons were located in the Fourier 3  difference maps, and there is no indication of agostic methylene C-H linkages with the central tungsten atom. The IR spectra of Nujol mulls of all the Cp'M(NO)R complexes 2  exhibit a strong absorption in the region 1630-1550 cm" attributable to a stretching 1  vibration of their NO ligand. Species containing the CH SiMe group also exhibit in their 2  3  Nujol mull IR spectra an absorption in the region 1240-1250 cm" which is assigned to 1  "Si-Me-  For instance, the IR spectrum of Cp*Mo(NO)(CH SiMe ) as a Nujol mull exhibits  a strong i/^o  2  a t  1595 cm" and a i^si-Me 1  a t  1243 cm" . 1  3  2  48 The H and C NMR spectra of the Cp'M(NO)R complexes are consistent with 1  B  2  their being monomeric species in solutions and can be interpreted in a straightforward manner. Particular spectral features worthy of note are that the *H NMR spectra of all the Cp'M(NO)R complexes display a single AY-pattern for the two magnetically equivalent 2  sets of diastereotopic methylene protons (i.e. R'CH H -M-CH .H .R') on the CH R' a  b  a  b  2  groups, (R' = SiMe , CMe , and CMe Ph). This is illustrated by the *H NMR spectrum of 3  3  2  •  ——  Cp Mo(NO)(CH SiMe ) in C D solution which is shown in Figure 2.1 (a). One doublet 2  3  2  6  6  of the AX-pattern is observed in the spectral region 2.5-1.8 ppm (2.20 ppm for Cp*Mo(NO)(CH SiMe ) ) while the second doublet appears upfield from TMS around 2  3  2  (1-2) ppm (-1.17 ppm for Cp*Mo(NO)(CH SiMe ) ), each with / 2  2  3  2  H a H b  between 10-12 Hz.  The two singlet proton resonances at 1.49 ppm and 0.37 ppm integrating for 15 and 18 protons, respectively, in the *H NMR spectrum of Cp*Mo(NO)(CH SiMe ) , [Figure 2.1 2  3  2  (a)], may be assigned to the methyl protons on the cyclopentadienyl ligand and the methyl protons on the alkyl groups. The most notable feature of the C NMR spectrum of all of B  the Cp'M(NO)R complexes is that a single carbon resonance is observed for the two 2  methylene carbon atoms. This is illustrated by the C NMR spectrum of B  Cp*Mo(NO)(CH SiMe ) in C D solution, shown in Figure 2.1 (b). A single carbon 2  3  2  6  6  resonance is observed at 66.5 ppm indicating that the two methylene carbon atoms are magnetically equivalent. The thermal stability of these 16-electron coordinatively unsaturated complexes, as suggested by Fenske-Hall molecular orbital calculations on the model compound CpMo(NO)Me , results from their lowest unoccupied molecular orbitals (LUMO's) being 2  non-bonding in character. The LUMO of these molecules is a metal-localized orbital 3b  which should also confer Lewis acid properties on these electron-deficient compounds. Consistent with this view is the fact that CpW(NO)(CH SiMe ) readily forms 1:1 adducts 2  3  2  with PMe in hexanes solution, however, CpW(NO)(CH SiMe ) does not coordinate the 3  2  3  2  more sterically demanding PMePh and PPh ligands. These results suggest that the 15  2  3  49  (a)  JUL '  1  I 3.0  '  I  u  ~1  JL 1  -1.0  5.0  (b)  1  JL 1—l—i—l " i "  1—r—i—i—|—i—i—i—i—j—i—i—i—i—j—i—i—i—i—|—i—i—i—i—j—i—i—r ~i—|—i—r—i—i—i—i—i—i—i—r—r • 40  2 0 PP  Figure 2.1 (a) 300 MHz *H and (b) 75 MHz C{ H} NMR spectra of B  1  Cp*Mo(NO)(CH SiMe )2 in QDg. [Signals designated by (*) are due 2  3  to a small amount of [Cp*Mo(NO)(CH SiMe3)]2-0*-O) as an impurity.] 2  T  50  accessibility of the LUMO to incoming bases will be restricted somewhat by the steric bulk of the hydrocarbon ligands on the dialkyl nitrosyl complexes. C.  Spectroscopic and Physical Properties and Molecular Structures of the  [Cp'Mo(NO)(CH SiMe )] fc-0) (Cp'= Cp or Cp*) Complexes.  19  2  3  2  Both of the new cyclopentadienylnitrosyl oxo alkyl complexes, [Cp'Mo(NO)(CH SiMe3)]2(/i-0), are air-stable as solids, but, in solution they are air2  sensitive. Furthermore, they appear to be stable to water in solution. They probably result from reactions of the Cp'Mo(NO)(CH SiMe3)2 complexes with either H 0 present in 2  2  solution or on the chromatography column materials [Alumina (UJ) or Florisil]. The solid-state molecular structure of [Cp*Mo(NO)(CH SiMe3)]20i-O), determined 2  by a single-crystal X-ray crystallographic analysis, confirms the dimetallic nature of these 115  new types of cyclopentadienylnitrosyl oxo complexes. A view of the molecular structure is presented in Figure 2.2, which shows the 3-legged piano-stool molecular geometry about each molybdenum center. Selected bond lengths and bond angles of one of the two independent molecules of the unit cell are presented in Tables 2-1 and 2-IL respectively. The nitrosyl ligand is bonded to the Mo atom in an essentially linear fashion (Mo-N-O = 169°), and the short Mo-N (=* 1.758(2) A) and long N-O (1.21-1.22 A) bond lengths suggest that there is considerable Mo-•NO back bonding in the molecule.  20  The most chemically interesting feature of the structure involves the orientation of the two Cp*Mo(NO)R-units with respect to each other about the Mo-O-Mo bond. The oxo ligand bridge between the two molybdenum centers is also essentially linear, one of the molecules in the unit cell having Mo-O-Mo 175.12°, while the other molecule in the unit cell has Mo-O-Mo 176.74°. The two Cp*Mo(NO)R-units are twisted by approximately 90° with respect to one another, as evidenced by the torsion angles N(l)-Mo(l)-Mo(2)-N(2) = 94.0°, C(21)-Mo(l)-Mo(2)-C(25) = 94.36°, and N(l)-Mo(l)-Mo(2)-C(25) = -3.53°, (calculated using a perfectly linear Mo-O-Mo bond). This, together with the observed  51  short Mo-O bond lengths (1.894(1) and 1.901(1) A ) which are intermediate between typical Mo-O and Mo=0 bond lengths, suggest that there is considerable interaction of the empty d-orbitals on each of the Mo centers with the lone pairs of electrons situated in />-orbitals on the oxygen atom. A strong Mo=O=Mo interaction is achieved by the empty orbital of one of the Cp*Mo(NO)R groups interacting with the p -orbital on the oxygen, and rotation x  of the other Cp*Mo(NO)R group by 90° allows the empty orbital on it to interact with the Py-orbital on the oxygen. This would therefore suggest that each Mo center is formally an 18-electron species.  C(16)  Figure 22  A view of the molecular structure of [Cp*Mo(NO)(CH SiMe3)]2(A« -O). 2  52 Table 2-1.  a  b  2  2  Mo(l) -O(l)  1.9001(15)  Mo(2) -O(l)  1.8945(15)  Mo(l) -N(l)  1.758(2)  Mo(2) -N(2)  1.758(2)  Mo(l) - C(21)  2.170(2)  Mo(2) - C(25)  2.165(3)  Mo(l) -Cp(l)  2.0528(12)  Mo(2) -Cp(2)  2.0695(12)  N(l)- 0(2)  1.218(3)  N(2)- 0(3)  1.224(3)  C(21) -• Si(l)  1.852(3)  C(25)•• Si(2)  1.858(3)  Cp(l) is the centroid of C(l)-C(5) and Cp(2) is the centroid of C(12)-C(17).  Table 2-II.  a  Selected Bond Lengths (A) in [Cp*Mo(NO)(CH SiMe3)] -&*-O).  Selected Bond Angles (°) in [Cp*Mo(NO)(CH SiMe )] -( t-0).  1  2  3  2  r  Mo(l) - N(l) - 0(2)  167.0(2)  Mo(2) - N(2) - 0(3)  168.8(2)  O(l) - Mo(l) -N(l)  101.75(9)  O(l)- Mo(2) -N(2)  103.47(9)  O(l) - Mo(l) -C(21)  106.45(8)  O(l)- Mo(2) -C(25)  104.05(10)  0(1) - Mo(l) -Cp(l)  122.31(6)  O(l)- Mo(2) -Cp(2)  122.59(6)  N(l) - Mo(l) -C(21)  93.93(10)  N(2)- Mo(2) -C(25)  94.64(11)  N(l) - Mo(l) -Cp(l)  119.00(8)  N(2)- Mo(2) -Cp(2)  118.88(8)  C(21) - Mo(l) -Cp(l)  109.27(8)  C(25) - Mo(2) -Cp(2)  108.80(8)  Mo(l) - C(21) - Si(l)  116.72(13)  Mo(2) - C(25) -Si(2)  117.02(14)  Mo(l) - O(l) -Mo(2)  175.12(9)  b  Cp(l) is the centroid of C(l)-C(5) and Cp(2) is the centroid of C(12)-C(17). The other molecule in the unit cell has Mo(3) - 0(4) - Mo(4) = 176.74(9)  53  The intramolecular dimensions of the Mo-O-Mo bridging unit in [Cp*Mo(NO)(CH SiMe3)]2(/i-0) may be compared to those of related structures in the 2  literature. The Mo-O-Mo angle in the crystallographically characterized complex [Cp*Mo(O) ] 0'-O), is 178° and has M-O bond lengths of * 1.86 A. In contrast to these 21  2  2  linear oxygen bridged systems, the bent M-O-M system is best represented by the Mo-OMo angle in the complex [CpMo(0)] (/i-0) which is 84.2°, and the complex also contains 22  2  2  a significantly longer M-O bond length of ~ 1.941(3) A. Furthermore, a related oxo-bridged nitrosyl complex has been structurally characterized, namely, [Ir(PPh3)(NO)] (/i-0), and 23  2  it contains an acute Ir-O-Ir angle of 82.3(3)° and long Ir-O bond lengths of =s 1.94 A. The IR spectral features exhibited by these two new types of cyclopentadienylnitrosyl oxo complexes are quite different from those exhibited by their corresponding parent dialkyl complexes, Cp'Mo(NO)(CH SiMe3) . The IR spectra of 2  2  [CpMo(NO)(CH SiMe )] 0i -O) and [Cp*Mo(NO)(CH SiMe )] 0i -O) as Nujol mulls 2  3  2  2  3  2  exhibit strong bands at 812 cm and 773 cm , respectively, which are diagnostic of Mo-O-1  -1  Mo vibrations. The nitrosyl-stretching frequencies exhibited by these two complexes (1603 cm and 1560 cm , respectively), are shifted to lower wavelengths by approximately 20-35 -1  -1  cm" than the t-No's exhibited by their parent dialkyl complexes (i.e. Cp'= Cp, V^Q 1622 1  Cp*,uxo 1595 cm" ). This is consistent with there being more electron density  cm" ; Cp'= 1  1  available on the molybdenum center for back-bonding into the antibonding orbitals of the nitrosyl ligand in these oxo complexes than in their parent dialkyl complexes. Therefore, this would suggest that the bridging oxo ligand is providing the molybdenum center, in addition to the 1 electron in the a-Mo-O link, some electron density associated with the lone pairs of electrons in its /?-orbitals. This is in agreement with the X-ray crystallographic results discussed above. The H and C{ H} NMR spectra of [Cp*Mo(NO)(CH SiMe )] 0x-O) in QDg X  B  1  2  3  2  solution are shown in Figure 2.3 and may be directly compared with the H and ^C^H} 1  NMR spectra exhibited by its parent dialkyl complex (Figure 2.1). The H NMR spectrum, 1  54  (a)  I  To"  2.5  I  1 2.0  1  1  T~  I  1.5  1  '  I  1.0  1  1  0.5  0.0  I  -1.0  1  I ' -1 . 5 PPM 1  '  (b)  I  I  I  I  I n . 1I »•»«. I  I  120  I  I  1 1 1  -i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—r—i—i—I—i—i—i—i—i—i—i—i—IIII—i—i—i—i—i—i i i ' i ' i i i ' " ~r~ 100 80 60 4b 2 0 PPM 6 r  Figure 23 (a) 300 MHz *H NMR and (b) 75 MHz ^C^rl} NMR spectra of [Cp*Mo(NO)(CH SiMe3)] -0i-O) in C D solution at ambient temperatures. 2  2  6  6  55  Figure 2.3 (a), exhibits an AB-pattern for the two diastereotopic methylene protons and singlet resonances at 1.59 ppm and 0.45 ppm (integrating for 15 and 9 protons, respectively) which may be assigned to the methyl protons on the Cp-ligand and the methyl groups attached to the silicon atom, respectively. The methylene carbon resonance in the ^C^H} NMR spectrum now appears at_36 ppm which is shifted upfield from the corresponding carbon resonance observed for its parent dialkyl complex (66.5 ppm, Figure 2.1 (a)). Evident from both the H and C NMR spectra of [Cp*Mo(NO)(CH SiMe )] (> J  B  2  3  2  O) and [CpMo(NO)(CH SiMe3)] (/j-0) in QDg solution is the occurrence of isomers in 2  2  solution in ratios of 9:1 and 3:1, respectively. The isomers have been assigned as depicted below. The molecular structure of [Cp*Mo(NO)(CH SiMe )] f>-0) (Figure 2.2) shows 2  3  2  that the two [Cp*Mo(NO)(CH SiMe )]-units are twisted by 90° with respect to each other 2  3  about the Mo-O-Mo bond. From the arguments above, two orientations should be possible, 90° and 270°. Isomer A (90°) is probably the dominant form in solution as molecular models show that Isomer B (270°) possess steric interactions of the Cp ligand on 1  one Mo with the bulky CH SiMe3 group on the other Mo center. Consistent with this are 2  the observed ratios listed above, being dramatically dependent on the Cp' ligand, i.e. Cp versus Cp*. The observation of distinct isomers for the two complexes [Cp'Mo(NO)(CH SiMe3)] (/i-0) suggests that these complexes maintain their dimeric 2  2  natures in solution.  Isomer A (90°)  IsomerB  (270°)  56  D.  Syntheses of a Series of Novel 16-electron Cyclopentadienyl Dioxo Alkyl Complexes,  Cp'M(0) R (Cp' = Cp or Cp*, R = CHjSiMe^ CH CMe3, CH CMe Ph, CH Ph, or Me) 2  2  2  2  2  of Molybdenum and Tungsten. Treatment of solutions of the 16-electron dialkyl compounds, Cp'M(NO)R , in 2  various organic solvents with an excess of dioxygen at ambient temperatures and pressures produces novel dioxo alkyl complexes, Cp'M(0) R, which can be isolated in reasonable 2  yields (45-60%). The general transformation is summarized in eq. 9, the conversions being relatively rapid and straightforward in most cases. excess 0  Cp'M(NO)R  2  2  •  6-12 h,20°C  Cp'M(0) R  (9)  2  [M= Mo or W; Cp'= Cp or Cp*; R= CH SiMe or CH ] [M= W; Cp'= Cp or Cp*; R= CH Ph] [M= W; Cp'= Cp; R= CH CMe or CH CMe Ph] 2  3  3  2  2  3  2  2  Interestingly, the presence of adventitious water during the occurrence of reactions 9 is a help and not a hindrance, the isolated yields of the dioxo alkyl products generally increasing by 10-15% under these conditions. The reason why this is so is not clear at the present time, although a rationale will be presented later. Furthermore, reactions 9 can be effected in the solid state. Exposure of the dialkyl complexes, Cp'M(NO)R , as analytically 2  pure red to purple solids to an atmosphere of molecular dioxygen over a period of a few days results in their conversion to the dioxo alkyl complexes in high yields. In fact, in some cases the yields are nearly quantitative. The unique reactants in reaction 9 are the bis(benzyl) complexes Cp'M(NO)(CH Ph) (Cp*= Cp and Cp*; M= Mo and W). Unlike most of their dialkyl 2  2  analogues, all the bis(benzyl) complexes are stable to air in the solid state for at least 2 months. Furthermore, only the tungsten bis(benzyl) complexes react in solution with 0 to 2  57  afford dioxo benzyl complexes, Cp'Wf^O^CH^Ph), albeit in low isolated yields. The limited reactivity of these bis(benzyl) complexes with molecular oxygen appears to be related to the unique molecular structures exhibited by these complexes in solution and in the solid state. As a result, the characteristic properties and reactivity of these bis(benzyl) complexes is considered as a separate story in Chapter 3 of this thesis. Those Cp'M(NO)R2 complexes that are very air- and moisture-sensitive give rise to additional products during reactions 9. For example, reaction of Cp*Mo(NO)(CH2SiMe3)2 with O2 gives the desired dioxo alkyl complex Cp*Mo(0)2(CH2SiMe3) and [Cp*Mo(O) ]20i-O), eq. 10. 2  Cp*Mo(NO)(CH SiMe ) 2  3  excess O2,6-12 h, 20 °C 2  Cp*Mo(0) (CH SiMe ) 2  2  + [Cp*Mo(0) ]2^-0)  3  2  (10)  If freshly prepared and analytically pure Cp*Mo(NO)(CH SiMe3)2 is used in reaction 10 2  the desired dioxo alkyl complex can be prepared in 53% isolated yield and the yields of [Cp*Mo(O)2]20*-O) are about 10%. If, however, impure staring material is used, significant quantities of [Cp*Mo(6)2]2v -0) (20-25% yields) are isolated as a by-product u  from the reaction, and the yields of Cp*Mo(0)2(CH SiMe3) are greatly reduced. 2  Presumably the [Cp*Mo(O)2]20*-O) by-product results from the reaction of O2 with [Cp*Mo(NO)(CH SiMe )]20i-O), the thermal decomposition product of 2  3  Cp*Mo(NO)(CH SiMe ) . 2  3  2  In many cases the dioxo alkyl complexes may also be prepared by the reaction of the dialkyl nitrosyl complexes with aqueous hydrogen peroxide in Et20 solutions, as outlined in eq. 11.  58  Cp'M(NO)R  30% H2O2 (aq.) 2  Et 0,1-2 h, 20 °C  *  Cp'M(0) R  (11)  2  2  [Cp'= Cp or Cp*; M= Mo or W; and R= CH SiMe or CH ] 2  3  3  In general, conversions 11 are more rapid than those involving 0 (eq. 9), usually being 2  complete in less than 1 h and producing comparable isolated yields of the Cp'M(0) R 2  products. The stoichiometry of the oxidizing agent must be carefully controlled, however, in order to avoid conversion of the desired Cp'M(0) R compounds to their Cp'Mfo 2  2  0 )(0)R derivatives (Chapter 5). Furthermore, occasionally the desired products of 2  reactions 11 are difficult to isolate due to the formation of fine white, insoluble metal-oxide powders as by-products. These powders are difficult to separate by filtration techniques from the dioxo complexes. The unique case where reaction 11 does not occur involves the reaction between Cp*Mo(NO)(CH SiMe ) and H 0 . The major product in this reaction is 2  3  2  2  2  [Cp*Mo(O) ] 0*-O) in 20% yield, and no significant quantities of Cp*Mo(0) (CH SiMe ) 2 2  2  2  3  are obtained. This probably is a result of reaction of Cp*Mo(NO)(CH SiMe ) with H 0 2  3  2  2  2  to first form [Cp*Mo(NO)(CH SiMe )] (/i-0) and further reaction with H 0 affords the 2  3  2  2  2  [Cp*Mo(0) ] (/i-0) product. As a result, reactions 9 are generally the method of choice 2 2  for the preparation of the Cp'M(0) R complexes. 2  The dialkyl reactants required for reactions 9 and 11 need not be isolable species, but may be generated in situ by the alkylation of their halo precursors with either alkylhthium or -Grignard reagents. For example, the relatively unstable Cp'M(NO)Me (Cp' = 2  Cp and Cp*; M= Mo and W) complexes may be generated in situ, and subsequently treated with either 0 or HC>2 as outlined in reactions 9 and 11 to afford a series of dioxo 2  2  methyl complexes. At present, the principal limitation of conversions 9 and 11 is the  59  availability of the requisite dialkyl reactant complexes which so far is restricted to compounds containing alkyl ligands that do not possess p -hydrogens. Two of the Cp'M(0)2R compounds prepared in this work, namely Cp*W(0)2(CH SiMe3) and Cp*W(0)2Me, have been recently prepared by Faller and Ma 2  via a different synthetic route. This route, outlined in equations 12 and 13, involves 24  initial oxidation of the group 6 metal carbonyl complexes with molecular oxygen. The two dioxo alkyl complexes are subsequently generated by the alkylation of Cp*W(0)2Cl (eq. 21  13), unfortunately in very low isolated yields (_ 10-15%). This route has so far been limited to only the Cp*-tungsten compounds and the alkyl groups where R= Me or CH2SiMe3. CHCI3  [Cp*M(CO) ] 2  2  + 0  [Cp M(O) ] 0i-O) + Cp M(0) C1  2  2  2  2  (12)  21  [M= Mo orW] Cp*W(0) Cl + RM' 2  R= Me orCH SiMe 2  3  • Cp W(0) R 2  [M'= LiorMgCl]  +  M'Cl  (13)  24  In this work, Cp W(0) C1 was prepared in a similar manner by exposing a CH2CI2 2  solution of Cp*W(NO)Cl2 to an excess of dioxygen overnight, as outlined in eq. 14. CH C1 2  [Cp W(NO)Cl ] 2  2  + 0  2  2  ^CpW(0) Cl (21% yield) 2  24h,25°C  (14)  The analytical and spectroscopic data for this complex are included in Tables 2-III and 2IV. Alkylation of the Cp*W(0) Cl complex thus prepared proved unsuccessful at affording 2  the desired dioxo alkyl, Cp*W(0)2R, complexes in better than 10-15% yields. This is presumably due to the competing reactions involving the reaction of the alkyl-lithium or alkyl-Grignard reagent with the terminal oxygen atom(s) of the Cp*W(0) Cl complex. 2  60  There is a precedent for this, as reaction of Cp*W(0)2Cl with 3 or more equivalents of MeLi does afford small quantities of Cp*W(0)Me3.  21  It appears, therefore, that the synthetic routes used in this work, outlined in eqs. 9 and 11, are superior to those of eqs. 12 and 13, since a series of Cp'M(0)2R complexes with a wide range of alkyl groups for both molybdenum and tungsten may be prepared. Furthermore, the Cp'M(0) X (Cp'= Cp or Cp*, M= Mo or W, X= CI or Br) complexes 25  2  have been known for some time now, and to the best of my knowledge no successful alkylations of these complexes, other than those in eq. 13, have been reported to date. E.  Some Physical Properties of the Cp'M(0)2R Complexes (Cp' = Cp or Cp*, M = Mo or  W, R = CH SiMe3, CH CMe3, CH CMe Ph, CH Ph, or Me). 2  2  2  2  2  The physical properties of the series of dioxo alkyl complexes of both molybdenum and tungsten isolated during this work are summarized in Tables 2-IH and 2-IV. All twelve of the dioxo alkyl complexes are white, diamagnetic solids which are very soluble in common organic solvents such as benzene, Et2<I), CH Ci2, and THF, and are somewhat less 2  soluble in hexanes. Species containing the Cp* and Me3SiCH2 ligands are the most soluble members of this class of compounds. Unfortunately, solutions of the various dioxo alkyl compounds (particularly in CH2CI2) slowly decompose and develop a yellow coloration over a few days even when maintained at -20 °C under N2. All the Cp'M(0)2R compounds are quite thermally stable as solids, however, and are unaffected by exposure to air under ambient conditions. Many of them melt reversibly without decomposition, and samples of CpM(0) (CH SiMe ) (M = Mo or W) may be sublimed unchanged at 80 °C and IO mm -3  2  2  3  onto a water-cooled probe. Heating a sample of CpW(0)2(CH2SiMe3) in refluxing dioxane for three days leaves most of the sample unchanged, but a small portion is converted to CpW(0)2Me, probably resulting from hydrolysis of the methylene carbonsilicon bond. The formulations of these complexes as monomelic, 16-electron entities are supported by their elemental analysis and low-resolution mass spectral data.  61  Table 2-III. Analytical and IR Data for the Dioxo Alkyl Complexes analytical data (%) C  Complex  IR data (Nujol, cm' ) 1  "M=0  H  mp.°C  calcd found  calcd  found  asvm svm  112-115  24.34 24.19  2.72  2.69  951  910  110-111  29.44 29.34  4.36  4.34  948  907  CpW(0) (CH CMe )  34.11 34.00  4.58  4.49  952  906  CpW(0) (CH CMe Ph) 154 dec  43.50 43.25  4.38  4.45  951  907  CpW(0) (CH Ph)  163 dec  38.70 38.43  3.25  3.26  947  901  Cp*W(0) Me  119-120  36.08 36.40  4.95  4.87  943  899  31.07 31.36  3.91  3.86  937  897  CpW(0) Me 2  CpW(0) (CH SiMe ) 2  2  2  3  2  2  3  2  2  2  2  2  Cp*W(0) Cl 2  Cp*W(0) (CH SiMe )  118-119  38.36 38.20  5.98  5.96  937  901  Cp*W(0) (CH Ph)  148-150  46.17 45.81  5.01  4.79  939  899  CpMo(0) Me  95 dec  34.64 34.53  3.88  3.95  926  902  CpMo(0) (CH SiMe )  95-97  38.57 38.56  5.75  5.70  924  895  Cp*Mo(0) Me  104-106  47.49 47.43  6.52  6.62  918  887  104-106  48.12 48.19  7.50  7.60  912  893  2  2  2  3  2  2  2  2  3  2  Cp*Mo(0) (CH SiMe ) 2  2  3  "Si-Me  1248  1242  1246  1242  62 Table 2-IV. Mass Spectral and *H and C{ H} NMR Data for the Dioxo Alkyl Complexes 13  1  low-resolution mass spectral data m/z  J  H NMR data  3  Complex  D  s  +  +  CpW(0) (CH SiMe ) 2  2  3  2  2  3  5.49 (s, 5H, C ^ ) 1.14 (s, 3H, CH V ^ - 12Hz)  8.6 (CH , / = 144.5 Hz)  5.69 (s, 5H, C 5 / / 5 )  109.5  0.88 (s, 2H, Cff , = 12 Hz) 0.28 (s, 9H, SiO/3)  19.2 (CHj, / = 135 Hz) 0.7(Si(CH3)3)  5 JO (s, 5H, C ^ ) 2.19 (s, 2H, CH , J = 11.4 Hz)  50.4 (C// , V c w = 144.4 Hz)  130 (s, 9H, 0 ( 0 / 3 ) 3 )  333 (C(CH3)3)  109.2  353, [P-Me] 337, [P-2Me] +  1  c w  2  352, [P] 337,[P-Me] 295,[P-2Me] +  +  (C5H5)  1  c w  109.5  (C5H5) 2  2  2  +  (C5H5) 3  y  +  CpW(0) (CH CMe )  1  6  296, [P] 281,rP-Me]  2  C{ H}NMR data  (C 6>  b  CpW(0) Me  13  HW  32.8(C(CH3)3) CpW(0) (CH CMe Ph) 2  2  2  414, [P] 399, [P-Me] 296, [PCMe Ph] 281, [PCHjCMe^h]  7.5-7.2 (m, 5H, C ^ ) 5.45 (s, 5H, C / y 252 (s, 2H, CH , J = 11.8 Hz)  +  +  2  2  c w  1.67 (s, 6H, 0 ( 0 / 3 ) 2 )  2  4  2  7.22-7.18 (m, 5H, C ^ )  372, [P]  +  2  5  5  1  HW  2  2  6  5  2  +  CpW(0) (CH Ph)  128-125.9 (m, C H ) 109.5 (C H ) 50.5 (CH , / = 145.4 Hz) 39.75 (CMe Ph) 30.8 (C(CH^ )  5.44 (s, 5H, C 5 / / 5 )  3.42 (s, 2H, CH ) 2  1453 (quaternary PhC) 136.9,129.8,128.5, 125.0 (Ph Q 109.8  (C5H5)  36.8 (CHj, = 134.8 Hz) 366, [P] 351, [P-Me]  Cp*W(0) Me  +  2  +  1.64 (s, 15H, C (Of3) ) 0.88 (s, 3H, O / 3 , J = 11.2 Hz) 5  5  2  Wff  116.7 (C (CH3) ) 13.85 (CH , V c w = 140.7 Hz) 10.22 (C (CH^ ) 5  5  3  S  Cp*W(0) (CH SiMe ) 2  2  3  423, [P-Me]  +  1.67 (s,15H, C ( 0 / 3 ) 0.41 (s, 2H, OZ , 2 / ^ = 10.7 Hz) 5  2  036 (s, 9H, SiO/3)  5  )  S  1165 (C (CH3) ) 245 (CH , / = 135 Hz) 5  5  2  1  c w  10.5  (05(^3)5)  0.97 (SiCHj)  63 Table 2-IV. continued low-resolution mass spectral data m/z  J  H NMR data  13  C{ H}NMR data 1  8  Complex  b  Cp*W(0) Cl  386, [P]  Cp*W(0) (CH2Ph)  442, [P] 351, [ P - C ^ ]  2  1.74(s, 15H,C (CW3) )  +  5  736(<L2H,a-C ^ )  +  2  5  6  5  4  6.87 (t, lH,p-Cg^ ) 2.84 (s, 2H, CH , = 13.8 Hz) 1.64 (s, 15H, C (CHJ ) 5  2  S  CpMo(0) Me  210, [P]  2  CpMo(0) (CH SiM ) 2  2  e3  +  267, [P-Me]  +  S  280, [P] 265, [P-Me] +  2  +  5  109.3 (C H ) 11.6 (CH^  5.61 (s, 5H, C ^ ) 1.20 (s, 2H, CH ) 0.32 (s, 9H, SiCff^  109.5 (C H ) 24.6 (CH ) 0.6 (SiCHj)  1.57 (s, 15H, C ^ C K ^ ) 0.90 (s, 3H, CH3)  117.4 (C (CH3) ) 15.8 (CHj) 10.4(C (CH3) )  5  2  2  3  352, [P]  +  5  2  Probe temperatures 100-150°C. species (e.g. ^Mo, W etc.).  fl  1 8 4  5  2  1.59 (s, 15H, C (CHj) ) 0.63 (s, 2H, CH ) 037 (s, 9H, SiCtfj) 5  5  5  5  5  Cp*Mo(0) (CH SiMe )  $  5.42 (s, 5H, C ^ ) 1.19 (s, 3H, CHJ  2  Cp*Mo(0) Me  146.0 (quaternary PhC) 129.9,128.4,124.8 (aromatic Q 117.1 (C (CHJ ) 44.0 (CHj, = 133.2 Hz) 10.4 (C^CH^)  5  5  117.4 (C (CH3) ) 28.8 (CHj) 10.5 (C (CRJ ) 0.8 (SiCH^ 5  5  5  S  Assignments involve the most abundant naturally occurring isotopes in each  64 The spectroscopic properties of all the Cp'M(0) R compounds (Tables 2-m and 22  IV) are consistent with their possessing the familiar three-legged piano-stool molecular structures having mirror symmetry at the central metal atom, i.e.,  a fact that has been established for CpW(0) (CH SiMe3) in the solid state by a single2  2  crystal X-ray crystallographic analysis. This analysis has also determined that the 4  tungsten-oxygen bond lengths of 1.716 (5) and 1.723 (5) A are consistent with the existence of W=0 double bonds in this compound. Accordingly, IR spectra of Nujol mulls of all the Cp'M(0) R complexes exhibit two strong absorptions in the region 960-885 cm  -1  2  attributable to the asymmetric and symmetric stretching vibrations of their M( = 0 )  2  groups. The " M O frequencies exhibited by these compounds are similar to those =  observed previously for the related CpM(0) X and [CpM(0) ] (/i-0) ' complexes (M 25  21  2  = Mo or W, X = halide). The f  M = 0  2  26  2  values of the dioxo alkyl compounds are relatively  insensitive to the alkyl group present, but do decrease by ca. 20 cm in going from W to -1  Mo which is as expected on the basis that stronger metal-oxygen bonding is anticipated for tungsten. Furthermore, these v  values of the dioxo alkyl compounds decrease by ca.  27  M = Q  10 cm" in going from Cp to Cp*. The H and C{ H} NMR spectra of the Cp'M(0) R 1  J  13  1  2  complexes (Table 2-IV) are relatively simple and can be interpreted in a straightforward manner. Particular spectral features worthy of note are that the H NMR spectra of the 1  Cp'W(0) (CH SiMe ) compounds display normal methylene singlets which generally 2  exhibit 14%  2  1 8 3  3  W satellites with / H W being in the range of ca. 11-13 Hz, as expected for a 2  monomelic species. In a complementary manner, the C{ H} NMR spectra of these 28  13  1  same compounds contain resonances due to the methylene carbons in the range 6 20-50  65  ppm, the Vcw values being 130-145 Hz. Furthermore, the C{ H} NMR spectra of 13  1  Cp'M(0)2Me species exhibit methyl carbon signals in the range S 8-16 ppm. Consequently, the physical properties of the dioxo alkyl compounds indicate that their monomeric molecular structures persist both in solutions as well as in the solid state.  F.  Possible Mechanisms for the Reactions of the Cp'M(NO)R2 Complexes with  Molecular Oxygen. The most striking feature of reactions 9, especially in so far as the Cp'W(NO)(CH2SiMe )2 complexes are concerned, is that they apparently bear no 3  resemblance whatsoever to the reactions undergone by these reactants when exposed to the heavier chalcogen, sulfur. Treatment of the precursor dialkyl nitrosyl complexes with elemental sulfur results in the sequential insertion of sulfur into the W-C a bonds as summarized in eq. 15 (M = Cp'W(NO) and R = CH SiMe3), each of the intermediate 2  complexes shown having been isolated and fully characterized for both the C p -  3 , 2 9 a  and  Cp*-containing systems. There is no evidence in either of the reactions 15 for the 29b  R  R  I  I  s  M-R  » M-  R S  R s  l/S  SR — • M  — SR  I  » M—SR  ( ) 15  formation of Cp'W(S)2(CH SiMe3) (Cp = Cp or Cp*), the dithio analogues of the dioxo 1  2  products formed in reactions 9. Conversely, there is also no evidence during reactions 9 involving the tungsten reactants for the formation of alkoxy complexes such as Cp'W(NO)(OCH SiMe3)2 (the formal oxo analogues of the final dithiolate products of 2  reactions 15). This is so despite the fact that such alkoxy complexes are known to result from the treatment of other 16-electron dialkyl complexes with elemental oxygen. ' For 30a b  example, reactions of Cp2ZrR2 complexes (isoelectronic with CpM(NO)R2) with O2 results  66  in the formation of bis(alkoxide) products, eq. 16. Similarly, insertions of molecular oxygen into group IV metal-carbon methyl bonds proceeds to form bis(methoxide) products in very high yields, eq. 11. Cp2ZrR  2  + 0  (tritox) M(Me) 2  2  300  ^Cp Zr(r -OOR)R  ^Cp Zr(OR)  2  2  2  + 0  7  2  :  2  [tritox = OC( Bu) ]  (16)  2  (tritox) M(OMe) 2  (17)  2  [M= Ti, Zr, Hf]  t  3  To check whether the nitrosyl bis(alkoxide) complexes, Cp'M(NO)(OR) , are 2  indeed intermediates in reactions 9, attempts were made to prepare these complexes via a different route. Reactions of several dialkyl nitrosyl complexes of tungsten in C^D^ solutions with various alcohols (eg. MeOH, BuOH and /?-MePhOH), as outlined in eq. 18, l  were performed in NMR tubes. Only the reaction involving CpW(NO)(CH SiMe3) and 2  Cp'W(NO)R  2  + R'OH  2  - Cp W(NO)(OR')R ,  -RH  1  + R'OH ^ Cp'W(NO)(OR') 2  (18)  2  -RH  1 Cp'= Cp or Cp*; R= CH SiMe and CH Ph; R'= Me, Bu and p-MePh 2 Cp'= Cp; R= CH SiMe ; R'= Me l  2  2  3  2  3  MeOH proceeds to completion. Furthermore, as the latter reaction proceeds, decomposition occurs as evidenced by the formation of brown precipitates. This likely reflects the instability of CpW(NO)(OMe) as *H signals attributable to CpH, Me SiCH R 2  3  2  (R=H or OH) and excess MeOH are observed. In all the other reactions in eq. 18, the reaction proceeds slowly to form the Cp'W(NO)(OR')R product and then stops. Both *H and C NMR data are collected for a number of these Cp'W(NO)(OR')R complexes in B  67  the Experimental Section, and they appear to be stable in solution for as long as six months. The rate of conversion of the dialkyl complexes to the Cp'W(NO)(OR')R complexes, the first step in eq. 18, depends greatly on the alcohol used, the alkyl substituents on the dialkyl complex and on the cyclopentadienyl ligand. The different rates observed in these reactions are consistent with steric factors governing the ease with which the alcohol can approach the metal center to expel the alkane and form the alkoxide. At no point, however, are proton resonances attributable to CpW(0)2R, (R= C H or 3  CH SiMe3) observed in these reactions or from decomposition of the alkoxide complexes. 