Luận án tiến sĩ: Synthesis, structure, and reactivity of terminal cobalt imido complexes

List of Figures Figure 1.1 Simplified bonding description of terminal imido complexes--- 1 Figure 1.2 Molecular orbital diagrams of an octahedral metal imido complex depicting the π bond

SYNTHESIS, STRUCTURE, AND REACTIVITY OF TERMINAL COBALT

IMIDO COMPLEXES

by Daniel Travis Shay

A dissertation submitted to the Faculty of the University of Delaware in partial fulfillment of the requirements for the degree Doctor of Philosophy in Chemistry and Biochemistry

Fall 2006

Copyright 2006 Daniel Travis Shay

All Rights Reserved

UMI Number: 3247713

3247713 2007

UMI Microform Copyright

All rights reserved This microform edition is protected against unauthorized copying under Title 17, United States Code.

ProQuest Information and Learning Company

300 North Zeeb Road P.O Box 1346 Ann Arbor, MI 48106-1346

by ProQuest Information and Learning Company

SYNTHESIS, STRUCTURE, AND REACTIVITY OF TERMINAL COBALT

IMIDO COMPLEXES

by Daniel Travis Shay

I certify that I have read this dissertation and in my opinion it meets the academic and professional standard required by the University as a dissertation for the degree of Doctor of Philosophy

Signed: _

Klaus H Theopold, Ph.D

Professor in Charge of dissertation

I certify that I have read this dissertation and in my opinion it meets the academic and professional standard required by the University as a dissertation for the degree of Doctor of Philosophy

Signed: _

Charles G Riordan, Ph.D

Member of dissertation committee

I certify that I have read this dissertation and in my opinion it meets the academic and professional standard required by the University as a dissertation for the degree of Doctor of Philosophy

Signed: _

Joseph M Fox, Ph.D

Member of dissertation committee

I certify that I have read this dissertation and in my opinion it meets the academic and professional standard required by the University as a dissertation for the degree of Doctor of Philosophy

Signed: _

Roger Grey, Ph.D

Member of dissertation committee

Acknowledgements

I would first like to thank my advisor Klaus H Theopold for giving me the

opportunity to work in his laboratory on such an interesting research project He has been a wonderful mentor offering me guidance throughout the past five years I am proud to hold the degree of Doctor of Philosophy in Chemistry from his research group at the University of Delaware

I would also like to thank all the students and professors in the Chemistry

Department at the University of Delaware who have helped me during my time at the University I would like to thank the Theopold group (both past and present) for their help and input for my projects Also, I would like to thank both Dr Riordan and Dr Fox and their respective groups for allowing me to use both their time and their student’s time for helpful discussions, and of course their lab supplies

I would like to thank the departmental staff for helping me keep the lab running smoothly Our electronic and glass shop have always responded very quickly when their services were required

Last but not least, I would like to thank my wife Ni Yan who has given me the support and encouragement to complete my Ph.D in chemistry She has pushed me each time I thought the road was long and for that I am grateful

I would also like to thank myself for sticking with it and seeing it to the end

Dedicated to my loving wife Ni Yan

Table of Contents

List of Tables - IX List of Figures - XI List of Schemes - XII Abstract - XIV

Introduction - 1

Synthesis, Structure, Reactivity of Terminal Imido Complexes-Results and Discussion Synthesis and Structure - 14

