The stabilization of tungsten(VI) alkyls, alkylidenes, and hybrides using a bidentate bis-amide ligand


Material Information

The stabilization of tungsten(VI) alkyls, alkylidenes, and hybrides using a bidentate bis-amide ligand
Physical Description:
Vanderlende, Daniel D
Publication Date:

Record Information

Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 21566647
oclc - 33343963
System ID:

Table of Contents
    Title Page
        Page i
        Page i-a
        Page ii
        Page iii
    Table of Contents
        Page iv
        Page v
        Page vi
        Page vii
    Chapter 1. Background and introduction
        Page 1
        Page 2
        Page 3
        Page 4
        Page 5
        Page 6
        Page 7
        Page 8
        Page 9
        Page 10
        Page 11
        Page 12
    Chapter 2. Synthesis of BIS-AMIDE chelate ligands and ligand metal complexes
        Page 13
        Page 14
        Page 15
        Page 16
        Page 17
        Page 18
        Page 19
        Page 20
        Page 21
        Page 22
        Page 23
        Page 24
        Page 25
        Page 26
    Chapter 3Synthesis and reactivity of W9VI) alkyls and alkylidenes
        Page 27
        Page 28
        Page 29
        Page 30
        Page 31
        Page 32
        Page 33
        Page 34
        Page 35
        Page 36
        Page 37
        Page 38
        Page 39
        Page 40
        Page 41
        Page 42
        Page 43
        Page 44
        Page 45
        Page 46
        Page 47
        Page 48
        Page 49
        Page 50
        Page 51
        Page 52
        Page 53
        Page 54
    Chapter 4. Formation of W(VI) hydrides from the BIS-alkyl complexes
        Page 55
        Page 56
        Page 57
        Page 58
        Page 59
        Page 60
        Page 61
        Page 62
        Page 63
        Page 64
        Page 65
        Page 66
        Page 67
        Page 68
        Page 69
        Page 70
        Page 71
        Page 72
        Page 73
        Page 74
    Chapter 5. Experimental
        Page 75
        Page 76
        Page 77
        Page 78
        Page 79
        Page 80
        Page 81
        Page 82
        Page 83
        Page 84
        Page 85
        Page 86
    Appendix A. Tables of NMR data
        Page 87
        Page 88
        Page 89
        Page 90
        Page 91
        Page 92
        Page 93
        Page 94
        Page 95
        Page 96
        Page 97
        Page 98
        Page 99
        Page 100
        Page 101
    Appendix B. Tables of crystallographic data
        Page 102
        Page 103
        Page 104
        Page 105
        Page 106
        Page 107
        Page 108
        Page 109
        Page 110
        Page 111
        Page 112
        Page 113
        Page 114
        Page 115
        Page 116
        Page 117
        Page 118
        Page 119
        Page 120
        Page 121
        Page 122
        Page 123
        Page 124
        Page 125
        Page 126
        Page 127
        Page 128
        Page 129
        Page 130
        Page 131
        Page 132
        Page 133
        Page 134
        Page 135
        Page 136
        Page 137
        Page 138
        Page 139
        Page 140
        Page 141
        Page 142
        Page 143
        Page 144
        Page 145
        Page 146
        Page 147
        Page 148
        Page 149
        Page 150
        Page 151
        Page 152
    Biographical sketch
        Page 153
        Page 154
Full Text







The people deserving thanks from the author for the completion of this dissertation are innumerable. A great debt is owed to Dr. Jim Boncella, who has guided the author through this adventure. The lessons taught by Jim will have an influence on the rest of the author's life. He not only challenged and motivated the author, he also brought out the best in the author's golf game, teaming with Larry Villanueva and William Vaughan to win the 1993 Analytical Open.

Special thanks goes to Dr. Khalil Abboud. Dr. Abboud solved or helped the author solve all of the crystal structures reported in this dissertation. He taught the author everything the author knows about crystallography. The author is grateful for the patience and enthusiasm Dr. Abboud showed throughout the research which was the basis for this dissertation, exemplified by the eleventh hour structure included in Chapter 4.

There were many people who passed through the Boncella lab during the author's tenure. Everyone of these people played a role in the completion of this dissertation. Those who have since moved on, Larry, Laura Blosch, Scott Gamble, and Gaines Martin, were positive role models who taught by example how to work hard in a fun group environment. Will Vaughan has been a constant source of fun, excitement, and intellectual stimulation over the past four plus years.' The author would also like to remember all the other members of the Boncella group who have made research so stimulating; Percy Doufou, Jerrold Miller, Mary Cajigal, Justine Roth, Jon Penney, Steve Wang and Faisal Shafiq. Who could ever forget Tegan Eve and Melissa Booth?

Special thanks also goes to Mike Cruskie and Chris Marmo who have been great friends and fellow chemists over the years. A special thanks also goes to the Talham group, especially John Pike and Houston Byrd, who both shared their unique perspectives ii

on life with the author. The author is also indebted to the people who made the research possible on a daily basis; Dr. King, Charlie Cromwell, Rudy and Vern.

None of this would have been possible without my parents who instilled in their son the desire to never be satisfied with past accomplishments. The author would like to acknowledge the love and devotion of the "Coach". He was always there for the author, displaying undying devotion and loyalty. Lastly, the author would like to acknowledge his wife Michelle, who has been the driving force behind the completion of this dissertation. Michelle has always believed that there is nothing her husband cannot accomplish, and that belief has motivated the author to strive to be more than he ever thought he could be, because maybe she is right.


ACKNOWLEDGEMENTS ......................................................................... ii

A B STR A CT ......................................................................................... vi


1 BACKGROUND AND INTRODUCTION ............................................. 1

1.1 High Oxidation State Transition Metal Chemistry Involving Ligand Metal
Multiple Bonds ....................................... ........... 1
1.2 Olefm Metathesis and Olefin Metathesis Polymerization .............................. 4
1.3 Chelate Stabilized Alkylidenes .......................................................... 8
1.4 Polydentate, Polyanionic Ligands ....................................................... 11

METAL COMPLEXES ................................................................ 13

2.1 Preparation of Bidentate Ligands ........................................................ 13
2.2 Synthesis of New Ligand-Metal Complexes ....................................... 16

ALKYLIDENES ........................................................................... 27

3.1 Background Information on W(VI) Alkyls and Alkylidenes ...................... 27
3.2 Synthesis of W(VI) Bis-Alkyl Complexes .......................................... 29
3.3 Alkylidene Formation From Bis-Alkyl Complexes ................................ 36
3.4 Metathesis Activity of W(NPh)(CHCMe3)(PMe3)[(Me3SiN)C6H4], 36 .......... 46

COMPLEXES .......................................................................... 55

4.1 High Oxidation State Transition Metal Hydride Complexes ...................... 55
4.2 Preparation of W(VI) Hydride Complexes ......................................... 55
4.3 Reactivity of the Dihydrides .......................................................... 69

5 EXPERIMENTAL ..................................................................... 75

APPENDICES .................................................................................... 87

A TABLES OF NMR DATA ................................................................ 88

B TABLES OF CRYSTALLOGRAPHIC DATA ..................................... 103

REFEREN CE S .................................................................................... 147


BIOGRAPHICAL SKETCH .................................................................... 153

Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy


Daniel D. VanderLende

December, 1994

Chairman: James Boncella
Major Department: Chemistry

The synthesis of a number of W(VI) complexes stabilized by bis-amide chelate ligands was achieved with the goal of preparing new olefin metathesis polymerization catalysts. Addition of Li2[(NSiMe3)2C6H4] 2, to W(NPh)C14(OEt2) yields W(NPh)C12[(NSiMe3)2C6H4] 14. A single crystal diffraction study of 14 reveals that it crystallizes in the space group P 21/n with a = 10.294(2) A, b = 17.859(3) A, c = 12.565A, 3 = 104.15(2)0, V = 2384.6(8) A3, Z = 4. The structure of 14 is unique in that the ligands phenyl ring is in close contact with the metal center, with a fold angle of 530. Addition of PMe3 to 14 affords the purple mono-adduct, W(NPh)C12(PMe3)[(NSiMe3)2C6H4] 21. A single crystal diffraction study of 21 reveals that it crystallizes in the space group P-1 with a = 9.562(1) A, b = 10.277(1) A, c = 14.920(2) A, a = 82.15(1)0, 3 = 80.18(1)o, y = 80.41(1)0, V = 1415.6(3) A3, Z = 2. Compound 14 can be alkylated to give the corresponding bis-alkyls, W(NPh)R2[[(NSiMe3)2C6H4]. The 14 electron bis-alkyl compounds show no evidence of a-agostic W-H-C interactions and have 1'JCoa-H values between 120 and 130 Hz.


W(NPh)(CH2CMe3)2[(NSiMe3)2C6H4] 25 can be heated in the presence of excess PMe3 to give the new alkylidene complex, W(NPh)(CHCMe3)(PMe3)[(NSiMe3)2C6H4], 36. Compound 36 crystallizes in the space group P 21/c with a = 16.116(3) A, b = 11.340(2) A, c = 17.960(4) A, 3 = 106.28(2)0, V = 3151(1) A3, Z = 4. The x-ray structure of 36 reveals a unique square pyramidal structure with the alkylidene carbon in the apical position. Compound 36 is an active ROMP catalyst, polymerizing twenty five equivalents of norbornene in ten minutes at room temperature. The reactivity of the catalyst can be altered by the addition of excess PMe3 to the reaction mixture.
W(NPh)(CH2CMe3)2[(NSiMe3)2C6H4] 25 reacts with molecular hydrogen in the presence of PMe3 to give the seven-coordinate dihydride complex, W(NPh)H2(PMe3)2[(NSiMe3)2C6H4] 39. The dihydride reacts with two equivalents of ethylene or styrene to give the W(IV) olefin complexes W(NPh)(7I2C2H4)(PMe3)2[(NSiMe3)2C6H4] 46, and W(NPh)(i2CH2CHPh)(PMe3)2[(NSiMe3)2C6H4] 47. The olefin complexes react with H2, hydrogenating the olefin and forming the dihydride, 39.


BACKGROUND AND INTRODUCTION 1,1: HiLyh Oxidation State Transition Metal Chernist[y Involving Metal Ligand Multiple Bonds.

There has long been an interest in the isolation of high oxidation state transition

metal complexes. 1 The term high oxidation state transition metal refers to transition metals which are in their highest or nearly highest oxidation state. Throughout this discussion, high oxidation state transition metals will usually refer to transition metals with zero d electrons, or dO complexes. 'Me applications of high oxidation state transition metal complexes are innumerable. A plethora of reactions are catalyzed by such compounds 1,2 oxidations, metathesis, polymerizations and many organic transformations. Therefore, the synthesis of novel compounds and investigation of their properties and reactivity are essential in order to expand our understanding of high oxidation state transition metal chemistry.

High oxidation state transition metals were being used in many applications long before the intrinsic characteristics or the structure of discrete molecules were known.3 In other words, high oxidation state transition metal complexes have been used as heterogeneous catalysts and homogeneous catalysts for decades, but, in order to expand our understanding of the reactions catalyzed by these compounds, the identity of the discreet molecules must be known. Often times, the active species in the reaction is different that the starting compound, and hence the structure of the active species is unknown or unproven.3 Many high oxidation state transition metal complexes contain multiply bonded atoms; nitrogen, oxygen, and carbon are the most common.2 Also, the active species in many catalytic processes are postulated to contain metal-ligand multiple



bonds.3,4 Often, these multiply bonded ligands are the key to the reactivity or stability of the molecules. Therefore, the synthesis and structural elucidation of new multiply bonded ligand-transition metal complexes will always be relevant.

Examples of transition metal oxo5 [0]2-, imido6 [NR]2- and nitride7 [N13complexes are numerous. Books and reviews on these types of ligands can be found throughout the literature. The chemistry of high oxidation state transition metal-carbon multiple bonds, alkylidenes [CHR]2- and alkylidynes [CR]3-, are rapidly becoming better understood, yet relatively few examples compared to oxo or imido complexes are known.8 The bonding in oxo, imido, and alkylidene complexes share some common characteristics. They all involve a metal ligand a-bond and one or more it-bonds between the ligandp orbitals and the empty metal d orbitals.1,8,9 In oxo and imido complexes, a lone pair of electrons can also form a second 7r-bond to the metal center. In imido complexes, this is evident in the fact that for most of the structurally characterized compounds, the M-N-C bond is nearly linear, greater than 1600.1,8 The bond is considered a six-electron donor and a di-anionic contribution from the ligand, creating a formal bond order of 3, Figure 1.1. In alkylidenes, there is not an extra pair of electrons, therefore, the bond is considered as one a and one y-bond, a four-electron [-2] donor, with a formal a bond order of 2, Figure

1.2. This interaction between the ligand 7t and metal d

Figure 1.1. Figure 1.2.
Bonding in imido complexes. Electronic interactions in alkylidenes.

orbitals leads to a stabilization of the complex due to electron donation to the empty orbitals on the dO metal center. L9 For this reason, many of the known high oxidation state


compounds contain one of the multiply bonded ligands mentioned. Figure 1.3 shows how the number of publications in this field of research has grown in just a few decades.1 It should be recognized that the table does not include oxo complexes, work on which is published at roughly an order of magnitude greater than the others combined.


*ZU Imidos
0 Alkylidenes

4 100I-~z

1960 1965 1970 1975 1980 1985
Period Ending

Figure 1.3. Graph of imido and alkylidene publication rates.

It would be difficult and unnecessary to give a complete review of high oxidation state transition metal complexes with multiply bonded ligands in this introduction. Therefore some restraints will be put on the background information. The information relevant to the research that was carried out pertains to W(VI) oxo, imido, and alkylidene compounds. Therefore this discussion will be limited to examples of these and other closely related compounds for direct comparison to the tungsten derivatives. Although numerous new W(VI) imido and a few oxo compounds were isolated as a result of this work, in every instance they behave as spectator ligands. They play an important role in the electronic stabilization of the complexes but do not participate directly in any reactions. This will become more evident in the discussion of the research results.


1.2 Olefin Metathesis and Olefin Metathesis Polymerization

The initial goal of this project was the isolation of thermally stable, coordinatively unsaturated alkylidene complexes. These new complexes were then intended to catalyze specialized metathesis polymerization reactions. And although this goal was attained in part, many novel, unrelated aspects of high oxidation state chemistry were observed along the way.

In just the last two decades, the important role of transition metal alkylidene

complexes in olefin metathesis reactions has been thoroughly investigated.10,11 The olefin metathesis reaction, in general, can be described as the net breaking of carbon-carbon double bonds, and forming two new carbon-carbon double bonds Figure 1.4. The

R H H R"
transition metal H
+ catalyst H +
R' R...
Figure 1.4. Scheme showing the net conversion observed in olefin metathesis reactions.

overall result is the exchange of substituents on the olefins. It was the suggestion by Herrison and Chauvin12 in 1970 that the mechanism for olefin metathesis involves an alkylidene bond Figure 1.5. An olefin can then coordinate to the electrophilic metal center. An intermediate metallacyclobutane is formed which can either cleave to give the original alkylidene/olefin complex or cleave to give a new olefin complex.


R" R[
H [MH H + H

Figure 1.5. The Chauvin mechanism for metathesis involving metal alkylidenes with
arrows depicting the direction of electron flow for productive metathesis.


This mechanism gained wide acceptance and seemed quite plausible for systems such as the popular olefin metathesis catalyst, WOC14/EtA1Cl2, where the formation of an initial tungsten ethylidene complex3 is easy to rationalize Figure 1.6.

Cl EtAlC12 -(A1C13)~ C -CH3CH3 C H1
I/+ 2 EtA1C12 -(c-)<|C/
Cl Cl Cl 1 CH3
Figure 1.6. The proposed formation of an ethylidene complex in the WOC14/EtA1Cl2 system.

Postulation and speculation about alkylidene formation officially ended on July 27, 1973, when Schrock13 at I. E. du Pont Central Research synthesized Ta(CHCMe3)(CH2CMe3)3. Since that time, numerous other alkylidene complexes have been isolated and have been shown to be active catalysts for metathesis and metathesis polymerization reactions. In the early 1980s, tungsten and molybdenum alkylidenes received special attention since they are highly active olefin metathesis catalysts. The fields of olefin metathesis and olefin metathesis polymerization received a boost in 1989 when Schrock published the detailed synthesis14 of W(NAr')(CH-t-Bu)(OR)2 (Ar' = 2,6-iPr2Ph; OR = O-t-Bu, OCMe2(CF3), OCMe(CF3)2). The W(NAr')(CH-t-Bu)(OCMe(CF3)2)2 derivative has proven to be the most active alkylidene catalyst to date. Recently, the molybdenum t-butoxide derivative has become available commercially from Strem Chemical, through Catalytica, albeit for a hefty price.

Schrock's tungsten and molybdenum catalysts have some unique features which lend to the high reactivity which has been observed. In order to design a new or better catalyst, certain features must be retained. Obviously Schrock's catalyst involves Group

(VI) metals in the 6+ oxidation state. That the catalysts are only four coordinate is essential since coordination of the olefin to the metal center is the initial step in metathesis. The imido moiety's role is crucial. There are examples of oxo-alkylidene's where the role of


the oxo is identical to the imido.15 Of the known tungsten and molybdenum alkylidenes, examples without an imido or oxo ligand are less common. 16

The imido functionality offers a great electronic stabilization to the metal center. The bonding can be thought of as a a bond and two nt bonds, the second n bond arising from donation of the lone pair of electrons on the imido nitrogen to an empty metal d orbital. The imido functionality can also stabilize the complex by adding steric bulk to the metal center. Most of the known alkylidene complexes contain 2,6-disubstituted aryl imido ligands.8 The other substituents also contribute to the steric bulk around the metal center. All of the alkoxide substituents are rather bulky, and the alkylidene substituent is usually a bulky alkyl group. The steric bulk is necessary since these complexes are known to dimerize forming bridging alkylidenes.8 The bulky metal center reduces the probability of this transformation. The steric bulk also plays a role in the formation of the alkylidene itself. A large percentage of the known alkylidenes are formed through aX-hydrogen abstraction reactions from dialkyl precursors or intermediates.8 The steric bulk helps promote the oa-abstraction, usually of a bulky alkyl substituent. The reactivity of the alkylidene catalysts is observed to increase as the electron withdrawing nature of the alkoxide is increased, O-t-Bu < OCMe2(CF3) < OCMe(CF3)2).8 This is intuitive since the electron withdrawing nature of the alkoxide would make an already electron deficient molecule more so, thereby increasing the olefin affinity, which, as was mentioned before, is the initial step in olefin metathesis.