2  In light of these results, the elucidation of the mechanisms operative during reactions 9 is of fundamental significance and interest. Plausible mechanistic pathways for reactions 9 using Cp'W(NO)(CH SiMe3)2 (Cp' = Cp or Cp*) as representative reactants 2  are presented in Scheme 2-1. The first step involves coordination of molecular oxygen to the coordinatively unsaturated 16-electron metal center in the dialkyl nitrosyl reactant, a demonstrated Lewis acid, to form a simple 1:1 adduct. The second step involves migratory 3  insertion of the bound nitrosyl group into one of the W-C a bonds, a process for which there is ample precedent in the chemical literature. [The alternative insertion of the 31  bound dioxygen into one of the W-C a bonds to form an alkyl peroxo complex would be expected to lead to a bis(alkoxy) complex as the final product.] The resulting peroxo 30  alkylnitroso complex could then expel the nitrosoalkane from the metal's coordination sphere in the final step and rearrange concomitantly to the final dioxo alkyl product. If such a mechanism is indeed operative, the by-product of reactions 9 involving the Cp'W(NO)(CH SiMe )2 reactants should then be the nitrosoalkane, Me SiCH NO. It is 2  3  3  2  unlikely that this by-product would persist long enough to be detected in the reaction mixture since such nitrogen oxides are known to be very useful oxidizing agents for the preparation of transition-metal organometallic oxo compounds, and it has been 32  established independently that PhNO in Et 0 consumes the tungsten dialkyls completely 2  even at -60 °C to produce 20-25% isolated yields of Cp'W(0) (CH SiMe ) as well as other 2  2  3  68  Scheme 2-1.  R=CH SiMe 2  3  69 products whose identities remain to be ascertained. In other words, it is expected that any nitrosoalkane produced during reactions 9 would itself be capable of functioning as the oxidant for these transformations, at least to some extent. Consequently, I have also included this pathway in Scheme 2-1. Regrettably, reactions 9 are not sufficiently clean to permit a straightforward kinetic analysis to be performed on them. Therefore, in an attempt to gain some direct evidence for the mechanistic possibilities outlined in Scheme 2-1,1firsteffected the reactions of Cp'W(NO)(CH SiMe )2 with 0 in benzene-*^ and monitored the progress of the 2  2  3  reactions by H NMR spectroscopy and then performed the reaction of 1  Cp*W(NO)(CH SiMe ) in hexanes with labelled 0 and analyzed the organometallic 1 8  2  3  2  2  product mixture by IR spectroscopy. Each of these studies will now be considered in turn. The H NMR monitoring only reveals the conversion of the dialkyl nitrosyl 1  reactants to the dioxo alkyl product complexes, no signals attributable directly to intermediate Cp'W-containing species or Me SiCH NO being evident. The H NMR 1  3  2  spectra of thefinalmixtures do, however, also exhibit several signals in the 6 0.1-0 ppm range attributable to alkyl protons. Figure 2.4 shows a series of H NMR spectra taken 1  during the course of the reaction between Cp*W(NO)(CH SiMe ) and dioxygen in the 2  3  2  presence of trace amounts of H 0 . At the completion of the reaction, which appears to be 2  quantitative, two extra proton signals are observed at 0.09 ppm and 0.02 ppm (integrating for 2 and 9 protons, with / . i= 6.6 Hz, respectively). Although this does not prove that 2  H  S  RNO is the by-product, it does account for all the protons and may well involve "Me SiCH OH" and/or "Me SiOH" species, resulting from degradation of the RNO 3  2  3  product. Traces of H 0 seem to give cleaner reactions, and, as seen previously, aid in 2  affording the dioxo alkyl complexes in higher isolated yields (=; 80-85%). This may be due to the H 0 destroying the RNO species generated which in turn would eliminate the 2  competing reaction between RNO and the dialkyl starting material.  33  70  (a)  [ (*) = solvent ]  (b)  (c)  (d)  ll 6 Figure 2.4  5  H  3  a  I  o  -i ppw  80 MHz *H NMR spectra at various times during the reaction between  Cp*W(NO)(CH SiMe3)2 and O2 in Q D solution in the presence of trace amounts of 2  6  H 0 ; (a) time = 0 min, (b) time = 1 h, (c) time = 11/2 h, (d) time = 2 h. 2  71  Table 2-V. Assignment of the i /  w = 0  Values to the Cp*W(0) (CH SiMe ) Isotopomers. 2  2  3  W=0 (Nujol mull, cm )  V  -1  isotopomer  i  2  v  Cp*W( 0) (CH SiMe )  v  939  901  Cp*W( 0)( 0)(CH SiMe )  926  868  Cp*W( 0) (CH SiMe )  891  858  16  2  16  2  3  18  2  3  18  2  2  3  378  355  A Nujol mull IR spectrum of the dried final reaction mixture of the labelling experiment (Figure 2.5 (a)), exhibits six absorptions of varying intensities at 939, 926, 901, 891, 868, and 858 cm attributable to ^ -1  W = Q  which may be compared with that of authentic  Cp*W(0) (CH SiMe ), Figure 2.5 (b). These absorptions may be assigned to the various 2  2  3  isotopomers of the Cp*W(0) (CH SiMe ) product by essaying the approximation that the 2  2  3  organic ligands contribute little to the W=0 vibrations and by treating the complex as a simple triatomic, 0=W=0, having  symmetry according to the method outlined by  Herzberg. If this is done, the assignments listed in Table 2-V are obtained. The validity 34  of these assignments may be confirmed by agreements of better than 1% between the observed and calculated isotope effects. Given these assignments, the ratios of the 35  various isotopomers in the final reaction mixture may be calculated from the intensity ratios of these absorptions in the final Nujol mull IR spectrum. In this manner, it can be determined that the final residue contains ca. 12% of Cp*W( 0) (CH SiMe ), ca. 35% of 16  2  2  3  Cp*W( 0)( 0)(CH SiMe ), and ca. 53% of Cp*W( 0) (CH SiMe ). In other words, 16  18  18  2  3  2  2  3  the majority of the dioxo alkyl product formed during the labelling study does indeed contain two labelled oxo ligands, as would be expected if the principal mechanistic pathway presented in Scheme 2-1 is operative. The fact that appreciable amounts of unlabelled or  72 o  ID  WAVENUMBERS  CCM-I>  • •  Figure 2.5. Spectral region 1510-520 cm" of the IR spectrum of (a) analytically pure 1  Cp*W(0) (CH SiMe ) as a Nujol mull and (b) the solid residue from the final reaction 2  2  3  mixtures of the 0 labelling experiment as a Nujol mull 1 8  2  73  partially labelled dioxo products are formed during the reaction involving labelled dioxygen can be attributed to some of them being formed by the reaction with the Me3SiCH NO by2  product (as shown in Scheme 2-1) and the rest resulting from slower exchange reactions of the labelled species with adventitious 0 and H 1 6  2  16 2  0 . That such exchange reactions do  occur can be demonstrated by simply exposing an Et 0 solution of the final labelled 2  reaction mixture to 0 for 2 d. An IR spectrum of the dried reaction residue as a Nujol 1 6  2  mull after this time clearly exhibits increased intensities of thei/w o' ^ s  u et 0 t n e  =  16  ^"  containing dioxo alkyl complexes (Table 2-V). In summary, then, the basic conclusion that emanates from these NMR and IR studies is that they are fully in accord with the mechanistic proposals outlined for reactions 9 in Scheme 2-1.  Summary The series of previously known 16-electron dialkyl nitrosyl complexes of molybdenum and tungsten, Cp'M(NO)R has been extended in this work to obtain a more 2  comprehensive class of complexes with a variety of alkyl substituents. These coordinatively unsaturated dialkyl nitrosyl complexes, Cp'M(NO)R , are very useful starting materials for 2  the preparation of a series of novel cyclopentadienyl dioxo alkyl complexes, Cp'M(0) R, 2  for both molybdenum and tungsten for a variety of alkyl groups. In general, reactions of the Cp'M(NO)R compounds with either 0 or H 02 in solution affords in moderate yields 2  2  2  the Cp'M(0) R complexes which are formally 16-electron species containing the central W 2  or Mo atom in its highest oxidation state of +6. There are a few exceptions. Most notably, the nitrosyl bis(benzyl) complexes, Cp'M(NO)(CH Ph)2, exhibit limited reactivity with 2  dioxygen. As a result, an extensive investigation into the molecular structures of these complexes both in solution and in the solid state was undertaken, and the results are discussed in the next chapter. Furthermore, reactions of Cp*Mo(NO)(CH SiMe3) with 2  2  molecular oxygen give in addition to Cp*Mo(0) (CH SiMe3), considerable amounts of 2  2  74  [Cp*Mo(O)2]20i-O) as a by-product. Reactions of Cp*Mo(NO)(CH SiMe ) with H 0 2  3  2  2  give [Cp*Mo(O) ] 0'-O) as the only product. Both of the Cp'Mo(NO)(CH SiMe ) 2 2  2  3  2  2  complexes are thermally- and moisture-sensitive in solution and decompose to form new types of cyclopentadienylnitrosyl oxo bridged dimolybdenum complexes, namely [Cp'Mo(NO)(CH SiMe )] 0i-O). 2  3  2  4 l NMR monitoring experiments and 0 -labelling studies of the reactions 18  2  between the Cp'M(NO)R compounds with 0 were performed during this work to gain 2  2  some direct evidence for the mechanism by which the dioxo alkyl complexes are formed from these reactions. The results of these studies are fully in accord with the reaction proceeding via direct attack of the 0 molecule at the 16-electron metal center in the 2  Cp'M(NO)R complex. Furthermore, the principal mechanistic pathway towards the 2  formation of the Cp'M(0) R complex involves both oxygen atoms of the 0 molecule 2  2  forming the two M = 0 linkages, as suggested in Scheme 2-1. The most interesting result of these studies is that the reactions with 0 bear no resemblance whatsoever with the 2  transformations undergone by the same dialkyl nitrosyl reagents when exposed to sulfur.  29  The characteristic reactivities of these new types of high oxidation state organometallic oxo complexes, Cp'M(0) R, are the subjects of Chapters 4,5 and 6 of this 2  thesis.  75  References and Notes:  (1)  Taken in part from: (a) Legzdins, P.; Phillips, E.C.; Sanchez, L. Organometallics 1989,5,940. (b) Legzdins, P.; Phillips, E.C.; Rettig, S. J. manuscript in preparation.  (2)  Sheldon, R. A.; Kochi, J. K. Metal-Catalyzed Oxidation of Organic Compounds; Academic: New York, 1981.  (3)  (a) Legzdins, P.; Rettig, S. J*; Sanchez, L. Organometallics 1988,7,2394. (b) Legzdins, P.; Rettig, S. J.; Sanchez, L.; Bursten, B.E.; Gatter, M.G. /. Am. Chem. Soc. 1985,107,1411.  (4)  Legzdins, P.; Rettig, S. J.; Sanchez, L. Organometallics 1985,4,1470.  (5)  Drezdzon, M.A.; Shriver, D.F. The Manipulations of Air-Sensitive Compounds; 2  (6)  nd  Ed.; John Wiley and Sons : New York, 1986.  Experimental Organometallic Chemistry, A Practicum in Synthesis and Characterization, Darensbourg, M.Y.; Wayda, A.L. Ed. Am. Chem. Soc. 1987.  (7)  Armarego, W.L.F.; Perrin, D.D.; Perrin, D.R. Purification of Laboratory Chemicals; 2  (8)  nd  Ed.; Pergamon Press : Oxford, 1980.  (a) Kharasch, M.S. Grignard Reactions of Non-Metallic Substances; Prentice Hall: New York, 1954. (b) Sommer, L.H.; Whitmore, F.C. /. Am. Chem. Soc. 1946,68, 481.  (9)  Spectroscopic data for all of the Cp'M(NO)X2 complexes used in this work, suggest that they are all monomeric in solution and therefore are represented as such throughout this chapter; (see also reference 12).  (10)  Bray, J.; Kita, W.G.; M^leverty, J.A.; Seddon, D. Inorg. Synth. 1976,16,24.  (11)  Legzdins, P.; Martin, D.T.; Nurse, C.R. Inorg. Chem. 1980,19,1560.  (12)  Dryden, N.H.; Einstein, F.W.B.; Jones, R.H.; Legzdins, P. Can. J. Chem. 1988, 66,2100.  76 (13)  Nurse, C R . Ph.D. Dissertation, The University of British Columbia, 1983.  (14)  Dryden, N.H.; Legzdins, P. manuscript in preparation.  (15)  Martin, J.T. Ph.D. Dissertation, The University of British Columbia, 1987.  (16)  The perhydrocyclopentadienyl analogue, [CpMo(NO)I]2 has been known for some time, see: James, T.A; M Cleverty J.A /. Chem. Soc. (A) 1971,1068. c  (17)  (a) Herring, F.G.; Legzdins, P.; Richter-Addo, G.B. Organometallics 1989,8, 1485. (b) Richter-Addo, G.B. Ph.D. Dissertation, The University of British Columbia, 1988.  (18)  Note, the following discussion does not pertain to the nitrosyl bis(benzyl) complexes Cp'M(NO)(CH Ph) , since they possess 18-electron metal centers 2  2  and coordinatively saturated molecular structures, see Chapter 3 of this thesis. (19)  The complex [Cp*Mo(NO)(CH SiMe )] 0i-O) exhibits interesting 2  3  2  electrochemical behavior in THF and CH C1 solutions. The full details will be 2  2  reported elsewhere.  lb  (20)  It is interesting that nearly identical intramolecular dimensions are observed in the W-NO linkage in the complex CpW(NO)(CH SiMe ) , i.e. W-N-O = 169.5°, 2  3  2  W-N = 1.757(8) A and N-O = 1.226 A, see reference 3. (21)  Faller, J.W.; Ma, Y. /. Organomet. Chem. 1988,340,59.  (22)  Couldwell, C ; Prout, K. Acta Crystallogr. B43 1978, 933.  (23)  Carty, P.; Walker, A ; Mathew, M.; Palenik, G J . /. Chem. Soc, Chem. Commun. 1969,1374.  (24)  Faller, J.W.; Ma, Y. Organometallics, 1988, 7,559.  (25)  (a) Cousins, M.; Green, M.L.H. /. Chem. Soc. 1964,1567. (b) Cousins, M.; Green, M.L.H. /. Chem. Soc A 1969,16.  (26)  Herberhold, M.; Kremnitz, W.; Razavi, A ; Shoellhorn, H.; Thewalt, U. Angew. Chem., Int. Ed. Engl 1985,24, 601.  (27)  (a) Cotton, F.A.; Wing, R.M. Inorg. Chem. 1965,4,867. (b) Callaham, K.P.; Wing, R.M. Inorg. Chem. 1969,8, 871.  (28)  Legzdins, P.; Martin, J.T.; Einstein, F.W.B.; Willis, A.C. /. Am. Chem. Soc. 1986,108,7971.  (29)  (a) Legzdins, P.; Sanchez, L. /. Am. Chem. Soc. 1985,107,5525. (b) Legzdins, P.; Martin, L.R. unpublished observations.  (30)  (a) Brindley, P. B.; Scotton, M. J. /. Chem. Soc., Perkin Trans. 2 1981,419. (b) Blackburn, T. F.; Labinger, J- A.; Schwartz, J. Tetrahedron Lett. 1975,3041. (c) Lubben, T. V.; Wolczanski, P.T. /. Am. Chem. Soc. 1987,109, 424 and references therein.  (31)  Seidler, M.D.; Bergman, R.G. /. Am. Chem. Soc. 1984,106, 6110 and references therein.  (32)  Bottomley, F.; Sutin, L. Adv. Organomet. Chem. 1988,28, 339.  (33)  Such reactions of alkylnitroso compounds with organometallic complexes, rarely give clean and quantitative reactions, (see reference 32).  (34)  Herzberg, G. Molecular Spectra and Molecular Structure. II. Infrared and Rama Spectra of Polyatomic Molecules', Van Nostrand: Toronto, 1966.  (35)  Reference 34, pp 228-231.  Chapter 3  Synthesis, Characterization, Physical Properties and Unusual Solid-State Molecular Structures of a Series of Nitrosyl Bis(Benzyl) Complexes of Molybdenum and Tungsten: Cp'M(NO)(CH Ph') , 2  [Cp = Cp (»? -C H ) or Cp* 0? -C Me ), Ph' = C H , or C ^ ^ ^ - M e , ) ] 5  5  5  5  s  5  6  5  2  1  79 Introduction As discussed in the preceding chapter, a large number of 16-electron dialkyl nitrosyl complexes of molybdenum and tungsten, Cp'M(NO)(R) , react with molecular oxygen at 2  ambient temperatures to afford a series of dioxo alkyl complexes, Cp'M(0) R, in good 2  yields, eq l .  2  Cp'M(NO)R  excess 0 : %.T.  2  7  •  Cp'M(0) R +  "RNO"  2  (1)  [M= Mo or W; Cp'= Cp or Cp*; R= CH SiMe or CH ] [M= W; Cp'= Cp or Cp*; R= CH Ph] [M= W; Cp'= Cp; R= CH CMe or Cr^CMe^h] 2  3  3  2  2  3  Proton NMR and 0 labelling studies, outlined in Chapter 2, are consistent with the 1 8  2  reaction proceeding via initial attack of the oxygen atom at the coordinatively unsaturated metal center. Unique members in the class of dialkyl nitrosyl complexes that exhibit limited reactivity to dioxygen in eq. 1 are the nitrosyl bis(benzyl) complexes, Cp'M(NO)(CH Ph') 2  2  (Cp' = Cp or Cp*; M= Mo or W; Ph' = C H or C H -2,4,6-Me ). As a result of this, a 6  5  6  2  3  more detailed investigation of the molecular structures of these bis(benzyl) complexes, both in solution and in the solid state, was undertaken. The first part of this chapter then, describes the full details of the preparation, characterization and spectroscopic properties of CpMo(NO)(CH Ph) (la), CpMo(NO)(CH C H2-2,4,6-Me ) (lb), 2  2  2  3  6  2  Cp*Mo(NO)(CH Ph) (2), CpW(NO)(CH Ph) (3), and Cp*W(NO)(CH Ph) (4). All of 2  2  2  2  2  2  these bis(benzyl) complexes possess unique Cp'M(NO)(r? -CH Ph)(r7 -CH Ph) solid-state 1  2  2  2  molecular structures and they all exhibit interesting temperature dependent dynamic processes in solution. The details of these studies and a rationale for the reactivity of the bis(benzyl) complexes with dioxygen are presented in the latter part of this chapter.  80 Experimental Section All reactions and subsequent manipulations involving organometallic reagents were performed under anhydrous and anaerobic conditions using conventional Schlenk-tube techniques ' or in a Vacuum Atmospheres Corp. Dri-Lab Model HE-43-2 drybox, under 3 4  an atmosphere of dinitrogen or argon. The general procedures employed in this study were the same as those described in the Experimental Section of the preceding chapter. The benzyl chloride (PhCH Cl) reagent was obtained from Aldrich and was dried over MgS0 2  4  before use. The organic compound, a -chloroisodurene (2,4,6-Me C H CH Cl), was also 2  3  6  2  2  obtained from Aldrich and was purified by vacuum distillation. The alkyl-Grignard reagents PhCH MgCl and 2,4,6-Me C H CH MgCl, were prepared according to known 2  3  6  2  2  procedures. The methanol and /-butyl alcohol reagents were distilled from CaH and 5  2  stored over 3A molecular sieves. The dioxygen, 0 , used during this work was purchased 16  2  from Medigas in 99.5% purity and was not further purified before use. The CpM(NO)I complexes (M= Mo and W ), and Cp*W(NO)I were prepared 6  7  8  9  2  2  by literature methods. The complex Cp*Mo(NO)I was prepared by treatment of 2  Cp*Mo(NO)(CO) with I in a similar manner as its perhydro analogue. The 10  2  2  Cp'M(NO)Cl complexes (Cp'= Cp or Cp*; M= Mo, or W), were synthesized by treating 2  the appropriate Cp'M(NO)(CO) complex with an equimolar amount of PC1 in E ^ O . 2  11  5  The purity of the above reagents was checked by elemental analysis before use. All other chemical reagents were either purchased from commercial suppliers or prepared by published procedures. All *H NMR spectra were obtained on a Varian Associates XL-300 or Bruker WP80 spectrometer with reference to the residual proton signal of the deuterated solvent employed (usually benzene-^). The C NMR spectra were recorded at 75 MHz and 25 B  MHz on a Varian Associates XL-300 spectrometer and a Bruker WP-80 spectrometer, respectively, with reference to the C signal of the solvent employed. All *H and C B  B  chemical shifts are reported in parts per million downfield from Me Si. The low 4  81 temperature *H and C NMR spectra were obtained on the instruments listed above using B  standard cooling apparatus. The standard low-power Waltz-16 broad band proton decoupling technique (^C^H}) was used. Collection of proton-coupled C NMR B  spectra, with excellent signal-to-noise and resolution in reasonable times (1-6 h), was accomplished using gated decoupling experiments [^H}) off during data acquisition (=*0.8 s) and on between acquisition (1.6 s)]. Standard Varian Attached Proton Test (APT) experiments allowed the collection of C NMR spectra which distinguished - C H and -CH B  3  type carbons from those of the type TC and -CH . In all such spectra in this work, -C and 2  C H carbons are plotted in the positive direction, while those of the type the -CH , and 2  3  CH are plotted in the negative direction. The melting point determinations were performed on a Gallenkamp Melting Point Apparatus, and the temperatures reported are uncorrected. Preparation of the Complexes CpMo(NO)(CH Ph) (la), CpMo(NO)(CH C H 2  2  2  6  2  2,4,6-Me3) (lb), Cp*Mo(NO)(CH Ph) (2), CpW(NO)(CH Ph) (3), and 2  2  Cp*W(NO)(CH Ph) (4). 2  2  2  2  All these syntheses involved treating the appropriate  2  Cp'M(NO)X precursors (Cp' = Cp or Cp*; M = Mo or W; X = CI or I) with the 6  2  requisite amount of PhCH MgCl at low temperature. The synthesis of 2  CpW(NO)(CH Ph) from CpW(NO)I has been previously reported from our 2  2  2  laboratories. All the nitrosyl bis(benzyl) complexes may be prepared in an analogous 12  manner from the nitrosyl diiodo precursors, but they are all best prepared from the nitrosyl dichloro species. The synthetic method for the preparation of Cp*Mo(NO)(CH Ph) (2) is 2  2  described below as a representative example. To a stirred, olive-green solution of Cp Mo(NO)Cl (1.5 g, 4.5 mmol) in THF (200 2  mL) at 0 °C was added dropwise via an addition funnel an I ^ O solution (17.4 mL, 0.52 M, 9.04 mmol) of PhCH MgCl, diluted with 20 mL of Etp. 2  The color of the reaction mixture  changed from yellow-green to red and then to a dark red color as the addition proceeded.  82 After the addition of the Grignard reagent was complete, the reaction mixture was stirred for 0.5 h at room temperature, and then 0.5 mL of deaerated H 0 was added by syringe. 2  This resulted in the formation of a sticky white precipitate, but no change in the color of the solution. The yellow white precipitate (MgCy was removed by filtration of the resulting mixture through Celite supported on a medium-porosity glass frit. The filtrate was taken to dryness in vacuo. Then CH C1 (30 mL) was added, and the resulting orange2  2  red mixture was chromatographed on an alumina column (Fisher neutral, 80-200 mesh, activity 3, 6 x 3 cm) made up in hexanes using CH C1 as eluant. The single orange-red 2  2  band that developed was collected, and the eluate was concentrated to approximately 10 mL under reduced pressure. An equal volume of hexanes was added, causing 0.69 g of orange-red crystals of Cp*Mo(NO)(CH Ph) to slowly crystallize from solution at room 2  2  temperature. These crystals were collected by filtration and dried under vacuum. Concentration of the mother liquor and cooling of the resulting solution to -20 °C overnight gave an additional 0.63 g of the desired product (66% total isolated yield), mp 152-155 °C The analytical, mass spectral, IR and H and C NMR data for this complex and all X  B  the other new bis(benzyl) complexes synthesized during this work are presented in Tables 3-V and 3-VI. In a similar manner 2.73 g (50 % yield) of orange-red CpMo(NO)(CH Ph) (la), 2  2  mp 146-149 °C, were prepared from CpMo(NO)I (6.5 g, 7.3 mmol) and PhCH MgCl (22.0 2  2  mL, 1.37 M, 30.0 mmol) at 0 °C in THF (150 mL). Large red crystals of CpMo(NO)(CH C H -2,4,6-Me ) (lb), 0.84 g (48 % yield), 2  6  2  3  2  mp 159-161 °C, were prepared from CpMo(NO)Cl (1.00 g, 1.91 mmol) and Me -2,4,62  3  CgHjCHjMgCL (68.2 mL, 0.11 M, 7.6 mmol, diluted with 20 mL of Et^O) in THF (200 13  mL) at 0 °C.  83 Similarly, 0.96 g (53% yield) of red crystals of CpW(NO)(CH Ph) (3), 2  2  mp 143-145 °C (dec), were prepared from CpW(NO)Cl (1.37 g, 1.96 mmol) and 2  PhCH MgCl (15.1 mL, 0.52 M, 7.85 mmol) at 0 °C in a THF (150 mL) and E^O (25 mL) 2  solvent mixture. Furthermore, 1.22 g (64% yield) of Cp*W(NO)(CH Ph) (4), mp 160-163 °C (dec), 2  2  as red crystals, were prepared from Cp W(NO)Cl (1.5 g, 1.8 mmol) and PhCH MgCl (11.1 2  2  mL, 0.65 M, 7.21 mmol) at 0 °C in a THF (150 mL) and EtjO (25 mL) solvent mixture.  X-ray Crystallographic Analysis of the Complexes la, lb, 2,3, and 4.  A suitable  X-ray quality crystal of each of the nitrosyl bis(benzyl) complexes was obtained by very slow crystallizations over many days from E^O solutions at -20 °C. All the X-ray structural determinations of the nitrosyl bis(benzyl) complexes were performed in a similar manner. The X-ray structure analyses of complexes la, 2,3, and 4, were performed by Drs F.W.B. Einstein and R.H. Jones at Simon Fraser University and complex lb's structure was determined by Dr. J. Trotter and Ms. V.C. Yee of our department. A summary of the pertinent crystallographic results for each complex are provided in Table 3-1. Final positional and equivalent isotropic thermal parameters (U^ = 1/3 x trace diagonalized U) for the complexes la, lb, 2,3, and 4, are reported elsewhere. Bond lengths (A) and bond 115  angles (°) for all the nitrosyl bis(benzyl) complexes are listed in Tables 3-II and 3-JJL respectively. Views of the solid-state molecular structures of CpMo(NO)(CH Ph) (la), 2  2  CpMo(NO)(CH C H -2,4,6-Me ) (lb), Cp*Mo(NO)(CH Ph) (2), CpW(NO)(CH Ph) 2  6  2  3  2  2  2  2  2  (3), and Cp*W(NO)(CH Ph) (4) are given in Figures 3.1,3.2,3.3,3.4 and 3.5, respectively. 2  2  84  Reactions of the Cp'M(NO)(CH Ph) Complexes (la, 2,3, and 4) with 0 . All of 2  2  2  these reactions involved treating solutions of the bis(benzyl) complex with an excess of dioxygen. The details of the reactions of the Cp'W(NO)(CH Ph)2 complexes with an 2  excess of dioxygen have already been presented in Chapter 2, the dioxo benzyl complexes, Cp'W(0)2(CH Ph), being isolated from these reactions in low to moderate yields. 2  Similar reactions of the molybdenum bis(benzyl) complexes Cp'Mo(NO)(CH Ph)2, 2  were attempted. The initial orange Et^O or toluene solution containing the Cp'Mo(NO)(CH Ph) complex remained unchanged after 6-12 h during the reaction with 2  2  0 . After 2 d the solution had developed a green coloration and a large amount of a green 2  precipitate had formed. No discernible proton resonances were observed in a H NMR 1  spectrum of the green residue in CD C1 . An IR spectrum of the green residue as a Nujol 2  2  mull displayed many bands in the spectral region 650-980 cm , characteristic of M=0, M-1  O and M-O-M vibrations. A low resolution mass spectrum of the green product showed no obvious mass peaks above m/z 194. HPLC studies were performed on the solution mixtures left after the completion of the reaction. At least 8 highly to moderately polar organic compounds were collected during these HPLC studies, and benzaldehyde and benzyl alcohol both appeared to be present. The identities of each of the fractions have not yet been confirmed. No Cp'Mo(0) (CH Ph) complexes were isolated from these 2  2  reactions.  Reaction of the Cp'W(NO) (CH Ph) Complexes with Alcohols. These reactions 2  2  were performed in NMR tubes in a similar manner and were monitored by *H and C B  NMR spectroscopy.  85 (a) Reaction of Cp*W(NO)(CH Ph) (4) with MeOH. 2  2  A CD C1 solution of 2  2  Cp*W(NO)(CH Ph) (0.060 g, 0.113 mmol) and MeOH (4.6 uL, 0.113 mmol) was sealed in 2  2  an NMR tube under a slight vacuum. An initial *H NMR spectrum displayed only proton resonances due to the two starting materials. The NMR tube was kept in an oil bath heated to 45 °C. After 3 weeks a H NMR spectrum showed that approximately one-half of 1  the Cp*W(NO)(CH Ph) starting material remained. New *H resonances slowly appeared, 2  2  and these could be attributed to free C H C H : *H NMR S 7.2-6.8 (m, Ph-#), 2.13 (s, 3H, 6  5  3  C H C ¥ ) ; "C^H} NMR * 129.2*128.1,125.6 (s, aromatic Ph-Cs), 21.5 (s, C H CH ) 6  5  3  6  5  3  and Cp*W(NO)(OCH )(CH Ph). 3  2  NMR data for Cp*W(NO)(OCH )(CH Ph): *H NMR s 7.32-6.8 (m, ?h-H), 2.80 3  (d, 1H, / 2  2  = H.0 Hz, C¥ H Ph), 2.51 (d, 1H, J 2  H a H b  a  b  fUHb  ~  11.0 Hz, CH tf Ph), 2.34 (s, a  b  3H, W(OG// )), 1.95 (s, 15H, C(C# ) ). C{ H> NMR 6 163.5,145.5,129.3,128.5,123.8 I3  3  3  1  5  (s, aromatic Ph-Cs), 112.9 (s, C(CH ) ), 42.6 (s, / 1  3  5  c w  = 99.8 Hz, CH Ph), 21.5 (s, 2  W(OCH )), 9.8 (s, C(CH ) ). 3  3  5  (b) Reaction of CpW(NO) (CH Ph) (3) with 'BuOH. 2  2  A CD C1 solution of 2  2  CpW(NO)(CH Ph) (0.060 g, 0.13 mmol) and 'BuOH (24.7uL, 0.26 mmol) was sealed in 2  2  an NMR tube under a slight vacuum. An initial *H NMR spectrum displayed only proton resonances due to the two starting materials. The NMR tube was then heated at 45 °C in an oil bath for one month, after which time there was no evidence of a reaction taking place. Furthermore, there was no indication that decomposition of the CpW(NO)(CH Ph) starting material had occurred. 2  2  86 Results and Discussion A.  Synthesis and Some Physical Properties of the Nitrosyl Bis(benzyl) Complexes:  Cp'M(NO)(CH Pli ) , (Cp'= Cp or Cp*; M= Mo or W; Ph'= C H or C H -2,4,6-Me3). l  2  2  fi  5  6  2  The method for the preparation of these complexes is analogous to that previously described for the synthesis of the series of thermally stable, 16-electron dialkyl nitrosyl complexes Cp'M(NO)R , (Cp = Cp or Cp*; M = Mo or W; R= Me, CH SiMe , 1  2  2  CH CMe , or CH CMe Ph). ' 12  2  3  2  14  2  3  Treatment of either the nitrosyl diiodo or dichloro  precursors, Cp'M(NO)X (Cp = Cp or Cp*; M = Mo or W; X = CI or I) with the 6  1  2  requisite amount of a benzyl Grignard reagent, Ph'CH MgCl (Ph'= C H or C H -2,4,62  6  5  6  2  Me ), at 0 °C affords the nitrosyl bis(benzyl) complexes in good isolated yields (50-65%), 3  i.e. eq. 2.  Cp'M(NO)X + 2 Ph'CH MgCl 2  2  1. THF/Et-0,0 °C 2. HjO.R.T.  Cp'M(NO)(CH Ph') 2  2  + 2 CIMgX (2)  There is no evidence in either of the reactions 2 for the formation of the intermediate iso-nitrosyl complexes, [Cp'M(NO)(CH Ph') ]MgXCl. '  12 14  2  2  During the course  of the reactions, the reaction mixtures change from brownish-green to a dark red color and a white to yellow solid precipitates from the reaction solution. Furthermore, the H 0 2  destroys the excess Grignard reagent present after the completion of the reaction. In order to simplify the work-up, all of the final reaction rnixtures are first filtered through Celite to remove the insoluble CIMgX by-product. Then, the reaction mixture is chromatographed through alumina to separate the remaining CIMgX by-product from the desired bis(benzyl) complexes. Recrystallization of the solids obtained after the work-up from CH Cl /hexanes at -20 °C affords the bis(benzyl) complexes as analytically pure crystals. 