Spectroscopic Characterization - 36

Reactivity - 48

Conclusion - 87

Experimental Section - 88

General Techniques - 88

Synthesis of TptBu,MeCoNAd - 89

Synthesis of TptBu,MeCoNtBu - 89

Synthesis of TptBu,MeCoNMe - 90

Synthesis of TptBu,MeCoNEt - 91

Synthesis of BptBu,Me(Me-pz-CMe2CH2N(Ad)H)Co - 92

Synthesis of BptBu,Me(Me-pz-CMe2CH2N(Ad)H)CoI - 93

Synthesis of BptBu,Me(Me-pz-CMe2CH2N(tBu)H)Co - 93

Synthesis of TptBu,MeCo(Lut)BARF - 94

Synthesis of TptBu,MeK-d9 - 95

Synthesis of TptBu,MeCo(N3) - 97

Reaction of TptBu,MeCoNAd with HCl - 98

Reaction of TptBu,MeCoNAd with H2O - 98

Reaction of TptBu,MeCoNAd with CO - 98

Reaction of TptBu,MeCoNtBu with CO - 99

Reaction of TptBu,MeCoLutBARF with O2 - 99

Adamantyl Aziridine - 100

Adamantyl Piperidine - 100

Methyl Azide - 101

Azo Adamantane - 102

Kinetic Isotope Effect Measurements - 102

Crystal Structure Determinations - 103

Appendix A Figure A1.1 Field dependence of magnetic susceptibility of 1 104

Crystal Data and Structure Refinement for TptBu,MeCoNAd-1/2 C5H12 (1) - 105

TptBu,MeCoNtBu-1/3 C5H12 (2) - 106

TptBu,MeCoNMe (3) - 107

TptBu,MeCoNEt (4) - 108

TptBu,MeCoLutBARF-1/2 C5H12 (5) - 109

BptBu,Me(Me-pz-CMe2CH2N(Ad)H)Co (6) - 110

BptBu,Me(Me-pz-CMe2CH2N(Ad)H)CoI (7) - 111

BptBu,Me(Me-pz-CMe2CH2N(tBu)H)Co (8) - 112

Table A1 9 Temperature (oC) vs Chemical Shift (ppm) for 1 - 113

Table A1.10 Temperature (oC) vs Chemical Shift (ppm) for 1d 27 114 References - 115

List of Tables

Table 1.1 Interatomic distances in Å for TptBu,MeCoNAd, (1) - 16

Table 1.2 Interatomic angles in degree for TptBu,MeCoNAd, (1) - 17

Table 1.3 Interatomic distances in Å for TptBu,MeCoNtBu, (2) - 22

Table 1.4 Interatomic angles in degree for TptBu,MeCoNtBu, (2) - 23

Table 1.5 Interatomic distances in Å for TptBu,MeCoNMe, (3) - 28

Table 1.6 Interatomic angles in degree for TptBu,MeCoNMe, (3) - 29

Table 1.7 Interatomic distances in Å for TptBu,MeCoNEt, (4) - 32

Table 1.8 Interatomic angles in degree for TptBu,MeCoNEt, (4) - 33

Table 1.9 List of chemical shifts for TptBu,MeCoNAd and TptBu,MeCo15Nad - 46

Table 1.10 Interatomic distances in Å for [TptBu,MeCo(Lut)][BARF], (5) - 54

Table 1.11 Interatomic angles in degree for [TptBu,MeCo(Lut)][BARF], (5) - 55

Table 1.12 Interatomic distances in Å for BptBu,Me(Me-pz-CMe2CH2N(Ad)H)Co, (6) - 62

Table 1.13 Interatomic bond angles in degree for BptBu,Me(Me-pz-CMe2CH2N(Ad)H)Co, (6) - 63

Table 1.14 Interatomic distances in Å for BptBu,Me(Me-pz-CMe2CH2N(Ad)H)CoI, (7) - 69

Table 1.15 Interatomic bond angles in degree for BptBu,Me(Me-pz-CMe2CH2N(Ad)H)CoI, (7) - 70

Table 1.16 Interatomic distances in Å for BptBu,Me(Me-pz-CMe2CH2N(tBu)H)Co, (8) - 74

Table 1.17 Interatomic angles in degree for BptBu,Me(Me-pz-CMe2CH2N(Ad)H)Co,

(8) - 75

Table 1.18 Rate Constants (S-1) and isotope effects for the interconversion of

1 to 6 - 85

Table A1.1 Crystal data and structure refinement for TptBu,MeCoNAd⋅1/2 C5H12, 1 105

Table A1.2 Crystal data and structure refinement for TptBu,MeCoNtBu⋅1/3 C5H12, 2 106

Table A1.3 Crystal data and structure refinement for TptBu,MeCoNMe, 3 - 107

Table A1.4 Crystal data and structure refinement for TptBu,MeCoNEt, 4 - 108

Table A1.5 Crystal data and structure refinement for [TptBu,MeCoLut][BARF]

Table A1 9 Temperature (oC) vs Chemical Shift (ppm) for (1) - 113

Table A1.10 Temperature (oC) vs Chemical Shift (ppm) for 1d 27 - 114

List of Figures

Figure 1.1 Simplified bonding description of terminal imido complexes - 1

Figure 1.2 Molecular orbital diagrams of an octahedral metal imido complex depicting the π bonding interaction - 3

Figure 1.3 Examples of terminal cobalt (III) imido complexes - 4

Figure 1.4 The multiple step synthesis of a terminal nickel II imido species 7

Figure 1.5 Reaction of an electrophile with an iridium t-butyl imido complex- 9

Figure 1.6 Nucleophilic attack of triphenyl phosphene with a chromium aryl imido complex resulting in ligand transfer - 9