As was mentioned earlier, these alkylidene catalysts can facilitate the polymerization of olefins. 10,11 The most common metathesis polymerization reaction is Ring Opening Metathesis Polymerization (ROMP).10 The ROMP reaction is driven by the relief of ring strain in cyclic olefins. Since olefin metathesis reactions are actually series of equilibria, the opening of a strained ring prohibits the reverse reaction, driving the polymerization.10 The overall ROMP reaction is shown in Figure 1.7. A large volume of work has been done on the ROMP reaction. Norbomene, NBE, is commonly used as the olefin since it is


readily polymerized by a large number of catalysts. Schrock and Grubbs have demonstrated that many ROMP reactions are 'living' polymerizations. 17

A unique olefin metathesis polymerization reaction which was developed here at the University of Florida involves acyclic dienes. 18 The net reaction for acyclic: diene

R + R ) M R

Figure 1.7. The ring opening metathesis polymerization of a cyclic olefin.

metathesis polymerization (ADMET) can be seen in Figure 1.8. The equilibrium is driven toward polymer formation by the removal of ethylene, or another small, volatile, olefinic molecule, from the reaction m ixture. ADMET polymerization has long been the driving force in the research efforts of the group. ADMET reactions are best carried out in neat monomer in order to maximize the olefm concentration and to prevent side reactions. One of the major hindrances in this chemistry is precipitation of the reaction mixture before the polymerization can be carried to high molecular weight. As an example, when poly(octene) reaches twelve connections in neat 1 ,9-decadiene, precipitation occurs. Performing the reaction at a temperature above the melting point of the polymer would alleviate this problem. However, at this point, the thermal stability of the catalyst becomes crucial. Schrock's catalyst is thermally unstable, so, although the Schrock catalyst has proven efficient for ADMET, achieving high molecular weight polymer is difficult.

M= +

Figure 1.8. The ADMET Polymerization of 1,9-Decadiene


The principles of ADMET have also been applied to depolymerizing unsaturated polymers.19 Since all the steps of ADMET are reversible, reacting the polymer, another olefin, and catalyst should depolymerize a poly-ene such as polybutadiene. Success has been found when polybutadiene is depolymerized using Schrock's catalyst and end-capped with silylenes. This reaction will be discussed more in Chapter 3.

Although the Schrock catalyst described above represents the ideal at this time, there are shortcomings involved in its preparation and reactivity. From first-hand observations, it has been observed that the preparation of Schrock's catalyst is a lengthy, patience-trying procedure. Not only are there multiple steps involved, but the yields are low and some of the reagents are far from inexpensive. Secondly, once synthesized, the catalysts, especially the fluorinated alkoxides, are particularly air/moisture sensitive. This aspect of its reactivity also translates into an intolerance of certain functional groups. A common phrase heard in polymer laboratories is "The catalyst was poisoned...". Adding to the drawbacks of Schrock type catalysts is their thermal 'instability'. There are polymer systems in which heating the reaction mixture would be advantageous toward achieving maximum yields or molecular weights; however, Schrock's catalysts general decompose over the temperature range of 60-80 oC.20 So, although the introduction of Schrock's catalysts opened up vast areas in metathesis and polymerization, there is still a great need for new catalysts which are easier and less expensive to prepare.

1.3 Chelate Stabilized Alkylidenes

An intuitive approach to the synthesis of an alkylidene compound with a greater

thermal stability would be to use chelating ligands somewhere in the molecule. The use of a chelating ligand may also produce interesting stereochemical properties in the metathesis reactions. This approach has been investigated by Schrock, Grubbs, Boncella, VanKoten


and others. There are a number of options as to where to apply 'chelates' in these complexes. Schrock has made a series of diolate complexes21 of the type Mo(CHCMe2Ph)(NAr)(diolate) Figure 1.9. Although no comment is made about

R/ / Sgat-B t-Bu
R SiMe2Ph tB
O. CHMe2Ph Ph. yCIMcPh
o o
SiMe2Ph t-Bu t-Bu
(R4tart)Mo(NAr)(CHMe2Ph) BINO(SiMePh)2Mo(NAr)(CHMe2Ph) Biphenol(t-Bu)4Mo(NAr)(CHMe2Ph) R = phenyl, naphthyl

Figure 1.9. Bidentate diolate complexes prepared by Schrock.

the thermal stability, it is assumed that they are more stable than the complexes with monodentate alkoxides. These compounds do, however, allow for stereochemical control of ROMP reactions. Grubbs has taken an interesting approach by synthesizing osubstituted aryl alkylidenes22, where the o-substituent has a-donor properties and can chelate to the metal center, stabilizing the alkylidene Figure 1.10. VanKoten's alkylidenes are stabilized by chelating a-donors as well.23 In the tungsten(VI) alkylidene complex, W(NPh)(C6H4-o-CH2NMe2)(CHSiMe3)(OSiPh3), the -NMe2 group on the ortho-methylene group acts as a a-donor, adding a chelate effect Figure 1.11.

Me' oiiO

11N H_ N OSiPh3
OR' = OCCH3(CF3)2 C
R = H, Me, i-Pr Si
Figure 1.10. Grubbs' Catalyst Figure 1.11. VanKoten's Catalyst.

Work done here at Florida by Blosch, Gamble, and Vaughan in the research group of James Boncella24 has focused on the use of the tris-chelating, mono-anionic ligand hydro(tris)pyrazolylborate, Tp. Six-coordinate alkylidene complexes of the type


TpM(NAr)(CHR)(X) (M = W, Mo; R = CMe3, CMe2Ph; X = Cl, Br, OTf, OMe, NHPh) have been isolated Figure 1.12. These complexes show remarkable air and thermal stability, although they only show metathesis activity in the presence of a Lewis acid, which generates a vacant coordination site. Once again, the importance of coordinative unsaturation is observed.


M=W; Mo
Y = NAr O
X = C; Br; OTf; OMe; NHPh
X R R =Me; Ph
y H
Figure 1.12. Tridentate chelate complexes using the hydro(tris)pyrazolylborate ligand.

The use of chelate ligands has provided two things to the chemistry of alkylidenes. First, it has produced more thermally stable alkylidenes, even air-stable in the case of the Tp alkylidenes. Secondly, the chelates ligands have greatly reduced the reactivity of the alkylidenes towards olefins. The ultimate goal of this work would be to design a ligand which provided stability while maintaining a high olefin affinity at the metal center. Obviously this would mean a coordinatively unsaturated, electron deficient molecule. A look back at the some of the known chelated alkylidene compounds reveals an undesirable trend; the chelate ligand involves a neutral a-donor interaction. In simple terms, all this does is clog the coordination sphere of the molecule. A more pragmatic approach would eliminate neutral a-donor ligands to a great extent and concentrate on the use of anionic ligands as chelates. The Tp ligand for example is a tridentate, mono-anionic ligand. Van Koten's chelate could be considered a bidentate, mono-anionic ligand.


1.4 Polvdentate, Polvanionic Chelate Ligands

An ideal ligand would be a polyanionic, polydentate ligand. Either a bidentate, dianionic ligand, or a tridentate, tri-anionic ligand. There are a number of examples in the literature of multidentate, multi-anionic ligands. Schrock and Cummings have used a tetradentate, tri-anionic ligand to prepare some novel high oxidation state titanium,25 vanadium25 and tantalum26 compounds. The ligand, (Me3SiNCH2CH2)3N was used to stabilize an interesting terminal phosphinidene complex as well as some other high oxidation state early transition metal complexes Figure 1.13. Verkade27 first used methyl derivatives of this ligand, (MeNCH2CH2)3N, to stabilize some group (IV) and (V) compounds Figure 1.14. Gade28 has used a similar ligand, H3CC(CH2NHR)3 (R = Me, Et, iPr, SiHMe2, and SiMe3), to stabilize titanium complexes. This ligand has a carbon in the bridgehead position, avoiding the a-interaction which the nitrogen had with the electrophilic metal center, creating a tridentate, tri-anionic ligand, thereby decreasing the metal centers coordination number by one.
Me3Si\ Y-..I. M=VTa
N--Y-M-----e3 M = V, Ta
Ta SiMe3 I Y =O,NMe
Me3Si N/ Y Z =O,NR

N 7N,,

Figure 1.13. Phosphinidene Figure 1.14. Gade's tetradentate
stabilized by Verkade's ligand. tris-anionic ligands.

Wilkinson29 has employed the use of o-phenylenediamine in the stabilization of tungsten (V) and (VI) compounds. One negative aspect of this particular bidentate, dianionic ligand is the tendency for rearrangement of the bis-amide to an imido-amine. This problem could be easily overcome by synthesizing N,N'-disubstituted derivatives. This premise is where the present study begins. A number of novel N,N'-disubstituted


derivatives of o-phenylenediamine and 1,8-diaminonapthalene were synthesized and their application as bidentate, di-anionic ligands were investigated. All of the work reported here involves tungsten (VI) phenylimido or oxo complexes exclusively. The high-yield synthesis of various tungsten (VI) phenylimido ligand stabilized complexes allows a convenient route into the reaction chemistry of these complexes. This work, involving the preparation of starting materials, will be covered in Chapter 2. Simple alkylation reactions allow the isolation in good yield of a series of mono- or di-alkyl complexes. Isolation of stable cis-bisalkyl complexes offers a look into the reaction chemistry of these compounds, and will be the focus of Chapter 3. Heating the bis neopentyl derivative in the presence of PMe3 induces oa-hydrogen abstraction, forming an alkylidene and one equivalent of neopentane. The neopentylidene complex is an active catalyst for the ROMP of norbomene. The activity of the catalyst can be tailored by the addition of excess ligand to the reaction mixture. This will also be covered in Chapter 3. A majority of the chemistry found in Chapters 2 and 3 has previously been published.30 An interesting feature of the alkyl complexes is that they react at room temperature with molecular hydrogen to form high oxidation state hydride complexes. The addition of a a-donor ligand accelerates the reaction tremendously as well as aides in the stabilization of the molecule. The sevencoordinate dihydride species formed reacts with ethylene, hydrogenating one equivalent, while the reduced metal species forms a tungsten (IV) ethylene complex. The reactivity of the alkyls towards hydrogen and olefins will be the focus of Chapter 4.


2.1: Preparation of Bidentate Ligands.

In order to pursue the use of 1,2-phenylenediamine as a bidentate, di-anionic

ligand, an accessible route to the synthesis of bulky N,N'-disubstituted derivatives was desired. A thorough review of the literature reveals surprisingly few examples of such compounds. The only known bisalkyl example is N,N'-dimethyl-1,2phenylenediamine,31 which is prepared through a tedious, dangerous, multi-step synthesis. There are examples of other disubstituted derivatives such as -S(O)2tolyl (tosyl)31 and

-C(O)Ph.32 A slight discrepancy in the literature was discovered for the case of N,N'bis(trimethylsilyl)- 1,2-phenylenediamine, 1,2-(Me3SiNH)2C6H4, 1. Before this was discovered, however, a nearly quantitative one-step synthesis of 1 was discovered eq

2.1. O-phenylenediamine was dissolved in Et2O on as large a scale as available Me3Si\
1. 2 eq. Me3SiCl .3H /NH2 2. 2 eq. NEt3 / H
I eq 2.1
NH Et2O,00 C I H
a1~NH2 N


glassware would allow. A slight excess of two equivalents of Me3SiCl was added, forming a white precipitate, presumably the hydrochloride salt A slight excess of two equivalents of NEt3 was added to the slurry. The solution became yellow amidst the solid. Filtering and removing solvent gave a bright yellow solid in greater than 95% yield. It is



important to note that the slightest impurity causes formation of a yellow oil, partly due to the low melting point (29 'C) of 1.
The preparation of 1 described above differs greatly from the literature methods. Compound 1 was first reported in the literature in 1960 by Birkofer.33 Birkofer refluxed o-phenylenediamine, two equivalents of Me3SiCl and NEt3 in toluene followed by a fractional distillation to give a moderate yield of 1. In 1970, West34 reported refluxing ophenylenediamine, hexamethyldisilazane, and a catalytic amount of Me3SiCl in THF for 24 hours. Fractional distillation using a spinning band column gave an 80% yield of 1. West also reported that addition of MeLi to 1 in THF solvent caused rapid 1,4 anionic rearrangements to occur.34 This finding was an important consideration when solvents for the ligation chemistry were chosen. In 1985, Lappert reported heating ophenylenediamine, Me3SiC1, and NEt3 in toluene (with no mention of Birkofer).35 Lappert then treated 1 with MgBu2 to give the deprotonated dimer, [Mg{ I.tN(SiMe3)C6H4N(SiMe3)-o } (OEt2)]2. Maatta36 reports another preparation in 1992. Here, o-phenylenediamine is deprotonated with two equivalents n-BuLi, followed by addition of two equivalents of Me3SiC1. The product, 1, is isolated as a yellow oil by vacuum distillation in 80% yield. One concern that this report brought out was that when (Me3SiNH)2C6H4, 1, was allowed to react with WC16, two equivalents of HC1 and two equivalents of Me3SiCl were lost, forming a bridging di-imide eq 2.2.

N, H 1. CH2C12
+ WC16 2. THF Cl N NCl C e
W .w1 W eq2.2

From Lappert's account, 1 should be susceptible to deprotonation. Addition of two equivalents of n-BuLi to 1 afforded the white salt Li2(Me3SiN)2C6H4, 2. Interestingly, the salt was soluble in C6D6 and the 1H NMR of 2 verifies that there were no N-H


protons. The salt, however, was extremely moisture sensitive and spontaneously ignited upon exposure to air, hence prolonged storage was difficult. Before the application of 1 and/or 2 as a ligand will be addressed, the synthesis of other potential ligands will be discussed.

Although the high yield synthesis discussed for 1 might seem applicable for a series of silyl chlorides, it did not prove to be. However, when refluxing hexanes were used as the solvent instead of Et20, a series of silylated compounds were isolated in high yield. This general route applies to making the -SiMe2Ph, 3; -SiMePh2, 4; or -SiMe2-t-Bu, 5, derivatives of o-phenylenediamine. Another derivative of o-phenylenediamnine was prepared in this manner, 4,5-dimethyl-1,2-(Me3SiNH)2C6H2, 6. This route was also utilized to prepare 1,8-(Me3SiNH)2C10H4, 7, in high yield and on a large scale from 1,8diaminonaphthalene and two equivalents of both Me3SiC1 and NEt3. An asymmetric disubstituted o-phenylenediamine derivative, 1-(PhNH)-2-(Me3SiNH)C6H4, 8, was synthesized similarly from N-phenyl-o-phenylenediamine, Me3SiC1 and NEt3.

Attempts to synthesize dialkyl derivatives of o-phenylenediamine proved less successful. In an attempt to synthesize 1,2-(iPrNH)2C61H4, o-phenylenediamine was slurried with excess sodium acetate in a cold acetic acid/acetone/water mixture. Excess NaBH4 was added slowly. After neutralizing the solution with NaOH and isolation of products, a 1:1 mixture of products was formed. They were separated by flash chromatography and analyzed. 1,2-(iPrNH)2C6H4, 9, was isolated as a colorless oil. The other product was a heterocyclic compound, 10, which is shown in eq 2.3. The heterocycle most likely forms by an attack of one imine nitrogen on the other imine carbon, followed by a 1,3 proton shift. Altering the reaction conditions did not change the relative yields of 9 and 10. The 1H NMR of 10 is shown in Figure 2.1. Since formation of the 7-membered heterocylce seemed unavoidable in this system, the same reaction was attempted using 1,8-diaminonaphthalene. Here, the formation of a 6-membered heterocycle might be less likely due to the strain involved in one imine attacking


NH2 1. acetic acid/H20 CMe
2. sodium acetate
3. acetone
N[ N CMe

1. NaBI-4 H.... H
2. NaOH + H eq 2.3
9 H H


the other. A mixture was not observed. A 90% yield of a 6-member heterocycle, 11, was the only compound isolated eq 2.4. This compound probably arises from attack of the imine carbon on the lone pair of electrons from the nitrogen, followed by a 1,3 proton shift. Changes in the reaction conditions did not afford any of the desired diamine.

1. acetic acid/H20 NN
2. sodium acetate
3. acetone eq 2.4
4. NaBH4
S / 5. NaOH


2,2: Synthesis of New Ligand Metal Complexes.

Addition of these new ligands to metals was the next step in the project. Two starting materials were initially chosen as trial compounds, WOC14, which was readily available, and W(NPh)C14(OEt2). W(NPh)C14(OEt2) can be prepared in high yield from the addition of PhNCO to WOC14, resulting in the loss of CO2. The ligand chosen for the majority of the work reported was 1,2-(Me3SiNH)2C6H4, 1. There are two simple routes available for the addition of the ligand to the metal center. First, the ligand could be doubly



Figure 2.1. The IHN MR spectrum of the 7-membered heterocycle, 10.


deprotonated and then added to the metal center in a simple metathesis reaction, forming the ligand complex and two equivalents of a chloride salt. The other route would be to add the diamine ligand directly to the metal, losing HC1 either spontaneously or through addition of a base such as NEt3. The first route proved the more successful, and the second route was attempted with minimal success.

An Et2O solution of WOC14 was added to 2 at -78 'C. Work-up afforded a

moderate yield of WOC12(Me3SiN)2C6H4, 13, eq 2.5. Studies of the reaction chemistry of this compound were not undertaken since isolating pure 13 is extremely Me3Si\ Me3Si\ 0
0l NCl Li N, WI C
C1W + Li C f "C eq 2.5

Cl Cl K'~ C1
/ /
Me3Si Me3Si 13

difficult. Altering the reaction conditions did not alleviate this problem. Attempts are currently underway to alter the starting W=O complex and allow cleaner isolation of products, thereby allowing a thorough study of the reactivity of 13a.

In a similar reaction, W(NPh)C14(OEt2) was allowed to react with 2 at -78 'C, which afforded W(NPh)C12(Me3SiN)2C6H4, 14, as an orange-red powder. The bisamido complex, 14, was also prepared on a large scale by deprotonating the diamine in situ. This method proved the most successful in preparing 14 in high yield eq 2.6. With an easy, large scale, high yield synthesis, 14 was an excellent starting point to investigate the reactivity of these chelated complexes. The electron count of the tungsten is formally considered to be 14 electrons, considering the imido as a 6 electron donor with the amide bonds contributing 1 electron each to the total electron count. The electronic donation of

a William Vaughan has undertaken the synthesis of other W=O complexes to be used as precursors for the addition of the ligand, 1. These W=O complexes have 'softer' substituents and would presumably be more tolerant to metathesis.


the bidentate ligand is unclear, the nature of which could make the molecule a 14e-, 16e- or 18e- complex.

