211 al-0, 1 - digital library/67531/metadc332212/m2/1/high... · 211 m9/j al-0, 3/ /1 reduction...

194
211 M9/J Al-0, 3/ REDUCTION PATHWAYS IN CYCLOPENTADIENYL RHENIUM DICARBONYL DIBROMIDE DERIVATIVES AND INDENYL RHENIUM TRICARBONYL: SYNTHESIS, STRUCTURE, AND REACTIVITY OF ANIONIC CYCLOPENTADIENYL RHENIUM COMPLEXES. RING ATTACK VS. METAL-HALOGEN EXCHANGE. DISSERTATION Presented to the Graduate Council of the University of North Texas in Partial Fulfillment of the Requirements For the Degree of DOCTOR OF PHILOSOPHY By Sang Woo Lee, B.S., M.S. Denton, Texas December, 1989

Upload: truongdat

Post on 04-May-2018

216 views

Category:

Documents


1 download

TRANSCRIPT

211 M9/J

Al-0, 3/ /1

REDUCTION PATHWAYS IN CYCLOPENTADIENYL RHENIUM DICARBONYL

DIBROMIDE DERIVATIVES AND INDENYL RHENIUM TRICARBONYL:

SYNTHESIS, STRUCTURE, AND REACTIVITY OF ANIONIC

CYCLOPENTADIENYL RHENIUM COMPLEXES.

RING ATTACK VS. METAL-HALOGEN EXCHANGE.

DISSERTATION

Presented to the Graduate Council of the

University of North Texas in Partial

Fulfillment of the Requirements

For the Degree of

DOCTOR OF PHILOSOPHY

By

Sang Woo Lee, B.S., M.S.

Denton, Texas

December, 1989

Lee, Sang Woo, Reduction Pathways in Cvclopentadienvl

Rhenium Dicarbonvl Dibromide Derivatives and Indenvl Rhenium

Tricarbonvl; Synthesis. Structure, and Reactivity of Anionic

Cvclopentadienvl Rhenium Complexes. Ring Attack vs. Metal-

Haloaen Exchange. Doctor of Philosophy (Chemistry),

December, 1989, 178 pp., 19 tables, 30 figures, 74 titles.

The reactions of diagonal and lateral Cp'Re(CO)2Br2

(where Cp' = n5-C5H5, rj5-C5Me5) and (r/

5-CgH7)Re(CO)3 with

reducing agents have been examined. Hydride reduction at

-78 °C is observed to occur at the Cp ring in both

CpRe(CO)2Br2 isomers, affording a thermally unstable

4 -[(rj -C5Hg)Re(CO)2Br2] complex. The product of hydride ring

attack has been characterized by low-temperature IR and *H

NMR measurements in addition to 13C NOE and heteronuclear 2D

NMR measurements. Reaction of lateral CpRe(CO)2Br2 with

either MeLi or PhLi affords both Cp-ring attack and metal-

halogen exchange, [CpRe(CO)2Br]" (1) while t-BuLi reacts

exclusively via metal-halogen exchange. diag-CpRe(CO)2Br2

reacts with the above lithium reagents to yield the same

metal-halogen exchange anion. Analogous reactions using

diag- and lat-Cp*Re(CO) 2Br2 (where Cp* = r^-CgMe^) afford

only the corresponding rhenium metal-halogen exchange anion,

[Cp*Re(CO)2Br] (2). The molecular structures of

l-[Li/15-Crown-5] and 2-PPP were established by X-ray

crystallography. l-[Li/15-Crown-5] crystallizes in the

monoclinic space group P2j with a = 10.860(4) A, b -

13.116(5) A, C = 7.417(3) A, 0 = 105.26(3)°, V = 1018.7(3)

A , and Z = 2. 2-PPP crystallizes in the orthorhombic space

group Pbca with a = 20.646(5) A, b = 17.690(5) A, c -

17.553(3) A, and z = 8.

Solution FT-IR studies of 2 in THF reveal the presence

of only solvent-separated io

Li+, K+, or PPP+ from -70 °C

room temperature displays a

>n pairs when the gegencation is

to room temperature. 2-Na at

39:61 mixture of carbonyl

oxygen-sodium and solvent-separated ion pairs, respectively.

These ion pairs reveals a reversible temperature-dependent

equilibrium. The equilibrium constant has been determined

by IR band shape analysis over the temperature range -70 °C

to room temperature and values of AH and AS are reported.

The reaction of the ring-attacked complex,

diag-[(rj4-C5H6)Re(CO)2Br2]" with PPh3, P(OPh)3, or Me3CNC

leads to the formation of the CpRe(CO)2L. Treatment of

[Cp'Re(CO)2Br]" with methyltriflate, TFA, and magic ethyl

yields the corresponding diag-Cp'Re(CO)2Br(R) (R = CH3, H,

C2H5) complexes based on in situ IR analysis. All of these

functionalized complexes decomposed in solution over a

period of days to give Cp'Re(CO)3 as the only isolable

product (20-30 %).

The reaction of the [Cp,Re(C0)2Br]" with Bu3SnH at

60 °C leads to the formation of diag-Cp'Re(CO)2(SnBuj)2,

which was also synthesized independently by the

deprotonation of diag-Cp'Re(CO)2H2 with EtjN in the presence

of BugSnBr at room temperature.

The reaction of Cp'Re(CO) 2Br2 with Bu-jSnH at room

temperature was discovered to afford the dihydride in

excellent yield and, thus represents an improved synthetic

route for the synthesis of diag-Cp'Re(CO)*>H2.

The hydride reduction of (rj5-CgH7)Re(CO)3 at room

temperature leads to the immediate formation of

5

l(W -CgH7)Re(CO)2H] complex, which has been characterized

by IR analysis and *H and 13C NMR spectroscopy.

ACKNOWLEDGMENTS

The author wishes to express his sincere gratitude to

Professor Michael G. Richmond for his kind guidance,

encouragement, and assistance in directing my research, and

for serving as the chairman of my advisory committee. I

appreciate Dr. M. Schwartz and S. P. Wang for helpful

discussions and the use of their band area program, and

Messrs. Don Ellington and George Delong for NMR assistance.

Drs. K. E. Daugherty, R. D. Thomas, and D. A. Kunz are

thanked for serving on my advisory commitee.

I am especially grateful to my parents and my parents-

in-law for their dedication and support throughout the

years.

X am very grateful to my wife, Oc—Kee, sons, Leonard

and Joseph for their patience, never-ending encouragement,

and special love during the past years and throughout this

work.

Finally, I thank the University of North Texas for

support in the form of a teaching assistantship and the

Robert A. Welch Foundation of Houston, Texas for partial

financial support.

i n

TABLE OF CONTENTS

page

ACKNOWLEDGMENTS ii;L

LIST OF TABLES i x

LIST OF FIGURES # # . x i

CHAPTER

I. INTRODUCTION . # ±

II. EXPERIMENTAL

A. General Procedures 20

1. Techniques .. 20

2. Instrumentation 20

B. Materials 21

1. Solvents 21

2. Reagents .. . . 21

C. Methods 22

1. Attached Proton Test (APT) 22

2. Heteronuclear Proton-Carbon Chemical

Shift Correction (HETCOR) 23

3. Crystallographic Analysis 23

4. Band-Shape Analysis 24

D. Preparation of Compounds 25

1. Preparation of the Sealed NMR Tube Reactions 25

IV

page

2. General Procedure for the Preparations of

Aryl Rhenium Tricarbonyl Derivatives,

Cp*Re(CO)3 (Cp' « Cp, MeCp, Cp*, CgH7) .. 26

a. CpRe(CO)3 26

b. MeCpRe(CO)3 27

c. Cp*Re(CO)3 27

d. CgH7Re(CO)3 28

3. General Procedure for the Preparations

of Cyclopentadienyl Rhenium Dicarbonyl

Dibromide Derivatives, Cp'Re(CO)2Br2

(Cp» = Cp, MeCp, Cp*) 29

a. CpRe(CO) 2Br2 29

b. MeCpRe(CO)2Br2 3 0

c. Cp*Re(CO)2Br2 31

d. (d5-Cp)Re(CO)2Br2 3 2

4. diag-[(r?4-C5H6)Re(CO)2Br2][Li] 32

5. [CpRe(CO)£Br][Li] 33

6. [Cp*Re(CO)2Br][Li] 33

7. [CpRe(CO)2Br][M]