2  2  87 All the nitrosyl bis(benzyl) complexes prepared in this work are high melting (143-163 °C), orange-red diamagnetic solids which are soluble in most common organic solvents such as benzene, toluene, and CH C1 . They are, for the most part, insoluble in hexanes or 2  2  pentane, but the tungsten complexes show some solubility in these non-polar solvents, and they are all moderately soluble in Et^O. As expected, the Cp* compounds exhibit greater solubilities than their Cp analogues. Unlike their dialkyl analogues Cp'M(NO)R (M= Mo or W; R= CH SiMe , 2  2  3  CH CMe , and CH CMe Ph) which are air-sensitive in solutions and in the solid state,  2  2  3  2  2  these bis(benzyl) complexes, Cp'M(NO)(CH Ph') exhibit limited reactivity with dioxygen. 2  2  That is, they are all stable to air in the solid-state for at least 2 months and only the tungsten analogues react with 0 in solution to afford low isolated yields of the dioxo 2  benzyl complexes, Cp'W(0) (CH Ph'). Furthermore, reactions of Cp"Mo(NO)(CH Ph') 2  2  2  2  complexes with dioxygen do not afford the Cp'Mo(0) (CH Ph') complexes, and no 2  2  organometallic molybdenum complexes are isolable from the final reaction mixtures. Because these bis(benzyl) compounds display such limited reactivity with dioxygen, the molecular structures of la, lb, 2,3, and 4 have been determined, and the results of these studies will be discussed in the following section. A discussion of the spectroscopic properties of these bis(benzyl) complexes, as they pertain to the molecular structures, will follow.  B.  Typical Benzyl Ligand Bonding Modes to Transition-metals Reported to Date in the  Literature. The benzyl ligand, C H C H , is a particularly interesting alkyl group because of its 2  6  5  potential ability to engage in both metal-carbon a bonding and metal-ligand r -bonding, the latter involving arene p * -orbital overlap with empty d x-orbitals on the transition-metal center. Huckel calculations on the density of the charge in a benzyl radical have shown 15  88 that approximately 4/7 of the available charge resides on the methylene carbon atom of th  the benzyl group, while l/l^ of the available charge resides on each of the two orthocarbon atoms and the single />ara-carbon atom. Therefore, the results suggest that the 16  metal-to-methylene-carbon linkage will be the dominant interaction between the metal and the benzyl group. The i/wo-carbon and the two mera-carbon atoms have no negative charge contributions in the bonding of the benzyl ligand, and therefore are not expected to interact significantly with the metal center. As depicted below, five types, of bonding modes of the benzyl ligand to metal centers have been reported in the Uterature, to date. The bonding mode of the benzyl ligand  M—CH  M  2  £  H  2  M—CH  2  M—CH 2  differs primarily in the extent of the arene p r -orbital overlap with the metal d* -orbitals. This interaction is strongly dependent on the nature of the M L fragment to which the n  benzyl ligand is bonded, as will be demonstrated again in this work. The r? - C H C H 1  2  6  5  bonding mode, the most common, involves only a strong metal-carbon a interaction. This mode of linkage is found in many electronically sufficient, coordinatively saturated transition-metal benzyl complexes.  17  The ?j -bonding mode, on the other hand, is typically 3  found in coordinatively unsaturated, electron deficient complexes, and is seen for many middle and late transition-metal complexes. '  18 19  In these complexes, the benzyl ligand is  considered to be a 3-electron donor to the metal center. In some demonstrated cases, the benzyl ligand can act as a trapping ligand, shifting from rj - to J? -, thereby stabilizing the 1  3  89  16-electron coordinatively unsaturated complex generated as a result of ligand loss (eg. C O or phosphine) via donation of the extra arene p x-electron density to the metal center.  19  Unlike the allyl ligand, C3H5, which is known to bind in either a symmetrical or unsymmetrical 77 •'-fashion, the r? -benzyl group always binds to the metal in an 20  3  unsymmetrical fashion, consistent with the results of the molecular orbital calculations, discussed above. Structural studies of many high valent, early transition/actinide metal tri- and tetrabenzyl complexes (i.e. those of Ti, > Z r , ' Hf, 21a  c  21b  c  21c  Th  1 9  ^ and U  21e  ) have revealed some  unusual benzyl linkages. In these multi-benzyl group complexes, most of the benzyl ligands are ^-bonded to the metal, but one or two are unique in that they exhibit M-C(i ) pso  distances that are within the range typically found for M - C single bonds. As a result of the Huckel calculations, weak interactions of the central metal with the two orf/to-carbon 15  atoms have been proposed. A trivial amount of electron density is thought to be carried by the /pso-carbon atom, and therefore, the M-C(i ) distances were considered to be short in ps0  order to get the two M-C( i! ) distances to be within bonding distances. As a result of ort  0  these proposed weak metal to orf/io-carbon interactions, some of the benzyl groups in these early metal complexes are classified as »? -benzyl ligands. It is, however, difficult to predict 4  the extent of the M-C( h ) interactions, as many of the M-C( }, ) distances in these ort  0  ORT  0  complexes are longer than typical M - C single bond lengths and values for the van der Waals atomic radii are not known for most transition-metals. It is possible that, if there are truly M-C( h ) interactions in these complexes, they may be electrostatic in nature. ort  0  In any event, at the limit where the M-C( h ) distances are too long to be bonding, ort  0  but where the M - C ^ Q ) distance is within normal bonding ranges, an r? -benzyl ligand 2  would result. The only documented example of this to date occurs in the complex [Cp2Zr(77 -CH2C H5)(NCCH3)] [BF ]-, recently reported by Jordan and co-workers. 2  +  6  22  4  The nitrosyl bis(benzyl) complexes described in this work will be added to this rare class of unusual rj -benzyl complexes. These types of benzyl ligands are unusual because the 2  90  previous Huckel calculations on the benzyl radical or anion do not explain the metal-benzyl interactions observed. Recently, an example of the r; -bonding mode of the benzyl ligand has been 5  proposed as the excited state generated during the rapid equilibrium in solution for the rhodium complex Rh(r -CH2C Me )[P(OPr )3]2. 3  i  ?  6  5  18(f)  Precedence for this »j -mode of 5  bonding has been observed in the structurally characterized complex, CpFefy 5  CH^QMes), which may alternatively be viewed as containing a fancy r? -cyclohexadienyl 23  5  ligand.  C.  Solid-State Molecular Structures of the Bis(benzyl) Complexes la, lb, 2,3 and 4. A single crystal of CpMo(NO)(CH Ph) , (la), suitable for an X-ray crystallographic 2  2  analysis was obtained from slow crystallization of the compound from Et20 solutions. The structure determination, the pertinent crystallographic results of which are summarized 115  in Table 3-1, establishes that the organometallic complex la is not a 16-electron coordinatively unsaturated compound as previously found for its CpW(NO)(CH2SiMe3)2 analogue.  12  Exchanging the bulky alkyl groups of the CpM(NO)R2 complexes (R =  CH SiMe , CH CMe3, or CH CMe2Ph) for a benzyl ligand allows the metal center to 2  3  2  2  accept the extra electron density available on this ligand, and to therefore, adopt a thermodynamically stable 18-electron configuration. However, it is the way in which the benzyl group is bound to the metal center, see below, that warrants a discussion of the molecular structure of these bis(benzyl) complexes. The intramolecular dimensions extant in these bis(benzyl) complexes will also be compared to related benzyl complexes in the literature. The bis(benzyl) complex la, as seen in the thermal ellipsoid plot in Figure 3.1, possesses a pseudo-4-legged piano-stool molecular structure, which consists of an essentially linear nitrosyl ligand (Mo-N-O angle of 174.1(3)° ), a "normal" f? -CH Ph ligand, 1  2  91  Table 3-1. Pertinent Crystallographic Data for the Complexes Cp'M(NO)(CH Ph') (Cp' = Cp or Cp*; M= Mo or W; Ph'= C H , or CgHyMej), la, lb, 2,3, and 4. 0  2  6  Complex  la  2  5  lb  2  3  4  formula  C H NOMo  C^r^NOMo  C^H^OMo  C H NOW  W o w  fw  37331  457.47  445.43  46122  53135  Cryst system  orthorhombic  triclinic  monoclinic  orthorhombic  monoclinic  space group  Pbca  PI  P2 /c  Pbca  Pljc  (A)  10.122(1)  9.4113(19)  13.969(5)  10.110(1)  13.916(4)  b(A) c(A)  15.625(2)  83499(18)  16.125(3)  15.614(2)  16.101(3)  20.770(2)  15.4533(18)  9.778(5)  20.654(2)  9.757(3)  a (deg)  ....  98.004(18)  —  —  —  0 (deg)  —  100.408(15)  110.67(4)  —  110.08(2)  7 (deg)  —  110511(16)  —  ....  —  V(A )  3284.9  10913(4)  2060.7  3260.4  20533  Z  8  2  4  8  4  0.36 x 0.38  0.23x0.24  0.15x0.15 x032  0.34x037  0.34x0.20  0.029  0.025  a  19  3  cryst dimens (mm )  19  xO.34  x0.21  3  0.024 R  0.031  0.031  0.039  s  f  resid dens  (e/A ) 3  1.218 0.29(5)  19  x032 0.028  0.030  1312  1.179  0.813  052(7)  1.286 0.9(2)  1.2120 0.80(11)  d  d  Enraf-Nonius CAD4-F diffractometer, Mo Ka radiation graphite monochromator. b  R =E I I o I l c I I / I o I • R =[Sw(| F j - I F j ) / E F ] ' / F  _  E  F  F  F  c  2  2  wF  2  0  d  w= [(CT(F)) + 0.0003 F ]- .  6  w= [a (F)]- .  2  2  2  1  1  f S (goodness of fit) = [Ew(| F | -1 F | ) / (No. of degrees of freedom)] / . 2  Q  xO.38 0.021  0.038  e  d  wF  x  19  c  1 2  d  92  Figure 3.1. A view of the molecular structure of la.  93  and a second benzyl ligand bonded to the molybdenum atom in an unusual fashion. The pertinent intramolecular dimensions for la are listed in Tables 3-II and 3-IU. The ligands of the piano stool are bent down from the plane bisecting the Mo atom and parallel to the plane of the Cp ligand, as evidenced by the angles Cp(ccntroid)-Mo-C(21) 110°, Cp(centroid)-Cp(centroid)-Mo-C(ll)^ 111.3°, Cp(centroid)-Mo-C(12)= s 135.1°, Cp(centroid)Mo-N= 124°, andC(21)-Mo-C(ll)= 127.8°. Three of the ligands in the four-legged pianostool are essentially atrightangles to each other, C(ll)-Mo-N= 91.3(2)° and N-MoC(21)= 91.2(2)°. The "normal" r; -CH Ph group is bonded to the Mo atom with a Mo-C 1  2  single bond length of Mo-C(21) = 2.251(3) A and with nearly tetrahedral geometry around the methylene-carbon atom C(21), Mo-C(21)-C(22)= 114.4°, expected of an sp carbon 3  atom. The unique benzyl ligand in la appears to involve bonding of only the methyleneand the ipso- carbon atoms to the Mo center and therefore, is best represented as an n 2  CH Ph ligand. The Mo-CH bond length of 2.203(4) A is significantly shorter than the 2  2  Mo-CH distance in the »7 -CH Ph ligand, indicating a strong interaction between the 1  2  2  methylene-carbon of the benzyl group and the central Mo atom. The Mo-C(j ) bond pso  length of 2.433(3) A is well within the range commonly observed for Mo-C single bonds. The two Mo-C^rtho) distances of 2.808(5) A and 3.236(4) A, however, are much longer than those typically found for normal Mo-C bonds. These two distances, being 14  significantly inequivalent, establish that the benzyl ligand is slightly twisted. All the carbon atoms within each of the two benzyl ligands, however, are coplanar. With an aim to determining the effect of steric and electronic factors on the mode of linkage of the fj -benzyl ligand in la, a single-crystal X-ray crystallographic analysis was 2  performed on each of the complexes CpMo(NO)(CH CgH -2,4,6-Me3) , (lb), and 2  2  2  Cp*Mo(NO)(CH Ph) , (2). Table 3-1 contains a summary of the pertinent crystallographic 2  2  details for these complexes. A view of the molecular structure of each of the complexes lb and 2 is given in Figures 3.2 and 3.3, respectively. Furthermore, with a view to gaining  Table 3-II. Bond Lengths (A) for the Complexes CyM(NO)(CH Ph) (Cp'= Cp or Cp*; M= Mo or W) (la, 2,3, and 4), and CpMo(NO)(CHC H -2,4,6-Me ) , (lb). 2  i  Complex M-N M-C(l) M-C(2) M-C(3) M-C(4) M-C(5) M-C(ll) M-C(12) M-C(13) M-C(17) M-C(21) N-O C(l)-C(2) C(l)-C(5) C(l)-C(6) C(2)-C(3) C(2)-C(7) C(3)-C(4) C(3)-C(8) C(4)-C(5) C(4)-C(9) C(5)-C(10) C(ll)-C(12) C(12)-C(13) C(12)-C(17) C(13)-C(14) C(14)-C(15) C(15)-C(16) C(16)-C(17) C(21)-C(22) C(22)-C(23) C(22)-C(27) C(23)-C(24) C(24)-C(25) C(25)-C(26) C(26)-C(27)  lb'  la 1.755(2) 2303(4) 2303(4) 2358(4) 2.410(4) 2374(4) 2.203(4) 2.433(3) 3236(4) 2.808(5) 2251(3) 1203(4) 1399(6) 1366(8)  1.764(2) 2305(4) 2292(4) 2360(5) 2.433(5) 2395(5) £196(3) 2.483(3) 3.172(3) 3.101(3) 2259(2) 1207(3) 1364(6) 1369(6)  1361(9)  1375(6)  1374(9)  1.409(7)  1399(8)  1382(6)  1.445(6) 1.411(5) 1398(6) 1351(6) 1367(7) 1361(9) 1380(9) 1.472(4) 1383(4) 1382(5) 1381(9) 1340(9) 1371(9) 1394(6)  1.452(4) 1.422(4) 1.422(4) 1384(5) 1374(5) 1391(4) • 1385(4) 1.486(3) 1.409(4) 1.412(3) 1386(4) 1382(5) 1373(4) 1389(4)  6  2  2  3 2  3*  2 1.773(3) 2321(4) 2314(4) 2390(3) 2.448(3) 2.422(3) 2.191(4) 2493(4) 3250(4) 2.949(5) 2231(3) 1208(4) 1.435(5) 1.433(5) 1.495(5) 1.416(5) 1.492(6) 1.415(5) 1308(6) 1.423(5) 1308(6) 1.481(5) 1.457(6) 1.413(6) 1393(6) 1375(7) 1384(7) 1377(7) 1389(7) 1.497(5) 1397(6) 1393(5) 1392(6) 1377(7) 1365(7) 1385(6)  1.755(5) 2370(7) 2302(7) 2306(8) 2365(7) 2409(8) 2.179(7) 2444(6) 3209(16) 2.938(14) 2223(5) 1215(7) 1388(12) 1379(12) 1.409(11) 1369(14) 1386(13)  1.441(10) 1.400 1.461 1364 1.430 1356 1.410 1.486(7) 1389(8) 1377(9) 1379(10) 1348(14) 1357(13) 1394(10) c  c  c  c  c  c  4 1.752(3) 2311(3) 2318(4) 2398(4) 2.449(4) 2.417(4) 2.173(4) 2510(4) 3.261(4) 3.015(4) 2.223(4) 1239(5) 1.423(5) 1.441(5) 1512(5) 1.427(5) 1.497(6) 1.429(5) 1.493(5) 1.423(5) 1.483(6) 1.482(5) 1.483(6) 1.405(6) 1389(7) 1370(7) 1394(8) 1377(7) 1394(8) 1505(5) 1.414(6) 1396(5) 1381(6) 1387(7) 1384(7) 1392(6)  * Average bond distances for PhC-CH, being 1504(7)A and 1508(15)A. W-C(m) = 3280(24), WC(H7)= 2668(19), C(12)-C(113)= 1.447, C(12)-C(117) = 1355, C(113)-C(114) = 1337, C(114)-C(115) = 1385, C(115)-C(116)= 1385, C(116)-C(117)= 1372. a restraint was applied to this bond b  c  95 Table Mil. Bond Angles (deg) for the Complexes Cp'M(NO)(CH Ph) (Cp'= Cp or Cp*; M= Mo or W) (la, 2,3, and 4), and CpMo(NO)(CH C H -2,4,6-Me3) , (lb). 2  2  Complex C(ll)-M-N C(12)-M-N C(12)-M-C(ll) C(21)-M-N C(21)-M-C(ll) C(21)-M-C(12) M-N-O C(5)-C(l)-C(2) C(6)-C(l)-C(2) C(6)-C(l)-C(5) C(3)-C(2)-C(l) C(7)-C(2)-C(l) C(7)-C(2)-C(3) C(4)-C(3)-C(2) C(8)-C(3)-C(2) C(8)-C(3)-C(4) C(5)-C(4)-C(3) C(9)-C(4)-C(3) C(9)-C(4)-C(5) C(4)-C(5)-C(l) C(10)-C(5)-C(l) C(10)-C(5)-C(4) C(12)-C(ll)-M C(ll)-C(12)-M C(13)-C(12)-M C(13)-C(12)-C(ll) C(17)-C(12)-M C(17)-C(12)-C(ll) C(17)-C(12)-C(13) C(14)-C(13)-C(12) C(L5)-C(14)-C(13) C(16)-C(15)-C(14) C(17)-C(16)-C(15) C(16)-C(17)-C(12) C(22)-C(21)-M C(23)-C(22)-C(21) C(27)-C(22)-C(21) C(27)-C(22)-C(23) C(24)-C(23)-C(22) C(25)-C(24)-C(23) C(26)-C(25)-C(24) C(27)-C(26)-C(25)  cravc^-cra  la  lb  127.7(1) 91.8(1) 174.1(3) 107.8(4)  9338(2) 100.6(1) 35.48(10) 94.80(11) 123.4(1) 88.04(9) 1665(2) 108.7(4)  108.0(5)  108.6(4)  108.9(5)  1073(4)  1073(5)  106.9(4)  108.2(4)  108.4(4)  80.8(2) 633(2) 111.9(2) 1243(4) 90.1(2) 119.8(4) 115.6(4) 122.0(4) 120.1(4) 121.2(5) 118.9(4) 1223(4) 114.4(2) 1213(3) 121,7(3) 117.0(3) 121.4(4) 121.1(4) 119.4(4) 1203(5) *21.0(4)  83.1(2) 61.42(15) 105.1(2) 120.7(3) 101.7(2) 1215(3) 1173(3) 1193(3) 123.1(3) 117.7(3) 1223(3) 119.7(3) 121.0(2) 1213(2) 120.6(2) 1183(2) 119.6(3) 122.4(3) 117.6(3) 1225(3) H95(3)  913(2) 92.9(1) 35.9(2)  912(2)  6  2  2  2  2 933(2) 945(1) 355(1) 92.9(2) 1243(2) 893(1) 170.0(3) 107.7(3) 125.4(3) 1263(3) 107.7(3) 126.1(4) 126.0(4) 108.6(3) 124.4(4) 1265(4) 108.4(3) 125.1(4) 1263(4) 1075(3) 125.7(3) 126.4(4) 83.6(2) 60.9(2) 1095(3) •121.9(4) 945(2) 119.6(4) 118.1(4) 1203(4) 1205(4) 120.6(4) 1193(4) 1213(4) 1215(3) 1225(4) 121.6(4) 115.9(4) 1215(4) 1203(4) 118.8(4) 120.7(4) 122,4(4)  4 93.7(2) 94.9(1) 36.0(2) 94.0(2) 123.0(2) 87.1(0) 169.7(3) 1085(3) 1253(4) 125.6(3) 1073(3) 126.7(4) 125.9(4) 108.8(3) 125.1(4) 125.8(4) 1073(3) 1253(4) 126.6(4) 107.6(3) 125.6(3) 1265(4) 845(2) 595(2) 109.6(3) 1213(4) 97.1(3) 1193(4) 1183(4) 120.9(4) 1195(4) 120.7(4) 119.7(4) 120.4(4) 120.4(3) 122.6(3) 120.6(4) 1163(4) 120.7(4) 1215(4) 118.7(4) 1203(4)  92.1(3) 935(2) 35.7(3) 91.7(2) 1263(3) 905(2) 174.7(5) 108.6(8)  107.1(8)  107.7(9)  108.9(8)  107.8(9)  823(4) 62.1(4) 1103(9) 120.1 943(6) 124.4 1153(8) 1203 1233 1195 117.7 * 124.0 1143(4) 121.0(6) 121.7(5) 117.1(6) 121.7(8) 120.1(7) 119.9(7) 120.7(8) 1205^ b  fc  b  b  b  b  mm  C(113)-C(12)-W= 1123(12), C(113)-C(12)-C(ll)= 129.4, W-C(H7)-C(12)-W-= 83.9(9), C(117)-C(12)C(ll)= 106.7, C(117)-C(12)-C(113)= 1233(11), C(H4)-C(113)-C(12)= 1223, C(115)-C(114)-C(113)= 1143, C(116)-C(115)-C(114)= 122.9, C(117)-C(116)-C(115)= 124.0, C(116)-C(117)-C(12)= 112.6. arestraintwas applied to this angle.  a  b  96  Figure 32. A stereo-view of the molecular structure of lb.  97  C<6)  C(14)  Figure 3.3. A view of the molecular structure of 2.  98 insight into the difference in the reactivity observed for the tungsten versus the molybdenum bis(benzyl) complexes with molecular oxygen, structural determinations were performed on the corresponding tungsten bis(benzyl) complexes, namely CpW(NO)(CH Ph) , (3) and 2  2  Cp W(NO)(CH Ph) , (4). A view of each of the molecular structures possessed by these 2  2  tungsten complexes, 3 and 4, is shown in Figures 3.4 and 3.5, respectively. From the Figures 3.1 through 3.5, it can be seen that all the nitrosyl bis(benzyl) complexes prepared in this work are isostructural, even though they crystallize in different space groups (Table 3-1). The intramolecular bond lengths and bond angles for all the complexes la, lb, 2,3, and 4 are summarized in Tables 3-II and 3-JJI, respectively. The intramolecular dimensions in the molecular structures of each complex are consistent with each being formulated as (^'MtNOX^-CH^hX^-CH^h), i.e.  A rationale for their being formulated as diamagnetic 18-electron species will be presented later. As observed previously in la, three of the 4 legs in the piano-stool structure in lb, 2, 3, and 4, are essentially atrightangles to each other (that is C(ll)-M-N and C(21)-M-N remain near 90° from la to 4). The nitrosyl ligand in each structure is essentially linear, the largest deviation from linearity being 166.5° in lb, which is likely due to either the result of more electron density at the metal center, consistent with the CH C H -2,4,6-Me group 2  6  2  3  being a better a -donor than CH Ph, or perhaps steric interactions of the NO ligand with 2  the two triply-substituted phenyl rings. One of the benzyl ligands in each of the complexes  Figure 3.4. A view of the molecular structure of 3.  100  cas)  CU2>  CC16)  C<11)  C(?>  CC25)  Figure 3.5. A view of the molecular structure of 4, showing the planarity of the »? -CH Ph ring with respect to the central tungsten atom. 2  2  101  is normal by virtue of the observed M-CH single bond length between 2.22 and 2.26 A, 2  and the nearly tetrahedral angle for M-C(21)-C(22) angle in each structure (114.4°, 121.0°, 121.5°, 114.2°, and 120.4°). The »? -benzyl ligand in each structure is characterized by, (a) 2  the significantly shorter M-CH bond length (0.02-0.04 A) as compared to that of the r? 1  2  CH Ph group, (b) a significantly shorter CH -C(i ) bond length than in the r? -CH Ph 1  2  2  pso  2  group (for instance C(ll)-C(12)= 1.445(6) and 1.457(6) A and C(21)-C(22)= 1.472(4) and 1.497(5) A in complexes la and 2 respectively), (c) M-C(i ) distances that are within the ps0  range for normal Group 6 metal-carbon single bonds, (d) an angle at the methylenecarbon atom, C(12)-C(ll)-M (80.8-84.5 °), that is well below the value expected for a tetrahedral sp -carbon, and (e) two M-C( ]i ) distances (between 2.81 and 3.26 A) that 3  ort  0  are longer than the range observed for Group 6 metal-carbon single bonds. D.  Comparison of the Intramolecular Dimensions of the Bis(benzyl) Complexes la, lb,  2, 3, and 4, with Those Found in Related Transition-Metal-Benzyl Complexes. It is of interest to compare the intramolecular dimensions of the M-(r? -benzyl) 2  linkage in the molecular structure of each complex la, lb, 2,3, and 4 with those of structurally characterized metal n -, n -, and r? -benzyl linkages reported in the literature. 2  3  4  Such a comparison might permit a better understanding of why the bonding is the way it is in the different complexes and how this might affect the subsequent reactivity of these complexes. Table 3-IV summarizes some of the metal to methylene-carbon, metal to ipsocarbon, and metal to orr/io-carbon bond lengths in several structurally characterized n n  benzyl (n= 2, 3, or 4) complexes. The structure depicted below provides a reference for 24  the labelling used in representing the metal-benzyl link.  M—CH  C  (ortho)  2  Ci(ortho*)  102 Table 3-IV. A Comparison of the Intramolecular Dimensions in Structurally Characterized Metal-Benzyl Complexes. 0  Benzyl Complex  Bond Lengths (A)  Of Late Transition Metals Ni(r> ^ C H ^ H ^ - M e J C ^ t P M e ^  1.930  2.050  2318  h  0.12  large  19  Pt^-CPhjXacac)  2.088  2.120  2.148  3.081  0.028  0.96  18d  Pd(r/ -CPh3)(acac)  2.105  2.154  2.200  3.132  0.046  0.998  18d  2.036  2.117  2.408  h  0.029  large  18e  2.128  2.246  2.453  h  0.21  large  18f  2.11  2.16  2.22  h  0.067  large  18b  2.24  232  234  h  0.019  large  18c  Mo(r? -CH Ph-/?-Me)Cp(CO)  2.27  2.36  2.48  331  0.115  0.945  18a  Mo(r? ^CHjPhXr/ -CH Fh)Cp(NO)  2.203  2.433  2.808  3.236  0375  0.803  Mo(r? -CYLpV)(n ^CHjPh'JCpCNO)  2.196  2.483  3.101  3.172  0.618  0.689  Mo(r? ^ O L / h ) ^ ^CHjP^Cp'CNO)  2.191  2.493  2.949  3.250  0.456  0.757  W(r? -CH2Ph)(rj ^ C H ^ C p T N O )  2.179  2.444  2.938  3.209  0.494  0.765  Wft -CH Yh)(j} -CH Fh)Cp*(NO)  2.173  2.510  3.015  3.261  0505  0.751  2344  2.648  3236  3.25  059  0.60  22  Th(CH2Ph) Cp*  2579  2.865  3352  3574  0.48  0.71  21d  ^(CHjPh^Cdmpey  253  2.86  331  3.44  0.43  058  21e  UCCHjPhJjMeCdmpeV  254  2.758  3.089  3.450  033  0.69  21e  3  Cofa -CH Ph)[P(OMe) ] 3  2  3  3  Rh(r? -CH C Me )[P(0-i-Pr)3] 3  2  6  5  2  Of Middle Transition Metals Fe (DPF)(CO) ' 2  5  Ru (DPF)(CO) 2  i 5  3  2  2  1  2  2  2  2  1  2  2  Of Early Transition/ Actinide Metals [Zr(r? -CH2Ph)Cp (N=CCHj)] 2  2  3  a  +  The data listed are with respect to the r; "-benzyl group (n= 2,3, or 4) in each of the complexes.  methylene carbon bond length. bond length or distance. [MC  (ipso)  -MCH ]. 2  8  e  c  The metal to ipso-carbon bond length.  d  The shortest metal to o/t/io-carbon  The longest metal to o/rto-carbon bond length or distance.  [MC^^-MCHj] - [MC^-MCrLJ.  h  f  [MC^j^-MCPLJ -  This bond distance was not reported (can be  calculatedfromthe atomic fractional coordinates reported for the complex). dmpe= Me^CHjCP^PMej.  The metal to  1  DPF= /i-ij ,r; -C(C H ) (C H ). 3  5  6  5  2  5  4  103  The first documented example of the benzyl group showing the capacity to interact with a transition metal's empty d r -orbitals via the p x-orbitals on the aromatic ring was observed in the r; -benzyl complex CpMo(CO)2(»? -CH2QH4-/>-Me), isolated by Cotton 3  3  and La Prade in 1968.  18a  Since then many other transition metal j? -benzyl complexes 3  have been structurally characterized.  18,19  In all such r? -benzyl complexes, the benzyl 3  group is bound to the metal in an unsymmetrical fashion, with the metal carbon bond lengths increasing in the order M-CH < M-C(i ) < M-C( h ), (Table 3-IV), the most 2  pso  ort  0  important features being that the M-C(; ) bond lengths typically range from 2.0 to 2.4 A pso  and the M-C( h ) bond lengths are slightly longer, i.e. 2.2 to 2.5 A. The non-bonding Mort  0  C(ortho) distances usually range from 3.0 - 3.3 A. The M-C(i ) bond lengths in la, lb, 2,3, and 4 (2.43-2.51 A) are similar to the Mpso  C(ortho) bond lengths in the »? -benzyl complexes (Table 3-IV). However, the closest 3  metal to orr/io-carbon distances in la, lb, 2, 3, and 4 (2.8 - 3.1 A) are significantly longer (0.3 - 0.6 A) than the M-C( h ) bond lengths (2.15-2.48 A), in typical fj -benzyl complexes 3  ort  0  (Table 3-IV). For instance, the closest Mo-C( h ) distance in la (2.808 A) is 0.40 A ort  longer than the corresponding M-C(  0  ) bond length in the most unsymmetrical »? -benzyl 3  ortno  complex known, namely, Co(r? -CH Ph)[P(OEt)3] , (Table 3-IV). The difference 3  18e  2  between the M-C(  ortno  3  ) bond length and the M-CH bond length, [(M-C( h ))-(M-CH2)], 2  ort  0  in the rj -benzyl complexes typically lies within the range of 0.1-0.4 A. This compares to 3  the range of 0.6-0.9 A for [(M-C(  ortho  ))-(M-CH )] in la, lb, 2,3, and 4. The metal to non2  bonding orr/io-carbon distance in »? -benzyl complexes is always longer than 3.0 A, the 3  difference [MC( h ')-MCH2] being close to or greater than 1.0 A. The significant ort  0  difference between the two values A and A', in Table 3-IV for the >?-benzyl complexes, is 3  indicative of the large twist of the benzyl group with respect to the transition-metal center. In complexes la, lb, 2,3, and 4 these two values are nearly equivalent, meaning the benzyl group is not twisted nearly as much with respect to the metal.  104 Also, in virtually all of the structurally characterized ??-benzyl complexes, bond 3  length alternation is observed in the aromatic ring of the benzyl group as a result of the arene * -electron interaction with the metal's empty d orbitals. The arene ring derealization is disrupted to such an extent that theringdimensions approach those of a os-l,3-butadiene group, which has a single C-C bond length of 1.47 A and C=C double bond lengths of approximately 1.34 A. M  CH,  In contrast, the arene C-C bond lengths in the r? -benzyl ligand in each of the complexes 2  la, lb, 2,3, and 4 do not exhibit this alternation in length. The two C^ j-C^ pso  ortho  ^ bond  lengths, C(12)-C(13) and C(12)-C(17) of the r? -benzyl ligand are slightly longer (~ 1.392  1.41 A) than the other four C-C bond lengths within the arene ring [ C(13)-C(14), C(14)C(15), C(15)-C(16), and C(16)-C(17) ], which are all approximately equal  1.35-1.38 A)  suggesting some degree of % electron delocalization over most of the ring. M—CH  2  The above features clearly establish the structural differences between the bis(benzyl) complexes prepared in this work and typical r/ -benzyl complexes reported in the literature. 3  Replacement of the two carbonyl ligands in CpMo(CO)2v7 -CH C H -/7-Me), 3  2  6  4  (Table 3-IV), with a nitrosyl group and an alkyl ligand (in this case another benzyl group) to give the isoelectronic complex la, results in a clearly different mode of binding of the benzyl group to the molybdenum center. This is interesting in light of what is expected based on electronic arguments. The presence of the nitrosyl group is expected to cause a  105  more electrophilic character to the molybdenum metal center in la than in the complex CpMo(CO) (r? ^CHzQrL^-Me) since the NO ligand is a stronger electron-withdrawing 2  group than the CO ligand. As a result, one might have anticipated observing a larger interaction of the molybdenum's empty d orbitals with the electron density on the arene ring of the benzyl ligand in compound la (i.e. an »? -benzyl group). Clearly, other factors 3  are significant in the bonding. The intramolecular dimensions of the M-(»7 -benzyl) linkage in each of the 2  structures la, lb, 2,3 and 4, closely resemble those observed in the only other rj -benzyl 2  complex [Cp Zr(N=CCH3)(r -CH Ph)] [BF4]-, recently reported by Jordan et al, (Table 2  2  ?  +  22  2  3-IV). The M-C(i ) bond length in each complex la, lb, 2,3 and 4 is significantly shorter ps0  than the corresponding bond length in the zirconium benzyl complex. The M-C( ii ) ort  0  distances however, are very similar. Furthermore, as listed in Table 3-IV, the intramolecular dimensions of the M-(»7 -benzyl) linkage in each of the structures la to 4, 2  are also similar to those observed in the »? -benzyl ligands of the early transition metals Ti, 4  Zr, and Hf and the actinide metals Th and U .  21  Some of the M-C(j ) bond lengths in pso  these early metal benzyl complexes are short enough to be consistent with M-C single bonds ([MC( )-MCH2] = 0.22 - 0.35 A). The true nature of the hapticity of some of the ips0  benzyl groups in these early metal complexes is not unambiguous since van der Waals atomic radii are not known for these metals, and therefore, the maximum expected M-C bond length(s) cannot be calculated. Moreover, the M-C( ]j ) distances in the »; -benzyl 4  ort  0  linkages are often longer than might be expected, and the differences in bond lengths [MC^rth^-MCHJ and [MC( i )-MC(ip )] usually range between 0.55-0.90 A and ort 10  SO  0.3-0.5 A, respectively (Table 3-IV). Unlike the »? -benzyl complexes, the values A and A' 3  are nearly equivalent in each of these early transition/actinide metal benzyl complexes as they are in la, lb, 2, 3 and 4, which are indicative of the benzyl ligand being nearly planar with respect to the central metal atom. Clearly the comparisons described above suggest that theoretical calculations are needed to determine which metal and benzyl group  106 molecular orbitals are involved in the bonding in these unusual transition metal rj - and r? -benzyl complexes. 4  A comparison of the intramolecular dimensions in each of the complexes la, lb, 2,3 and 4, reveals subtle variations in the linkage of the ??-benzyl fragment depending on the 2  Cp versus Cp* ligand, the Ph versus C H -2,4,6-Me group and on the Mo versus W atom. 6  The difference between the M - C and the M-C(  ortho  (ipso  2  3  ) and the M-CH bond lengths [(M-C^ 2  ^-(M-CHj)],  ) and the M-CH bond lengths [(M-C^ ))-(M-C(H ))], and the size of 2  ortho  2  the bond angles C(12)-C(ll)-M and C(17)-C(12)-M, increase as the Cp ligand is substituted with a Cp* group and the central metal atom is changed from Mo to W. These changes in the intramolecular dimensions are consistent with the benzyl group moving away from the metal atom at the ipso-caibon. This is also manifested by a slight increase in the angle between the nitrosyl ligand and the normal r/ -benzyl group (cf. C(21)-M-N for 1  la, lb, 2, 3, and 4 of 91.2, 94.8, 92.9, 91.7, and 94.0, respectively). Furthermore, the two metal otf/iocarbon distances of the unique benzyl ligand in each structure slowly become closer in length in going from la to 4. That is, the M-C^  ortno  ^ distances, 2.808(5) A and  3.236(4) A in la, lengthen to 3.015 A and 3.261 A in 4, suggesting that as the benzyl group moves away from the central metal atom, it also slowly becomes less twisted with respect to the metal atom.  