Me3Si. Me3Si. Me3Si i
NH, X NN4f.c
CNH2 eq n-BulLi CQ I 1W
H L + /W eq 2.6
/ OEt2 /
Me3Si Me3Si Me3Si 14
2- in situ

X-ray quality crystals were obtained by dissolving 14 in toluene and cooling to -10 'C. The thermal ellipsoid plot of 14 is shown in Figure 2.2, while selected bond lengths and angles are found in Table 2.1. The geometry of the molecule is square pyramidal with the imido nitrogen in the axial position. The W atom lies .58 A above the plane defined by the two amido nitrogens and the two chlorides. This is similar to the crystal structure of W(NPh)C14(OEt2), which is octahedral with the imido cis to all four chlorides. The imido nitrogen bond length is 1.730(10) A, which is well within the range of other imido complexes where the imido group is considered to be a six electron donor because of donation of the lone pair of electrons on the nitrogen to an empty d orbital on the metal center. One feature of the structure of the molecule that is quite surprising is the 'orientation' of the ligand. The phenyl group of the bis-amide ligand is distorted and bent toward the metal center. The dihedral angle between the plane of the C1-C6 ring and W, Ni, N2 is only 1300, as if there were a metal-olefin type interaction between the W and the phenyl ring of the ligand. The complement of this angle is referred to as the 'fold angle', 500 and is quite diagnostic when compared to other compounds. The interaction appears quite significant; the distances between W and Cl and W and C2 are only 2.58(1) A. Although this is greater than the W-C bond length in high oxidation state tungsten alkyl complexes, it is still within the vanderWaal's radii. In the few compounds that are known with this type of ligand, the distances are much greater (>2.80 A).29,37 There are only a


Figure 2.2. Thermal ellipsoid plot W(NPh)CI2[(Me3SiN)2C6H4], 14. The protons and silyl
methyl groups have been omitted for clearity.


few examples of structures of o-phenylenediamido-type ligands.29,37,38 None of these structures appear to have an interaction between the ring carbons and the metal center. If the ring is in fact acting as a two electron donor, the electron count on the metal would now be 16e-.

Table 2.1: Selected Bond Lengths (A) and Angles (0) for compound 14.

1 2 3 1-2 1-2-3

Cll W C12 2.383(4) 82.9(2)
C12 W N1 2.387(4) 150.5(3)
N1 W N2 1.951(11) 83.9(4)
N2 W N3 1.952(11) 110.2(5)
N3 W Cl 1.730(10) 137.4(4)
C1 W C2 2.582(13) 31.9(4)
C2 W Cll 2.582(13) 117.2(3)
N1 Sil C13 1.768(10) 108.4(7)
N2 Si2 C16 1.781(12) 106.0(7)
C1 NI W 1.42(2) 98.8(7)
C2 N2 W 1.40(2) 99.4(8)
C7 N3 W 1.39(2) 166.2(9)

Analogies can be made to some other types of compounds where similar bonding exists. Peterson39 solved the crystal structure of a Cp2Zr chelated bis-amido complex in which the distances between the P-carbons and the zirconium are 2.612(3) A and 2.603(3) A. Peterson claims that this close interaction is due to donation from the filled it-orbital of the C=C bond to the empty dz2 orbital on the zirconium as shown in Figure 2.3. Rothwel140 observed similar results with quite similar bond lengths, in the 2.40-2.60 A range, for other enediamido and enamidolate chelate compounds of zirconium, titanium and tantalum as shown in Figure 2.4. These compounds, including 14, not only have close contact distances, but also have abnormally large 'fold angles'. The compounds prepared by Rothwell and Peterson have fold angles between 35 and 50 0. Rothwell also measured the AG's for the barrier to 'flip' these enediamido metallacycle rings. The AGt's were in the 13-16 kcal/mol range.40 This type of 'flip' would not be observed in a molecule such


as 14 since the molecule does not have a mirror plane through the metal center. Both Rothwell and Peterson also point out the fact that although the M-Cp bond lengths are longer than normal for similar M-R complexes, they are within the range of M-C bond lengths in M-Cp complexes. Typically, W-Cp metal-carbon bond lengths are between

Cp N R
Zr Si Me OAr M = Ti, Zr, Hf
\ Me R = CH3, CH2Ph
Cp Me OAr R' = xy, Ph, tBu
Figure 2.3: Figure 2.4:
A zirconacene enediamido complex. Enediamido complexes of group 4 metals.

2.3 and 2.45 A. This is shorter than the 2.58 A observed for 14, and does not fit as well with the comparisons suggested by Rothwell and Peterson. Lappert has observed similar behavior for some bidentate, di-anionic o-xylidene complexes of some bis-cyclopentadienyl group 4 and 5 metals.41 The metallacycles in these compounds also displayed a significant interaction with the metal center, having fold angles between 410 and 53 o

Continued investigation of the use of these chelating bis-amide ligands led to the

synthesis of other new compounds. There were two goals for preparing new derivatives of these bis-amide derivatives, less solubility and more crystallizability. To this end, 1,2(Me2PhSiNH)2C6H4, 3, was deprotonated in situ and reacted with W(NPh)C14(OEt2) to yield W(NPh)C12(Me2PhSiN)2C6H4, 15. This compound was somewhat less soluble but recrystallization proved unfruitful. Yellow crystals of 1,8-K2(Me3SiN)2C10H4, 16 or 1,8Li2(Me3SiN)2C10H4, 17 were isolated when 1,8-(Me3SiNH)2C10H4, 7, was deprotonated with two equivalents of KH or n-BuLi. These salts react readily with W(NPh)C14(OEt2) to give W(NPh)C12(Me3SiN)Cl10H4, 18, as a dark powder which was markedly less soluble than 13, 14, or 15 eq 2.7. Nonetheless, a suitable solvent could


not be found for recrystallization. The same reaction using 4,5-dimethyl-1,2(Me3SiNH)2C6H2, 6, yielded W(NPh)C12[4,5-Me2-1,2-(Me3SiN)2C6H2], 19. This

Me3Si Me3Si I

N \ / NK \N. I 0,1Cl
Cl C1 N K /" N C eq 2.7
W + Y Cl
Cl t cI
OEt2 Me3Si Me3Si 18

complex certainly simplified the 1H NMR spectrum, but did not show any greater ease in isolation or crystallization. Attempts were made at synthesizing tungsten phenyl imido derivatives using the other ligands mentioned; however, although results were encouraging, full characterization of these derivatives was not obtained.

It was also possible to substitute the chloride atoms in 14 with a more labile

substituent. This was desirable if alkylation of the dichloride proved unsuccessful. When 14 was allowed to react with two equivalents of AgOTf, the bistriflate complex, W(NPh)(OTf)2[(Me3SiN)2C6H4] (OEt2), 20, was isolated as a bright orange powder. This complex must be more electron deficient than 14 since it forms an etherate complex, whereas 14 does not.

Since all these compounds are five-coordinate, electron deficient molecules, they would be expected to form adducts with o-donor ligands. When PMe3 was added to a red Et20 solution of 14, the solution immediately turned purple. Addition of pentane followed by slowly cooling the sample to -10 'C yielded dark purple crystals of W(NPh)C12(PMe3)[1,2-(Me3SiN)2C6H4], 21. Integration of the 1H NMR and combustion analysis confirmed the stoichiometry of 21. Although 21 appears as a discreet mono-adduct, the 31P NMR spectrum shows a very broad singlet for the PMe3 ligand, nearly 250 Hz wide, and suggests that at room temperature an equilibrium between 21 and free PMe3 was established. More evidence will be given for this ligand exchange later.


Purple, crystalline a-adduct complexes were also formed when 14 was exposed to THF, 22; 3-picoline, 23; or CH3CN, 24, eq 2.9. Addition of these a-donors increases the

Me3Si 14 Me3Si

L = PMe3, 21; THF, 22; 3-picoline, 23; CH3CN, 24 formal electron count of the molecules to 16e- with no it-donation from the folding of the ligand at this point. The donation of this electron density, as well as now having a 6coordinate complex, would clearly have an impact on the ligand 'folding' which was observed in the structure of 14. Recrystallization of the PMe3 adduct, 21, by slowly cooling a pentane solution to -10 'C gave purple, x-ray quality crystals.

The structure of 21, shown in Figure 2.9, has some unique features. Selected bond lengths and angles for the structure of 21 can be found in Table 2.2. The PMe3 adds to the molecule trans to imido nitrogen, creating an octahedral geometry, with the amido nitrogens and the chlorides mutually cis in the basal plane. In a comparison between the structures of 14 and 21, one very general thing that stands out. Because of the added electron density and steric bulk of the PMe3, all the bonds to the metal center are longer in 21. For instance, even the chlorides are 0.06 A or more further away. Most significantly, the fold angle has increased from 50 0 to only 28 . The W-Cring distances have increased from 2.58 A each in 14 to 2.79 and 2.78 A in 21. It is interesting to note that the geometry around the nitrogen atoms in the bis-amide ligands in both 14 and 21 is virtually planar. The amide bond lengths in 21 are only slightly longer, 2.010(5) A and 1.990(5) A than in 14, 1.951(11) A and 1.952(11) A. One feature of this structure which is quite


Table 2.2: Selected Bond Lengths (A) and Angles (0) for compound 21.

1 2 3 1-2 1-2-3

Cll W C12 2.449(2) 92.43(7)
Cli W P 75.86(7)
C12 W P 2.443(2) 75.70(7)
P W N1 2.720(2) 87.2(2)
P W N2 89.2(2)
P W N3 160.3(2)
N1 W N2 2.010(5) 80.8(2)
N2 W N3 1.990(5) 105.8(3)
N3 W C1 1.747(6) 124.6(2)
C1 W C2 2.797(6) 29.7(2)
C2 W Cll 2.785(6) 137.4(2)
NI Sil 1.781(6)
Cl N1 W 1.402(8) 108.8(4)
C2 N2 W 1.387(9) 109.8(5)
C7 N3 W 1.388(9) 164.3(5)

unusual is the extremely long W-P bond length, 2.720(2) A. This bond appears to be at

least 0.2 A longer than most W-P bond lengths in W(VI) complexes. This weak

interaction is supported by the afore mentioned broad singlet observed in the 31p NMR

spectrum. This long bond length may be due to trans influence of the imido nitrogen,

which is a strong trans influencing ligand.

A number of interesting new W(VI) imido and oxo complexes have now been

prepared. These new complexes have interesting structural characteristics which will play

an important role in influencing the chemistry associated with them.

Figure 2.9. Thermal ellipsoid plot W(NPh)Cl2(PMe3)((Me3SiN)2C6H4], 21. The protons and silyl methyl groups have been omitted for clearity.


3.1: Background Information on W(VI) Alkyls and Alkylidenes.

In order to pursue the project goal of creating a new olefin metathesis catalyst,

alkylation of the W(VI) dichloride was investigated. There are surprisingly few examples of group(VI) do alkyl complexes in the literature. 1,2,42,43 There are numerous examples of d group (IV) alkyl and dialkyl complexes, most of which are used as Ziegler-Naatta type catalysts. Homoleptic alkyls, such as WMe6 are well-known.42 Other alkyls are less prevalent. Schrock has isolated a number of W(VI) imido alkyl complexes.44 Many of these were isolated in attempts to find precursors for alkylidene complexes. It was observed that W(NPh)C12(CH2CMe3)2 was not isolable. However, if one or two of the chlorides are substituted with t-butoxide ligands, dialkyls can be isolated, W(NPh)Cl(O-tBu)(CH2CMe3)2 and W(NPh)(O-t-Bu)2(CH2CMe3)2. In these compounds, the alkyl groups are oriented cis to one another in a trigonal bipyramidal geometry. There are also examples in which the alkyl groups are trans to one another. Schrauzer prepared a variety of W(VI) dioxo complexes45 of the type W(O)2R2(bipy), where R = Me, Et, n-Pr, and CH2CMe3. These compounds are quite stable due to the chelate effect of the bipyridine ligand, which also serves to 'lock' the alkyls trans to one another, limiting reductive elimination or a-hydrogen abstraction reactions. The bonding in do alkyl complexes is rather straight-forward. The alkyl ligand acts as a 2e- donor ligand forming a a-bond with the metal center. The most common means of preparing alkyl compounds is by simple metathetical exchange reactions.42 Grignard reagents and lithium, zinc, or aluminum alkyls are commonly used as alkylating agents.



There are many possible reactions that can take place when a dO metal is alkylated, prohibiting isolation of a transition metal alkyl complex. If the alkyl group has 3-protons, the metal can undergo a -hydrogen elimination reaction.2,42 Another reaction that takes place, especially with bulky alkyl groups, is a-hydrogen abstraction.2 An alkylidene is formed as a result. There are two mechanisms proposed for this reaction, neither of which has been proven. One is initial a-elimination to form an alkylidene-hydride complex. Evidence for this mechanism comes from the chemistry of later transition metals. This addition is not possible for a do metal center because the reaction involves oxidation of the metal center. The other mechanism proposes a three-center, two-electron transition state which eliminates alkane, forming the alkylidene. Both a-abstraction mechanisms can be seen in Figure 3.1. Since the discovery of the first alkylidene complex, other routes have been discovered for preparing alkylidenes.46


CH2R R ----- H CHR
II s
MCH2R M C H + RCH3 (2)
M -CH2R M M------- C-H IN
Figure 3.1. The two a-abstraction mechanisms for alkylidene formation.

Despite these developments, a-abstraction reactions are still the most common method for preparing alkylidene complexes. Neopentylidene (=CHCMe3) and neophylidene (=CHCMe2Ph) complexes are the most prevalent, due to their steric bulk and lack of [3-protons. Reactions that proceed through a-abstraction can be divided into two general categories, proximal a-abstraction and ligand induced a-abstractiona. Proximal aa Proximal a-abstraction and ligand induced a-abstraction are not commonly used in the literature, however coining these terms is useful for the discussion.


abstraction refers to reactions in which an alkylidene is formed by a-abstraction immediately upon alkylation. This type of reaction was observed for the addition of excess neopentyl grignard to TaC15, forming Ta(CHCMe3)(CH2CMe3)3.13 The mere steric bulk of the neopentyl groups induces ca-abstraction, eliminating neopentane. The second type of reaction, ligand induced a-abstraction, is characterized by the addition of a a-donor ligand, most commonly PMe3, to a cis bis-alkyl complex, inducing a-abstraction and elimination of an alkane. A classical example is the addition of PMe3 to W(NPh)(CH2CMe3)2(PMe3)C12, which results in the formation of the alkylidene complex W(NPh)(CHCMe3)(PMe3)2C12 and neopentane.44 These principles will be discussed as they apply to the synthesis of the new alkylidene complexes that were synthesized during this study.

3.2: Synthesis of W(VI) Bis-Alkyl Complexes.

Most of the reaction chemistry was performed on compound 14, since it was the first compound isolated and was available in large quantities. When 14 was allowed to react with two equivalents of C1MgCH2CMe3 in Et20 at -78 'C, the bis-alkyl complex W(NPh)(CH2CMe3)2[(Me3SiN)2C6H4], 25, was isolated as a dark red crystalline solid eq 3.1. The yield for this reaction was usually about 75-80%, and seems only to be

Me3Si Me3Si I

+ 2 eq. CIMgCH2CMe3 2-eg 3.1 N,-780C ~e .
/ /
Me3Si 14 Me3Si 25

limited by the extreme solubility of 25 in hydrocarbon solvents. There are many interesting characteristics of this compound. The 1H NMR, shown in Figure 3.2,


showed that there was a plane of symmetry in the molecule. The neopentyl methyl groups were equivalent as were the -SiMe3 methyls. The methylene protons were observed as diastereotopic protons at 2.13 ppm and 2.29 ppm. The 2JHH was 10 Hz, while there were 183W satellites observed at 11 Hz from 2JWH. The aromatic region also shows the ligands protons resonating in an AA'BB' spin system, stemming from the symmetry plane in the molecule. These observations are consistent with the square pyramidal structure drawn in eq 3.1 for 25. Since 25 is only a 14e- complex (the degree of interaction of the metal center with the phenyl ring of the bis-amide ligand is not known, so an electron count of 16e- might also be possible), and coordinatively unsaturated, an agostic interaction of one of the neopentyl methylene protons and the metal center is conceivable. It has been established that the magnitude of the coupling of the methylene proton with the methylene carbon is diagnostic of an agostic interaction. If the coupling constant is less than 120 Hz, then an agostic interaction is likely.47 The 1JCH for 25 was 123 Hz, and is consistent with a "normal" metal alkyl a-bond. An interesting point is that 25 was isolable as a cis (bis)-neopentyl complex. Recalling the structure of 14, the steric crowding around the metal center is overwhelming. It would seem likely that the bis-neopentyl complex would undergo a proximal a-hydrogen abstraction upon alkylation. The chelation of the bisamido ligand may hinder the a-abstraction since rearrangement to a tetrahedral, fourcoordinate alkylidene is necessary.

Other bis alkyl complexes can be prepared in an analogous manner. The neophyl derivative W(NPh)(CH2CMe2Ph)2[(Me3SiN)2C6H4], 26, was isolated in 85% yield as a brownish powder which is less soluble in hydrocarbons than compound 25. For the case of the dimethyl derivative, W(NPh)(CH3)2[(Me3SiN)2C6H4], 27, isolation was more difficult. The compound was extremely soluble in pentane, and could only be isolated by cooling a concentrated solution of 27 in Et2O (1.2 grams in 2 ml) to -78' C, which gave a red solid after several days. The 1H NMR spectrum of 27 is shown in Figure 3.3 and clearly shows the 3:1 ratio of -SiMe3 to W-Me2 peaks. Coupling of 183W to the methyl

N,, f

Me3Si 25

2. 51 2.601 2.25 2.201 2.5 2. 10 PPM. 2.(

Figure 3.2. The IH NMR Of W(Nph)(CH2CMe3)2[(Me3SiN)2C6H-4], 25.


protons was also observed and gives rise to the satellites of the methyl peak with 2JWH = 6 Hz. The 1JC.H for 27 was 123 Hz as well, indicating that an agostic interaction is unlikely. When 14 was treated with two equivalents of ClMgEt, a red oil was isolated. The red oil appeared by NMR to be pure W(NPh)(CH2CH3)2[(Me3SiN)2C6H4], 28. The 1H NMR spectrum revealed methylene resonances as multiplets at 1.91 and 2.31 ppm, while the ethyl-CH3 protons resonated as a singlet at 1.86 ppm. The 1JC-H for 28 was 120 Hz, consistent with the other bis-alkyls isolated. It was interesting that an electron deficient bis-alkyl complex with 13-protons was isolated. Often times these types of alkyls decompose by 13-hydrogen elimination making them difficult to isolate. The dibenzyl complex, W(NPh)(CH2Ph)2[(Me3SiN)2C6H4], 29, was prepared using benzyl grignard as the alkylating agent. The 1H NMR was interesting because the benzyl methylene protons did not appear to be diastereotopic, as were the methylene protons in 25, 26 and 28. The methylene protons resonated as a singlet at 2.78 ppm and integrate 2:9 to the

-SiMe3 peak. However, when 29 was heated to 80 'C in C7D8, the resonance becomes what appeared to be a triplet. Cooling the sample only broadened the singlet. The methylene protons are considered diastereotopic, yet coincidentally have identical chemical shifts, obscuring any coupling.

