Study and Utilization of Group 6 Metal Complexes

STUDY AND UTILIZATION OF GROUP 6 METAL COMPLEXES

ANDREA DENISE MAIN

A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY

UNIVERSITY OF FLORIDA

2000 Not many appreciate the ultimate power and potential usefulness of basic knowledge accumulated by obscure, unseen investigators who, in a lifetime of intensive study, may never see any practical use for their findings but who go on seeking answers to the unknown without thought of financial or practical gain.

-Eugenie Clark ACKNOWLEDGEMENTS

First, I am especially thankful for my partner. Dee Dee Mathews, for being there through thick or thin. I would also like to thank Linda Mathews for all of her love and support.

I thank my advisor. Dr. Lisa McElwee-White, for her support and encouragement through the years. Without her knowledge of organometallic chemistry, I would still be explaining metal-mediated reactions as some magical phenomena.

I would like to thank my friends who have helped to make this an extraordinary period of my life: Stacey Huber, Dr. Sue Troutman Lee, Dr. Sunday Brooks, Damon

Storhoff, Ann Storhoff, Holly Newman, Dr. Patricia Bottari, Dr. Margaret Kerr, Dr.

Yingxia He, Dr. Christine Nixon Lee, Kirsten Johnson, Fang "Annie" Qian, Dr. Mark

Tess and Dr. Rene Vieta.

Special thanks go to Larry Lee for all the hours of therapy. Although I think he should have been a psychologist, I am sure he will make a wonderful dentist.

I would like to thank my parents, George and Linda Main, for their love and support throughout my life. I am here today because of their dreams, which helped me to find my own.

I thank my Savior, the Lord Jesus Christ, for being there when things looked bleak. He showed me that life is not to be squandered.

Ill 21

TABLE OF CONTENTS

page

ACKNOWLEDGMENTS iii

LIST OF TABLES vii

LIST OF FIGURES viii

LIST OF SCHEMES ix

ABSTRACT xi

CHAPTERS

1 INTRODUCTION 1 General Overview of Low-Valent Metal Carbynes and Metal-Catalyzed

Carbonylation Reactions 1 Bonding 2 Protonation of Low-Valent Metal Carbynes 4 Protonation in the Presence of Alkynes 10

Photochemistry 1 Carbonylation with Transition Metals 20 Catalytic Carbonylation of Amines with Metal Carbonyl Complexes 21 Nickel Carbonyl Complexes 21 Palladium Carbonyl Complexes 22 Platinum Carbonyl Complexes 22 Cobalt Carbonyl Complexes 23 Rhodium Carbonyl Complexes 25 Iron Carbonyl Catalyst 26 Ruthenium Carbonyl Complexes 28 Manganese Carbonyl Complexes 30 Rhenium Carbonyl Complexes 3 Chromium, Molybdenum, and Tungsten Carbonyl Complexes 32

IV Conclusion 37

2 PHOTOPHYSICS OF LOW VALENT CARBYNES 38 Introduction 38 Photophysics of Metal Carbyne Complexes 39

Synthesis of Cp{P(OMe) 3 }(CO)WsCR Complexes 41

Electronie Strueture of Cp{P(OMe)3 }(CO)M=CPh (W=37, Mo = 38) 42 Eleetronie Absorption Spectra 44 Luminescence Studies 48 Transient Absorption Spectroseopy 53 Conclusion 58

3 SYNTHESIS AND PHOTOOXIDATION OF A TUNGSTEN CYCLOHEXENYL CARBYNE COMPLEX 59 Introduction 59 Synthesis of (Ti5-C5H5)(CO){P(OMe)3}W=C(c-C6H9) (44) 61 Photooxidation of (r|5-C5H5)(CO){P(OMe)3}W=C(c-C6H9) (44) 63

Oxidation of (T]5.C5H5)(CO){P(OMe)3}WHC(c-C6H9) (44) in the Presenee of Unsaturated Substrates 69 Other Reaetions with Unsaturated Substrates 71 Conclusion 71

4 CATALYTIC OXIDATIVE CARBONYLATION OF AMINES WITH W(CO)e 73 Introduction 73 Carbonylation of Primary Amines 75 Funetional Group Compatibility Study 79 Conclusion 85

5 ALTERNATIVE SYNTHESIS OF CYCLIC UREA HIV-1 PROTEASE INHIBITORS 87 Introduction 87 Carbonylation of Diamine 85 91 Carbonylation of Disubstituted Diamine 91 92 Conclusion 93

6 EXPERIMENTALS 94 General 94 General Instrumentation 94 Photophysieal Methods 95 Syntheses 95 [(C0)5WC(0)(2-napthyl)][NMe4] (42) 95

V Cl (CO){P(OMe)3}3W=C-(2-naphthyl) (41) 96 Cp(CO){P(OMe)3}W=C-(2-naphthyl) (40) 96 (CO)5\V-C(OCH3)(c-C6H9) (45) 97

[CO(Br) {P(OMe) 3 } 3W=X(c-C 6H9)] (49) 97 [Cp(CO){P(OMe)3}W=X(c-C6H9)] (44) 98 CpCl{P(OMe)3}W[Ti3-CH(c-C6H9)] (51) 98 Photooxidation of 44 99 General Procedure for the Catalytic Carbonylation of Benzylamines with W(CO)6 99 Procedure A 99 Procedure B 100 Procedure C 100 Preparation of Preparation of N,N’-dibenzylurea 55 100 Preparation of N,N’-Bis(4-chlorobenzyl)urea 57 101 Preparation of N,N’-Bis(4-bromobenzyl)urea 59 101 Preparation of N,N’-Bis(3-iodobenzyl)urea 61 101 Preparation of N,N’-Bis(4-methoxybenzyl)urea 63 101 Preparation of N,N’-Bis(4-methylthiobenzyl)urea 65 102 Preparation of N,N'-Bis(4-hydroxymethylbenzyl)urea 67 102 Preparation.3- of N,N’-Bis(ethyl 4-carboxylbenzyl)urea 71 102 Preparation of N,N’-Bis(4-carboxylic acid benzyl)urea 73 103 1.3- Preparation of N,N’-Bis(4-ethenylbenzyl)urea 75 103

Preparation1.3- of N,N’-Bis(4-nitrobenzyl)urea 77 104 Preparation of N,N’-Bis(4-cyanobenzyl) urea 79 104 Preparation of N,N’-Bis(4-aminobenzyl)urea 81 104 Preparation of 3,4-Dihydro-2(l//)-quinazolinone 83 104 Preparation of (4R, 5R)-2,2-Dimethyl-4,5-bisformyl- dioxolane N,N-Dimethylhydrazone 88 105 Preparation of (3R, 4S, 5S, 6R)-2,2-Dimethyl-4,5-bisformyl- dioxolane-3,6-bis(phenylmethyl) N,N-Dimethylhydrazine 89 105 1.3- Preparation of (2R, 3S, 4S, 5R)-2,2-Dimethyl-3,4-bisformyl-

1 dioxolane-2,5-bis(phenylmethyl)diamine 85 106 Preparation of (4R, 5S, 6S, 7R)-Hexahydro-5,6-(9-

isopropylidene-4,7-bis(phenylmethyl)-2//- 1 ,3-diazapin-2-one 86 106 Preparation of (2R, 3S, 4S, 5R)-2,2-Dimethyl-3,4-bisformyl-l,3- dioxolane-2,5-bis(phenylmethyl) N,N-Dibenzyldiamine 91 106 Attempted preparation of (4R, 5S, 6S, 7R)-Hexahydro-5,6-0- isopropylidene-4,7-bis(phenylmethyl)-N,N-dibenzyl- diazapin-2-one 90 107

REFERENCES 108

BIOGRAPHICAL SKETCH 120

VI LIST OF TABLES

Table page

2.1 Absorption Spectra - of Cp(CO){P(OMe)3 }M=C-R (37 40) in THF Solution 47

2.2 Luminescence and Transient Absorption Data

for Cp(CO){P(OMe) 3 }M=CR 49

4. 1 Catalytic carbonylation of primary amines in single/ biphasic solvent systems 78

4.2 Catalytic carbonylation of substituted benzyl amines 81

vii 1

LIST OF FIGURES

Figure page

I . I Partial diagram for MnsC-Me]'' MO [Cp(CO)2 3

1 Radical cation A prefers bent geometry 20

2. 1 Orbital diagram for mixing Cp(CO){P(OMe)3 } W=C-Ph (37) 43

2.2 HOMO and LUMO of Cp(CO){P(OMe) 3 }WsC-Ph (37) 45

2.3 Absorption spectrum of Cp(CO){P(OMe) 3 }WsC-Ph in THF solution at room temperature 46

2.4 Emission and excitation spectra: (a) for 37 and 39 and (b) for

40 . Emission spectrum is at right and excitation at left 5

2.5 Transient absorption difference spectra following laser excitations (355 nm, 10 ns fwhm, 10 mJ/ pulse) 56

3.1 Comparison of '^C NMR data between 51 and known 52 67

3.2 Electrospray mass spectral data of 51 68

3.3 Cyclic voltammogram of 44 . Sweep rate: 100 mV/s 70

3.4 Investigated reactivity of 53 with unsaturated substrates 71

5.1 HIV Protease inhibitors 87

viii 94671

LIST OF SCHEMES

Scheme page

1 . 1 Protonation of Cp(CO)2W=CTol (7) with HBF4 6

1 .2 Protonation of Cp(CO)2W=C-Tol in the presence of alkynes 1

1 .3 Suggested precursors to naphthol complex 18 12

1 .4 Photochemical migration of CO to carbyne carbon in Cp(CO)2W=C-Tol (7) 13

1.5 Photochemical isomerization of cis, cis- X(CO)2(PR3)2M=C-Ph 1

1 .6 Metal-centered reactivity of photooxidized metal carbynes 1

1 .7 Selected organic products from the photooxidation of carbynes 1

1.8 Mechanisim of photooxidation of Cp(CO){P(OMe)3 }MosC(c-C 3 H 5 ) 18

1 .9 Proposed formation of olefins 1

1.10 Reaction for mechanism the carbonylation of amines with Fe(CO)5 28

1.11 Proposed two step reaction mechanism 3

1.12 Postulated mechanism for the carbonylation of

primary amines in the presence of Mn2 (CO)io 32

1.13 Stoichiometric carbonylation of amines in the presence of a tungsten carbamoyl complex 34

1.14 Proposed mechanism for the carbonylation of amines in the presence of 32 36

IX 3.1 Proposed mechanism for generation of metal carbenes upon photooxidation of metal carbynes 60

3 .2 Synthesis of (r| ^-CsHjXCO) {P(OMe) 3 } W^CCc-CgHg) (44) 62

3.3 Proposed mechanism for the protonation of carbynes 65

5.1 Synthesis of protease inhibitor core structure via an amino acid 88

5.2 Preparation of the diamine acetonide 85 89

5.3 Synthesis of core structure starting from the ester of L-tartaric acid 90

5.4 Dupont process scale-up of cyclic urea 90 91

X 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

STUDY AND UTILIZATION OF GROUP 6 METAL COMPLEXES

Andrea Denise Main

December 2000

Chairman: Lisa McElwee-White Major Department: Chemistry

This dissertation describes the preparation, reactivity, and photochemistry of a

series of low-valent == == carbyne complexes, Cp(CO){P(OMe) 3 }MsC-R (M Mo, W; R

aryl, alkyl). These carbynes were prepared from Mo(CO)6 or W(CO)6 respectively.

Photolysis of these carbyne complexes in chlorinated solvents leads to the generation of

highly reactive 17 e" species, which are formed by one-electron transfer from the

complexes to the solvent. The precursors to these 17 e' species are the excited state of the carbynes. Investigations of the excited states were performed using absorption, emission, and transient absorption spectroscopy.

XI 7

Previous work done by the McElwee-White group had demonstrated that the photooxidation of carbyne complexes of the type [(rj^-C5H5)(CO){P(OMe)3]M=C-R]

(M = Mo, W; R = alkyl) results in rearrangement and decoordination of the carbyne ligand to yield organic products. Mechanistic studies suggested that following formation

of the 1 e' species mentioned above, H-abstraction took place at the carbyne carbon to

yield a carbene. The first direct evidence for the formation of carbene complexes in these

reactions was provided by the study of the cyclohexenyl complex [(r|^-

C5H5)(CO){P(OMe)3]W=C-C6H9] Ifom W(CO)6 . Photooxidation of the carbyne complex afforded the carbene complex [(r|^-C5H5)(Cl){P(OMe)3]W=CH-C6H9].

Following the use of W(CO)g in the preparation of carbyne complexes, it was then explored as a catalyst for the carbonylation of amines to ureas. Previous McElwee-White

group members had demonstrated that W(CO)6 serves as a catalyst for carbonylation of primary and secondary amines to ureas. Primary and secondary diamines can be converted to cyclic ureas by the same process. This dissertation describes the study of functional group compatibility with the catalyst during the carbonylation reaction. A series of substituted benzyl amines were synthesized and carbonylated to the corresponding ureas under the W(CO)e system. Single and biphasic solvent systems were investigated which revealed that higher yields could be obtained in the presence of the biphasic solvent system, CH2CI2 and H2O.

This dissertation also describes the synthetic application of the W(CO)6 system towards the synthesis of the cyclic urea core structures of HIV protease inhibitors DMP

323 and DMP 450 .

xii CHAPTER 1 INTRODUCTION

General Overview of Low-Valent Metal Carbvnes and Metal-Catalyzed Carbonvlation Reactions

Metal carbyne complexes have been extensively explored since their first introduction into the scientific realm in 1973. The first carbyne complexes were low-

valent and of the type X(CO)4MsCR (X = Cl, Br; M = Cr, Mo, W; R = Ph). Although carbynes of groups 5, 7, 8, and 9 have since then been reported as well as carbynes with various R groups, group 6 metal carbynes remain the most common. The constant interest in metal carbyne complexes is due to their potential as organic synthons and their ability to mimic intermediates in catalytic reactions. ^ Three general reviews by H.

Fischer et al.,^ Kim and Angelici,^ and Mayr and Hoffmeister^ have comprehensively covered the synthesis, characterization, and reactivity of metal carbynes up until 1990.

Other shorter reviews have covered more specific subjects: low valent group 6 metal

^ O' carbynes,6-9 bridging carbyne ligands, 12 metal carbynes with carbaborane ligands, 13 ruthenium and osmium metal carbynes, 14 photochemistry,! 3 high valent metal

»16-1 carbynes,! 8 and a survey of the coupling reactions of n-ligands including carbyne

ligands by Mayr and Bastos.19 The most current survey of metal carbyne chemistry is the 1995 review by Engel and Pfeffer.20 Due to the extensive reviews already available

1 2

for metal carbynes, this chapter will cover literature after 1995 with emphasis on the reactivity and photochemistry of metal carbyne complexes.

There are a number of different synthetic routes to metal carbyne complexes. One

route incorporates a metal carbonyl complex as the starting material.^ 1 Metal carbonyl complexes have not only been useful in preparation of organometallic complexes but also have been found to transform organic substances into other species. Hence, this chapter will also include a review of metal carbonyl complexes that have demonstrated their ability to carbonylate organic compounds, particularly amines.

Bonding

In order to understand the reactivity of low-valent metal carbyne complexes, it is useful to look at the nature of the chemical bonding in these complexes. Fenske analyzed the M=C-R triple bond in neutral and positively charged low-valent carbyne complexes using Fenske-Hall22 approximate SCF calculations.23-25 projj^ tpggg calculations, the a bonds were found to be strong and the n bonds degenerate or nearly degenerate between the metal (i.e. chromium, manganese, and iron) and the CR ligand. Kostic and Fenske ascertained from these calculations that the carbyne ligand was a stronger n acceptor than

CO resulting in increased electron density on the carbyne carbon.24 a typical MO diagram of n interactions is shown in Figure 1.1. The cyclopentadienyl ligand mixes into the former d orbitals to give three non-degenerate molecular orbitals in the metal fragment. However, the HOMO of the resulting complex is still primarily metal-based, and the LUMO is still a n* antibonding orbital. 3

Further literature includes an ab initio study of a metal carbyne complex, that of

tra«5-Cl(CO)4Ci=CH, along with correlation energies published in 1986 by Poblet et al.

These studies showed that the degenerate Cr-C n bond is essentially nonpolar, while the

CT bond is polarized toward the carbyne carbon, which leads to a partial negative charge at

the carbyne ligand.26 The negative polarity of the carbyne carbon explains the facile

protonation of many complexes at the carbyne carbon. Protonation of metal carbynes is said to be charge-controlled. However, in some instances, nucleophiles have been known to attack the carbyne carbon. This mode of reactivity is said to be controlled by frontier molecular orbitals. The HOMO of the nucleophile donates its electrons into the LUMO

of the carbyne complex, even though the LUMO is slightly negatively charged.27

Figure 1.1 Partial MO diagram for [Cp(CO)2MnsC-Me]^ 4

Protonation of Low-Valent Metal Carbvnes

Following the synthesis of the first metal carbyne in 1973 , it was four years

before material was published describing the reaction of metal carbyne complexes with

electrophilic reagents. Bottrill and Green reported the protonation of the electron rich

carbyne [Mo(=CCH2‘Bu){P(OMe)3}2(r|-C5H5)] ( 1 ) with HBF4 which led to the formation

the of cationic hydrido species [MoH(=CCH2^Bu){P(OMe)3}2(ri-C5H5)][BF4] (2) (see

equation 1 . 1).28 The authors were unable to determine the intial site of protonation.

Protonation could occur directly on the molybdenum center, or the carbyne carbon could

be protonated first, followed by hydrogen transfer to the metal, otherwise known as an a-

hydride elimination.

-1+ BF,

\ HBF^ M^C-CH‘Bu (MeO)3P"^o= C-CH‘Bu H- ( 1 . 1 ) (MeO)3P (MeO)3P^ P(OMe)3

Numerous reports of protonation reactions followed, showing not only

protonation at the metal but also at the carbyne carbon. For example. Holmes and

Schrock reported the protonation of [W(=CH)(PMe3)4Cl] ( 3 ) with CF3SO3H which gave

the ligand protonated product [W(=CH2)(PMe3)4Cl][ CF3SO3] (4) (equation 1.2).

However, by changing the ancillary ligands to the bridging diphosphine

Me2PCH2CH2PMc2, protonation occurred at the metal center rather than at the ligand as was in seen the former case (equation 1 . 3).29 5

H ' CF 3 S03 I H. H + C ( 1 . 2 ) ,PMe, ^PMc3 CF3SO3H Mc3l^ r ..w.: " vsx"' MC3P" PM63 Me,P' PMe, I Cl Cl

' CF S03 H H n 3

Mej MC2 MC2 Men CF3SO3H P'*//,.

C C 1 . 3 *p ( ) P^ I ^P Men T MC2 Men Men Cl H

Subsequent chemistry revealed that changing the counterion of the acid could greatly influence the course of the reaction. Stone protonated Cp(CO)2 WsCTol (7) (Tol

= tolyl) with HBF4 to yield the cationic dinuclear p-alkyne complex 10. The reaction was proposed to proceed through the cationic carbene [Cp(CO)2 W=C(H)Tol]^ (8) (see

Scheme 1.1). 6

Scheme 1.1 Protonation of Cp(CO)2WsCTol (7) with HBF4

+ BF. Tol \ qqi.^W=C-To1 HBF. OC ""V H oc CO

+ + BF. Tol. BF. \ /Tol Tol \ / 0O/-.,„.^w=cf ^ Cp(CO)2W^ /W(CO)2Cp CO H H Cp(CO)2W=C-Tol

The intermediate 8 was never detected but was inferred by the formation of

Cp(CO)2(I)W=C(H)Tol (11) when the reaction was run in the presence of iodide

(equation 7

(1.4)

However, protonation of Cp(CO)2 WsCTol in the presence of HCl instead of HBF4

resulted in double protonation of the carbyne carbon to form an alkyl ligand that

subsequently yielded the r)^-acyl complex 12 upon CO insertion (equation 1.5).32,33

\ 2 HCl OC'^^^^C-Tol (1.5) CH2T01 O'" Cl OC Cl

Another example of counterions affecting the reaction pathway is the protonation of Cp{P(OMe)3}2MosCCH2‘Bu ( 1 ) with CF3COOH. Unlike the reaction of 1 with HBF4

(equation 1.1), CF 3 COO' coordinates to the molybdenum center resulting in loss of the carbyne ligand.24

It was also discovered that protonation could take place across the face of the carbyne complex resulting in an agostic interaction of the C-H bond with the metal (for

1.6).25,36 example, see equation This type of reactivity is better understood by taking into account that the probability of hydrogen migration from the carbyne carbon to the 8

metal center increases when there is little n-acceptor character in the ligands. If

hydrogen migration did occur in complexes with strong 71 acceptor ligands like CO, the

formation of the localized M-H bond would eliminate n backbonding from the metal

center back into the n acid ligand, resulting in ligand loss. Thus, 71 -acceptor ligands

inhibit full migration of the hydrogen to the metal center due to competition for the d

electrons resulting in agostic interactions.^

CO CO ,^PMC3 .j.PMe3 ^ Me C1-^>V=C—Me HCl .W; ( 1 . 6 ) Me3P Me3P PMe, H Cl

13 14

Consistent with the explanation above, McElwee-White reported the protonation

of the carbyne CpL MosCBu [L = 2 (15) CO, P(OR) 3 ] with HBF4 (equations 1.7-1.9)37 which resulted in a systematic shift of the hydrogen from the metal center to the face to the carbyne ligand as the number of carbonyl groups were increased from zero to two.