(M = Li (15-Crovm-5), PPP) 34

8. [MeCpRe(CO)2Br] [PPP] 34

9. [Cp*Re(CO)2Br] [PPP] 35

10. Reaction of Cp»Re(CO)2Br2 with One

Electron Reducing Agents (Cp1 = Cp, Cp*) .. 36

page

11. Reaction of Cp'Re(CO) 2Br2 with Grignard Reagents (Cp* = Cp, Cp*) 37

• CgHgCpRe (CO) 2 3-7

13. General Procedure for the Preparations

of Cyclopentadienyl Rhenium Dicarbonyl

Alkyl Bromide Derivatives,

CpRe(CO)2(R)Br (R « H, CH3, CgHg) 3 8

a. diag-CpRe(CO)2(H)Br 39

b. diag-CpRe (CO) 2 (CH3) Br 39

c. diag-CpRe (CO) 2 (C2H5) Br 39

d. diag-Cp*Re(CO)2(H)Br 4 0

e. diag-Cp*Re(CO)2(CH3)Br 4 0

f. diag-Cp*Re(CO)2(C2H5)Br 4 i

14. [C9H7Re(CO)2H] [Li] 4 1

15. diag-Cp'Re(CO)2H2 (Cp« = Cp, Cp*) 4 2

16. Cp'Re(CO)2(SnBu3)2 (Cp' = Cp, Cp*) 42

17. CpRe(CO)2L (L = PPh3, P(OPh)3,

(CH3)3CNC) 4 3

18. BrRe (CO) 5 4 4

19. (d5-Cp)Tl 4 4

20. (dg-Cp) Re (CO) 3 4 5

E. Thermodynamic Measurements 46

1. The Ion Pairing Eguilibrium in 2-Na, SSIP vs. CIP 4 6

VI

page

F. Kinetic Studies 47

1. Reaction of diag-CpRe(CO)2Br2 with

2. Isomerization of lat-CpRe(CO)2Br2 to diag-CpRe(CO)2Br2 in the

Presence of EtjSiH 4 7

III. RESULTS AND DISCUSSION

A. Aryl Rhenium Tricarbonyl Derivatives,

Cp'Re(CO)3 (Cp» = Cp, MeCp, Cp*, CgH7) 49

B. Cyclopentadienyl Rhenium Dicarbonyl Dibromide Derivatives, Cp'RefCOUBr? (Cp» = Cp, MeCp, Cp*) 5 5

C. Bromorheniumpentacarbonyl, BrRe(CO)g 62

D. Deuterated Cyclopentadienyl Rhenium Tricarbonyl, (d5-Cp)Re(CO)3 6 3

E. Deuterated Cyclopentadienyl Rhenium

Dicarbonyl Dibromide, (d5-Cp)Re(CO)2Br2 64

F. Hydride Reduction of diag-CpRe(CO)2Br2. Synthesis and Characterization of the Thermally Unstable Ring-Attacked Product, diag-[(n4-C5H6)Re(CO)2Br2]" 65

G. Reduction of Diagonal and Lateral CpRe(CO)2Br2: Ring Attack vs. Metal-Halogen Exchange 74

X-Ray Crystallographic Structure of [CpRe(CO)2Br][Li/l5-Crown-5] 8 3

I. Reactivity and Stability Studies of Anionic Cyclopentadienyl Complexes 94

Vll

page

J. Reduction of Diagonal and Lateral Cp*Re(CO)2Br2 with RLi, RMgX, and Trialkylborohydrides 113

K. X-Ray Crystallographic Structure of [Cp*Re(CO)2Br][PPP] 1 2 2

L. Reactivity and Stability Studies of Anionic Pentamethylcyclopentadienyl Rhenium Complex, [Cp*Re(CO)2Br][Li] 137

M. Reduction Studies Using One-Electron

Reducing Agents About Cp'Re(CO)2Br2 (Cp1 = Cp, Cp*) 138

N. Thermodynamic Study of the Ion Pairing

Equilibrium for [Cp*Re(CO)2Br][Na] 142

0. Reaction of Cp'Re(CO)2Br2 with Bu3SnH

and Et3SiH (Cp' = Cp, Cp*) 148

P. Hydride Reduction of Indenyl Rhenium Tricarbonyl: Synthesis and Character-ization of [(n -CgH7)Re(CO)2H]" 162

REFERENCES 167

Vlll

LIST OF TABLES

T a b l e Page

1. XR and Proton NMR Data for Aryl Rhenium

Tricarbonyl Derivatives, Cp»Re(CO)3 50

13 2. c NMR Chemical Shifts for Aryl Rhenium

Tricarbonyl Derivatives, Cp'Re(CO)3 51

3. IR and Proton NMR Data for Cyclopentadienyl Rhenium

Dicarbonyl Dibromide Derivatives, Cp'Re(CO)2Br2 .. 57

13

4. C NMR Chemical Shifts for Cyclopentadienyl

Rhenium Dicarbonyl Dibromide Derivatives,

Cp 'Re (CO) 2Br2 58

5. Anionic Cyclopentadienyl Rhenium Complexes with Reducing Agents: Ring Attack (R.A) vs. Metal-Halogen Exchange (M) 78

6. X-Ray Crystallographic and Data Processing

Parameters for [CpRe(CO)2Br][Li/15-Crown-5] ... 84

7. Table of Positional Parameters and Their Estimated Standard Deviations for

[CpRe(CO)2Br] | Li/15-Crown-5] 85

8. Bond Distances in Angstroms for

[CpRe(CO)2Br][Li/15-Crown-5] 8 9

9. Bond Angles in Degrees for

[CpRe(CO)2Br][Li/l5-Crown-5] 9 0

10. X—Ray Crystallographic and Data Processing

Parameters for [Li/15-Crown-5][Re04] 99

ix

13.

14.

Table Page

11. Table of Positional Parameters and Their Estimated Standard Deviations for [Li/15-Crown-5] [Re04] . 1 0 Q

12. Bond Distances in Angstroms for

[Li/15-Crown-5][Re04] 1 Q 3

Bond Angles in Degrees for [Li/15-Crown-5][Re04]. 104

X-Ray Crystallographic and Data Processing Parameters for [Cp*Re(CO)2Br] [Ph4PJ 1 2 3

15. Positional Parameters for Non-Hydrogen Atoms

for [Cp*Re(CO)2Br][Ph4P] with Estimated

Standard Deviations in Parenthesis 1 24

16. Bond Distances in Angstroms for

[Cp*Re(CO)2Br][Ph4P] 1 2 9

17. Bond Angles in Degrees for [Cp*Re(CO)2Br][Ph4P] . 131

18. Equilibrium Parameters for the Conversion of 2-Na from CIP into SSIPa

• • • 145

19. Selected Bond Dissociation Energies 160

LIST OF FIGURES

Figure P a g e

1. Methane and Carbon Dioxide Redox Coupling Scheme for the Production of Acetic Acid. Pathway (A): methane oxidative addition; pathway (B): carbon dioxide insertion; pathway (C): reductive elimination and repeat catalytic cycle 10

2. Infrared spectra of the carbonyl region for (a)

CgH/Re^OJs anc* (k) CPRe(CO)3. Both spectra

were recorded at 25 °C in cyclohexane. 53

3. *H NMR spectrum of CgHyRefCO)^ at 25 °C in

CDC13 54

4. Infrared spectra of the carbonyl region for (a)

diag-CpRe(CO)2Br2 and (b) lat-CpRe(CO)2Br2.

Both spectra were recorded at 25 °C in CH2C12. ... 61

5. Infrared Spectra of the carbonyl region

(a) [r?4-C5H6)Re(CO)2Br2] [Li] and (b)

C (-*7 -C5Hg)Re(CO)2Br2] [Li] with 5 equiv. of 15-Crown-5. Both spectra were recorded at

-70 °C in THF 66

6. NMR spectra of (a) [ (J74-C5Hg)Re(CO) 2Br2]"

and (b) [ (rj4-C5H5D)Re(CO)2Br2]~ at -70 °c

in d8-THF 6 8

7. (a) *^C{*H} NMR spectrum and (b) *3c NMR spectrum

using spin-echo J-modulation spectroscopy (APT)

of diag-[ (rj4-C5H6)Re(CO) 2Br2]". All spectra

were recorded at -70 °C in dg-THF 69

X I

Figure P a g e

8. Heteronuclear chemical shift correlation

spectrum of diag-[ (t?4-C5H6)Re(CO) gBr-]' at

-70 °C in dg-THF 7 1

9. Long-range heteronuclear chemical shift

correlation of diag-[ (r?4-C5H6)Re(CO)2Br2]" at

-70 °C in dg-THF. . 7 3

10. Infrared spectrum of the carbonyl region for

[CpRe(CO)2Br]" at -70 °C in THF 75

11. Infrared spectra of the carbonyl region for

(a) lat-[(T74-C5H6)Re(CO)2Br2]"

(b) lat-[ (n*-C5H5Me)Re(CO)2Br2]"

(c) lat-[ (r?4-C5H5Ph)Re(CO)2Br2]"

All spectra were recorded at -70 °C in THF 79

12. Infrared spectra of the carbonyl region for

(a) lat-[ (»?4-C5H6)Re(CO)2Br2] [Li]

(b) lat-[ (n4-CgHg)Re(CO) 2Br2] [Li] with 5 equiv.

of 15-Crown-5. Both spectra were recorded at •70 °C in THF. 80

13. Infrared spectra of the carbonyl region for (a)

[CpRe(C0)2Br][Li] and (b) [CpRe(CO)2Br][MgBr]

Both spectra were recorded at -70 °C in THF. ... 82

14. Perspective view (ORTEP plot) of

[CpRe(CO)2Br] [Li/15-Crown-5] showing the atom labeling (hydrogen atoms are omitted for clarity) . 8 7

15. Top view (ORTEP plot) of [CpRe(CO)2Br]"

showing the Cp ring-rhenium bond lengths 92

16. Infrared spectrum of the carbonyl region for

diag-CpRe(CO)2(SnBu3)2 at 25 °C in THF 97

Xll

Figure Page

17. Perspective view (ORTEP plot) of

[Li/l5-Crown-5][ReO^] showing the atom labeling

(hydrogen atoms are omitted for clarity). .... 102

18. Infrared spectrum of the carbonyl region for [Cp*Re(CO)2Br]- at -70 °C in THF. 1 1 5

13 1 19. C{ H) NMR spectrum of [Cp*Re(CO)2Br] [PPP]

at 25 °C in CD2C12 1 1 7

20. Infrared spectra of the carbonyl region for (a)

[Cp*Re(C0)2Br][Li] and (b) [Cp*Re(CO)2Br][MgBr].