The synthesis of other bis-alkyl complexes has been investigated and show

promise, although most of the products have not been completely characterized. Reacting two equivalents of allyl magnesium chloride with 14 gave a mixture of compounds. It appeared as though the major product was a bis-allyl complex where one allyl group was bound in an ri1 manner while the other was 13. More characterization is necessary to determine the identity of the compounds. The substitution of the chlorides on 14 with aryl groups was also investigated. A bright red powder was isolated when 14 is allowed to react with two equivalents of PhLi. The 1H NMR confirmed the identity of the compound as W(NPh)Ph2[(Me3SiN)2C6H4], 30. This chemistry is currently being investigated by other members of the Boncella research group. Since the reactivity of 14 has shown so



me3sl 27

1. is 1 14 1. 13 1. 12 1. 11 1. 10 1.01) 1. OA

7 6 4 3 2

Figure 3.3. The IH NMR of W(NPh)(CH3)21(Me3SiN)2C6H4], 27.


much promise, alkylation of other dichloride derivatives was investigated. Allowing W(O)C12[(Me3SiN)2C6H4], 13, to react with two equivalents of neopentyl grignard does not afford isolation of the bis-neopentyl complex as expected. However, the 1H NMR spectrum of the brown solid shows resonances similar to the diastereotopic methylene protons of 25. This result was not unexpected since the chemistry of the oxo complex was consistently less clean than the imido compounds. A similar result was observed when W(NPh)C12(Me3SiN)2CO10H4, 18, was allowed to react with neopentyl grignard. Full characterization was not achieved but the spectral data were consistent with a bis-neopentyl complex. Better results were achieved when W(NPh)C12(Me2PhSiN)2C6H4, 15, and W(NPh)C12[4,5-Me2-1,2-(Me3SiN)2C6H2], 19, were allowed to react with neopentyl grignard. The two new bis-neopentyl complexes, W(NPh)(CH2CMe3)2(Me2PhSiN)2C6H4, 31 and W(NPh)(CH2CMe3)2[4,5-Me2-1,2(Me3SiN)2C6H2], 32, were isolated as dark red solids and have spectral properties that are similar to 25.
Heretofore, only bis-alkyl or bis-aryl compounds were isolated. However, adding only one equivalent of alkyl to the metal center should also be possible. When one equivalent of neopentyl grignard was allowed to react with 14, the mono-neopentyl chloride complex W(NPh)Cl(CH2CMe3)(Me3SiN)2C6H4, 33 was isolated as a red powder eq 3.2. The methylene protons of the neopentyl groups were observed as

'II...,,Cl 3Et2i

W + 1 eq. C1MgCH2CMe3 -78 0C K jN Cl -78oC N C1 eq 3.2
N Cl N
I /
Me3Si 14 Me3Si 33

diastereotopic doublets at 1.93 and 2.08 ppm respectively (2JH-H = 10 Hz). The 1H NMR spectrum of the mono-neopentyl complex, 33, differs from the bis-alkyls due to the


absence of a plane of symmetry. The -SiMe3 resonances were observed as inequivalent singlets, whereas they were equivalent in the bis alkyl complexes. In the aryl region, an AA'BB' spin system was no longer observed for the bis-amide ligand, each ligand aryl proton was inequivalent (two doublets and two triplets).

There are few examples of mixed alkyl complexes, so 33 would be a prime candidate to allow asymmetric substitution. When 33 was allowed to react with one equivalent of MeLi in an NMR tube, W(NPh)(CH3)(CH2CMe3)(Me3SiN)2C6H4 was generated. The 1H NMR showed four singlets in the alkyl region, in a 3:3:3:1 ratio, corresponding to the two-SiMe3 groups, the neopentyl group, and the methyl group. Diastereotopic methylene protons are also observed. This compound was not isolated on a preparatory scale. Another asymmetric compound was isolated when 33 was allowed to react with one equivalent of LiNMe2, forming W(NPh)(CH2CMe3)(NMe2)(Me3SiN)2C6H4, 34, eq 3.3.

Me3Si i Me3Si I

C1aN Me
/ / Me
Me3Si 33 Me3Si 34

The 1H NMR spectrum of 34 reveals diastereotopic methylene protons as doublets at 1.33 and 2.60 ppm respectively. There is lack of rotation about the W-NMe2 bond, since two singlets were observed at 3.19 and 3.68 ppm. This lack of rotation is due to the 7rdonation of the lone pair of electrons on the amide nitrogen to the metal center. This type of interaction is quite common in electron deficient molecules such as 34. Taking into account the electronic donation from the bis-amide ligand, this molecule should be considered as being electronically saturated.


3.3: Alkylidene Formation From Bis-Alkyl Complexes.

In Section 3.1, the two mechanisms of alkylidene formation from bis-alkyl complexes were discussed. The fact that bis-alkyl complexes were isolated does not eliminate proximal a-abstraction as a route to alkylidene formation. Without adding a adonor ligand, a-abstraction could be thermally induced. Although there are no examples of thermally induced a-abstraction reactions, this route should be viable. Many proximal aabstraction reactions are performed at reduced temperatures, and warmed to room temperatures where a-abstraction takes place. Since the bis-alkyl complexes are thought to exist at low temperatures in these reactions, a-abstraction occurs at or near room temperature. Heating a C6D6 solution of 25 in a sealed NMR tube gave a very interesting result. Neopentane formation was observed in the 1H NMR, however, it was not a result of a-hydrogen abstraction. The 1H NMR spectrum reveals what appears to be a metallacyclobutane complex formed from the y-abstraction of a proton from a neopentyl methyl group eq. 3.4. The 1H NMR spectrum of the P,D'-dimethylmetallacyclobutane complex, W(NPh)(CH2CMe2CH2)(Me3SiN)2C6H4, 35, is consistent with other

Me3Si\ Me3Si

W 2-3 aCys + CMe4 eq3.4
v / 2-3 days N

Me3Si 25 Me3Si 35

metallacyclobutane complexes in the literature.48 There are two doublets, at -1.7 and 2.1 ppm respectively (2JH-H = 9 Hz), corresponding to the methylene protons of the metallacycle. The large difference in chemical shift of the protons is consistent with other metallacycles.48 The methyl protons of the metallacycle resonated at 0.54 and 0.57 ppm respectively. They were inequivalent since they lie above and below the plane of the


metallacycle. This compound has not been fully characterized since it was difficult to isolate and purify. The difficulty in isolating the metallacycle results from the 2-3 days of heating which were required for the complete thermolysis of 25 to 35. During this time, metallacycle formed earlier in the reaction started to decompose to unidentifiable products. However, 35 is formed by other routes and will be discussed at length later in Chapter Four. It seems unusual that 25 would undergo a y-abstraction, however there are examples of this type of reaction in the literature. For example, Marks49 thermally decomposes Cp*2Th(CH2CMe3)2 and observes the loss of neopentane, giving the 13,'dimethylcyclobutane thorium complex, Cp*2Th(CH2C(Me)2CH2).

Further investigation into this reaction using other bis-alkyls proved to be

unfruitful. The bis-neophyl complex, 26, was interesting since it possesses at least three thermal decomposition routes that involve proton abstraction reactions. These include: ahydrogen abstraction, forming a neophylidene; y-abstraction similar to 25 to give a Pmethyl, 1'-phenyl metallacyclobutane; or ortho-metallation, forming a five-membered metallacycle. Examples of each of these types of reactions have been reported in the literature. The a-abstraction and y-abstraction reactions have already been discussed. Orthometallation reactions are observed for compounds in which the abstraction of a proton from the ortho position of an aryl group forms a five-membered ring.2,50 A good example is the thermolysis of CH3Rh(PPh3)3, shown in Figure 3.4, in which the ortho proton of Ph

A ph, P Rh[P(C6H5)312
[(C6H5)P]3RhCH3 A Ph- bRh[P(C6H53]2

Figure 3.4. Orthometallation of (Ph3P)3RhMe

one of the phenyl groups is abstracted, releasing methane and forming a metallacycle.51 However, when 26 was thermally decomposed at 90 'C for 3 days, no discernible products were observed. Interestingly, the 1-methyl, '-phenyl-metallacyclobutane


complex will be isolated and discussed by an alternate route in Chapter 4. Similarly, only complete decomposition to indiscernible products was observed when the dibenzyl and diethyl compounds were heated. The dimethyl complex was unique in that it was thermally stable over a period of 10 days at 90 'C.

Since no thermal pathways seemed to lead to alkylidene formation, ligand-induced ct-abstraction was attempted. In a sealable NMR tube, one equivalent of PMe3 was added to a C6D6 solution of the bis-neopentyl complex, 25. Over a period of ten days at room temperature, no reaction was observed by 1H NMR. Interestingly enough, no peak shifts were observed corresponding to coordination of PMe3. Even upon cooling a C7D8 solution of 25 and PMe3 gives no evidence of ligand coordination by observation of the 1H NMR. This behavior was contradictory to the behavior of the dichloride precursor, 14, which forms adducts with c-donors quite readily. The bis-neopentyl complex, 25, is isoelectronic with 14, but the steric environment at the metal center in 25 would obviously be much more hindered. The tube was then warmed to 70 'C in an oil bath. After only three hours, neopentane formation was observed by 1H NMR. Additionally, a doublet began to grow in at 9.62 ppm, indicative of an alkylidene proton coupled to a bound phosphine. The reaction proceeded nearly to completion at this temperature in 3 days; however, it was found that the reaction proceeds faster and much cleaner with two or three equivalents of phosphine present and a temperature of ca. 90 'C, eq 3.5.

Me3Si~ N Me3Si\H

I,, + 3 PMe3 , P +CMe4 eq3.5
Me3Si 25 Me3Si/ 36

In order to prepare and characterize this new alkylidene on a preparatory scale, a problem had to be overcome. In an open system, PMe3, which has a high vapor pressure and boils


at 39 'C, would be lost when the reaction was heated to 90 'C. Additionally, heating a solution in a closed system using conventional Schlenkware would be dangerous due to a possible buildup of pressure. This problem was overcome by dissolving 25 in toluene in a tube fitted with a Young's joint with a Teflon seal. Three equivalents of PMe3 were added and the reaction was heated to 90 'C for 12 hours. The color of the solution changed from dark green to bright orange over the time of the reaction. After cooling to room temperature, the solution was transferred to a Schlenk flask where the solvent was removed. Extracting with pentane and cooling to -10 0 C gave orange crystals of the new alkylidene complex, W(NPh)(CHCMe3)(PMe3)[(Me3SiN)2C6H4], 36.

In the 1H NMR of 36, a doublet was observed at 9.62 ppm (3JP-H = 4 Hz)

corresponding to the alkylidene proton. Tungsten satellites were also observed for this resonance (2JW.H = 11 Hz). The alkylidene carbon was observed at 277.4 ppm (1Jc-H4 = 110 Hz). This coupling is consistent with an agostic interaction between the alkylidene proton and the metal center. The PMe3 resonance was observed as a doublet at 0.98 ppm (1JpH = 9 Hz) in the 1H NMR spectrum. There are three other singlets in the alkyl region of the 1H NMR spectrum of 36, all in a 1:1:1 ratio. One is the neopentyl methyl group,

1.39 ppm, while the other two are the silyl methyl groups, 0.38 and 0.41 ppm. The aromatic region also confirms that the molecule no longer has a plane of symmetry which would make the silyl methyls equivalent.

The geometry of the five-coordinate alkylidene complex, 36, cannot be deduced merely from the spectroscopic data. Since this was a five-coordinate complex, there were numerous square pyramidal and trigonal bipyramidal complexes which would have fit the spectral data. Therefore determination of the crystallographic structure was essential, not only to determine the geometry of the molecule, but also to gain insight into the catalytic activity of the molecule which will be discussed later in this chapter. Slowly cooling a pentane solution of 36 to -10 'C afforded single orange crystals which were suitable for diffraction.


The thermal ellipsoid plot of 36 is shown in Figure 3.5. Selected bond lengths and angles can be found in Table 3.1. Upon examining other group (VI) alkylidene complexes,8 it was found that the structure of 36 was very unique. The geometry was square planar with the alkylidene in the apical position. The tungsten atom lies 0.61 A above the square plane defined by the imido nitrogen, the PMe3 phosphorous atom, and the two amide nitrogens. The average deviation of N1, N2, N3, and P1 from the square plane was only 0.03 A. Re-examining the issue of the interaction between metal center and the x-system of the bis-amide ligand gives an expected result. Since the electron count of the molecule has been increased by two by the addition of the ligand, a decrease in the foldangle would be expected. The fold angle was 40 0, a 10 0 increase from the dichloride structure, 14, yet 12 o less than the fold angle in the six-coordinate PMe3 adduct of the dichloride, 21. The 40 0 fold angle suggests a weak interaction between the aryl ring of the bis-amide ligand and the metal center. The angle may be due, in part, to the great steric bulk around the metal center. Thus, the folding of the bis-amide ligand may be necessary to relieve steric interactions.

It was apparent that the chelating nature of the ligand has dictated the observed geometry of the molecule. The literature reveals that five-coordinate imido-alkylidene complexes of tungsten and molybdenum prefer to adopt trigonal bipyramidal, rather than square pyramidal structures. The compounds, anti-W(trans--CHCH=CHMe)[N-2,6C6H3(iPr)2] (OCMe(CF3)2]2(quinuclidine)52 and W(CHCH=CHPh2)[N-2,6C6H3(ipr)2](OCMe(CF3)232[P(OMe)3]53 are good examples of the preferred trigonal bipyramidal structure of these types of compounds Figure 3.6. Although 36 and these two alkylidenes have similar substituents, the geometry's are much different. The geometric constraints of the chelating bis-amide ligand must be responsible for the



Si3 S12
V C25 C26

N3 N2
C11 C24


Figure 3.5. The thermal ellipsoid plot of W(NPh)(CHCMe3)(PMe3)[(Me3SiN)2C6H4], 36. The protons and the silyl methyls have been omitted for clarity.


Table 3.1: Bond Lengths (A) and Angles (0) for the non-H atoms of compound 36.

1 2 3 1-2 1-2-3

P1 W NI 2.502(4) 82.8(4)
P1 W N2 148.1(3)
P1 W N3 81.2(3)
NI W N2 1.789(9) 98.4(4)
NI W N3 140.1(4)
N1 W C13 103.6(5)
N2 W N3 2.095(10) 77.8(4)
N2 W C13 113.0(5)
N3 W C13 2.067(10) 114.6(4)
N2 Si2 C18 1.736(11) 108.6(6)
N3 Si3 C21 1.761(9) 111.8(6)
C1 N1 W 1.387(14) 160.8(9)
C7 N2 W 1.39(2) 106.6(8)
W N2 Si2 129.8(5)
C12 N3 W 1.40(2) 105.6(7)
C8 C7 N2 128.6(13)
C12 C7 N2 1.43(2) 114.6(11)
C14 C13 W 1.50(2) 148.4(9)
C13 W P1 1.884(13) 97.5(4)

unique geometry of 36. The bis-amide ligand of 36 should be able to coordinate in a axialequatorial ligation, allowing the molecule to adopt a trigonal bipyramidal geometry, but it is

not observed.

r Ph N

P(OMe)3 V

Figure 3.6. Examples of trigonal bipyramidal alkylidenes.

Although the solid state structure of the compound has been solved, the structure in

solution was actually the key to the reactivity of the molecule. When the alkylidene proton

was irradiated in an nOe experiment, a 6% enhancement was observed for both the silyl


methyl groups. No significant enhancement was observed for the imido aryl protons, or the PMe3 methyls. When the neopentyl methyls were irradiated, the ortho-arylimido protons and the PMe3 protons where enhanced by 2.5% and 3.7% respectively, while no significant enhancement was observed for the silyl methyl groups. This geometry will be referred to as syn, where the neopentylidene t-butyl group is syn to the imido group.

There are two reactions which should be considered at this time to better understand the role of the PMe3 in the formation of the alkylidene. First when Cu(I)Cl was added to a C6D6 solution of the alkylidene, 36, formation of the metallacyclobutane complex, 35, was observed by 1H NMR eq 3.6. Cu(I)Cl forms an insoluble adduct with PMe3 and effectively removes it from the solution. The four-coordinate alkylidene then undergoes a rearrangement to the metallacycle. The rearrangement is effectively a 1,3 shift of a proton

Me3Si H, MezSi

Ni Pe3 + CU(I)C1 -6D WV + CuCl(PMez)x eq 3.6

Me3Si 36 Me3Si' / 35
Mzi 35

from a y-methyl to the alkylidene carbon, as well as forming the new W-C bond. Although the nature of this rearrangement is not known, the result is undeniable and will be discussed further in Chapter 4. Secondly, when PMe3 was added to a C6D6 solution of the metallacycle, 35, complete conversion to the alkylidene, 36, was observed by 1H NMR eq 3.7. These two reactions show the relationship between the alkylidene, 36, and the metallacycle, 35.

Insight into the mechanism by which the alkylidene, 36, is formed from the bisneopentyl complex, 25, eq 3.5, is gained by these reactions. The thermolysis of 25 in the absence of PMe3 might initially give a four-coordinate alkylidene, which quickly rearranges to give the five-coordinate metallacycle, 35. In the presence of PMe3, the fourcoordinate alkylidene is 'trapped' by the phosphine, forming 36. The intermediate four-


Me3Si Me3Si\ H'C

+ PMe3 C6D6 P eq 3.7
K2(N (N \PMe3

Me3Si 35 Me3Si 36

coordinate alkylidene could be tetrahedral, and coordination of the PMe3 would result in rearrangement to the observed geometry of 36, with the alkylidene in the apical position. This mechanism can be seen in Figure 3.7. Regardless of the actual mechanism, the observed net result was still a-abstraction induced by PMe3.

Me3Si /
Me3Si 25 Me3Si 35

I PMe3

MeSi Hj H'C

N4 N

N PMe3
Me3Si 36
Figure 3.7. Scheme for alkylidene and metallacycle formation showing an intermediate tetrahedral alkylidene. Although the application of the PMe3 ligand to induce an a-abstraction reaction might be widely applied to the other bis-alkyl compounds isolated, the bis-neopentyl appears to be a unique case. The neophylidene compound can be isolated, however, a


longer reaction time is necessary. When the dimethyl complex, 27, was heated to 90 oC in the presence of excess PMe3, no reaction was observed over 10 days, not even decompostion of the dimethyl complex. Once again, it was astonishing that no PMe3 adduct formation was observed. Adduct formation would be expected since the methyl substituents are not much bigger than the chlorides in 14. When the diethyl complex, 28, was heated to 90 o C with PMe3, decomposition to indistinguishable products was observed. Surprisingly, when the dibenzyl complex, 29, was heated with excess PMe3 the compound appeared quite stable. No benzylidene formation was observed over time in the 1H NMR, although after a week at 90 C, decomposition occurred. Though disheartening, it is not uncommon to see a unique behavior when dealing with neopentyl and neophyl compounds. The first as well as the most common examples of alkylidenes are either neopentyl or neophyl.
The nature of the a-donor ligand necessary to induce ox-abstraction was also investigated. It would be advantageous to be able to prepare alkylidenes from the bisneopentyl compound using the weakest a-donors available. This would facilitate easy removal of them either to isolate a four-coordinate alkylidene or the in situ dissociation of the ligand forming a transient four-coordinate intermediate. When the bis-neopentyl complex, 25, was heated in the presence of PEt3 in a sealable NMR tube, the alkylidene complex W(NPh)(CHCMe3)(PEt3)[(Me3SiN)2C6H4], 37, was observed. However, when trying to isolate 37 on a preparatory scale, a mixture of the alkylidene and the metallacyclobutane complex was isolated. The lack of phosphorous coupling to the alkylidene proton in the 1H NMR of 37, which was observed as a broad singlet at 9.82 ppm, was evidence of the weak coordination of the PEt3 ligand. Weak coordination was also evident from the 31P NMR spectrum. The PEt3 was observed as a broad singlet at

-11.42 ppm; no 183W satellites were observed as was the case with the PMe3 adduct of the alkylidene. As expected, when PMe3 was added to a C6D6 solution of 37, immediate formation of the PMe3 adduct, 36, was observed by 1H NMR. Using other phosphines


did not lead to the isolation of new alkylidenes, only decomposition or formation of the metallacyclobutane complex was observed by 1H NMR. These reactions were attempted using PMePh2, PCy3, and PPh3. When quinuclidine, N(CH2CH2)3CH, was heated in a C6D6 solution of the bis-neopentyl product, the metallacycle was the only product observed. This was curious since quinuclidine serves as a a-donor for other electron deficient organometallic compounds, including a number of alkylidenes.52 Only decomposition was observed when THF or DME were utilized as the a-donor ligands.