E.  Spectroscopic Properties of the Nitrosyl Bis(benzyl) Complexes. The analytical and spectroscopic data for the new bis(benzyl) complexes la, lb, 2,  S, and 4, are summarized in Tables 3-V and 3-VI. The parent ion for each complex is 25  observed in its low-resolution 70 eV mass spectrum. The IR spectral features exhibited by these bis(benzyl) complexes are worthy of discussion in light of their solid-state molecular structures. All of the bis(benzyl) complexes exhibit f o ' N  s m  their IR spectra that fall  within the range found for linear M-N-O linkages, i.e. 1640-1550 cm" . The most 1  26  107  Table 3-V. Analytical, IR and Mass Spectral Data for the Nitrosyl Bis-Benzyl Complexes Low-resolution Analytical Data (%) Mass Spectral Data C H N m/z calcd found calcd found calcd found  IR Data v*r ( ) c m  !  n  8  Complex  b  CpMoCNOJCO^CgHj)^  Nujol CHjClj  375, [P ]  61.13  61.10  5.13  5.10  3.75  3.84  1601  1601  459, [P ]  65.64  65.80  6.83  6.90  3.06  3.01  1572  1585  Cp'MotNOXCH^Hj)^  445, [P ]  65.01  64.91  659  6.68  3.16  3.06  1580  1578  CpW(NO)(CH C H )  d  461, [P ]  49.48  49.38  4.15  4.16  3.04  2.81  1570  1570  CpV^CCHjC^)/  531, [P ]  54.25  54.45  550  5.45  2.64  2.80  1566  1553  +  CpMo(NO)(CH C H -Me ) 2  2  fl  6  6  2  5  3  +  2  +  +  2  +  Probe temperatures 100-150°C.  species (e.g. ^Mo,  1 8 4  Assignments involve the most abundant naturally occurring isotopes in each  W etc.). The ^ c  N  m O  ^ ^ spectra of this complex as a Nujol mull, or KBr pellet, or in e  CJ^Clj solution is difficult to assign due to two nearly equally intense absorptions at 1601 and 1587 cm for the -1  nitrosyl ligand and the phenyl group. ^ In both the Nujol mull and CH^Ci^ solution IR spectra of this complex, the v  N Q  absorption contains a shoulder at 1595 cm , attributable to vibrations of the phenyl ring. -1  108 Table 3-VI. Variable Temperature *H and C{ H} NMR Data for the Nitrosyl Bis-Benzyl Complexes 13  1  H NMR data (CDjCy i  ^C^U} NMR data (CD C1 )  X  Complex  8  CpMo(NO)(CH C H ) 732(m,6H,CH C i/ ) 6.80 (m, 4H, CKf^) 537 (s, 5H, C ^ ) 2.40 (d, 2H, Cff H ^ ' J 1.22 (d, 2H, CHK, / 2  6  5  2  300 K  2  2  (  2  130.2,129,127.9 (aromatic PhQ 1013 (C H ) 35.91 (CHj)  5  5  5  = 6.6 Hz) = 6.6 Hz)  H a H b  2  H a H b  183 K 7.85-6.10 (m, 10H, Cllf^f ) 5.36 (s, 5H, C J/ ) 3.19 (d, IH, u W ^ C g l l , / 2.67 (d, IH, ^ - C H / / C H / 1.62 (d, IH, i j < » H C H , / -lj08(d,lH,i7'-CH 7f C H ,  2  5  2  a  b  6  5  H a H  2  a  x  6  a  x  s  6  = 4.2Hz) „ = 42 Hz) = 8.7 Hz) = 8.7 Hz) HaHx  H a H b  2  2  H a H x  s  6  2  3  2  2  2  6  6  2  3  2  3  5  3  6  3  5  2  6  s  / H a H b  2  HaHb  6  5  129.7 (aromatic PhC) 101.9 (C H ) 32.7 (CH/^Hj-) 21.4(C H -(CH3)3) 20.8 (C H -(CH ) )  208 K 6.29-6.15 (m, 4H, CHJZJIf) 458 (s, 5H, CJi ) = 5.7 Hz) 2.83 (d, IH, q W l ^ C j H j - , 2.71 (d, IH, nUM&Cjlf, / = 5.7Hz) 1.64 (d,lH, ti'-CHJIJZflf, 'HaHx = 83 Hz) 1.89,1.67,1.60,155,1.47, and JH),CH C H -(Ci/^ ) 1.27 (s, 6x(3H), 83 Hz) -2.45 (d, IH, 2  5  2  3  2  3  6  5  2  2  5  6  5  h  1  1.77(s,15H,C (C/n ) 5  (5  2  5  6  2  3  3  300 K  2  2  3  149.2 (quaternary r ^ - C H ^ H j ) 148.9,144.9,143.6 (quatern. PhC) 132.9,132.1,130.1,130.0, 127.8,1275 (aromatic PhC) 1043 (quaternary n -CH C H -) 1013 (C H ) 32.8 (r^-CHjC^-) 30.2(^-01/:^-) 21.6,21.0,20.8, 20.26,20.01, 19.47 (CH C H -(CH ) ) 2  Cp*Mo(NO) (CH2C H ) 7.16(m,6H,CH C ff ) 6.68 (d, 4H, C H j C X ) 231 (d,2H,CHH ,V HaHb  6  300 K  CpMo(NO) (CH C H -2,4,6-Me ) 6.25 (m, 4H, CUf^) 4.62 (s, 5H, C / y 2.298 ( br s, 4H, CHJZJfy 1.82 (s, 12H, CH C H -(Ci/ ) ) 156 (s, 6H, CH C H -(CfY ) ) 2  152.7 (quaternary ri -CH C HJ 139.6,132.9,129.6,127.5,126.2, 1235,1215 (aromatic PhQ 111.2 (quaternary r ^ - C H ^ H j ) 100.4 (C H ) XA^-CHjCgH;) 323 (^-CUf^) i  s  "  7  5  H  Z  )  5  131.0,128.0,126.7 (aromatic PhC) 108.97 ^ ( C H ^ ) 44.12 (CHj) 1037 (C (CH^ ) S  S  036 (d, 2H, C H ^ ? / ^ = 75 Hz) continued..  109 TABLE 3-VI. continued ^C^H} NMR data (CD C1 )  *H NMR data (CD CL,)  2  2  2  Complex  8  Cp Mo(NO)(CH2C H ) 7.75 (t, 2H, CHjCgHj) 7.6-6.6 (m, 6H, CKf^) 5.53 (d, 2H, C H j C ^ ) 6  5  183 K  2  3.31 (d, 1H, r j J - O / H ^ , V 1.81 (d, 1H, ^ - C H ^ C ^ , /  H a H b  2  H a H b  1.68 (s, 15H,  4030(772-0120^5)  1.16 (d, 1H, n -CH U C H , J = -1.75 (d, 1H, r ; - C H i / C H , J a  x  6  5  HaHx  2  6  6  5  93 Hz) 930 Hz)  2  2  2  = 6.6 Hz) = 6.6Hz)  0,(0/3)5) f  CpW(NO)(CH C H )  152.8 (quaternary ^-CRf^) 139.4,131.6,1393,127.6,126.9, 1213, (aromatic PhC) 115.0 (quaternary f^-CIL^Hj) lOS^rC^CH^) 44.19 (r^-CH^Hj)  5  m V x  =  9.46 (C (CHJ ) S  300 K  2  7.26 (m, 6H, C H j C ^ ) 6.75 (m, 4H, CHjC^j)  130.4,128.6,127.66 (aromatic PhC)  5.33 (s, 5H, C 5 / / 5 )  32.97 (CH , V ^ , = 59.0 Hz)  100.0 CC5H5) 2  2.15 (d, 2H, CHl^, / 1.15(d,2H,CH // , /  = 7.8 Hz) = 7.9Hz)  2  H a H b  2  a  b  HaHb  183 K 8.0 (t, 2H, CUf^) 7.6-6.6 (m, 6H, CUf^) 6.25 (d, 2H, CHjC^j) 5.46 (s, 5H, Cj/Zj) 3.13 (d, 1H, i ? - C H H C H , W= > 2.64 (d, 1H, r ^ - O y / ^ H j , ? H a H b = - ) 137 (d, 1H, i | - f f l H C ,6H5 >„ H a H x ) -0.723 (d, 1H, f7'-CH //C H , ' J ^ ' 10.0 Hz) 4  b  6  5  s  3 H z  5  4 H z  1523 (quaternary f^-CH^Hj) 1405,133.5,1303,128.1,127.4,126.7, 126.0,121.8, (aromatic PhC) 10856 (quaternary f ^ - O L ^ H j ) 9955 (T,H ) 35.05 (rj^-CHjC^) 5  2  y  H  a  6  6  5  2  2.24 ( d , 2 H , 0 / H , 2 > a  b  =  1  a  a  Cp= >7 -C H and Cp" = ^-CjMe^ 5  5  5  b  b  0H z  3  4  9  6  ft'CH^Hj)  5  133.6,132.1,128.4,127.1 (PhC) 108.2 (C (CHJ ) 42.91 (OL,) S  H a H b  = 9.11 Hz)  H a H b  5  10.29(05(013)5)  1.81 (s, 15H, 0 5 ( 0 / 3 ) 5 )  0.55 (d, 2H, C H / / / /  0  300 K  Cp W(NO)(CH C H ) ° 7.16 (m, 6H, CHjC^/j) 6.82 (d, 4H, CH^CgH.) 2  $  = 9.11 Hz)  Cooling the CDjClj solution of this compound to 183 K did not achieve a  limiting spectrum; indeed, coalescence was maintained down to this temperature.  110  chemically interesting feature of the IR spectrum of each Cp'M(NO)(CH2Ph')2 complex as a Nujol mull is that the V^Q is always observed at a higher wavenumber than the v^  Q  exhibited by the complexes' 16-electron dialkyl analogue(s) Cp'M(NO)R2, (R= CH SiMe , 2  3  CH CMe , and CH CMe2Ph). For instance, the observed V^Q at 1601 cm in the IR -1  2  3  2  spectrum of CpMo(NO)(CH Ph) , (la), as a Nujol mull, is higher in wavenumber than the 2  VNO  a t  2  1587 cm exhibited by CpMo(NO)(CH SiMe )2. Furthermore, the observed V^Q -1  2  3  at 1570 cm in the IR spectrum of CpW(NO)(CH Ph)2, (3), as a Nujol mull is higher in -1  2  wavenumber than the V^Q exhibited by each of the complexes CpW(NO)(CH SiMe )2, 2  3  CpW(NO)(CH CMe ) , and CpW(NO)(CH CMe2Ph) , (i.e. 1541,1560 and 1540 cm' , 1  2  3  2  2  2  respectively). This suggests that there is less electron density available on the metal 12  center for x-backbonding to the antibonding orbitals on the nitrosyl ligand in the 18-electron bis(benzyl) complex than in its 16-electron dialkyl analogue(s). It is not immediately clear why this is so. F.  Variable Temperature NMR Studies of the Bis (benzyl) Complexes la, lb, 2,3, and 4. The H and ^ C NMR spectral data for all of the bis(benzyl) complexes are 1  collected in Table 3-VI. Both variable temperature H and ^ C NMR spectroscopy of all 1  of the bis(benzyl) complexes in CD2CI2 solutions establish that the complexes are fluxional, exhibiting the dynamic process depicted below.  Ill  At room temperature, the rate of the fluxional process is fast on the NMR time scale, and a time average proton or carbon spectrum is observed that is consistent with the intermediate structure shown above. Therefore, at ambient temperatures the benzyl ligands appear equivalent, resulting in two magnetically equivalent sets of diastereotopic protons on the methylene-carbon atoms. At some lower temperature, the rate of the above fluxional process becomes slow on the NMR time scale. As a result, two clearly different methylenecarbon and phenyl group environments are observed in the H and 1  NMR spectra at  these lower temperatures. The room temperature and low temperature H and C NMR 1  U  spectra of the bis(benzyl) complexes will be considered in turn below. Representative room temperature H NMR spectra of all the bis(benzyl) 1  complexes, namely those of CpMo(NO)(CH QH -2,4,6-Me3) (lb) and 2  2  2  Cp*W(NO)(CH QH ) (4), are shown in Figures 3.6 (a) and 3.7 (a), respectively. Most of 2  5  2  the spectra display, in addition to proton resonances due to either the Cp or Cp* ligand and the aromatic phenyl protons, a single AY-pattern for the four methylene protons of the bis(benzyl) complex which was assigned using the appropriate proton-proton decoupling experiments. One doublet of thc^Y-pattern appears in the spectral region 6 2.4-2.0 ppm while the other doublet appears upfield in the range S 1.2-0.3 ppm, and the /naHx 2  coupling constant varies between 6.6 and 9.1 Hz, depending on the complex. This AXpattern is diagnostic of two magnetically equivalent sets of diastereotopic methylene protons, i.e. [Ph'CHaHx-M-CHa-Hx-Ph ], consistent with the two benzyl groups being 1  equivalent. This is further exemplified by the H NMR spectrum of 4 shown in Figure 3.7 1  (a). Interestingly, the 300 MHz H NMR spectrum of lb at ambient temperatures, Figure X  3.6 (a), does not exhibit the expected ASf-pattern for the methylene protons, likely due to coalescence. Only two proton resonances are observed for the six methyl groups on the two benzyl ligands. These resonances may be assigned to the two equivalent ^ara-methyl and four equivalent ort/io-methyl substituents on the benzyl ligands, respectively.  jo Bjpads H W N  Hdd »  '  '  {H > X  3  Ot  «  1 l _ J 1 1—1—'  '  •  •  Z H J V SZ.  n  09  1—1—J—l—1—1—1—1—L-I_1_J  • (  (q) pn* H  I  X  OB 1 1—1—1—1—1—1—1  1  1 1 1  '9*e wngij  Z H W OOC (*)  001 1  1 1 1  1  1 1 L_  . i  i  i  1 i  i  i  t  OKI 1 i  i  t  i  1 i  i  t  i  1  •  (  i  1  1  (q)  Idd E • ' ' • • ' 1  21  • ' • •  1  1  1 i I i > i • 1 i i i i  P •• • • • ' • f' 1  1  i  f  i i i H I I I  : I t  F  1  • I I I  F  I 1  in  («)  Zll  t  1 ,  i  •sajruBJsduiavJusiquiB jo Bjpads HWN (H } 3 x  nr.,,  JL  suonmps \3 aD z  a  (fr) ( H 3 H3)(ON)M, 3  z  <z  S  9  Z  d  ZHW SZ, (q) V™ H ZHW 00£ 00 'L'Z amSu T  001  09 ,  11 •  i  i  l  I  »  »  •  I  1  _J  I  I  I  1_  T  (q)  Bdd 0 I I I I I I I I I I I I I  nr  (»)  en  114  The room temperature  NMR spectrum of each bis(benzyl) complex exhibits  only 3 or 4 carbon resonances for all of the 12 carbon atoms of the two benzyl groups, Table 3-VI. Two representative examples, namely those of CpMo(NO)(CH QH -2,4,62  2  Me ) (lb) and Cp*W(NO)(CH C H ) (4), are shown in Figures 3.6 (b) and 3.7 (b) 3  2  2  6  5  2  respectively. The exception is complex lb, Figure 3.6 (b), where only 1 carbon resonance is observed in the spectral region 160-115 ppm. The low number of Ph-C resonances observed in the  NMR spectrum for each complex is probably due to rapid rotation of  the two magnetically equivalent phenyl rings causing a carboni-carbor^ axis of symmetry that is identical for both benzyl groups in solution at room temperature. Furthermore, each spectrum contains a single carbon resonance for the two methylene-carbon atoms in the spectral region 6 44-33 ppm. A gated decoupled C NMR spectrum of la and lb B  establishes that the one-bond coupling constant between the two equivalent methylenecarbon atoms and the two protons, VQH. * 142.2 and 140.0 Hz, respectively. s  27  The spectral features of the H and ^ C NMR spectra of each of the bis(benzyl) X  complexes are remarkably changed, however, upon cooling of the NMR tube containing the complex in CD C1 solution to approximately 183 K (Table 3-VI). Compare, for 2  2  instance, the 300 MHz H and 75 MHz C { H} NMR spectra of complex lb obtained at J  B  1  208 K shown in Figure 3.8 with those obtained at room temperature shown in Figure 3.6. Dramatic changes are observed as the aromatic proton and carbon regions become much more complex and all six methyl groups on the two benzyl groups become inequivalent as the complex is cooled to low temperatures. At low temperatures, the *H NMR spectra of each complex in CD C1 solution 2  2  exhibits, in addition to proton resonances due to either the Cp or Cp* ligand and the phenyl protons (Table 3-VI), four doublets which integrate for one proton each for the methylene protons of the two benzyl groups. Simple proton-proton decoupling experiments can be used to assign these doublets to a low field AB-pattern in the spectral region 3.31-2.29 ppm (with coupling of the two protons /naHb between 4.2 and 6.6 Hz) and a high field 2  115  (a)  I I | I I 1I | I I I I | I I 1I | I I TI | I I I I | I I T T f T  T 1 1 | TI PT"] T I I 1 | l l T I ] l l l l | l l l l j l 1 1  X  I I I | I IT T 1 1 I I I | I I 1 1 I I I 2 - 3 PPI  TT~ri  (b)  ' ' ' '.Jo' " " " ' 'J,'  1 I  I '  I I  I  I I  I I I I I 60  —i—j—i—i—i—i—]—i—i—i—i—J—i—i—i—r—|—i—r—i—i—f 40 M PPM  1  100  Figure 3.8. NMR spectra of CpMo(NO)(CH C H -2,4,6-Me ) , (lb), in CD C1 solution 2  6  2  3  2  2  2  at 208 K: (a) 300 MHz *H NMR spectrum, (b) 75 MHz C ^H} NMR spectrum, B  (c) 75 MHz gated decoupled C NMR spectrum, (d) 75 MHz (APT) C NMR B  D  spectrum. continued on next page  116  Figure 3.8. continued from previous page.  117  AY-pattern in the spectral region 1.64 - (-2.45) ppm (with coupling of the two protons 2  ./HaHx between 8.3 and 10 Hz). An illustration of these proton-proton decoupling  experiments using CpMo(NO)(CH2QH5) (la) as a representative example, is given in 2  Figure 3.9. Similar decoupling experiments can be used to find the fourth doublet in the low temperature *H NMR spectrum of lb, Figure 3.8 (a). Irradiation of the high field doublet at -2.45 ppm causes a singlet to appear at ^ 1.64 ppm among the envelope of methyl resonances in the spectral region 1.89-1.47 ppm. The observed AB and AX" patterns, therefore, establish that the two benzyl ligands of each of the bis(benzyl) complexes are very different in solution at low temperatures and that the two protons attached to each methylene-carbon atom are diastereotopic. Unfortunately, the structure of the bis(benzyl) complex in solution can not be assigned unambiguously from the H NMR data alone. For X  that, the C NMR data are also required. B  The low-temperature gated decoupled C NMR spectra for complexes la, lb, and B  2 are also consistent with the two benzyl groups being different at low temperatures. Not only are two carbon resonances observed for the methylene-carbon atoms of the benzyl groups, but two very different carbon-proton coupling constants are exhibited by the different methylene-carbon atoms. The low field methylene-carbon resonance in the C D  NMR spectra of complexes la, lb, and 2, exhibits / 1  C H  = 152.0,150.1, and 148.2 Hz,  respectively, while the high field methylene-carbon resonances exhibit V C H 131.6,130.7, =  and 125.5 Hz, respectively (Table 3-VI). The implication of these observed coupling constants concerning the bonding of the benzyl group to the metal center will be discussed later. However, the above features clearly establish that the two methylene-carbon environments are very different at low temperatures. Furthermore, the low temperature solution gated decoupled, as well as attached proton test (APT)  NMR experiments, can be used to determine the chemical shifts of  the two i/wocarbon atoms of the bis(benzyl) complexes. Using the spectra obtained for complex lb as an example [Figures 3.8 (c) and (d)] the z/wo-carbon atom of the f; -CH Ph' 1  2  118  Figure 3.9. 80 MHz *H NMR spectra of CpMo(NO)(CH C H ) , (la) in CD C1 solutions 2  6  5  2  2  2  at 183 K. Proton-proton decoupling experiments: (a) methylene-proton region of undecoupled spectrum, (b) irradiation at -1.08 ppm, (c) irradiation at 1.62 ppm, (d) irradiation at 2.67 ppm, and (e) irradiation at 3.19 ppm.  119  group is assigned to the observed carbon resonance at 149.2 ppm, and the other carbon resonances in the spectral region 143-148 ppm are assigned to the other phenyl carbon atoms in the CH2QH2-2,4,6-Me3 ligands. The chemical shift of the i/wo-carbon of the rj 1  Cr^Ph' group in each of the other bis(benzyl) complexes is always at =* 152 ppm and appears to be largely independent of the CpM(NO) fragment. The ipso-carbon atom of the r7 -CH Ph' group, on the other hand, is observed upfield in the spectral region 115-104 2  2  ppm, and its chemical shift shows a large dependence on the CpM(NO) fragment to which it is attached. The  CP/MAS solid-state NMR spectrum of each of the bis(benzyl) complexes  la, lb, 2, 3 and 4, were obtained at ambient temperatures by Drs. C. Fyfe and H. Gies of this Department. The similarity of the low-temperature of the bis(benzyl) complexes with their respective  { H} NMR spectrum of each 1  CP/MAS solid-state NMR spectrum  at ambient temperatures establishes unambiguously that the molecular structure of the low temperature form of each of the bis(benzyl) complexes in CD2CI2 solutions is identical to its solid-state molecular structure. This can be illustrated by the spectra obtained for complex la as a representative example. The low temperature 25 MHz spectrum in CD Ci2 solution and the 2  { H} NMR 1  CP/MAS solid-state NMR spectrum at ambient  temperature of complex la, are shown in Figures 3.10 (a) and (b), respectively. The two spectra are very similar. Most important is the correspondence of the chemical shifts of the two (pso-carbon resonances and the methylene-carbon resonances in the solution spectrum with the chemical shifts observed for these carbon atoms in the solid-state  NMR  spectra. The additional carbon resonances observed in the phenyl-carbon region of the solid-state spectrum of the bis(benzyl) complexes are most probably due to the lack of rotation of the benzyl ligand in the solid-state removing the carbonx-carbor^ plane of symmetry, which exists to some extent in solution.  (a)  A  LJLilJi 150  100  Li  50  (b)  Figure 3.10. (a) 25 MHz ^C^H} NMR spectrum of CpMo(NO)(CH C H ) , (la), 2  6  5  CD C1 solution at 193 K, and (b) 100 MHz C CP/MAS NMR spectrum of B  2  2  CpMo(NO)(CH C H ) , (la), at ambient temperature. 2  6  5  2  2  121  G.  Determination of the Free Energy of Activation for the Fluxional Process. A series of H NMR spectra at various temperatures was obtained for all the 1  bis(benzyl) complexes la, lb, 2,3 and 4, a portion of those for complex la being shown in Figure 3.11 as a representative example. Each complex exhibits coalescence of the methylene proton resonances at some well defined temperature (Table 3-VTJ). The free energy of activation, A G * , for the fluxional process of each bis(ben2yl) complex may be determined from the temperature of coalescence, T , and the difference in the chemical c  shift (in Hz) between the methylene proton signals undergoing coalescence, Av , at T , c  c  using the equation below (Kg being Boltzman's constant).  28  AG* =-RT ln[(xh/y2K )(Ai/ /T )] B  c  The values for Av (i.e. Ai/^2)c c  a n  c  c  d ^(2,4)c)»which correspond to the difference in  chemical shift of the two halves of the AB and AX methylene patterns observed in the low temperature spectrum, were estimated according to the method outlined by Fryzuk and MacNeil. The same coalescence temperature, T was used for both Ai/( ) and 29  0  Av  (2,4)c»  30 a Q  1>3  c  d therefore, for each complex, two values for AG* were determined. The  A G * value determined for each complex indicates that the fluxional process is very facile in solution at low temperature. Furthermore, an interesting trend is evident from the results presented in Table 3-VII. The temperature of coalescence varies dramatically (as much as 83°C) depending on the Cp'M(NO)- fragment to which the benzyl ligands are attached. The molybdenum bis(benzyl) complexes undergo coalescence at higher temperatures than their tungsten analogues. The results clearly illustrate that the fluxional process at low temperature is much more facile for the tungsten bis(benzyl) complexes than those of molybdenum. It can be assumed that at room temperature the tungsten bis(benzyl) complexes are more fluxional than their molybdenum analogues.  Figure 3.11. 80 MHz H NMR spectra of la in CD C1 solution at various temperatures. X  2  2  123 Table 3-VII. Determination of the Free Energy of Activation for the Fluxional Process. T  Compound  AG  c (Kcal/mol)  (K)  CpMo(NO)(CH C H -2,4,6-Me ) 2  6  2  3  CpMo(NO)(CH C H ) 2  6  5  2  Cp*Mo(NO)(CH C H ) 2  6  5  CpW(NO)(CH C H ) 2  6  5  6  2  2  Cp*W(NO)(CH C H ) 2  2  5  2  Vz4)  (Kcal/mol)  266(4)  12.0(2)  11.2(2)  232(3)  10.9(3)  10.5(3)  203(3)  9.3(3)  9.1(3)  210(3)  9.8(3)  9.5(3)  183(5)  8.4(5)  8.2(5)  This nonrigidity parallels the difference in reactivity of these bis(benzyl) complexes with molecular oxygen in solution. The tungsten bis(benzyl) complexes react with 0 to 2  afford low yields of Cp'W(0) (CH Ph) complexes, while the molybdenum analogues do 2  2  not react with 0 to afford Cp'Mo(0) (CH Ph) complexes, undergoing decomposition 2  2  2  after prolonged reaction times. If the mechanism proposed for these reactions in Chapter 2 is indeed operative, the reactions of the dialkyl nitrosyl complexes with 0 seems to be 2  governed by the ease with which the dioxygen molecule can approach the metal center. The fact that these bis(benzyl) complexes are sterically more congested about the molybdenum or tungsten center than their 16-electron dialkyl analogues, Cp'M(NO)R  2  (R = Me, CH SiMe , CH CMe , and CH CMe Ph) explains why they are inert to 0 as 2  3  2  3  2  2  2  solids and why they show limited reactivity in solution. The fact that the metal center is indeed protected somewhat from incoming bases is exemplified by the difference in the reactivity of the Cp'W(NO)(CH Ph) and 2  2  Cp'W(NO)(CH SiMe ) complexes with alcohol reagents. As seen in Chapter 2, reaction 2  3  2  of Cp*W(NO)(CH SiMe ) with MeOH to form Cp*W(NO)(OMe)(CH SiMe ) occurs 2  3  2  2  3  rapidly (i.e. 1-3 h), while the analogous reaction of Cp*W(NO)(CH Ph) with MeOH to 2  2  124  form Cp*W(NO)(OMe)(CH Ph) requires many weeks at room temperature. Furthermore, 2  reaction of CpW(NO)(CH SiMe ) with BuOH affords CpW(NO)(0- Bu)(CH SiMe ) l  2  3  t  2  2  3  over the course of several days, but no reaction of CpW(NO)(CH Ph) with BuOH is l  2  2  apparent even after six months. H.  A Formulation of the Bonding in the (rj -CH Ph*) Ligand of the Bis(benzyl) 2  2  Complexes la, lb, 2,3 and 4. 4  The bonding of the rj -benzyl group to the molybdenum or tungsten metal centers 2  in these nitrosyl bis(benzyl) complexes is currently formulated as involving a 3 electron linkage as a result of some interaction of the metals empty d-orbitals with the arene electron density at the //wo-carbon atom of the aromatic ring. Such a 3-electron linkage is invoked since the complexes are not paramagnetic species, based on their *H and ^ C NMR spectroscopic properties. It is difficult, however, to understand exactly how these i? 2  benzyl ligands can be 3-electron donors. In valence bond terms, one possible simplistic bonding representation is depicted below.  The X-ray crystallographic results have established that the M-CH bond length in 2  the M-(r? -benzyl) interaction is always significantly shorter (*0.05 A ) than the M-CH 2  bond length in the M-(»/^benzyl) interaction. Furthermore, the CHj-C^j  2  ^ bond length in  the r? -benzyl group is always significantly shorter (a; 0.03-0.04 A ) than the same bond 2  length in the normal TJ ^benzyl group. Low temperature gated decoupled C NMR U  spectroscopy has established that the methylene-carbon atom of the 77-benzyl group is 2  primarily sp - in character ( V Q J = 152-148 Hz compared to V 2  31  C  H  = 131.6-125 Hz in the  125  normal q -benzyl group). This is also supported by the X-ray crystallographic data which reveal that the angle M-C(ll)-C(12) (80.8 - 84.5°) is well below the expected tetrahedral value expected for an sp -hybridized methylene-carbon atom. This is more in accord with 3  the bonding of an olefin in an t? -fashion to a metal center. Consistent with this is the 2  C(ll)-M-C(12) in the range of 35.5-36.0°. In addition the C NMR spectroscopic data B  (CP/MAS and low temperature solution) support the existence of a strong interaction between the ipso-carbon atom and the metal center as this carbon resonance appears in the spectral range 8 115-104 ppm which is shifted considerably upfield from where resonances for aromatic quaternary carbon atoms are typically observed ( S 160 -145 ppm). Clearly the results of this work suggest that molecular orbital calculations are required, not only to understand the bonding of the unique benzyl ligand found in these nitrosyl bis(benzyl), complexes, but in a more general sense, to rationalize the various modes of bonding of the benzyl ligand to early and late transition-metals. Summary This work has demonstrated that the Cp'M(NO)(CH Ph')2 complexes (Cp' = Cp or 2  Cp*; M= Mo or W; Ph'= QH5 or C6H -2,4,6-Me3) are unique members in the class of 2  cyclopentadienylnitrosyl dialkyl complexes of molybdenum and tungsten. They are formally coordinatively saturated, 18-electron complexes possessing an unusual Cp'M(NO)(r; -CH Ph')(T7 -CH Ph') structure both in the solid-state and in low-temperature 2  1  2  2  solutions. Variable temperature H and ^ C NMR spectroscopy of the complexes in X  CD C1 solutions establishes that the complexes under go a facile fluxional process, the two 2  2  benzyl ligands interchanging their modes of attachment to the molybdenum or tungsten metal center. This fluxional process is much more facile for the tungsten bis(benzyl) complexes than the molybdenum analogues and suggests that during reactions of any Cp'M(NO)R complex with molecular oxygen, initial attack by C^ at the metal center is 2  required to eventually form the Cp'M(0) R complexes. 2  126 References a n d Notes:  (1)  Taken in part from; (a) Legzdins, P.; Phillips, E.C.; Fyfe, C ; Gies, H.; Einstein, F.W.B.; Jones, R.H. manuscript submitted for publication in Angew. Chem. (b) Einstein, F.W.B.; Jones, R.H.; Legzdins, P.; Phillips, E.C. manuscript submitted for publication in Organometallics. (c) Legzdins, P.; Phillips, E . C ; Trotter, J.; Yee, V . C manuscript in preparation.  (2)  Legzdins, P.; Phillips, E . C ; Sanchez, L. Organometallics 1989,8, 940.  (3)  Drezdzon, M.A.; Shriver, D.F. "The Manipulations of Air-Sensitive Compounds"; 2  (4)  nd  Ed.; John Wiley and Sons : New York, 1986.  "Experimental Organometallic Chemistry, A Practicum in Synthesis and Characterization", Darensbourg, M.Y.; Wayda, A.L. Editors Am. Chem. Soc. 1987.  (5)  (a) Kharasch, M.S. "Grignard Reactions of Non-Metallic Substances"; Prentice -Hall: New York, 1954. (b) Sommer, L.H.; Whitmore, F.C /. Am. Chem. Soc. 1946, 68, 481.  (6)  Spectroscopic data for all of the Cp'M(NO)X2 complexes (X= CI or I), used in this work suggest that they are all monomeric in solution and therefore are represented as such throughout this chapter, (see also reference 9).  (7)  Bray, J.; Kita, W.G.; M^leverty, J.A.; Seddon, D. Inorg. Synth. 1976,16, 24.  (8)  Legzdins, P.; Martin, D.T.; Nurse, CR. Inorg. Chem. 1980,19,1560.  (9)  Dryden, N.H.; Einstein, F.W.B.; Jones, R.H.; Legzdins, P. Can. J. Chem. 1988, 66, 2100.  (10)  Nurse, CR. Ph.D. Dissertation, The University of British Columbia, 1983.  (11)  Dryden, N.H.; Legzdins, P. manuscript in preparation.  (12)  Legzdins, P.; Rettig, SJ,; Sanchez, L. Organometallics 1988, 7,2394.  (13)  Me -C H CH C1 (5.0 g, 29.6 mmol) and Mg turnings (4.5 g, 185 mmol) were 3  6  2  2  reacted in an Et^O (250-300 mL) solution and after the reaction was complete the  127  mixture was filtered through Celite supported on a medium-porosity glass frit. The resulting solution of Grignard reagent was standardized with a 0.1 M HC1 solution. See Chapter 2 of this thesis. (a) Borden, W.F. "Modern Molecular Orbital Theory for Organic Chemists"; Prentice-Hall: Englewood Cliffs, NJ, 1975 ; p 99. (b) McWeeny, R. "Coulson's Valence", 3 ed.; Oxford University Press : London, 1979 ; p 248. rd  The same results are obtained for a benzyl anion: see reference 15. (a) Chappell, S.D.; Cole-Hamilton, DJ. J. Chem. Soc, Dalton Trans. 1983,1051. (b) Jacob, v.K.; Thiele, K.-H.; Keilberg, C ; Niebuhr, R. Z. Anorg. Allg. Chem. 1975,415,109. (a) Cotton, F.A.; LaPrade, M.D. /. Am. Chem. Soc. 1968, 90,5418. (b) Behrens, U.; Weiss, E. /. Organomet. Chem. 1975, 96,399. (c) Behrens, U.; Weiss, E. /. Organomet. Chem. 1975, 96,435. (d) Bailey, P.M.; Maitlis, P.M.; Sonoda, A. /. Chem. Soc., Dalton Trans. 1979, 347. (e) Bleeke, J.R.; Burch, R.R., Coulman, C.L.; Schardt, B.C. Inorg. Chem. 1981,20,1316. (f) Burch, R.R.; Muetterties, E.L.; Day, V.W. Organometallics 1982,1,188. (g) Chappell, S.D.; Cole-Hamilton, DJ.; Galas, A.M.R.; Hursthouse, M.B.; Walker, N.P.C. Polyhedron 1985,4,121. (h) Latesky, S.L.; McMullen, A.K.; Niccolai, G.P.; Rothwell, LP. Organometallics 1985, 4, 902. Carmona, E.; Marin, J.M.; Paneque, M.; Poveda, M.L. Organometallics 1987,6, 757. Greenhough, T. J.; Legzdins, P.; Martin, D. T.; Trotter, J. Inorg. Chem 1979,11, 3268. (a) Bassi, I.W.; Allegra, G.; Scordamaglia, R.; Chioccola, G. /. Am. Chem. Soc. 1971, 93,3787. (b) Davies, G.R.; Jarvis, J.AJ.; Kilbourn, B.T.; Pioli, AJ.P. /. Chem. Soc, Chem. Commun. 1971, 677. (c) Davies, G.R.; Jarvis, J.AJ.; Kilbourn, B.T. /. Chem. Soc, Chem. Commun. 1971,1511.  128  (d) Mintz, E.A.; Moloy, K.G.; Marks, TJ.; Day, V.W. /. Am. Chem. Soc. 1982, 104,4692. (e) Edwards, P.G.; Anderson, R.A.; Zalkin, A. Organometallics 1984, 3,293. (22)  Jordan, R.F.; Lapointe, R.E.; Bajgur, C.S.; Echols, S.F.; Willet, R. /. Am. Chem. Soc. 1987,109, 4111.  (23)  Hamon, J.-R.; Astruc, D.; Rom n, E.; Batail, P.; Mayerle, JJ. /. Am. Chem. Soc. 1981,103,2431.  (24)  Previously, Edwards, Anderson and Zalkin used this method to quantify the bond length differences found in late transition-metal ??-benzyl complexes and early 3  transition-metal r? -benzyl complexes, (see reference 21e). 4  (25)  Some of the data for this compound has be reported previously, see reference 12.  (26)  Shoulders on the I/^JQ bands around 1595 cm" due to vibrations of the phenyl rings, 1  are sometimes intense and can make assignments difficult. (27)  As we will see later, these values of  (142.2 and 140.0 Hz) are exactly half of the  sum of the two values of V C H for the different methylene-carbon resonances that are observed in the low temperature gated decoupled C NMR spectra of la and B  lb (i.e. for la:  = 152.0 and 131.6, for lb: V C H = 150.1 and 130.7).  (28)  Thomas, W. A. Annu. Rev. NMR Spectrosc. 1968,1,43.  (29)  Fryzuk, M. D.; MacNeil, P. /. Am. Chem. Soc 1984,106, 6993.  (30)  Note, these two sets of methylene protons do not coalescence at precisely the same temperature, and therefore, the calculated values for A G * will not be identical.  (31)  (a) Levy, G. G; Lichter, R. L.; Nelson, G. L. Carbon-13 Nuclear Magnetic Resonance Spectroscopy, 2nd ed.; John Wiley and Sons: New York, 1980. (b)  Mann, B. E.; Taylor, B. F . C NMR Data for Organometallic Compounds; Academic 13  Press: London, England, 1981. (c) Becker, E. B. High Resolution NMR: Theory and Chemical Applications, 2nd ed.; Academic Press: New York, 1980.  Chapter 4 New Types of Cyclopentadienyl Oxo Trialkyl Complexes, CpW(0)Rand Oxo Alkylidene Complexes, CpWfOXsROR, of Tungsten  1  130 Introduction This chapter describes some of my investigations into the characteristic reactivity towards electrophiles of the dioxo alkyl complexes, Cp'W(0) R, whose synthesis and 2  characterization is discussed in Chapter 2. The remarkable aspect of the reactivity exhibited to date by these complexes is that their metal-alkyl linkages remain unaffected by reagents that would normally be expected to cleave them. For instance, reactions of the Cp'W(0) R complexes with protonic acids such as hydrogen halides would normally be 2  expected to proceed as outlined in eq. I, the anticipated Cp'W(0) X products (M= Mo or 2  2  W, X= Br or CI) being well known.  3  Cp'W(0) R + HX  »- Cp'W(0) X + RH  2  (1)  2  Instead, it is the W=0 groups of the Cp'W(0) R complexes that are the preferred sites of 2  reactivity. This is particularly well illustrated by the behavior of some of the tungsten dioxo alkyl complexes upon treatment with hydrogen chloride to afford interesting new oxo dichloride complexes, namely Cp'W(0)(Cl) R. These complexes are also preparable from 4  2  reactions of the Cp'W(0) R compounds with various other chlorinating agents. The full 2  details of this reactivity constitute the first part of this chapter. The important consequence of the preparation of these new oxo dichloro species is that a novel synthetic route to cyclopentadienyl oxo trialkyl, Cp'W(0)R , and oxo 3  alkylidene, Cp'W(0)(=R')R, complexes of tungsten has been found. Therefore, the 5  second part of this chapter describes my efforts to prepare a number of such Cp'W(0)R  3  and Cp'W(0)(=R')R complexes from alkylation reactions of the Cp'W(0)(Cl) R 2  compounds. All the new high oxidation state organometallic oxo complexes prepared in this work have been completely characterized by conventional analytical and spectroscopic techniques.  131 Experimental Section All reactions and subsequent manipulations were performed under anaerobic and anhydrous conditions. The general procedures employed in this study were the same as those described in the Experimental Sections of the preceding two chapters. The HC1 gas was purchased from Matheson Gas Company in 99.0% purity and was not purified further before use. The two alkoxides, NaOMe and KC^Bu, were purchased from Aldrich. The alkyl-Grignard reagent, PhCH MgCl was prepared by the literature procedure. The 6  2  Me SiCH MgCl (1.0M solution in E^O), CH Li (1.4M solution in Et^O), PC1 and 3  2  3  5  Me SiCl reagents were all purchased from Aldrich. The cyclopentadienyl dioxo alkyl 3  complexes, Cp'W(0) R (Cp'= Cp or Cp*, M= Mo or W; R= C H or CH SiMe ), used as 2  3  2  3  reagents in this work were prepared by the methods outlined in Chapter 2 of this thesis. Preparations of Cp*W(0)(Cl) (CH SiMe ) by Employing HC1, PC1 or Me SiCl as 2  2  3  5  3  the Chlorinating Agents. (a) With HC1. To a stirred, colorless solution of Cp*W(0) (CH SiMe ) (0.85 g, 1.9 2  2  3  mmol) in E^O (20 mL) was added by microsyringe a 6 M solution of HC1 in E^O (1.0 mL, 6.0 mmol HQ).  7  The reaction mixture immediately became yellow; stirring of the mixture  was continued for 12 h whereupon a yellow precipitate also formed. Removal of the solvent in vacuo and recrystallization of the resulting yellow solid from a 3:1 E^O/hexanes solvent mixture at -20 °C afforded 0.87 g (91% yield) of Cp*W(0)(Cl) (CH SiMe ) as a 2  2  3  yellow crystalline solid which was isolated by filtration. (b) With PC1 . To a stirred, white suspension of PC1 (0.20 g, 0.96 mmol) in 5  5  CH C1 (15 mL) at 0°C was added solid Cp*W(0) (CH SiMe ) (0.42 g, 0.96 mmol). 2  2  2  2  3  Immediately, the reaction mixture became dark yellow, but the color lightened to yellowgreen as stirring was continued for 0.5 h. At the end of this time, the mixture was allowed to warm to room temperature, and the solvent was then removed under reduced pressure  132 to obtain a yellow-green solid. Crystallization of this solid from E^O at -20 °C produced yellow-green microcrystals of Cp*W(0)(Cl) (CH SiMe ) (0.32 g, 68% yield). 2  2  3  (c) With M^SiCl. To a stirred, colorless solution of Cp*W(0) (CH SiMe ) (0.61 2  2  3  g, 1.4 mmol) in E t ^ (10 mL) at room temperature was added Me SiCl (1.0 mL, 7.9 mmol) 3  by syringe. The reaction mixture became yellow immediately. After being stirred for 48 h, the reaction mixture consisted of a yellow microcrystalline precipitate and a yellow supernatant solution. Volatiles were then removed in vacuo, and the yellow solid remaining was recrystallized from 1:1 E^O/hexanes at -20 °C to obtain 0.67 g (97% yield) of yellow Cp*W(0)(Cl) (CH SiMe ). 2  2  3  The analytical, mass spectral, IR, and H and ^ C NMR data for X  Cp W(0)(Cl) (CH SiMe ) and all the other new oxo complexes prepared during this work 2  2  3  are presented in Tables 4-1 and 4-II. Preparation of Cp*W(0)(Cl) (Me). This complex was obtained by treating 2  Cp*W(0) Me with an excess of Me SiCl in E^O in a manner analogous to that outlined 2  3  above in method (c) for the preparation of Cp*W(0)(Cl) (CH SiMe ). Yellow 2  2  3  Cp W(0)(Cl) Me was isolated in 60% yield by recrystallization of the reaction residue 2  from E^O at -20 °C. Preparation of CpW(0)(Cl) (CH SiMe ). This complex was prepared by utilizing 2  2  3  HC1 as the chlorinating agent in a manner analogous to that outlined above in method (a) for its permethylcyclopentadienyl analogue. This reaction was performed over 4 h at 0 °C, however, and the final product compound precipitated as a yellow crystalline solid in 54% isolated yield when the final reaction mixture was maintained at -20 °C for 2 d.  133 Attempts at the Preparation of the Molybdenum Analogues: Cp*Mo(0)(Cl) R, (R= 2  Me or C^SiMej). (a) R= Me.  To a stirred solution of Cp*Mo(0) Me (0.11 g, 0.4 mmol) in Et^O 2  (10 mL) at 0 °C was added Me SiCl (0.25 mL, 2.0 mmol) using a 1 mL glass syringe. The 3  initially clear, colorless reaction mixture turned pale yellow immediately and continued to darken slowly within the first 5 min of the reaction. After 6 h, the solution had developed a slight blue coloration. The Et20 solvent was removed in vacuo to obtain a greenish-blue residue. An IR spectrum of this residue as a Nujol mull exhibited absorption bands at 922 and 844 cm" which may be respectively assigned to a v 1  Mo=Q  ,  and 3 « , of Cp  Cp*Mo(0)(Cl) Me. The residue did not, however, completely redissolve in Et^O. 2  (b) R= CHjSiMej.  This reaction was performed utilizing HC1 as the chlorinating  agent in a manner analogous to that outlined for its tungsten analogue above. To a stirred solution of Cp*Mo(0) CH SiMe in E^O (15 mL) at -35°C was added HC1 in E^O. The 2  2  3  reaction mixture immediately became yellow, and a yellow precipitate slowly formed. After approximately 1 h the solution quickly developed a blue coloration and the yellow precipitate was consumed. Reactions of Cp*W(0)(Cl) (CH SiMe3) with NaOMe and KO'Bu. 2  (a) With NaOMe.  2  To a stirred suspension of NaOMe (0.055g, 1 mmol) in EtjO  (10 mL) at room temperature was added by cannulation a yellow solution of freshly prepared Cp*W(0)(Cl) (CH SiMe ) (0.25g, 0.5 mmol) in E^O (10 mL). The resulting 2  2  3  reaction mixture was stirred for 12 h after which time the solution was greyish green in color. The solvent was removed in vacuo, and the resulting solid residue was dried at 20 °C for 2 h. Addition of E^O (2 x 10 mL) and filtration of the resulting mixture through Celite (6x4 cm) supported on a medium porosity glass frit gave a yellow filtrate, with a bluish residue remaining on top of the Celite. A yellow solid was obtained from the filtrate by  134 solvent removal, and both the IR and H NMR spectra of this solid confirmed that it was 1  Cp*W(0) (CH SiMe ) (crude weight -0.17 g, 80% yield). 2  2  3  (b) With KOteu.  To a stirred suspension of KO Bu (0.23g, 2.03 mmol) in Etp l  (15 mL) at 0°C was added by cannulation a yellow solution of freshly prepared Cp*W(0)(Cl) (CH SiMe ) (0.50g, 1.01 mmol) in EtjO (30 mL). Immediately the reaction 2  2  3  mixture changed from yellow to red-brown, and a yellow-white precipitate formed. The reaction mixture was stirred for 30 min at 0°C, after which time its volume was reduced in vacuo to ~ 10 mL and then hexanes (20 mL) were added. The resulting mixture was filtered through Celite (6x4 cm) supported on a glass frit. All volatiles were removed from the filtrate in vacuo to give a red oil. An IR spectrum of the neat oil revealed bands at 1242 (s) (^ . ), 1171 (s), 943 (vs, b), 905 (s), 852,833 (s) (y . ), and 559 (m) cm" . 1  si  Me  Cp  H  Recrystallization of the red oil from pentane at -20 °C gave dark red-green crystals that were highly air- and moisture-sensitive. Indeed, any attempts to characterize this product further led to decomposition, and all further spectra obtained suggested that Cp W(0) (CH SiMe ) was the decomposition product. This was illustrated by the fact 2  2  3  that upon exposure of the above red oil to air it turned yellow, and an IR spectrum of this yellow oil was diagnostic of the presence of Cp*W(0) (CH SiMe ), i.e. 1242 (s) (v i_ ), 2  939 and 901 (s) (^ ,)>  a  n  d  w=c  2  3  S  Me  850, 833 (s) ^ . ) cm"- The green-red crystals dissolved in 1  Cp  H  C D to give a yellow solution whose *H NMR spectrum was consistent with the presence 6  6  of Cp*W(0) (CH SiMe ). Furthermore, the product complex was so air-sensitive that an 2  2  3  elemental analysis of the greenish crystals gave carbon and hydrogen content consistent with Cp*W(0) (CH SiMe ). 2  2  3  Preparation of CpW(0)(CH SiMe3) . A stirred yellow solution of freshly prepared 2  3  CpW(0)(Cl) (CH SiMe ) (0.16 g, 0.38 mmol) in Et^O (25 mL) at approximately -40 °C 2  2  3  (maintained by a saturated aqueous CaCl /Dry Ice bath) was treated dropwise with a 1.0 2  M Et^O solution of Me SiCH MgCl (0.78 mL, 0.78 mmol). The reaction mixture 3  2  135 immediately developed a brown color, but it was stirred at this temperature for 30 min before being permitted to warm to room temperature to ensure completion of the reaction. The final mixture was filtered through a short (4x2 cm) column of Florisil supported on a medium porosity frit. The filtrate was taken to dryness in vacuo, and the remaining yellow solid was crystallized from pentane at -20 °C to obtain 0.08 g (40% yield) of yellow needlelike crystals of CpW(0)(CH SiMe ) . The H NMR spectrum of the product complex in 1  2  3  3  C D at 20 °C (Table 4-II) also exhibits small signals due to the corresponding oxo 6  6  alkylidene compound, CpW(0)( = CHSiMe )(CH SiMe ) (Table 4-JJ), and Me Si which 3  2  3  4  increase in intensity with time. Preparation of Cp*W(0)(Me) (CH SiMe ). To a stirred, yellow solution of 2  2  3  Cp*W(0)(Cl) (CH SiMe ) (0.42 g, 0.85 mmol) in E^O (30 mL) at approximately -40 °C 2  2  3  was added dropwise a 1.4 M solution of MeLi in E^O (1.4 mL, 1.96 mmol). Immediately the solution became green in color, and a white precipitate formed. After being stirred for 1 h, the reaction mixture was permitted to warm slowly to room temperature by removing the surrounding cold bath. The solvent was then removed from the final reaction mixture to obtain an oily blue residue. This residue was extracted with hexanes (2 x 10 mL), and the blue-green extracts were chromatographed on a Florisil column (2x4 cm) using EtjO as the eluant. The single yellow band that developed was eluted from the column and collected. Solvent was removed from the eluate under reduced pressure, and the resulting oily yellow solid was crystallized from pentane at -20 °C to obtain 0.26 g (62% yield) of Cp*W(0)(Me) (CH SiMe ) as analytically pure, large yellow crystals. 2  2  3  Preparation of Cp*W(0)(Me)(CH Ph) . To a stirred, yellow solution of 2  2  Cp*W(0)(Cl) Me (0.22 g, 0.52 mmol) in Etp (30 mL) at ca. -40 °C was added dropwise a 2  1.1 M solution of PhCH MgCl in Et^O (0.99 mL, 1.09 mmol). The reaction mixture 2  immediately became cloudy and developed a green-brown coloration. After being stirred for 1 h at this temperature, the mixture was permitted to warm to room temperature and  136 was then filtered through a short (2x4 cm) column of Florisil supported on a medium porosity frit. The yellow filtrate was taken to dryness in vacuo, and the remaining residue was recrystallized from 1:1 E^O/hexanes at -20 °C overnight to obtain 0.12 g (44% yield) of Cp*W(0)(Me)(CH Ph) as a yellow crystalline solid. 2  2  Preparation of CpW(0)(=CHSiMejXCH^SiM^). A stirred, yellow solution of CpW(0)(CH SiMe ) (0.30 g, 0.57 mmol) in benzene (15 mL) was maintained at 50 °C for 2  3  3  two days. Removal of all volatiles from the final reaction mixture in vacuo produced a yellow solid which was recrystallized from EtjO to obtain 0.24 g (95% yield) of CpW(0)( = CHSiMe )(CH SiMe ) as analytically pure, yellow needles which were 3  2  3  collected by filtration. This conversion was also effected in benzene-^ while being monitored by  NMR  spectroscopy. This monitoring confirmed the clean conversion of the oxo trialkyl compound into the oxo alkylidene complex, the only by-product being Me Si. 4  Preparations of Cp*W(0)( = R')(CH SiMe3) (R' = CHSiMe3 or CHPh). Both of 2  these complexes were synthesized and isolated in a similar manner. The experimental procedure, using the case when R' = CHSiMe as a typical example, was as follows. 3  A yellow solution of Cp*W(0)(Cl) (CH SiMe ) (0.45 g, 0.91 mmol) in Etp (50 2  2  3  mL) was cooled to ca. -40 °C by employing a saturated aqueous CaCl /Dry Ice bath. To 2  this stirred solution was added dropwise a colorless solution containing 2.5 mL of a 0.82 M solution (2.0 mmol) of Me SiCH MgCl in E^O diluted with an additional 15 mL of Et^O. 3  2  The reaction mixture instantly became green in color, but then slowly changed to brown as the addition of the Grignard reagent continued. After this addition was complete, the final reaction mixture was stirred for an additional 30 min at ca. -40 °C before being permitted to warm slowly to room temperature by removal of the cold bath. Solvent was then removed in vacuo, and the remaining brown solid was extracted with hexanes (2 x 10 mL). The brown hexanes extracts were concentrated to ca. 5 mL under reduced pressure and  137 were chromatographed on a short (3x2 cm) column of Florisil with E^O as eluant. The single brown band that developed was eluted from the column and collected. Solvent was removed from the eluate in vacuo and the remaining red-brown solid was recrystallized from pentane at -20 °C to obtain 0.27 g (58% yield) of highly air-sensitive Cp*W(0)(=CHSiMe )(CH SiMe ) as large, black needles. 3  2  3  Yellow-green needles of Cp*W(0)(=CHPh)(CH SiMe ) were obtained in an 2  3  analogous manner in 69% isolated yield when PhCH MgCl was used in place of 2  Me SiCH MgCl in the above procedure. 3  2  Preparations of Cp*W(0)(=R') (Me) (R' = CHSiMej or CHPh). Both of these complexes were generated by gently warming (50-60 °C) C D solutions of their oxo trialkyl 6  6  precursors (i.e. Cp*W(0)(Me) (CH SiMe ) and Cp*W(0)(Me)(CH Ph) , respectively) in 2  2  3  2  2  NMR tubes. The oxo alkylidene product complexes were characterized spectroscopically (Table 4-II), but were not isolated.  138 Results and Discussion A.  Synthesis, Spectroscopic and Some Physical Properties of the Cp'W(0)(Cl) R 2  Complexes (Cp'= Cp or Cp*, R= C H or CI^SiMej). 3  Treatment of representative members of the 16-electron dioxo alkyl complexes, namely Cp'W(0) R (Cp'= Cp or Cp*, R= Me or CH SiMe ), with a slight excess of HC1 2  2  3  in E^O results in their clean conversion to the corresponding Cp'W(0)(Cl) R compounds 2  as summarized in eq. 2.  Cp'W(0) R + xsHCl 2  • Cp'W(0)(Cl) R + H 0 2  2  (2)  There is absolutely no evidence during the occurrence of conversions 2 for the formation of any of the Cp'W(0) Cl products even though these are well-known species that might 4  2  reasonably be expected to result from the reaction of a transition-metal alkyl with a hydrogen halide. The net organometallic conversions shown in eq. 2 can also be effected 8  with PC1 or Me SiCl as the chlorinating agents. The possible mechanistic pathways by 5  3  which these reactions could occur will be discussed later. The congeneric molybdenum dioxo alkyl complexes initially appear to react with HC1 (or Me SiCl) in a manner similar 3  to that shown in eq. 2, but they then convert to other products whose identities have yet to be determined (see Experimental Section). The tungsten oxo dichloro products are isolable in high yields (>90%) from reactions 2 as yellow, crystalline solids which may be handled in air for short periods of time without the occurrence of noticeable decomposition. These solids are less soluble in common organic solvents than are their dioxo alkyl precursors. Their mass, IR, and *H and ^ C NMR spectra (Tables 4-1 and 4-II) are consistent with their possessing monomeric  139  Table 4-1.  Analytical and IR Data for the New Oxo Complexes Isolated in this Work. Analytical Data (%)  IR Data (Nujol, cm") 1  C Complex  H  calcd  found  CpW(0)(Cl) (CH2SiMe3)  75 dec  2555  25.80  3.81  3.92  951  1244  Cp*W(0)(Cl) (CH SiMe )  83 dec  34.09  34.14  531  538  939  1244  Cp*W(0)(Cl) Me  102  3138  31.67  431  438  939  84 dec  38.78  38.48  7.22  7.64  941  1259,1242  Cp'wXOXMe^CI^SiMej)  106 dec  42.48  42.85  7.13  7.45  933  1240  Cp*W(0)(Me)(CH Ph)  99 dec  56.41  56.66  6.06  6.17  935  CpW(0)(=CHSiMe^XCHjSiMej)  74 dec  35.61  35.80  5.93  6.06  956  1255,1240  Cp*W(0)(=CHSiMegXCHjSiMe^)  127 dec  42.51  42.88  7.13  733  949  1250,1238  Cp*W(0)(=CHPh)(CH SiMe )  107 dec  49.22  49.40  6.29  6.40  939  1238  2  2  2  3  2  CpW(0)(CH SiMe ) 2  3  2  3  2  2  3  found  M=0 "Si-Me  m .°C D  calcd  U  asvm  Table 4-II. Mass Spectral and *H and C{ H} NMR Data for the Oxo Complexes U  low-resolution mass spectral data m/z  1  J  H NMR data  2  2  371, [P-H,CLMe] 352, [P-2C1]  3  1  s  b  CpW(0)(Cl) (CH SiMe )  C{ H}NMR data  <W  8  Complex  13  113.1 (C H ) 60.4 (CH SiMe3) 1.95 (Si(CH^  5.47 (s, 5H, C^f ) 2.80 (d, 1H, CW H , 2/^ = 10.7 Hz) 1.26 (d, 1H, CH tt ,  +  5  s  +  2 /  HH =  1 0  -  A  B  A  B  7 H z  5  2  )  0.42 (s,9H, SiQCHJJ Cp*W(0)(Cl) (CH SiMe ) 2  2  3  456, [P-H,C1] 426, [P-H,CL2Me] +  2.23(d,lH,a/ H ,2/ = 9.8 Hz, = 6.3 Hz) 1.75 (s, 15H, C (CHJ,) 0.80 (d, l H . C H ^ , / ^ = 9.8 Hz, = 63 Hz) 0.49 (s, 9H, $(0/3)3)  122.1 (C (CHJ ) 633 (CHj, ^ = 72 Hz) 12.0(C (CH3) )  1.78 (s, 3H, 0 / , 2 / ^ = 5.5 Hz) 1.70 (s, 15H, C (CHJ )  121.5 (C (CH3) ) 52.2 (CH , ^ = 90.5 Hz) 115 (C (CHJ )  A  +  X  HH  $  2  385, [P-H,C1] 371, rP-H,Cl,Me]  Cp*W(0)(Cl) Me  +  2  3  +  S  S  5  S  5  5  2.2(Si(CH3)3) 5  5  3  $  CpW(0)(CH SiM ) 2  e3  511, [P-Me] 439, [P-CH SiMe ] +  3  2  3  +  539 (s, 5H, C ^ ) 1.41 (d,2H,0/ H , 2/^=13.9 Hz) 1.22 (d, 2H, C H ^ , / = 13.9Hz) 0.64 (s, 2H, OZ Si, 2 / ^ = 7.9 Hz) 036 (s, 9H, Si(CH^ 0.28 (s, 18H, 2 S K O ^ j ) A  B  A  B  2  H H  2  Cp*W(0)(Me) (CH SiMe ) 2  2  3  437, [P-Me] 407, [P-3Me] +  +  Isomer A: trans-Me Groups 1.47(s,15H,C (0/3) ) 135 (s, 2H, OZ Si) 133 (s, 611,20/3) 050(s,9H, Si(CH^ 5  2  5  S  1073 (C H ) 385 (O^Si, ^ = 75 Hz) 323 (2 O^Si, Hz) 5  5  45 ($(013)3)  2.2 (2 S^OlJj)  112.7 (C (CH3) ) 39.4 (O^Si, =61.7 Hz) 31.2 (013, / = 66.7Hz) 10.48 (C (CH3) ) 4.63 (Si(CH^ 5  5  5  5  1  c w  continued  141 Table 4-II continued  low-resolution mass spectral data m/z  *H NMR data  8  Complex 2  2  1  &  Isomer B: cis-Me Groups L49(s,15H,C (CH ) ) 130 (s, 3H, CHJ 0.74 (d, IH, CH /f Si, / =4.2 Hz) 0.47 (s, 9H, S\(CH^ -0.09 (s, 3H, C / y  c  3  C{ H}NMRdata  <W  b  Cp*W(0)(Me) (CH SiMe )  13  5  3  A  s  x  2  H H  112.3 (C (CHJ ) 39.44 (O^Si, / = 61.7Hz) 3633 (CH , = 60.5 Hz) 33.76 (CH , J = 57.6 Hz) 10.60 (C (CHJ ) 2M(Si(Ctl£) S  S  1  c w  3  3  1  CW  S  Cp*W(0)(Me)(CH2Ph)  440, [P-Me,Ph]  +  2  5  758(d,4H,2o-Phtf, 148.1 (quaternary PhC) 2 / ^ = 73 Hz) 130.9,127.2,124.5 7.13(t,4H,2ro-Phff, (aromatic C) 2/ = 7.6Hz) 112.8 (C (CHJ ) 6.87(t,2H,2/?-Phff, 57.6 (CHj, ^ = 64 Hz) 2/ = 73Hz) 337 (d, 2H, 2Cff H Ph, 435 (CH3) 95Hz, / = 12.1 Hz) 10.7(C (CH3) ) 238 (d, 2H, 2CH i/ Ph, / = 9 5 H z , / = 4.5Hz: HH  S  5  HH  A  B  2  H W  5  A  2  5  B  2  HW  H H  H W  1.44(s,15H,C (af3) ) 0.29 (s, 3H, CHJ 5  5  1034 (d, IH, =CH, / = 10.0 Hz) 539 (s, 5H, Cji/j) 0.85 (d, 1H, Ci/ H Si, ^ = 12.9 Hz) 0.68 (d, IH, CH tf Si, / = 12.9 Hz) 0.49 (s, 9H, CHSitCff^  CpW(0)(=CHSiMe-jXCH^iMe^  2  H W  A  A  B  246.6 (=CH) 103.6 (C H ) 5.71 (C^Si) 5  5  1.78 (CHSi(CH3)3)  130 ( C H ^ C H ^  B  2  H H  0.16 (s, 9H, C H ^ K C f f ^ Cp*W(0)(=CHSiMe )(CH SiMe ) 3  2  3  6  8.89 (s,lH, =CH, J = 7.1Hz) 1.62 (s, 15H, C (CHJ ) 0.49 (s, 9H, CHSKCffj)^ 036 (s, 9H, CHjSiCCff^j) 0.10 (s, 2H, Cff Si) 2  HW  S  2  S  2395 (=CH) 111.8 (C (CH3) ) 163 (CHj, / = 119Hz) 11.0(C (CH3) 2.16 ( C H S K C H ^ 1.6 (CI^SiCCHj)^ 5  5  1  c w  5  5  continued.  142 Table 4-II  continued  low-resolution mass spectral data m/z*  'H NMR data  " C ^ I f l N M R data  S  s  8  Complex  9.74 (s,lH, =CH,  Cp*W(0)(=CHPhXCHjSiMe^  J  2  mf  = 11.0 Hz)  727 (d, 2H, C ^ ) 7.19-7.08 (m, 2H, C H ) 6.86 (t, IH, Cgff ) 257 (<L 2H, Of Si, 6  S  5  2  ^ HH 2  J  mf  =  3  ,  5 H  z  >  = 27.8 Hz) S  0.18 (s, 9H, SiQCHJJ 9.69 (s,lH, =CH,  e  J  2  mf  = 11.6 Hz) 3  5  9.86 (s,lH, =CH,  e  J  2  HW  = 8.6 Hz)  7.9(d,2H,2o-Phtf, 2 / ^ = 75 Hz) 731(t,2H,2m-Ph/?, / = 75Hz) 7.0(t,lH,l/>-Phff,) 1.64 (s, 15H, C (CHJ ) 0.86 (s, 3H, CH 2 / ^ = 9.6 Hz) 2  H H  5  S  2433 (=CH, / = 157 Hz) 111.7 (C (CHJ ) 31.65 (CH3) 10.7(C (CH3) ) 1.92 ( S K C H ^ c w  r  Cp W(0)(=CHPh)(Me)  5  1  1.64 (s, 15H, C (Cfl ) ) 0.75 (s, 3H, CHJ 054 (s,9H, SiCCff^ 5  5  c w  S  e3  S  1  156 (s, 15H, C (CHJ )  Cp W(0)(=CHSiM )(Me)  2493 (=CH, 1 ^ = 161 Hz) 152.9 (quaternary PhQ 129.1,128.1,123.2 (aromatic Q 111.9 (C (CHJ ) 37.74 (CHj, / = 119.1Hz) 10.69 (C (CHJ ) 1.19 ( S K O ^  S  $  5  S  5  252.9 (=CH, ^ = 150.1 Hz) 145.2 (quaternary PhQ 1295,1293,128.5, 125.6 (aromatic Q 111.6 (C (CU^ ) 31.9 (CH3) 10.9(C (CH3) ) 5  5  S  5  y  a  Probe temperatures 100-150°C. ^ Assignments involve the most abundant naturally occurring isotopes in  each species (e.g. ^Mo,  1 8 4  W etc.).  c  The C / / H proton resonance of this compound was not observed due A  X  to overlap with other resonances. ^ Reliable mass spectral data for this complex could not be obtained because of its extreme air- and moisture-sensitivity.  e  This complex was not isolated but was simply generated by  heating a solution of its oxo trialkyl precursor in an NMR tube.  143 four-legged piano stool molecular structures with cis chloride ligands, i.e.  The cis orientation of the chloride ligands in these complexes is established by the methylene proton and carbon resonances in their *H and ^C^H} NMR spectra (Tables 4I and 4-II). As representative examples, the *H and C{ H} NMR spectra of 13  1  Cp*W(0)(Cl) (CH SiMe ) in C D are shown in Figure 4.1 (a) and (b), respectively. The 2  2  3  6  6  methylene protons of the CH SiMe ligand give rise to an AX pattern in the H NMR 1  2  3  spectrum with ^JJJJ = 6.3 Hz, a feature consistent with their being diastereotopic in nature. Furthermore, the gated decoupled C NMR spectrum of Cp*W(0)(Cl) (CH SiMe ) in 1 3  2  2  3  C D , Figure 4.1 (b), exhibits a low-field resonance at S 63.3 ppm assignable to the 6  6  methylene carbon atom of the CH SiMe group, and this resonance displays coupling to 2  two inequivalent protons ( / J  CH  3  = 116 and 129 Hz). If the molecular structure involved  trans chloride ligands, then equivalent methylene proton environments would be expected. Attempts to grow suitable single crystals of Cp*W(0)(Cl) (CH SiMe ) for a confirmation 2  2  3  of its solid-state molecular structure by an X-ray crystallographic analysis have so far been unsuccessful due to the proclivity of the compound to crystallize as needle-like platelets. In any event, like all the other oxo complexes described in Chapter 2, the Cp'W(0)(Cl) R 2  compounds appear to be monomeric 16-electron species containing the tungsten metal in its highest oxidation state of + 6.  144  (a)  11111 2 . 6  111 1 1 1 1 1 1 1 1 1 T T J T T 2 2 2 4  I  I  1 1 I I 2 . 0  I  I I I I I I I I I  • n y r r  rrp-r I-M  1 . 4  1. 6  1 8  i I i  iii| iiil|i 1. 2  i  o. a  i l 0  6  i| llii|iill|lll  0  4  0  2 P P M  (b)  I 140  1  J .  -'-  -'80  2 0  P P M  Figure 4.1 (a) 300 MHz *H and (b) 75 MHz gated decoupled C NMR spectra of n  Cp*W(0)(Cl) (CH SiMe ) in C D solution at ambient temperatures. 2  2  3  6  6  0  145 Since oxidative addition to the tungsten(VI) center in the dioxo alkyl reactant is unlikely, reactions 2 probably proceed via initial addition of HCl across one of the W=0 bonds as shown in eq. 3.  The resulting hydroxo chloro compound is expected to react quickly with a second equivalent of HCl to afford the final organometallic product, water being extruded as the by-product. Regrettably, monitoring of the various conversions 2 by *H NMR spectroscopy fails to reveal any evidence pro or con these putative intermediates. When Me3SiCl is used as the chlorinating agent, the extruded H2O presumably, 9  reacts with it to generate the necessary HCl in an autocatalytic fashion. Alternatively, Bercaw and Parkin have proposed that the reaction of Me3SiCl with Cp* W(0) proceeds 2  via direct attack of Me3SiCl at the W=0 bond. If such a mechanism is operative in the 10  Me3SiCl reactions of this work, the latter reactions would proceed as summarized in eq. 4.  146 Unfortunately, monitoring of the reaction between Cp*W(0) (CH SiMe ) and Me SiCl by 2  2  3  3  both H and ^Si NMR spectroscopy fails to reveal any evidence for the intermediates in X  eq. 4 even though these reactions with Me SiCl are much slower than those involving 3  excess HCl. The proton signals due to the starting materials slowly decrease in intensity over time, while proton signals assignable to Cp*W(0)(Cl) (CH SiMe ) and Me Si-02  2  3  3  SiMe increase in intensity. 3  In a manner similar to the reactions involving HCl, reactions of the dioxo alkyl complexes with PC1 may also involve direct attack at the W = 0 bond as outlined in eq. 5. 5  Addition of C1 P-C1 across one of the W=0 bonds could result in an adduct complex as the 4  intermediate, which could immediately rearrange to deliver an additional CI atom to the W center and eliminate as the phosphoryl trichloride, C1 P = 0. The isolated yields of the 3  desired oxo dihalides from these reactions are somewhat reduced than those from the reactions involving HCl or Me SiCl. This may be due to a possible competing reaction 3  involving the oxo dichloride product with either the excess PC1 reagent or with the 5  C1 P=0 species generated. No attempts were made in this work to see ifaCp W(C1) R 3  4  species could be prepared by using 2 mol equivalents of PC1 , although complexes of this 5  type would be of interest.  147 B.  Characteristic Reactivity of the CpW(0)(Cl) R Complexes (Cp'= Cp or Cp*, R= CH, 2  or C^SiMej). Some characteristic reactivity of the dichloro oxo alkyl complexes, Cp'W(0)(Cl) R, 2  has been investigated. Reactions of Cp*W(0)(Cl) R (R= CH SiMe or Me) with excess 2  2  3  H 0 in E^O solution regenerate the Cp*W(0) R starting materials in approximately 55% 2  2  yield. Treatment of Cp*W(0)(Cl) (CH SiMe ) with two equivalents of either NaOMe or 2  2  3  KO Bu does not generate the expected bis(alkoxide) complexes, l  Cp*W(0)(OR') (CH SiMe ) (R' = Me or *Bu) as isolable products. Color changes, 2  2  3  indicating that a reaction takes place, occur upon mixing of these reagents, however tractable products cannot be isolated. In the case involving reaction of Cp*W(0)(Cl) (CH SiMe ) with KOteu, a dark red -green crystalline product is obtainable 2  2  3  from a pentane extraction of the reaction mixture. This product complex is extremely airand water-sensitive and any attempt to characterize it by rR and *H NMR spectroscopy gives spectra that are consistent with Cp*W(0) (CH SiMe ) being the decomposition 2  2  3  product. The dichloro oxo alkyl products, Cp'W(0)(Cl) R, (Cp'= Cp or Cp*, R= CH SiMe 2  2  3  or Me) prepared in this work are very useful synthetic precursors since they provide a reliable synthetic route to oxo trialkyl and oxo alkylidene complexes, this route being depicted below, the cyclopentadienyl ring shown representing either C H or C Me . 