Both spectra were recorded at -70 °c in THF. .. u s

21. Perspective view (ORTEP plot) of [Cp*Re(CO)2Br]" showing the atom labeling (hydrogen atoms are omitted for clarity) 1 2 7

22. Perspective view (ORTEP plot) of PPP+ gegenion showing the atom labeling (hydrogen atoms are omitted for clarity) 1 2 8

23. Top view of (a) lat-Cp*Re(CO)2I2 and (b)

[Cp*Re(CO)2Br]" showing the Cp* ring-rhenium bond lengths (A). Distances for the former complex are taken from ref. 32. 134

24. Infrared spectra of the carbonyl region for

[Cp*Re(CO)2Br][Na] in THF as a function of

temperature: (1) -70 °C, (2) -30 °c, and (3)

25 C. The inset displays the asymmetric

carbonyl region (1825-1725 cm"1) for the experimental spectrum ( ) of 2-Na recorded

at -70 °C, resolved bands (....) for the contact and solvent-separated ions, and the theoretical curve ( ) for both ion pairs 1 4 4

x m

Figure Page

25. Plot of In Keg + 1 vs. 1/T for the

contract solvent-separated ion pair

equilibrium for [Cp*Re(CO)2Br] [Na] in THF 146

26. Infrared spectra of the carbonyl region for

(a) diag-CpRefCOJgfSnBu^

(b) lat-Cp*Re(CO)2(SnBu3)2

Both spectra were recorded at 25 °C in

cyclohexane 152

11 0 27. The Sn NMR spectrum of diag-CpRe(CO)2(SnBuj)^

in CDClj solution (0.1 M) at 25 °C 153

28. Absorbances changes for lat-CpRe(CO)2Br2 in

toluene at 40 °C in the presence of EtgSiH. .. 156

29. Absorbances changes for diag-CpRe(CO)2Br2

with EtjSiH (10-fold excess) at 40 °C

in toluene 158

30. *H NMR spectrum of [ (r}5-CgH7)Re(CO) 2H]" at 25 °C

in dg-THF 164

xiv

CHAPTER I

INTRODUCTION

The objective of these investigations, which were

carried out in partial fulfillment for a Ph.D degree in

chemistry and are reported in this dissertation, comprises

the synthesis, structure, and reactivity of anionic

cyclopentadienylrhenium compounds and the reduction pathways

in cyclopentadienylrhenium dicarbonyl dibromides. The

chemistry of organo-transition-metal compounds pertinent to

this work and the literature on hydrocarbon C-H bond

activation and C02 insertion into the M-H bond are reviewed

in this section.

One of the most interesting goals of homogeneous

organo-transition-metal chemistry is the possibility of

carrying out selective chemical transformations on, or

functionalizing, unreactive materials such as saturated

hydrocarbons. The homogeneous activation of C-H bonds by

metal complexes, especially at saturated carbon centers, has

long been recognized as a great interest and challenging

objective.*

Initial observations of oxidative addition to arene C-H

bonds in 1965 provided the impetus to look for complexes

that would activate the weaker sp3-hybridized alkane C-H

bonds.2 However, this goal was not achieved until nearly

two decades later and was accompanied by an intermediate

period in which much confusion occurred concerning the

thermodynamic feasibility of the C-H bond activation

3

reaction. in this account, mechanistic studies with a

series of homogeneous rhodium organometallic complexes are

summarized and have provided for the first time a

comparative evaluation of the relative equilibrium constants

and rates of reaction for both alkane and arene hydrocarbon

activation (Eq. 1).

R-H m

\ Since Chatt observed the first clear example of simple

oxidative addition of the C-H bond of naphthalene to a

ruthenium metal center,2 hydrocarbon activation has become

the subject of many transition-metal studies. From the

point of view of quantity and availability, alkanes (satura-

ted hydrocarbons) would also be very attractive feedstocks

for catalytic synthesis of organic molecules since they are

major constituents of natural gas, petroleum, and coal

3

liquefaction processes. However, they have not received

much use in synthetic processes and as feedstocks in

catalytic reactions.

Saturated hydrocarbons are among the most ubiquitous

and chemically stable of all organic materials due to the

strong C-H bond (~ 99 kcal/mole) and C-C bond (~ 83 kcal/

mole) energies, but they can be activated in special cases.

As a result of their relative inertness, saturated hydro-

carbons have a long history of activation by nonmetallic

reagents and methods. Hydrocarbon thermal reactions and

combustion have been studied, and there are well-known free

radical reactions (e.g., autoxidation, photochlorination)

which serve to functionalize these materials.^ More

recently, reagents such as ozone, superacids, fluorine, and

H2O2 have been used to activate hydrocarbons.® However,

these reactions often require large amounts of energy

(either light, heat, or chemical) and are usually very

unselective. Therefore, attention has been focused on the

discovery of soluble organo-transition-metal complexes

capable of inserting a metal center into the C-H bond of

alkanes (Eq. 1). During the past 15 years many examples of

intramolecular C-H oxidative addition [that is, the metal

and reacting C-H bond are located in the same molecule

(Eq. 2)] have been discovered.^ Some specific examples of

this process are depicted in Equations 3-5.

M - CHgR — — ^ H - M - CHR (2)

/ CPPh3)3l

rC1 ( P P| l 3, 2 ( c l ) I/^__^ (3)

•I CH2CHe3

CH2CHe3

CMe4 LgPt^ L (4)

-Bu) /H ( t

HC13 + (t-Bu)2P(CH2)5P(t-Bu)2 H ' — j — ( 5 ) ^fe(t-Bu)

CM = Ir, Rh)

The largest group involves so called ortho-metallation

processes, in which insertion takes place into the C-H bond

of an aromatic ring attached to an atom directly bound to

the metal (Eq. 3). Some cases are also known of insertion

into a C-H bond of an alkyl chain located in the same

molecule as the metal (Eq. 4 and 5). Obviously, the

proximity of the reacting C-H bond to the metal center is a

critical factor favoring such cyclometallation processes.

Many examples of homogeneous intermolecular activation

of saturated hydrocarbons by metal complexes are also

reported.

In the late 1960's and early 1970's Shilov and his

group reported reactions involving both alkane H/D exchange

and conversion to chlorides and acetates catalyzed by

soluble platinum salts (Eq. 6).*e>^

DOAc RD

Ptncl„ / ,B. KB — ^ (6>

Pt'»Cln RC1

However, somewhat elevated temperatures (100-120 °C) are

required for these reactions. Little else is known about

their mechanism including whether they are most likely

colloidal and involving metal (Pt) cations similar to the

superacid reactions.

In early 1980*s Crabtree's group® at Yale, and Felkin

and coworkers at Gif-Sur-Yvette in France^ reported two very

interesting and unique iridium- and rhenium-induced

dehydrogenation processes shown in Eq. 7 and Eq. 8.

Crabtree has provided convincing evidence that these

reactions are completely homogeneous and has proposed that

they are initiated by insertion of the metal center into an

alkane C-H bond, but it has so far been difficult to

6

[lrH 2s 2L 2]+

+ t-BuCH=CH 2 [CpIrHL 2]+ + t - B u C H 2 C H 3 (7)

(S = Acetone, L = PPh 3)

L2ReH7 + t-BuCH=CH 2 CpR«L 2H 2 + t-BuCHgCHj (8)

determine this conclusively since the reactions are quite

complicated. These reactions involve multiple hydrogen loss

in the saturated hydrocarbon substrate and require an added

alkene as a hydrogen acceptor.

In solution, a few relatively electron-rich complexes

have been demonstrated to undergo insertion into C-H bonds

activated by adjacent functional groups of organic compounds

having C-H bonds with low bond energy or high acidity (Eq. 9

and 10); however, the metal centers in these molecules

.H CH 3(CO)CH 3

(dmpe)^^ ^ (dmpe)2M^ o - ArH ^ C H 2 - C - C H 3 (9)

(N = Fe, Ru)

h v J ^ C H 2 s i M ® 3 C p 2 W H 2 + M e 4 S i — y C ° (10)

7

apparently react with C-H bonds in their own ligands more

rapidly than with saturated hydrocarbons.lc>10

Recently, a few examples of the reaction shown in Eq. 1

have been demonstrated between organo-transition-metal

complexes and completely saturated hydrocarbons (alkanes) in

homogeneous solution under photolytic conditions (Eq. 11).

11-13 In 1982 Bergman and Janowicz were the first to

[CpMLH2] o r

hv RH jr [CpHL] Cp(L)Ji^

H (11)

[CpML(CO)] —

(Cp» = Cp, Cp*? M = I r , Rh? L as CO, PMe3)

synthesize an iridium complex which successfully converts

alkanes into hydrido(alkyl)metal complexes in high yield at

room temperature. Evidence has been obtained that this C-H

insertion (oxidative addition) reaction proceeds through a

simple three-center transition state and does not involve

organic free radicals as intermediates. In accordance with

this, the intermediate [Cp*IrL] reacts most rapidly with C-H

bonds such as those at primary carbon centers, in small

organic rings, and in aromatic rings. Reductive elimination

of the hydrocarbon from the hydrido(alkyl)metal complexes

can be induced photochemically or by heat, regenerating the

reactive intermediate [Cp*ML], which is then capable of

attacking the C-H bond of other hydrocarbons. Oxidative

8

addition of the corresponding rhodium complexes [Cp*RhL] to

alkane C-H bonds has also been observed, although the

products formed in this case are much less stable, and

undergo facile reductive elimination. These recent

observations provide an incentive for reexamining the

factors which have been assumed to control the rate of

reaction of transition metal complexes with C-H bonds,

notably the need for electron-rich metals and the close

proximity of reacting centers.