One property of the new alkylidene which was promising was the thermal stability of the molecule. A NMR tube containing a C7D8 solution of the PMe3 adduct, 36, can be heated to 90 'C over days with no observed decomposition in the 1H NMR spectrum. Even though more alkylidenes could not be prepared, the isolation of this new alkylidene offers a starting point for a study into the metathesis activity of this chelated alkylidene. Although it would have been desirable to be able to prepare more derivatives of alkylidenes using this new o-phenylenediamine ligand system it is not unusual to see this type of reactivity. The formation of alkylidenes by means of a-abstraction of an alkyl proton is a delicate reaction. There are many factors which influence the reaction products. The sterics of the ancillary ligands and bis alkyl groups play important roles, as well as the electronics of the ancillary ligands. These factors play important roles not only in the aabstraction reaction, but also in the stability of the alkylidene itself. There are many decomposition pathways available for alkylidenes, and the balance of steric and electronic factors of all the substituents must be carefully controlled for isolation of a stable alkylidene.54

3.4. Metathesis activity of W(NPh)(CHCMe3)(PMe3)[(MegSiN)C6H41., 36.

Before examining the metathesis reactivity of the alkylidene, the mechanism of this reaction should be considered as well as how it applies to the known structure of 36.


Schrock has done extensive studies8 which have concluded that in olefin metathesis reactions the olefin prefers to attack the C-N-O face in the alkoxide alkylidenes. This would translate to the C-Nimido-Namido face in the chelated alkylidene. The open coordination site in 36 is trans to the alkylidene carbon, where, if the olefin coordinated, rearrangement of a six-coordinate complex would be necessary. The most likely mechanism for olefin metathesis with this new alkylidene involves prior dissociation of the PMe3 ligand, followed by attack of the olefin on the vacant coordination site at the CNimido-Namido face of the molecule. Support for this mechanism will be offered throughout the discussion.

When a pentane solution of W(NPh)(CHCMe3)(PMe3)[(Me3SiN)2C6H4], 36, was refluxed with one equivalent of diphenylacetylene, a metallacyclobutene complex, W(NPh)[C(H)(t-Bu)C(Ph)C(Ph)] [(Me3SiN)2C6H4], 38, was isolated as a red oil eq 3.8. In the 1H NMR spectrum a singlet was observed at 2.82 ppm (J2w-H = 8 Hz)

Me3Si Me3Si H
C Ph M N
WNe + pentane i Ph eq 3.8
N PMe3 Ph / Ph

Me3Si 36 Me3Si 38

corresponding to the a-proton of the metallacyclobutene complex. Metallacyclobutene complexes of tungsten are quite common, arising from the metathesis of alkylidenes with diphenylacetylene.55 Although the reaction took place under relatively mild conditions by heating to 35 'C, it is important to note that the reaction must be carried out in an open system. A different result is observed if the phosphine is not liberated from the reaction mixture. In a sealable NMR tube, a C6D6 solution of 36 and diphenylacetylene was warmed to 50 'C for 3 days. The 1H NMR reveals the formation of 38 within 10 hours, however, after 3 days, a singlet was apparent at 5.50 ppm, as well as a much smaller


singlet at 5.45 ppm. A broad singlet was observed at 0.90 ppm, presumably a new PMe3 resonance. The new product was apparently a vinyl alkylidene, arising from the PMe3 induced ring opening of the metallacycle, eq 3.9.

Me3Si 1 H Me3Si
S N eS N PMe3
N,,... II PMe3 N II/ Ph eq 3.9
W / Ph 50e 0C N"N H

/ Ph /P
Me3Si 38 Me3Si

The observation of two new olefmic resonances was most likely due to formation of both cis and trans isomers of the vinyl alkylidene. This reaction has been observed for other metallacyclobutene complexes when a-donor ligands are added.55

The ring opening metathesis polymerization (ROMP) of norbornene, NBE, was chosen in order to investigate the olefin metathesis activity of 36. The alkylidene, 36, polymerized 25 equivalents of NBE in 10 minutes. Analysis of the polymer using GPC techniques reveals a very high molecular weight for the polynorbomene. The molecular weight was determined to be 61,000 g/mol versus a polystyrene standard. However, Schrock and Grubbs have determined that a conversion factor of 2.2 is appropriate for polynorbomene versus polystyrene.17 This would make the corrected Mn = 27,000 g/mol. This was still enormous compared to the molecular weight which would be expected if all the catalyst was active. If 100% of the catalyst was initiated and propagating, the molecular weight should be near 2500 g/mol. This result showed that less than 10% of the catalyst was active. This result was consistent with the prediction that phosphine must be lost in order for the olefin to coordinate to the metal center. Also, in order to observe such a high molecular weight, the rate of the propagation step must be much faster than the rate of the initiation step. Figure 3.8 demonstrates this as well as the dependence the reaction would have on the presence of PMe3 in the reaction mixture.


Grubbs studied the effect PMe3 had on the ROMP of cyclobutene56 by Schrock's catalyst, W(NAr)(CH-t-Bu)(O-t-Bu)2, Ar = 2,6-diisopropylphenyl. The observations made by Grubbs were consistent with the ROMP of NBE by the catalyst, 36. In order to observe the effect of added PMe3, 36 was dissolved in toluene in the presence of ten equivalents of PMe3. A toluene solution of 150 equivalents of NBE was then added and the mixture was stirred for one hour. Precipitation with methanol gave a 80% yield of polynorbornene. The lack of consumption of the monomer, even over the longer reaction time, demonstrates the effect of PMe3 on the rate of the reaction. Analysis of the polymer using GPC techniques versus a polystyrene standard gave a corrected Mn = 35,000 g/mol. The theoretical Mn, 14,124 g/mol, was roughly 50% of the observed Mn, compared to less

kFK [W + PMe3

PMe3 A B

WF/X + f6ki (2[W

+ PMe3 [(3)
C PMe3 D

[W + [W (4)
Figure 3.8. Kinetic scheme for the polymerization of norbornene by 36.

than 10% for the uninhibited polymerization reactions. All of the GPC data for the NBE polymerization reactions can be found in Table 3.2. The percentage of active catalyst has been increased dramatically by the addition of PMe3. It is important to note how the phosphine is affecting the polymerization. The added PMe3 must be hindering propagation to a much greater extent than it is hindering initiation. Grubbs measured the Keq of PMe3


binding to both the uninitiated catalyst, A, and the propagating alkylidene, D.56 It was observed that the PMe3 binds much more strongly to the propagating alkylidene than the uninitiated catalyst. This large difference in binding energy virtually stalls the propagation of the polynorbornene, allowing much more alkylidene to initiate. For the polymerization of NBE using 36 as the catalyst in the presence of PMe3, the reaction was slowed to such an extent that after an hour, the reaction only reached 80% completion. But, judging from the Mn of the polynorbene polymers formed, the percentage of active catalyst was still well below 100%.
The percentage of active catalyst was normally observed to be between 40% and

60% when ten equivalents of PMe3 were used. So, in the inhibited polymerization of NBE using 36, although the added phosphine makes k3 >> k-1, k-1 was still much larger than k1. If k_1 > k1, then the observed percentage of active catalyst was understandable. At room temperature, an equilibrium was observed between A and B for Grubbs' system, with the equilibrium favoring B.56 At room temperature, no equilibrium was observed for 36, the PMe3 was tightly bound with distinctive 183W satellites (1Jwp = 128 Hz).

The nature of the propagating alkylidene was also observed in an NMR tube

reaction. Five equivalents of PMe3 and five equivalents of NBE were allowed to react with 36 in C6D6. After one hour the 1H NMR spectrum of the reaction revealed that only 80% of the NBE was consumed. Also, ca. 40% of 36 was converted into propagating alkylidene. Two broad resonances were observed at 9.09 and 9.31 ppm. These peaks correspond to the syn and anti isomers of the propagating alkylidene species. 17,56

One property that was not mentioned about the poly(norbornene) polymers

catalyzed by 36 was the molecular weight distribution. The polydispersity index, PDI, is a measure of distribution of the molecular weight of the polymer chains and is calculated by Mw/Mn where Mw is the weight-averaged molecular weight and Mn is the number-average molecular weight. All of the polymers which were prepared displayed very narrow polydispersities, regardless of whether or not the polymerizations were inhibited by PMe3.


A comprehensive theoretical study was done by Gold57 in order to understand the relationship between the observed PDI values and the relative rates of propagation and initiation, kp/ki. This study, simplified by others,58 allows a determination of the theoretical Mw/Mn based on kp/ki. The formula takes into account the initial concentrations of catalyst and monomer, and the concentrations of catalyst and monomer at any given time.

From NMR experiments, after 30 minutes the uninhibited NBE polymerization has 10% active catalyst with 40% of the monomer consumed. This gives a relative kp/ki of 115. This was extremely fast although the high molecular weights observed agree with this calculation. In a similar experiment inhibited by 5 equivalents of PMe3, the relative kp/ki was observed to be 30. The effect of the PMe3 can be seen on the relative rate, however, a kp/ki of less than 1 would serve to initiate all the catalyst and give better control of the molecular weight of the polymerization. From Gold's calculations,57 relative kp/ki's of this magnitude should give PDI values between 1.005 and 1.12.

Table 3.2. Polymerization of NBE using 36, inhibited vs. uninhibited.

Inhibited with 10 equivalents of PMe3
eq's of NBE time Mn/ correcteda PDI Mn/ theoretical % yield
150 1 hr 35,000 g/mol 1.01 14,124 g/mol 80
1000 1 hr 104,000 g/mol 1.00 94,217 g/mol 65
163 3 hrs 94,000 g/mol 1.07 15,357 g/mol 85
70 4 hrs 126,000 g/mol 1.07 6,592 g/mol 88
25 5 min 28,000 g/mol 1.12 2,354 g/mol 90
100b 5 min 37,000 g/mol 1.07 9,517 g/mol 90
342 10 min 120,000 g/mol 1.03 32,202 g/mol 90
52 1 hr 115,000 g/mol 1.04 4,888 g/mol 95
a. A correction factor of 2.2 was applied for polynorbornene vs. polystyrene.
b. The polymer was end-capped with benzaldehyde after 5 minutes.


The catalytic reactivity of 36 has been examined for other systems as well. The alkylidene, 36, catalyzes the polymerization of cyclooctene, however, only at elevated temperatures. This supports the mechanism of dissociation of PMe3 before the olefin can attack. The ROMP of cyclooctene has a higher activation barrier than NBE since there was much less ring strain to be relieved and the NBE is much smaller, allowing NBE monomer to coordinate more readily to an open coordination site of the catalyst. The ADMET polymerization of 1,9 decadiene has also been investigated. It was already discussed that 36 is stable to thermal decomposition in refluxing toluene, therefore, it should be a candidate for a thermally stable ADMET catalyst. Observations did not agree with this assertion. The alkylidene does catalyze the ADMET oligomerization of 1,9-decadiene, however, the reaction never proceeded past dimers and trimers. It has been suggested that 36 has a preference for internal olefins, causing unproductive metathesis, however this or any other explanation has not been substantiated.

One aspect of the ADMET reaction which was observed was the poisoning of the catalyst by the ethylene produced in the reaction. When ethylene was bubbled into a C6D6 solution of 36, the 1H NMR initially reveals what appears to be a metallacyclobutane complex with an ct-t-butyl group. There were multiplets at 2.40 ppm, 2.62 ppm, and 3.76 ppm in a 1:1:1 ratio. A new singlet at 1.20 was also observed in a 9:1 ratio with the singlets. Over time these resonances diminished as resonances grew in which correspond to a unsubstituted metallacyclobutane complex. The 1H NMR reveals multiplets at 1.80 ppm, 1.91 ppm, 2.20 ppm, 2.58 ppm and 2.75 ppm in a 1:1:1:2:1 ratio. Figure 3.9 shows the scheme by which the c-t-butyl metallacyclobutane is formed and then further reacts to give the unsubstituted metallacyclobutane complex. Although there was no other spectral data to support this mechanism, this appears analogous to the behavior displayed by Schrock's catalyst.48,59 The peak positions of both metallacycles correspond to observations made for the alkylidene, W(NAr)(CH-t-Bu)(OR)2 and ethylene.59 Examination of the room temperature 1H NMR over the course of the reaction did not


reveal the intermediate methylidene complex, which must be formed before the unsubstituted metallacycle can be formed.

[W] [W]

+ PMe3

[w] [w] < H
-PMe3 H
H H PMe3
Figure 3.9. Proposed mechanism for formation of an unsubstituted
metallacycle from the reaction of 36 and excess ethylene.

Recently, the applications of ADMET have been expanded to the depolymerization of unsaturated polymers. Wagener has utilized Schrock's catalyst in order to depolymerize 1,4-polybutadiene, end-capped with a silylene.60 Although 36 proved unsuccessful for the polymerization of 1,9-decadiene, possibly because of a preference for internal olefins, its activity as a depolymerization catalyst was investigated. 500 equivalents of 1,4polybutadiene was dissolved in minimal toluene in the presence of 36. After 24 hours, 500 equivalents of the end-capping group were added. GPC analysis revealed that the depolymerization was successful. Only monomer and dimer were present. Currently, further work is being done on this reaction.

In this chapter, the substitution of the chlorides in 14 led to the isolation of

numerous new alkyl complexes. The chemistry of the bis-neopentyl derivative, 25, led to the isolation of a new metallacycle and new alkylidene complex. The metathesis reactivity of the alkylidene, 36, was investigated and shown to be inhibited by the addition of PMe3.


The reactivity of the bis-neopentyl derivative, 25, involving molecular hydrogen will be the focus of Chapter 4, and will draw on many of the same principles discussed in this chapter.


4.1. High Oxidation State Transition Metal Hydride Complexes

The synthesis of high oxidation state transition metal hydride and polyhydride

complexes is a highly active area of chemical research.61,62 Complexes containing M-H bonds have long been known to be important intermediates in a plethora of catalytic processes.62,63 The hydrogenation of olefins, both catalytic and stoichiometric, is one of the most important functions of transition metal hydride complexes.61,64 Recently, Rothwell even demonstrated the catalytic hydrogenation of benzene using a tantalum(V) hydride complex. There are a number of methods that have been utilized in the preparation of these hydride complexes. One common preparative method is the high pressure hydrogenation of metal-alkyl bonds.61,62 Another means of preparing hydride complexes is the oxidative addition of hydrogen to lower oxidation state compounds such as W(IV) and Ta(III).61,62,66 Transition metal hydride complexes have also been prepared by utilizing hydride reagents such as n-Bu3SnH or LiBEt3H.67 Because of its small size, many hydrides have coordination numbers greater than six. Also, most monomeric hydrides are stabilized by strong a-donors ligands, such as phosphines.

4.2. Preparation of W(VI) Hydride Complexes.

There have been many reports of the hydrogenation of high oxidation state alkyl complexes to give hydrides, however, frequently these reactions are performed under forcing conditions (extremely high pressures of hydrogen in the presence of phosphine ligands at elevated temperatures).61,62 Rothwell observed the reaction of hydrogen and



Ta(R)2(OAr)3 to give Ta(H)2(OAr)3(PR3) in the presence of phosphine.68 This reaction was carried out at 90 oC under 8300 kPa of hydrogen for 24 hours eq 4.1. These

R'".,Ta OAr + PMe2Ph 2 H2 (1200 psi) a PhMe2- Ta- OAr eq 4.1

R I 90oC, 24 hrs I
R = CH2C6H4-p-Me;
OAr = OC6H3-2,6-Pri2

conditions are typical for the hydrogenation of alkyl compounds and are quite harsh. Similar conditions were not necessary for the hydrogenation of some of the bis-alkyl complexes discussed in Chapter 3. When W(NPh)(CH2CMe3)2(Me3SiN)2C6H41, 25, was placed under two atmospheres of hydrogen in the presence of two equivalents of PMe3 at room temperature, the conversion of the bis-neopentyl to a new dihydride complex was complete in less than two hours. The dark brown solution turned magenta, and small red crystals precipitated from solution eq 4.2. The new complex was the seven coordinate dihydride, W(NPh)(H)2(PMe3)2[(Me3SiN)2C6H4], 39.

of Me3Si Me3Si Me3Si\ Me3P
N,,.. NNIf
< ~ W +X +2PMe3 H2 Y m H eq 4.2
hexanes ~I
N/ room temp N H

Me3Si 25 / Me3P
Me3Si 39

Compound 39 was characterized by multinuclear NMR and X-Ray

crystallography. At room temperature, the 1H NMR spectrum reveals the hydride resonances as a broad triplet at 9.28 ppm with a peak separation of 38 Hz. The two PMe3 groups appeared as a singlet at 1.04 ppm, while the silyl methyls were inequivalent singlets at 0.79 and 0.81 ppm, respectively. The two phosphines were observed as a singlet at


-24.46 ppm (1JWH = 188 Hz) in the 31P NMR spectrum. When the 1H NMR spectrum was taken at -50 'C, numerous changes were observed. First of all, the PMe3 resonance resolved into a triplet with a peak separation of 3 Hz at 0.84 ppm. Secondly, the silyl methyls separated further, resonating at 0.69 and 0.75 ppm. The hydrides appeared as a doublet of doublets at 8.92 ppm and 9.06 ppm respectively. The apparent coupling was 37 Hz.