Figure 1.7 Copper catalyzed aziridination of olefins - 10

Figure 1.8 One of the more common ligands employed for the catalytic aziridination of olefins - 10

Figure 1.9 Mechanistic picture for the aziridination of olefins - 11

Figure 1.10 C-H activation of an inferred cobalt imido complex - 13

Figure 1.11 Molecular structure of TptBu,MeCoNAd, complex 1 - 15

Figure 1.12 Molecular structure of TptBu,MeCoNtBu, complex 2 - 21

Figure 1.13 Molecular structure of TptBu,MeCoNMe, complex 3 - 27

Figure 1.14 Molecular structure of TptBu,MeCoNEt, complex 4 - 31

Figure 1.15 1H NMR spectrum of TptBu,MeCoNAd, 1, in benzene C6D6 at 295K - 38

Figure 1.16 d orbital splitting pattern of degenerate metal d orbitals in C3V symmetry- 39 Figure 1.17 Magnetic susceptibility and moment as a function of temperature of complex 1 - 42

Figure 1.18 Magnetic susceptibility and moment as a function of temperature of

complex 1 after Hgo - 43

Figure 1.19 Variable temperature 1H NMR analysis of 1 in C6D6 - 45

Figure 1.201H NMR of 1:1 mixture of TptBu,MeCo14NAd and TptBu,MeCo15NAd in benzene C6D6 at 295K - 47

Figure 1.21A The molecular structure and of [TptBu,MeCo(Lut)][BARF], 5 - 52

Figure 1.21B The molecular structure and of [TptBu,MeCo(Lut)][BARF], 5 - 53

Figure 1.22 The molecular structure of BptBu,Me(Me-pz-CMe2CH2N(Ad)H)Co, 6 - 61

Figure 1.23 The molecular structure of BptBu,Me(Me-pz-CMe2CH2N(Ad)H)CoI, 7 68

Figure 1.24 Molecular structure of BptBu,Me(Me-pz-CMe2CH2N(tBu)H)Co, (8) - 73

Figure 1.25 Interconversion of 1 to 6 at 335.1(3) K - 76

Figure 1.26 Erying Analysis of TptBu,MeCoNAd from 40-90oC - 81

Figure A1.1 Field dependence of magnetic susceptibility of 1 - 104

List of Schemes

Scheme 1.1 Possible organic transformation products of TptBu,MeCoNAd with ethylene 60 Scheme 1.2 Synthesis of pinacolone d3 - 82

Synthesis, Structure, and Reactivity of TptBu,MeCoNR (TptBu,Me = hydrotris(3-t

Bu-5-Me- pyrazolyl)borate) (R = Me, Et, tBu, Ad)

Kinetic studies of the thermal decomposition of TptBu,MeCoNAd which undergoes C-H activation of the Tp ligand yielding BptBu,Me(Me-pz-CMe2CH2N(Ad)H)Co, have been closely monitored by 1H NMR spectroscopy Determination of the rate of C-H insertion of the imido ligand of TptBu,MeCoNAd at a variety of temperatures led to the generation of an Erying Plot which showed curvature Kinetic isotope measurements were also conducted showing a temperature dependent kinetic isotope effect indicating that breaking of the C-H bond was involved in the rate determining step

transfer reactions.3 They have also gained interest because of their role as the

catalytically active species in olefin aziridination.4 Metal imido complexes can be

categorized into two different classes.5 The first category is linear, i.e., when an imido complex exhibits a M-N-R bond angle in the range of 160o to 180o.5 The second

category is considered to be bent; here the M-N-R bond angle is between 130o to 150o.5,6 Valence bond theory suggests the principal bonding modes consist of one σ and one to two π bonds depending on the interaction of the nitrogen’s lone pair electrons with the metal.7 Figure 1.1 represents a simplified view of the two different bonding modes of the terminal imido ligand with a transition metal.8