On the other hand, further functionalization of these

hydrocarbon activated complexes, Cp*M(CO)(R) (H), as a route

to fine organic chemicals has not been very successful to

date. These complexes can potentially react with C02 to

afford the corresponding metallocarboxylic acid,

Cp*M(CO)(R)(CO2H), via CO2 insertion into the metal-hydride

bond. This is attractive since the reductive elimination of

the methyl and metalloacid groups can be envisioned to give

the commodity chemical acetic acid and regenerate the metal

catalyst in the case of R = methyl (coordinatively unsatur-

complex, [Cp*M(CO) ]). Such an insertion reaction

(i.e., M-H + CO2 * M—CO2H) is well documented in other

metal-hydride complexes,^ but no analogous Cp*M(CO)(CHg) (H)

insertion chemistry has been studied in depth. Further

reaction of this complex with additional methane and CO2

ensures a cyclic process for this overall methane oxidation

9

and CC>2 reduction scheme as shown in Figure 1.

This directed synthesis of acetic acids is of commer-

cial importance due to its use as a major industrial

chemical which is necessary for the oxidation of para-xylene

to terephthalic acid (nylon polymer precursor), production

of polyvinyl alcohol and polyvinyl acetate polymers, pharma-

ceuticals, pesticides, dyes, and acetic anhydride synthesis

which is used to prepare cellulose acetate and aspirin.^

Collectively, these uses of acetic acid represent a multi-

million dollar industry for the whole world. Although

acetic acid has been produced in relatively large quantities

by fermentation and sundry catalytic processes, it was not

until Monsanto discovered the low pressure carbonylation of

methanol using rhodium catalysts in 197116 that an

industrial process showed such high yields (> 90%) and

reaction rates in the enzymatic range.^ However, the

Monsanto process suffers from being a petroleum-based

17 18

process. * Therfore, an alternative process based on

abundant reagents remains desirable. The use of methane and

carbon dioxide as chemical feedstocks remains highly

attractive due to their inexpensive nature and vast

abundance.

Therefore, the main goal of my research was to

synthesize the analogous rhenium-hydride complexes [i.e.,

Cp*Re(C0)2(CHg) (H) ] which are isoelectronic and isolobal^

10

coordlnately unsaturated intermediate

Cp*M(CO)2

-CO hv

[Cp*M(CO)]

1 *ch3-h

CHj-H

[Cp*M(CO)]

acetic acid

Cp*

,.M,

oc'y

ch3

Cp"

"iM\ o i \ y o c " 7 C-OH

CH,

CHr-C-OH

Figure 1. Methane and Carbon Dioxide Redox Coupling Scheme for the Production of Acetic Acid. Pathway (A): methane oxidative addition? pathway (B): carbon dioxide insertion; pathway (C): reductive elimination and repeat catalytic cycle.

11

to the rhodium and iridium systems (Scheme 1) and to

investigate the CO^ insertion chemistry and catalytic

properties associated with these complexes. Unfortunately,

the results of the reaction of CpRe(CO)2Br2 with reducing

Scheme l

l - m -

M as R h , I r M S R e

d 8 - M L 4 ^ d 6 - M L 5

agents such as LiEt3BH and RLi (R = CH3, Ph, and t-Bu)

suggest that direct metathetical style reductions in

CpRefCOJgBrg as a route to rhenium derived hydrocarbon

activated templates are not possible unlike those observed

in the corresponding rhodium and iridium systems [e.g.,

CpMLBrg • CpMLH(R)].^ Therefore, reduction studies of

cyclopentadienylrhenium dicarbonyl dibromide derivatives

were conducted and the reactivity associated with the

resulting anionic cyclopentadienylrhenium complexes explored

and reported herein.

The cyclopentadienyl ring ligand has played an

important role in the development of organometallic

12

chemistry since the discovery of ferrocene in 1952. The

significance of the cyclopentadienyl ligand has stimulated

interest in investigating the chemistry of other closely

related ligands, such as rings in which one or more of the

hydrogen atoms are substituted by other groups.^

In 1958 the cyclopentadienylrhenium tricarbonyl, c

CpRe(CO)j (Cp = r? -C5H5), was first prepared from the

refluxing of rhenium decacarbonyl in dicyclopentadiene by

77

Green and Wilkinson. The same compound has also been

obtained from the reaction of Re(CO)gX (X = Br, CI) in THF

or benzene with CpNa or CpTl. The aryl rhenium

tricarbonyl derivatives (aryl » Cp, MeCp, Cp*, and CgHy24)

obtained from these procedures can be purified by column

chromatography over silica gel with petroleum ether or by

slow sublimation under vacuum while maintaining a small

temperature gradient. Aryl rhenium tricarbonyl derivatives

are white solids or a yellow solid in the case of the

indenyl complex (CgHy) and are air stable compounds. The structure of CpRe(CO)^ has been determined by single crystal

25

X-ray analyses. The cyclopentadienyl ring is almost

perfectly planar (largest deviation, 0.003 A). The small

variation of C-C bond lengths within the ring indicates some

tendency for delocalized bonding, which is quite expected

for molecules of this type which possess effectively

13

cylindrical symmetry.

Synthetic pathways to the deuterated cyclopentadienyl

rhenium tricarbonyl, (d5-Cp)Re(CO)3, have recently been

described in detail.26 The reaction of (d5-Cp)Tl and

BrRe(CO)g in benzene proceeds readily at 60-70 °C overnight

to give excellent yields of (d5-Cp)Re(CO)3.

Although at first glance the permethylation of

cyclopentadienyl rings might appear to be a relatively minor

alteration, it has proven to be a useful perturbation for

mechanistic: studies and has uncovered a significantly

different chemistry in many systems.27 Some of the changes

in behavior of pentamethylcyclopentadienyl complexes may be

attributed to the steric protection provided by the five

ring methyl groups. Another very important factor is the

apparent increased electron density and donor strength of

the permethy1ated ring relative to the parent Cp ring.

Therefore, methylcyclopentadienyl, pentamethylcyclopenta-

dienyl, and mdenyl coordinated rhenium carbonyl derivatives

have been prepared and the chemistry associated with these

compounds investigated and reported herein.

Cyclopentadienyl metal derivatives of the type CpMX^

are formed by many d^ transition metals of the 4d and 5d

transition series including Nb(I), Ta(I), Mo(II), W(II), and

• Four—legged piano—stool complexes of the form

14

CpMX2^2 m a¥ ®xist as two nonequivalent stereoisomers that

are commonly referred to as cis and trans isomers. We have

adopted King's nomenclature to describe the isomeric

dibromides discussed in this paper. Here the descriptors

lateral and diagonal correspond to the cis and trans

stereoisomers, respectively.

In 1969 Nesmeyanov and his coworkers first prepared

CpRe(CO>2Br2 from the reaction of bromine and CpRe(CO) in

trifluoroacetic acid,2® and subsequently the product was

shown to consist of cis and trans isomers which could be

separated.29 The diiodide complex CpRe(CO)2I2 was not

reported until 19813®'31 and may also be synthesized in cis

and trans forms from the direct reaction of CpRe(CO)3 with

I2 in dimethyl sulfoxide. The dichloride CpRe(CO)2Cl2 was

only recently reported.32 Of these dihalides, the dibromide

has received much attention in terms of chemical investiga-

33 34 ti°n ' and as a precursor in the synthesis of other four-

legged piano-stool complexes based on the CpRe fragment.35"

42

However, synthetic pathways to the corresponding

pentamethylcyclopentadienylrhenium dicarbonyl dihalides,

CP*Re(CO)2X2 (where Cp* = ^"CgMeg), have only recently been

described in detail. While the diiodide Cp*Re(CO)2I2,

initially obtained from the reaction of Cp*2Re2(CO)j with

J2' w a s t h e first compound of the Cp*Re(CO)2x2 prepared,33

15

it was not until the work of Sutton et al. that reliable and

stereoselective synthesis of these dihalide compounds were

r e p o r t e d . 3 2 , 3 3 . 4 3 , 4 4 A s such^ the reactivity of these

dihalides remains to be explored and established.

In related reactivity studies using the isomeric

dibromides, Cp'Re(CO)2Br2 (CP' ~ CP*)/ w© have observed

that a direct metathetical replacement of the bromide

ligands with RLi or RMgX reagents (where R = Me, Ph, or

t-Bu) to yield the dialkyl (aryl) complexes Cp'Re(CO)2R2

does not occur. The reaction of CpRe(CO)2Br2 with Grignard

reagents has been reported in 1974 to give compounds of the

form CpRe(CO)2(R) (X) and CpRe(CO)2R2 (where R = alkyl or

34

aryl; X = halide) . However, no mention was made of the

relationship between the product dependence and initial

dibromide stereochemistry. Furthermore, while the reported

physical data dealing with such compounds as

CpRe(CO)2Br(Me), CpRe(CO)2I(Me), and CpRe(CO)2(Me)2 are not

in question, we do not believe that these represent products

of direct alkyl/bromide exchange as we observe metal-halogen

exchange and cyclopentadienyl ring attack prior to the

formation of the above complexes (vide infra). The relative

amounts of these products (metal/halogen exchange and ring

attack) are dependent on the nature of the reducing agent

and the stereochemistry of the initial dibromide. That is,

16

diag-CpRe(CO)2Br2 and Cp*Re(CO)2Br2 (either isomer) react

exclusively with RLi and RMgX reagents to afford

[Cp'Re(CO)2Br] as a result of metal—halogen exchange while

lat-CpRe(CO)2Br2 gives both [CpRe(CO)2Br]" and 4

[ (*? -CgHgR)Re(CO)2Br2] . This latter complex derives from

RLi (R - Me or Ph) attack on the cyclopentadienyl ligand.