These results are consistent with determination of the final protonation site by the

electron density at the metal. As mentioned before, when 7i-backbonding to the carbonyl ligands decreases the electron density at the metal, the protonation site shifts from the metal toward the carbyne carbon. 9

+ BF,

\ HBF^ .M CBu (MeO)3P'^¥ ^CBu (MeO)3P"y (1.7) H i (MeO)3P^ (MeO)3P

15a 16a

16c 15c 10

Protonation in the Presence of Alkvnes

Metal carbenes have found numerous applications in organic synthesis38-41 anj

one important class of reactions is one that occurs between carbene complexes and

alkynes. The product isolated depends on the metal, the ligands, the substituents on the

carbene, and the reacting alkyne. Shown in equation 1 . 1 0 is a typical example of a metal

carbene reacting with an alkyne.

The reaction of metal carbenes with alkynes has been extended to the same type of chemistry with metal carbynes. As mentioned earlier, protonation of metal carbynes often leads to the intermediacy of a metal carbene. Analogous to the reaction in equation

1.10, metal carbynes have been protonated in the presence of alkynes to form addition

products. For example, when Cp(CO)2WsC-Tol (7) reacts with HBF4 and acetylenes, depending on the R groups on the acetylene and the presence of carbon monoxide, q^- allylidene (17) or q"-naphthol (18) products result (Scheme 1.2).42 Jhe naphthol 11

complex 18 is thought to form by either the r|'‘-ketenyl complex 19a or the metallacycle

19b (Scheme 1.3), both of which are derived from protonation of the metal carbyne at the

ligand, followed by insertion of the alkyne into the MsC bond. This is still a relatively new field of ehemistry and could lead to a number of different products, depending on the carbyne, the aeid, and the unsaturated substrates used.

Scheme 1.2 Protonation of Cp(CO)2 W=C-Tol in the presence of alkynes

18 12

Scheme 1.3 Suggested preeursors to naphthol complex 18.

Photochemistry

Relatively little is known about the photochemical properties of metal carbyne complexes. The electronic character of ligands attached to metals can undergo significant change by photoexcitation, especially if the excited states involving those ligands are populated. Carbyne complexes should have a rich photochemistry due to the unsaturated character of the metal-carbyne moiety and the presence of long-lived and low-lying metal-to-carbyne ligand-charge transfer (MLCT) excited states.43 There are a number of examples in literature where photoexcitation of metal carbyne complexes has led to various reaction pathways that would be unavailable to the carbyne in the ground 13 state. The most common photochemical processes among metal carbyne complexes

involve carbonyl ligands. Mayr et al.44 and Geoffroy et al.45 have established that carbonyl-carbyne coupling is an important photochemical reaction pathway in some

systems. Photolysis of the electron rich Cp(CO) 2 \VsC-Tol in the presence of phosphines

- gives r\ and r| -ketenyl complexes, depending on the phosphine employed (Scheme

1.4).

Scheme 1.4 Photochemical migration of CO to carbyne carbon in Cp(CO)2 \V=C-Tol (7)

Cp(PPh3)(CO)W—

7 Cp(CO)(dppe)W- 20 . Tol PPhj 21 hv

[Cp(CO) W=C-Tol] Cp(CO)W— 2 C.^ r\

r\ -ketenyl MLCT Excited 22 Cp(CO)2W— Tol

O O Cp(CO)W— C— Tol CO

23 24 14

The mechanism is not completely understood, but the increased electron density at the carbyne carbon in the MLCT state is thought to facilitate migration of the CO ligand and subsequent formation of a 16 e’ ketenyl complex. Further reaction of the ketenyl complex with phosphines leads to the products 20 and 21. In the photoassisted cis-trans isomerization of complexes of the type X(CO)2L2MsC-Ph (X = Cl, Br; L =

= PMes, PPhs; M Mo, W), Mayr has deduced the presence of q ^-ketenyl intermediates

(Scheme 1.5). In the photolysis of cis 25, the reaction forms a five-coordinate ketenyl species in which the phosphines subsequently rearrange to be trans to each other, minimizing steric forces, followed by migration of the carbonyl back to the metal center to give trans 26.

Scheme 1.5 Photochemical isomerization of cis, cis- X(CO)2(PR3 )2M=C-Ph

CO 15

Another interesting reaction found in literature is the photolysis of

Cl(CO) 2 (PMe)3 WsCPh with excess phenylacetylene which gives the p'-ketenyl complex

W(Ti'-PhCCO)Cl(CO)(PhCCH)(PMe3)2 45

Photochemical electron transfer has been shown to activate 18 e" complexes

toward reactivity. Due to differences in reactivity between open- and closed-shell

organometallic complexes, reaction pathways unavailable to neutral 1 8 e' complexes can

often be accessed by the ensuing 17 e' radical cations.46-48 -phis strategy has proven

effective for metal carbynes. Over the past several years, the McElwee-White laboratory

has been investigating photooxidation reactions involving irreversible electron transfer

from a series of tungsten and molybdenum aryl- and alkylcarbynes to halogenated

solvents.49 Early experiments involved photolysis of the carbynes CpLiL2MsCR 27 [Li,

L = = = 2 CO, P(OMe)3 ; R alkyl, aryl; M Mo, W] in CHCI3 solutions. When the reactions

were run in the presence of PMe 3 to stabilize any unsaturated intermediates, new cationic carbyne complexes resulted as shown in Scheme 1.6.49,50 Mechanistic studies showed the primary photoprocess to be electron transfer from the excited state of the carbyne to

CHCI3. Upon reduction, CHCI3 undergoes fragmentation to form chloride ion and dichloromethyl radical resulting in irreversible electron transfer. The ensuing 17-electron

27"^* cationic carbyne then undergoes metal-centered reactivity involving ligand exchange

with the more strongly donating PMe3 to form 28 followed by halogen abstraction to yield the final cationic carbyne 29. 16

Scheme 1.6 Metal-centered reactivity of photooxidized metal carbynes.

hv + + Cl- + 27 •CHCI2 CHCI3 Li 27

-I- PMe3 cr 27’

PMe3

CHCI3 28 cr + •CHCI2

In the absence of strongly nucleophilic species such as PMea, the photolysis of carbyne

complexes results in very different reactivity. If the carbyne substituent is a primary or secondary alkyl group, photooxidation yields carbyne ligand-centered reactivity resulting in organic products (Scheme 1.7). 17

Scheme 1.7 Selected organic products from the photooxidation of carbynes.

Mechanistic studies revealed that the reactions in Scheme 1.7 share the same first two mechanistic steps. Scheme 1.8^^ shows these first two steps in the transformation of of Cp(CO){P(OMe)3}MosC(c-C3H5) (30) into cyclopentenone. After photoexcitation of

30, the first step involves one electron transfer from the excited state of the carbyne to the solvent, forming the 17-electron radical cation A. In the second step, the radical cation A abstracts a hydrogen from the reaction medium to form the cationic carbene B which undergoes ring expansion to form the metallacycle C. The mechanism then proceeds with insertion of CO into the metal-alkyl bond and coordination of Cl’ to form D.

Reductive elimination of the organic ligand produces the cyclopentenone. 18

Scheme 1.8 Mechanisim of photooxidation of Cp (CO){P(OMe) 3 }MosC(c-C 3 H 5 )

"H* \ MoczC OC' / P(OMe)3 B 19

Similarly, in the formation of 1 -butene from the photolysis of

Cp(CO){P(OMe) 3 }MosC-«-C4H9 (31), the first step is electron transfer from the excited

state to the solvent, which forms the seventeen-electron radical cation A, followed by H-

abstraction at the carbyne carbon to yield the cationic carbene complex B (Scheme

1.9).52 However, at this point, B undergoes an H-shift rather than CO insertion to form

the cationic rj^-olefin complex C, which subsequently eliminates to form the terminal

olefin.

Scheme 1.9 Proposed formation of olefins

hv ,.Mo^C CH2C^7 Mcf=C— CH7C3H7 — (MeO)3P' (MeO)3P'Y CHCI3 y CO CO 31 A

+ RH -R*

+ 1 H H X H-shift (MeO)3P"^^o^— n .MctrC. (MeO)3P i ^CH2C3H7 CO A H C3H7 CO c L B J

It should be noted that the H-abstraction process (A^B) for an organometallic radical is highly unusual. Reactivity at an organic ligand is known to occur in metal radicals but reaction at the metal center is much more common.^^ When ligand-centered reactions do occur, they are almost always dimerization processes. In order to explain the ligand’s affinity for the hydrogen, INDO calculations were carried out on the carbyne 20

radical cation A as the angle (|) was varied from 180° to 120°.^2 These calculations

determined that the radical cation A favors a bent carbyne configuration (Figure 1 .2) over

the original linear carbyne geometry. The geometry change causes spin density to reside

on the carbyne carbon.

A Preferred geometry

Figure 1.2 Radical cation A prefers bent geometry

This type of chemistry has resulted in the formation of a variety of olefins, dienes, dienals and other cycloalkenones from the photooxidation of various alkyl metal carbynes.

Carbonylation with Transition Metals

Carbonylation of organic compounds in the presence of transition metal catalysts has become a topic of increasing interest. One facet of this chemistry is carbonylation involving metal carbonyl complexes.^4,55 There are numerous reactions throughout the literature that invoke metal carbonyl complexes in the catalytic carbonylation of organic

compounds. There are reviews of this type of chemistry56,57 and a multitude of articles describing catalytic transformations such as olefins to aldehydes^S or carboxylic acids,^9 21

arenes to aromatic aldehydes,^^ aryl halides to aromatic acids, alkyl halides to

62 esters, nitroaromatics to carbamates,63 and alkynylben2ylalcohols64 or epoxides^^ to

cyclic carbonates. Typically, late transition metals, groups 8-10, are used to catalyze

these transformations, although there has been one gold(I) carbonyl catalyst (group 12)

reported.66

Catalytic Carbonylation of Amines with Metal Carbonvl Complexes

Further example of these reactions is the catalytic carbonylation of amines with

metal carbonyl complexes to form isocyanates, ureas, and carbamates. There is

continued interest in deyeloping new synthetic routes for such compounds for multiple

reasons: isocyanates are used as precursors for polyurethanes and polyureas;67 ureas are used in the pharmaceutical industry as well as in agricultural products;68 and carbamates not only act as chiral auxilaries, but also are found in pharmaceuticals.69 Furthermore, interest in new synthetic routes continues in an effort to find altematiyes to the traditional method of carbonylating amines using phosgene, a highly toxic and corrosiye gas.20

Nickel Carbonyl Complexes

Initial studies of the transition metal catalyzed carbonylation of amines date back to World War II where nickel salts were added to amines and the catalytically actiye species were undoubtedly metal carbonyls.21"23 Howeyer, subsequent carbonylation 22

reports of amines using Ni catalysts are few with very low product yields,^ 1 and none of

the catalysts are carbonyl complexes.

Palladium Carbonyl Complexes

Palladium catalysts are the most commonly used transition metal catalysts for the

carbonylation of amines. Although the actual catalysts are not carbonyl complexes, a

significant amount of the reactions are thought to proceed through palladium carbonyl

intermediates. For instance, palladium acetate promoted by iodine can carbonylate

amines in the presence of alcohols to carbamates and is thought to proceed through a

"74 "75 Pd(0)(carbonyl) complex. In the absence of alcohols, ureas are formed.

Mono- and di-methoxycarbonyl complexes of palladium of formula [PdL2Cl2 -

= 1 n(COOCH3 )n][n or 2; L2 = bipyridine or 1 , 1 0-phenanthroline] have been found to carobnylate amines to ureas at room temperature under CO pressure. In the presence of

CuCl2 , the reaction generates carbamates rather than ureas.^^

Furthermore, a number of palladium catalysts that also carbonylate amines to ureas and carbamates have been reported by Kanagasabathy and Chaudhari,”77 Alper et al.,^8,79 Valli and Alper.^^

Platinum Carbonyl Complexes

The one example in the literature using platinum carbonyls to carbonylate amines is the photolytic reaction of Pt(CO)2 (PPh3)2 with morpholine and cyclohexyl iodide under 23

1 atm of carbon monoxide at room temperature which gives the corresponding amide in

75% yield (see equation 1.12).81

75% 1 atm CO

(1.12'l

Cobalt Carbonyl Complexes

The earliest report of a cobalt carbonyl complex carbonylating amines was in

1953 by Sternberg who generated formamides in 60-78% yields from the carbonylation of secondary amines, dimethylamine and piperdine, with Co2(CO)8 under 200 atm of

carbon monoxide at 200 °C (equation 1.13).^2 i^gy intermediate in this reaction is assumed to be a carbamoyl species H(CO)nCo-CONR2 which produces the formamide upon reductive elimination.

+ [Co”(piperdine)6 2 ] [Co2(CO)g] 200 atm CO 78% (1.13)

Other reports of Co2(CO)s-catalyzed processes include the carbonylation of allyl amine under even more severe conditions with 300 atm of carbon monoxide at 280 °C which afforded 2-pyrrolidine (equation 1.14). Although run at room temperature, \

photolysis of this reaction later proved even more inefficient with low yields, and product mixtures containing 2-pyrrolidone, N,N'-diallylurea, and N-allyl-3-butenamide. The photochemical cleavage of C-N bonds of amines coordinated to Co was proposed.^3

+ CO (1.14) (300 atm) — - H

There are a few examples of a metal carbonyl compound catalyzing the ring- opening and carbonylation of aziridines. The ring strain present in three-membered ring compounds suggests ring-opening/ carbonylation could take place under relatively mild conditions. One of those examples is the treatment of 1,2-disubstituted aziridines with 33

of carbon atm monoxide and Co2(CO)8 in DME (1,2-dimethoxyethane) for 24 h at 100 °C to yield P-lactams in up to 95% yield. The reaction is regiospecific with insertion of a

carbonyl into the least substituted of the two carbon-nitrogen bonds in the ring.^'^

R R' R. .R"

'.II R. Co2(CO)g, DME, CO (33 atm) (1.15) N 100 °C, 24 hr / N— CH2CH2Ph PhCH2CH2 O 25

Rhodium Carbonyl Complexes

When subjected to the rhodium catalyst [Rh(CO)2Cl]2, primary amines are

converted to the corresponding formamides or ureas, depending on the conditions. For

instance, «-butyl amine forms a mixture of Wbutylformamide and di-«-butylurea in the

presence of the catalyst under 59 atm of carbon monoxide at 160 °C (equation 1.16).

However, when the reaction is carried out in the presence of PMes, Wbutylformamide is

formed in quantitative yield.

[Rh(CO)2Cl]2

+ 1 . BuNHCHO BuNHCONHBu ( 16 ) BuNH2 60 atm CO 35 % 65% 160

In contrast to the severe temperatures and pressures required for the Co2(CO)g- catalyzed carbonylation of allyl amines, Rh4(CO)i2 and [Rh(CO)2Cl]2 carbonylated butenyl amine under milder conditions to give 2-piperidinone, exclusively (equation

1.17). The conditions used in these rhodium-catalyzed reactions include 400 psi of

CO/H2 and temperatures of 40-50 °C. Yields varied from 85-87%.86

+ CO/H2 [Rh] (1.17) 26

Pyrrolidinones were isolated as the only products when aliphatic iV-allyl- or N-

(methylallyl)amine was carbonylated in the presence of HRh(CO)(PPh3 )3 . The reaction

requires the addition of NaBH4 and isopropanol, and takes place under 35 atm of carbon

monoxide at 100 °C. Specifically, N-allylcyclohexylamine afforded N-cyclohexyl-2-

pyrrolidinone in yield 78% (equation 1.18). The yield decreases appreciably ifNaBH4 is

substituted by other borohydrides.^^

CO, HRh(COXPPh3)3

(1.18) NaBHq, i-PiOH, CH2CI2 35 atm, lOO'^C

Additional Rh reports include the catalyst Rh(CO)2(acac) which carbonylates

aniline under 41 atm of a CO/ O2 gas mixture, giving A,iV-diphenyl urea (74%

selectivity) and A-phenyl carbamate (24.5% selectivity). When Nal, acting as a

promoter, is added to the reaction, there is an increase both in the activity of the catalyst

and the selectivity for the carbamate.

Iron Carbonyl Catalyst

Formamide complexes can be obtained by the reaction of piperidine with Fe(CO)s

(equation 1.19). However, this reaction is limited to piperidine. When pyrrolidine or 27

pyridine is reacted under the same conditions, iron-amine complexes form.^^

Furthermore, when the catalyst reacts with ammonia, urea is formed.

(1.19)

Dombek and Angelici later repeated this chemistry with the addition of CO and

found that they could carbonylate primary (cyclohexylamine) and secondary (piperidine)

amines to the corresponding formamides under 95 atm of CO at 200 °C in 66-83% yields,

respectively. Their proposed mechanism is shown in Scheme 1.10. Reaction A involves

the formation of a carbamoyl complex anion, Fe(C0)4[C(=0)NR2]'. This reaction is

rapid and reversible at room temperature with piperdine, pyrrolidine, and «-butyl amine.

Upon heating at 50 °C for 50 hr, the carbamoyl complex to decomposes Fe(CO)4 (HNR2) and the formamide, HCONR2 . The authors propose that this decomposition occurs via steps B, C, and D. Step B involves proton transfer from the ammonium cation to the iron atom giving the hydride complex which upon reductive elimination, generates the formamide in step C. The resulting coordinatively unsaturated species Fe(CO )4 then reacts with excess amine to the give the final observed product, Fe(CO)4 (HNR2 ). Further

reaction of Fe(CO)4 (HNR in the catalytic reaction 2 ) with CO could generate Fe(CO)5 as is shown in step E.^O 28

Scheme 1.10 Reaction mechanism for the carbonylation of amines with Fe(CO)5

A 2HNR2 Fe(CO)s — OC— Fe— CO H2NR2 ^

oc I CO

E B CO -HNR2

NR2 Fe[CO]4 H

CO D CO HNR2

Fe(CO)4(HNR2)

Ruthenium Carbonyl Complexes

As described earlier, Co2(CO)8 and Fe(CO)s can catalyze the carbonylation of piperdine to the formamide, but under severe reaction conditions. The ruthenium catalyst, Ru3(CO)i2, can convert piperdine to the formamide under much milder

conditions (75 °C, 1 atm CO).^l

Typically, secondary amines are more easily converted to formamides than primary amines. Stringent conditions are usually required for primary amines. For 29

example, Kealy and Benson reported that carbonylation of the primary amine,

cyclohexylamine, in the presence of allene gave A^-cyclohexylformamide using

Ru3(CO)i 2 as the catalyst. This transformation, however, required extremely high carbon

monoxide pressure 92 (1000 atm). Watanabe also used Ru3 (CO)i 2 to catalyze the

carbonylation of various primary amines to the corresponding formamides with much

lower levels of CO but rather high temperatures (39 atm, 180 °C).93 Again, a carbamoyl

species is suggested as the key intermediate in the formation of formamides. It is postulated that the carbamoyl complex is formed by either an intermolecular nucleophilic attack of the amine on the metal carbonyl ligands or by an intramolecular 1,2-shift reaction between coordinated carbon monoxide and amine. The authors could not isolate the carbamoyl species but its presence was evident in the IR.

Aniline can be carbonylated in the presence of various ruthenium carbonyl complexes, such as Ru(CO)3l3, Bu4N[Ru(CO)3l3], and Ru(CO)2(py)Cl2^^’94 to give N- benzylcarbamate or N,N'-diphenylurea. The carbamate is formed in moderate yields with

a 13:1 mixture of CO and O2 in methanol at 170 °C.