5  5  5  + 2[M]CI  R=CH SiMe 2  3  ; [M]=l_i ; R'=Me  R=CH SiMe ; [M]=MgCI ; R'=CH SiMe 2  3  2  R=Me ; [M]=MgCI ; R*=CH Ph 2  3  or CH Ph 2  6  5  (6)  148  crR^R'-H)  tf\  CfR (R-H)  (a)  N  (b)  product R=CH SiMe ; R'=CH SiMe or CH Ph (a) R=CH SiMe ; R'=Me (b) R=Me ; R'=CH Ph (a) 2  3  2  3  2  3  2  2  Thus, treatment of the tungsten dichloro oxo compounds with 2 equiv of an alkyl-lithium or -Grignard reagent affords the oxo trialkyl complexes which, when isolable, may be obtained in ca. 60% yields from reactions 6 as yellow, crystalline solids. The presence of the three alkyl ligands in these compounds imparts to them a greater solubility in common organic solvents than that exhibited by their dichloro precursors. Indeed, their solubilities are even greater than those of the related dioxo alkyl species, e.g. Cp*W(0)(Me) (CH SiMe ) is very soluble in pentane. When exposed to air and moisture 2  2  3  both as solids or in solutions, the isolable oxo trialkyl complexes (Tables 4-1 and 4-II) appear to be reasonably stable for short periods of time. The spectroscopic properties of the Cp'W(0)R complexes (Tables 4-1 and 4-II) 3  indicate that they possess normal monomeric four-legged piano stool molecular structures. Their monomeric nature is suggested by their characteristic IR and mass spectra. The complexes as Nujol mulls exhibit a single i /  in the IR spectral region 941 - 935 cm' . 1  w  =  0  The most abundant mass ion peak observable in the low-resolution mass spectra of the oxo trialkyl complexes which are isolable is assignable to a species corresponding to the parent ion minus a methyl group.  11  The intramolecular dimensions of one of these oxo trialkyl species, namely CpW(0)(CH SiMe ) , have been established by single-crystal X-ray crystallography, 12  2  3  3  which also confirmed its normal monomeric four-legged piano stool molecular geometry.  149 The most chemically interesting aspect of the structure involves the W-O bond distance of 1.664 (8) A, which is consistent with the existence of a W=0 double bond.  13  As a result,  the oxo trialkyl complexes may be formulated as 16-electron species again containing the tungsten atom in its highest oxidation state of +6. Worthy of note is the fact that the W=0 bond length in CpW(0)(CH SiMe ) is significantly shorter than the W=0 linkages in 2  3  3  CpW(0) (CH SiMe ) (1.716 (5) A and 1.723 (5) A ) , which might suggest some triple 13  2  2  3  bond character in the W=0 link of the oxo trialkyl compound. When two of the alkyl ligands are different from the third in these oxo trialkyl species, as in Cp*W(0)(Me) (CH SiMe ) and Cp*W(0)(Me)(CH Ph) , their piano stool 2  2  3  2  2  molecular structures should result in their existing as cis and trans isomers. This feature is readily evident in the solution NMR spectra of Cp*W(0)(Me) (CH SiMe ) (Table 4-II), 2  2  3  the two isomers being designated as A (trans-Me groups) and B (cis-Me groups), as depicted below.  CH ' '^V'''CH SiMe3 O CH  Me SiCH -''^ v' 'CH3 O CH  1,  3  V  2  3  3  ,  2  3  B  A  Figure 4.2 shows the *H NMR spectrum of Cp*W(0)(Me) (CH SiMe ) in benzene-^ at 2  2  3  a30 °C. It is apparent from the NMR spectrum (Figure 4.2) and the NMR data (Table 4II) that the molecular structure involving trans methyl groups is the dominant form in solution, the ratio of the two isomers in benzene-**^ at =*30 °C is A:B = 63:37. The isomers are assigned in a manner similar to that discussed above for Cp*W(0)(Cl) (CH SiMe ). 2  2  Thus, the molecular structure involving trans methyl groups gives rise to equivalent methylene proton environment, whereas a cis methyl group structure generates diastereotopic methylene protons on the CH SiMe group. 2  3  3  150  JL  LA  | I I I I | I I I I I I I I I | I I I I I I I II | I I I I | I I I ! | I I I I I I I I I  !.!  2.0  i.B  Figure 42  1.6  1.4  I I I I I I I I I I | M I I I I I I I | f I M | I I I I | 1.2 1.0 O.B  A  JN  M i l  0.6  1111111111111111111111111111111111111111 O.S 0.0 -0.2 PPM -0.4  300 MHz *H NMR spectrum of Cp*W(0)Me (CH SiMe ) 2  in C D solution at ambient temperatures. 6  6  2  3  151  Interestingly, Cp*W(0)(Me)(CH Ph)2 does not exhibit similar isomerism, existing 2  exclusively as the trans-(CH Ph)2 isomer under identical conditions, i.e. 2  O  CH Ph 2  The *H NMR spectrum of this compound in benzene-^ at ^30 °C is shown in Figure 4.3. The trans nature of the benzyl groups is indicated by the observed AB pattern for two magnetically equivalent methylene proton environments on the CH Ph groups, the set of 2  H protons being cis with respect to the oxo ligand while the Hj, protons are cis with a  respect to the methyl group. This is further substantiated by the fact that the two doublets of the AB pattern exhibit different coupling to the tungsten center ( /vvH = 12.1 Hz and 2  a  2/wHb = 4.5 Hz). Presumably steric factors play the principal role in determining the thermodynamically most stable isomers of these oxo trialkyl complexes. Steric factors also appear to be dominant in determining whether the oxo trialkyl product complexes of reactions 7 are sufficiently thermally stable to be isolable in the first place under ambient conditions. When the coordination environment around the central metal is sufficiently congested, as for Cp*W(0)(CH SiMe )3 and Cp*W(0)(CH SiMe3)(CH Ph)2, the oxo 2  3  2  2  trialkyl complexes do not persist at temperatures above -20 °C. Instead, alkane elimination (presumably via intramolecular a-H abstraction) occurs as shown in eq. 7 and the new oxo alkylidene complexes, Cp*W(0)(=CHSiMe )(CH SiMe ) and 3  2  3  Cp*W(0)(=CHPh)(CH SiMe3), result. Furthermore, a H NMR spectrum of J  2  CpW(0)(CH SiMe3)3 in Q D exhibits small proton signals due to the corresponding oxo 2  6  alkylidene compound, CpW(0)(=CHSiMe )(CH SiMe3) (Table 4-II), and Me Si which 3  increase in intensity with time.  2  4  Figure 4.3  300 MHz 4i NMR spectrum of Cp*W(0)Me(CH Ph) 2  in C D solution at ambient temperatures. 6  6  2  153 For those oxo trialkyl complexes which are isolable at ambient temperatures, reactions 7 may be induced by gently warming solutions of them to 50-60 °C. For example, X  H NMR monitoring of the thermal conversion of Cp*W(0)(Me)(CH Ph) in C D at 2  2  6  6  50 °C indicates the clean formation of Cp*W(0)(Me)(=CHPh), the conversion being complete within 3 days. An APT C NMR spectrum of the product complex in C D B  6  6  establishes that the carbon resonances at 252.9 ppm and 31.9 ppm have one and three proton atom(s) attached to them, respectively. This feature is consistent with the existence of the benzylidene complex which appears to be the thermodynamically favored product from the thermolysis reaction. No signals attributable to Cp*W(0)(=CH )(CH Ph) are 2  2  evident in the NMR spectra. Under identical experimental conditions, Cp*W(0)(Me) (CH SiMe ) similarly 2  2  3  converts to Cp*W(0)(Me)(=CHSiMe ), the notable feature being that the trans isomer 3  converts more rapidly than its cis analogue. This is probably a kinetic phenomenon, since the proton undergoing  a-H elimination from the CH SiMe group can couple with either 2  3  of the two methyl groups of the trans-methyl isomer to expel C H .  Again, no direct  1 4  4  evidence for the methylidene complex, Cp*W(0)(=CH )(CH SiMe ), is evident in the 2  2  3  NMR spectrum although this complex might be expected to be formed as a result of intramolecular a-H elimination of C H from the cis-methyl isomer of 4  Cp*W(0)(Me) (CH SiMe ). 2  2  3  Whether external heating is required or not, reactions 7 afford the final oxo alkylidene products in high yields (40-70%). These complexes are much more sensitive to air and moisture, both as solids and in solutions, than are their oxo trialkyl precursors. The physical properties of these complexes (Tables 4-1 and 4-II) are consistent with their having the molecular structures shown in eq. 7. As a result they are again formulated as 16electron species containing the tungsten in its highest oxidation state of +6. It is still of interest to obtain suitable single crystals of one of the isolable Cp'W(0)(=R')R complexes  154 in order that an X-ray crystallographic analysis may be performed on it. There are few structurally characterized oxo alkylidene complexes reported in the literature to date. C.  Future Studies It is hoped that the new Cp'W(0)(a) R compounds prepared in this work will be 2  viable synthetic precursors to some other new types of organometallic oxo complexes. For instance, it is possible that reactions of these complexes with hydride reagents, such as LiAlH or NaBH^ will afford oxo hydride complexes, Cp'W(0)(H) R (R= CH SiMe or 4  2  2  3  Me). In addition, reductions of the oxo dichloro complexes with Na/Hg in the presence of olefins or acetylenes may afford a variety of oxo olefin or oxo acetylene complexes. Indeed, examples of the expected oxo acetylene complexes from these reactions are known. For example, the complexes Cp W(0)(r7 -C H )R, (Cp' = Cp or Cp*; R = Me or C(O)Me) ,  2  2  2  have been isolated and fully characterized from the reactions of Cp'W(CO) [CH = CHC(6)Me] and Cp W(CO) (r? -C H )R with NO in toluene solutions ,  2  2  2  2  2  at -78 "C. It is also hoped that the molybdenum analogues of the tungsten dichloro oxo 15  alkyl complexes prepared in this work might be preparable at low temperatures. In addition, aspects of the characteristic chemistry of the oxo alkylidene complexes should be the focus of future work in these laboratories. Specifically, it is of interest to determine whether these complexes react with olefins in solution at room temperature or at some elevated temperature to undergo the transformation outlined in eq. 8 below. Perhaps a Lewis acid, such as A1C1 , may be required as a co-catalyst. 3  CpW(0)(=R)R + R*HC=CHR*  •Cp W(0)(=R*)R + R'HC=CHR* ,  Coordination of the Lewis acid to the oxygen atom of the W=0 link via the aluminum, would render the metal center more electron deficient and this might allow for easier electrophilic attack of the metal-alkylidene link on the reactant olefin. For instance,  (8)  155 Osborn and co-workers have shown that extremely active metathesis catalysts are obtained from M(0)X(CH CMe ) (M = Mo or W; X = CI or Br) and a Lewis acid such as AlBr . 2  3  3  3  The adduct complexes produced, i.e. [MX(CH CMe ) ]-(M-0)-[AlBr ], are 2  3  3  3  photochemically activated by the loss of Me C to generate the alkylidene complexes 4  [MX(=CHCMe )(CH CMe )]-(/i-0)-[AlBr ] which are extremely active metathesis 3  2  3  3  catalysts.  16  Furthermore, the oxo alkylidene complexes prepared in this work may have some synthetic utility in organic synthesis. Reactions with unsaturated organic reagents such as C 0 , ketenes, alkyl isocyanates, etc. are of interest, as new types of organometallic 2  complexes may be preparable.  Summary  Reactions of representative examples of the dioxo alkyl complexes of tungsten, Cp'W(0) R, with chlorinating agents such as HC1, Me SiCl and PC1 afford cis-dichloro 2  3  5  oxo alkyl complexes, Cp'W(0)(Cl) R, several of which have been isolated and 2  characterized in this work. These Cp'W(0)(Cl) R complexes are important and useful 2  organometallic synthons since treatment of these complexes with either alkyl-lithium or alkyl-Grignard reagents affords a series of oxo trialkyl complexes, Cp'W(0)R , which in 3  turn thermally convert to novel oxo alkylidene complexes of tungsten, Cp'W(0)(=CHR')(R). All the spectroscopic data for these new types of oxo complexes are consistent with their possessing normal piano stool molecular structures. Furthermore, they are all formally 16-electron species containing the tungsten atom in its highest oxidation state of + 6. Other aspects of the characteristic chemistry of these dichloro oxo alkyl and oxo alkylidene complexes will be the focus of future work in these laboratories.  156 References and Notes:  (1)  Taken in part from: (a) Legzdins, P.; Phillips, E.C; Rettig, SJ.; Sanchez, L.; Trotter, J.; Yee, V . C Organometallics 1988, 7,1877. (b) Legzdins, P.; Phillips, E.C; Sanchez, L. Organometallics 1989,5,930.  (2)  Kochi, J.K. Organometallic Mechanisms and Catalysis; Academic: New York, 1978.  (3)  (a) Cousins, M.; Green, M. L. H. /. Chem. Soc. 1964,1567. (b) Cousins, M.; Green, M. L. H. / . Chem. Soc. A 1969,16.  (4)  Similarly, it is the W=0 link that is transformed in reactions of the dioxo alkyl complexes with hydrogen peroxide, affording novel peroxo alkyl Cp'M(r? -0)(0)R 2  complexes, see Chapter 5 of this thesis. (5)  As mentioned in the introduction section of Chapter 2, the first examples of these complexes namely, CpW(0)(CH SiMe ) and CpW(0)( = CHSiMe )(CH SiMe ), 2  3  3  3  2  3  were obtained in very low yields from some of the early chemical studies of the complex CpW(NO)(CH SiMe ) . 2  (6)  3  2  (a) Whitmore, F. C ; Sommer, L. H. /. Am. Chem. Soc. 1946, 68,481. (b) Kharasch, M.S. Grignard Reactions of Non-Metallic Substances; Prentice-Hall: New York, 1954.  (7)  This solution was prepared by bubbling HCl gas through dry, deaerated E^O and was standardized by titration with 1.0 M aqueous NaOH.  (8)  Powell, P. Principles of Organometallic Chemistry, 2nd ed.; Chapman and Hall: London, 1988.  (9)  This organic liquid is extremely difficult to keep dry as it is extremely moisture sensitive, and therefore, almost always contains some dissolved HCl.  (10)  Bercaw, J.E.; Parkin, G /. Am. Chem. Soc. 1989, 111, 391.  (11)  The Si-CH bond is quite susceptible to cleavage under the conditions of the 70 eV 3  mass spectrum obtained.  157 (12)  Legzdins, P.; Rettig, S. J.; Sanchez, L. Organometallics 1985,4,1470.  (13)  Feinstein-Jaffe, I.; Gibson, D.; Lippard, S J.; Shrock, R.R.; Spool, A. /. Am. Chem. Soc. 1984,106,6305 and references therein.  (14)  Conversely, the cis-methyl isomer would have only one methyl group cis to the a-protons on the CH SiMe group, and therefore, the transformation to expel C H 2  3  is expected to be slower. (15)  (a) Alt, H.G.; Hayen, H.I. Angew. Chem. Int. Ed. Engl 1985,24,497. (b) Alt, H.G.; Hayen, H.I. /. Organomet. Chem. 1986,316,105.  (16)  (a) Kress, J.R.M.; Osborn, J.A.; Russell, MJ.M.; Wesolek, M.G. /. Chem. Soc, Chem. Commun. 1980,431. (b) Kress, J.R.M.; Le Ny, J.-P.; Osborn, J.A.; Wesolek, M.G. /. Chem. Soc, Chem. Commun. 1981,1039.  4  Chapter 5  Novel Organotransition Metal Peroxo Alkyl Complexes of Molybdenum and Tungsten: Cp'M(O) (»? -0 )R 2  2  [Cp = Cp (r? -C H ) or Cp* (i? -C Me ), 5  5  5  5  5  5  M= Mo or W, R= CH SiMe3 or CH ] 2  3  1  159  Introduction Transition-metal peroxide complexes possessing the linkage M^-Oj) are considered to be potential oxygen-atom transfer agents and represent an important class of reactive intermediates in catalytic oxidation processes of hydrocarbons. There are 2  numerous examples of inorganic transition-metal peroxo complexes reported in the literature, ' those of the group 6 metals being the most common and possibly the best 3 4  characterized; however, only a few of these are reactive towards olefins. This is possibly due to their being relatively insoluble in common organic solvents and the fact that most transition-metal peroxo complexes are coordinatively saturated. Organometallic peroxo complexes, on the other hand, are considerably more soluble in organic solvents and are commonly invoked as key intermediates in the oxygen atom transfer step of the oxidation processes (e.g. epoxidation). Only recently have a few examples of these organometallic peroxo complexes been isolated and characterized by conventional spectroscopic methods and X-ray crystallography. *' *'' ' Investigations into 1  1  5 6  the characteristic chemistry of these complexes may provide details of the actual oxygenatom transfer step in metal-mediated oxidation processes which is as yet not well understood.  3  The first part of this chapter describes the synthesis of the first examples of a new class of formally 16-electron organometallic peroxo alkyl complexes of both molybdenum(VI) and tungsten(VI), Cp'M(0)(t7 -0 )R, [Cp'= (»? -C H or »? -C Me , M = 2  5  2  Mo or W, R = CH SiMe or C H ] . 2  3  3  l a , l b  5  5  5  5  5  These new types of oxo complexes are conveniently  prepared from reactions of H 0 with the cyclopentadienyl dioxo alkyl complexes, 2  2  Cp'M(0) R, whose synthesis and characterization are discussed in Chapter 2. A discussion 2  of the spectroscopic and physical properties of the peroxo species is also presented. It should be noted that at the same time as this work was being done, two of the Cp*W(0)(r? 2  O^R compounds, namely Cp*W(0)(r? -0 )(CH SiMe ) and Cp*W(0)(f? -0 )Me, were 2  2  2  2  3  2  160 prepared via a different route and reported by Faller and Ma. Recently, additional 5  examples of these types of organometallic peroxo complexes have appeared involving the Group 5 metals T a and Nb, i.e. Cp^Ta^-O^R (R = Me, Et, Pr, Ph or CH Ph) and 6a  7  n  2  Cp* Nb(r -0 )X (Cp* = C H SiMe ; R = CI or Me). 2  2  ?  2  5  4  3  The second part of this chapter describes my preliminary investigations into the possible utility of these coordinatively unsaturated organometallic peroxo compounds as oxygen-atom transfer agents or catalysts for the epoxidations of olefins. In particular, the reactions of Cp*W(0)(r? -0 )(CH SiMe ) with phosphines (PPh ) and olefins 2  2  2  3  3  (tetracyanoethylene) are described. In addition, the spectroscopic properties (including UV-vis absorption studies) and the results of a single crystal X-ray crystallographic analysis of an interesting charge-transfer complex between Cp*W(0)(r? -0 )(CH SiMe ) and 2  2  2  3  tetracyanoethylene are presented. To the best of my knowledge this is the first structurally characterized example of a transition-metal peroxo complex functioning as an electron donor in a charge-transfer complex.  Experimental Section All reactions and subsequent manipulations were performed under anaerobic and anhydrous conditions. The general procedures employed in this work were as described in the Experimental Sections of the preceding chapters. The tetracyanoethylene (TCNE) was obtained from Eastman as a white crystalline solid and was purified by sublimation in vacuo (0.01mm) onto a cold water probe before use. The cyclopentadienyl dioxo alkyl complexes, Cp'M(0) R (Cp'= Cp or Cp*, M= Mo or W; R= C H or CH SiMe ), used as 2  3  2  3  reagents in this work were prepared by the methods outlined in Chapter 2 of this Thesis. The solution H and C NMR spectra were obtained on a Varian XL-300 FT NMR 1  B  spectrometer with reference to the residual proton of the deuterated solvent used. The solution P ^H} NMR spectra were obtained at 121 MHz on Varian XL-300 FT NMR 3 1  161  spectrometer with reference to 85% H3PO4 in D2O. The IR spectra were recorded on a Nicolet 5DX-FT IR spectrometer, internally calibrated with a Ne/He laser. The UV-vis. absorption spectra were measured with the assistance of Ms. Karen Rupitz on a PYE UNICAM PU 8800 spectrometer kindly made available to me by Dr. S. Withers of this Department. Elemental analyses were performed by Mr. P. Borda of this Department. Preparations of the Cp'M(0)(r? -0 )R Complexes (Cp' = Cp or Cp*, M = Mo or W, 2  2  R = Me or CH SiMe3). Only in the case of Cp*W(NO)(CH SiMe3)2 did the treatment of 2  2  the Cp'M(NO)R2 complexes with an excess of H2O2 result in a high yield of the corresponding Cp'M(0)(r? -02)R species. In general, these latter compounds were best 2  synthesized by employing the Cp'M(0) R (Cp'= Cp or Cp*, M= Mo or W; R= C H or 2  3  CH SiMe3) complexes as the synthetic precursors. In this manner, and as a lb  2  representative example, the preparation of CpW(0)(r? -02)(CH.2SiMe3) was as detailed 2  below. To a stirred, colorless solution of CpW(0) (CH SiMe ) (0.30 g, 0.81 mmol) in Et 0 2  2  2  3  (25 mL) at ambient temperature was added by microsyringe a 30% by weight aqueous solution of H 0 (0.32 mL, 4.1 mmol H2O2). After 5 h, there had been no apparent 2  2  change in the appearance of the reaction mixture; it was taken to dryness in vacuo to obtain a pale yellow solid. Recrystallization of this solid from Et20 at -25°C gave colorless plates of analytically pure CpW(0)(r? -0 )(CH SiMe ) (0.25 g, 80% yield). 2  2  2  3  The reaction to prepare the permethylcyclopentadienyl analogue, Cp*W(0)(r/ 2  02)(CH SiMe3) (0.41 g, 52% yield), in this manner was monitored during the course of the 2  reaction by periodically removing aliquots from the reaction mixture and taking them to dryness in vacuo. Nujol mull IR spectra of the dried residues showed the gradual diminution in intensity with time of the two vw=o' due s  t 0  Cp*W(0)2(CH SiMe3) at 937 2  (asym) and 901 (sym) cm" and the appearance and growth of new absorptions due to 1  Cp*W(0)(n -02)(CH SiMe3) at 941 ( « / 2  ) , 868 (i/ -o). and 567 and 559 («/ _ ) cm" . 1  2  w = 0  0  w  0  162 The analytical, mass spectral, IR and ^ and  NMR data for these and all the  other peroxo alkyl complexes prepared during this work are presented in Tables 5-1 and 5H. Using a method similar to that described above, white Cp*W(0)(r; -02)Me (0.07 g, 2  71 % yield) and CpW(0)(r? -02)Me (0.055 g, 59 % yield), and pale-yellow Cp*Mo(0)(r; 2  2  0 )(CH SiMe ) (0.075 g, 70 % yield), CpMo(0)(fj -02)(CH SiMe ), (0.06 g, 63 % yield), 2  2  2  3  2  3  Cp*Mo(0)(r -0 )Me (0.073 g, 69 % yield), and CpMo(0)(i7 -0 )Me (0.045 g, 56 % yield), 2  ?  2  2  2  were prepared from their requisite starting Cp'M(0) R complexes. The yields reported 2  above refer to isolated crystalline products. Yields calculated for crude powders from the reactions indicate that the conversions are nearly quantitative (=;90%). Reaction of Cp*W(0)(r? -0 )(CH SiMe3) with PPh .  In E^O (10 mL) white,  2  2  2  3  analytically pure Cp*W(0)(r? -0 )(CH SiMe ) (0.055 g, 0.12 mmol) and white PPh (0.034 2  2  2  3  3  g, 0.13 mmol) were mixed, and the resulting clear colorless solution was stirred at room temperature for ~ 1 h. The Et^O solvent was then removed in vacuo to afford a white residue which was dried under reduced pressure for ~30 min. An IR spectrum of this solid residue as a Nujol mull exhibited absorption bands at 937 and 901 cm" which are 1  attributable to Cp*W(0) (CH SiMe ) (see Table 2-m) and absorption bands at 1190, 2  2  3  1121, 721, 696, and 540 cm" which are assigned to Ph P=0. Furthermore, the IR 1  8  3  spectrum did not exhibit absorptions due to Cp*W(0)(r? -0 )(CH SiMe ) (vide supra). A 2  2  1  2  3  H NMR spectrum of the reaction residue in C D exhibited phenyl proton resonances in 6  6  the region 8.0-6.5 ppm and signals at 6 1.75 (s, 15H, C (CH ) ), 5  0.48 (s, 2H, CH^) and 0.42  3 5  (s, 9H, CH Si(C/7 ) ) which are diagnostic of Cp*W(0) (CH SiMe ). In addition, a P 3 1  2  3  3  2  2  3  NMR spectrum of the reaction residue in CDC1 exhibited a singlet at -28.64 ppm 3  (due to Ph P=0) and a singlet at +6.00 ppm (due to residual Ph P). ' 9  3  3  10  163 Preparation of [Cp*W(0)(r; -0 )(CH SiMe3)] -M-[(CN) C=C(CN) ] and 2  2  2  2  [CpW(0)(r? -0 )(Me)] -/*-[(CN) C=C(CN) ]. 2  2  2  These complexes were prepared in a  2  2  2  2  similar manner from reactions of the appropriate Cp'W^X^-O^R complex with tetracyanoethylene. The method for the preparation of [Cp*W(0)(f? -02)(CH SiMe )]22  2  3  /i-[(CN) C=C(CN) ] is described below as a representative example. 2  2  In E^O (20 mL), white crystalline Cp*W(0)(f? -0 )(CH SiMe ) (0.10 g, 0.22 mmol) 2  2  2  3  and white TCNE (0.028 g, 0.22 mmol) were mixed. The reaction solution immediately changed from colorless to orange-yellow in color. The Schlenk tube was then wrapped in Al foil to shield the reaction solution from light. After 12 h a pale brown solution was filtered from a white precipitate that had formed. An IR spectrum of this white solid as a Nujol mull established that it was [Cp*W(O) ] 0*-O) ("w = o 11  2  '  9 3 1  2  8  8  3 cm  " )- ^ 1  E t  2°  solvent was removed from the filtrate under reduced pressure to afford an orange brown solid. The solid was washed with hexanes (2x5 mL) and was recrystallized from EtjO at 20 °C to give 0.091g of orange-brown crystals of [Cp*W(0)(r7 -0 )(CH SiMe )] -/i2  2  2  3  2  [(CN) C=C(CN) ] (80 % yield, based on W). 2  2  Anal. Calcd. for C ^ H ^ O g N ^ i ^ : C, 39.39 ; H, 5.06 ; N, 5.40. Found : C, 39.49 ; H, 5.05 ; N, 5.34. IR (Nujol mull) v cm" ; *  si  1  w  =  0  551 (m) cm" ;  3  1  (d, 1H, / a - H b 2  =  12  H  Ha-Hb=  12  1  0  > ^  2  5  1242 (ms)  -1  943 (s)cm" ;i/ . 860 (s) cm" ; d * 850,829 (s) cm" ; v  1  2/  2255, 2218,2145 (w) cm ; x/ .  c = N  Me  1  0  Cp  W Q 2  577,561,  380 (m) cm" . H NMR (C D )« 1.57 (s, 15H, C ( G ¥ ) ) , 0.93 1 X  0  =  w  =  > ^ 5  0  2/  6  Ha-w=  i  a  ^  i  2/  Ha-Si=  7  3  6  ^  5  3  5  Ci/ H SiMe ), 0.52 (d, 1H, a  b  3  W = 73 Hz; / . = 7.2 Hz, CH tf SiMe ), 0.46 (s, 9H, / . = 2  2  Ha  si  a  b  3  Ha  si  6.4 Hz, Si(C/Y ) . ^C-^H} NMR (C D )d 117.2 (s, C (CH ) ), 108.8 (s, C (CN) ), 108.2 3  3  6  (s, C2(CN) ), 33.5 (s, / . = 1  4  c  w  6  5  3  5  2  4  Hz, CH H SiMe ), 10.11 (s, C (CH ) ), 1.83 (s, Si(CH ) ). a  b  3  5  3  5  3  3  Low-resolution mass spectrum (probe temperature 120 °C) m/z 584 [P -Cp*W(0)(n +  0 )(CH SiMe )]. 2  2  3  2  164 Similarly, yellow-orange crystals of the analogous [CpW(0)(i7 -02)Me] -M2  2  [(CN) C=C(CN)2] complex were prepared in approximately 55% yield. 2  Anal. Calcd. for C H 0 N W : C, 28.75 ; H, 2.14 ; N, 7.45. Found : C, 28.89 ; H, l g  1 6  2.13 ; N, 7.65. IR (Nujol mull) i/ _ w  6  v _ Cp  4  2  3115, 3098 (m) cm ; -1  H  _1  1  -1  c=N  964 (s) cm ;j/Q.Q 864 (s) cm" ; s _ 835 (s) cm" ; y  0  2257,2222 (w) cm ;  i>  578,565 (m) cm" .  1  Cp  H  1  W 0 2  Low-resolution mass spectrum (probe temperature 120 °C) m/z 312 [CpW(0)(r/ 2  0 )Me] . +  2  X-ray Crystallographic Analyses of the Complexes CpW(0)v7 -0 )(CH SiMe3) and 2  2  [Cp*W(0)(r? -0 )(CH SiMe3)] - /* -[(CN) C=C(CN) ]. 2  2  2  2  2  2  2  A suitable single crystal of each  of the complexes CpW(0)(r -0 )(CH SiMe ) and [Cp*W(0)(r/ -0 )(CH SiMe )] -/i2  2  ?  2  2  3  2  2  3  2  [(CN) C = C(CN) ] was obtained from crystallizations over many days from E^O solutions 2  2  at -20 °C. The X-ray structure determinations were performed in a similar manner by Drs J. Trotter and S. Rettig and Ms. V.C. Yee of our Department. ' The pertinent 12 13  crystallographic data, the refinement details and the final positional and equivalent isotropic thermal parameters ( £ 7 ^ = 1/3 x trace diagonalized U) for each complex are reported elsewhere. Bond lengths and bond angles for CpW(0)(f? -0 )(CH SiMe ) and 10  2  2  2  3  [Cp*W(0)(r -0 )(CH SiMe )] -/^-[(CN) C=C(CN) ] are listed in Tables 5-m and 5-IV, 2  ?  2  2  3  2  2  2  respectively. Views of the solid-state molecular structures of CpW(0)(r; -0 )(CH SiMe ) and 2  2  2  3  [Cp*W(0)(r -0 )(CH SiMe )] -/x-[(CN) C=C(CN) ] are given in Figures 5.2 and 5.3, 2  ?  2  2  3  2  2  2  respectively. Determination of the Formation Constant for [Cp*W(0)(n -0 )(CH SiMe3)] -/*2  2  2  2  [(CN) C=C(CN) ] in CH C1 Solution. Typically, the charge-transfer (CT) complex was 2  2  2  2  prepared in situ by mixing appropriate weighed amounts of Cp*W(0)(f7 -0 )(CH SiMe ) 2  2  2  and TCNE in methylene chloride in a 5- or 10-mL volumetric flask. The volume of  3  165  solution was assumed to remain unchanged during the UV-vis absorption measurements. Figure 5.5 illustrates the change in absorption with the changes in concentration of Cp*W(0)(f? -02)(CH SiMe3) and TCNE. The concentrations of the two reagents 2  2  indicated in Figure 5.5 from top to bottom were as follows; (A) ^1.4xl0- MCp*W(O)(r; . 2  2  0 )(CH SiMe ) with 0.101 M, 0.083 M, 0.064 M, 0.056 M, 0.043 M, 0.026 M, 0.015 M, 2  2  3  0.0069 M, and 0.000 M TCNE;  (B) * 1.55 x 10* M TCNE with 0.035 M, 0.029 M, 0.020 2  M, 0.014 M, 0.010 M, 0.008 M, 0.006 M, and 0.00 M  Cp^(0)(r, -02)(Crl SiMe ). 2  2  3  166 Results and Discussion A.  Synthesis and Some Physical Properties of the Cp'M(0)(r? -0 )R Complexes (Cp' = 2  2  Cp or Cp*, M = Mo or W, R = Me or CI^SiMe^. Treatment of representative members of the 16-electron dioxo alkyl complexes Cp'M(0) R of either molybdenum or tungsten with 30% H 0 (aq) in E^O results in then2  2  2  clean conversion to the corresponding Cp'M(0)(»7 -0 )R compounds as summarized in 2  2  eq. 1, where M = Mo or W, R = Me or CH SiMe and the cyclopentadienyl ring 2  3  represents either C H or C Me . 5  5  5  5  The peroxo alkyl product complexes are white (M = W) to pale yellow (M = Mo) airstable solids which are isolable from reactions 1 in moderate to high yields (ca. 55 - 80%). These solids are slightly less soluble in common organic solvents than are their dioxo alkyl precursors, and most of them melt explosively. The spectroscopic properties of the peroxo alkyl complexes, which are summarized in Tables 5-1 and 5-U, are consistent with their possessing the molecular structures shown in eq. 1 both in solutions and in the solid state. Their Nujol mull IR spectra exhibit strong bands in the regions 960-930, 875-850, and 575-555 cm" , assignable to v 1  M = Q  , i / _ , and 0  0  "M-0' respectively (Table 5-1). In most cases the parent ion is observable in their mass spectra, the most abundant daughter ion being attributable to [P-OJ , which corresponds +  167  Table 5-1. Analytical and IR Data for the Peroxo Alkyl Complexes Analytical Data (%)  mp.°C  CpMo(0)(r? -0 )Me 2  2  1  H  c Complex  IR data (Nujol, cm' ) "M=O  "o-o  "Si-Me  "M-O  calcd  found  calcd  found  32.17  32.49  3.60  3.86  949  878  575565 1238 569,557  CpMo(0)(r? ^C^XCr^SiMe-j)  85 dec  36.49  36.37  5.44  5.40  943  875  Cp*Mo(0)(r? -0 )Me  97 dec  44.90  44.87  6.17  6.13  934  875  565  CpW(0)(r/ -0 )Me  127 dec  23.10  23.36  2.58  2.70  955  864  569  CpW(0)(»?  107-108  28.15  27.95  4.20  433  955  852  1238 563  116 dec  34.57  34.55  4.75  4.70  949  860  571  110-111  37.01  37.11  5.77  5.80  941  868  2  2  2  2  2  -OJ(CH£rMeJ  Cp*W(0)(r? -0 )Me 2  2  Cp*W(0)(r? -0 )(CH SiMe ) 2  2  2  3  1238 567,559  168 Table 5-II. Mass Spectral and *H and ^CrH} NMR Data for the Oxo Complexes low-resolution mass spectral data m/z  'H NMR data  13  C{ H}NMR data 1  ( W  3  Complex  s  b  226, [P] 210, [P-0]  CpMo(0)(r? -0 )Me 2  CpMo(0)(r? -0 )(CH SiM ) 2  2  2  e3  +  282, [P-0] 267,[P-0,Me]  1093 (C H ) 24.9 (CH3)  5.26 (s, 5H, C ^ ) 1.