Although indenyl rj5-complexes of transition metals are

analogous to r?5-cyclopentadienyl species, they exhibit some

specific properties. In an indenyl ligand, a benzene ring

fused with a 5-membered ring takes part in distribution of

jr-electron density.4®

The reactivity of the rj®-indenyl rhenium tricarbonyl

has been previously studied in exchange reactions of the

coordinated CO with other p-donor ligands.46 The reactions

of (f?5-C9H7)M(CO)3 (M • Mn, Re) with strong acids (A1C13*HC1

and FSOjH) have also been studied4^ and shown to involve a

metallotropic rj5 -• rj6 rearrangement in which the metal

carbonyl group migrates from the five- to the six-membered

ring of the aromatic ligand. The reverse 17® -*• rj®

rearrangement is initiated by bases. In this study we

investigated the reduction pathways of (r?5-CgH7)Re(CO) 3 with

LiEt3BH.

Diagonal cyclopentadienylrhenium dicarbonyl dihydride,

diag-CpRe(CO)2H2 was first observed by Graham and Hoyano

17

from the photolysis of CpRe(CO)3 in the presence of H2.41a

Bergman and Yang later synthesized it in good yield by the

reduction of CpRe(CO)2Br2 with zinc and acetic acid.37b The

former method is known to give low yields while the latter

method in our hands has proven inconvenient as the yields

are dependent upon the work-up conditions. However, the

reaction of Cp'Re(CO)2Br2 (either isomer, Cp' = Cp, Cp*)

with BujSnH (2 equiv.) at room temperature was discovered to

afford the dihydride in excellent yield and, thus,

represents a new and an improved synthetic route for the

synthesis of diag-Cp'Re(CO) 2H2-

48

Graham recently reported the synthesis and properties

of a series of rhenium compounds of the type CpRe(CO)2x2

(X • germanium and tin species).31 The bis(triphenyltin)

rhenium complex, diag-CpRe(CO)2(SnPh3)2, w a s synthesized in

very poor yield (7 %) from the photolysis of CpRe(CO)3 in

the presence of HSnPhj. However, the reaction of the metal-

halogen exchange product, [CpRe(CO)2Br]", with BUjSnH at

60 °C leads to the formation of the bis(tributyltin) rhenium

complex, diag-CpRe(CO)2(SnBu3)2, which was also

independently synthesized in excellent yield by the

deprotonation of diag-CpRe(CO) 2H2 with EtjN in the presence

of Bu^SnBr at room temperature.

Recently there has been considerable interest in the

18

solution structure of the alkali metal salts of various

transition metal carbonylate anions. In particular, the

chemistry associated with cyclopentadienyl carbonylate

anions is of both mechanistic and synthetic interest. These

compounds are among the most versatile of the transition

metal anions and have been widely used to prepare many novel

and significant organometallic complexes of the transition

49

metals. Extensive studies by Edgell and coworkers on the

sodium tetracarbonyl cobaltate were the first to illustrate

the importance of ion pairing in such systems.®® It was

shown that an equilibrium existed between a tight ion pair involving a sodium-carbonyl oxygen interaction and a solvent

separated ion pair. In 1974 Pribula and Brown reported

similar ion pairs for the sodium pentacarbonylmanganate

51

system. Other workers have been concerned with the

redistribution of electronic charge in the carbonylate

itself upon changing the electrostatic potential of the

countercations,49-53 or interested in ion-pair site specifi-

city in non-symmetric carbonylates.52 The ability of v(CO)

infrared spectroscopy to detect very small structure and

electronic changes as well as the availability of a

substantial arsenal of appropriate solid—state structural

studies of metal carbonylates has promoted the development

of a new subdiscipline in ion-pairing phenomena.

Darensbourg and coworkers have extended the account of

19

ion pairing by studying the propensity of [CpM1(CO)^]" (M' =

Cr, Mo, W) to form contact ion pairs as a function of the

countercation (Li+, Na+, K+, Me^N+) and solvent (EtgO,

Et20/HMPA, THF/HMPA, and CH3CN).53 Of the three possible

proposed interactions between the metal carbonylate and the

cation shown below, and they did not observe any indication

of cation penetration of the coordination sphere yielding an

ion-paired structure (III).

° nCr V

M ^ C 0 0 C V M \ C 0 M' o c - > M ^ c o OC OC •—M OC

(i) (II) (in)

In order to establish the nature of ion pairs in

solution for [Cp*Re(C0)2Br][Na] complexes, the effect of

temperature on the equilibrium between solvent-separated ion

pairs and carbonyl oxygen contact ion pairs over the

temperature range -70 °C to room temperature was examined.

This system reveals a reversible temperature-dependent

equilibrium involving both anionic species. The equilibrium

constant for these ion pairs has been determined by IR band

shape analysis and values of AH and AS have been determined.

20

CHAPTER II

EXPERIMENTAL

A. General Procedures

1. Techniques

All manipulations were conducted under an inert

atmosphere of argon, either on a high vacuum line with

modified inert Schlenk techniques55 or in a nitrogen filled

Vacuum Atmosphere DL Series inert-atmosphere Dri-box.

2. Instrumentation

Infrared spectra were recorded on a Nicolet 20SXB FT-IR

Spectrometer in 0.1 mm NaCl cells. Low-temperature IR

spectra were recorded on the same spectrometer with a Specac

Model P/N 21.000 variable-temperature cell equipped with

inner and outer CaF£ windows. Dry ice/acetone was used as

coolant, and the reported cell temperatures, taken to be

accurate to ± 1 °c, were determined with a copper-constantan

thermocouple. *H (300 MHz) and 13C (75 MHz) NMR spectra

were obtained using a Varian VXR-300 MHz NMR Spectrometer

21

while 90 MHz *H NMR spectra were recorded on a JEOL FX90Q

NMR Spectrometer. 2D (13.1 MHz), (22.5 MHz), *19Sn (33.7

MHz) NMR spectra were recorded on the JEOL FX90Q NMR

Spectrometer. The C and H analyses were performed by

Atlantic Microlab, Atlanta, 6A.

B. Materials

1. Solvents

Tetrahydrofuran, benzene, and toluene were distilled

from purple solutions of sodium benzophenone ketyl under

argon. Aliphatic hydrocarbon solvents (CgH^, C^Hj^),

CH2CI2, Et20, and HMPA were distilled from calcium hydride

under argon.

Deuterated solvents (dg-THF and dg-benzene) were vacuum

distilled from calcium hydride while CDgC^ and CDCI3 were

distilled from P2O5 under argon and stored in Schlenk

vessels.

2. Reagents

Rhenium decacarbonyl was used as received. Deuterated

cyclopentadienylrhenium tricarbonyl was prepared from known

26 cc procedure. Pyridinium hydrobromide perbromide , sodium

naphthalide (0.5 M in THF), cobaltocene (0.25 M in THF)^,

22 CO

and t-Butyl isocyanide30 were prepared from known literature

procedures. Triphenylphosphine (Alfa) was recrystallized

from absolute ethanol. Triphenylphosphite (Aldrich) was

distilled from calcium hydride under argon. Magic ethyl was

purchased from Alfa and stored under argon in a Schlenk

vessel while methyl triflate was prepared according to the . 5 5 50

published procedure. Bu3SnH was prepared from known

literature procedures. EtjSiH was purchased from Lancaster

Synthesis Ltd. and used as received.

CH3Li (1.4 M in EtgO), PhLi (2.0 M in CgHjg/EtgO;

70/30), t-BuLi (1.7 M in pentane), sec-BuLi (1.3 M in cyclo-

hexane), MeMgBr (1.5 H in toluene/THF? 75/25), t-BuMgCl (2.0

M in THF), LiEt3BH(D) (1.0 M in THF), NaEtjBH (1.0 M in

THF), and K-Selectride (1.0 M in THF) were all purchased

from Aldrich and used as received.

C. Methods

1. Attached Proton Test (APT)

This J-modulation spin-echo experiment utilized the

standard PD-90°-D1-180°-D2-1800-D3-ACQUISITION pulse

sequence that is equipped with the Varian VXR-300

Spectrometer. The broad-band proton decoupler was gated off

during the D1 delay period and then turned back on for the

23

rest of the pulse sequence. All spectra were obtained

employing quadrature phase detection and automatic base line

correction. Alternate FIDs were corrected with 180° phase

shifts to remove DC bias between the two receiver channels.

2. Heteronuclear Proton-Carbon Chemical Shift

Correlation (HETCOR)

The heteronuclear chemical shift correlation spectrum

was acquired by using the pulse sequence of Freeman and

Morris6®, modified to provide quadrature detection in the

second frequency domain as described by Bax and Morris61.

The data was acquired as a 128 (zero filled to 512) x 2K

matrix to yield a 512 x IK F ^ data matrix that was

processed by using 2D Gaussian Apodization prior to the

second Fourier transformation.