Cyclic amines can be carbonylated to the formamides with ruthenium cluster

compounds, [HRu3 (CO)n]‘ and [H 3 Ru4(CO)i 2 ]’. Both carbon monoxide and a mixture of hydrogen and carbon dioxide gasses are used. A neutral carbamoyl cluster complex was isolated and thought to be a possible intermediate.^® 30

Manganese Carbonyl Complexes

The manganese carbonyl complexes, Mn2 (CO)io and CH Mn(CO) have also 3 5 ,

been explored for the carbonylation of amines. Again, stringent conditions are

required with reactions run between 180 and 200 °C under 130 atm of CO. In these

reactions, primary amines gave dialkylureas exclusively, contrary to previous reactions

which primarily form formamides. Also, it should be noted that ammonia, secondary and

tertiary amines, and aniline react very slowly, with the corresponding urea being the

primary product in each case with little formation of the formamide. Calderazzo

proposed that the poor yields of formamides in these reactions were due to the over-all

reaction occurring in two steps with ^2 being probably much larger than k\ (Scheme

1.11). If k\ were larger than ki, little or no dialkylurea would be formed. Calderazzo

concluded that since both steps occur in the coordination sphere of the metal, the

condition for the formation of the urea is that the formamide formed in the first step

remains on the metal and undergoes a fast nucleophilic attack on the formyl carbon atom by a second molecule of amine.

Dombek and Angelici later explored the reaction mechanism more thoroughly.^'^

By studying several stoichiometric reactions, they proposed the following mechanism which is detailed in Scheme 1.12. The authors admit that there are possible alternative mechanisms for individual steps but overall, given the supporting evidence. Scheme 1.12 best describes the carbonylation of primary amines with Mn2(CO)io. There are three key steps in the mechanism. The first describes the formation of the carbamoyl complex

from the reaction of the amine with Mn 2 (CO)io. The second step depicts the reaction of 31

the amine and CO with the carbamoyl complex to give the urea via the organic

isocyanate. In the third step, Mn2(CO)io is regenerated from Mn(CO)5’ and RNHs^.

In addition, HMn(CO)5, an intermediate in the reaction, was also found to catalyze the

reaction, probably by a similar mechanism.

Finally, Srivastava and coworkers have used the manganese complex, (t]-

- CH3C5H4)Mn(CO)3, to carbonylate primary amines by irradiation with UV light for 1 00

250 hours. The non-catalytic reaction uses the manganese complex simply as a source of

C0.72

Scheme 1.11 Proposed two step reaction mechanism

Stepl: H2N-R + CO —- HCONHR

+ Step 2 : HCONHR H2N-R b. ^ RNHCONHR + H2

Rhenium Carbonyl Complexes

The only report of a rhenium complex carbonylating amines is the one described

by Hieber and Schuster.96 They invoke the use of Re(CO)5Cl in the stoichiometric carbonylation of ammonia to formamide. Such carbonylations with third row transition metals are rare. 32

Scheme 1.12 Postulated mechanism for the carbonylation of primary amines in the

presence of Mn2(CO)io.

NH2R NH2R OC XO (CO)^Mn-Mn(CO:b Mn(CO)5' + .0 OC^ I ^C' -NH3R+ CO NHRI 1/2 H2 CO -NH2R NH3R+ HMn(CO]6 -NH2R

CO oc; ^^co

oc^l ~^C^

NH3R+ -NH2R CO I O NHR

II RHN— C—NHR NH2R -NH3R+

NH2R OC^ COco - Mn o Mn(C0)5‘ + R—f^G=o /I ^c"^ OC CO T NR-

Chromium. Molybdenum, and Tungsten Carbonyl Complexes

There are very few examples of Group 6 metals carbonylating amines. Doxsee and Grubbs have described the preparation of dimethylformamide from the stoichiometric reaction of Cr(C0)6 with LiNMc2.97 This example involves nucleophilic 33

attack at transition-metal-bound carbon monoxide followed by reduction by molecular

hydrogen under exeeptionally mild conditions (equation 1 .20).

O LiNMc2 OLi

Cr(CO)6 (CO)5Cr= ( 1 . 20) NMe-i

Angeliei has reported the reaction of [CpM(CO)4 ]PF6 (M = Mo, W) with excess

CH 3NH2 to give 1,3-dimethylurea.^^ As shown in Scheme 1.13, when the reaetion is run in the presenee of two equivalents of amine, the earbamoyl eomplex

CpM(CO) 3 (CONHCH 3 ) was obtained whieh suggested that the first step was a nucleophilic attack on a carbonyl ligand. Further reaetion of the carbamoyl complex with exeess amine gave the corresponding urea. When the reaction is run in excess trimethylamine, an organic isocyanate is obtained. The authors postulate a mechanism similar to the manganese mechanism depicted in Scheme 1.12. The first equivalent of amine deprotonates the earbamoyl ligand produeing an isoeyanate, which then reacts with another equivalent of primary amine to yield urea.

Finally, the McElwee-White laboratory has reported the earbonylation of seeondary amines whieh produce formamides in the presenee of the iodo-bridged dimer

[(CO)2W(NPh)l2]2 (32) (equation 1.21).^^ This stoichiometric reaction proceeds at room temperature for a variety of seeondary amines although reported yields are low. When 32 was reacted with primary amines under the same reaction conditions, primary amines 34

Scheme 1.13 Stoichiometric carbonylation of amines in the presence of a tungsten

carbamoyl complex

R-N-C=0

were selectively carbonylated to the corresponding 1,3-disubstituted ureas (equation

1.22). Furthermore, when 32 was reacted with a,oo-diamines under stoichiometric oxidative carbonylation conditions, cyclic ureas were formed. Further research resulted in the carbonylation reactions being catalytic in 32 when the reaction was run under

excess carbon monoxide. A possible mechanism for this transformation is shown in 35

Ph / R N OO/, v\CO \ JI H w> R OC CO N CH2CI2

( 1 . 22) [Ox] Ph 32

Scheme 1.14. IR spectra of the reaction mixtures were consistent with the presence of a carbamoyl intermediate. Following formation of 33 with four equivalents of amine, the carbamoyl complex 34 was proposed to form after a nucleophilic attack by the amine on a carbonyl ligand followed by deprotonation with an additional equivalent of the amine.

Conversion of 34 to the isocyanate 35 is postulated, and is thought to proceed by oxidation at the metal followed by deprotonation with excess amine. However, no free or coordinated isocyanates have been observed in the reaction mixtures. Nucleophilic attack

on the isocyanate complex or free isocyanate by another equivalent of amine is thought to 36

form the urea and the cationic complex 36. Complex 36 now has a free coordination site that in the presence of excess CO would render the reaction catalytic.

Scheme 1.14 Proposed mechanism for the carbonylation of amines in the presence of 32

r r”

OO/,, I 2 eq NH R ,.'K, ^ 2 nc, o(^ i^ir^co OC^ I^'^NH2R , ^ Ph 32

1 eq NH2R

r CO N OC,„ .j,*III ,sNH2R oa„. ,„sNH2R w’ I^^nh2R 0(T I NH 2R*

36 33

2 eq NH2R

[Ox] oa,.f ,„vNH2R leqNH2R T H 34 37

Conclusion

Although the use of organometallic complexes as organic synthons has been explored for many years, there are multitudes of avenues yet to be investigated. The remainder of this text will foeus on investigations of reaetive intermediates in the photolysis of metal carbyne complexes as well as examine further applications of metal carbonyl complexes in the catalytie carbonylation of amines. 38

CHAPTER 2 PHOTOPHYSICS OF LOW-VALENT CARBYNES

Introduction

There is increasing interest in discovering transition metal complexes that will not only luminesce, but also be useful in the preparation of chemical products. Such species are rare. Typically, photolysis of organometallic complexes with carbonyl ligands results in ligand loss as the primary photoprocess due to population of the metal-ligand

antibonding orbital via excitation of the d-d manifold. However, the "ligand loss" photoprocess is not the only photoprocess available to metal carbyne complexes. There are a number of examples where photolysis of metal carbynes has led to the conversion

of the carbyne moiety to other organic ligands.45,101-103 ^s mentioned in the previous chapter, metal carbyne complexes containing carbonyl ligands are known to undergo migration of the carbonyl to the carbyne ligand, resulting in the formation of r|^'ketenyl

complexes (equation 2.1).45 The migration is described as the result of increased electron density on the carbyne carbon due to a MLCT transition.

Prior to the research described herein, it was found that one-electron transfer from the MLCT excited state of a series of tungsten and molybdenum alkylcarbynes to halogenated solvents resulted in unprecedented ligand-centered radical processes.^ 1 >52, 1 04- 1 08 39

\ PMC3 hv

This is highly unusual. Not only is reactivity at an organic ligand in metal radicals far

less common than reactions at the metal center,^^ ^^t also, if this were 1 e' oxidation of the MLCT excited state, oxidation would take place at the ligand and the ensuing radical

would be metal centered. Namely, following oxidation, reactivity would be expected at the metal center.

Photophysics of Metal Carbvne Complexes

In order to understand the unusual reactivity of metal carbyne complexes, there have been a number of photophysical studies of complexes that contain the metal- arylcarbyne functionality, M=C-Ar, where M = W(IV), Mo(IV), and Os(IV). 1 09 family of complexes displays remarkably similar absorption and luminescence properties, despite having a rather wide variety of ancillary ligands in the coordination environment.

In general, the absorption of M=C-Ar complexes is characterized by a weak band in the

10^ 400-500 nm region ( e » M'^cm ') and a more intense band between 300 and 350 nm

(e « 10'^ M‘’cm'^). Theoretical and spectroscopic studies indicate that the low energy

band is associated with the HOMO LUMO transition, which is dominated by the d(M)

10-1 —> n (M=C-Ar) configuration.23,24,43,49,1 12 intensity of the HOMO 40

LUMO transition indicates that it is forbidden; the transition is likely parity forbidden but

spin-allowed (i.e. a singlet-singlet transition). The transition is parity forbidden owing to

significant degree of d-character in the HOMO and LUMO. The more intense near-UV

band is associated with a spin-allowed Tr(MsC-Ar) -> 7i*(MsC-Ar) transition, which bears some analogy to the n^n transition of the Ph-CsC-Ph chromophore.

There are also several reports of luminescence from complexes that contain the

10,1 1 MsC-Ar funetionality.4^’49,1 1 Pqj. example, complexes of the type

X(CO)2 (L)2W=C-Ph and [Ph-C=Os(NH3)s]^^ exhibit weak luminescence in the 600-650

nm region which has been attributed to the d(M) 7r*(M=C-Ar) excited state. The emission lifetimes of these complexes range from 50 to 250 ns, with quantum yields typically on the order of 5 x 10''*, The comparatively long excited state lifetimes coupled

with relatively low radiative rates (kr« 10^ s"') strongly imply that the luminescent state has triplet spin character. This assertion has been eonfirmed in several cases by the observation of triplet energy transfer from a carbyne excited state to organic triplet state

110,111 acceptors. jpg triplet character of the luminescent state in the carbyne complexes explains the large Stokes shift of the emission relative to the low intensity absorption in the mid-visible region, which is assigned to the spin-allowed (singlet- singlet) absorption.

This chapter provides further insight into the excited states of metal carbyne complexes of the type Cp(CO){P(OMe) 3 }M=C-R, where M = Mo or W and R = aryl

(complexes 37-40). These investigations are to provide further understanding of the ligand-centered reactivity at the carbyne ligand. Considering MLCT excitation does not

usually lead to reactivity at the ligand, the question to be answered is whether this 41

photoprocess is best described as MLCT or something else. Answers can be found by

examining the electronic structure, the excited states, the emissive states, and lifetimes of

these states.

\ _ OC Ar

P(OMe)3

37: M = W, Ar = phenyl 39: M = W, Ar = o-tolyl

= 38: M Mo, Ar - phenyl 40: M = W, Ar = 2-naphthyl

Synthesis of CptCOllPtOMel^lMsC-R Complexes

The carbyne complexes Cp(CO){P(OMe) 3 }MsC-Ph (W = 37, Mo = 38) and

Cp(CO){P(OMe)3}WsC(o-Tol) (39) were prepared as described previously.49,1 13-1 15

Preparation of the napthylcarbyne complex Cp(CO){P(OMe)3}W=C(2-Np) (40) involved an analogous route. The tris(phosphite) complex Cl(CO){P(OMe)3}3\V=C(2-Np) (41) was prepared by reaction of [(CO)5W(CO{2-Np})][NMe4] (42) with oxalyl chloride, followed by addition of excess trimethyl phosphite to give the impure tris(phosphite) complex 41. Reaction of 41 with fresh sodium cyclopentadienide (CpNa) led to displacement of the chloride anion and two P(OMe)3 ligands to yield 40. The crude 40 42

required more extensive purification than is necessary for 37-39, but was ultimately

isolated in 20% yield.

Electronic Structure of CDtCOMPtOMet^lMsC-Ph tW = 37. Mo = 38t^

As an aid to interpreting the spectra (vide infra) of 37-40, extended Huckel

calculations 1 16 were performed on 37 and 38 . jhg results of these one-electron

calculations are useful in visualizing the coefficient distributions in the frontier orbitals

and providing a rough sequence of orbital energies for 37-40.

Figure 2.1 shows a partial molecular orbital diagram for the formation of 37 from

the fragments Cp(CO){P(OMe)3 }W and CPh^, calculated in the same manner as 37. The

orbital orderings are similar to those previously reported for the model compound

Cp{P(OH)3}2Mo=CPh,49 although the energetic splittings are somewhat different.

These calculations show that the HOMO of the complex (Figure 2.2) is comprised largely

of the metal dx^-y^ orbital but is partially devoted to back-bonding into the n orbital of

the CO ligand. Below it are the two metal-carbon n bonds, their degeneracy broken by

conjugation of the lower one with the phenyl ring. The LUMO is the orbital identified as

the MsC 71 orbital that is conjugated into the phenyl 7r-system. However, inspection of

Figure 2.2 reveals that it too is primarily composed of a metal d orbital delocalized into

the 71* orbital of the CO ligand. The NLUMO is a nearly pure M=C tt* orbital. Extended

Huckel calculations were also performed on the molybdenum compounds 38, and the

TTie work in this chapter was done in conjunction with Thomas K. Schoch. Helpful discussions and analysis of spectral data were contributed by Dr. Kirk S. Schanze. 43

-55' n* (W-C)

-65.

n* (W-C-Ph)

£? C UJ d -4f^' T3 V V \ <• ^ x' '3 -75« d(W-CO)^|<'il U*c3 -4f dx2-y2

-85. 71 (W-C)

p -t- Ph 7t

IT 7t (W-C-Ph)

95.

*" (MeO) 3F*"’y (MeO)3P""^ oc OC

Figure 2.1. Orbital mixing diagram for Cp(C0){P(0Me)3} WsC-Ph (37). Orbital energies are derived from extended Hiickel calculations. 44

frontier orbitals obtained were very similar to those calculated for 37. Assignment of the

HOMO as a metal d orbital that is nonbonding with respect to the carbyne ligand is

consistent with structural information obtained by X-ray crystallography on

Br(dmpe) 1 2 W =C-Ph and its e’ oxidized congener [Br(dmpe)2W=C-Ph][Pp6].^

Oxidation results in only a slight shortening of the W-C(carbyne) and W-Br bonds (0.024

and 0.042 A) while the W-P bonds are lengthened somewhat. Upon this evidence, the

HOMO is assigned as a nonbonding orbital that is primarily dx^-y^ in character. Similar

observations have been made for Cl(dppe)2 MosC(>-Tol) and [Cl(dppe)2 Mo=C(p-

To1)][PF6].117

This picture of the electronic structure of carbyne complexes 37 and 38 suggests

that the lowest energy transition involves the frontier orbitals depicted in Figure 2.2. This

is the transition identified in prior literature as d(M) n*(MsC-Ar) MLCT in

10,1 1 nature.4^’4^>l 1 Although this transition does involve some charge transfer to the

carbyne ligand, the earlier designation of "MLCT" suggests a higher degree of charge transfer and higher band intensity than is actually observed {vide infra). Both the HOMO and the LUMO are primarily composed of the metal d orbitals and the carbonyl n orbitals. As a result, the lowest energy transition has a considerable degree of d-d

character that is consistent with its low extinction coefficient.

Electronic Absorption Spectra

Figure 2.3 illustrates the absorption spectrum of tungsten carbyne 37 in THF solution, which is typical for the series of metal carbynes examined herein, and Table 2.1 45

Figure 2.2. HOMO and LUMO of Cp(CO){P(OMe)3}W=C-Ph (37). Coefficients are derived from extended Huckel calculations. Hydrogens and phosphite methoxy groups are omitted for clarity. Key: (top) LUMO; (bottom) HOMO. 46

summarizes the data and assignments for the entire set of complexes. The absorption

spectra of the aryl-substituted carbynes 37-39 exhibit two primary UV/visible absorption

features: a strong band at approximately 330 nm and a weaker broad absorption at about

480 nm. The absorption spectrum of naphthylcarbyne 40 showed slight red shifts of both

bands, but otherwise was similar.

The absorption features in Table 2.1 are very similar to those described above for

the structurally related arylcarbyne complexes. These features include a weak absorption

with X,max located between 475 and 500 nm which is assigned to the parity forbidden,

spin-allowed (i.e. singlet-singlet) d(M) n (MsC-Ar) transition as well as a moderately

intense absorption in the near-UV between 325 and 350 nm which is assigned to the spin-

allowed 7r(M=C-Ar) ->7i*(MsC-Ar) transition.

250

200 CO

150

100 o 3

1 50

2.3. Figure Absorption spectrum of Cp(CO){P(OMe)3 }WsC-Ph in THF solution at room temperature. 47

The low-energy absorptions are clearly related to those assigned as d -> ti* MLCT

for the similar compounds X(CO)2 L2W=C-R (R = Ph, ‘Bu; X - Cl, Br; L2 = TMEDA, 2

Py» dppe).^ These transitions in 37-40 are best described as d to d transitions rather

than MLCT. The d-d character of the low-energy absorption in 37-39 (and by inference

in 40) is underscored by previous studies from the McElwee-White laboratory which

demonstrated that the absorption band is not solvatochromic.49 The more intense

* absorption at 330 nm is assigned to 71 ^ 71 transitions of the M=C-Ar chromophore.

While there is a significant energy gap between the visible and near-UV transitions, the

extended Hiickel calculations are in qualitative agreement with these assignments. The

slight red shifts for the naphthylcarbyne can be attributed to mixing of the larger n system

with the LUMO, lowering its energy.

Table 2.1. Absorption Spectra of Cp(CO){P(OMe) 3 }M=C-R (37-40) in THE Solution®

M R A,n,ax/nm(10^cm‘') emax/M''crn'’ assignment

37 W Ph 329 (30.4) 8000 t:( W=C-Ph)^7i*( W=C-Ph)

483 (20.7) 50 d(W)^7i*( WsC-Ph)

38 Mo Ph 328 (30.5) 4000 7t( MosC-Ph)->7t‘( Mo=C-Ph)

477 (20.9) 60 d(Mo)->7i*( MosC-Ph)

39 W o-Tol 331 (30.2) n( WsC-o-Tol)^7i*( Mo=C-o-Tol)

476 (21.0) d(W)^7i*( WsC-o-Tol)

40 W 2-Np 348 (28.7) 6000 7t( W=C-Np)->7i*( W=C-Np)

490 (20.4) 70 d(W)^7t*( W=C-Np)

®Data for 37-39 were previously reported.49 Data for 40 were obtained in this work. 48

Luminescence Studies

Detailed steady state and time resolved luminescence studies were carried out on

complexes 37-40. The fluorescence detection system was corrected for monochromator

and photomultiplier response to 800 nm, but owing to the sharp decrease in the efficiency

of the detector at wavelengths greater than 800 nm, luminescence data collected beyond

this wavelength were not useful. This point is significant, because the emission of the W

and Mo carbyne complexes extends well into the near-IR {vide infra).

The luminescence features of the tungsten and molybdenum carbynes studied are also similar to those of related arylcarbyne complexes. Each of the tungsten carbyne complexes luminesces weakly in solution at 298 K with emission lifetimes ranging from

60 to 170 ns. The molybdenum phenylcarbyne complex 38 is not luminescent at room temperature in fluid solution; however, red luminescence is observed when the compound is frozen in a glassy matrix at 77 K. Table 2.2 provides a summary of the emission parameters, including wavelength maxima (Xmax) for the corrected emission bands, and

emission quantum efficiencies and lifetimes (

- = by the expressions d>em/xem and kn (1/xem - K). Note that by applying these expressions it is assumed that the luminescent state is reached with unit quantum efficiency. 49

Table 2.2. Luminescence and Transient Absorption Data for Cp(CO){P(OMe) sjMsCR®

298 K'’ 77 K'

M R ^em Vs'' kjs'' Wns A|nax/nm TeJps

(10^ cm-') (10^) (10^ cm-'

37 W Ph 747(13.4) 141 6.9x10"' 4.9 7.8x10^' 129 741(13.5) 3.2

<10"* 38 Mo Ph e e <2 2.0x10’ 49 787(12.7) 8.3

39 W o-Tol 745 (13.4) 170 3.6x10"* 2.1 5.9x10^ 735(13.6) 3.6

40 W 2-Np >780 (<12.7) 66 1.5x10’ 60

- emission maximum; = emission lifetime; = “^max ®em= emission quantum yield; radiative decay rate; /fcnr nonradiative decay rate; tta = transient absorption decay lifetime. '’Argon-degassed THF solutions. ‘ Argon -degassed

2-MeTHF glasses. ‘'Amax and Tem values for la-c previously reported.' Too weak for accurate determination.^^

In fluid solution at 298 K the most pronounced emission is observed from

tungsten complexes 37 and 39. Emission from these complexes is very similar,

appearing as a broad, structureless band in the red with Xmax « 745 nm (Figure 2.4a).49

In each case the luminescence efficiency is rather low, with Oem« 5 x 10‘‘* (Table 2.2).