% (s, 3H, 0/3)  +  2  6 5  5  536 (s, 5H, C j / y 2.06 (d, IH, C f f H , / =12.4Hz) 135 (d, IH, CH H , 12.4 Hz) 035 (s, 9H, 81(0/3)3)  +  +  A  B  2  HH  A  Cp*Mo(0)(r/ -0 )Me  2%, [P] 280, [P-0] 265, [P-O.Me]  1.67 (s, 3H, O/3) 1.47(s,15H,C (Of3) )  312, [P] 296, [P-0] 281, [P-0,Me]  535 (s, 5H, C ^ ) 156 (s, 3H, CH / = 8.7Hz)  +  2  2  B  +  5  118.1 (C (CH3) ) 29.9 (CH3) 9.81 (C (CH3) ) 5  5  +  CpW(0)(r? -0 )Me  5  +  2  2  +  CpW(0)(r? -0 )(CH SiMe ) 2  2  2  3  368, [P-0] 353, [P-O.Me]  +  5  +  2  382, [P] 366, [P-0] 351, [P-0,Me]  5.42 (s, 5H, C ^ ) 1.47 (<L1H[,Cff H , / = 13 Hz, = 13 Hz) 0.86 (d, IH, CH tt , J = 13 Hz, ^ = 8 Hz) 0.34 (s, 9H, 51(0/3)3)  109.6 (C H ) 28.1 (CHj, / =92.7Hz) 1.7(Si(CH3)3)  136 (s, 15H, C (CHJ ) 139 (s, 3H, CH ^ =8.4 Hz)  117.0 (C (CHJ ) 25.2 (CH , = 101.3 Hz) 9.7(C (CH3) )  2  A  +  2  B  B  HH  m  $  +  5  1  2  2  2  3  438, [P-0] 423, [P-0,Me] +  138 (s, 15H, C (CHJJ 0.93 (d, 1 H , C H H , V = 12.6 H z , / ^ = 10 Hz) 052 (d, l H . C H ^ , / ^ 12.6 Hz, = 6.9 Hz) 0.45 (s,9H, 81(0/3)3) A  B  H H  2  2  0  Probe temperatures 100-150 °C.  in each species (e.g. ^Mo, coupling to  1 8 3  184  W (I = 1/2).  W).  c  5  3  $  +  c  cw  5  Cp*W(0)(r? -0 )(CH SiMe )  5  S  S  y  +  c  c w  2  2  5  3  H W  +  5  109.0 (C H ) 20.1 (CH , V = 96Hz)  y  A  Cp*W(0)(r/ -0 )Me  5  5  117.2 (C (CH3) ) 33.1 (CHj, / = 97.4Hz) 9.9(C (CH3) ) 15(Si(CH3)3) 5  5  1  c  c w  5  5  Assignments involve the most abundant naturally occurring isotopes Approximately 14% of the resonance is a superimposed doublet showing  169  (b)  T—>  Figure 5.1.  |  T 1 T  100  -i—l—l—i—i—r-  -j—i—i—i—j—r  NMR spectra of CpW(0)(»? -0 )(CH SiMe ) in C D solution at -25 °C, 2  2  2  3  6  6  (a) 300 MHz *H NMR spectrum and (b) 75 MHz ^C^H} NMR spectrum.  170 to [Cp'M(0) R] . Figure 5.1 displays the H and C NMR spectra of CpW(0)(rj +  X  B  2  2  0 )(CH SiMe ) which are shown as representative examples for all the peroxo complexes. 2  2  3  Interestingly, the introduction of the extra oxygen atom into the metal's coordination sphere in the dioxo alkyl complex has a significant effect on the methyl and methylene signals in the 4 1 and C NMR spectra of the resulting peroxo complexes. The B  inequivalent methylene protons give rise to AB patterns in the H NMR spectra (Table 51  II), and both the methyl and methylene resonances occur some 0.5-1.0 ( H) and  10 ppm  1  (^C) downfield from those exhibited by their dioxo alkyl precursors (see Table 2-IV). Furthermore, the  values diminish by approximately 40 Hz, thereby reflecting the  different electron density at the metal center extant in the Cp'W(0)(»? -0 )R species. 2  2  A single-crystal X-ray crystallographic analysis of CpW(0)(»? -0 )(CH SiMe ) has 2  2  2  3  confirmed the monomeric nature of these new types of organometallic peroxo complexes in the solid-state. A view of the molecular structure of CpW(0)(r7 -0 )(CH SiMe ) is 2  2  2  3  presented in Figure 5.2, and illustrates that the complex possesses a slightly flattened threelegged piano stool molecular geometry. Selected bond lengths and bond angles of the molecule are listed in Table 5-III. The most chemically interesting features of the structure are those that describe the oxo and peroxo linkages. The observed intramolecular dimensions are fully in accord with W=0 and W(r; -0 ) linkages. As a result, all of these 2  2  Cp'M(0)(rj -0 )R complexes are formulated as 16-electron species containing the 2  2  molybdenum or tungsten atom again in its highest oxidation state of + 6. B.  Reactions of Cp*W(0)(r? -0 )(CH^iMe^ with Ph P and (CN) C = C(CN) . 2  2  3  2  2  The mechanism for the transfer of an oxygen atom from a transition-metal peroxo complex to an olefin has been the subject of many reports. Generally, the reaction is considered to proceed via 1,3-dipolar cycloaddition of the peroxo group to the olefin. This may be accomplished in a number of ways. Two possible pathways are depicted in Scheme 5-1. The first mechanism involves pre-coordination of the olefin to the unsaturated  171  Table 5-III. Selected Bond Lengths (A) and Bond Angles (deg) in CpW(0)(r7 -0 )(CH SiMe3). 2  2  a  2  Bond Lengths 1.68(3) 1.92(3) 1.87(3) 1.44(3) 2.15(4) 1.84(4) 1.81(5) 1.88(4)  W-O(l) W-0(2) W-0(3) 0(2)-0(3) W-C(6) C(6)-Si Si-C(7) Si-C(9)  2.09(3) 1.48(7) 1.44(7) 1.28(7) 138(6) 132(6) 1.81(4)  W-Cp C(l)-C(2) C(l)-C(5) C(2)-C(3) C(3)-C(4) C(4)-C(5) Si-C(8) &  Bond Angles 0(l)-W-0(2) 0(l)-W-0(3) 0(1)-W-C(6) 0(3)-W-C(6) 0(3)-W-Cp C(6)-W-Cp W-C(6)-Si C(6)-Si-C(8) C(6)-Si-C(9) C(7)-Si-C(8) C(8)-Si-C(9) C(7)-Si-C(9) b  b  a  99.7(13) 106.5(14) 99.2(14) 77.8(14) 132.4(11) 108.7(16) 119.(2) 110(2) 107(2) 113(2) 112(2) 106(2)  Estimated standard deviations are provided in parentheses. Cp refers to the centroid of the (rj -C H ) ring. 5  5  0(2)-W-0(3) W-0(2)-0(3) W-0(3)-0(2) 0(2)-W-C(6) 0(1)-W-Cp 0(2)-W-Cp C(6)-Si-C(7) C(l)-C(2)-C(3) C(2)-C(3)-C(4) C(3)-C(4)-C(5) C(4)-C(5)-C(l) C(5)-C(l)-C(2) b  b  44.7(10) 66(2) 70(2) 122.4(13) 118.2(11) 108.9(11) 108(2) 110(5) 107(5) 110(5) 107(4) 102(5)  173  Scheme 5-1.  174 transition-metal center, a process which renders the olefin electrophilic in nature. The next step would presumably involve electrophilic attack of the olefin at the peroxo group resulting in insertion of the olefin into one of the metal-oxygen bonds of the metal-peroxo linkage and formation of a five-membered peroxometallacycle. The alternate mechanism involves interaction of the olefin with one of the peroxo ligand oxygens of the complex and subsequent nucleophilic attack of the end carbon atom of the olefin at the metal center would form the peroxometallacyclic intermediate. Decomposition of the peroxometallacyclic intermediate in either case via a 1,3-dipolar cyclo-reversion mechanism would generate a transition-metal oxo complex and the epoxidized olefin. Examples of the peroxometallacycle have been isolated from reactions of Pt and Rh peroxo complexes with cyano-substituted olefins, but not in the case of such complexes containing Group 4-6 metals. Reports concerning spectroscopic, kinetic and theoretical studies which are consistent with either of the mechanistic pathways presented in Scheme 5-1 can be found in the literature. The route suggested depends on the transition-metal peroxo 4  complex and on the olefin used in the reactions (i.e. whether it is nucleophilic or electrophilic). The organometallic peroxo complexes prepared in this work are capable of transferring an oxygen atom to phosphines, such as PhP . The reaction of 3  Cp*W(0)(r/ -0 )(CH SiMe ) with PhP in E^O solution proceeds quantitatively over the 2  2  2  3  3  course of 1-2 h as written in eq. 2. Cp*W(0)(r, -0 )(CH SiMe ) + PhP 2  2  2  3  3  •Cp*W(0) (CH SiMe ) + PhP =0 (2) 2  2  3  3  The Cp*W(0) (CH SiMe ) and PhP =0 products in the reaction are identified by both IR 2  2  3  3  and *H and P NMR spectra of the final reaction mixtures. The success of this reaction 31  suggests that oxygen atom transfer to olefins might indeed be possible with these peroxo complexes.  175  Addition of tetracyanoethylene (TCNE) to an Et20 solution of the white organometallic complex Cp*W(0)(7? -02)(CH SiMe3)2 results in an immediate color 2  2  change from colorless to orange-brown. Surprisingly, there is no evidence for a peroxometallacycle or tetracyanoethylene oxide being formed in the reaction. Instead, two different products are obtained, as outlined in eq. 3. A small quantity of white [Cp*W(O) ] 0i-O) [IR (Nujol mull): 11  2  y  2  931,883 cm ; «/ -0-W 812 (798 sh) -1  w  =  0  W  cm" ], is isolated as a minor product (=• 1 % yield based on W), and is likely formed via the 1  slow decomposition of the Cp*W(0)(r7 -02)(CH SiMe3)2 starting material. After 2  2  fractional crystallization of the reaction solution from E t 0 at -20 °C, a 2:1 adduct of the 2  organometallic peroxo complex and the olefin is obtained as an orange-brown microcrystalline solid in ~80% yield (based on W).  Cp*W(0)(f -0 )(CH SiMe3)  + (CN) C = C(CN)  2  ?  2  2  2  ^  2  [Cp*W(0)(^ -02)(CH SiMe3)] [(CN)2C = C(CN) ] 2  2  2  2  +  [Cp*W(O) ] 0'-O)  (3)  2 2  The spectroscopic properties of this orange-brown solid, specifically a low-resolution mass spectrum and an elemental analysis for carbon, hydrogen and nitrogen, are consistent with it being formulated as [C28H5206Si2W][QN4]. An IR spectrum of the compound as 2  a Nujol mull exhibits absorption bands at 2255,2218,2145 (w) attributable  I^CN'S of the  tetracyanoethylene, and an absorption band at 860 cm" assignable to I/Q-O °f 1  group. Furthermore, the i^o-0 I/Q-O 13  m  t  h  e  m  m  e  a  n  W(r? -02) 2  product is shifted somewhat from the corresponding  Cp*W(0)(r? -02)(CH SiMe3) starting material (865 cm" ). The H and 2  1  1  2  C{ H} NMR spectra of this orange-brown product are relatively uninformative as the 1  observed *H and C chemical shifts are nearly identical to those exhibited by the B  respective starting materials in eq. 3. A similar yellow-orange 2:1 adduct complex is preparable from CpW(0)(Tj -02)Me and TCNE based on elemental analysis and IR 2  spectral data.  176  An X-ray crystallographic analysis performed on a single crystal of the orangebrown product from eq. 3 confirms that it is indeed a 2:1 adduct in the solid state.  12  However, as shown in Figure 5.3, the crystal structure reveals that the 2:1 adduct exists as a charge-transfer complex, involving a molecule of tetracyanoethylene bridged between the \V(r? -02) units of two molecules of the organometallic peroxo complex. The pertinent 2  crystallographic data for the complex will be detailed elsewhere. Selected bond lengths 10  and bond angles of the molecule are presented in Table 5-lV. The most significant feature of the crystal structure involves the mode of linkage of the peroxo units to the tetracyanoethylene ligand. Specifically, the TCNE remains planar and is perpendicular to the two peroxo units. Furthermore, the TCNE molecule occupies two possible orientations which are 90° with respect to each other, as depicted in Figure 5.4 (a). The two oxygen atoms of the f?-C>2 peroxo groups, however, are not bonded symmetrically to the C=C 2  double bond, but rather, they lie skewed across the middle of the C=C double bond of the TCNE ligand, as shown in Figure 5.4 (b). A comparison of the intramolecular dimensions in the Cp*W(0)(r; -02)(CH2SiMe3) 2  fragment of the charge-transfer complex (Table 5-IV) with those of CpW(0)(rj 2  0 )(CH SiMe3) (Table 5-III), shows slight variations in the bond lengths and angles in the 2  2  W(0)(r/ -02) unit. For instance, the W=0 bond length is slightly longer in the charge2  transfer complex (1.730(7) and 1.68(3) A). Furthermore, the W-O bond lengths on average are slightly longer in the W(r; -02) unit of the charge-transfer complex. The O-O bond 2  lengths, however, are identical in both complexes. This is in agreement with the IR spectral data as only a small shift in the ^o-O value (approximately 5 cm") is observed for 1  the O-O group of the peroxo unit upon formation of the adduct complex. Similarly, the yellow-orange solid [CpW(0)(r? -02)Me] [(CN)2C=C(CN)2] can be 2  2  prepared from CpW(0)(ry -02)Me and TCNE. This yellow-orange complex possesses 2  similar physical and spectroscopic properties to those of  177  C12  Figure 5.3.  An ORTEP plot of the molecular structure of  [(r -C Me )W(0)(r? -0 )(CH SiMe )] -/.-[(CN) C=C(CN) ]. 5  7  2  5  5  2  2  3  2  2  2  178 Table 5-IV. Selected Bond Lengths (A) and Bond Angles (deg) in [Cp*W(0)(77 -0 )(CH SiMe3)] - u -[(CN) C=C(CN) ]. 2  2  2  2  2  a  2  Bond Lengths 1.730(7) 1.902(8) 1.933(7) 2.158(1) 2.081(4) 1.849(11) 1.867(12) 1.84(2) 1.83(2) 1.427(11) 1.123(14) 1.45(3) 1.47(3) 1.32(4)  W-O(l) W-0(2) W-0(3) W-C(ll) W-Cp Si-C(ll) Si-C(12) Si-C(13) Si-C(14) 0(2)-0(3) C(17)-N(l) C(17)-C(15) C(17)-C(16) C(15)-C(15') b  C(l)-C(2) C(l)-C(5) C(l)-C(6) C(2)-C(3) C(2)-C(7) C(3)-C(4) C(3)-C(8) C(4)-C(5) C(4)-C(9) C(5)-C(10) C(18)-N(2) C(18)-C(15) C(18)-C(16) C(16)-C(16')  1.424(13) 1.402(13) 1.471(14) 1.433(D) 1.487(13) 1398(11) 1504(12) 1.403(11) 1518(13) 1.489(12) 1.129(13) 1.45(3) 1.49(3) 1.22(4)  Bond Angles 0(l)-W-0(2) 0(l)-W-0(3) 0(1)-W-C(11) 0(1)-W-Cp 0(2)-W-0(3) 0(2)-W-C(ll) 0(2)-W-Cp 0(3)-W-C(ll) 0(3)-W-Cp C(ll)-W-Cp C(ll)-Si-C(12) C(ll)-Si-C(13) C(ll)-Si-C(14) C(12)-Si-C(13) C(12)-Si-C(14) C(13)-Si-C(14) W-0(2)-0(3) N(l)-C(17)-C(15) N(2)-C(18)-C(15) C(17)-C(15)-C(15') C(18)-C(15)-C(15) C(17)-C(15)-C(18) b  b  b  b  110.2(5) 106.4(4) 96.2(4) 113.3(3) 43.7(3) 81.4(4) 134.1(3) 124.5(4) 108.2(3) 107.8(3) 105.8(6) 111.4(7) 1115(8) 108.1(7) 107.0(7) 112.7(12) 693(5) 163(2) 164(2) 116(3) 118(3) 126(2)  W-0(3)-0(2) C(2)-C(l)-C(5) C(2)-C(l)-C(6) C(5)-C(l)-C(6) C(l)-C(2)-C(3) C(l)-C(2)-C(7) C(3)-C(2)-C(7) C(2)-C(3)-C(4) C(2)-C(3)-C(8) C(4)-C(3)-C(8) C(3)-C(4)-C(5) C(3)-C(4)-C(9) C(5)-C(4)-C(9) C(l)-C(5)-C(4) C(l)-C(5)-C(10) C(4)-C(5)-C(10) W-C(ll)-Si N(l)-C(17)-C(16) N(2)-C(18)-C(16) C(17)-C(16)-C(16 ) C(18)-C(16)-C(16') C(17)-C(16)-C(18)  Estimated standard deviations are provided in parentheses. Cp refers to the centroid of the (rj -C Me ) ring. 5  5  5  1  67.0(4) 107.1(8) 124.7(10) 128.1(10) 108.2(8) 126.6(9) 125.0(9) 106.6(7) 125.6(8) 127.8(9) 1093(8) 124.6(8) 125.6(8) 108.6(8) 124.7(9) 1265(9) 1175(6) 160(2) 160(2) 118(3) 117(3) 125(2)  179  (A)  Figure 5.4.  (B)  Partial views of the molecular structure of [Cp*W(0)(r/ -02)(CH SiMe )]2- ^ 2  2  3  [(CN) C = C(CN) ], (A) a view illustrating the planarity of the bridging TCNE molecule 2  2  between the two r? -0 peroxo groups and (B) a view illustrating the skewed orientation of 2  2  the f? -0 unit with respect to the C=C double bond of the TCNE molecule. 2  2  180 [Cp*W(0)(r -0 )(CH SiMe )]2[(CN) C=C(CN) ], and therefore, [CpW(0)(r/ 2  2  ?  2  2  3  2  2  0 )Me] [(CN) C = C(CN) ] and [Cp*W(0)(r -0 )(CH SiMe )] [(CN) C=C(CN) ] are 2  2  2  2  2  ?  2  2  3  2  2  2  probably isostructural. The fact that these complexes are charge-transfer compounds and are isolable as air-stable solids indicates that there is sufficient electron density on the oxygen atoms of the (r/ -0 ) unit to function as an electron donor to the tetracyanoethylene 2  2  acceptor. To the best of my knowledge these are the first examples of transition-metal peroxo complexes functioning as electron donors in charge-transfer complexes.  C.  Spectrophotometric Determination of the Formation Constant Of the Charge  Transfer Complex Between Cp*W(0)(»? -0 )(CH SiMe3) and TCNE. 2  2  2  UV-vis spectral data exhibited by the charge-transfer (CT) complex involving the CH SiMe species can be used both to determine whether it exists as a 1:1 or a 2:1 complex 2  3  in solution and to measure the formation constant of the complex. Addition of TCNE to a CH C1 solution of the organometallic peroxo complex produces pale yellow to orange2  2  brown colored solutions, eq. 4. CH C1 2  {M}(„ -0 )  +  2  2  (CN) C=C(CN) 2  2  K{M}(r -0 )] [(CN) C=C(CN) ] 2  2  ?  2  n  2  2  (4)  [{M} = Cp*W(0)(CH SiMe ), n= 1 or 2] 2  3  The charge-transfer complex exhibits a UV-vis absorption at approximately 415 nm in solution. The intensity of this absorption depends linearly on both the concentration of Cp*W(0)(r? -0 )(CH SiMe ) ([D] ) and the concentration of TCNE ([Acc] ) over the 2  2  2  3  0  0  concentration ranges studied (Table 5-V). Figure 5.5 illustrates the dependence of the intensity of the absorbance of the charge-transfer complex generated in equation 4 under two conditions: (a) where the concentration of Cp*W(0)(r/ -0 )(CH SiMe ) ([D] ) is 2  2  2  3  Q  held relatively constant at ^ 1.4 x 10" M and the concentration of TCNE ([Acc] ) is varied 2  0  181  Table 5-V. Spectrophotometric Determination of the Formation Constant of the ChargeTransfer Complex between Cp*W(0)(r -0 )(CH SiMe ) and TCNE. 2  ?  2  Concentration Cp*W(0)(r -0 )(CH SiMe ) (M) 2  ?  a  2  2  3  2  fl  3  Absorbance TCNE (M)  A  m  a  x  =  415 nm  0.0143 0.0143 0.0138 0.0139 0.0142 0.0143 0.0140 0.0142 0.0147  0.0 0.0069 0.015 0.026 0.043 0.056 0.064 0.083 0.101  0.03 0.11 0.22 0.375 0.59 0.702 0.711 0.873 1.25  0.0 0.0060 0.0081 0.0101 0.0138 0.0204 0.0290 0.0350  0.0155 0.0158 0.0150 0.0150 0.0150 0.0158 0.0155 0.0150  0.00 0.102 0.117 0.160 0.22 0.322 0.414 0.562  0.0229 0.0231 0.0086  0.023 0.024 0.0086  0.54 0.52 0.075  In methylene chloride solution at 25 °C  182  Wavelength, (nm)  A Figure 5.5.  B  UV-vis absorption spectra of the charge-transfer complex between  Cp*W(0)(r? -0 )(CH SiMe ) and TCNE in methylene chloride solutions; (A) 0.0140 M 2  2  2  3  Cp*W(0)(r? -0 )(CH SiMe ) at various concentrations of TCNE (see Experimental 2  2  2  3  Section); (B) -0.0155 M TCNE at various concentrations of Cp*W(0)(r? -0 )(CH SiMe ) 2  2  2  3  (see Experimental Section). For comparison, the absorption spectra of solutions of 0.0143 M Cp*W(0)(r/ -0 )(CH SiMe ) and 0.0155 M TCNE alone in methylene chloride solution 2  2  are indicated by (  2  3  ) and ("••) respectively.  183  from 0.00-0.101 M (Figure 5.5 (a)), and (b) where the concentration of TCNE is held relatively constant at =• 1.55 x 10" M and the concentration of Cp*W(0)(fj 2  2  0 )(CH SiMe ) is varied from 0.00-0.035 M, (Figure 5.5 (b)). Plots of absorbance vs 2  2  3  concentration for both cases produce straight lines over the entire concentration ranges used in this experiment. At no point does the absorbance at 415 run due to the CTcomplex deviate from linearity when the concentration of one of the two reagents is held constant and the concentration of the other is varied to large excess. In other words, the absorption continues to increase in a linear fashion, the limiting factor being the solubility of either of the two reagents in the CH C1 solution. 2  2  The UV-vis spectral data obtained (specifically, the absorbance of the CT-complex at 415 nm at various concentrations of the two reagents in eq. 4) are listed in Table 5-V. The data have been analyzed for a 1:1 CT-complex in solution using the method developed by Benesi and Hildebrand. The dependence of the absorbance of the charge-transfer 14  band (ACJT) on the concentration of a 1:1 CT-complex in solution is given by [D]  • A  1  0  1  1  =  CT  (5)  + K  £  CT  [ACC]  erjr  0  under conditions in which the concentration of the electron acceptor [Acc] is in large 0  excess relative to the concentration of the donor [D] . The formation constant for eq. 4 is 0  K (in units of M" ) and ecr is the extinction coefficient at the monitoring wavelength (in 1  units rvi^cm" ). The Benesi-Hildebrand plot, [ D ] / A C T vs [Acc] , is shown in Figure 5.6, 1  _1  0  0  which contains all the data from Table 5-V. It should be noted that plotting only the data where [TCNE] >> [C^*W(0)(r7 -C^)(CH SiMe )] also produces a straight line plot. The 2  2  3  spectrophotometric data, therefore, exhibits Benesi-Hildebrand behavior and suggests a 1:1 charge-transfer complex between Cp*W(0)(rj -02)(CH SiMe ) and TCNE in solution. 2  15  2  3  The extinction coefficient (e ) at 415 nm for the 1:1 charge-transfer complexes is CT  approximately 209 M^cm" , which is calculable from the y-intercept (Figure 5.6). It should 1  be noted that  184  0.15 0.14  H  0.13 0.12  -  0.11  -  0.10  -  0.09 0.08  -  0.07  -  0.06  -  0.05  -  0.04  -  0.03  -  0.02  -  0.01  -  0.00  ~1— 120  —r~ 140  [TCNE]- , (M" ) 1  1  Figure 5.6. Spectrophotometric determination of the formation constant of the chargetransfer complex between Cp*W(0)(n -C>2)(CH2SiMe3) and TCNE in methylene chloride 2  at 25 °C according to the Benesi-Hildebrand method (eq. 5), a plot of [Cp*W(0)(n -02)(CH SiMe3)]/Absorbance versus [TCNE]- . 2  1  2  185 there is considerable uncertainty in the extrapolated extinction coefficient from the BenesiHildebrand y-intercept. A value of 5.3(3) M" for the formation constant K of equation 4 is 1  obtained from the slope of the plot and e  CT  in Figure 5.6. The magnitude of K indicates  that the CT-complex between Cp*W(0)(f? -0 )(CH SiMe ) and TCNE is moderately weak 2  2  2  3  in solution. The formation constant and extinction coefficient of the CT-complex 16  between Cp e  CT  W(0)(r7  -O^CH^iMe-j) and TCNE are very similar to the values of K and  observed for the electron donor-acceptor complexes between various diazoalkanes [e.g.  diazabicyclooctene (DBO) and diazabicycloheptene (DBH)] and T C N E . Furthermore, 17  the charge-transfer complex between DBO and TCNE exists as a 2:1 complex in the solid state. In acetonitrile solution, however, DBO and TCNE apparently form a weak 1:1 CTcomplex which has similarly been evaluated by the Benesi-Hildebrand method.  17  D.  Future Studies It is possible that the charge-transfer complex [Cp*W(0)(77 -0 )(CH SiMe )] -At2  2  2  3  2  [(CN) C=C(CN) ] may be an isolable intermediate on the way to epoxidation of the olefin 2  2  TCNE. This would suggest that some electron transfer occurs before a peroxometallacyclic intermediate is formed in the mechanisms depicted in Scheme 5-1. Obviously, some investigations of the chemical properties of this charge-transfer complex are required with two objectives in mind. Can the CT-complex be induced to form the peroxometallacycle of which there are none known for complexes containing transition-metals from Groups 4 to 6? Furthermore, both thermolysis and photolysis may induce this reaction to occur. Can tetracyanoethylene oxide be confirmed as the eventual product in these reactions? Thermolysis of [Cp*W(0)(r -0 )(CH SiMe )] -M-[(CN) C=C(CN) ] as a solid or 2  ?  2  2  3  2  2  2  in solution produces purple to black products which are completely soluble in CH Ci2 2  solutions, and the identities of these product(s) are currently under investigation. Very recently, a report by Kamata and Miyashi suggests that photolysis may be the 18  method most appropriate to induce the oxygen-atom transfer. Their work shows that  186 mixing (Ar' C) S compounds with TCNE in CH C1 results in the formation of a charge2  2  2  2  transfer complex between the two reagents, as depicted below in Scheme 5.2. Photolysis using radiation of wavelength near the charge transfer band of the CT-complex results in the coupling of the two reagents, and the tetrahydrothiophene is formed.  A y /  \  '  P h  +TCNE  (1)  TCNE "  8+  Ar'  Ph  CH a 2  8  2  (1)  hv,CT  (1) - TCNE" +  CN^/ I^CN CN CN  Scheme 5-2. The system is an organic analogue to the system studied in this work, the thiirane being analogous to the tungsten peroxo complex. The goal then would be to irradiate CH CL 2  solutions of Cp*W(0)(r7 -0 )(CH SiMe ) and (CN) C=C(CN) at 415 nm to see if the 2  2  2  3  2  2  peroxometallacycle can be formed or if tetracyanoethylene oxide is produced, as depicted in Scheme 5-3.  Scheme 5-3.  187 Alternatively, if these studies are not successful, investigations into the utility of the dioxo alkyl complexes, Cp'M(0) R, as catalysts for the epoxidation of olefins should be 2  undertaken, i.e. eq. 6.  (6)  It is possible that the active catalyst is a peroxo alkyl species, i.e.  OH and that a metal oxo peroxo compound, a number of examples of which have been isolated in this work, may form if no olefin is present.  Summary Reactions of representative examples of the dioxo alkyl complexes of molybdenum and tungsten, Cp'M(0) R, with 30% H 0 (aq) in EtjO solution afford novel 2  2  2  organometallic peroxo alkyl complexes Cp'M(0)(f? -0 )R, several of which have been 2  2  isolated and characterized in this work. The formulation of these complexes as monomeric 16-electron entities having a piano-stool molecular structure has been confirmed by a single-crystal X-ray crystallographic analysis of CpW(0)(n -0 )(CH SiMe ). Studies of the 2  2  2  3  chemistry of Cp*W(0)(»? -0 )(CH SiMe3) as a prototypal member of this series of 2  2  2  compounds have shown that transfer of one of the oxygen atoms to phosphines such as PPh is possible, the parent dioxo alkyl complex Cp*W(0) (CH SiMe ) being regenerated 3  2  2  3  in the process. Reaction with TCNE results in the formation of a 2:1 charge-transfer complex [Cp*W(0)(n -0 )(CH SiMe )] -^-[(CN) C=C(CN) ] whose solid-state 2  2  2  3  2  2  2  188 molecular structure has been determined. In solution however, the charge-transfer species between Cp*W(0)(r -0 )(CH SiMe ) and TCNE exhibits Benesi-Hildebrand behavior 2  ?  2  2  3  and is therefore formulated as a 1:1 adduct. Quantitative evaluation of the formation constant by the Benesi-Hildebrand method indicates that the charge-transfer complex is moderately weak in CH C1 solutions. Consequently, investigations to determine whether 2  2  these charge-transfer complexes are indeed isolable intermediates en route to the epoxidation of TCNE are currently in progress. Other aspects of the characteristic chemistry of the alkyl peroxo complexes, Cp'M(0)(n -0 )R, will also be the focus of future 2  2  work in these laboratories.  189  References and Notes:  (1)  Taken in part from: (a) Legzdins, P.; Phillips, E.C.; Rettig, SJ.; Sanchez, L.; Trotter, J.; Yee, V.C. Organometallics 1988, 7,1877. (b) Legzdins, P.; Phillips, E.C.; Sanchez, L. Organometallics 1989,5,930. (c) Legzdins, P.; Phillips, E.C; Trotter, J.; Yee, V.C. manuscript in preparation.  (2)  Sheldon, R.A.; Kochi, J.K. Metal-Catalyzed Oxidation of Organic Compounds; Academic Press: New York, 1981.  (3)  Mimoun, H. 'Transition-metal peroxides - synthesis and use as oxidizing agents",  Chapter 15; The Chemistry of Functional Groups, Peroxides; S. Patai (Ed.); John Wiley & Sons Ltd: 1983. (4)  J0rgensen, K.A. Chem. Rev. 1989, 431, and references therein.  (5)  Faller, J.W.; Ma, Y. Organometallics 1988, 7, 558.  (6)  (a) van Asselt, A.; Trimmer, M.S.; Herding, L.M.; Bercaw, J.E. /. Am. Chem. Soc 1988,270,8254. (b) Drouin, M.; Harrod, J.F. Can. J Chem. 1985,63, 353. (c) Mason, M.G.; Ibers, J.A. /. Am. Chem. Soc. 1982,104,5153. (d) van der Ent, A.; Onderdelinden, AX. Inorg. Chim. Acta 1973, 7, 203.  (7)  Antinolo, A.; Manotti Lanfredi, A.M.; Martinez de Ilarduya, J.; Otero, A.; Royo, P.; Tripicchio, A. /. Chem. Soc, Dalton Trans. 1988, 2685.  (8)  These absorption bands were compared to those observed in an IR spectrum of authentic samples of Cp*W(0)2(CH SiMe3) and Ph P=0 as a Nujol mulls. 2  (9)  3  The assignments are confirmed by comparing the observed chemical shifts with literature values for Ph P=0 (-26.0 ppm) and Ph P ( + 6.0 ppm) with respect to 85% 3  3  H3PO4 (see reference 10).  (10)  Denney, D.B.; Denney, D.Z.; Wilson, L.A. Tet. Lett. 1968, 85.  (11)  Faller, J.W.; Ma, Y. /. Organomet. Chem. 1988,340,59.  190 (12)  Crystals of CpW(0)(r -0 )(CH SiMe ) are orthorhombic, a = 6.531(6) A, b = 2  ?  11.659(2) (13)  2  2  3  A, c = 32.198(7) A, and D  = 2.08 g/cm , Z = 8, space group Pbca. 3  c  Crystals of [Cp*W(0)(r7 -0 )(CH SiMe )] - / . - [ ( C N ^ C ^ C N ^ are monoclinic, 2  2  2  3  2  a = 19.3259(17) A, b = 9.8313(13) A, c = 23.485(6) A, p = 107.249(11) (deg), and D = 1.62 g/cm , Z = 8, space group C2/c. 3  c  (14)  Benesi, H.A.; Hildebrand, J.H. /. Am. Chem. Soc. 1949, 71,2703.  (15)  It should be noted that the UV-vis spectral data have not been analyzed for a 2:1 CT-complex in solution, and therefore, a CT-complex of this stoichiometry may indeed exist in solution.  (16)  Blackstock, S.,C; Kochi, J.K. /. Am. Chem. Soc. 1987,109, 2484.  (17)  Person, W.B.; Mulliken, R.S. Molecular Complexes, A Lecture and Reprint Volume Wiley : New York, 1969.  (18)  Kamata, M.; Miyashi, T. /. Chem. Soc, Chem Commun. 1989,557.  Chapter 6  Reactions of the Dioxo Alkyl Complexes, Cp'M(0) R, with />-Tolyl Isocyanate 2  192 Introduction The study of transition-metal imido complexes (L M=NR, L = ligands, R = alkyl n  n  or aryl) and other compounds with metal-ligand multiple bonds, i.e. metal-oxo, -sulfido, alkylidene, -alkylidyne and related compounds, is currently in a phase of extremely rapid growth. ' These complexes are of interest as new reagents or catalysts for organic 1 2  synthesis. They also serve as model complexes for the active sites in such heterogeneous oxidation processes as the ammoxidation of propylene to acrylonitrile and the catalytic 3  carbonylation of organoazides and nitroaryl compounds. Furthermore, metal-imido 4  species appear to be involved in the catalytic cycles of a variety of metallo-enzymes.  5  In previous chapters of this thesis, the synthesis, characterization and reactivity of metal-oxo and -alkylidene complexes, specifically, those of the type 1 and 2 depicted below, have been discussed. These complexes have been shown to exhibit varied and interesting chemistry. It is, therefore, of interest to determine whether the analogous metal-imido complexes (type 3) can be prepared so that their chemistry can be studied.  R 1  R  R  2  3  The most common method for the synthesis of metal-imido complexes involves the reaction of alkyl- or arylisocyanates with metal-oxo complexes, a transformation represented in a general fashion in eq. I.  6  R  + RN=C=0  • L*M^ > = 0 (l)  •  L M=NR + n  C0  2  193  Reactions of this type are believed to proceed via a metallacyclo-carbamate intermediate, a few examples of which have recently been isolated and structurally characterized. This 6,7  chapter presents the results of my preliminary investigations to see whether the dioxo alkyl complexes, Cp*M(0)2R, discussed in Chapter 2 exhibit reactivity similar to that depicted in eq. 1. Specifically, this chapter describes the reactions of a few of the Cp*M(0) R 2  complexes with /Molyl isocyanate. It is interesting that, so far, no oxo imido complexes of the type Cp*M(0)(=NC6H4-/j-Me)R have been isolated from these reactions. Rather, complexes which are thought to derive from the corresponding bis(imido) complexes Cp M(=NQH^-Me^R, are isolated, i.e.  ^* s > —  The spectroscopic properties of these complexes and the results of a single crystal X-ray crystallographic analysis of C^V(=NQH4-/»-Me)[N(C H4-/7-Me)C(0)N(QH -/?6  4  Me)](CH SiMe3) are presented. Furthermore, reactions of the Cp*W(0) R complexes 2  2  (R= CH SiMe or CH ) with /?-MeC6H N = C=0 were monitored by H and C NMR A  2  3  3  B  4  spectroscopy. The results of these studies are presented with a view to deciphering a plausible mechanism for the formation of the imido product complexes. Experimental Section All reactions and subsequent manipulations involving organometallic reagents were performed under anaerobic and anhydrous conditions. The general procedures employed in this study were as described in the Experimental Sections of the previous chapters. The /?