3. Crystallographic Analysis

Data were collected on an Enraf-Nonius CAD-4 diffrac-

tometer at 24 ± 2 °C using graphite-monochromated Mo

radiation. The data were corrected for Lorentz and polari-

zation effects. The structure was solved by MULTAN6* or the

Patterson method and successive cycles of differrence

Fourier syntheses were followed by least-squares refinement.

Data with intensities less than 3a(I) were rejected, and a

non-Poisson contribution weighting scheme with an intensity

2 4

factor P set at 0.04 was used in the final stages of

refinement. P is used in the calculation of a(I) to down

weight intense reflections in the least-squares refinement.

The function minimized was S - (| FQ| -| FC|)2, where

2 - 4 ( F 0 ) 2 / [ 2 ( F 0 ) 2 ] 2 ,

S ( F Q ) 2 = [ S 2 ( C + R2B) + <P(F0)2}2]/Lp2, and S is the scan

rate, C is the total integrated peak count, R is the ratio

of scan time to background counting time, and Lp is the

Lorentz-polarization factor. Suitable crystals were grown

from a solution of THF and n-heptane (vide infra) and were

mounted in a thin-walled glass capillary and sealed under

nitrogen. All non-hydrogen atoms were refined anisotropi-

cally and scattering factors were taken from ref. 63.

4. Band-Shape Analysis

Since both the asymmetric carbonyl stretching bands of

the sodium-contact and solvent-separated ion pairs in

solution of [Cp*Re(C0)2Br][Na] complex exhibit significant

overlap, the infrared band shapes of these CO bands were

calculated using a numerical procedure in order to determine

the ratio of their areas. Absorbances were digitized from

1825 cm ^ to 1725 cm * by 2 cm~^ intervals and entered into

files on the university VAX 11 /85 computer. Following

baseline correction, the spectra were fit by a model

consisting of Lorentzian band-shapes, each characterized by

25

a peak frequency (v), maximum intensity (I), and half width

[FWHH] (A). Since the instrument resolution (2 cm'*) is far

less than the observed bandwidths (20 cm"*), it was unneces-

sary to convolute the model spectrum with a resolution

(slit) function. The parameters were varied to minimize the

squared deviation between the experimental and calculated

intensities using a standard non-linear regression

64

procedure . Given that the area of a Lorentzian peak is

proportional to the product of the bandwidth and the maximum

intensity, the area ratio of the different ion pairs is

calculated easily as A2/AJ - (I2 x A2)/(IJ x Aj) .

D. Preparation of Compounds

1. Preparation of the Sealed NMR Tube Reactions

Starting material and an appropriate deuterated solvent

(0.7 mL for 5 mm NMR tube or 3.5 mL for 10 mL NMR tube) were

transferred into the NMR tube under argon. After the tube

was shaken well, the reagent was added at -78 °C. The tube

was then freeze-pump-thaw degassed three times prior to

flame sealing.

26

2. General Procedure for the Preparations of Aryl

Rhenium Tricarbonyl Derivatives, Cp'Re(CO)3

(Cp» = Cp, MeCp, Cp*, C9H7)

A 100 mL of Fisher-Porter tube was charged with cyclo-

pentadiene dimer or freshly distilled indene, Re2(CO)10, and

a magnetic stir bar. The Fisher-Porter tube was heated up

to 210 °C over several hours employing a heating mantel as

the heat source. At this time, the pressure gauge

registered an increased pressure of 100 psi. At this time,

the heat source was turned off and the Fisher-Porter tube

was allowed to cool to room temperature. The liberated CO

and H2 gas were released from the Fisher-Porter tube. This

process was repeated until no pressure increase was

observed. The white solid material was transferred to a

fritted glass funnel and washed three times with cold

hexane. Air was pulled through the product for several

minutes to remove the hexane. The colorless powder was

analytically pure and was used directly in subsequent

reactions.

a. CpRe(CO)j

Cyclopentadiene dimer (7.4 mL, 56.0 mmole), Re2(CO)10

(10.4 g, 16.0 mmole), and a magnetic stir bar were placed

into the Fisher-Porter tube. This reaction was carried out

27

as described in general procedure. The total released

pressure of CO and H2 gas was 310 psi. The yield of

CpRe(C0)3 was 9.4 g (87 %).

IR (Cyclohexane) : 2030 (m), 1939 (s) cm"*.

^-NMR (CDCI3) : -CgHg (5.2 s, 5H) .

13C-NMR (CDC13) : -%H 5 (84.4, 5C) , -Q0 (194, 3C) .

b. MeCpRe(CO)3

Methylcyclopentadiene dimer (10.4 mL, 61.2 mmole),

Re2(CO)jQ (10.0 g, 15.3 mmole), and a magnetic stir bar were

placed into the Fisher-Porter tube. This reaction was

carried out as described in general procedure. The total

released pressure of CO and H2 gas was 300 psi. The yield

of MeCpRe(CO)3 was 9.5 g (89 %).

IR (Cyclohexane) : 2027 (m), 1935 (s) cm"*.

*H-NMR (CDCI3) : -CgU^Me (5.23 dd, J - 4.2 Hz, 4H), -CH3

(2.23 S, 3H) . 13C-NMR (CDCI3) : -£H3 (13.6), -£-Me (106.6),

-£H (83.8, 2C), -£H (83.2, 2C), -QO (194.6, 3C).

C. Cp*Re(CO)3

Freshly distilled pentametylcyclopentadiene (7.9 mL,

50.5 mmole), Re2(C0)|Q (10.0 g, 15.3 mmole), and a magnetic

stir bar were placed into the Fisher-Porter tube. This

reaction was carried out as described in general procedure.

28

The total released pressure of CO and H2 gas was 315 psi.

The yield of Cp*Re(CO)3 was 11.7 g (94 %).

IR (Cyclohexane) : 2013 (m), 1922 (s) cm"*.

*H-NMR (CDCI3) : -Qi3 (2.15 s, 15H) .

13C-NMR (CDC13) : -£H3 (10.7, 5C), -£-Me (98.4, 5C) , -Q0

(198, 3C).

d. CgH7Re(CO)3

Freshly distilled indene (6.3 mL, 53.6 xnmole),

Re2(CO)jQ (10.0 g, 15.3 mmole), and a magnetic stir bar were

placed into the Fisher-Porter tube. This reaction was

carried out as described in general procedure. The total

released pressure of CO and H2 gas was 280 psi. The yellow

crystalline material was purified by column chromatography

over silica gel with petroleum ether/C^C^ (4:1 v/v). The

yield of CgH7Re(CO)3 was 9.7 g (82 %).

IR (Cyclohexane) : 2028 (s), 1940 (s), 1933 (s) cm"*.

*H-NMR (CDC13) : -Ha (5.65 t, J - 3.2 Hz, 1H), -Hb (5.78 d,

J = 3.2 Hz, 2H), -Hc (7.10 dd, J - 3.2 Hz, 2H), -Hd (7.50

dd, J = 6.3 Hz, 2H).

13C-NMR (CDC13) : -S.0 (193, 3C) , -£j (91, 1C) , -£2 (71, 2C) ,

-C3 (108, 2C), -£4 (126, 2C), (123.5, 2C) .

29

3. General Procedure for the Preparations of

Cyclopentadienyl Rhenium Dicarbonyl Dibromide

Derivatives, Cp'Re(CO)2Br2 (Cp- - cp, MeCp, Cp*)

Cyclopentadienylrhenium tricarbonyl and pyridinium

hydrobromide perbromide were placed into a 250 mL of round

bottom flask with a magnetic stir bar, then trifluoroacetic

acid was transferred to the reaction flask. After stirring

the reaction mixture for 30 min at room temperature, the

reaction mixture was quenched by pouring it into 1000 mL of

water. The pale orange precipitate was filtered and washed

three times with water (100 mL portions). After drying

under aspirator suction, the crude product was chromato-

graphed over silica gel. Successive elution with petroleum

ether/CH2Cl2 (1:1 v/v) gave first unreacted rhenium

tricarbonyl as a colorless band, then diag-CpRe(CO)2Br2 as a

red band was separated, and finally lat-CpRe(CO)2Br2 as a

brown band was separated by elution with only CH2C12.

a. CpRe(CO)2Br2

CpRe(C0)3 (6.0 g, 18.0 mmole) and CgHgNHBr^ (6.1 g,

19.0 mmole) were placed into a 250 mL of round bottom flask,

and 55 mL of CFjCOgH was transferred to the reaction flask.

This reaction was carried out as described in general

procedure. Column chromatography gave unreacted CpRe(C0)3

30

(2.9 g, 48 % recovery), diag-CpRe(CO)2Br2 (2.8 g, 33 %

yield, 63 % conversion), and lat-CpRe(CO)2Br2 (0.9 g, 11 %

yield, 21 % conversion).

For diag-CpRe(CO)2Br2 : *H-NMR (CDClj) ; -C5H5 (5.75 s, 5H) .

13C-NMR (CDCI3) ; -£5H5 (93.9, 5C) , -£0 (182.6, 2C) .

IR (CH2C12) y 2067 (m), 2003 (s) cm"1.

For lat-CpRe(CO)2Br2 : H-NMR (CDCI3) ; -CgHg (6.15 s, 5H) .

13C-NMR (CDCI3) ; -£5H5 (94.6, 5C) , -£0 (196.4, 2C) .

IR (CH2C12) ; 2054 (s), 1983 (m) cm"1.

b. MeCpRe(CO)2Br2

MeCpRe(CO)3 (5.9 g, 16.8 mmole) and CgHgNHBr3 (5.6 g,

17.5 nunole) were placed into a 250 mL of round bottom flask,

and 50 mL of CF3C02H was transferred to the reaction flask.