Emission excitation spectra for 37 and 39 are also illustrated in Figure 2.4a and reveal that the red luminescence is produced by excitation of the visible and near UV absorption bands. Emission lifetimes for 37 and 39 are comparable, and indicate that the luminescent excited state has a lifetime in the 100-200 ns range at room temperature.

The emission of 37 and 39 was also examined at 77 K in a 2-MTHF solvent glass, and the maxima and lifetimes for the low temperature luminescence are listed in Table

2.2. The emission of both complexes at low temperature appears as a broad band which is similar in energy and bandshape to that observed in fluid solution at 298 K. Thus, luminescence "rigidochromism", which is typically observed for charge transfer excited 50

states, was not apparent for 37 and 39. This effect underscores the small degree of

MLCT character in the luminescent excited state for the carbyne complexes. The 77 K

emission lifetimes for the two tungsten complexes are similar, and are comparable to the

18-120 lifetimes observed at 77 K for d-d excited states in W(CO) 5 L complexes.!

At 298 K in fluid solution the 2-naphthylcarbyne complex 40 exhibits an emission

which is substantially red-shifted relative to that of the phenyl and o-tolyl complexes.

Owing to the large red shift, the instrument used is only capable of detecting the blue

edge of the band (Figure 2.4b), and A.max for 40 is apparently >800 nm. An emission

excitation spectrum of 40 obtained with Xgm = 790 nm is also illustrated in Figure 2.4b.

There is good agreement between the absorption and excitation spectra for 40, which confirms that the near-IR emission of 40 indeed emanates from the complex. The « lifetime of the near-IR emission of 40 was determined to be 66 ns, which is slightly shorter than the lifetimes for the phenyl and o-tolyl complexes.

Molybdenum carbyne 38 exhibits no detectable luminescence at 298 K.

However, at 77 K in 2-MTHF the complex features a broad, structureless emission band that is similar in bandshape but red-shifted compared to the emission of the tungsten analog 37 (Table 2.2). The emission band decreased in intensity as the temperature of the

MTHF solution is increased above the glass-to-fluid transition (100-120 K) at which

point the luminescence becomes too weak to detect (Oem < lO"'*). The emission lifetime of 38 at 77 K (8.3 ps) is slightly longer than that of the tungsten analog.

By analogy to previously reported arylcarbyne complexes, the luminescence observed from 37-40 is attributed to the d(M) -> 7i*(MsC-Ar) triplet state. The triplet 51

Wavelength / nm

Figure 2.4 Emission and excitation spectra: (a) for 37 and 39 and (b) for 40. Emission spectrum is at right and excitation spectrum at left. 52

assignment has been substantiated in 37 by previously reported triplet energy transfer

studies.'^^ The significant Stokes shift of the emission tfom 37, 39, and 40 relative to the

low-intensity visible absorption band (»7.0 x 10^ cm"’) is attributed to the fact that the

absorption is due to the singlet-singlet HOMO-LUMO transition. The Stokes shift

observed for these carbyne complexes is similar to that reported for ^d-d luminescence in

tungsten and molybdenum carbonyl complexes of the type M(CO) 5 L.l Of further

interest is the fact that the luminescence data presented herein argues against the

suggestion made earlier that the large Stokes shift may be partly due to geometry changes

associated with excitation of the M=C-Ar functionality. 1 ^ ^ ^ significant

difference in the emission energy or bandshape for 37 and 39 in fluid solution compared

to in rigid glass at 77 K implies that there is no substantial inner- or outer-sphere

reorganization associated with the optical excitation.

Despite the overall similarities of the absorption and luminescence properties of

37-40, there are important distinctions that provide insight concerning the nature of the

electronic states involved in the photophysics. First, the visible and near-UV absorption bands and the luminescence of the 2-Np complex 40 are noticeably red-shifted compared to those for the phenyl and o-tolyl analogs. The red-shift of both absorption bands and the luminescence in the 2-Np complex is expected, since the naphthalene ring provides extended conjugation for the n*(MsC-Ar) orbital, which is the LUMO of the complex and is the "acceptor" orbital for the visible and near-UV n, n transitions.

Second, the tungsten and molybdenum phenylcarbyne analogs 37 and 38 display remarkably similar ground- and excited-state absorption spectra. This likeness points to a similarity in the electronic structure of the two complexes. The one notable difference in 53

the photophysical properties of the two phenylcarbynes is the lack of observable

luminescence from 38 in fluid solution at room temperature. Since the excited state

lifetimes of 37 and 38 are not significantly different (i.e. from transient absorption 130

and 50 ns for 37 and 38, respectively vide infra), the lower emission yield of 38 might be

attributed to lower spin-orbit coupling (and therefore less singlet-triplet mixing) in the

molybdenum complex as compared to the tungsten analog.

Transient Absorption Spectroscopy

Nanosecond laser flash photolysis studies were carried out on carbyne complexes

and 40 in 37, 38, argon-degassed THF solutions. All experiments were performed using

the third harmonic output of a Nd:YAG laser for excitation (355 nm, 10 ns fwhm, 10

mJ/pulse). As described below, strongly absorbing transients were observed for each

complex, and the decay lifetimes for the transient absorptions were determined by global

kinetic analysis of the transient absorption decay at > 25 wavelengths. ^21

Figure 2.5a illustrates transient absorption difference spectra of 37 at delay times

ranging from 0 to 280 ns following laser excitation. The spectra are characterized by a

broad absorption band with A.max » 440 nm. There appears to be a hint of structure on the absorption band, with distinct (and reproducible) shoulders at 395, 420, and 460 nm.

Global analysis of the transient absorption data indicates that the transient decays with a lifetime of 129 ns, in good agreement with the previously reported emission lifetime

=141 ns). ('^em The correspondence of the transient absorption and emission lifetimes strongly implies that the transient absorption is due to the luminescent excited state. 54

Under the assumption that the lowest excited state 37 is populated with unit efficiency

following 355 nm excitation, it is possible to determine the difference molar absorptivity

(Ae) of this state by using the relative actinometry method. ^ 22 experiment was accomplished by using [/ac-2,2-bipyridine)Re(CO)3(4-benzylpyridine)]^ as an

actinometer (As = 1 1,200 M 'cm'' at 370 nm)123 and led to a value of As = 3000 M''cm'

' at 440 nm for 37.

Figure 2.5b illustrates transient absorption difference spectra of 40 at delay times

ranging from 0 to 120 ns following laser excitation. The spectrum of this complex is characterized by a moderately intense and broad absorption band with Xmax « 445 nm, with distinct (and reproducible) shoulders at 395 and 420 nm. Global analysis of the transient absorption data indicates that the transient decays with a lifetime of 49 ns. The close correspondence between the transient absorption spectra of 37 and 38 points to a similar electronic structure for the absorbing excited states in both complexes.

Figure 2.5c illustrates the transient absorption difference spectra of 38 at delay times ranging from 0 to 160 ns following laser excitation. The spectra are characterized by an intense but featureless transient absorption band with Xmax « 460 nm. Global analysis of the transient absorption data indicates that the transient decays with a lifetime of 60 ns in very good agreement with the emission lifetime of the complex (66 ns). The agreement between the transient absorption and emission decay lifetimes suggests that

for 40 the transient absorption arises from the luminescent excited state.

A distinguishing feature of this work is the observation of strong transient absorption by the d(M) —> n (M^C-Ar) excited states of the tungsten and molybdenum arylcarbynes. This is the first study of transient absorption spectra of photoexcited 55

Fischer carbyne complexes. There are several important ramifications of the observation of transient absorption signals for the excited states of this family of complexes. First, as noted above, it is clear that the excited state carbynes have comparatively large difference

molar absorptivities. This is in contrast with the low molar absorptivity that is typical of d-d excited states of organometallic complexes. ^ 24, 125 moderate to large difference

molar absorptivity of the excited states of the arylcarbyne complexes is more in accord with that of d7i(M) —> n (ligand) MLCT excited states in d^ metal diimine complexes

^’*^ such as Ru(bpy) and [(bpy)-Re'(CO) = 3 3 (py)] (bpy = 2,2'-bipyridine and py pyridine). 123, 126

An important question concerns the origin of the moderate to strong absorbance of the d ^ 71 excited states in the arylcarbyne complexes. The absorption of MLCT excited states in d^ metal diimine complexes has been assigned to spin-allowed n n

and 71 —> 7T transitions of the diimine radical anion chromophore that is produced by

127 MLCT excitation. gy analogy, the absorption of the d ^ n* excited state in the arylcarbyne complexes may be associated with an allowed n* n* transition of the

MsC-Ar chromophore. In this connection, it is important to note that the excited state absorption of phenylcarbyne complexes 37 and 38 is quite similar, while that of naphthylcarbyne 40 is slightly red shifted (and qualitatively appears to be more intense).

The red shift in the naphthyl system suggests the involvement of the ti to ti* levels of the aryl unit in the optical transition detected in the transient absorption experiment.

Finally, the substantial difference absorptivity of the excited state in the arylcarbyne complexes provides an exceedingly useful means to track the excited state dynamics (and reactivity). This is significant, because even though emission studies have 56

Figure 2.5 Transient absorption difference spectra following laser excitations (355 nm, 10 ns fwhm, 10 mJ/ pulse) of (a) 37 (b) 38 and (c) 40. 57

provided an opportunity to study the excited states of certain metal carbynes, the method

has limitations. For example, among 37-40 and related Cp(CO)2M=C-R and

Cp{P(OMe) 3 } 2MsC-R species, only the tungsten aryl carbynes with one phosphite ligand

and one CO ligand display easily observable luminescence. Even for these complexes,

the luminescence efficiency is low, as evidenced by the quantum yield for emission for

37 and 39. The bis(phosphite) arylcarbynes are not emissive in solution at room

temperature,49 and molybdenum complex 38 is also nonluminescent under the same

experimental conditions. Comparing the lifetime measurements obtained by absorbance

with those observed by the emission method provides assurance that the observed

transients are in fact the emissive states. Furthermore, by applying global analysis to the

transient absorption data, it is in principle possible to obtain very reliable excited state

decay rate data.^^l

It should be noted that after the work described in this chapter had been

completed, an additional study was undertaken by co-workers that investigated the role of photochemical electron transfer in the reactions of tungsten carbynes, namely, electron transfer from the excited state of Cp{P(OPh) 3 }(CO)W=CPh (43) to a series of electron acceptors. 128 examined the quenching of the d ^ ti* excited state of 43 with a series of pyridinium and nitroaromatic acceptors of varying reduction potential. Laser flash photolysis revealed that quenching is accompanied by the appearance of radical ion products, thereby providing evidence that quenching occurs via electron transfer. The dependence of the bimolecular quenching rate constant on the reduction potentials of the acceptors establishes that the d ^ ti* excited state of 43 is a potent reducing agent. 58

Conclusion

Low-valent carbyne complexes are well-established compounds in organometallic

chemistry, yet relatively few photophysical studies on these complexes have appeared in

the literature. = In this work, the complexes Cp(CO){P(OMe) 3 }MsC-R (M = Mo, W; R

Ph, o-Tol, 2-Np) (37-40) were examined in an effort to learn more about the excited

states of these molecules, as well as to provide some information relevant to the study of

photochemical reactions. Extended HUckel calculations and absorbance spectra indicate

that 37-40 possess low-lying exeited states which can be populated directly via a parity-

forbidden, singlet-singlet optical transition in the mid-visible region. This transition is

better described as a d-d transition rather than MLCT. Emission is observed from the triplet manifold of these excited states, and at room temperature in fluid solution the

lifetimes of the triplet state range from 50 ns for the molybdenum phenylearbyne 38 to

170 ns for the tungsten tolylcarbyne 39. Transient absorbance measurements performed on 37-40 in THE indicate that the triplet excited states can also be detected and monitored by absorption, since the lifetimes obtained in the absorbance experiments match those obtained through emission methods. The observation of strong excited state absorption provides an avenue for future work concerning the photophysics and photochemistry of arylcarbyne complexes. Such work may include the study of exeited state resonance

Raman to provide further information concerning the nature of the excited states. 59

CHAPTER 3 SYNTHESIS AND PHOTOOXIDATION OF A TUNGSTEN CYCLOHEXENYL CARBYNE COMPLEX

Introduction

As discussed in Chapter I, a variety of organic compounds including olefins, dienes, dienals and other cycloalkenones have been synthesized from the photooxidation

of various alkyl metal carbynes of the type Cp(CO){P(OMe)3 }M=CR [R= alkyl; M= Mo,

W].106 Mechanistic studies suggested that these reactions shared the following two

initial steps (Scheme 3.1): 1) photoinduced electron transfer from the carbyne to the halogenated solvent to form a 17 electron radical cation A and 2) hydrogen abstraction from the reaction medium to produce a 16 electron cationic carbene B.49, 5 1,1 07, 108, 129

Despite the usual tendency of organometallic radicals to undergo ligand exchange and halogen abstraction at the metal center as discussed previously,4^’47,130 geometric distortions of the carbyne radical cation cause radical reactivity to occur at the carbyne carbon and the H-abstraction takes place at that site.^^ The resulting cationic carbene complex B is the reactive species in the reactions that produce organic products.

Although mechanistic studies were consistent with the intermediacy of cationic carbene complexes in the formation of organic products from photooxidized carbynes, to this point, carbene complexes had never actually been observed in the reaction mixtures. 60

This work describes the first direct observation of a hydrogen abstraction product upon photooxidation of a tungsten cyclohexenyl carbyne complex.

Scheme 3 . 1 . Proposed mechanism for generation of metal carbenes upon photooxidation of metal carbynes

+•

^ hv nn W=OCHRiR2 \ / CHCI3 ocv"'^w=c-chr,R2 P(OMe)3 P(OMe)3

+ RH -R*

H \ / CHR1R2 P(OMe)3

B 61

Synthesis of fTi5-Cs H.¥COHPrOMel2lW^Crr-aHg) (441

The cyclohexenyl tungsten carbyne complex 44 was synthesized from the

tungsten carbene 45 as shown in Scheme 3.2. There is a literature preparation for the

tungsten carbene 45 which reports a yield of 52%. 131 However, for the current study,

carbene 45 was synthesized by modifications of a route that was used in the synthesis of

the molybdenum analogue of 40.132-134 xhese modifications resulted in a higher yield

than that reported in the literature. The modified route involves the lithiation of the

hydrazone 46 with r-butyl lithium to form the cyclohexenyl lithium reagent 47. The

immediate reaction of 47 with tungsten hexacarbonyl formed the cyclohexenyl tungsten

acyl anion 48. Anion 48 was methylated with trimethyl tetrafluoroborate which gave

carbene 45. Once isolated, the crude product 45 was not allowed to stand over-night but was chromatographed immediately on a -78°C chilled alumina column giving a purified yield of 75%. Carbene 45 was readily converted to the tris(trimethyl phosphite) carbyne complex 49 using Fischer’s method9,21 which involves step-wise addition of BBrs and

P(OMe)3, giving 49 in 65% yield. Reaction of CpNa with 49 and subsequent purification through a -78°C chilled alumina column produced carbyne 44 in 41% yield as a reddish- orange oil that can be stored for approximately one week at -30 °C. \

Scheme 3.2. Synthesis of (ii^-C 5 H 5 )(CO){P(OMe) 3 }W=C(c-C 6H9) (44)

1) 0°C W(CO)6

48 BBr3 CH2CI2

2) P(OMe)3

P(OMe)3 / CpNa \ I >CO — _ OC' (MeO)3P THF p(OMe)3 P(OMe)3

49 44 63

Photooxidation of (Ti^-Cs Hs¥CO')IPrOMe^2 ^^^^^^~^'^HQ) (44)

Photolysis of carbyne 44 at -45 °C in CDCI 3 while monitoring the reaction

progress by 'H NMR spectroscopy resulted in observation of downfield shifts of the alkyl

peaks and the formation of a new Cp peak. Since the reactivity of related carbyne

1 44'^’ complexes following photooxidation suggested that had probably abstracted a

deuterium atom tfom the CDCI 3 solvent, the experiment was repeated in CHCI 3 at -45°C

so that the abstracted H-atom could be located by *H NMR. Upon completion of the

photolysis as determined by disappearance of the initial carbyne 44 by TLC of the

reaction mixtures, the CHCI 3 was pumped off at -45°C, and the residue was dissolved in

cold CDCI 3 in order to obtain a *H NMR spectrum. The spectrum revealed a broad singlet at 515.35 which is characteristic of a proton on the carbene carbon of an electrophilic carbene complex. 1^5

One obvious candidate for the photooxidation product was the cationic carbene

complex 50. Electron transfer from 44 to the solvent followed by H-abstraction at the carbyne carbon of 44 would produce 50 (equation 3.1). However, IR spectra of the reaction mixtures failed to show a new signal for a terminal CO ligand. 136,137 Qnce 50 had been ruled out as the observable species, the next obvious possibility would be displacement of the CO in 44 with Cf to yield the neutral carbene 51 (equation 3.2).

Reaction mixtures from the photooxidation of 44 would contain Cf since it is generated upon reduction of the CHCI3 solvent by the excited state of carbyne 44.138 64

Complex 51 is formally the product of protonation of carbyne 44 at the carbyne

carbon and Cl substitution. Thus, it was not unreasonable to attempt its preparation by

reaction of 44 with HCl, although protonation of metal carbynes is a complicated issue

because proton attack can occur either at the metal or at the carbyne carbon.3,29,34,139

Reaction of related (r|5-C5H5)(CO){P(OMe)3}M=CR complexes with HCl generally leads

to Ti -acyl complexes of the type (ri^-C5H5)Cl2{P(OMe)3}MCOCH2R (for example, see

3.3).^2,33,49,5 1,106,107,129,140 equation However, Kreissl’s work on the mechanism of r] -acyl formation points to the intermediacy of a metal carbene intermediate which

then undergoes CO insertion (Scheme 3.3). This postulate was supported by the fact that when the aminocarbyne (Ti5-C5H5)(CO)2W^CNEt2 was reacted with HCl, the reaction stopped at the carbene. 140 Although in the Kreissl work, the carbene was only 65

HCl*Et20 ^/ff Mo=C-(CH2)3CH3 Mo~Cn riu p. cr CH2(CH2)3CH3 (MeO)3P / EtoO /i (3.3) OC Cl P(OMe)3

Scheme 3.3. Proposed mechanism for the protonation of carbynes.

M== Mo, W R= alkyl, aryl X- cr, CF3COO', CC13COO'

observable for the aminocarbyne case, which is electronically very different from 44, the presence of the cyclohexenyl group in 44 opened the possibility of preventing CO insertion into the carbene ligand of intermediate 50 by chelation of the double bond.

As desired, protonation of the vinyl carbyne 44 with HCl (equation 3.4) did not produce the n^-acyl complex, but rather yielded the carbene complex 51 as a yellow 66

powder which is thermally labile but stable at -45 °C. Comparison of the 'H NMR

spectrum of the HCl product to the 'H NMR spectra of the reaction mixtures from

photooxidation of carbyne 44 provided confirmation that both routes had produced the

same complex.

The 'H NMR, '^C NMR, and MS data for 51 support the proposed structure. In

-50° the *H NMR of 51 in CDCI 3 at C, the proton on the carbene carbon appears as a singlet at 5 15.35, which is characteristic of conjugated electrophilic carbene

complexes.135 -phe ‘^C NMR chemical shifts of 5 277.8 for the carbene carbon, 6 75.4 for the quaternary vinyl carbon, and 8 93.7 for the tertiary vinyl carbon strongly resemble the literature values for the tungsten ri^-vinyl carbene complex (ri

C 5 H5 )(CO)(I)W=C(Ph)C(Ph)CHTol (52) (Figure 3.1).42 xhe electrospray MS of 51 in

CH 3 CN exhibited a molecular ion peak at 502.1, with an isotopic distribution pattern that matched the simulation for the molecular formula of 51, C 15 H24CIO3 PW (Figure 3.2). 67

5 94.4 8 94 7

52 51

Figure 3.1. Comparison of '^C NMR data between 51 and known 52.