N

R M

Bonding Mode A Bonding Mode B

Figure 1.1: Simplified bonding description of terminal imido complexes

Structure A depicts the linear fashion with the N atom to be considered sp hybridized while structure B shows the bent bonding mode and considers the N atom to be sp2

hybridized In this simplified bonding description of the imido ligand, the interaction of

the N atom’s lone pair electrons with that of the metal d orbitals determines a linear or bent structure If empty d orbitals are available for bonding, one can image more

interaction of the N atom’s electrons representing a more linear M-N-R bond In more electron rich compounds bending of the M-N-R bond can occur and the lone pair

electrons will be more localized on the nitrogen (bonding mode B) However, bending of the M-N-R bond can occur in complexes with electron deficient metal centers One such example is Os(NtBu)4 The Os-N-C bonds are bent with an angle of 156.4o.9 Despite the interaction of the nitrogen’s electrons it is common to represent the metal to ligand bond

as M=NR This allows for a more simplified assignment of the metal’s oxidation state with the imido ligand acting as a di anionic 4 e- donor.10 Figure 1.2 represents the

bonding of the imido ligand with the d orbitals of an arbitrary transition metal in terms of molecular orbital theory depicting an octahedral metal complex that can form a triple bond with a N atom.11 Considering the principal axis (z) to lie along the M-N bond, the π interaction from the N atom’s p orbitals with the metal d orbitals is N px with M dxz and

N py with M dyz The M dxy is a non-bonding orbital Using both the simplified valence bonding depicted in Figure 1.1 along with the molecular orbital description for the π interaction of the N atom’s lone pair electrons accounts for the binding of an imido ligand with a transition metal The interaction of the lone pair electrons of the nitrogen with the metal also governs the M-N bond distance Stronger binding interaction of the imido ligand to a metal center indicated by better overlap of the nitrogen’s p orbitals with the metal d orbitals provides for short M-N bond distances Typical values for first row transition metal imido complexes M-N bond distances are in the range of 1.60-1.74 Å.12

N N

N N

N N N Co N R

R'

+

N

N Ar Ar

Co=NAd Ph-B

PPh 2

PPh2

Co=NTol P

Figure 1.3: Examples of terminal cobalt (III) imido complexes

These short bond distances indicate that the imido ligand forms multiple bonds with transition metals.13 In cobalt imido chemistry M-N bond distances are between 1.62 - 1.67Å.1d,14 Figure 1.3 depicts examples of cobalt imido complexes of 1.60-1.74 Å.12 These short bond distances indicate that the imido ligand forms multiple bonds with transition metals.13 In cobalt imido chemistry M-N bond distances are between 1.62 - 1.67Å.1d,14 Figure 1.3 depicts examples of cobalt imido complexes listing the relevant Co-N bond distances and Co-N-R bond angles.1d,14 While lone pair electron interaction does influence both the M-N bond distance and the M-N-R bond angle of metal imido complexes, it is not the only interaction that needs to be accounted for We must also consider the steric environment created by the ancillary ligands Steric repulsions of the imido ligand’s R group by the metal’s ancillary ligands can influence both the M-N bond distance and the M-N-R bond angle One example where we can see the steric influence

of the ancillary ligand of the metal on the M-N-R bond angle can be seen in the cobalt

imido complexes above in Figure 1.3 The M-N-R bond angle in example B is smaller than in examples A and C In example B, the bidentate ligand creates a little more room for the imido ligand as compared to the tridentate ligands in examples A and C Any greater bending of the M-N-R bond angles in either A or C would cause unfavorably steric interactions

Synthesis

While many examples of imido complexes exist for early transition metals, as one travels across the periodic table to the right hand side that number begins to decrease.15 This is due to the lack of empty metal d orbitals available to accept π donation from the imido ligand.1 Therefore the metal must have an accessible oxidation state to

accommodate the di anionic imido ligand and have empty metal d orbitals available to accept π donation

There are numerous synthetic routes that have been successful in obtaining

terminal imido complexes, ranging from deprotonation of amines to RN transfer using isocyanates.16 A few such reactions are represented below

Synthetic routes for the synthesis of terminal imido complexes

(1) O=WCl4 + RNCO → RN=WCl4 + CO2

(2) O=ReCl3(PPh3)2 + ArNH2 → ArN=ReCl3(PPh3)2 + H2O

(3) (Me3SiO)-ReO3 + 3 tBu(Me3Si)NH → (Me3SiO)-Re(=NtBu)3 + 3 Me3SiOH

In equation 1, the ligand exchange is believed to proceed through coordination of the isocyanate to tungsten followed by the expulsion of CO2 This is a convenient synthetic transformation of a metal oxo to metal imido complex with the release of CO2 providing the driving force needed to complete the reaction The examples presented in equations 2 and 3 are similar to 1 in that the amine first coordinates to the metal followed by the loss

of water or an alcohol A remarkable feature of examples 2 and 3 is that the imido

complexes can be successfully isolated despite the fact that they are present in solution with an equivalent of a protic substance such as water or an alcohol.17 The above three examples for the synthesis of terminal imido complexes proceed relativity smoothly and