This reaction was carried out as described in general

procedure. Column chromatography gave unreacted MeCpRe(CO)3

(2.6 g, 44 % recovery), diag-MeCpRe(CO)2Br2 (2.7 g, 33 %

yield, 60 % conversion), and lat-MeCpRe(CO)2Br2 (1.0 g, 12 %

yield, 22 % conversion).

For diag-MeCpRe(CO)2Br2 : -NMR (CDCI3) ; -CH3 (2.3 s, 3H),

-C5MeU4 (5.62 and 5.47 dd, each 2H). 13C-NMR (CDCI3) ;

-£H3 (13.2), -£-Me (109.3), -CH (94.5, 2C), -CH (93.4, 2C) ,

-£0 (183.6, 2C).

31

IR (CH2C12) ? 2063 (m), 2046 (m), 1998 (s) cm"1.

For lat-HeCpRe(CO)2Br2 : lH-NMR (CDClj) ; -CH3 (2.15 s, 3H),

-CgMeJfy (5.88 and 5.78 dd, each 2H). 13C-NMR (CDC13) ; -£H3

(14.1), -£-Me (111), -£H (97.9, 4C). IR (CHgClg) ; 2050

(s), 1978 (m) cm"1.

c. Cp*Re(CO)2Br2

Cp*Re(CO)3 (9.7 g, 24.0 mmole) and CgHgNHBr3 (8.0 g,

25.0 mmole) were placed into a 250 mL of round bottom flask,

and 70 mL of CF3C02H was transferred to the reaction flask.

This reaction was carried out as described in general

procedure. Column chromatography gave unreacted Cp*Re(C0)3

(5.0 g, 51 % recovery), diag-Cp*Re(CO)2Br2 (0.5 g, 4 %

yield, 8 % conversion), and lat-Cp*Re(CO)2Br2 (4.7 g, 37 %

yield, 75 % conversion).

For diag-Cp*Re(CO)2Br2 : -NMR (CDC13) ; -CE3 (1.98 s,

15H). 13C-NMR (CDC13) ; -£-Me (105, 5C), -£H3 (10.4, 5C) , -

CO (186.3, 2C) . IR (CH2C12) ; 2050 (m), 1980 (s) cm"1.

For lat-Cp*Re(CO)2Br2 : -NMR (CDC13) ? -CH3 (2.03 s, 15H).

13C-NMR (CDC13) ; -C-Me (106.8, 5C), -£H3 (10.4, 5C) , -CO

(201.5, 2C). IR (CH2C12) ; 2034 (s), 1959 (m) cm"1.

32

d. (d5-Cp)Re(CO)2Br2

(d5-Cp)Re(CO)3 (0.75 g, 2.2 mmole) and CgHgNHBrj (0.77

g, 2.4 mmole) were placed into a 100 mL of round bottom

flask, and 10 mL of CF3C02H was transferred to the reaction

flask. This reaction was carried out as described in

general procedure. Column chromatography gave unreacted

(dg-Cp)Re(CO)j (0.35 g, 47 % recovery),

diag-(d5-Cp)Re(CO)2Br2 (0.34 g, 32 % yield, 61 %

conversion), and lat-(d5-Cp)Re(CO)2Br2 (0.12 g, 11 % yield,

21 % conversion).

For diag-(d5-Cp)Re(CO)2Br2 : 13C-NMR (CDCI3) ; -£5Dg (182.5,

5C). IR (CH2C12) ; 2067 (m), 2003 (vs) cm-1.

For lat-(d5-Cp)Re(CO)2Br2 : 13C-NMR (CDCI3) ; -£5d5 (195.4,

5C). IR (CH2C12) ; 2054 (vs), 1983 (s) cm"1.

4. diag-[ (r?4-C5Hg)Re(CO)2Br2] [Li]

The thermally unstable ring-attacked anionic complex,

diag-[ (rj —<CgHg)Re(CO)2Br2] [Li], was synthesized from the

treatment of diag-CpRe(CO)2Br2 (46.7 mg, 0.1 mmole) with a

stoichiometric amount of 1.0 M LiEt3BH(D) (0.11 mL, 0.11

mmole) in 20 mL of THF at -78 ®C. The product solution

exhibited a yellow color.

1 H-NMR (dg-THF, -70 °C) : -CH2 (2.90 s, 2H), -CH (6.43 s,

33

2H), -CH (6.53 s, 2H).

13C-NMR (dg-THF, -70 °C) : -£H2 (42.6), -£H (134.3 S, 2C) ,

-£H (133.1 s, 2C), -£0 (208, 2C).

IR (THF, -70 °C) : 1884 (s), 1758 (m) cm"1.

5. [CpRe(CO)2Br][Li]

The thermally stable product of metal-halogen exchange,

[CpRe(CO)2®r][Li], was synthesized by treating

diag-CpRe(CO)2Br2 (46.7 mg, 0.1 mmole) with a stoichiometric

amount (l.o eq.) of RLi (0.11 mmole, e.g. R = Me, Ph, t-Bu)

in 20 mL of THF at -78 °C. The product solution exhibited a

yellow color.

^H-NMR (dg-THF) : -C5H5 (4.80 s, 5H).

13C-NMR (dg-THF) : -£5H5 (87.2 S, 5C) , -£0 (214.3 s, 2C) .

IR (THF) : 1882 (s), 1806 (s) cm"1.

6. [Cp*Re(C0)2Br][Li]

The thermally stable product of metal-halogen exchange,

[Cp*Re(CO)2Br][Li], was synthesized by treating

lat-Cp*Re(CO)2Br2 (53.7 mg, 0.1 mmole) with two equivalents

of LiEtjBH or one equivalent of RLi (R = Me, Ph, sec-Bu,

t-Bu) in 20 mL of THF at -78 °C. The product solution

exhibited a yellow color.

34

IR (THF) : 1862 (s), 1788 (m) cm'1.

7. [CpRe(CO)2BrJ[M] (M - Li(15-Crown-5), PPP)

diag-CpRe(CO)2Br2 (186.8 mg, 0.40 mmole) was dissolved

in 40 mL of THF( then 1.7 M t—BuLi (0.24 mL, 0.41 mmole) was

added to the reaction flask at -78 °c. After the reaction

mixture was allowed to warm to room temperature, a methanol

solution (1.5 mL) of PPP-C1 (165 mg, 0.44 mmole) or a 0.46 M

15-Crown-5/THF (3 mL) was added to the reaction flask,

followed by a layer of degassed n-heptane (10 mL). The

analytical sample and yellow crystals suitable for X-ray

diffraction analysis were obtained after l day at room

temperature. After the solution was decanted, the crystals

were dried under vacuum.

For [CpRe(CO)2Br][PPP], yield : 225 mg (78 %).

Anal, for CjjHgjBrOgPRe, found (calcd); C 51.00 (51.24); H

3.50 (3.44).

For [CpRe(C0)2Br][Li(15-Crown-5)], yield : 199 mg (81 %) .

8. [MeCpRe(CO)2Br][PPP]

diag-MeCpRe(CO)2Br2 (192.4 mg, 0.40 mmole) was

dissolved in 20 mL of THF, then 1.7 M t-BuLi (0.24 mL, 0.41

mmole) was added to the reaction flask at -78 °C. After the

reaction mixture was allowed to warm to room temperature, a

35

methanol solution (1.5 mL) of PPP-ci (165 mg, 0.44 mmole)

was added to the reaction flask. Degassed n-heptane (10 mL)

was gently layered on the THF reaction solution. Orange

crystals were formed immediately at the solution interface.

After solvent decantation, the crystals were collected and

dried under vacuum. Yield : 251 mg (85 %).

Anal, for C^I^BrC^PRe, found (calcd) ; c 51.27 (51.88); H

3.70 (3.65).

13C-NMR (CD2C12) : -£H3 (14.2, 1C) , -CH (79.4, 2C) , -CH

(79.0, 2C), quaternary (118.3 and 117.1 d, J_ - 90 Hz, p-c

1C), ortho and meta (130.9 and 134.7, 2C each), para (135.9,

1C), -£0 (209, 2C).

9. [Cp*Re(CO)2Br][PPP]

lat-Cp*Re(CO)2^r2 (200 mg, 0.37 mmole) was dissolved in

40 mL of THF, then 1.7 M t-BuLi (0.23 mL, 0.39 mmole) was

added to the reaction flask at -78 °C. After the reaction

mixture was allowed to warm to room temperature, a methanol

solution (2 mL) of PPP-C1 (153 mg, 0.40 mmole) was added to

the reaction flask. The solution was concentrated to 10 mL,

filtered under argon using a fine porosity frit, and then

degassed n-heptane (10 mL) was gently placed on top of the

reaction solution. The analytical sample and orange yellow

crystals suitable for X-ray diffraction analysis were

36

obtained after 2 days at room temperature. After the

solvent was decanted, the crystals were collected and dried

under vacuum. Yield : 254 mg (86 %).

Anal, for CjgHjjBrOgPRe, found (calcd); C 54.32 (54.27); H

4.43 (4.40).

IR (KBr) : 1850, 1773 cm"1, IR (CHgClg) : 1860, 1781 cm"1.

1h"NMR (CD2C12) : -CH3 (1.94 s, 15H), -CgHg (7.6-7.95 m,

20H). 13C-NMR (CD2C12) : -£H3 (11.0, 5C) , -£-CH3 (93.8,

5C), quaternary (117.3 and 118.5 d, Jp.c - 90 Hz, 1C), ortho

and meta (131.0 and 134.8, 2C each), para (136.1, 1C),

-S.0 (213.1, 2C).