Photolysis of carbyne 44 at -45 °C in CDCI 3 , containing decane as a standard, afforded carbene 51 in 25% yield as determined by integration of the Cp resonance in the

'H NMR spectrum. In order to rule out photochemical formation of HCl as the source of

51 in the photooxidation reactions, 44 was photolyzed in CDCI 3 to which the hindered base 2,6-di-tert-butylpyridine had been added in order to scavenge any acid. The addition of 2,6-di-/ert-butylpyridine is a technique that is used to distinguish radical

cation chemistry from acid-induced reactions.5U52,107,142,143 jhg yjgijj of the photolysis product was the same in the presence and the absence of the base, ruling out formation of 51 in the photolysis mixtures by reaction of 44 with HCl. 68

SIMULATION

ACTUAL

Figure 3.2 Electrospray mass spectral data of 51 69

Oxidation of (nS-CsHOrCOMPrOMel.^W^Cfr^-C.Ho^ (44) in the Presence of Unsaturated Substrates

Although high-valent carbyne complexes have found wide use in the areas of

olefin metathesis and alkyne polymerization, little is known about the reactivity of low

144- valent metal carbyne complexes with unsaturated substrates. 16, 146 Efforts to

'^’ provide further enlightenment on this reactivity involved studying the reactivity of 44

in the presence of unsaturated species. As discussed earlier, 44^’ is formed by photochemical electron transfer from 44 to . CHCI 3 However, upon reduction, CHCI 3

Iragments to Cl and •CHCb.l^^ The •CHCI2 is known to undergo unwanted Arbuzov

1^7 reactions with the P(OMe)3 ligand qj^ similar carbyne complexes which is evident by the formation of CH 3 CI in the reaction. 61 The use of outer sphere chemical oxidants at low temperatures has been studied and provides an alternative method in which low

yields and side reactions Irom •CHCI2 can be avoided.

A cyclic voltammogram of carbyne 44 taken in 0.1 M BU4NSO3 CF 3 /C2H4CI 2 under nitrogen showed the first oxidation wave at +0.55V vs. NHE to be completely irreversible regardless of scan speed or switching potential. The first oxidation peak corresponds to the formation of 44^*. The complete irreversibility of this peak indicates that 44 is a very reactive species and undergoes immediate chemical reaction on the cyclic voltammogram time scale. Based on the oxidation potential of 44, the chemical oxidant, acetyl ferrocenium, with an oxidation potential of +0.82Vl48 was chosen for the oxidation investigations. 70

> 120. 00 2. 200 1 . 000 E v». NHE (V)

Figure 3.3 Cyclic voltammogram of 44. Sweep rate: 100 mV/s

Initially, carbyne 44 was reacted with acetylferrocenium tetrafluoroborate

(AeFc BF4 ) at -90 °C and NMR spectra were taken in 10 °C intervals upon wanning to room temperature. The reaction mixture upon warming ehanged eolor from blue to green to brown. Formation of a hydrogen abstraction product was never observed, unlike the formation of as diseussed earlier in 51, the photolysis of 44 in CHCI3 . Considering the fact that hydrogen abstraction did not readily take place in the ehemieal oxidation reaetion, it seemed feasible to attempt to trap the ensuing 17-eleetron radical cation with an unsaturated substrate. Torraca et al. have reported the reaction of a butyl molybdenum carbyne with AcFe^BF4 in the presenee of phenyl acetylene resulted in C-C coupling of the acetylene to the carbyne earbon.149 Unfortunately, when carbyne 44 was reacted 71

with AcFc BF4 in the presence of various alkynes, olefins, ketenes, and imines, all

reactions resulted in intractable material.

Other Reactions with Unsaturated Substrates

The carbene 51 that is formed from the protonation of 44 with HCl is similar in

structure to a diene. Efforts to form [4 + 2] or [2 + 2] adducts were attempted in the

presence of various olefins, acetylenes, aldehydes, and ketenes. No reactivity was ever

observed other than the decomposition of 51 to intractable materials.

It is possible the chelation of the cyclohexenyl group to the metal was inhibiting

further reactivity of 51. Attempts to circumvent this problem included protonation of 44 with HBF4 with and without CO pressure in attempts to trap the putative 16-electron cationic carbene 53 (Figure 3.4). Again, all in situ reactions with unsaturated substrates resulted in intractable materials.

Figure 3.4 Investigated reactivity of 53 with unsaturated substrates.

Conclusion

The photooxidation of the cyclohexenyl carbyne 44 in CHCI3 yields carbene complex 51. This experiment represents the first direct observation of a carbene 72

intermediate following single electron oxidation of a carbyne complex and provides

further evidence for H-abstraction at the carbyne carbon as a common mechanistic step in

previously reported photooxidation reactions. Formation of 51 upon thermal reaction of

44 with HCl also provides additional support for the Kreissl mechanism for generation of

T) -acyl complexes from carbyne complexes and HCl by extending the range of observable carbene intermediates from amino derivatives to alkyl derivatives. 73

CHAPTER 4 CATALYTIC OXIDATIVE CARBONYLATION OF AMINES WITH W(CO)6

Introduction

Substituted ureas have been of reeent interest due to appearanee of this functionality in drug candidates such as HIV protease inhibitors, 150,151 FKBP12

152 153, inhibitors, CCK-B reeeptor antagonists 154 endothelin antagonists. 155 in addition, ureas have found widespread use as agrieultural ehemieals, resin precursors, dyes, and additives to petroleum eompounds and polymers. 156 Among the numerous methods for synthesis of N,N'-disubstituted ureas are the reaetions of primary amines with isocyanates, phosgene, or 157,158 phosgene derivatives (equation 4 . 1 ). Yields are usually good to excellent for these reactions, but other aspects can be problematic.

Phosgene itself is highly toxie and corrosive while phosgene derivatives can be expensive to use on a large seale. Additional impetus for replacement of phosgene derivatives comes from the standpoint of atom economyl59 by which using CO to install a carbonyl moiety into a urea would be preferable to a method using a typieal phosgene derivative such as 1,1-carbonyldiimidazole. The direet metal-catalyzed eonversion of amines and

CO to ureas provides an alternative to phosgene and its derivatives. 74

+ 2 HX (4.1)

H H

X = Cl, CI3CO, -^N

Although catalytic carbonylation has been investigated over many years,55iMil60 topic remains of interest. Oxidative conversion of primary amines into ureas has been reported for transition metal catalysts involving Ni,’71

Mn,73,94,95 Ru161 a^d, most commonly, Pd.^S, 162,163 jy[ain group elements such as

1^6- sulfiirl6^’165 and selenium 168 gan also serve as catalysts. However, neither the transition metal- nor the main group-catalyzed reactions are ideal. The transition metal- catalyzed reactions generally require high temperatures and pressures. In addition, yields for aliphatic amines are generally much lower than those for aromatic cases. Among the main group catalysts, selenium-catalyzed carbonylation reactions can produce high yields of ureas under mild eonditions. However, generation of hydrogen selenide as a byproduct and the need for stoiehiometric or exeess selenium for certain substrates l^^’l^^ are problematic.

Prior to this work, preliminary studies were reported on the catalytic oxidative carbonylation of primary amines to ureas using either [(CO)2W(NPh)l2]2 or W(CO)6 as the precatalyst and I2 as the oxidant.99,169 jt was found that the W(CO)6/l2 catalyst system also works for the conversion of secondary amines to tetrasubstituted ureas and the preparation of eyclic ureas directly from primary and secondary a,co- 75

diamines. Extensive optimization studies were performed on the oxidative

carbonylation of aliphatic primary amines to N,N’-disubstituted ureas using the

W(CO)6/l2 oxidative carbonylation system on the substrate «-propyl amine (equation

4.2). Variables such as catalyst, temperature, solvent, CO pressure and quantity of added

O W(CO)6 RNH2 CO /I2/ K2CO3 H H

base were examined. The optimal conditions for the carbonylation of w-propylamine

were determined to be 2 mole % W(CO)6, stoichiometric amounts of amine and I2, and

1.5 equiv of K2CO3 per equiv amine reacted in CH2CI2 at 90 °C under 80 atm CO

pressure. This chapter is a continuation of the work described above. It will include

further investigations of optimized procedures, applications of the optimized conditions

to other primary amines, as well as a functional group compatibility study of the

W(CO)6/l2 oxidative carbonylation system.

Carbonylation of Primary Amines

Although CH2CI2 was the best single solvent for the « -propylamine case,

solubility problems were encountered when CH2CI2 was used as solvent for various

substituted benzylamines {vide infra), for which either the amine starting material or the hydroiodide salt was insoluble. Note that the stoichiometry of oxidative carbonylation of 76

amines to ureas dictates that two equiv of the amine hydroiodide (RNH3I) will be

produced per equiv of urea (equation 4 . 3 ). In the absence of added base, these amine

salts are observed in the reaction mixtures. The presence of K2CO3 as a sacrificial base

increases the product yield by regenerating the free amine from the salt and returning it to

catalyst R. R

+ 4 . 3 4 RNf^ N N 2 [RNH3]Y ( ) I2/CO H H

the substrate pool. However, insolubility of the amine salts will hinder this deprotonation

since the K2CO3 and the salt will coexist as immiscible solids. To circumvent the

inability of solid K2CO3 to deprotonate insoluble amine salts, the two-phase solvent

system CH2CI2/H2O was investigated. Due to the solubility of the hydroiodide salts in basic water, they could be deprotonated and shuttled back into the methylene chloride layer as the free amine. An added advantage was that the base is soluble in the water layer and stirring problems associated with solid K2CO3 were eliminated.

For both w-propylamine and many of the substituted benzylamines used in the functional group compatibility studies {vide infra), room temperature yields increased when the solubility problems were resolved with the two-phase CH2CI2/H2O solvent system. Although there are several examples of two-phase carbonylation reactions in the

literature, a biphasic system for the carbonylation of amines is unprecedented. In fact, carbonylation of amines in aqueous medium has only been reported for conversion of amines to formamides with a water-soluble ruthenium carbonyl catalyst. 1^2 77

Although the carbonylation of «-propylamine at room temperature in the two-

phase CH2 CI2/H2 O solvent system resulted in 73% yield which was higher than the yield

for the single-solvent CH2CI2 system, repeating the reaction at 90 °C resulted in a 85%

yield which was comparable to the 90% yield achieved in the previously described

reaction run at 90 °C in CH2CI2 .

Therefore, a series of primary amines was reacted under the following conditions;

2 mole % W(CO)6 , stoichiometric amounts of amine I 1.5 and 2 , and equiv of K2CO 3 per

equiv amine reacted in CH2CI2 at 90 °C under 80 atm CO pressure. These reactions gave the corresponding N,N’-disubstituted ureas, as seen in Table 4.1. The ureas were formed in moderate to high yields, ranging from 90% for N,N’-di-«-propylurea to 53% for N,N’- di-/-propylurea. The yields are generally lower if the amine bears secondary or tertiary alkyl substituents. For comparison, the amines in Table 4.1 were also run in the

CH2CI2/H2O solvent system. Analogous to the decrease in yield of the «-propyl urea in the two-phase system at 90 °C, the other primary amines also suffered a decrease in yield

when carbonylated in the mixed solvent system. It is noteworthy that aniline is unreactive under the reaction conditions, considering aromatic amines are generally better substrates than their aliphatic analogues for the formation of ureas under transition metal-catalyzed carbonylation conditions.^^2,173 78

Table 4.1 Catalytic carbonylation of primary amines in single/ biphasic solvent systems

%Yield“'*’ %YieiF^ Amine Product

CH2CI2 CH2CI2/H2O

90 85

53 27

® Isolated yield of urea calculated per equivalent of amine

’’ Reaction conditions: amine (14.2 mmol), W(CO)6 (0.28 mmol), I 2 (7.1 mmol), K2CO3

(21.3 mmol), CH2CI2 (40 mL), 90 “C, 80 atm CO, 24 h.

The solvent was CH2 CI2 (35 mL) plus H2O (5 mL). Other conditions are as in footnote

b. 79

Functional Group Compatibility Study

In order to explore applications of this reaction in organic synthesis, a study of

functional group tolerance was undertaken. Various substituted benzylamines (equation

4.4) were reacted with CO under similar carbonylation conditions, as seen in Table 4.2.

Substituted benzylamines were chosen as model substrates for the compatibility study

because the benzene ring provides a rigid backbone which separates the functional group

W(CO)6

CO / I2 / K2CO3

from the reacting amine. The possibility of intramolecular interaction is eliminated in these substrates. This first screen of functional groups addresses interference by simple binding of the catalyst to the functional group or competitive reaction of the functional group by oxidation or carbonylation. More complex substrates, such as conformationally flexible amino alcohols in which carbonylation could lead to different types of products, are the subject of further investigation and will not be discussed here.

Initial studies with the parent benzylamine 54 established that N,N’-dibenzylurea

55 was formed in 63% yield upon reaction under the optimized reaction conditions for n-

propylamine (Table 4.2, CH2CI2 ). As discussed above, the low yields of ureas for many of the substituted benzylamines were attributed to the insolubility of their respective

hydroiodide salts in CH2 CI2 . As shown in Table 4.2, it was found that yields for many of

the substituted ureas were significantly higher when the 7:1 CH2CI2/H2 O solvent system

was used (Procedure C, Chapter 6). 80

The effects of aryl halides were examined first. The p-chloro-, p-bromo-, and m-

iodobenzylamines 56, 58, and 60 produced their respective N,N’-disubstituted ureas 57,

59, and 61 in surprisingly low 35%, 30%, and 39% yields when the reaction was run in

CH 2 CI2 . Although side reactions resulting from oxidative addition of the aryl halide bonds to the tungsten complex were considered, the main products of the reactions turned out to be the amine hydroiodides. Upon addition of water to the reaction mixtures, the yields of the ureas increased to 77%, 77%, and 70% respectively.

In experiments with ether and thioether compounds in CH 2 CI2 , the p- methoxybenzylamine 62 was carbonylated to urea 63 in 47% yield, while p- methylthiobenzylamine 64 was carbonylated to urea 65 in 24% yield. Under these conditions, the major products were the hydroiodide salts. The solubility problem was

again circumvented by addition of H2 O to the solvent system resulting in the ureas forming in 70% and 81% yields, respectively.

The hydroxymethyl compound 6 produced urea 6 67 with yields of 5% in CH 2CI2

in and 58% CH 2 CI2/H2 O. Remarkably, the carbonylation reaction was completely selective for the amine over the alcohol, within the limits of detection. None of the corresponding carbamate or carbonate could be observed in the reaction mixtures. This experiment suggests that the carbonylation of amines could be carried out in the presence of unprotected hydroxyl groups. In contrast, thiol 68 failed to form the corresponding urea 6 9, undoubtedly due to the sensitivity of thiols to oxidation. 81

Table 4.2 Catalytic carbonylation of substituted benzyl amines

roYield"’*’ %YiiiP Amine Product CH2CI2 CH2CI2/H2O 82

“/oYield^-" %Yield“’‘^ Amine Product CH2CI2 CH2CI2/H2O

O 72 O O 73 83

"/oYield^’” %Yield^" Amine Product CH2CI2 CH2CI2/H2O

28 14

NH2 8j

17 20

Reaction conditions: amine (7.1 mmol), W(CO)6 (0.14 mmol), I2 (3.5 mmol), K2CO 3

(10.7 mmol), CH2CI2 (20 mL), 70 °C, 80 atm CO, 24 h.

The solvent was CH2 CI2 (21 mL) plus H2O (3 mL). Other conditions are as in footnote b. 84

The ester-substituted benzylamine 70 was also successfully carbonylated to the

urea 71 in 36% yield in CH2CI2 with 55% yield obtained in CH2CI2/H2 O. The carboxylic

is acid analogue 72 insoluble in CH2 CI2 , which hindered efforts to explore the tolerance of the catalyst to the presence of carboxylic acids. However, the solubility of 72 in basic

H 2O made it possible to obtain the corresponding urea 73 in 37% yield from the

CH2CI2/H2 O system.

Compatibility of the catalyst system with terminal alkenes was then examined.

The olefin 74 was reacted under the carbonylation conditions to form the urea 75 in a

41% yield in CH2 CI2 . Although we anticipated possible problems with reactivity at the double bond, there were no identifiable products in which the double bond had been perturbed. The low yields could, however, be indicative of unidentified side reactions.

Unlike the previously described reactions of substituted benzylamines, higher yields were

obtained when the reaction using CH2 CI2 as the solvent was diluted to 0.03M in substrate and conducted at room temperature.

In addition, /?-nitrobenzylamine (76) was successfully carbonylated to urea 77 in a

45% yield in CH2 CI2 with 76% yield obtained in CH2CI2/H2 O. The cyano-substituted benzylamine 78 was also carbonylated to its respective urea (79) in 37% yield in CH2 CI2

and 68% yield in CH2 CI2/H2 O. Note that the possibility of the nitrile moiety serving as a ligand for the catalyst did not interfere with the reaction.

Further functional group studies included the reaction of the /7-aminobenzylamine

80 under the carbonylation conditions. If both the aliphatic and aromatic amines were to participate in the carbonylation reaction, oligomeric products would be formed. Primary aromatic amines, such as aniline were unreactive under the carbonylation conditions, so 85

formation of the simple urea 81 was expeeted to dominate. However the reaction mixture was quite complex and urea 81 was only isolated in 28% yield. In this case, addition of water to the solvent system decreased the yield to 14%.

The apparent participation of the aromatic amine in the reactions of 80 brought up the question of intramolecular carbonylation of the o-amino benzylamine 82 to form

3,4-dihydro-2(7//)-quinazolinone (83 ) (equation 4.5). Although aromatic amines are apparently not nucleophilic enough to initiate the reaction, they could be sufficiently

W(CO)6

CO/I2 /K2 C03

nucleophilic to complete the cyclization following initial reaction of the aliphatic amine.

In fact, under the carbonylation conditions 80 did form the cyclic urea, albeit in low yields (17 % yield in CH2CI2 and 20% yield in CH2CI2/H2 O). The results from 80 and 82

suggest that differentiation between aliphatic and aromatic amines is not sufficiently good to obtain selective reactivity in the presence of both.

Conclusion

In summary, aliphatic primary amines can be catalytically carbonylated to 1,3- disubstituted ureas in good to high yields using the commercially available, inexpensive and air stable W(CO)e as the catalyst. The preference of this catalyst for aliphatic amines 86

is complementary to the late transition metal catalysts, which are generally more effective

for aromatic amines than for aliphatic ones. Not only does this system provide an

alternative to phosgene and phosgene derivatives, but it is also compatible with a variety of functional groups, including unprotected alcohols. 87

CHAPTER 5 ALTERNATIVE SYNTHESIS OF CYCLIC UREA HIV-1 PROTEASE INHIBITORS

Introduction

Cyclic ureas have recently been of interest as the core structures of potent HIV protease inhibitors such as the clinical candidates DMP 323 and DMP 450.150,151,176

These compounds serve as inhibitors of the HIV- 1 -encoded aspartyl protease (HIV-Pr) which plays a fundamental role in the growth of immature virion proteins. [Kaltenbach,

1998 #1088] These cyclie ureas were designed to bind to HIV-Pr with the diol oxygens participating in symmetrical hydrogen bonding to the catalytic Asp 25/25'.

Figure 5.1 HIV Protease Inhibitors 88

There have been a few synthetic routes reported in the literature for compounds like DMP 323 and DMP 450, all of which involve problematic reagents or result in poor yieldsJ^^ Scheme describes the original cyclic urea core structure synthesis which used amino acid starting materials. Disadvantages to this synthesis include the

Scheme 5.1 Synthesis of protease inhibitor core structure via an amino acid

CbzHN CbzHN Me a b, c CbzHN NHCbz N. COOH H,C OMe O

CbzHN NHCbz

SEMO OSEM SEMO OSEM

SEM = 2-(TrimethylsiIyl)ethoxymethyl

g.h

HO OH

CORE STRUCTURE

Reagents; (a) j-BuOCOCl, A/'.O-dimethoxyhydroxylamine HCl; (b) LiAlH; (c) VCl3(THF)3, Zn-Cu;

(d) SEMCl; (e) cat. Pd(OH)2; (f) l.l'-carbonyldiimidazole; (g) Benzyl bromide, NaH; (h) HCl, dioxane/MeOH 89

tedious preparation of i?-configured amino acids and the intolerance of the vanadium

coupling reaction towards heteroatom-containing substituents. Disadvantages in terms of process plant scale-up are the inability to isolate intermediate products without the use of column chromatography, and the disposal of large quantities of the vanadium/zinc mixed

waste generated in the pinacol coupling step.^^^ Furthermore, the carbonylation step from the diamine to the cyclic urea uses the toxic and expensive phosgene derivative 1,1- carbonyldiimidazole.