Unfortunately, high quality single crystals of 39 have not yet been obtained. Numerous attempts were made to acquire crystallographic data on 39. Some data was collected although an accurate structure could not be determined. The data suggest the presence of trans-phosphines with the amido and imido nitrogens in the equatorial plane. In order to have equivalent phosphines and inequivalent hydrides, the hydrides probably lie cis to each other, in the plane of the imido and amido nitrogens. This geometry, shown in eq 4.2, should give an ABX2 resonance for the hydrides, which would be a doublet of triplets. The doublet of doublets observed at low temperature are very broad and may be due to a second order effect. A concerted effort is currently underway to obtain an X-ray structure of 39. The PMe2Ph derivative, W(NPh)(H)2(PMe2Ph)2[(Me3SiN)2C6H4], 40, was prepared in an analogous reaction. The hydride ligands were observed as a broad singlet at 9.80 ppm in the 1H NMR spectrum at room temperature. At -50 'C, a triplet at 9.56 ppm was observed with 39 Hz coupling, analogous to 39. The 31P NMR spectrum revealed a singlet at -22.48 ppm with 183 Hz 183W satellites. The PMe2Ph derivative was prepared in the hopes growing better crystals, since most of the structurally characterized hydrides are the PMe2Ph derivative.61,62,67

The seven-coordinate dihydride, 39, should serve as an excellent model to study the reactivity of dihydride complexes. Therefore, a more convenient route to the synthesis of this compound would be useful. Avoiding the alkylation step and adding hydride to the dichloride, 14 would be a viable route. Two equivalents of superhydride, LiBEt3H, were allowed to react with 14 in cold Et20. Work-up afforded 39 as a brown-red powder.


There was residual BEt3 which could not be removed from the compound. Although this reaction was important in showing the ability of 14 to add hydride, it was not employed to make the hydrides for the reactivity studies.

When two equivalents of PCy3 were utilized as the phosphine ligands, the steric bulk of the ligand prevented isolation of the analogous bis-phosphine complex. Although combustion analysis has not been performed because of the excess phosphine present, the spectral data are consistent with the isolation of the monophosphine complex W(NPh)(H)2(PCy3)[(Me3SiN)2C6H4], 41, eq 4.3. The 1H NMR spectrum of 41 reveals a doublet at 11.35 ppm (2JP-H = 83 Hz).

Me3Si Me3Si

/ hexanes Heq43
N room temp N |
Me3Si 25 / PCy3
Me3Si 41

Satellites due to the 183W 1JW-H coupling were observed at 64 Hz. The silyl methyl groups were observed as a singlet at 0.83 ppm. The 31P NMR spectrum of 41 reveals a singlet at 66.53 ppm (1JP-H = 83 Hz).

The chelating nature of the bis-amide ligand again dictates the geometry of the molecule. Rothwell prepared Ta(OAr)2CI(H)2(L)2 derivatives using PMe3, PMe2Ph, PMePh2.68 The structures of these compounds reveal that the phosphines are always trans to one another. In these compounds, the aryloxides are always trans to one another as well. This differed from 39 and 40 in which the amide nitrogens must be cis to one another because of the chelating nature of the ligand. In order to investigate the role of the trans phosphines, and to fine out whether or not cis phosphine complexes were stable, a chelating phosphine was selected. Diphenylphosphinoethane, DPPE, was utilized in a


reaction similar to eq 4.2. The new hydride was found to be W(NPh)(H)2(DPPE)[(Me3SiN)2C6H4], 42, eq 4.4.
The 1H NMR spectrum of 42 reveals a very symmetrical molecule. The silyl methyls were observed as a singlet at 0.49 ppm. The methylene protons of the DPPE ligand were observed as multiplets at 2.23 and 2.42 ppm. The hydrides were equivalent at all temperatures observed. At room temperature the hydrides appeared as a complex multiplet which can be viewed as the X part of an ABX2 spin system where the phosphines are A and B. Satellites due to 183W coupling were observed at 58 Hz for each of the peaks in the spectrum. The 1H NMR spectrum of the hydride region can be seen in Figure 4.1.

M %Me3Si Me3Si N N,. N H
."N.H2 H ,Ph
W+ DPPE W-P eq 4.4
(/ X hexanes N I\I Ph
N room temp N '
Me3Si 25 / Ph Ph
Me3Si 42

The phosphines were inequivalent in the 31P NMR spectrum, and appear as doublets, at

-12.70 and 30.43 ppm, respectively (2Jpp = 83 Hz). The geometry of this sevencoordinate dihydride can be inferred from the NMR data. Having one leg of the DPPE bisect the two hydrides is the only possible geometry giving both equivalent silyl methyls and hydrides as well as inequivalent phosphines.
There are few examples of high oxidation state hydride complexes where strong 0donor ligands are not coordinated to the metal center. Coincident with this is that there are few coordinatively unsaturated high oxidation state hydride complexes. One of the few examples resulted when Rothwell hydrogenated Ta(OAr)3(CH2C6H4-4-Me)2 under 1200 psi of hydrogen at 90 oC for 24 hrs to give Ta(H)(OAr)4 in low yield.68 The monohydride

ro,.,H .-Ph
I Ph

Me3Si 42

11:n ts:n l#:75 It:" 11.88 11:n 11:50 IIAS t1:2S to 'N
Figure 4.1. The 1H NMR spectrum of the hydride region of W(NPh)(H)2(DPPE)I(Mc3SiN)2C6H4],42


is presumably formed in a ligand exchange reaction. There are also metallocene derivatives of the type Cp2MIH2, but these will not prove insightful to this discussion.69

When phosphine was present upon hydrogenating the bis-neopentyl complex, 25, the reaction was complete in a matter of hours. When no phosphine was added to the reaction, the hydrogenation proceded much more slowly, and allows an examination of the mechanism of the reaction. When H2 gas was sealed in an NMR tube containing a C6D6 solution of 25, hydrogenation of the metal carbon bonds took place in a matter of days. The rate was dependent on the pressure of H2 gas. When a very low pressure of gas, < 20 psi was sealed in the tube, the reaction took nearly a week to reach completion. However, when 30 psi of H2 was used, the reaction was complete in about 36 hours. The product is only observable under an atmosphere of H2 gas. The product of the hydrogenation was presumably a dimer due to interpretation of the NMR data. The silyl methyl groups were observed as a singlet at 0.33 ppm in the 1H NMR spectrum. Two equivalents of neopentane were observed at 0.93 ppm. The spectrum also reveals a singlet at 15.68 ppm with 168 Hz 183W satellites corresponding to the two hydrides. The tungsten satellites were actually observed as doublets, J = 4 Hz. The observation of the tungsten satellites as doublets was the key evidence in proposing a dimeric structure. Cotton observed a similar coupling effect in the W2C14(NHCMe3)(PR3)2 system.69 The only observable satellites will be from a dimer with only one 183W, due to only 14% abundance. This would constitute an ABX spin system, resulting in a doublet of doublets. The hydride resonance is shown in Figure 4.2. The shift of the hydrides was in the normal range for high oxidation state group 5 and 6 complexes in the literature.61-68 Evidence for proposing bridging imido groups will be given later in the chapter. The reaction is shown in eq 4.5. When the sample was evaporated to dryness, the dihydride isomerized or rearranged, forming what appeared to be a bridging hydride. When the sample was redissolved in C6D6, the 1H NMR spectrum changed dramatically. At room temperature, the hydrides did not appear in the spectrum, although a broad increase in the integral was observed between


M i Me3Si SiMe3
,o.. / _Me3Si X 9N,,." /

Ix N- 5
H2, 20 psi WWe .
N NN_@ hexanes X /I q.
N\room temp H N H /
/ 36 hours N.
Me3Si 25 Me3Si SiMe3


11 and 13 ppm. When the sample was cooled to -25 'C, a sharp singlet at 12.58 ppm and a broad singlet at 15.80 ppm were observed in a 1:1 ratio. When the sample was cooled further, to -50 'C, the sharp singlet remained unchanged while the broad singlet split into two broad singlets at 15.28 ppm and 15.98 ppm respectively. The ratio of the three peaks was observed to be 2:1:1. The broad singlets are consistent with being bridging hydrides, while the sharp singlet appears to be the terminal hydrides. The sharp singlet actually had satellites at 53 Hz and 100 Hz. Although the true identity of the molecule cannot be confirmed, it is assuredly still a dihydride complex of sorts. When two equivalents of PMe3 were added to the sample, the bis-phosphine dihydride, 39, was formed by observation of the 1H NMR. There is not enough known at this time to make more substantial conclusions, but work is continuing in this area.

As was mentioned earlier, since the hydrogenation, at very low pressures of H2, takes a number of days, an opportunity to observe intermediates and deduce a mechanism was presented. In an NMR tube reaction, 25 was dissolved in C6D6 and sealed under an atmosphere of less than 20 psi H2. After twelve hours, the 1H NMR revealed four compounds in the reaction mixture. There was 25% starting material, neopentane, and two new compounds, the metallacyclobutane, 35, and a monohydride complex, 44 eq 4.6. The spectrum reveals the two doublets, at 1.7 and 2.1 ppm, which correspond to the metallacyclobutane complex discussed in Chapter Three.

Figure 4.2. The 1H NMR spectrum of the hydride resonance of 43.


Me3Si Me3Sk MeSi
N4 H, <20 psi e
4W Cs25 \H W eq 4.6
N room temp H
12 hours/

Me3Si/ 25 12hours Me3Si Me3Si 35

The monohydride, W(NPh)(CH2CMe3)(H)[(Me3SiN)2C6H4], 44, was characterized by a singlet at 18.43 ppm (1JW-H = 151 Hz). The methylene protons of the monohydrideneopentyl complex appeared diastereotopic with a doublet at 2.60 ppm and a broad singlet at 3.28 ppm. As the reaction was observed over the course of the following 3 days, the bis-neopentyl complex, 25, disappeared completely and the dihydride complex, 43, grew in. Over the course of a number of trials of this NMR tube reaction, it was observed that the bis-neopentyl reached at least 90% completion before dihydride formation initiated.

The metallacyclobutane complex was thought to be in an equilibrium with the

monohydride complex under an H2 atmosphere. The metallacycle was formed when H2 was lost from the monohydride by a net y-hydrogen elimination mechanism. The reverse reaction takes place when H2 was added across one of the W-C bonds of the metallacycle eq 4.7. This equilibrium was

Me3Si H3C Me3Si
CH3 -H2 "**... /
W CH3 +H2w eq 4.7
H+H2 aN aN/
/ /
Me3Si 44 Me3Si 35

demonstrated by adding H2 to an NMR tube containing a C6D6 solution of 25, then degassing the reaction after eight hours. The 1H NMR spectrum, which was taken immediately after degassing, revealed the three compounds expected, 25, 35, and 44 in roughly a 1:1:3 ratio. Over the course of two weeks, the reaction was degassed


periodically. During this time, the bis-neopentyl complex was completely consumed. Also, as the reaction proceeded, the monohydride decreased and was only observable in trace amounts. Figure 4.3 demonstrates how only a catalytic amount of hydrogen was necessary for complete conversion of the bis-neopentyl complex to the metallacycle.

Me3Si\ N


Me3Si 25


Me3Si II3C Me3Si\

MyNSi% +H3 ~f

Me3Si 44 Me3Si 35
Figure 4.3. Formation of the metallacycle, 35, by the addition of a
catalytic amount of H2 to the bis-neopentyl complex, 25.

All of the observations which have been made pertaining to the hydrogenation of 25 in the absence of PMe3 were from NMR tube experiments. When complete hydrogenation of 25 was carried out on a preparatory scale only the bridging hydride was isolated. It appears to be the same bridging hydride which was observed when 43 was evaporated to dryness and redissolved in C6D6. When a pentane solution of 25 was stirred under two atmospheres of H2 for 36 hours and cooled to -10 'C, dark crystals were formed which were suitable for X-ray diffraction.

The X-ray structure revealed a unique dimeric structure. The thermal ellipsiod plot of 45 is shown in Figure 4.4. The structure was interesting since there is a neopentyl

Me3Si SiMe3


Me3Si i


Figre4.. hemaEliSoi ltSfO5


group on one of the tungsten atoms and not on the other. Both of the imido groups were bridging and both the nitrogen's lie slightly closer to the tungsten without the neopentyl group, probably in order to relieve steric congestion. The bridging hydride was located and refined. It was found to be 2.07(10)A from W1 and 1.87(1 1)A from W2. Although the terminal hydrides were not found in the difference Fourier map, there appears to be open coordination sites which could accommodate the terminal hydrides, both on the unsubstituted tungsten. The open coordination sites are in the axial positions. The fold angle of the bis-amide ligand on the neopentyl substituted tungsten was 54 O. The fold angle of the bis-amide ligand on the unsubstituted tungsten was 0 . The planarity of the bis-amide ligand would be due to the seven-coordinate nature of the tungsten atom. The 1H NMR of 45 reveals that the molecule is quite fluxional, with all of the hydrides appearing as a very broad singlet at 12.9 ppm.

Hydrogenation of the bis-neophyl complex, 26, gave results similar to 25. However, the metallacycle was always observed in a much lower ratio than in the bisneopentyl hydrogenation. This was probably because the -methyl, f'-phenylmetallacyclobutane complex was much less stable than the 03,P'-dimethylmetallacyclobutane complex. Therefore, it loses H2 and reverts to the monohydride much more readily.

Another means of investigating this equilibrium was to examine the effect of added phosphine to the reaction mixture. Once the equilibrium was established and the reaction degassed, one equivalent of PMe3 was added to the reaction mixture. Surprisingly, the alkylidene, 36, was formed almost immediately. This reaction proceeds by an a-hydrogen abstraction from the monohydride complex, eliminating H2, or possibly rearrangement of the metallacycle. The H2 released then adds to the metallacyclobutane, forming more monohydride to react with the phosphine. The mechanism for this reaction can be seen in Figure 4.5. This would be the first example of an a-abstraction involving the loss of hydrogen to form an alkylidene. Undoubtedly, reductive elimination of neopentane would


be a more expected than a-abstraction. The implications of such a mechanism contradict many of the criterion which dictate a-abstraction reactions in high-oxidation state transition

Me3Si\ Me3Si H3C

Me3Si' 35 Me3Si' 44
H2 3

-H2 +PMe3


N PMe3
Me3Si' 36

Figure 4.5. Mechanism for the formation of the alkylidene by
0-abstraction from the equilibrium mixture of 35 and 44.

metals. Certainly no steric relief was gained by the elimination of H2. The W-Ca-Cp angle decreases from an alkyl to an alkylidene but that is overshadowed by the addition of PMe3 to the metal center.

This brings up the question of the hydrogenation of the alkylidene. When a C6D6 sample of 36 and one equivalent of PMe3 was placed under two atmospheres of H2, the dihydride, bis-phosphine complex, 39, was formed in a matter of hours eq 4.8. A different observation was made when the alkylidene was hydrogenated under an extremely low pressure of H2 in the absence of 'added' phosphine. Over the course of ten days, the alkylidene was completely consumed to give a one to one mixture of two products and one equivalent of neopentane. The two products were the dihydride, bis-phosphine complex, 39, and the metallacyclobutane complex, 35. The two products were most likely formed


by the disproportionation of the intermediates since there were less than two equivalents of H2 added.

MeMe3 C Me3Si\ Me3P O
N" 1 N%%IN, I N
PMe3 H2, 2 atm N eq 4.8
N PMe3 +
Me3Si/ 36 / Me3P
Me3Si 39

It cannot be denied that there appears to be a number of loose ends in the chemistry of these new hydride complexes, especially in the absence of phosphines. However, it is also undeniable that there was a wealth of information that lead to the conclusions which were made. Although isolation of a number of these compounds is unlikely due to their instability, crystallographic data on one or more of these compounds would be quite insightful. The discovery of an example a,-hydrogen abstraction involving the monohydride-neopentyl has enormous implications and will be investigated further.

4,3. Reactivity of the Dihydrides.

High oxidation state transition metal hydrides complexes are known to hydrogenate olefins both stoichiometrically and catalytically.62,63 Rothwell even demonstrated the hydrogenation of arenes using Ta(V) dihydrides.65 Coordination of the olefin was an essential step in the reaction. Olefins are 7c-acceptor ligands and require filled metal dorbitals to facilitate back-bonding. Therefore do olefin complexes are unlikely. Nonetheless d1 and d2 complexes should have sufficient orbitals to overlap with an olefin acceptor orbital, stabilizing a n-olefin complex. Ligands with a strong electronic donation to the metal center should aid in the isolation of coordinated 7t-olefin complexes. Imido ligands (-NR)2- are strong 7t-donors because of the lone electron pair and should aid in the


isolation of it-olefin complexes. Strong a-donors, such as phosphines should also aid in creating a suitable electronic environment for t-olefin complexes. Currently, there are very few examples of dI or d2 it-olefin complexes which have been thoroughly characterized.71 Therefore, the isolation of 7t-olefin complexes is essential to the advancement of this chemistry.

Initial investigation into the reactivity of some of the phosphine stabilized hydride complexes, 39 and 42, has begun. When a C6D6 sample of W(NPh)(H)2(PMe3)2[(Me3SiN)2C6H4], 39, in a sealable NMR tube was placed under an atmosphere of ethylene, ethane formation was observed by the observation of a singlet at

1.1 ppm in the 1H NMR spectrum. A second equivalent of ethylene was bound to the reduced metal center, forming the W(IV) ethylene complex, W(NPh)(1r2C2H4)(PMe3)2[(Me3SiN)2C6H4], 46, eq 4.9. The 1H NMR spectrum of the ethylene complex reveals that the compound is quite fluxional at room temperature.

Me3Si \ Me3 Me3Si PMe3

/ 2 C2H4 N.
N \ H N W + C2H6 eq 4.9

Me3 NMe3P

Me3Si 39 Me3Si 46

The room temperature 1H NMR spectrum is shown in Figure 4.6. The ethylene protons were observed as two multiplets at 1.96 and 2.19 ppm respectively. The coupled 13C NMR spectrum revealed a triplet at 36 ppm corresponding to the ethylene carbons. The 1JC-H coupling of the carbons was 156 Hz, typical for an ethylene complex, indicating that little 'metallacyclopropane' character was evident71. The PMe3 ligands were equivalent and were observed at 1.05 ppm in the 1H NMR spectrum. At room temperature, the 31p NMR spectrum revealed a broad singlet at -23.2 ppm while at -25 'C, a sharp singlet at
-21.77 ppm with 238 Hz 183W satellites was observed. The silyl methyls appeared as a

MC3Sf PMe3
Me3p, Me3SI 45


Figure 4.6. Ile 1H NMR spectrum of W(NPh)(il2-C21-I4)(PMe3)21(Me3SiN)2C61-141I 46, at 23 0C


broad singlet, nearly 120 Hz wide, at 0.50 ppm in the 1H NMR spectrum at room temperature. When the sample was cooled to -25 'C, two sharp singlets were observed at

0.39 ppm and 0.42 ppm. A tantalum ethylene complex prepared by Schrock72 and tungsten olefin complexes prepared by Nielson73 show that the ethylene prefers to coordinate cis to the imido. The ethylene also preferred to coordinate cis to phosphine ligands. In every case, the ethylene coordinated trans to an 'X' ligand. Nielson obtained crystallographic data on W(NPh)C12(Me2C=CH2)(PMe3)2 which clearly demonstrated this geometry.73 The spectral data for 46 were consistent with these compounds and support a trans phosphine structure. Difference nOe experiments on 46 showed that upon irradiation of the ortho-protons of the imido group, a 4.8% enhancement was observed for the coordinated ethylene. This substantiates the cis orientation of the imido and the ethylene.