in most instances in near quantitative yield In each case the oxo ligand is exchanged for the isoelectronic imido moiety What makes this synthetic route so appealing is the availability of so many metal oxo species.18 Indeed this route is widely utilized with early metal oxo species such as tungsten or molybdenum.16 Reactions with azo

compounds such as PhN=NPh and oxidation by the use of organic azides also provide pathways to terminal imido complexes of the transition metals.19,20 Below (equation 4) is

an example of a vanadium (II) species being oxidized by phenyl azide to produce the corresponding aryl imido complex.21 The reaction is believed to proceed through

coordination of the azide, followed by elimination of N2 to form the imido ligand

(4) Cp*

2V + PhN3 → Cp*

2V=NPh + N2

One of the more interesting approaches for the synthesis of late metal terminal

imido complexes has been demonstrated by Hillhouse et al starting from a nickel (II)

halide species.22

P

P Ni-Cl2

C8K

P P Ni P

P Ni

-NaN(TMS)2P

P Ni=NAr Cl

Figure 1.4: The multiple step synthesis of a terminal nickel II imido species

The initial Ni (II) dichloride complex was reduced with potassium graphite to produce

the bridging Ni (I) chloride species shown in Figure 1.5 The bridging dimer was then

reacted with an aryl lithium amide forming the amide complex, which was then oxidized

by tropylium hexafluorophosphate yielding the cationic Ni (II) amide species The final

step in the imido ligand synthesis was the deprotonation of the amide with a strong base

like sodium bis(trimethylsilyl) amide at -35oC Here we see an elegant multi-step

approach to the synthesis of a Ni (II) imido complex providing another synthetic route to

a terminal imido complex

While many examples for the synthesis of terminal imido complexes do exist,

each system must be analyzed individually to determine which synthetic route will be

suitable Metals in higher oxidation states tend to favor ligand exchange reactions such

as the case seen in equation 1 with the tungsten (VI) oxo complex Metals that start in low oxidation states with stable oxidation states two units greater tend to favor reactions with azides and azo compounds This is especially true for the synthesis of terminal cobalt imido complexes.1d,14 To date, the only successful synthetic procedure for the synthesis of cobalt imido complexes is the reaction of organic azides with cobalt (I) species In each example, the Co (I) species contains a labile ligand such as a phosphine

or coordinated toluene which can easily be replaced by an imido ligand The cobalt (I) center is oxidized by either an alkyl or aryl azide to produce a Co (III) imido complex One of the major driving forces for these reactions is believed to be the loss of N2

Reactivity

The reactivity of transition metal imido complexes depends on a variety of different factors such as the oxidation state of the metal, the ancillary ligands used to support the metal, and the metal itself.23 The nitrogens of imido complexes that have weak lone pair electron interaction with the metal d orbitals tend to be more nucleophilic

in nature.23 Here, the lone pair electrons tend to be localized on the nitrogen, allowing it

to react with electrophilies An example from Bergman et al using a Cp iridium imido

complex seen in Figure 1.6 shows the nucleophilic nature of the imido ligand.24

Ir N

t

Bu

CH3I

[Cp*IrI2]2 + tBuNMe3I

Figure 1.5: Reaction of an electrophile with an iridium tert-butyl imido complex

Terminal imido ligands can also act in an electrophilic manner.25 If the metal imido complex has a considerable amount of electron donation to the metal from the nitrogen’s lone pair electrons, it can react with nucleophiles The electrophilic nature of the imido ligand has been demonstrated in chromium chemistry seen in Figure 1.7.26

N

N Cr

N

N

N Cr

PPh 3 PPh3

TolN=PPh3+

Figure 1.6: Nucleophilic attack of triphenyl phosphine at a chromium aryl imido complex resulting in ligand transfer.

One of the more important and industrially useful reactions for transition metal imido complexes are ligand transfers to olefins.4 Transfer of an imido moiety to an olefin

such as ethylene produces an N-substituted aziridine.27 The most common commercial uses for functionalized aziridines are as hardeners in paints, varnishes, or other coatings One of the most heavily studied catalytic systems for the ligand transfer of an imido ligand is the aziridination of alkenes by copper imido species supported by nitrogen based ancillary ligands.28 Seen below is the catalytic cycle for the synthesis of tosyl aziridine from a Cu(III) imido complex