Attempts were made to find alternative protecting groups that would result in crystalline products and eliminate the need for column chromatography. The diamine acetonide was prepared from diol 84 by protection with 2,2-dimethoxypropane followed

by subsequent hydrogenolysis of the bis-Cbz moiety (Scheme 5.2). Problems arose when the carbonylation of diamine 85 resulted in only trace quantities of the cyclic urea

86. The low yield is attributed to the strain of forming a trans fused bicyclo[3.5.0] decane ring system. Under optimized conditions, cyclization can occur under high temperatures and high dilutions in a chlorinated solvent in 67% yield. However, the volume inefficiency of this cyclization and the requirement for high temperature,

chlorinated solvents has rendered this conversion unacceptable for process scale-up.

Scheme 5.2 Preparation of the diamine acetonide 85

85 86 90

An alternative synthesis to the acetonide diamine 85 has been published which employs the ester 87 of L-tartaric acid as the starting material rather than an amino acid.[Rossano, 1995 #1090; Confalone, 1999 #615] 87 is reduced with DIBAL-H to a

1 dialdehyde followed by addition of 1 , -dimethylhydrazine which affords the bis- hydrazone 88. Subsequent reaction with benzyl lithium proceeds stereospecifically to afford dihydrazine 89. Diamine 85 is obtained after hydrogenolysis of 89 with Raney

nickel. Again, cyclization is achieved by reaction of 85 with 1 , 1'-carbonyldiimidazole.

Disubstituted urea 90 is then formed by the deprotonation and alkylation of the urea

nitrogens. Although this route is undoubtedly

Scheme 5.3 Synthesis of core structure starting from the ester of L-tartaric acid

MejN NMej 'nh HN 1)DIBAL-H

2) 2-propanol 3) HjNNMcj

87 RaNi

250 psi H2 100 “C

NaH Ph Ph Ph BnBr

CD! = l.r-carbonyidiimidazole

90 86 85 91

more efficient than the amino acid route, the cyclization step to form cyclic urea 86 remains a problem. To date, DuPont's process plant scale-up utilizes the Scheme 5.3

synthesis to produce diamine 85. However, once diamine 85 is formed, it is converted to

secondary diamine 91 upon which cyclization takes place in the presence of phosgene to the disubstituted urea 90 (Scheme 5.4). 90 can be isolated in 75% yield. Phosgene is used due to the inability of fl'-carbonyldiimidazole to carbonylate secondary diamines.

Scheme 5.4 Dupont process scale-up of cyclic urea 90

This chapter discusses the use of the W(CO)6 / I 2 carbonylation system as a possible alternative to using phosgene or phosgene derivatives in the cyclization of the primary and secondary diamines 85 and 91 to the corresponding cyclic ureas 86 and 90.

Carbonylation of Diamine 85

Diamine 85 was synthesized according to the method described in Scheme 5.3.

Diamine 85 was reacted under the following conditions: 2 mole % W(CO)6, stoichiometric amounts of diamine and I2, and 2 equiv of K2CO3 relative to diamine reacted in CH2CI2 at room temperature under 80 atm CO pressure. The reaction gave the 92

corresponding cyclic urea as shown in equation 5.1. Unfortunately, the carbonylation of diamine 85 to the cyclic urea 86 at room temperature resulted in only a 23% yield.

Although cyclization could occur without phosgene or phosgene derivatives, this low yield was not an improvement over the yield obtained in Scheme 5.2. However, when the reaction described above was repeated at 80 °C, cyclic urea 86 was formed in 70% yield.

Considering these results are unoptimized, the W(CO)6 / U -catalyzed system is an attractive alternative to the synthesis described in Scheme 5.2. Clearly the cyclization to

a trans fused ring is not a problem for the W(CO)6 / CO system at 80 °C.

Furthermore, keeping in mind the functional group compatibility studies in

Chapter 4, it is speculated that future carbonylation of highly functionalized diamine substrates will proceed without tedious functional group protection and deprotection steps.

Carbonylation of Disubstituted Diamine 91

Similar to the preceding section, disubstituted diamine 91 was synthesized according to Scheme 5.4. Using the same W(CO)6 / U reaction conditions described for the carbonylation of diamine 85 in equation 5.1, diamine 91 was carbonylated in an attempt to form cyclic urea 90 (equation 5.2). The reaction was run at room temperature 93

as well as 80 °C. Unfortunately, in both cases, unreacted starting material was isolated after work-up. There was no indication of the presence of cyclic urea 90 in the IR, 'H

NMR or C'^ NMR. Further attempts of producing 90 included running the reaction at

90 °C and 1 10 °C as well as changing the solvent to the biphasic solvent system CH2CI2/

H2 O (90 °C). Again, in all cases, unreacted starting material was isolated. Possibly changing the reaction conditions such as solvent or temperature could render this reaction viable in the future.

Conclusion

DMP 323 and DMP 450 are representative illustrations of possible synthetic uses of catalytic carbonylation. The cyclic urea core structure 86 of these protease inhibitors can be synthesized in 70 % yield using the W(CO)e / U-catalyzed system which has the potential of being an alternative route to 86, rather than using the previously reported high reaction temperatures and phosgene derivatives. Future work could include applications of this catalytic carbonylation system to highly functionalized diamines as well as further investigations into the cyclization of disubstituted diamine 91 to cyclic urea 90 under catalytic carbonylation conditions. 94

CHAPTER 6 EXPERIMENTALS

General

Standard inert atmosphere techniques were used in the syntheses of the metal

carbynes and metal carbene. Et20, THE, and toluene were distilled from Na/Ph2CO.

Hexane, CHCI3 (ethanol free unless stated), and methylene chloride were distilled from

CaH 2 . CDCI3 was degassed by three freeze-pump-thaw cycles and stored over 3 A molecular sieves. All other starting materials were purchased in reagent grade and used without further purification. The carbyne complexes 37-39 were synthesized by literature

^3-1 methods.^^’^ 15 Syntheses of 85-86 and 88-91 were based on published routes^^^ for which no experimental information was provided.

General Instrumentation

'H, '^C, and ^'P NMR spectra were recorded on Varian XL-300 and VXR-300

NMR spectrometers. 'H NMR spectra are referenced to the residual protons of the deuterated solvent. ^'P NMR spectra are referenced to 85% H3PO4 and are proton

decoupled. IR spectra were recorded on a Perkin-Elmer 1 600 spectrometer. A Hewlett-

Packard 8452A diode array spectrophotometer was used to obtain UV-visible absorption spectra. Photolyses were performed at -45 °C in 5 mm NMR tubes by irradiation with a

Hanovia medium pressure mercury vapor lamp in a Pyrex immersion well. 95

Photophysical Methods

Corrected emission and emission excitation spectra were obtained with a Spex

Industries F-112A spectrophotometer. Emission correction factors were generated by using a 1000 W tungsten primary standard lamp. Emission quantum yields are reported

^'^ relative to aqueous Ru(bpy) 3 (<3>em = 0.055). All room temperature measurements were carried out on argon-degassed THE solutions and studies at low temperature were conducted on argon-degassed 2-methyltetrahydrofuran (MTHF) solutions.

The instrumentation for the emission lifetime measurement on 39 has been described previously.1^^’1^1 Other emission lifetimes were measured on a

Photochemical Research Associates time-correlated single photon counting

spectrophotometer, which relies on an H2 gas-filled spark gap for an excitation source.

For emission decay experiments excitation light was filtered using a colored glass filter

(Schott, UG-1 1) and emission light was filtered using interference filters at 650, 700 or

750 nm. Emission decay analysis was carried out using the DECAN deconvolution software. Laser flash photolysis experiments were carried out on a system that has been described previously. Global analysis of the multiwavelength transient absorption data was effected using the SPECFIT factor analysis and software. ^^l

Syntheses r(COLWaOV2-napthvnirNMeJ (42)

2-Naphthyl bromide (4.0 g, 19 mmol) was dissolved in 60 mL of THF and cooled to -78° C. «-Butyllithium (2.0 M in pentane, 10 mL) was added under nitrogen and a

yellow suspension was observed. After the solution was allowed to stir for 1 h, the 96

entire solution was transferred by cannula to a flask containing W(CO)6 (6.75 g, 19

mmol) in THF at -78° C. The resulting red solution was allowed to stir for 2.5 h at -78°

C, and then the THF was removed under vacuum, leaving a red oily-solid which was

recrystallized from Et20/ hexanes to yield a red crystalline solid. The solid was dissolved

in 30 mL of degassed H 2 O, and the mixture was filtered through a fritted funnel into a

solution containing NMe4 Br (7.4 g, 48 mmol). The solution separated into two layers,

and the top layer was removed via cannula. The remaining red oil was placed under

vacuum to remove any additional H2O and used without further purification (9.3 g, 88%).

'H NMR (CDCI3): 6 7.4 - 8.0 (m, 7H), 2.25 (s, 12H).

Cl fCOHP(OMef2hW=C-t2-naphthvn t41f

Acyl complex 42 (9.3 g, 17 mmol) was dissolved in 70 mL of CH2CI2 and cooled

-95° to C. Oxalyl chloride (2.2 g, 17 mmol) was added, the solution was allowed to warm to -20° C, and evolution of gas was observed. After the solution was cooled once

-95° more to C, trimethyl phosphite was added (16.7 g, 135 mmol), and when it warmed to room temperature, more effervescence occurred. The mixture was then refluxed for 24

h, after which the solvent and excess P(OMe)3 were removed under vacuum. The remaining red oily solid was dissolved in Et20 and passed through a fritted funnel containing neutral alumina. The Et20 was removed under vacuum, leaving 41 as a red

oil in 58% yield. 'H NMR (CDCI 3 ): 5 7.4 - 7.9 (m, 7H), 3.6 (m, 27H).

Cpfcot f PtOMeL 1 W=C-t2-nanhthvn r40f

Tris (phosphite) complex 41 (2.9 g, 3.8 mmol) was dissolved in 45 mL of THF, 97

and CpNa (2.0 M in THF, 4.0 mL) was added. The dark red solution was refluxed for 24

h, after which THF and P(OMe)3 were removed under vacuum. The resulting dark oil

was dissolved in EtaO and passed through a short alumina column to remove unreacted

CpNa. After the solvent was removed, the red oil was placed on a longer alumina

column and washed with hexane. The yellow eluent was discarded, and further elution

with 1 :2 Et20/ hexane gave a red solution which was concentrated to give 40 as a red oil

in 20% yield. 'H NMR (CDCI 3 ): 5 7.81 (s, IH), 7.76 (d, IH), 7.69 (d, IH), 7.59 (d, IH,

(m, 3H), (s, 7.42 5.57 5H, Cp), 3.61 (d, Jhp = 12.0 Hz, 9H, [P(OMe) 3 ]). ‘^C NMR

(CDCI3 ): 6 286.5 (d, /cp = 18.2 Hz, W=C), 235.3 (d, Jcv = 9.7 Hz, CO), 150.0 (s, =C-Q,

133.0, 131.9 (bridge), 127.8, 127.7, 127.2, 126.9, 126.8, 126.4, 125.8 (Np), 90.0 (s, Cp),

(d, = 52.4 Jcp 2.4 Hz, P{OMe} 3 ). ^'P NMR (CDCI 3 ): 5 173.0 (Jwp = 662 Hz). IR

(CH2 CI2 ): 1888 cm ' (vwco)- UV (THF): 348 (s = 6000), 490 (e = 70) nm; HRMS

(FAB) m/z calcd for M"' (C20H21 O4PW), 540.0687; found, 540.0619.

(CO)4W=CrOCHAfc-C.Hg) (451

Carbene 45 was prepared by a slight modification of the literature method. It was

found that the yield could be increased by 20% if trimethyl tetrafluoroborate is used as the methylating reagent rather than methyl triflate, and if the product is purified by

column chromatography on alumina (hexane) under N2 atmosphere with eluent and column cooled to -78° C.

rCO(Br){P(OMe)3 )2W=Crc-C.HQftr49f

The carbene complex (CO) 5 W=C(OCH 3 )(c-C6H9) (45) (0.84 g, 1.9 mmol) was 98

-78° dissolved in 30 mL of CH2CI2 and cooled to C. A 1.0 M solution of BBrs in CH2CI2

(2.85 mL, 2.85 mmol) was added and the solution was stirred at -78° C for 10 min. The

-45° -78° solution was warmed to C, then cooled back down to C and P(OMe)3 (1.34 mL,

15.2 mmol) was added. The solution was refluxed for 10 hr. Removal of solvent in

vacuo and washing the product with Et20 yielded 49 as an orange-yellow oil (0.94 g,

65.3%). 'H NMR (CDCI 3 ): 5 1.63-1.95 (m, 4H), 2.00-2.40 (m, 4H), 3.64 (d, 9H, J=12

Hz), 3.95 (virtual triplet, 18H), 6.23 (hr, IH); IR (CH2 CI2 ): vco 1943 cm''.

rCp(CO) (P(OMe) 3 l WsCfc-aHoll (44)

Complex 49 (0.94 g, 1.2 mmol) was dissolved in 30 mL of THF and 2.0 eq of

CpNa (0.21 g, 2.4 mmol) was added. The solution was refluxed for 5 hours and the solvent was removed in vacuo. Chromatography of the resulting orange oil on neutral

4:1 alumina with hexane/Et20 followed by 100% Et20 yielded 44 (0.24 g, 40.5%): 'H

NMR (CDCI 3 ): 5 1.40-1.60 (m, 4H), 1.73-1.90 (br, 2H), 2.17-2.20 (br, 2H), 3.52 (d, 9H,

JPH= 12 Hz), 5.43 (s, 5H), 6.00 (s, IH); '^C (CDCI3 ): 5 21.5, 22.2, 25.6, 26.6, 52.3

= [P(OMe) 3 ], 89.5 (Cp), 129.8 (y-vinyl), 151.4 (P-vinyl), 235.0 (CO), 291.4 (d, W=C, Jpc

17 Hz); ^'P{'H} NMR (CDCI 3 ) 6 175.1 (Jwp-673 Hz); IR (CH2 CI2 ): Vco 1878 cm'';

HUMS (FAB) m/z calcd for M^ (C 16H23O4PW) 494.0843, found 494.0802.

CpCUP(OMeLlWrp^-CHrc-C.HQl1 (51)

Carbyne complex 44 (0.10 g, 0.20 mmol) was dissolved in 2 mL of Et20 and cooled to -10 "C. Upon addition of 1 equiv of ethereal HCl as a 1.0 M solution (2.2 mL),

51 immediately precipitated as a yellow powder. After further purification by washing 99

with cold Et 51 was obtained in yield. 20 , 28 % Note: complex 51 decomposes rapidly if

not kept below -45 °C during purification and sample preparation. 'H NMR (CDCI 3 ): 5

1.60-2.00 (m, 4H), 2.46-2.80 (m, 4H), 2.82-3.00 (br, IH), 3.78 (d, 9H, Jph = 1 1 Hz), 5.73

(s, 5H), 15.35 (s, IH); (CDCI 3 ): 6 20.5, 24.2, 30.7, 54.1 [P(OMe)3 ], 75.4 (P-vinyl),

(y-vinyl), = 93.7 94.9 (Cp), 277.8 (W=C); NMR (CDCI 3 ) 5 113.0 (Jwp 409 Hz);

MS (electrospray, CH 3 CN) m/z found 502.1 (M+). Isotope distribution pattern matched

simulation for C 15 H24CIO 3 PW.

Photooxidation of 44

Carbyne complex 44 (0.12 g, 0.24 mmol) was dissolved in 10 mL of CHCI 3 and cooled to -45 °C. The solution was photolyzed for 5 hr at -45 °C while the disappearance of 44 was monitored by TLC. Upon completion of the photolysis, the solvent was

removed in vacuo at -45 °C leaving a red residue. CDCI 3 that had been cooled to just above its freezing point was cannulated into the remaining residue. The solution was then transferred to a chilled NMR tube and a 'H NMR spectrum was taken at -50 °C.

General Procedures for the Catalytic Carbonylation of Benzylamines with

W ( CO ),

Procedure A

To a stirred solution of W(CO)6 (49 mg, 0.14 mmol) in 20 mL of CH2CI2 in the glass liner of a Parr high-pressure vessel was added 50 equiv of p-methoxybenzylamine

(1.0 mL, 7.1 mmol), 75 equiv of K2CO3 (1.47 g, 10.7 mmol) and 25 equiv of iodine (0.89 g, 3.5 mmol). The vessel was then charged with 85 atm of CO, heated to 70 °C and left 100

to stir under pressure for 24 h. The pressure was released and the maroon solution was filtered. The solid collected on the filter paper was heated in CH2CI2 , filtered, and both

filtrates were combined. The combined CH2 CI 2 layers were washed with a 1.0 M HCl

solution followed by washing with a saturated Na2 S03 solution. The resulting pale

yellow solution was then dried with Na2 S04 and filtered. The solution was concentrated

to yield a pale yellow solid, which was then rinsed with 1 5 mL of ether. The urea was

obtained as a white solid (0.50 g, 1.7 mmol, 47% yield). The solid was identified as

N,N’-bis(4-methoxybenzyl)urea by comparison with literature values.

Procedure B

Procedure B is identical to that of procedure A except the reaction was diluted to

O. 03 M in substrate and run at room temperature.

Procedure C

is Procedure C identical to that of procedure A except 2 1 mL of CH 2CI2 and 3 mL

of H2 O were employed as solvent. The product urea precipitated out of the reaction mixture, was filtered, washed with Et20 and collected.

Preparation of N.N’-dibenzvlurea 55

Procedure A afforded 55 in 63% yield. Procedure C afforded 55 in 73% yield.

Urea 55 was identified by comparison with literature values. ^ 101

Preparation of N.N’-Bis(4-chlorobenzynurea 57

Procedure A afforded 57 in 35% yield. Procedure C afforded 57 in 77% yield:

'H NMR (DMSO-de) 5 7.37 (d, J = 7.6 Hz, 4H), 7.26 (d, J = 7.8 Hz, 4H), 6.56 (br s, 2H),

4.20 (d, J = 5.4 Hz, 4H); '^C NMR (DMSO-d6) 5 158.0 (C=0), 140.1, 131.0, 128.8,

128.1, 42.3; IR (Nujol) 1567 cm'’; HRMS (FAB) calcd for C 15 H 15 CI 2N2 O (M+H^)

309.0561, found 309.0561.

Preparation of N.N’-Bis(4-bromobenzvBurea 59

Procedure A afforded 59 in 30% yield. Procedure C afforded 59 in 77% yield:

'H NMR (DMSO-de) 5 7.50 (d, J = 8.1 Hz, 4H), 7.20 (d, J = 8.2 Hz, 4H), 6.53 (br s, 2H),

= 4.19 (d, J 5.9 Hz, 4H); '^C NMR (DMSO-d6 ) 6 158.0 (C=0), 140.5, 131.0, 129.2,

119.5, 42.3; IR (Nujol) 1561 cm''; HRMS (FAB) calcd for Ci 5 Hi 5 Br2N20 (M+H^)

396.9551, found 396.9560.

Preparation of N.N’-Bis(3-iodobenzvnurea 61

Procedure A afforded 61 in 39% yield. Procedure C afforded 61 in 70% yield:

'H NMR (DMSO-de) 6 7.60 (s, 2H), 7.56 (d, J = 7.7 Hz, 2H) 7.25 (d, J = 7.5 Hz, 2H),

7.10 (t, 2H), 6.49 (t, 2H), 4.19 (d, J = 5.9 Hz, 4H); '^C NMR (DMSO-de) 5 158.0 (C=0),

143.8, 135.5, 135.2, 130.5, 126.4, 94.8, 42.2; IR (Nujol) 1584 cm’’; HRMS (FAB) calcd

for C 15 H 15 I2N2O (M+H^) 492.9274, found 492.9271.

Preparation of N.N’-Bis('4-methoxvbenzvBurea 63

Procedure A afforded 63 in 47% yield. Procedure C afforded 63 in 70% yield. 102

Literature reports on 63 do not eontain spectral data. *H NMR (DMSO-de) 5 7.15 (d,

J = 8.4 Hz, 4H), 6.85 (d, J = 8.2 Hz, 4H), 6.31 (br s, 2H), 4.13 (d, J = 5.4 Hz, 4H), 3.71

(s, 6H); NMR (DMSO-de) 6 158.04 (C=0), 157.98, 132.8, 128.3, 113.6, 55.0, 42.4;

IR (Nujol) 1578 cm'*; HRMS (FAB) calcd for C17H21N2O3 (M+H^) 301.1552, found

301.1557.