-tolyl isocyanate reagent (p-MeC6H4N=C=0), was obtained from Aldrich, and was not further purified before use. The cyclopentadienyl dioxo alkyl complexes, Cp'M(0) R 2  [Cp'= Cp and Cp*, M= Mo and W, R= CH3 and CH SiMe3], used as reagents in this work 2  were prepared by the' methods outlined in Chapter 2 of this thesis.  Preparation of Cp W( = NC H -p-Me) [-N(C H -p-Me)C(0)N(C H -p6  Me)] (CH SiMe3>, 3a.  4  6  4  6  4  To a stirred, clear colorless solution of Cp*W(0) (CH SiMe )  2  2  2  3  (0.331 g, 0.75 mmol) in toluene (20 mL) was added a 10-fold excess ofp-MeC H N = C=0 6  4  (1.0 mL, 7.9 mmol) using a 1 mL glass syringe. After the addition was complete, the reaction mixture was stirred at room temperature. The mixture changed to a yellow color in a few minutes and then to an orange color after 3 h. The reaction flask was placed in an oil bath, and the toluene solution was refluxed overnight, whereupon the solution became dark red. The reaction mixture was cooled to room temperature, and the toluene solvent was removed in vacuo. The resulting orange-red solid was washed with cold hexanes (1 x 10 mL, -20 °C) and cold Etp (2 x 5 mL, -20 °C). The solid was then dried in vacuo. An IR spectrum of this solid material as a Nujol mull was devoid of any W=0 stretching vibrations in the spectral region 950-850 cm . The orange-red solid was recrystallized -1  from a 1:1 CH Cl /hexanes solution mixture at -25 °C to obtain 0.145 g of red crystals of 2  2  Cp*W(=NC H -^-Me)[-N(C H -/?-Me)C(0)N(C H -/?-Me)](CH SiMe ), 3a, (26 % yield). 6  4  6  4  6  4  2  3  Anal. Calcd for <^H N OSiW: C, 57.67; H, 6.32; N, 5.60. Found: C, 57.66; H, 47  3  6.29; N, 5.75. IR (KBr pellet): v cm ; J ^ -1  1663 (s) cm" ; v  1  Q=c  si  1  1653 (s) cm' ; also 1505 (s) and  1  2  2  1  Q=Q  1304 (vs) cm' . rl NMR (C D ) 6 7.99 (d, 2 / ^ = 8.6 Hz), 7.43 (d, J 1  x  6  mi  8.4 Hz), 6.99 (d, J =  8.6 Hz), 6.91 (d, /  2  mi  = 8.4 Hz), 7.13  2  6  2  1242 (m)  Cp  1300 (vs), 814 (m) and 798 (m) cm' . IR (CH C1 ): v  (d, J =  Me  1135 (m) cm' ; « » 850, 825 (m) cm' ; also 1504 (s), 1448 (m), 1344 (m), 1  w = n  1609 (m) cm- ;j/ .  1  Q=Q  2  HH  = 8.5 Hz), 6.80 (d, J 2  H H  WH  = 8.5 Hz)  (12 H, 3 C // -CH ), 2.25,2,19 (s, 2 x (3H), C(0)[NC H -Ctf ] ), 1.96 (s, 3H, =NC H 6  4  3  6  4  3  2  CH ), 1.75 (s, 15H, C (Ctf ) ), 1.48 (d, IH, C// H SiMe , / - H b 2  3  5  IH, CH /f SiMe , / - H b 2  a  b  3  3  5  =  1  a  1  3  8 Hz  H a  )> " ° - & 16  9 H  b  3  6  =  1  1  3  8H z  H a  ' (Cff ) ). si  3  3  13  )' ° -  4  ('  1 9 5  d  C{ H} NMR (C D ) 6 1  6  6  164.8 (s, C(0)(NC H -CH ) ), 153.7 (s, =N-C ipso), 144.1,142.5 (s, 2 x C(0)[-N-C ipso), 6  4  3  2  137.5,131.5,129.3,129.2 (s, ipso C's, C H -CH ), 129.1,129.0,128.9,127.6,123.5 (s, C-H of 6  4  3  C H ), 117.0 (s, C (CH ) ), 40.39 (s, C H S i M e , / = 60.5 Hz), 21.15 (=NC H -CH ), 1  6  4  5  3  5  2  3  wc  6  20.86, 20.78 (s, C(Q)[NC H -CH ] ), 11.5 (s, C (CH ) ), 3.29 (s, Si(CH ) ). Low6  4  3  2  5  3  5  3  3  4  3  195 resolution mass spectrum (probe temperature 120 °C) m/z 749 [P] , 616 (P-[p+  MeC H NCO]) . +  6  4  Reactions of the Cp*W(0) R Complexes [R= C H or CH^iMej] with n 2  3  equivalents of /?-MeC H N=C=O, [n = 1,2 or 3]. 6  All these reactions were performed in  4  a similar manner. A microsyringe was used to add />-MeC H N=C=0 to deuterated 6  4  benzene solutions of the appropriate dioxo alkyl complex, Cp*W(0) R (R= CH , or 2  3  CH SiMe ), in ratios of Cp*W(0) R : /7-MeC H N = C=0 equal to 1:1,1:2, and 1:3. The 2  3  2  6  4  resulting reactions were monitored by H and C NMR spectroscopy over a 4-month X  B  period. The H and ^C^H} NMR data for the dioxo alkyl reagents in C D solutions, are X  6  fi  listed in Table 2-IV. NMR data for />-MeC H N = C = 0 : % NMR (C D ) S 6.68 (d, 2H, C # -Me), 6  4  6  6  6  4  6.56 (d, 2H, C H -Me), 1.94 (s, 3H, C H -C// ); C{ H} NMR (C D ) 6 135.3 (s, ipso B  6  4  6  4  1  3  6  6  carbon of C H ), 130.2,124.6 (s, C H 's), 20.7 (s, C H -CH ). 6  4  6  4  6  4  3  (a) R= CH SiMe and n = 1. To a solution of Cp*W(0) (CH SiMe ) (0.07 g, 2  3  2  2  3  0.16 mmol) in C D (1 mL) was added/>-MeC H N = C=0 (22uL, 0.17 mmol) using a 6  6  6  4  microsyringe. The resulting solution was then transferred to an NMR tube, and the tube was flame-sealed under a slight vacuum. Initial H and ^C^H} NMR spectra of the 1  reaction solution displayed only proton and carbon signals due to the two starting materials. The reaction tube was placed in an oil bath and heated to 55 °C. After four months the reaction mixture was orange-red. The proton and carbon resonances due pMeC H N=C=0 diminished completely, but signals due to Cp*W(0) (CH SiMe ) 6  4  2  2  3  remained. Two product complexes were produced in a ratio of 2:1. The major product was identified as compound 3a by comparison of the H and ^C^H} NMR spectra to those of J  the authentic sample (vide supra). The other resonances in the spectra of thefinalreaction  196 mixture were assigned to Cp*W(=NC H -/?-Me) (CH SiMe ), 2a, based on proton 6  4  2  2  3  integration. NMR data for Cp*W(=NC H - p-Me)[-N(C H -/7-Me)C(0)N(C H -/76  4 j  6  4  6  4  Me)](CH SiMe ), 3a, was the same as presented in (a) above. 2  3  NMRdataforCp*W(=NC H -/?-Me) (CH SiMe ),2a: 6  4  2  2  H NMR (C D ) 5 7.93-  1  3  6  6  6.81 (d, 12H, 3 x C /f -Me), 2.14 (s, 6H, 2 x = NC H -C// ), 1.82 (s, 15H, C (C/f ) ), 0.63 6  4  6  4  3  5  3  5  (s, 2H, / _ = 8.4 Hz, CrY SiMe ), 0.25 (s, 9H, Si(Cr7 ) ). ^C^H} NMR (C D ) 6 2  w  H  2  3  3 3  6  6  155.7 (s, =N-Cipso), 131.9,130.4 (s, ipso C's, C H -CH ), 130.1,128.9,124.6,124.3,123.9 6  4  3  (s, C-H of C H ), 113.77 (s, C (CH ) ), 19.78 (s, W-CH Si), 20.97 (s, =NC H -CH ), 10.75 6  4  5  3  5  2  6  4  3  (s, C (CH ) ), 2.64 (s, Si(CH ) ). 5  3  5  3  3  (b) R= CH SiMe and n = 3. This reaction was performed in a manner 2  3  analogous to that described in (a) above. A C D (1 mL) solution of Cp*W(0) CH SiMe 6  6  2  2  3  (0.06 g, 0.15 mmol) and p-MeC H N=C=0 (60 ^L, 0.47 mmol) was cannulated into an J  6  4  NMR tube and the tube was flame-sealed under a slight vacuum. The NMR tube was placed in an oil bath and heated to 55 °C. Over the course of four months, the reaction mixture turned deep red. Monitoring the reaction by *H and a  NMR spectroscopy »i  revealed the slow conversion of the starting dioxo alkyl complex into Cp W(=NC H -p6  4  Me)[-N(C H -/»-Me)C(0)N(C H - p-Me)](CH SiMe ), 3a. No proton or carbon 6  4  6  4  j  2  3  resonances were observed that could be assigned to either Cp*W(0)(=NC H -/> 6  4  Me)(CH SiMe ), la, or Cp*W(=NC H -/?-Me) (CH SiMe ), 2a, as possible intermediates. 2  3  6  4  2  2  3  NMR data for Cp*W(=NC H -/7-Me)[-N(C H -/;-Me)C(0)N(C H -/?6  4  6  4  6  Me)](CH SiMe ), 3a: H NMR (C D ) 6 7.93 (d, 2H , / X  2  2 H  b ' Ha-Hb= 2/  8  6  - H )' 6  2  6.91 (s, 2H , / . H b = 2  a  H a  8  7  H a  -Hb  1  A  2  3  3  9  6  a  (> a> V - H b = s  2H  - Hz). ^ 5  6  -  8  1  (> d  2 H  -  8  and  5  0  b ' Ha-Hb= - ) 2/  8  5m  =  8  (> d  2 H  6 H z  4  >» ^  6  -  9 8  b ' Ha-Hb= 2/  8  ( ' d  5  [assignable to 12H of 3  x C /f -Me], 2.24, 2.20 (s, 2 x 3H, C(0)[NC H -CrY ] ), 1.98 (s, 3H, =NC H -C/7 ), 1.81 (s, 6  4  6  15H, C (CH ) ), 1.45 (d, IH, / - H b = 2  5  3  5  H a  1  1  2 H  4  z  3  2  6  4  3  ' C# H SiMe ), 0.20 (d, IH, / . H b = 2  a  b  3  H a  1  1  2  197  Hz, CHaflbSiMes), -0.18 (s, 9H, Si(Ci/ )3). C{ H} NMR (QD ) 6 164.8 (s, 13  1  6  3  C(0)(NC H4-CH )2), 153.65 (s, =N-Cipso), 144.1,142.45 (s, 2 x C(0)[-N-C ipso]), 137.5, 6  3  131.35 (s, ipso C H4-CH ), 129.0,128.8,128.2,127.6,123.5 (s, C-H o f Q H ^ , 117.14 (s, 6  3  C (CH ) ), 40.2 (s, ^ . 0 = 60.5 Hz, W-CH Si), 21.18 (s, ^NQaj-CH^, 20.87,20.80 (s, 5  3  5  2  C(0)[NC H -CH ] ), 11.4 (s, C5(CH ) ), 3.30 (s, S^CH^). 6  4  3  2  3  (c) R= C H and n = 2.  5  This reaction was performed in a manner similar to that  3  described in (a) above. To a clear colorless solution of Cp*W(0) Me (0.67 g, 0.18 mmol) 2  in Q D (1 mL) was added /^-MeQrL^N=C=O (46 uL, 0.37 mmol) using a microsyringe. 6  The resulting reaction mixture was cannulated into an NMR tube, and the tube was then flame-sealed under a slight vacuum. Initial *H and C{ H} NMR spectra of the solution 13  1  displayed only proton and carbon signals due to the two starting materials (see above). The sample tube was then placed in an oil bath and heated to 55 °C. Over the next four months the reaction solution turned from colorless to orange-red, and the signals due to the starting materials decreased in intensity as new resonances grew in. The resonances , I  that grew in were assigned to two product complexes in a ratio of 4:1, Cp W(=NC6H4-p; , Me)[-N(C H4-p-Me)C(0)N(C H -/7-Me)](CH ), 3b, as the major product and 6  6  4  3  Cp*W(0)(=NQH4-p-Me)(CH ), lb, as the minor product. 3  I  1  NMR data for Cp W(=NC H4-/j-Me)[-N(QH4- p-Me)C(0)N(C H -^-Me)](CH ), 6  j  6  4  3  3b: *H NMR (QDg) 6 7.90 (d, 2H, / -H= 8.3 Hz, Q/^-Me), 7.08 (s, 4H, Cgfy-Me), 2  H  7.06 (d, 2H, / -H= 8.3 Hz, Qfy-Me), 6.90 (d, 2H, / -H= 8.5 Hz, Q/^-Me), 6.84 (d, 2H, 2  2  H  2  H  8.5 Hz, Cg/^-Me), 2.20 (s, 6H, C(0)[NC6H -C/Y ] ), 2.02 (s, 3H, ^ N Q ^ - C i ^ ) ,  JH-H=  4  3  2  1.71 (s, 15H, C5(Ctf ) ), 1.04 (s, 3H, W-Ctf ). ^C^H} NMR (QD ) 5 166.15 (s, 3  5  3  6  C(0)(NC H -CH ) ), 153.7 (s, =N-C ipso), 143.8,143.2 (s, 2 x C(0)[-N-C ipso]), 136.6, 6  4  3  2  133.1 (s, ipso Cs of QH4-CH3), 129.0,128.9,128.6,124.8,125.0, (s, C-H of Q H 4 ) , 116.3 (s, C (CH ) ), 35.28 (s, ^ . 0 = 58.8 Hz, W-CH ,), 21.1 (s, =NQH4-CH ), 20.98,20.89 (s, 5  3  5  3  C(0)[NQH4-CH ] ), 11.1 (s, C5(CH ) ). 3  2  3  5  3  198 NMR data for Cp*W(0)(=NC H - p-Me)(CH ), lb: H NMR (C D ) s 7.83 (d, X  6  4  J  3  6  6  1H, / . = 8.45 Hz, C # -Me), 7.11 (s, 1H, C # -Me), 6.98 (d, 1H, / . = 8.4 Hz, C # 2  2  H  H  6  4  6  4  H  H  6  4  Me), 6.81 (d, 1H, / . = 8.5 Hz, C # -Me), 2.16 (s, 3H, =NC H -Cff ), 1.63 (s, 15H, 2  H  H  6  4  6  4  3  C (C// ) ), 1.34 (s, 3H, W-C/f ). ^C^H} NMR (C D ) S 142.06,141.0 (s, =N-C ipso), 5  3  5  3  6  6  134.1,131.5,128.8,126.3, (s, C-H of C H ), 118.3 (s, C (CH ) ), 41.5 (s, W-CH ,), 21.0 (s, 6  4  5  3  5  3  =NC H -CH ), 10.93 (s, C (CH ) ). 6  4  3  5  3  5  Preparation of Cp*Mo(=NC H -p-Me) [-N(C H -p-Me)C(0)N(C H -p-Me)] (CH ), 6  4  6  4  6  4  3  3c. A toluene solution (25 mL) of Cp*Mo(0) Me (0.071 g, 0.25 mmol) and p2  MeC H N=C=0 (0.16 mL, 1.276 mmol) was refluxed for 24 h, after which time the 6  4  solution was red. The solvent was removed in vacuo to give an orange solid which was washed with hexanes (2 x 10 mL) and Et20 (1 x 10 mL). An IR spectrum of this orange solid as a Nujol mull was devoid of any Mo = 0 stretching vibrations in the spectral region 930-870 cm* which could be attributed to the Cp*Mo(0) Me starting material. 1  2  Recrystallization of the solid material from 1:1 CH Cl /hexanes at -20 °C over 1 week gave 2  2  0.045 g of Cp*Mo(=NC H -/j-Me)[-N(C H -p-Me)C(0)N(C H -/;-Me)](CH ), 3c, as deep 6  4  6  4  6  4  3  red crystals (30% yield). Anal. Calcd for C ^ H ^ O M o : C, 67.22; H, 6.67; N, 7.13. Found: C, 63.20; H, 6.03; N, 6.65. IR (KBr pellet): v 8  1111 (m) cm' ; 6 1  Cp  1661 (vs) cm" ; z/ _ 1630, 1607 (m) cm' ; * 1  c  =  Q  1  ph  H  M o = N  1178,  820 (s) cm' ; also 1504 (vs), 1454 (m), 1379 (m), 1288 (vs), 1024 (m), 1  *  935 (m), 727 (m), 521 (m) and 449 (m) cm" . H NMR (CD C1 ) 6 8.16 (d, 2H, / 1  X  2  2  2  H H  =  8.25 Hz, C // -CH ), 7.26-7.08 (m, 10 H, C # -CH ), 2.38 (s, 3H, =NC H -Ctf ), 2.28 (s, 6  4  3  6  4  3  6  4  3  2x (3H), C(0)[NC H -C/Y ] ), 1.64 (s, 15H, C (C// ) ), 1.50 (s, 3H, Mo-C# ). 6  4  3  2  5  3  5  3  199  X-ray Crystallographic Analysis of Cp*W(=NC H4-p-Me)[-N(C H -pi 6  Me)C(0)N(QH4-/>-Me)](CH SiMe3), (3a). 2  6  4  A suitable single crystal of this compound  was obtained by crystallization from a 1:1 CH^C^/hexanes solution at -25 °C. The X-ray structure determination was performed by Ms. V.C. Yee and Dr. J. Trotter of this Department. The pertinent crystallographic data, refinement details and final positional and equivalent isotropic thermal parameters (£/  eq  = 1/3 x trace diagonalized U) for the  complex will be reported elsewhere. Bond lengths and bond angles for the complex are 9  listed in Table 6-1, and views of the solid-state molecular structure are given in Figure 6.2,  200 Results and Discussion  A.  Synthesis and Some Physical Properties of the CpTVl( = NC H -/?-Me) [-N(C H -/>6  4  6  4  Me)C(0)N(C H -/>-Me)]R Complexes [M = W, R = CH SiMe3 (3a), C H (3b); M = Mo, 6  4  2  3  R = CH (3c)]. 3  Reactions of representative members of the 16-electron dioxo alkyl complexes, Cp*M(0) R, with /7-tolyl isocyanate (/?-MeC H N = C = 0) result in their conversion to the 2  6  4  corresponding Cp*M(=NC H -p-Me)[-N(C H -^-Me)C(0)N(C H p-Me)]R compounds 6  4  6  4  6  4:  as summarized in eq. 2, where R'= /?-MeC H , and for M= W, R= CH SiMe (3a), or 6  4  2  3  C H (3b), and for M = Mo, R = C H (3c). 3  3  + 2 C0  2  (2)  Conversions 2 are most conveniently performed in refluxing toluene over a period of 12-24 h. In this manner, 3a and 3c are isolable as air-stable, orange-red crystalline solids in moderate yields. Alternatively, complexes 3a and 3b can be generated by heating NMR tubes containing benzene-^ solutions of the appropriate Cp*W(0) R starting material and 2  />-MeC H N=C=0 over the course of 2-4 months. Reactions 2 are much slower when 6  4  performed in NMR tubes, presumably due to minimal agitation of the reaction mixture and retardation of the reaction by the build-up of the C 0 by-product in the NMR tube. 2  Complexes 3a and 3c are much less soluble in common organic solvents than are their dioxo alkyl precursors. For instance, they are insoluble in hexanes and E^O, but, they are quite soluble in benzene, toluene and CH C1 . The IR and mass spectral 2  2  201  properties of the new imido complexes are consistent with their possessing the solid-state molecular structures shown in eq. 2, a fact that has been confirmed by an X-ray crystallographic analysis of compound 3a (see below). The Nujol mull IR spectra of 3a and 3c exhibit a strong absorption band in the region 1665-1660 cm which is assignable to a -1  VQ Q. =  Other bands observed in the IR spectra appear in the regions 1505-1500, 1450-1445,  1300-1288,1170-1110, and 815-795 cm , but cannot be assigned with certainty. -1  Comparison of the Nujol mull spectra of these complexes with other tungsten alkyl- and aryl-imido complexes reported in the literature leads to the conclusion that the bands at 1170-1110 cm and 815-795 cm are attributable to »>w=N ^ d •'w-N > respectively. The -1  -1  parent ion is observable in the low-resolution mass spectrum of 3a, the most abundant daughter ion at m/z 616 being attributable to [P-^-MeCgHjNCO)]" ", which corresponds to 1  [Cp* W(=NC H -^-Me)2(CH SiMe3)] . +  6  4  2  The solution IR spectrum of 3a and the H and ^ C NMR spectra of the imido 1  complexes 3a, 3b and 3c suggest that their molecular structures are maintained in solution. For instance, the IR spectrum of 3a in CH Cl2 solution exhibits an absorption band in the 2  region 1653 cm due to the VQ=0- Furthermore, the *H and C NMR spectra of these -1  W  complexes exhibit resonances due to two magnetically different /7-MeQHtN-environments. To illustrate this, the H and C NMR spectra of Cp* W(=NQH -p-Me) 1  D  4  [-N(QH -/>-Me)C(0)N(QH -/>-Me)](CH SiMe ) (3a) are shown as representative 4  4  2  3  examples in Figure 6.1. The phenyl protons exhibit three sets of AB patterns in the spectral region 8.0-6.8 ppm. Two methyl resonances, in a ratio of 2:1, are observed in the i  spectral regions 2.3-2.1 and 2.0-1.9 ppm, which correspond to the two W-N(QH -/?4  Me)C(0)N(QH -p-Me) and W^NCgH^-p-Me environments, respectively. All the 4  complexes exhibit a resonance in the ^ C NMR spectral region 168-163 ppm which is attributable to the carbon of the carbonyl group of the metallacyclourea.  i Interestingly, the =NC H -/?-Me and -N(C H -p-Me)C(0)N(C H -/?-Me) groups 6  4  6  4  6  4  have a more dramatic effect on the H and ^ C NMR chemical shifts of the methyl and 1  202  (a)  JLJLWI T~r-i--r-!-f-T i i i  i  i  r  j i 'i  T  i"i  T  T  " ' " ' J  *" "*  1  1  I t-n-r^"i-|-T i T T H ' T j - n n  | ii  T  T  J r i vi-'| r n  " ~J" "" " L  F  R  1  ''  1 ' ' '  r  1~  r  (b)  I 22.9  I "  I  23.0  I  I  rn I  21.9  I I I I  21.0  I  20.5  I M I  I  20.0  I I  ri I  10.9  I I I I  10.0  I  I I I I  I  I I  II  W  i 160  lio  ISO  100  80  60  40  20  Figure 6.1. (a) 300 MHz *H and (b) 75 MHz ^C^H} NMR spectra of Cp*W(=NC H -/)-Me)[-N(C H -/;-Me)C(0)N(C H -/7-Me)](CH SiMe3) 6  4  6  4  (3a) in CgDg solution at ambient temperatures.  6  4  2  0  PPM  203 methylene groups (M-CH or M-CH SiMe ) in the imido metallacyclourea complexes than 3  2  3  do the oxo ligands in the Cp*M(0) R complexes. The methylene protons in complex 3a 2  are inequivalent and give rise to an AB pattern in the H NMR spectrum, Figure 6.1(a). 1  Both the methyl and methylene resonances of the imido complexes occur some 0.2-0.6 (*H) and ^20 ppm ( C ) downfield from those exhibited by their dioxo alkyl precursors (Table 2B  IV). Furthermore, the  value diminishes by approximately 70-80 Hz which reflects the  different electron density extant at the metal center in the imido complexes than in their dioxo alkyl precursors. A plausible mechanistic pathway for reactions 2 is presented in Scheme 6-1. The first step involves nucleophilic attack by one of the oxo ligands of the dioxo alkyl complex at the electrophilic carbon atom of the isocyanate molecule leading to the formation of a metallacyclocarbamate intermediate, which subsequently expels C 0 and forms the oxo 2  imido complex 1. In an analogous manner, the second step involves a reaction of the oxo ligand of the imido complex 1 with another equivalent of />-MeC H N = C=O which would 6  4  result in the formation of a bis(imido) alkyl complex 2. The final step involves trapping of the bis(imido) alkyl complex, 2, with another equivalent of />-MeC H N = C=0 to form 3, 6  4  presumably via nucleophilic attack by the N-atom of the imido group at the electrophilic carbon of the isocyanate. The NMR spectroscopic studies suggest that in the presence of sufficient quantities of the isocyanate, the dioxo alkyl complexes react with the p-  „1  MeC H N = C=0 to incorporate three /?-MeC H N-groups and the final Cp M(=NC H 6  4  6  1  4  6  4  /7-Me)[-N(C H -p-Me)C(0)N(C H -/?-Me)]R product is formed. Initially in the reactions, 6  4  6  4  neither the oxo imido or the bis(imido) complexes accumulate to an extent where they may be detected spectroscopically. It is thought that once the oxo imido complex of the type la is produced in the reaction mixture (Scheme 6-1), a subsequent reaction with two more equivalents of />-MeC H N=C=0 rapidly occurs. This may explain the observation that, 6  4  in the absence of sufficient quantities of the isocyanate reagent, the dioxo alkyl complexes  204 Scheme 6-1.  205 react with it to form a product which has incorporated only one />MeC H N-group (the 6  4  oxo imido intermediate) or a product which has incorporated two />-MeC H N-groups (the 6  4  bis(imido) intermediate). The />MeC H N = C=0 reagent reacts more quickly with 1 and 6  4  2 to eventually form 3 than it reacts with the dioxo alkyl starting complex to form more of the intermediates 1 and 2. Regardless of the ratio of the />-MeC H N=C=0 and Cp*W(0) R reactants used, 6  4  2  i.e. 1:1, 2:1, or 3:1, the first product complexes detectable by *H and  NMR  spectroscopy are the final imido species of eq. 2, that is, complexes 3a or 3b above. The signals due to the imido metallacyclourea complexes (3) continue to increase in intensity at the expense of the proton and carbon resonances due to the two starting materials. However, when the reactions with less than three equivalents of the />MeC H N = C=0 6  4  reagent are near completion (i.e. nearly all the />-tolyl isocynate is consumed), additional proton and carbon resonances begin to appear in the NMR spectra. In the case where one equivalent of p-to\y\ isocynate is reacted with the Cp*W(0) (CH SiMe ), the new proton 2  2  3  and carbon signals in the NMR spectrum can be assigned to Cp*W(=NC H -/?6  4  Me) CH SiMe (2a) and the eventual products exist in a ratio of 3a:2a = 2:1 while a 2  2  3  considerable amount of the dioxo starting material remains in the mixture. When two equivalents of /Holyl isocynate are reacted with Cp*W(0) (CH ), the new signals that 2  3  grow in are attributable to Cp*W(0)(=NC H -p-Me)(CH ) (lb), and 3b and lb finally 6  4  3  exist in a ratio of 3b:lb = 4:1. These compounds 2a and lb correspond to the Cp*W(0)(=NC H -/>Me)R and Cp*W(=NC H -/?-Me) R intermediates depicted in 6  4  6  4  2  Scheme 6-1. B.  Solid-State Molecular Structure of Cp*W( = NC H -/>-Me) [-N(C H -^-Me)C(0)N(C H -^-Me)]CH SiMe3 (3a). 6  4  6  4  6  4  2  An X-ray crystallographic analysis on a single crystal of Cp*W(=NC H -/?-Me)[6  4  N(C H -p-Me)C(0)N(C H -p-Me)](CH SiMe ), 3a, confirmed the molecular structure of 6  4  6  4  2  3  206 this new imido complex prepared from reaction 2. Views of the molecular structure of 3a are presented in Figure 6.2. Figure 6.2 (a) illustrates that the complex possesses a normal four-legged piano stool molecular geometry. Selected bond lengths and bond angles of the molecule are listed in Table 6-1. The imido ligand is bonded to the W atom in an essentially linear fashion, the angle W-N(l)-C(16) being 171(4)°. The W-N(l) bond length is short (1.778(13) A ) and is consistent with a W=N double bond. Cotton and Shamshoum have suggested that the bond lengths 1.71-1.72 A and 1.84 A best describe 10  pure WsNPh triple and W=NPh double bonds, respectively. As a result, the observed W^imido  D o n d  distance  m  compound 3a suggests that this linkage possesses some partial  triple bond character. This is likely due to some extra N-»W * -bonding in addition to the 7T electrons already present in the W=N double bond. It is worth noting that the intramolecular dimensions of the W=N-C H -p-Me group in compound 3a compare well 6  4  with those found in WCl (=N-Ph)(PhC=CPh)(PMe ) (i.e. W - N ^ j ^ = 1.770(14) A and 2  W-N-C  3  2  = 176.3(1.5)°) which is proposed to contain a WsN-Ph group, the phenyl11  i m i d o  imido ligand acting as a 4-electron donor to the central W atom. Further structural features worth noting in the molecular structure of 3a are those that describe the W-[-N(C H -p-Me)C(0)N(C H -/?-Me)] unit (Figure 6.2 (b)). The torsion angle about 6  4  6  4  atoms N(3)-W-N(2)-C(ll) is 2.7(12)° which establishes that these atoms are co-planar (the oxygen atom is out of this plane by approximately 10°). The observed tungsten-nitrogen bond lengths are consistent with the existence of W-N single bonds. However, the metallacycle is not symmetrical as evidenced by the two different W-N (2.218 and 2.117 A ) and C-N (1.32 and 1.40 A ) bond lengths, respectively.  207  Figure 62. Views of the solid-state molecular structure of C^*W(=NC H -^-Me)[-N(C H -/?-Me)C(0)N(C H -/;-Me)](CH SiMe3), 3a: 6  4  6  4  6  4  2  (a) side-view of molecule, (b) view of molecule down the Cp *^  centI0  (Cp* atoms omitted for clarity).  i ^ - W axis a  208 Table 6-1. Selected Bond Lengths (A) and Bond Angles (deg) in Cp*W(=NC H -/>-Me) [-N(C H -^-Me)C(0)N-(C H -/?-Me)] (CH SiMe3). 6  4  6  4  6  4  2  Bond Lengths 1.778(13) 2.218(13) 2.117(12) 2.24(2) 2.023(5) 1.87(2) 1.88(3) 1.83(3) 1.91(3) 1.18(3)  W-N(l) W-N(2) W-N(3) W-C(12) W-Cp Si-C(12) Si-C(13) Si-C(14) Si-C(15) O-C(ll) b  N(l)-C(16) N(2)-C(ll) N(3)-C(ll) C(16)-C(17) C(16)-C(21) C(17)-C(18) C(18)-C(19) C(19)-C(20) C(19)-C(22) C(20)-C(21)  1.37(2) 1.32(3) 1.40(3) 1.41(5) 1.29(4) 1.40(3) 135(4) 136(4) 1.56(3) 1.41(3)  Bond Angles N(l)-W-N(2) N(l)-W-N(3) N(l)-W-C(12) N(l)-W-Cp N(2)-W-N(3) N(2)-W-C(12) N(2)-W-Cp N(3)-W-C(12) N(3)-W-Cp C(12)-W-Cp C(12)-Si-C(13) C(12)-Si-C(14) C(12)-Si-C(15) C(13)-Si-C(14) C(13)-Si-C(15) C(14)-Si-C(15) W-N(l)-C(16) W-N(2)-C(ll) W-N(2)-C(23) b  b  b  b  a  124.8(13) 96.0(7) 89.3(9) 121.7(12) 57.9(5) 74.5(7) 1D.4(4) 125.1(7) 112.3(4) 110.6(6) 114.9(11) 108.9(11) 1093(13) 109.2(14) 106.8(15) 107(2) 171(4) 99.2(12) 144.6(9)  C(ll)-N(2)-C(23) W-N(3)-C(ll) W-N(3)-C(30) C(ll)-N(3)-C(30) 0-C(ll)-N(2) 0-C(ll)-N(3) N(2)-C(ll)-N(3) W-C(12)-Si N(l)-C(16)-C(17) N(l)-C(16)-C(21) C(17)-C(16)-C(21) C(16)-C(17)-C(18) C(17)-C(18)-C(19) C(18)-C(19)-C(20) C(18)-C(19)-C(22) C(20)-C(19)-C(22) C(19)-C(20)-C(21) C(16)-C(21)-C(20)  Estimated standard deviations are provided in parentheses,  k Cp refers to the centroid of the (r; -C Me ) ring. 5  5  5  116.1(14) 101.4(12) 1303(9) 128.2(14) 135(2) 124(2) 101.4(18) 121.8(10) 112(3) 131(3) 117(2) 120(2) 124(2) 113(2) 119(2) 128(2) 124(2) 122(2)  a  209 C.  Future Studies The characteristic chemical reactivity of the imido products from reaction 2 is of  interest and, therefore, the scope of such reactions must be determined. Thermolysis of these products may afford the bis(imido) complexes by expelling a />-MeC H N=C=0 6  4  group, or may afford the aryl-nitrogen analogue of the oxo peroxo complexes considered in Chapter 5, i.e.,  v  CH SiMe3 2  The utility of complexes such as 3a and 3b in reactions for the formation of new carbonnitrogen bonds should be investigated. Specifically, reactions of these complexes with olefins and acetylenes are of interest.. Preliminary reactions of Cp*W(0) (CH SiMe ) with diphenyl ketene 2  2  3  1 2 , 1 3  have  been performed during this work. These reactions were performed in a manner similar to those described above with /7-tolyl isocyanate, and a reaction takes places immediately upon mixing the two reagents without solvents present. Three products in a ratio of 1:1:1 have been observed spectroscopically, but the identity of each complex remains to be ascertained. These reactions certainly look extremely promising, and it is possible that oxo alkylidene complexes of the type discussed in Chapter 4 may be preparable by this route as illustrated below. Ph  +  co  2  210  References and Notes: (1)  Nugent, W. A.; Haymore, B. L. Coord. Chem. Rev. 1980,31,123.  (2)  Nugent, W. A.; Mayer, J. M. Metcd-Ligand Multiple Bonds; Wiley-Interscience: New York, 1988.  (3)  Maata, E. A.; Du. Y. /. Am. Chem. Soc. 1988,110, 8249, and references therein.  (4)  Sharp, P. R.; Ge, Y. -W. Organometallics, 1988, 7,2234, and references therein.  (5)  (a) New Trends in the Chemistry of Nitrogen Fixation; Chatt, J.; da Camara Pina M.; Richards, R. L; Eds.; Academic Press: London, 1980. (b) Rocklage, S. M.; Turner, H. W.; Fellmann, J. D.; Schrock, R. R. Organometallics, 1982,1, 703.  (6)  Geoffroy, G. L.; Jernakoff, P.; Geib, S. J.; Rheingold, A. L. /. Chem. Soc, Chem. Commun. 1987,1610, and references therein.  (7)  (a) Glueck, D.S.; Hollander, FJ.; Bergman, R.G. /. Am. Chem. Soc. 1989, 111, 2719. (b) Kiisthardt, U.; Herrmann, W.A.; Ziegler, M.L.; Zahn, T.; Nuber, B.; /. Organomet. Chem. 1986,311,163.  (8)  The ratio of C:H:N is as expected; however, this analysis may be low due to combustion problems. Attempts are being made to obtain a better analysis.  (9)  Legzdins, P.; Phillips, E.C.; Trotter, J.; Yee, V.C. manuscript in preparation.  (10)  Cotton, F.A.; Shamshoum, E.S. /. Am. Chem. Soc. 1984,106,3222.  (11)  Clark, G.R.; Neilson, AJ.; Rickard, C.E.F.; Ware, D.C. /. Chem. Soc, Chem. Commun. 1989,343.  (12)  Diphenyl ketene may be prepared from the reaction of bromo-diphenylacetyl bromide with PPh in benzene (see reference 13). The orange-yellow liquid is best 3  purified by vacuum distillation, and it should be stored under a dinitrogen or an argon atmosphere at -20 °C. (13)  Darling, S.D.; Kidwell, R.L. /. Org. Chem. 1968,33,3974.  211 Spectral Index  Page  IR Spectrum of Cp*Mo(NO)(CH SiMe ) as a Nujol Mull.  212  IR Spectrum of [Cp*Mo(NO)(CH SiMe )] -0i -O) as a Nujol Mull.  213  IR Spectrum of Cp*Mo(NO)(CH Ph) as a Nujol Mull.  214  IR Spectrum of Cp*Mo(0) Me as a Nujol Mull.  215  IR Spectrum of Cp*W(0) CH SiMe as a Nujol Mull.  216  IR Spectrum of CpW(0)(r? -0 )(CH ) as a Nujol Mull.  217  IR Spectrum of Cp*W(C% -0 )(CH SiMe ) as a Nujol Mull.  218  2  3  2  2  3  2  2  2  2  2  2  3  2  2  3  2  2  2  3  IR Spectrum of Cp*W(0)(r? -0 )(CH SiMe )- /i -[(CN) C=C(CN) ] as a KBr Pellet.  219  IR Spectrum of CpW(0)Cl (CH SiMe ) as a Nujol Mull.  220  IR Spectrum of Cp*W(0)Me (CH SiMe ) as a Nujol Mull.  221  2  2  2  2  3  2  2  2  3  2  2  3  75 MHz C{ H} NMR Spectrum of Cp*W(0)Me (CH SiMe ) B  1  2  2  3  as a C D Solution. 6  222  6  IR Spectrum of Cp*W(0)Me(CH Ph) as a Nujol Mull. 2  75 MHz C{ H} NMR Spectrum of Cp*W(0)Me(CH Ph) 13  223  2  1  2  2  as a C D Solution. 6  224  6  IR Spectrum of Cp*W(0)( = CHSiMe )(CH SiMe ) as a Nujol Mull.  225  IR Spectrum of Cp*W(0)( = CHPh)(CH SiMe ) as a Nujol Mull.  226  3  2  3  2  3  300 MHz H NMR and 75 MHz ^C^H} NMR Spectra of X  Cp*W(0)(=CHPh)(CH SiMe ) as a C D Solution. 2  3  6  6  227  IR Spectrum of Cp*W(=NC H -^-Me)[-N(C H -/»-Me)C(0)N6  4  6  4  (C H -/>Me)]CH SiMe as a KBr Pellet. 6  4  2  3  228  212 IR Spectrum of Cp*Mo(NO)(CH SiMe ) as a Nujol Mull. 2  3  2  0  o Q  IZB'ES  592 'ES  B S S 'Zt-  2 5 8 "IE  3 3 N V ±1I  9fr T "12  WSNVyj.%  QW "OX S 9 8 2 "•—  CI. 2703  o III  12. Q38  XTRANSMITTANCE 23. 802 35.588 47.334  58.100  70. 888  WW. I°f N. * ^ (o-»0- [( 3WTS H3)(ON)ow,d3] jorarupsdsHI n  3  €  z  217  TQO 'BV  eil'Ot'  SZ.T "ZE E E S "f Z I B S *ST 33NVJ.XI WSNVaXX  IBt'E "8  S S O f TJ  220  222 75 MHz ^C^H} NMR Spectrum of Cp*W(0)Me (CH SiMe ) as a C D Solution. 2  2  3  6  6  224 75 MHz C{ H} NMR Spectrum of Cp*W(0)Me(CH Ph) as a C D Solution. B  1  2  2  6  6  -•.4251  O  •  szz  8. 1472  ZTRANSMITTANCE 18.719 25.292 33.884  42.438  51.008  227 300 MHz *H NMR and 75 MHz ^C^H} NMR Spectra of Cp*W(0)(=CHPh)(CH SiMe ) as a C D Solution. 2  3  ii  6  6  

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.831.1-0060263/manifest

Comment

Related Items