PhI

[L2Cu]+

[L2CuNTs]+

RR

Ts N

In this example, the nitrene source for the aziridination of alkenes is

[N-(p-toluenesulfonyl)imino]phenyliodinane Though the imido moiety has not been isolated, due to its reactive nature, its presence has been postulated as the key active species in olefin aziridination on the basis of hybrid density functional theory calculations

(B3LYP).29 Kinetic experiments show that the rate for the functionalization of an olefin

is first order in both the metal complex and alkene.29 It is believed that the first step in

the reaction shown in Figure 1.10 is loss of the coordinated ethylene The next step is the reaction of the naked Cu (I) species with PhI=NTs that proceeds through the homolysis of the PhI=NTs to form the transient copper imido species as shown in Figure 1.10.30 The rate determining step is believed to be insertion of the alkene into the metal nitrogen bond

to form the coordinated aziridine adduct.29 The weakly coordinated aziridine can then be replaced by another molecule of ethylene to complete the cycle It is interesting to note that the reaction shows little substituent effect on the rate of aziridination when

employing a variety of para substituted styrenes Indeed, olefin aziridination of

p-nitrostyrene proceeded with almost an identical rate to that of styrene

NN

Cu

N

N Cu-

Figure 1.9: Mechanistic picture for the aziridination of olefins

Over the past few years the interest in late transition metal complexes that contain metal to ligand multiple bonds has increased substantially.30 This is especially true for the imido ligand mainly because it is isoelectronic to the oxo ligand It is this

isoelectronic relationship of the imido ligand with the oxo ligand that initially provoked our interest in cobalt imido chemistry A major area of focus in our laboratory has been the synthesis and characterization of metal oxo species that employ the sterically

hindered tris(pyrazolyl)borate ligand TpR,R’ (Tp = Tris(pyrazolyl)borate).31 However, we have been unsuccessful in isolating a cobalt oxo compound Every attempt in isolating a

TpR,R’Co=O species was meet will failure as the oxo species would inevitable abstract a hydrogen atom to yield TpR,R’Co-OH.31 Presently there are no examples of isolated terminal cobalt oxo species This led our investigation of cobalt oxo species to cobalt imido complexes in the hope that light could be shed on the reactive nature of the

former.32

Our initial studies of cobalt imido complexes began with the attempted synthesis

of the trimethylsilyl imido complex TptBu,MeCoNTMS.32 When 1.0 equivalents of

trimethylsilyl azide was reacted with TptBu,MeCo(N2), the color changed from brown to red along with the evolution of roughly 2.0 equivalents of nitrogen gas However,

workup of the reaction did not produce the terminal imido complex TptBu,MeCo=NTMS Unexpectedly, the inferred trimethylsilyl imido complex had abstracted a hydrogen atom from one of the tert-butyl groups of the Tp ligand to form an amido complex Figure 1.11 shows the reaction path and final outcome from the reaction of TptBu,MeCo(N2) with trimethylsilyl azide The trimethylsilyl imido intermediate abstracted a hydrogen atom from the tert-butyl group of the Tp ligand followed by the formation of a carbon-cobalt bond which stabilized the carbon radical.32 This hydrogen abstraction reactivity was very reminiscent of our elusive [TptBu,MeCo=O] intermediate because we were unable to isolated a stable terminal imido species.31b Despite the difficulties in isolating the

terminal trimethylsilyl imido complex TptBu,MeCoNTMS, it was believed that variation of the R group of the imido ligand could facilitate the isolation of a terminal cobalt imido complex

N Co TMS(H)N H B N

N 2

Figure 1.10: C-H activation of an inferred cobalt imido complex

Results and Discussion

Synthesis and Structure

When 1.0 equivalent of adamantyl azide was added as a solid in one portion to a stirred solution of TptBu,MeCo(N2) in pentane or THF, the result was a rapid color change from brown to green along with the simultaneous evolution of nitrogen gas After allowing the reaction to stir for 60 minutes the solution had turned to a dark green color and no more nitrogen evolution was observed Subsequent work up of the reaction mixture and

crystallization from pentane at –35o C for 24 hours produced green plates of

TptBu,MeCoNAd, 1, in 78% yield.33 Figure 1.12 shows the result of a structure

determination by X-ray diffraction; complex 1 is a four coordinate terminal imido

complex of the type TptBu,MeCo=NR of C3V symmetry Two distinguishing features of 1

are the relatively short Co-N(7) bond distance of 1.655(2) Å (Table 1.1), and the N(7)-C(25) bond angle of 178.3(2)o (Table 1.2) (for crystallographic parameters see Appendix A) These bonding parameters are in good agreement with other reported terminal imido complexes of cobalt (shown in Figure 1.3) and indicate the multiple bonding nature of the imido ligand.1d,14 Indeed, a bond length of 1.655 Å is far too short for a typical cobalt-nitrogen single bond which is on the order of 2.05Å.34 Typical metal-nitrogen distances for terminal imido complexes are in the range of 1.60-1.74 Å.1,11,14,33 The four coordinate environment around the metal center is a slightly distorted