Preparation of N.N’-Bis(4-methvlthiobenzvl')urea 65

Procedure A afforded 65 in 24% yield. Procedure C afforded 65 in 81% yield:

'H NMR (DMSO-de) 5 7.19 (m, 8H), 6.42 (br t, 2H), 4.16 (d, J= 6.0 Hz, 4H), 2.44 (s,

6H); '^C NMR (DMSO-d6) 5 158.0 (C=0), 137.7, 135.9, 127.8, 126.1, 42.5, 15.0; IR

(Nujol) 1584 cm'*; HRMS (Cl) calcd for C17H21N2OS2 (M+H^) 333.1095, found

333.1106.

Preparation of N.N'-Bis(4-hvdroxvmethvlbenzvBurea 67

Procedure A afforded 67 in 5% yield. Procedure C afforded 67 in 58% yield. *H

NMR (DMSO-de) 5 7.41 (m, 8H), 6.56 (t, 2H), 5.31 (t, 2H), 4.65 (d, J- 5.4 Hz, 4H), 4.39

(d, J= 6.0 Hz, 4H); ‘^C NMR (DMSO-de) 5 158.0 (C=0), 140.8, 139.2, 126.8, 126.4,

62.7, 42.8; IR (Nujol) 3331, 1555 cm''; HRMS (FAB) calcd for C17H21N2O3 (M+H^)

301.1552, found 301.1533.

Preparation of N.N’-Bis(ethvl 4-carboxvlbenzvBurea 71

Procedure A afforded 71 in 36% yield. Procedure C afforded 71 in 55% yield:

'H NMR (DMSO-de) 6 7.91 (d, J = 8.2 Hz, 4H), 7.38 (d, J = 8.2 Hz, 4H), 6.65 (br t, 2H), 103

4.31 (s, 4H), 4.30 (qt, 4H), 1.32 (t, 6H); NMR (DMSO-de) 6 165.7, 158.1 (C=0),

146.7, 129.1, 128.2, 127.0, 60.6, 42.8, 14.2; IR (CH2CI2 ) 1708, 1678 cm'’; HRMS (FAB)

calcd for C 21 H25N2O 5 (M+H^) 385.1763, found 385.1760.

Preparation of N.N’-Bis(4-carboxvlic acid benzvBurea 73

Procedure C was employed except that the product 73 did not precipitate out until the reaction mixture was acidified with IN HCl. Urea 73 was obtained in 37% yield; 'H

NMR (DMSO-d6) 5 12.89 (s, 2H), 7.88 (d, J= 7.6 Hz, 4H), 7.35 (d, 7.8 Hz, 4H), 6.62

(t, 2H), 4.29 (d, J= 5.9 Hz, 4H); NMR (DMSO-dg) 5 167.2, 158.1(NHC=0), 146.2,

129.3, 126.9, 109.4, 42.8; IR (Nujol) 1684, 1561 cm"'; MS (Electrospray) for C 17 H 16N2O5

(M+H^) 329.3, calcd 329.3.

Preparation of N.N’-Bis(4-ethenvlbenzvnurea 75

When procedure A was employed, no product was isolated. The viscous material obtained suggested the formation of oligomers. Procedure B afforded a 69% yield of 75.

When procedure C was employed, 75 was only isolated in 19% yield: 'H NMR (DMSO-

de) 6 7.40 (d, J = 7.9 Hz, 4H), 7.21 (d, J = 7.9 Hz, 4H), 6.70 (qt, 2H), 6.45 (br s, 2H), 5.78

(d, J = 17.7 Hz, 2H), 5.21 (d, J = 10.9 Hz, 2H), 4.21 (d, J = 5.6 Hz, 4H); NMR

5 (DMSO-de) 158.1 (C-0), 140.7, 136.4, 135.5, 127.2, 126.0, 113.7, 42.7; IR (CH2 CI2 )

1678 cm'*; Anal. Calcd for C 19H20N2 O; C, 78.05; H, 6.89; N, 9.58. Found: C, 74.70; H,

6.95; N, 9.03. .

104

Preparation of N.N’-Bis('4-nitrobenzyllurea 77

Procedure A afforded 77 in 45% yield. Procedure C afforded 77 in 76% yield:

'H NMR (DMSO-de) 6 8.20 (d, J = 8.5 Hz, 4H), 7.51 (d, J = 8.5 Hz, 4H), 6.85 (br t, 2H),

4.34 (d, J = 6.0 Hz, 4H); '^C NMR (DMSO-ds) 5 158.0 (C-0), 149.3, 146.2, 127.8,

123.4, 42.6; IR (Nujol) 1584, 1537 cm'*; Anal. Calcd for C 15 H 14N4O 5 : C, 54.54; H, 4.27;

N, 16.96. Found: C, 54.07; H, 4.23; N, 16.66.

Preparation of N.N’-Bis(4-cyanobenzvn urea 79

Procedure A afforded 79 in 37% yield. Procedure C afforded 79 in 68% yield:

'H NMR (DMSO-d6) 5 7.80 (d, J = 7.3 Hz, 4H), 7.44 (d, J - 7.0 Hz, 4H), 6.79 (br s, 2H),

4.31 (d, J = 4.9 Hz, 4H); '^C NMR (HMSO-dg) 5 158.0 (C=0), 147.1, 132.2, 127.7,

119.0, 109.2, 42.7; IR (CH2CI 2 ) 2214, 1678 cm’'; HRMS (FAB) calcd for Ci 7H, 5N40

(M+H^) 29 1 . 1 246, found 29 1 . 1 249

Preparation of N.N’-Bis(4-aminobenzvnurea 81

When procedure A was employed, the reaction formed a solid black mass.

Procedure B afforded the urea 81 in 28% yield after recrystallization in EtOH. Procedure

C afforded 81 in 14% yield. The urea 81 was identified by comparison with literature values.

Preparation of 3.4-Dihvdro-2niT)-quinazolinone 83

Procedure A afforded 83 in 17% yield. Procedure C afforded 83 in 20% yield.

The urea 83 was identified by comparison with literature values, 105

Preparation of (4R. 5R')-2.2-DimethvI-4.5-bisformvl-1.3-dioxolane N.N- Dimethvlhvdrazone 88

To a -40 °C chilled solution of dimethyl 2,3-0-isopropylidene-L-tartrate (2.18 g,

10 mmol) in toluene was added DIBAL-H (22.0 mL, 22.0 mmol) dropwise and allowed

to stir at -40 °C for 1 hr. Upon addition of methanol (0.89 mL, 22.0 mmol), the solution was warmed to -10 °C. N,N-dimethylhydrazine (1.67 mL, 22.0 mmol) was then added

and allowed to stir for 1 hr at -10 °C. The reaction was quenched with 30 mL of H2O, extracted with Et20, dried over MgS04, and concentrated to give 88 as a white solid

(2.35 g, 97 %). 'H NMR (CDCI3) 5 6.37 (s, 2H), 4.46 (s, 2H), 2.85 (s, 12H), 1.47 (s,

6H).

Preparation of (3R. 4S. 5S. 6RV2.2-Dimethvl-4.5-bisformvl-1.3-dioxolane-3.6- bis(phenvlmethvl') N.N-Dimethvlhvdrazine 89

To a 100 mL Schlenk flask under N2 was added toluene (12.8 mL, 120 mmol) and

5ec-butyl lithium (16.0 mL, 20 mmol). The solution was cooled to -20 °C and THF (5.0 mL, 60 mmol) was added dropwise. Upon formation of yellow crystalline mass, the reaction was allowed to stir at room temperature for 2 hr. The reaction was then cooled

to -15 °C and 88 (1.78 g, 7.0 mmol), which was dissolved in 10 mL of toluene, was added

dropwise. The reaction stirred for 1 hr at -15 °C and then was quenched with H2O, followed by extraction with Et20. The organic layer was dried over MgS04 and fumaric

acid (0.86 g, 7.0 mmol) was added. The mixture stirred overnight upon which the fumaric acid salt of 89 had precipitated. The precipitate was collected, dissolved in H2O, and deprotonated by the addition of a saturated NaHCOs solution. The aqueous mixture was then extracted with Et20, dried over MgS04, and concentrated to give 89 as a

colorless oil (2.12 g, 71 %). 'H NMR (CDCI3) 5 7.25 (m, lOH), 4.1 (s, 2H), 3.1 (m, 2H), 106

2.7 (m, 4H), 2.1 (s, 12H), 1.4 (s, 6H); NMR (CDCI3) 5 139.73, 129.60, 128.47,

126.29, 107.72, 77.53, 59.07, 47.92, 37.61, 27.56.

Preparation of ('2R. 3S. 4S. 5R')-2.2-Dimethvl-3.4-bisformvl-1.3-dioxolane-2.5- bistphenvlmethvndiamine 85

To a solution of 89 (0.6 g, 1.4 mmol) in methanol contained in a Parr high

pressure vessel was added 3 g of wet Raney Ni (washed once with MeOH). The vessel

was charged with 500 psi H2 and heated to 80 °C for 18 hr. The reaction was then filtered

through celite and concentrated to an oil. The oil was columned on alumina with 99:1

EtOAc/NEts followed by 30:10 EtOAc/MeOH to give 85 (0.11 g, 23 %). 'H NMR

(CDCI3) 5 7.20 (m, lOH), 4.05 (s, 2H), 3.10 (m, 2H), 2.80 (m, 2H), 2.61 (m, 2H), 2.50 (s,

4H), 1.41 (s, 6H); ‘^C NMR (CDCI3) 5 138.48, 129.23, 128.55, 126.47, 108.86, 79.60,

53.66,41.34, 27.32.

Preparation of t4R. 5S. 6S. 7R')-Hexahvdro-5.6-0-isopropvlidene-4.7-bis(phenvlmethvlV 2//-1.3-diazapin-2-one 86

Procedure A was employed except that 25 equiv of diamine 85, 50 equiv of

K2CO3, and 25 equiv of I2 were used. Procedure A at room temperature afforded 86 in

23% yield. Procedure A at 80 °C afforded 86 in 70% yield. The urea 86 was identified by comparison with literature values.

Preparation of t2R. 3S. 4S, 5R')-2.2-Dimethvl-3.4-bisformvl-1.3-dioxolane-2.5- bistphenylmethyll N.N-Dibenzvldiamine 91

Diamine 85 (50 mg, 0.15 mmol) was refluxed with benzaldehyde (0.03 mL, 0.29

mmol) in toluene for 1 hr. The reaction was cooled followed by the addition of acetic 107

acid (0.02 mL, 0.29 mmol) and NaBH(OAc)3 (75 mg, 0.35 mmol). The reaction was heated for 2 hr at 35 “C and then quenched with a saturated NaHCOs solution. The

mixture was then extracted with EtOAc, dried over MgS04 and concentrated to give 91

as a white solid (76 mg, 100 %). *H NMR (CDCI 3 ) 5 7.38 (m, 20H), 4.29 (s, 2H), 3.85

(m, 2H), (m, 3.71 4H), 2.8 (m, 4H), 1.60 (s, 6H); ‘^C NMR (CDCI 3 ) 5 150.54, 139.46,

129.25, 128.31, 128.10, 126.67, 126.01, 107.87, 77.44, 57.85, 51.53, 38.21, 27.30.

Attempted preparation of 14R. 5S, 6 S. 7RVHexahvdro-5.6-(9-isopropvlidene-4.7- bistphenvlmethvlVN.N-dibenzvl- 1 .3-diazapin-2-one 90

The carbonylation of diamine 91 was attempted using procedure A at room temperature, 80 °C, 90°C, and 110 °C. No evidence for product 90 was observed in the

IR, 'H NMR, or NMR.

The carbonylation of diamine 91 was also attempted using procedure C at 90 °C.

Again, no evidence for product 90 was observed in the IR, 'H NMR, or ’^C NMR. 108

REFERENCES

(1) Schrock, R. R. Accounts ofChemical Research 1986, 19, 342-348.

(2) Zaera, F.; Somorjai, G. A. Journal of the American Chemical Society 1984, 106, 2288-2293.

(3) Fischer, H.; Hofmann, P.; Kreissl, F. R.; Schrock, R. R.; Schubert, U.; Weiss, K. Carbyne Complexes', VCH: Weinheim, Germany, 1988.

(4) Kim, H. P.; Angelici, R. J. Advances in Organometallic Chemistry 1987, 27, 51-

111 .

(5) Mayr, A.; Hoffmeister, H. Advances in Organometallic Chemistry 1991, 32, 227- 324.

(6) Mayr, A. Comments Inorg. Chem. 1990, 10, 227-266.

(7) Fischer, E. O. Angew. Chem. 1974, 18, 651.

(8) Fischer, E. O.; Schubert, U.; Fischer, H. Pure and Applied Chemistry 1978, 50, 857-870.

(9) Fischer, E. O.; Schubert, U. J. Organomet. Chem. 1975, 100, 59-81.

(10) Jeffery, J. C.; Went, M. J. Polyhedron 1988, 7, 775-790.

(11) Stone, F. G. A. Abstracts ofPapers of the American Chemical Society 1985, 189, 2-INOR.

(12) Stone, F. G. A. Abstracts ofPapers of the American Chemical Society 1988, 195, 557-INOR.

(13) Dahlke, P.; Jeffery, J. C.; Mortimer, M. D.; Stone, F. G. A. Polyhedron 1992, 11, 1587-1596.

(14) Gallop, M. A.; Roper, W. R. Advances in Organometallic Chemistry 1986, 25, 121-198. 109

(15) Pourreau, D. B.; Geoffroy, G. L. Advances in Organometallic Chemistry 1985 , 24, 249-352.

(16) Schrock, R. R. Journal of Organometallic Chemistry 1986, 300, 249-262.

(17) Schrock, R. R. Science 1983 , 219, 13-18.

(18) Buhro, W. E.; Chisholm, M. H. Advances in Organometallic Chemistry 1987, 27, 311-369.

(19) Mayr, A.; Bastos, C. M. Prog. Inorg. Chem. 1992 , 40, 1.

(20) Engel, P. F.; Pfeffer, M. Chemical Reviews 1995 , 95, 2281-2309.

(21) Fischer, E. O.; Kreis, G.; Kreiter, C. G.; Muller, J.; Huttner, G.; Lorentz, H. Angewandte Chemie-International Edition in English 1973, 564-565.

(22) Hall, M. B. Inorg. Chem. 1976 , 21, 179.

(23) Kostic, N. M.; Fenske, R. F. Journal of the American Chemical Society 1981, 103, 4677-4685.

(24) Kostic, N. M.; Fenske, R. F. Organometallics 1982 , 1, 489-496.

(25) Kostic, N. M.; Fenske, R. F. Journal of the American Chemical Society 1982, 104, 3879-3884.

Poblet, J. Strich, (26) M.; A.; Wiest, R.; Benard, M. Chemical Physics Letters 1986 , 126, 169-175.

(27) Green, M.; Orpen, A. G.; Williams, 1. D. Journal of the Chemical Society-

Chemical Communications 1982 , 493-495.

(28) Bottrill, M.; Green, M. Journal ofthe American Chemical Society 1977, 99, 5795- 5796.

(29) Holmes, S. J.; Schrock, R. R. Journal of the American Chemical Society 1981, 103, 4599-4600.

(30) Jeffery, J. C.; Laurie, J. C. V.; Moore, 1.; Stone, F. G. A. Journal of

Organometallic Chemistry 1983 , 258, C37-C40.

(31) Howard, J. A. K.; Jeffery, J. C.; Laurie, J. C. V.; Moore, L; Stone, F. G. A.;

Stringer, A. Inorganica Chimica Acta-Articles and Letters 1985 , 100, 23-32. 110

(32) Kreissl, F. R.; Sieber, W. J.; Keller, H.; Riede, J.; Wolfgruber, M. Journal of Organometallic Chemistry 1987, 320, 83-90.

(33) Kreissl, F. R.; Sieber, W. J.; Wolfgruber, M.; Riede, J. Angewandte Chemie- International Edition in English 1984, 23, 640-640.

(34) Bottrill, M.; Green, M.; Orpen, A. G.; Saunders, D. R.; Williams, I. D. Journal of the Chemical Society-Dalton Transactions 1989, 511-518.

(35) Bastos, C. M.; Lee, K. S.; Kjelsberg, M. A.; Mayr, A.; Van Engen, D.; Koch, S.

A.; Franolic, J. D.; Klooster, W. T.; Koetzle, T. F. Inorganica Chimica Acta 1998, 279, 7-23.

(36) Feng, S. G.; White, P. S.; Templeton, J. L. Journal of the American Chemical Society 1990, 112, 8192-8193.

(37) Torraca, K. E.; Ghiviriga, I.; McElwee-White, L. Organometallics 1999, 18, 2262-2266.

(38) Dotz, K. H. Pure and Applied Chemistry 1983, 55, 1689-1706.

(39) Ddtz, K. H. Angew. Chem., Int. Ed. Engl. 1975, 14, 644.

(40) Ddtz, K. H. Transition Metal Carbene Complexes', Verlag Chemie: Deerfield Beach, FL, 1983.

(41) Ddtz, K. FI. Reactions of Coordinated Ligands', Plenum: New York, 1986.

(42) Garrett, K. E.; Sheridan, J. B.; Pourreau, D. B.; Wu, C. F.; Geoffroy, G. L.; Staley, D. L.; Rheingold, A. L. Journal of the American Chemical Society 1989, 777,8383-8391.

(43) Bocarsly, A. B.; Cameron, R. E.; Rubin, H. D.; McDermott, G. A.; Wolff, C. R.; Mayr, A. Inorganic Chemistry 1985, 24, 3976-3978.

(44) Mayr, A.; Kjelsberg, M. A.; Lee, K. S.; Asaro, M. F.; Hsieh, T. C.

Organometallics 1987, 6, 2610-2612.

(45) Sheridan, J. B.; Pourreau, D. B.; Geoffroy, G. L.; Rheingold, A. L.

Organometallics 1988, 7, 289-294.

(46) Oganometallic Radical Processes', Trogler, W. C., Ed.; Elsevier: Amsterdam, 1990; Vol. 22.

(47) Baird, M. C. Chemical Reviews 1988, 88, 1217-1227. Ill

(48) Astruc, D. Accounts ofChemical Research 1991, 24, 36-42.

(49) Carter, J. D.; Kingsbury, K. B.; Wilde, A.; Schoch, T. K.; Leep, C. J.; Pham, E. K.; McElwee-White, L. Journal of the American Chemical Society 1991, 113, 2947-2954.

(50) Leep, C. J.; Kingsbury, K. B.; McElwee-White, L. Journal of the American Chemical Society 1988, 110, 7535-7536.

(51) Kingsbury, K. B.; Carter, J. D.; Wilde, A.; Park, H.; Takusagawa, F.; McElwee- White, L. Journal of the American Chemical Society 1993, 115, 10056-10065.

(52) Schoch, T. K.; Orth, S. D.; Zemer, M. C.; Jorgensen, K. A.; McElwee-White, L. Journal of the American Chemical Society 1995, 117, 6475-6482.

(53) Torraca, K. E.; McElwee-White, L. Coord. Chem. Rev. , in press.

(54) Falbe, J. New Syntheses with Carbon Monoxide', Springer-Verlag: Berlin, 1980.

(55) Colquhoun, H. M.; Thompson, D. J.; Twigg, M. V. Carbonylation: Direct Synthesis of Carbonyl Compounds', Plenum: New York, 1991.

(56) Ojima, I. Chemical Reviews 1988, 88, 101 1-1030.

(57) Alper, H. Pure and Applied Chemistry 1988, 60, 35-38.

(58) Kalck, P.; Serra, C.; Machet, C.; Broussier, R.; Gautheron, B.; Delmas, G.; Trouve, G.; Kubicki, M. Organometallics 1993, 12, 1021-1022.

(59) Xu, Q.; Nakatani, H.; Souma, Y. Journal of Organic Chemistry 2000, 65, 1540- 1543.

(60) Rosini, G. P.; Boese, W. T.; Goldman, A. S. Journal of the American Chemical Society 1994, 116, 9498-9505.

(6 1 ) Wu, H. S.; Tan, W. H. Journal of Catalysis 1998, 1 78, 604-6 1 0.

(62) Weston, W. S.; Lightfoot, P.; Cole-Hamilton, D. J. Journal of Organometallic Chemistry 1998, 553, 473-476.

(63) Gargulak, J. D.; Berry, A. J.; Noirot, M. D.; Gladfelter, W. L. Journal of the American Chemical Society 1992, 114, 8933-8945.

(64) Yoneda, E.; Kaneko, T.; Zhang, S. W.; Takahashi, S. Tetrahedron Letters 1998, 59, 5061-5064. 112

(65) Alper, H.; Calet, S. Tetrahedron Letters 1988 , 29, 1763-1766.

(66) Xu, Q.; Imamura, Y.; Fujiwara, M.; Souma, Y. Journal of Organic Chemistry

1997 , 62, 1594-1598.