A W(IV) olefin complex was also formed when two equivalents of styrene was allowed to react with the dihydride, 39. In an NMR tube experiment, formation of ethylbenzene was observed along with formation of the styrene complex, W(NPh)(T12CH2CHPh)(PMe3)2[(Me3SiN)2C6H4], 47. The olefinic peaks were observed as sharp multiplets at 2.62 and 2.80 ppm and a triplet at 3.87 ppm in a 1:1:1 ratio. The orientation of the coordinated styrene is not known, although it appears by NMR that only one orientation is adapted.

When the catalytic hydrogenation of ethylene and cyclooctene was attempted using the dihydride, 39, mixed results were observed. Cyclooctene and H2 were added to an NMR sample of 39. Over time, even with mild heating, it did not appear as though any cyclooctene was hydrogenated by observation of the NMR. It was already known that one equivalent of ethylene is stoichiometrically hydrogenated by 39, so detection of catalysis by NMR proved difficult. It appeared as though, over days, that the ratio between the ethylene and the ethane appeared to decrease. Further work will be focused on this reaction.


The reaction of ethylene and the DPPE dihydride, 41, did not yield an ethylene complex. Instead, no reaction was observed at room temperature over three days. However, when a purple C6D6 solution of 41 and ethylene was heated to 90 0C for 12 hours, the color changed to yellow. The 1H NMR revealed that DPPE had been lost from the metal center. The resonances corresponding to the hydrides and free ethylene were also no longer present. The product formed appears to be a metallacyclopentane complex formed by the addition of two equivalents of ethylene to the metal center.74 The overall reaction can be seen in eq 4.10. The complex appeared to have

Me3Si I Me3Si

W, XS C2H4 +DPPE eq 4.10

Me3Si/ Ph Ph Me3Si

a plane of symmetry due to the observation of a singlet at 0.31 ppm in the 1H NMR spectrum and two AA'BB' multiplets at 7.11 ppm and 7.41 ppm. The metallacyclopentane protons were observed as three multiplets, at 1.58 ppm, 2.41 ppm, and 2.89 ppm in a 2:2:4 ratio.

A reaction that is important to mention at this point is the reaction of excess ethylene with the bis-neopentyl complex, 25. When ethylene was allowed to react with a C6D6 solution of 25, the 1H NMR revealed that 1-butene was released. The formation of 1butene was very slow but very clean, a 60:40 mixture of ethylene and 1-butene was observed after 14 days at room temperature. Throughout the reaction, the concentration of 25 remained virtually unchanged while formation of other organometallic products was not observed. After 14 days, the sample was heated to 80 'C for 8 hours. The 1H NMR spectrum revealed that all of the ethylene had been consumed. There was no bis-neopentyl complex observable in the NMR, although the large amount of 1-butene obscured much of


the spectrum. The sample was degassed thoroughly. The 1H NMR revealed that the organometallic product was identical to the proposed metallacyclopentane complex from the reaction of 42 and ethylene.

The fact that 1-butene formation was observed without observation of a new

organometallic compound suggests that a very small amount of an active catalyst was being produced at room temperature. At higher temperature, all of the bis-neopentyl must be converted to the active catalyst. The formation of 1-butene from the metallacyclopentane complex would go by a simple 1-hydrogen elimination reaction. Therefore, the formation of the metallacyclopentane is the key step of the reaction. The formation of metallacyclopentane compounds from addition of two equivalents of ethylene to a low oxidation state metal center is well known.2,74,75 These metallacyclopentane complexes are also known to undergo -hydrogen elimination reactions, forming 1-butene. The formation of a reduced metal species, W(IV), in this case, would be formed by the reductive coupling of the two neopentyl groups. The mechanism for this reaction can be seen in Figure 4.7.

Most of the work reported in this chapter involves very recent results, and has brought up many new questions as well as areas for future work. Other members of the research group will continue this work, and hopefully continue to make great strides in this area.

xs C2H4 rW] IV
P-hydrogen elim.


Figure 4.7. The catalytic formation of 1-butene from the addition of ethylene to the bis-neopentyl complex, 25.


Unless otherwise noted, all procedures were performed under dry argon

atmosphere using standard Schlenk techniques or in a nitrogen atmosphere dry box. All solvents were dried according to established literature procedures.

1H, 13C and 31P NMR spectra were recorded on a Varian VXR-300 (300 MHz), a General Electric QE-300 (300 MHz), or a Varian Gemini-300 (300 MHz) spectrometer. Chemical shifts were referenced to the residual protons of the dueterated solvents and are reported in ppm downfield of TMS for 1H and 13C NMR spectra. 31P NMR were referenced to an external H3PO4 standard. Elemental analysis were performed by Atlantic Microlabs, Inc., Norcross, GA.

Preparation of N,N'-bis(trimethylsilyl)-o-phenylenediamine, 1:

Trimethylsilylchloride (12.7 mL, 0.10 mol) was added slowly to a solution of ophenylenediamine (5.00 g, 0.047 mol) in 50 mL of Et20. A white precipitate formed upon addition. Triethylamine (13.9 mL, 0.10 mol) was then added slowly, ensuring that stirring continued throughout the addition. After stirring for 3 hours at room temperature, the mixture was filtered and the solid was washed twice with 15 mL of Et20. The filtrate was stripped of solvent under reduced pressure to give 10.2 grams of a yellow solid; yield, 86.8%. M.P.; 29.5 'C.



Preparation of Li2rl1.2-(NSiMe32CH4. 2.

A solution of N,N'-bis(trimethylsilyl)-o-phenylenediamine (5.26 g, 20.9 mmol) in 75 mL of pentane was cooled to -78 'C. To this solution, 2 equivalents of n-BuLi (16.8 mL, 41.9 mmol, 2.5 M sol. in hexanes) was added slowly. A white precipitate formed as gas was evolved. Upon addition, a bubbler was attached and the reaction was stirred at room temperature for 2 hours under a flow of argon. The mixture was filtered and the solid dried under reduced pressure. The volume of the filtrate was reduced to 25 mL under reduced pressure and cooled to -15 oC to give colorless crystals. Yield; 4.93 g (combined), 89.5%.

Preparation of N.N'-bis(dimethylphenylsilyl)-o-phenylenediamine, 3:

O-phenylenediamine (1.58g, 14.64 mmol) was slurried in 50 mL of hexanes. Two equivalents of both dimethylphenylsilylchloride (4.90 mL, 29.29 mmol) and triethylamine (4.08 mL, 29.29 mmol) were added via syringe. The mixture was then refluxed for 12 hours. Upon cooling, a white salt precipitated from solution. The mixture was filtered. The yellow solution was stripped of solvent under reduced pressure to give 4.86 grams of a yellow/red solid; yield, 87.0%. Anal. Calc'd for C22H28N2Si2: C, 70.15; H, 7.49; N,

7.44. Found: C, 69.89; H, 7.28; N, 7.21.

Preparation of NN'-bis(methyldiphenylsilyl)-o-phenylenediamine, 4:

O-phenylenediamine (0.64 g, 5.94 mmol), methyldiphenylsilylchloride (2.50 mL, 11.81 mmol) and triethylamine (1.82 mL, 13.00 mmol) were reacted as described above for 3 to give 2.09 grams of a reddish solid; yield 70.3%. Anal. Calc'd for C32H32N2Si2: C, 76.75; H, 6.44; N, 5.60. Found: C, 76.39; H, 6.23; N, 5.27.


Preparation of NN'-bis(trimethylsilvl)-4.5-dimethyl- 1 .2-diaminobenzene, 6:

4,5-dimethyl-1,2-diaminobenzene (5.11 g, 37.52 mmol), trimethylsilylchloride (9.52 g, 75.05 mmol), and triethylamine (10.53 mL, 75.05 mmol) were reacted as described above for 3 to give 9.12 grams of a yellow solid; yield, 88.53%. Anal. Calc'd for C14H28N2Si2: C, 59.93; H, 10.06; N, 9.99. Found: C, 60.13; H, 10.29; N, 10.21.

Preparation of NN'-(trimethylsilyl)-1.8-diaminonapthalene. 7:

1,8-diaminonapthalene (10.61 g, 67.07 mmol), trimethylsilylchloride (17.87 mL, 140.84 mmol), and triethylamine (19.63 mL, 140.84 mmol) were reacted as described for 3 above to give 18.15 grams of a red solid; yield, 89.4%. Anal. Calc'd for C16H26N2Si2: C, 63.49; H, 8.67; N, 9.27. Found: C, 63.17; H, 8.39; N, 9.08.

Preparation of N-phenyl, N'-trimethylsilyl-o-phenylenediamine. 8:

N-phenyl-o-phenylenediamine (1.53 g, 8.28 mmol), trimethylsilylchloride (1.18 mL, 8.50 mmol), and triethylamine (1.08 mL, 8.50 mol) were reacted as described above for 3 to give 1.64 grams of a reddish solid; yield; 77.2%. Anal. Calc'd for C15H20N2Si: C, 70.24; H, 7.87; N, 10.93. Found: C, 69.91; H, 7.68; N, 10.73.

Preparation of 1.2-(iPrNH )2C6H4 9 and 1.2-[N(H)C(Me)2CH2CH(Me)N(H)1-CaH41

O-phenylenediamine (2.5 g, 23.12 mmol) and sodium acetate (7.21 g, 87.9 mmol) were slurried in 15 mL of acetic acid, 20 mL of acetone, and 40 mL of H20. The mixture was stirred for 30 minutes in an ice bath. NaBH4 (14.97 g, 277.4 mmol) was added slowly. Upon addition, the reaction was allowed to warm to room temperature and was stirred for one hour. NaOH (6 M solution) was added until slightly basic by litmus test. The mixture was then extracted with Et20 (2 x 20 mL). The Et2O was removed in vacuo to


yield a red oil. The oil was separated by flash chromatography on a Silica column using hexanes. The heterocyclic compound 1,2-[N(H)C(Me)2CH2CH(Me)N(H)]-C6H4 10, was eluted first, solvent was striped in vacuo to give a white powder (1.31 g, 41.1%). Anal. Calc'd for C12H20N2: C, 75.80; H, 9.47; N, 14.74. Found: C, 75.71; H, 9.64; N, 14.75. The diamine, 1,2-(iPrNH)2C6H4 9, was eluted last and was isolated by cooling the hexanes solution to -78 'C to yield a white powder, which melted to a colorless oil upon warming (0.87 g, 29.3 %).

Preparation of 1,8-[N(H)C(Me)2N(H)C1jnH4_ 11

1,8-Diaminonapthalene (2.57 g, 16.25 mmol) and sodium acetate (5.06 g, 61.75 mmol) were slurried in 10 mL of acetic acid, 20 mL of acetone, and 40 mL of H20. After stirring in an ice bath for 30 minutes, NaBH4 (15.95 g, 285.9 mmol) was added slowly. Upon addition, the reaction was allowed to warm to room temperature and was stirred for one hour. NaOH (6 M solution) was added until slightly basic by litmus test. The mixture was then extracted with Et20 (2 x 20 mL). The Et20 was removed in vacuo to yield red powder (2.96 g, 91.9%). Anal. Calc'd for C13H14N2: C, 78.76; H, 7.12; N, 14.13. Found: C, 78.22; H, 7.09; N, 14.49.

Preparation of WOC912[(NSiMe32CL41. 13:

WOC14 (1.00 g, 2.93 mmol) and 1.1 equivalents of Li2[1,2-(NSiMe3)2Cl6H4 2 (0.85 g, 3.20 mmol) were combined in a Schlenk tube and cooled to -78 'C. 50 mL of Et20 which had been cooled to -78 OC was added. The reaction was warmed to room temperature and stirred for 8 hours. Solvent was removed under reduced pressure. The solid was extracted with 50 mL of pentane and filtered through a Celite pad. The dark solution was cooled to -78 C to afford a dark red powder (0.39 g, 25.6% yield). Anal. Calc'd for C12H22N2OC12Si2W: C, 27.65; H, 4.26; N, 5.37. Found: C, 27.61; H, 4.29; N, 5.41.


Preparation of W(NPh)Clr2[(NSiMe312C_j1. 14:

N,N'-bis(trimethylsilyl)-o-phenylenediamine 1 (3.83 g, 15.16 mmol) was

dissolved in 30 mL of Et2O and cooled to -78 'C. Two equivalents of n-BuLi (12.13 mL, 30.32 mmol, 2.5 M soln in hexanes) were then added. The reaction was warmed to room temperature and stirred for one hour. The reaction was recooled and .95 equivalents of W(NPh)C14(OEt2) (7.10 g, 14.4 mmol) in 20 mL of Et20 was added. The reaction was stirred for three hours, then filtered through a celite pad, which, in turn, was rinsed with more Et20. Solvent was removed under reduced pressure. The dark red solid was washed with pentane until an orange powder remained. The powder was dried to yield 7.31 g of 14; 85.1%. Anal. Calc'd for C18H27N3C12Si2W: C, 36.25; H, 4.56; N, 7.05. Found: C, 35.91; H, 4.78; N, 6.78.

Preparation of 1,8-K2(Me3SiN)C1oHI4 16 and 1,8-Li2(MeSiN)C0loH4fi17

1,8-Diaminonapthalene was dissolved in pentane and cooled to 0 oC. Two

equivalents of either KH or n-BuLi were then added. The reactions were allowed to warm and stirred for one hour. Yellow precipitate formed during the reaction. Cooling the reaction mixtures to -10 C afforded nearly quantitative yields of the corresponding salts as yellow crystals.

Preparation of W(NPh)C12[ l.8-(NSiMe3)C1oH6L.18

1,8-(NHSiMe3)2C10H6 7 (1.28 g, 4.23 mmol) was dissolved in 25 mL of Et20

and cooled to 78 C. Two equivalents n-BuLi (3.38 mL, 8.46 mmol, 2.5 M in Et20) were added via syringe. The reaction was allowed to warm to room temperature and stirred for one hour. Over this time the color of the solution changed from red to yellow. The solution was recooled and .95 equivalents of W(NPh)C14(OEt2) (2.02 g, 4.10 mmol) in a 25 mL Et20 solution were added. The reaction was allowed to warm to room temperature


and was stirred for 6 hours. The mixture, which had turned dark red, was filtered through a celite pad and was washed with Et20 until colorless. The solvent was removed under reduced pressure and dried for 4 hours. The solid was then washed 3 times with 20 mL of pentane and dried under reduced pressure overnight. 2.13 grams of a dark red solid were isolated; yield, 80.4%. Anal. Calc'd for C22H29N3C12Si2W: C, 40.88; H, 4.52; N, 6.50. Found: C, 40.59; H, 4.21; N, 6.19.

Preparation of W(NPh)Cl2[4,5-(CH3)2-1,2-(NSiMe3)2C6H21,19:

N,N'-bis(trimethylsilyl)-4,5-dimethyl-1,2-diaminobenzene 6 (1.29 g, 4.70 mmol), W(NPh)C14(OEt2) (2.29 g, 4.65 mmol), and n-BuLi (3.76 mL, 9.40 mmol, 2.5 M in in Et20)were reacted as described above for 14. 1.89 g of a red solid were isolated; 64% yield. Anal. Calc'd for C20H31N3C12Si2W: C, 38.47; H, 5.00; N, 6.73. Found: C, 38.29; H, 4.78; N, 6.48.

Preparation of W(NPh)(OSO2CF3)[3iN)C i4i(OEt2) 20:

W(NPh)C12[(NSiMe3)2C614] 14 (0.50 g, 0.84 mmol) and two equivalents of

Ag(OSO2CF3)2 (0.43 g, 1.68 mmol) were combined in a Schelnk tube and dissolved in 25 mL of 0 oC Et20. After stirring at room temperature for 8 hours, the Et20 was removed under reduced pressure. The solid was extracted with Et20 and filtered through Celite until colorless. The reddish solution was concentrated to 10 mL and cooled to -10 'C to give

0.59 grams of 20 as an orange solid, yield: 85.0%. Preparation of W(NPh)C2(PMe3) [(NSiMe3)2C6Li41.21:

W(NPh)C12[(NSiMe3)2C6H4] 14 (0.68 g, 1.14 mmol) was slurried in 30 mL of pentane. Excess PMe3 (4.55 mL, 2.00 mmol, .44 M in toluene) was added to the reaction. The reaction immediately turned from redish to deep purple. The reaction was cooled to

-78 oC to give 21 as 0.71 grams of purple crystals, yield; 90.5%.


Preparation of W(NPh)Cl(L)[(NSiMe3)2C6I4L. L = THF, 22: 3-Picoline. 23: CH3CN, 4:

W(NPh)C12[(NSiMe3)2C6H4] 14 was dissolved in a minimum amount of the

solvent, L. The deep purple solution was then added slowly to a stirring pentane solution, which immediately turned purple. The solutions were then cooled to -78 'C to give purple crystals of the mono-adduct.

Preparation of W(NPh)(CH2C(CH33,r2[(NSiMe3C i114 25.:

W(NPh)C12[(NSiMe3)2C6H4] (2.78g, 4.66 mmol) was dissolved in 30 mL of

Et20 and cooled to -78 'C. Two equivalents of C1MgCH2C(CH3)3 (7.37 mL, 9.32 mmol,

1.27 M soln in Et20) were then added. The reaction was allowed to warm to room temperature after 30 minutes. After one hour, solvent was removed under reduced pressure. The solid was extracted with pentane until clear and filtered through a Celite pad. The solution was concentrated to a total volume of about 10 mL and cooled in an -78 'C cold bath to yield dark crystals of 25; 2.19 g (yield 70.1%). Anal. Calc'd for C28H47N3Si2W: C, 50.52; H, 7.12; N, 6.31. Found: C, 50.36; H, 7.04; N, 6.14.

Preparation of W(NPh)(CH2C(CH3)2Ph)r[(NSiMe3}2H4L 26:

W(NPh)C12[(NSiMe3)2C6H4] (2.04 g, 3.42 mmol) was dissolved in 30 mL of Et20 and cooled to -78 oC. Two equivalents of ClMgCH2C(CH3)2Ph (6.55 mL, 6.84 mmol, 1.045 M soln in Et20) were then added. The reaction was allowed to warm to room temperature after 30 minutes. After one hour, solvent was removed under reduced pressure. The solid was extracted with pentane until clear and filtered through a Celite pad. The solution was concentrated to a total volume of about 15 mL and cooled in an -78 'C cold bath to yield 2.21 g of 26 as a light brown solid. Yield: 82.4%. Anal. Calc'd for C38H45N3Si2W: C, 58.23; H, 5.79; N, 5.36. Found: C, 57.95; H, 5.58; N, 5.29.


Preparation of W(NPh)(CH32[(NSiMe3)2C6i41, 27:

W(NPh)C2[(NSiMe3)2C6H4] (1.03 g, 1.73 mmol) was dissolved in 20 mL of

Et20 and cooled to -78 oC. Two equivalents of MeLi (2.47 mL, 3.46 mmol, 1.4 M soln in Et20) were then added. The reaction was allowed to warm to room temperature after 15 minutes. After 30 minutes, solvent was removed under reduced pressure. The solid was extracted with pentane until clear and filtered through a Celite pad. The solution was concentrated to a total volume of about 5 mL and cooled in an -78 'C cold bath to yield 680 mg of 27 as a gold-brown solid. Yield: 70.8%. Anal. Calc'd for C20H33N3Si2W: C, 43.24; H, 5.99; N, 7.56. Found: C, 42.89; H, 5.94; N, 7.29.