Co(1)-tetrahedron of C3V symmetry The tert-butyl groups in the 3 position of the pyrazolyl rings also provide the steric bulk required to only permit a monomeric terminal imido

Figure 1.11: Molecular structure of TptBu,MeCoNAd, (1)

Table 1.1 Selected interatomic distances in Å for TptBu,MeCoNAd, (1)

Table 1.2 Selected interatomic angles in degree for TptBu,MeCoNAd, (1)

species The N(7)-C(25) distance of 1.441(3) Å allows the adamantyl cage to sit directly within the pocket formed by the tert-butyl groups of the Tp ligand This also accounts for the nearly linear Co(1)-N(7)-C(25) bond angle of 178.3(2)o; any greater deviation from linearity would promote unfavorable steric interactions of the adamantyl cage with the tert-butyl groups of the ligand A molecule of pentane cocrystallizes with every two

molecules of 1 and can be seen in the 1H NMR spectrum (see Figure 1.16)

The synthesis of the tert-butyl imido complex, TptBu,MeCoNtBu, is significantly different from that of its adamantyl counterpart Whereas an equivalent of adamantyl azide can be added in one portion to a room temperature solution of TptBu,MeCo(N2) to produce the terminal imido species in relatively good yield, addition of tert-butyl azide under these conditions yields almost exclusively the azide complex TptBu,MeCo(N3).35 The difference in the reactivity of adamantyl azide and tert-butyl azide with

TptBu,MeCo(N2) may lie in the difference in the C-N bond strengths of the tBu-N and

Ad-N bonds The C-I bond dissociation energy for tert-butyl iodide is 50.0 kcal/mol versus 50.6 kcal/mol for the C-I bond of adamantyl iodide This slightly weaker tert-butyl heteroatom bond could account for the homolytic cleavage of tert-butyl azide to form the azide adduct.36

We have seen the formation of the azide complex TptBu,MeCo(N3) in other

attempted syntheses of terminal imido complexes when reacting benzyl and diphenyl methyl azide with TptBu,MeCo(N2) When one equivalent of benzyl azide Ph-CH2-N337 or diphenyl methyl azide (Ph)2-CH-N3 38 was added to a room temperature pentane solution

of TptBu,MeCo(N2), the result was an instant color change from brown to blue along with the simultaneous evolution of nitrogen gas Upon workup of the reaction mixture, the

sole paramagnetic species obtained was TptBu,MeCo(N3) Modification of the reaction procedure had no effect on the formation of the azide adduct Because the benzyl radical

is much more stable than the adamantyl radical, the reaction will always proceed through homolysis of the R-N3 bond to form the azide adduct.39

Modification of the reaction procedure for tert-butyl azide with TptBu,MeCo(N2) does have an effect on the outcome of the reaction Consider the following for the

formation of either the azide adduct or terminal imido complex TptBu,MeCoNtBu in the addition of tert-Butyl azide to TptBu,MeCo(N2) The assigned numbering system for tert-butyl azide is seen below

N=N=N

1 2 31

The most likely reaction scenario is that nitrogen 3 of tert-butyl azide first

coordinates to the metal center releasing the coordinated nitrogen molecule from the cobalt The next step can proceed in one of two different directions First, the

coordinated tert-butyl azide can in a concerted fashion rearrange to coordinate N(1) while undergoing homolysis of the N(1)-N(2) bond which will release a second molecule of

nitrogen This would lead to the terminal tert-butyl imido species 2 The second option,

which is seen in the room temperature addition of tert-butyl azide to the metal, is a

cleavage of the carbon (1)-nitrogen (1) bond facilitating the formation of the azide adduct

TptBu,MeCoN3. A fast room temperature addition of the azide always leads to the

formation of the azide adduct; however, modifying the reaction conditions of tert-butyl azide with TptBu,MeCo(N2) can produce the terminal tert-butyl imido complex in high yield