(67) Ryan, A. J.; Stanford, J. L. Comprehensive Polymer Science', Pergamon Press: Oxford, 1989; Vol. 5.

(68) Vishnyokova, T. P.; Golubeva, I. A.; Glebova, E. V. Russ. Chem. Rev. 1985, 54, 249-261.

(69) Adams, P.; Baron, F. A. Chem. Rev. 1965 , 65, 567.

(70) Moling, P.; Tarranga, A. Comprehensive Organic Functional Group Transformations', Pergamon Press: Oxford, 1995.

(71) Giannoccaro, P.; Nobile, C. F.; Mastrorilli, P.; Ravasio, N. Journal of

Organometallic Chemistry 1991 , 419, 251-258.

(72) Srivastava, S. C.; Shrimal, A. K.; Srivastava, A. Journal of Organometallic

Chemistry 1991 , 414, 65-69.

(73) Bassoli, A.; Rindone, B.; Tollari, S.; Chioccara, F. Journal ofMolecular Catalysis

1990 , (50,41-48.

Pri-Bar, I.; J. (74) Schwartz, Journal ofOrganic Chemistry 1995 , 60, 8124-8125.

(75) Pri-Bar, I.; Alper, H. Canadian Journal of Chemistry-Revue Canadienne De Chimie 1990, 68, 1544-1547.

(76) Giannoccaro, P. Journal of Organometallic Chemistry 1994, 470, 249-252.

(77) Kanagasabapathy, S.; Thangaraj, A.; Gupte, S. P.; Chaudhari, R. V. Catalysis

Letters 1994 , 25, 361-364.

(78) Alper, H.; Hartstock, F. W. Journal of the Chemical Society-Chemical

Communications 1985 , 1141-1142.

(79) Alper, H.; Vasapollo, G.; Flartstock, F. W.; Mlekuz, M.; Smith, D. J. H.; Morris,

G. E. Organometallics 1987 , 6, 2391-2393.

(80) Valli, V. L. K.; Alper, H. Organometallics 1995 , 14, 80-82.

(81) Kondo, T.; Sone, Y.; Tsuji, Y.; Watanabe, Y. Journal of Organometallic

Chemistry 1994 , 473, 163-173. 113

148-3 1 (82) Sternberg, H. J. Am. Chem. Soc. 1953 , 75, 3 52.

(83) Mori, S.; Matsuyoshi, H.; Kudo, K.; Sugita, N. Chemistry Letters 1991 , 1397- 1400.

(84) Piotti, M. E.; Alper, H. Journal ofthe American Chemical Society 1996, 118, 111- 116.

(85) Sheldon, R. A. Chemicals from Synthesis Gas: Catalytic Reactions of CO and H\ Reidel Pub. Co.: Boston, 1983.

(86) Anastasiou, D.; Jackson, W. R. Journal of Organometallic Chemistry 1991, 413, 399-410.

(87) Zhou, J. Q.; Alper, H. Journal ofOrganic Chemistry 1992, 57, 3328-3331.

(88) Mulla, S. A. R.; Gupte, S. P.; Chaudhari, R. V. Journal of Molecular Catalysis

1991 , 67,L7-L10.

(89) Hieber, W. Chem. Ber. 1958 , 91, 2223-2233.

(90) Dombek, B. D.; Angelici, R. J. Journal of Catalysis 1977 , 48, 433-435.

(91) Byerly, J. J.; Rempel, G. L.; Takebe, N.; James, B. R. J. Chem. Soc. Chem.

Comm. 1971 , 1482-1483.

(92) Kealy, T. J.; Benson, R. E. J. Org. Chem. 1961 , 26, 3126.

(93) Tsuji, Y.; Ohsumi, T.; Kondo, T.; Watanabe, Y. Journal of Organometallic

Chemistry 1986 , 309, 333-344.

(94) Dombek, B. D.; Angelici, R. J. Journal of Organometallic Chemistry 1977, 134, 203-217.

(95) Calderazzo, F. Inorganic Chemistry 1965 , 4, 293-296.

(96) Hieber, W.; Schuster, L. Z. Anorg. Allg. Chem. 1956 , 287, 214-222.

(97) Doxsee, K. M.; Grubbs, R. H. Journal of the American Chemical Society 1981 , 103, 7696-7698.

(98) Jetz, W.; Angelici, R. J. Journal of the American Chemical Society 1972 , 94, 3799-3802.

(99) McCusker, J. E.; Abboud, K. A.; McElweeWhite, L. Organometallics 1997, 16, 3863-3866. 114

(100) Geoffrey, G. L.; Wrighton, M. S. Organometallic Photochemistry, Academic Press: New York, 1976.

(101) Beevor, R. G.; Freeman, M. J.; Green, M.; Morton, C. E.; Orpen, A. G. Journal of

the Chemical Society-Dalton Transactions 1991 , 3021-3030.

(102) Mayr, A.; Asaro, M. F.; Glines, T. J. Journal of the American Chemical Society

1987 , 109, 2215-2216.

(103) Green, M. Journal of Organometallic Chemistry 1986, 300, 93-109.

(104) McElwee-White, L.; Kingsbury, K. B.; Carter, J. D. Journal of Photochemistry

and Photobiology A-Chemistry 1994 , 80, 265-270.

(105) McElwee-White, L.; Kingsbury, K. B.; Carter, J. D. Advances in Chemistry Series

1993 , 335-349.

(106) McElwee-White, L. Synlett 1996, 806-814.

(107) Mortimer, M. D.; Carter, J. D.; McElwee-White, L. Organometallics 1993 , 12, 4493-4498.

(108) Carter, J. D.; Schoch, T. K.; McElwee-White, L. Organometallics 1992 , 11, 3571- 3578.

(109) McElwee-White, L. Inter-American Photochemical Society Newsletter 1999, 22, 25-31.

(1 10) Bocarsly, A. B.; Cameron, R. E.; Mayr, A.; McDermott, G. A. In Photophysics and Photochemistry of Coordination Compounds', Yersin, H., Vogler, A., Eds.; Springer-Verlag: Berlin, 1987; pp 213-216.

(111) Trammell, S.; Sullivan, B. P.; Hodges, L. M.; Harman, W. D.; Smith, S. R.;

Thorp, H. H. Inorganic Chemistry 1995 , 34, 2791-2792.

(1 12) Manna, J.; Gilbert, T. M.; Dallinger, R. F.; Geib, S. J.; Hopkins, M. D. Journal of

the American Chemical Society 1992 , 114, 5870-5872.

(113) Fischer, E. O.; Massbol, A. Chemische Berichte 1967 , 100, 2445-2456.

(114) Mayr, A.; Dorries, A. M.; McDermott, G. A.; Vanengen, D. Organometallics

1986 , 5, 1504-1506.

(115) McDermott, G. A.; Dorries, A. M.; Mayr, A. Organometallics 1987 , 6, 925-931. 115

(116) Schoch, T. K. , PhD Dissertation, Stanford University, 1994.

(117) Mortimer, M. D. , PhD Dissertation, University of Bristol, 1991.

(118) Lees, A. J.; Adamson, A. W. Journal of the American Chemical Society 1982, 704,3804-3812.

(119) Lees, A. J. Journal ofthe American Chemical Society 1982, 104, 2038-2039.

(120) Lees, A. J. Chemical Reviews 1987, 57, 71 1-743.

(121) Stultz, L. K.; Binstead, R. A.; Reynolds, M. S.; Meyer, T. J. Journal of the American Chemical Society 1995, 777, 2520-2532.

(122) Carmichael, L; Hug, G. L. Journal of Physical and Chemical Reference Data 1986, 15, 1-250.

(123) Lucia, L. A.; Schanze, K. S. Inorganica Chimica Acta 1994, 225, 41-49.

(124) Creutz, C.; Chou, M.; Netzel, T. L.; Okumura, M.; Sutin, N. Journal of the American Chemical Society 1980, 102, 1309-1319.

(125) Langford, C. H.; Malkhasian, A. Y. S.; Sharma, D. K. Journal of the American Chemical Society 1984, 106, 2727-2728.

(126) Ohno, T.; Yoshimura, A.; Malaga, N. Journal of Physical Chemistry 1990, 94, 4871-4876.

(127) Watts, R. J. Journal ofChemical Education 1983, 60, 834-842.

(128) Cavalheiro, C. C. S.; Torraca, K. E.; Schanze, K. S.; McElwee-White, L. Inorganic Chemistry 1999, 38, 3254-3257.

(129) Kingsbury, K. B.; Carter, J. D.; McElwee-White, L. Journal of the Chemical Society-Chemical Communications 1990, 624-625.

(130) Astruc, D. Electron-Transfer and Radical Processes in Transition-Metal Chemistry, VCH: New York, 1995.

(131) Wulff, W. D.; Bauta, W. E.; Kaesler, R. W.; Lankford, P. J.; Miller, R. A.; Murray, C. K.; Yang, D. C. Journal of the American Chemical Society 1990, 112, 3642-3659.

(132) Cusack, N. J.; Reese, C. B.; Risius, A. C.; Roozpeikar, B. Tetrahedron 1976, 32, 2157-2162. 116

(133) Chamberlin, A. R.; Stemke, J. E.; Bond, F. T. Journal of Organic Chemistry

1978 , 43, 147-154.

(134) Harvey, D. F.; Grenzer, E. M.; Gantzel, P. K. Journal of the American Chemical

Society 1994 , 116, 6719-6732.

(135) Casey, C. P.; Miles, W. H.: Tukada, H. Journal of the American Chemical Society

1985 , 707,2924-2931.

(136) Main, A. D.; Kingsbury, K. B.; Carter, J. D.; Schoch, T. K.; MeElwee-White, L. Unpublished results.

(137) Schoch, T. K.; Main, A. D.; Burton, R. D.; Lucia, L. A.; Robinson, E. A.;

Schanze, K. S.; McElwee-White, L. Inorganic Chemistry 1996 , 35, 7769-7775.

(138) Buehler, R. E. In The Chemistry of the Carbon-Halogen Bond', Patai, S., Ed.; John Wiley: London, 1973; pp 795-864.

(139) Howard, J. A. K.; Jeffery, J. C.; Li, S.; Stone, F. G. A. Journal of the Chemical

Society-Dalton Transactions 1992 , 627-634.

(140) Kreissl, F. R.; Sieber, W. J.; Wolfgruber, M. Journal of Organometallic

Chemistry 1984 , 270, C45-C47.

Kreissl, F. R.; Keller, (141) H.; Schutt, W. Journal of Organometallic Chemistry 1992 , 441, 75-80.

(142) Mirafzal, G. A.; Bauld, N. L. Organometallics 1991 , 10, 2506-2508.

(143) Gassman, P. G.; Singleton, D. A. Journal ofthe American Chemical Society 1984, 106, 7993-7994.

(144) Grubbs, R. H. In Comprehensive Organometallic Chemistry, Wilkinson, G., Stone, F. G. A., Eds.; Pergamon Press: Oxford, 1982; Vol. 8; pp 499.

(145)(148) Wallace, K. C.; Liu, A. H.; Davis, W. M.; Schrock, R. R. Organometallics 1989 , 8, 644-654.

(146) Ginsburg, E. J.; Gorman, C. B.; Marder, S. R.; Grubbs, R. H. Journal of the

American Chemical Society 1989 , 111, 7621-7622.

(147) Brill, T. B.; Landon, S. J. Chemical Reviews 1984 , 84, 577-585.

Connelly, N. G.; Geiger, W. E. Chemical Reviews 1996 , 96, 877-910. 117

(149) Torraca, K. E.; Abboud, K. A.; McElwee-White, L. Organometallics 1998, 17, 4413-4416.

(150) Lam, P. Y. S.; Jadhav, P. K.; Eyermann, C. J.; Hodge, C. N.; Ru, Y.; Bacheler, L.

T.; Meek, J. L.; Otto, M. J.; Rayner, M. M.; Wong, Y. N.; Chang, C. H.; Weber, P. C.; Jackson, D. A.; Sharpe, T. R.; Erickson-Viitanen, S. Science 1994, 263, 380-384.

(151) Lam, P. Y. S.; Ru, Y.; Jadhav, P. K.; Aldrich, P. E.; DeLucca, G. V.; Eyermann,

C. J.; Chang, C. H.; Emmett, G.; Holler, E. R.; Daneker, W. F.; Li, L. Z.; Confalone, P. N.; McHugh, R. J.; Han, Q.; Li, R. H.; Markwalder, J. A.; Seitz, S. P.; Sharpe, T. R.; Bacheler, L. T.; Rayner, M. M.; Klabe, R. M.; Shum, L. Y.; Winslow, D. L.; Komhauser, D. M.; Jackson, D. A.; EricksonViitanen, S.; Hodge, C. N. Journal ofMedicinal Chemistry 1996, 39, 3514-3525.

(152) Dragovich, P. S.; Barker, J. E.; French, J.; Imbacuan, M.; Kalish, V. J.; Kissinger, C. R.; Knighton, D. R.; Lewis, C. T.; Moomaw, E. W.; Parge, H. E.; Pelletier, L.

A. K.; Prins, T. J.; Showalter, R. E.; Tatlock, J. H.; Tucker, K. D.; Villafranca, J. E. Journal ofMedicinal Chemistry 1996, 39, 1872-1884.

(153) Semple, G.; Ryder, H.; Rooker, D. P.; Batt, A. R.; Kendrick, D. A.; Szelke, M.; Ohta, M.; Satoh, M.; Nishida, A.; Akuzawa, S.; Miyata, K. Journal of Medicinal Chemistry 1997, 40, 331-341.

(154) Castro, J. L.; Ball, R. G.; Broughton, H. B.; Russell, M. G. N.; Rathbone, D.; Watt, A. P.; Baker, R.; Chapman, K. L.; Fletcher, A. E.; Patel, S.; Smith, A. J.; Marshall, G. R.; Ryecroft, W.; Matassa, V. G. Journal of Medicinal Chemistry 1996, 39, 842-849.

(155) vonGeldem, T. W.; Kester, J. A.; Bal, R.; WuWong, J. R.; Chiou, W.; Dixon, D. 968-981. B. ; Opgenorth, T. J. Journal ofMedicinal Chemistry 1996, 39,

(156) Vishnyakova, T. P.; Golubeva, I. A.; Glebova, E. V. Russian Chemical Reviews (English Translation) 1985, 54, 249-261.

(157) Hegarty, A. F.; Drennan, L. J. In Comprehensive Organic Functional Group Transformations', Katritzky, A. R., Meth-Cohn, O., Rees, C. W., Eds.; Pergamon:

Oxford, 1995; Vol. 6; pp 499-526.

(158) Guichard, G.; Semetey, V.; Didierjean, C.; Aubry, A.; Briand, J. P.; Rodriguez, M. Journal of Organic Chemistry 1999, 64, 8702-8705.

(159) Trost, B. M. Angewandte Chemie-International Edition in English 1995, 34, 259- 281. ,

118

(160) Wender, L; Pino, P. Organic Syntheses via Metal Carbonyls', Wiley Interscience: New York, 1968.

(161) Mulla, S. A. R.; Rode, C. V.; Kelkar, A. A.; Gupte, S. P. Journal of Molecular

Catalysis a-Chemical 1997 , 122, 103-109.

(162) Dahlen, G. M.; Sen, A. Macromolecules 1993 , 26, 1784-1786.

(163) Gupte, S. P.; Chaudhari, R. V. Journal of Catalysis 1988 , 114, 246-258.

(164) Franz, R. A.; Applegath, F.; Morriss, F. V.; Baiocchi, F. Journal of Organic

Chemistry 1961 , 26, 3306-3308.

(165) Franz, R. A.; Applegath, F.; Morriss, F. V.; Baiocchi, F.; Bolze, C. Journal of

Organic Chemistry 1961 , 26, 3309-3312.

(166) Sonoda, N.; Yasuhara, T.; Kondo, K.; Ikeda, T.; Tsutsumi, S. Journal of the

Americal Chemical Society 1971 , 93, 6344.

(167) Yoshida, T.; Kambe, N.; Murai, S.; Sonoda, N. Bulletin of the Chemical Society

ofJapan 1987 , 60, 1793-1799.

(168) Yoshida, T.; Kambe, N.; Murai, S.; Sonoda, N. Tetrahedron Letters 1986 , 27, 3037-3040.

(169) McCusker, J. E.; Logan, J.; McElwee-White, L. Organometallics 1998 , 17, 4037- 4041.

(170) McCusker, J. E.; Qian, F.; McElwee-White, L. Journal ofMolecular Catalysis A- Chemical in press.

(171) McCusker, J. E.; Grasso, C. A.; Main, A. D.; McElwee-White, L. Organic Letters

1999 , 1, 961-964.

(172) Khan, M. M. T.; Halligudi, S. B.; Shukla, S.; Abdi, S. H. R. Journal ofMolecular

Catalysis 1989 , 51, 129-135.

(173) Wilkinson, G.; Stone, F. A.; Abel, E. W. Comprehensive Organometallic Chemistry, Pergamon Press: New York, 1982; Vol. 8.

(174) Buffin, B. P.; Poss, M. J.; Arif, A. M.; Richmond, T. G. Inorganic Chemistry

1993 , 32, 3805-3806.

(175) Poss, M. J.; Arif, A. M.; Richmond, T. G. Organometallics 1988, 7, 1669-1670.

(176) De Lucca, G. V. Journal ofOrganic Chemistry 1998, 63, 4755-4766. 119

(177) Confalone, P. N.; Waltermire, R. E. In Process Chemistry in the Pharmaceutical Industry, Gadamasetti, K. G., Ed.; Marcel Dekker: New York, 1999; pp 201-219.

(178) Nugiel, D. A.; Jacobs, K.; Worley, T.; Patel, M.; Kaltenbach, R. F.; Meyer, D. T.; Jadhav, P. K.; DeLucca, G. V.; Smyser, T. E.; Klabe, R. M.; Bacheler, L. T.; Rayner, M. M.; Seitz, S. P. Journal ofMedicinal Chemistry 1996, 39, 2156-2169.

(179) Harriman, A. Journal of the Chemical Society-Chemical Communications 1977, 777-778.

(180) Tro, N. J.; Haynes, D. R.; Nishimura, A. M.; George, S. M. Journal of Chemical Physics 1989, 91, 5778-5785.

(181) Haynes, D. R.; Helwig, K. R.; Tro, N. J.; George, S. M. Journal of Chemical Physics 1990, 93, 2836-2847.

1990. (182) De Roeck, R.; Boens, N.; Dockx, J. ; 1.0 ed.; K. U. Leuven: Belgium,

(183) Wang, Y. S.; Schanze, K. S. Chemical Physics 1993, 176, 305-319.

(184) Atanassova, I. A.; Petrov, J. S.; Mollov, N. M. Synthetic Communications 1989, 19, 147-153.

(185) Bigi, F.; Frullanti, B.; Maggi, R.; Sartori, G.; Zambonin, E. Journal of Organic Chemistry 1999, 64, 1004-1006.

(186) Leung, M. K.; Lai, J. L.; Lau, K. H.; Yu, H. H.; Hsiao, H. J. Journal of Organic Chemistry 1996, 61, 4175-4179. BIOGRAPHICAL SKETCH

Andrea Denise Main was bom on September 24, 1971, and grew up in Winter

Garden, Florida. She was home-schooled by her parents from 8th to 12th grades. Upon

graduation, she entered the University of Central Florida in Orlando, Florida, in 1989.

She spent her spare time working for Dr. John Gupton, her organic research advisor. In

1993, after completing her Bachelor of Science degree in chemistry, she traveled to the

Netherlands and spent the summer working for a pharmaceutical company. In the fall of

1993, she entered graduate school at the University of Rochester. Deciding it was extremely cold and weathering Rochester's worst winter in ten years, she transferred to

the University of Florida in Gainesville, Florida. She decided to pursue the study of organometallic chemistry under the direction of her advisor. Dr. Lisa McElwee-White.

She received her Ph.D. in organic/organometallic chemistry in the summer of 2000. She

has obtained a postdoctoral research position at Pacific Northwest National Laboratories

in Richland, Washington where she will explore "green" chemistry under the direction of

Dr. John Linehan.

120 I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy.

Lisa McElwee-White, Chairman Professor of Chemistry

I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy.

Stewart Associate Professor of Chemistry

J. Eric Enholm Associate Professor of Chemistry

1 certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy.

James Boncella Associate Professor of Chemistry

Michael Kau:Cau^an Professor or Materi! cience and Engineering This dissertation was submitted to the Graduate Faculty of the Department of Chemistry in the College of Liberal Arts and Sciences and to the Graduate School and was accepted as partial fulfillment of the requirements for the degree of Doctor of Philosophy.

August 2000 Dean, Graduate School