Preparation of W(NPh)(CH2CH3)2[(NSiMe3).2C6H4, 28:

W(NPh)C12[(NSiMe3)2C6H4] (1.12 g, 1.88 mmol) was dissolved in 25 mL of

Et20 and cooled to -78 'C. Two equivalents of EtMgCl (1.88 mL, 3.76 mmol, 2.0 M soln in Et20) were then added. The reaction was allowed to warm to room temperature after 15 minutes. After 30 minutes, solvent was removed under reduced pressure. The solid was extracted with pentane until clear and filtered through a Celite pad. The solvent was removed under reduced pressure to yield 0.98 g of 28 as a thick red oil.. Yield: 89.3%. Preparation of W(NPh)(CH2Ph)2[(NSiMe3)2C641,29

W(NPh)C12[(NSiMe3)2C6H4] 14 (2.02 g, 3.39 mmol) was dissolved in 25 mL of Et20 and cooled to -78 oC. Two equivalents of C1MgCH2Ph (6.77 mL, 6.77 mmol, 1.0 M soln. in Et20) were then added. The reaction was allowed to warm to room temperature after 15 minutes. After 30 minutes, solvent was removed under reduced pressure. The solid was extracted with pentane until clear and filtered through a Celite pad. The solution was concentrated to 10 mL and cooled to -78 'C to give 1.47 grams of 29 as a dark solid.


Yield: 61.3%. Anal. Calc'd for C32H41N3Si2W: C, 54.31; H, 5.84; N, 5.94. Found: C, 53.98; H, 5.61; N, 5.69.

Preparation of W(NPh)(C1)(CH2CMe3)[(NSiMe 133:

W(NPh)C12[(NSiMe3)2C6H4] 14 (1.65 g, 2.77 mmol) was dissolved in 25 mL of Et20 and cooled to -78 'C. One equivalent of ClMgCH2CMe3 (2.31 mL, 2.77 mmol, 1.2 M soln in Et20) was then added. The reaction was allowed to warm to room temperature after 15 minutes. After 45 minutes, solvent was removed under reduced pressure. The solid was extracted with pentane until clear and filtered through a Celite pad. The solution was concentrated to 5 mL and cooled to -78 'C to give 1.09 grams of 33 as a red solid. Yield: 62.3%. Anal. Calc'd for C23H38N3CISi2W: C, 43.71; H, 6.06; N, 6.65. Found: C, 43.38; H, 5.81; N, 6.09.

Preparation of W(NPh)(CHCMe3)(PMe ) r(NSiMe3g2641. 36:

In a 200 mL glass tube fitted with a teflon Young's joint,

W(NPh)(CH2CMe3)2[(NSiMe3)2C6H4] (1.25 g, 1.87 mmol) was dissolved in 25 mL of toluene. Five equivalents of PMe3 (.968 mL, 9.35 mmol) were then added and the tube was sealed. The reaction was then heated to 70 oC for 24 hours. The solution was transferred to a round-bottom Schlenk were solvent was removed under reduced pressure. The brown oil was extracted with pentane and the volume of the filtrate was concentrated to about 15 mL. The solution was cooled to -10 oC to yield 0.83 g of 36 as orange crystals; yield 66.0%. Anal. Calc'd for C26H46N3PSi2W: C, 46.49; H, 6.90; N, 6.26. Found: C, 46.23; H, 6.81; N, 6.05.

Preparation of W(NPh)[CH(t-Bu)C(Ph)C(Ph)lr[(Me3Si NbCLi4138

W(NPh)(CHCMe3)(PMe3)[(NSiMe3)2C6H4] 36 (0.21 g, 0.31 mmol) and

diphenylacetylene (0.06 g, 0.34 mmol) were dissolved in 25 mL of pentane. The reaction


was then refluxed for 20 hours. Upon cooling, solvent was stripped in vacuo to give a red oil which appeared pure by 1H NMR.

Preparation of W(NPh)(FI)2(PMe3}2L(MSpda). 39: Method 1:

In a glass tube with a Teflon Young's joint, W(NPh)(CH2CMe3)2(TMS2pda) (1.66 g, 2.48 mmol), 25 was dissolved in 25 mL of hexanes. PMe3 (0.64 mL, 6.20 mmol) was added via syringe. The solution was then placed in liquid nitrogen while a vacuum was applied. Once the solution was frozen solid under vacuum, the flask was sealed. Hydrogen gas was then purged through the neck of the flask. Once the neck was purged, the H2 hose was wired securely to the flask. The flask was opened until the H2 reached a pressure of 10 PSI. The flask was then resealed and the H2 line removed. The reaction was then allowed to warm to room temperature. After four hours of stirring, the color of the solution changed from brown to magenta. Magenta crystals had also precipitated from solution. The solution was transferred to a Schlenk tube and cooled to -10 'C to give magenta crystals. The mother liquors were concentrated and cooled to give more of the same. Total Yield of 39: 1.48 g (87.8%). Anal Calcd for C24H47N3P2Si2W: C, 42.41; H, 6.97; N, 6.18. Found: C, 42.18; H, 6.79; N, 6.03.

Method 2:

W(NPh)C12(TMS2pda) (1.08 g, 1.81 mmol), 14 was dissolved in 20 mL of Et20 and cooled to -78 'C in an isopropanol/dry ice bath. Two equivalents of PMe3 (0.38 mL, 3.62 mmol) were added via syringe. Two equivalents of LiBEt3H (3.62 mL, 3.62 mmol,

1.0 M solution in THF) were added slowly. After warming to room temperature and stirring for four hours, the solvent was removed under reduced pressure. The solid was extracted twice with 10 mL of pentane. The brown solution was concentrated to 5 mL and


cooled to -10 'C. 0.87 g of a brown-red solid precipitated from solution. 1H NMR confimned the formation of 39 with about 15% BEt3 impurity present.

Method 3:

W(NPh)(CHCMe3)(PMe3)(TMS2pda) 36 (50 mg, 0.07 mmol) was dissolved in C6D6 in a NMR tube fitted with a teflon Young's joint. One equivalent of PMe3 (7 p.L,

0.07 mmol) was added via microliter syringe. The NMR tube was then fitted with a Schlenk adapter. The NMR tube was frozen in liquid nitrogen and placed under vacuum. The tube was then sealed frozen, under vacuum. Hydrogen gas was then purged through the Schlenk adapter for five minutes. The teflon seal was opened to allow the 112 to fill the vacuum in the NMR tube. The NMR tube was charged with about 15 PSI of H2. Over a period of less than 2 hours, complete conversion of 36 to 39 was observed in the 1H NMR.

Preparation of W(NPh)(H)9(DPPE)(TMS~pda), 42:

In a glass tube with a Teflon Young's joint, W(NPh)(CH2CMe3)2(TMS2pda) (0.65 g, 0.98 mmol) and DPPE (0.39 g, 0.98 mmol) were dissolved in 30 mL of hexanes. The solution was then placed in liquid nitrogen while a vacuum was applied. H2 gas was introduced as described for 39. The reaction was then allowed to warm to room temperature. After eight hours of stirring, the color of the solution had changed from brown to purple. Purple solid had also precipitated from solution. The solution was transferred to a Schlenk tube, concentrated to 10 mL, and cooled to -10 'C to give a purple solid. Total Yield: 0.71 g (78.2%). Anal. Calcd. for C44H53N3Si2P2W: C, 57.08; H,

5.77; N, 4.54. Found: C, 57.33; H, 5.84; N, 4.54.


Preparation of W(NPh)(l2-C2_H4 (PMe32(TMS2pda), 46:

In a glass tube with a Teflon Young's joint, W(NPh)(H)2(PMe3)2(TMS2pda), 39, (0.64 g, 0.94 mmol) was dissolved in 20 mL of pentane. The solution was then placed in liquid nitrogen under vacuum. Once the solution was frozen solid under vacuum, the flask was sealed. Ethylene was purged through the neck of the flask and then the hose was wired on. The flask was opened to a low pressure of ethylene (2 PSI) for about ten seconds. Then flask was then resealed and allowed to warm to room temperature. After 12 hours, the color of the solution had lightened. The solution was transferred to a Schlenk tube and cooled to -10 'C. Purple-brown solid formed and was isolated. The mother liquors were reduced in volume under reduced pressure to about 5 mL and recooled to yield more solid. Total yield; 0.59 grams, 89.1%. Anal. Calcd. for C26H49N3Si2P2W: C, 44.25; H, 7.00; N, 5.95. Found: C, 43.91; H, 6.79; N, 5.78.

Polymerization Experiments:

All of the ROMP experiments were carried out under an inert atmosphere in

deolifinated toluene solutions. The reactions were terminated by transfering the mixtures to stirring methanol with a trace of BHT added to insure against radical reations upon precipitation. The precipitated polymers were dried in vacuo and analyzed by GPC.


Spectroscopic Data

Table A-i1. IH NMR Data

compound 6,1212 muLlt Jli Hzt assignt
1,2-[(CH3)3SiNH2C6L4 1 0.18 s 18 -SiM 3
3.00 br s 2 -NH
6.83 m2 aromatic
6.89 m2

1,2-Li2[(CH3)3SiN]2C6H4 2 0.22 s 18 -SiAe3
6.59 m 2 aromatic
6.87 m 2 aromatic

1,2-[(CH3)2PhSiN{]2C6H4 3 0.33 s 12 -SiMe2?Ph
3.28 br s 2 -NH
6.69 m 2 aromatic
6.88 m 2 aromatic
7.20 m 6 aromatic
7.57 m 4 aromatic

1,2-[(CH3)Ph2SiNH]2C6H[4 4 0.61 s 6 -SiAePh2
3.57 br s 2 -NH
6.58 11 2 aromatic
6.92 m 2 aromatic
7.18 m 12 aromatic
7.58 m 8 aromatic

4,5-Me2-1,2-(Me3SiN-)2C6H4 6 0.19 s 18 -SiMe3
2.12 s 6 4,5-Me2
2.95 br s 2 -NH
6.80 s 2 aromatic

l,8-(Me3SiNII)2C10H6 7 0.17 s 18 -SiNMe3
5.38 br s 2 -NH
6.70 d 7 2 p-CflOH6
7.19 t 8 2 m-CiOH6
7.33 d 7 2 o-CI0H6

1,2-Ph[(CLI3)3SiNH]C6114 8 0.11 s 9 -SiAe3
4.09 br s 1 -NH
4.43 br s 1 -NHf
6.49 d 8 2 aroamrtic
6.73 t 8 2 aromatic
6.96 t 8 2 aromatic
7.07 m 3 aromatic

1,2-(iPrNH)2C6H4 9 0.98 d 6 12 -CHMe2
3.08 br s 2 -NIH
3.32 quin 6 2 -CH~e2
6.69 m 2 aromatic
6.93 m 2 aromatic



1 ,2-IIN(H)C(Me)2CH2CH(Me)N(H)]C6H4 10 0.83 d 6 3 -CHIAg
0.89 s 3 -CM&(Me)
1.07 s 3 -CMefte)
1.25 dd 6 1 -CIH
1.62 t 6 1 -Clii
2.90 br s 2 -NH
3.17 qd 4 1 -CHTMe
6.45 m 2 aromatic
6.80 m 2 aromatic
1,8-IiN(H)C(Me)2N(H)]ClOH4 11 0.98 s 6 -CM~f,
3.42 br s 2 -NHl
6.26 d 7 2 aromatic
7.23 d 7 2 aromatic
7.35 t 7 2 aromatic

W(O)C12[(Me3SiN)2C6H4] 13 0.39 s 18 -SiM3
6.80 m 2 aromatic
6.95 m 2

W(NPh)C12[(Me3SiN)2C6H4] 14 0.35 s 18 -SiMe3
6.71 t 8 1 p-NPh-H
6.94 m 2 aromatic
7.01 t 8 2 m-NPh-H
7.14 m 2 aromatic
7.31 d 8 2 o-NPh-H

W(NPh)C12[(Me2PhSiN)2C6H4] 15 0.62 s 6 -SiMe2Ph
0.68 s 6 -SiMe9)Ph
6.62 m 2 aromatic
6.65 t 8 1 p-NPh-H
6.92 t 8 2 m-NPh-H
to aromatic

W(NPh)C1211,8-(Me3SiN)2CIOHi6] 18 0.41 s 18 -SiMe3
6.39 d 7 2 p-CIOH6
6.72 t 8 1 p-NPh-H
6.93 d 7 2 o-CIOIH6
6.98 t 8 2 m-NPh-H
7.13 t 7 2 m-Cj0H6
7.48 d 8 2 o-NPh-H

W(NPh)C12[4,5-Me2-1,2- 0.41 s 18 -SiMe3
(Me3SiN)2C6H4] 19
1.99 s 6 4,5-Me9Ph
6.73 t 8 1 p-NPh-H
7.02 t 8 2 m-NPh-H
7.07 s 2 aromatic
7.35 d 8 2 o-NPh-li


W(NPh)(OSO2CF3)2[(Me3SiN)2C6H4J- 0.32 s 18 -SiM3
(GEt2) 20 &
1.09 t 7 6 -GEt2
3.25 q 7 4 -GEt2
6.55 t 8 1 p-NPh-II
7.02 t 8 2 m-NPh-H
7.09 m 2 aromatic
7.30 m 2 aromatic
7.38 d 8 2 o-NPh-H

W(NPh)C12(PMe3)[(Me3SiN)2C6H4] 0.42 s 18 -SiMe3
0.79 d J2P-H=9 9 -PMe3
6.63 t 8 1 p-NPh-H
6.83 m 2 aromatic
7.01 m 2 aromatic
7.17 t 8 2 m-NPh-H
7.49 d 8 2 o-NPh-Hf

W(NPh)C12(=H)[(Me3SiN)2C6H4] 22 0.41 s 18 -SiMe-3
0.92 qnt 3 4 -THIF
3.63 t 6 4 -THF
6.62 t 8 1 p-NPh-H
6.85 m 2 aromatic
7.03 m 2 aromatic
7.10 t 8 2 m-NPh-H
7.47 d 8 2 o-NPh-Hf

W(NPh)C12(3-Me-py)[(Me3SiN)2C61-4] 0.42 s 18 -SiMe3
2.17 s 3 py-Me
6.21 t 2 1 py-5-H
6.33 d 8 1 py-4-H
6.54 m 2 aromatic
6.65 t 11 1 p-NPh-I1
6.76 m 2 aromatic7.18 t 9 2 m-NPh-H
7.61 d 7 2 o-NPh-Hf
8.72 s 1 py-1-H8.75 d 6 1 py-6-H

W(NPh)C12(NCCH3) [(Me3SiN)2C6H4] 0.38 s 18 -Si-M3
0.49 s 3 -NCCMe
6.67 t 7 1 p-NPh-H
6.93 m 2 aromatic
7.04 t 7 m-NPh-H
7.14 m 2 aromatic
7.34 d 7 2 o-NPh-H

W(NPh)(CH2CMe3)2[(Me3SiN)2- 0.54 s 18 -SiMO
C6H4] 25
1.00 s 18 -CMW3


2.13 d 10 2 -CH2CMe3
J2WH= 11
2.29 d 10 2 -CH2CMe3
J2 W= 1
6.83 t 7 1 p-NPh-H
6.86 m 2 airomatic7.19 t 7 2 m-NPh-H
7.25 11 2 aromatic
7.59 d 7 2 o-NPh-H

W(NPh)(CH2CMe2Ph)2[(Me3SiN)2- 0.42 s 18 -SiMkQ3
C6H4] 26
1.37 s 9 -CMe2,Ph
1.38 s 9 -CW2Ph
1.95 d 11 2 -CHj2C
J2~ H1 1
2.94 d 11 2 -CI C
J2 W1= 11
6.83 t 7 1 aromatic
6.94 t 7 2
6.97 m 2
7.01 t 7 4
7.06 t 7 2
7.18 d 7 4
7.19 d 7 2
7.30 m 2
W(NPh)(CH3)2[(Me3SiN)2C6H4] 27 0.31 s 18 -SiMe3
1.12 s J2W-H=6 6 W-Me
6.86 t 8 1 p-NPh-H
7.06 m 2 aromatic
7.11 t 8 2 m-NPh-H
7.32 d 8 2 o-NPh-H
7.35 m 2 aromaticW(NPh)(CH2CH3)2[(Me3SiN)2C6H4] 0.29 s 18 -SiMe3
1.86 s 6 -CH2CH3
1.91 m 2 -CH2CH3
2.31 m 2 -CH2CH3
6.86 t 8 1 p-NPh-H
7.04 m 2 aromatic
7.11 t 8 2 m-NPh-H
7.31 d 8 2 o-NPh-Hf
7.37 m2 aromatic

w(NPh)(CH2Ph)2r(Me3SiN)2C6H4] 29 0.09 s 18 -SiMe3
2.78 s 4 -CHj2Ph
6.81 t 8 2 p-~CH2Ph-li
6.88 t 8 1 p-NPh-H
to aromatic
7.30 d 8 2 o-NPh-H


7.37 m 2 aromatic
spectrum taken at 80 'C in C7D8
2.71 m 7 4 -CHj2Ph

W(NPh)Ph2[(Me3SiN)2C6-L] 30 0.11 s 18 -SiMe3
6.82 t 7 2 aromatic
6.86 t 7 1
7.04 t 7 4
7.18 t 7 2
7.21 m 2
7.39 m 2
7.52 d 7 2
7.62 d 7 4

W(NPh)(CH2CMe3)2[(Me2PhSiN)2C6E14] 31 0.81 S 12 -SiMe2)Ph
1.03 s 18 -CH2CMe3
2.04 d 10 2 -CH2CMe3
J2 H1 1
2.78 d 10 2 -Cjj2CMe3
J2WH= 11
6.59 aromatic

W(NPh)C1(CH2CMe3)[(Me3SiN)2- 0.24 s 9 -SiN493
C6H4] 33
0.42 s 9 -SiMe3
1.23 s 9 -CH2CMe3
1.93 d 10 1 -CH2C -e
2.08 d 10 1 -CH2CMe3
6.76 t 7 1 p-N~h-I1
to aromatic
7.41 d 8 2 o-NPh-H

W(NPh)(CII2CMe3)(NMe2)[(Me3SiN)2 0.31 s 9 -Si.M3
C6H4] 34
0.46 S 9 -SiMe3
1.03 s 9 -CH2CMei
1.33 d 10 1 -CH9,C~e3
2.60 d 10 1 -CH2CMe3
3.19 s 3 -NMR2
3.68 s 3 -NYM&
to aromatic

W(NPh)(CH2C(Me)2CH2)[(Me3SiN)2CHfl 35 -1.70 d 9 2 -Cf2
0.39 s 9 -SiM.93
0.54 s 3 -CH2CMe20.57 s 3 -CH2CMe-2-