Slow addition of a 0.33 M solution of tert-butyl azide in pentane via a syringe pump to a stirred solution of TptBu,MeCo(N2) in pentane at –10 oC resulted in a gradual color change from brown to green The rate of azide addition was maintained at no greater than 1 drop per second to allow for complete diffusion of the reagent into solution before the next drop was added Under these reaction conditions, nitrogen was not

observed evolving from the solution After the addition of the azide solution was

complete the reaction was slowly warmed to room temperature and the solvent was removed under reduced pressure The contents of the flask were dissolved in a minimal amount of THF, loaded on a celite column, and eluted with 100 % pentane As the column was eluted, two fractions were observed running down the column The first fraction which ran faster in pentane as a green band amidst the white celite was the desired product TptBu,MeCoNtBu The second blue band which ran slower than the

desired product TptBu,MeCoNtBu was the trace amount of the azide adduct TptBu.MeCo(N3) Subsequent work up of the collected green fractions by removing the solvent, re-

dissolving the residue in a minimal amount of pentane, and cooling to -35oC for 24 hours

produced green plates of 2, TptBu,MeCoNtBu, in 74% yield 1H NMR spectroscopy of the

Figure 1.12: Molecular structure of TptBu,MeCoNtBu, 2

Table 1.3 Selected interatomic distances in Å for TptBu,MeCoNtBu, (2)

Table 1.4 Interatomic angles in degree for TptBu,MeCoNtBu, 2

green fraction in C6D6 before crystallization confirmed the separation by use of the celite column, showing no trace of TptBu,MeCo(N3) The solid state structure of 2, as determined

by X-ray diffraction, can be seen in Figure 1.13 2 possesses many of the same

characteristics as 1 The Co(1)-N(7) bond distance of 1.660(3) Å and the

Co(1)-N(7)-C(25) bond angle of 179.4o provide evidence once again that the imido ligand is multiply bonded to the metal These coordination parameters are in good agreement with those of

1 as well as the other reported terminal cobalt imido complexes by Peters, Warren, and

Meyer et al.1d,14 2 also exists as a slightly distorted tetrahedron of C3V symmetry

Further bending of the Co(1)-N(7)-C(25) bond angle would result in unfavorable steric repulsions with the tert-butyl groups of the Tp ligand

After the synthesis and characterization of stable tertiary alkyl imido complexes were accomplished, our attention turned to the synthesis of secondary and primary alkyl imido complexes In an attempt to synthesize a secondary alkyl imido complex,

isopropyl azide40 was added to a pentane solution of TptBu,MeCo(N2) The reaction

proceeded with the evolution of nitrogen along with a color changed from brown to dark red Work up of the reaction consisted of filtering the solution, stripping off the solvent, dissolving the crude solid in a minimal amount of pentane, and cooling the solution to -

35oC for 24 hours After 24 hours the solution was filtered leaving behind a red tar which was washed with cold pentane Many attempts in isolating crystals suitable for X-ray diffraction studies failed 1H NMR studies of the red isolated tar at room temperature in

C6D6 showed a variety of unidentified paramagnetic resonances providing no structural information Reactivity studies of the red tar were conducted in an attempt to distinguish

if the isopropyl imido complex was synthesized If the imido complex was present, it is

plausible that it would react with carbon monoxide transferring its ligand and forming isopropyl isocyanate.41 When an NMR sample of the red tar in C6D6 was allowed to react at room temperature with an atmosphere of CO, there was no reaction observed Heating the sample to 80oC for days at a time had no effect on the reaction except to slowly produce a variety of new unidentified paramagnetic species

A continuation of the synthesis of terminal alkyl imido complexes directed attention to the use of methyl and ethyl azide.42 Because these reagents are unstable and not available commercially, they needed to be synthesized starting from sodium azide and either dimethyl or diethyl sulfate

(CH3O)2SO2 + NaN3 → MeN3 + (CH3O)SO3Na (CH3CH2O)2SO2 + NaN3 → EtN3 + (CH3CH2O)SO3Na

Due to the toxic and explosive nature of these reagents the use of both methyl and ethyl azide was discontinued shortly after their initial use in the synthesis of TptBu,MeCoNR (R

= Me, Et) and only the synthesis and structural characterization of the corresponding imido complexes were conducted.43 Note: Extreme care should be exercised when handling these reagents and alternative procedures should be attempted when possible Alkyl azides have been known to spontaneously explode upon distillation and ALL work should be performed behind a blast shield.42,43

When one equivalent of methyl azide in a cold pentane solution was added in one portion to a pentane or THF solution of TptBu,MeCo(N2), an immediate color change from brown to green occurred along with the simultaneous evolution of nitrogen gas After