10. Reaction of Cp'Re(CO)2Br2 with One Electron

Reducing Agents (Cp1 « Cp, Cp*)

In a typical experiment, a THF solution (20 mL) of

Cp'Re(CO)2Br2 (0.05 mmole, either isomer) was treated with

0.5 M sodium naphthalide (0.2 mL) or 0.25 M cobaltocene

(0.40 mL) at -78 °C. Red-brown solutions of Cp'Re(CO)2Br2

(either isomer) react instantaneously with added sodium

naphthalide to give a yellow solution containing a mixture

of carbonyl oxygen-sodium contact ions and solvent-separated

ion pairs, while cobaltocene gives only solvent-separated

ions.

37

IR (THF, -78 °C) : For [CpRe(CO)2Br][Na] (CIP) ; 1882, 1793

cm"1. [CpRe(CO)2Br]" (SSIP) ; 1882, 1808 cm'1.

For [Cp*Re(CO)2Br][Na] (CIP) ; 1863, 1776 cm"1.

[Cp*Re(CO)2Br]" (SSIP) ; 1863, 1791 cm"1.

11. Reaction of Cp'Re(CO)2Br2 with Grignard Reagents

(Cp» - Cp, Cp*)

A THF solution (20 mL) of Cp'Re(CO)2Br2 (0.05 mmole,

either isomer) was treated with 1.5 M MeMgBr (0.04 mL) or

2.0 M t-BuMgCl (0.03 mL) at -78 °C. Red-brown solutions of

Cp'Re(CO)2Br2 (either isomer) react sluggishly at -78 °C but

react instantaneously with the added Grignard reagents at

room temperature to give a yellow solution containing only

carbonyl oxygen-magnesium contact ions, [Cp'Re(CO)2Br][MgX].

IR (THF, -78 °C) : For [CpRe(CO)2Br][MgX] ; 1890, 1757 cm"1.

For [Cp*Re(CO)2Br][MgX] / 1872, 1734 cm"1.

12. C6H5CpRe(CO)3

lat-CpRe(CO) 2Br2 (467 mg, 1.0 mmole) was placed into a

Schlenk flask, then 40 mL of distilled THF was cannulated

into the Schlenk flask. This reaction mixture was next

cooled to —78 ®C. 1.0 M PhLi (1.1 mL, 1.1 mmole) was added

to the Schlenk flask using a syringe. After stirring the

reaction mixture for 20 min, the vessel was allowed to warm

38

to room temperature. The solvent was removed under

aspirator suction and the residue was dissolved in minimum

amount of CJ^C^. The crude product was column

chromatographed over silica gel. Successive elution with

petroleum ether/CHgClg (4:1 v/v) gave PhCpRe (CO) 3 (123 mg,

30 % yield) which was characterized spectroscopically.

^H-NMR (CDCI3) : -C6H5 (7.40 m, 5H) , -CH (5.43 m, 2H) , -CH

(5.79 m, 2H).

13

C-NMR (CDClj) : meta (126.6, 2C), ortho (129.0, 2C), para

(128.8, 1C), quaternary (132.0, 1C), -£H (82.1, 2C), -£H

(84.5, 2C), -£-Ph (108.8, 1C). IR (Cyclohexane) : 2028 (m), 1939 (s) cm'1.

13. General Procedure for the Preparations of Cyclo-

pentadienyl Rhenium Dicarbonyl Alkyl Bromide

Derivatives, CpRe(CO)2(R)Br (R = H, CH3, C2H5)

To a THF solution (30 mL) of the [CpRefCO^Br] [Li]

anion freshly prepared from the reaction of

diag-CpRe(CO)2Br2 (46.7 mg, 0.1 mmole) and a stoichiometric

amount (1.0 eg.) of t—BuLi was added trifluoroacetic acid,

methyl triflate (CF3SO3CH3), or magic ethyl (FSO3C2H1J) at

-78 °C, respectively. This reaction was monitored in the

carbonyl region by IR spectroscopy.

39

a. diag-CpRe(CO)2(H)Br.

The desired hydridobromide complex was immediately

formed when trifluoroacetic acid (0.10 mL, 1.25 mmole) was

added to the THF solution of [CpRe(CO)2Br][Li] anion (0.10

mmole) at -78 °c. The resulting hydridobromide complex was

shown to possess diagonal stereochemistry based on IR

spectral measurements.

IR (THF, -78 °C) : 2032 (s), 1962 (s) cm'1.

]H-NMR (dg-THF, -78 °C) : -CgHg (6.06 S, 5H) , -Re-fi (-9.03

S , 1H).

b. diag-CpRe(CO)2(CH3)Br

Methyl triflate (0.02 mL, 0.20 mmole) reacted

immediately with the [CpRe(CO)2Br][Li] anion (0.10 mmole) in

THF solution at -78 °C to afford the desired methylbromide

complex with diagonal stereochemistry.

IR (THF, -78 °C) : 2028 (m), 1954 (s) cm"1.

c. diag-CpRe(CO)2(C2H5)Br

Magic ethyl (0.02 mL, 0.20 mmole) reacted immediately

with the [CpRe(CO)2Br][Li] anion (0.10 mmole) in THF

solution at —78 to afford the desired ethylbromide

complex with diagonal stereochemistry.

40

IR (THF, -78 °C) : 2032 (m), 1959 (s) cm"1.

d. diag-Cp*Re(CO)2(H)Br

To a THF solution (30 mL) of [Cp*Re(C0)2Br][Li] anion

prepared from the reaction of lat-Cp*Re(CO)2Br2 (53.7 mg,

0.1 mmole) and a stoichiometric amount (1.0 eq.) of t-BuLi

was added trifluoroacetic acid (0.10 mL, 1.25 mmole) at

-78 °C. The desired hydridobromide complex was formed

immediately and was shown to possess diagonal stereo-

chemistry by IR measurements.

IR (THF, -78 °C) : 2018 (s), 1945 (vs) cm"1.

^H-NMR (dg-THF) : -CH3 (2.01 s, 15H), -Re-fl (-9.98 s, 1H) .

e. diag-Cp*Re(CO)2(CH3)Br

To a CH2C12 solution (10 mL) of [Cp*Re(C0)2Br][PPP] (40

mg, 0.05 mmole) was added 1.0 M methyl triflate/CH2d2 (0.06

mL, 0.06 mmole) at -78 ®C. The desired methylbromide

complex was formed immediately and was shown to possess

diagonal stereochemistry based on IR measurements.

IR (CH2C12, -78 °C) : 2023 (m), 1942 (s) cm"1.

41

f. diag-Cp*Re(CO)2(C2H5)Br

To a CH2C12 solution (10 mL) of [Cp*Re(C0)2Br][PPP] (40

mg, 0.05 mmole) was added magic ethyl (0.006 mL, 0.06 mmole)

at -78 °C. The desired ethylbromide complex was formed

immediately and was shown to possess diagonal

stereochemistry based on IR measurements.

IR (CH2C12, -78 °C) : 2020 (m), 1944 (s) cm"*.

14. [C9H7Re(CO)2H][Li]

To a THF solution (25 mL) of CgH7(CO)3 (100 mg, 0.26

mmole) was added two equivalents of LiEtjBH at room

temperature. The resulting thermally stable hydrido rhenium

anion, [CgHyRe(CO) 2H] , was characterized in situ

spectroscopically.

IR (THF) : 1992, 1906 cm"1.

JH-NMR (dg-THF) : -Re-H (-5.5 s, 1H), -Ha (6.56 t, J=3.3 Hz,

1H), -Hb (5.64 d, J—3.2 HZ, 2H), -J^ (6.42 dd, J=6.0 Hz,

2H), (7.22 dd, J= 5.9 Hz, 2H) .

13C-NMR (dg-THF) : -£0 (200, 2C), -£j (91), -£2 (71, 2C),

-£3 (HO, 2C) , -£4 (120, 2C) , (114, 2C) .

42

15. diag-Cp»Re(CO)2H2 (Cp' = Cp, Cp*)

To a benzene solution (40 mL) of Cp'Re(CO)2Br2 (2.14

mmole, either isomer) was added BU3S11H (1.20 mL, 4.40 mmole)

at room temperature. The reaction mixture was stirred

overnight at room temperature, with monitoring by IR. The

solvent was then removed under vacuum and the crude oily

product was flash column chromatographed over silica gel

under nitrogen. After the filtrate was dried under vacuum,

the pure product was crystalized in pentane at -78 °c using

a dry ice/acetone bath.

For diag-CpRe(CO)2H2 : IR (Cyclohexane) : 2022 (m), 1953 (s)

cm"1. ^-NMR (CgDg) : -Re-H (-9.68 S/ 2H) , -C^g (4.35 d,

5H) • For diag-Cp*Re(CO)2H2 : IR (Cyclohexane) : 2026 (m),

1936 (s) cm"1. !H-NMR (CgDg) : -Re-H (-9.25 s, 2H), -CH3

(1.81 s, 15H).

16. Cp'Re(CO)2(SnBU3)2 ^Cp' ™ C p'

To a benzene solution (40 mL) of Cp'Re(CO)2Br2 (2.2

mmole) was added BujSnH (1.2 mL, 4.5 mmole) at room

temperature. After the reaction mixture was stirred for

several hours with monitoring by IR, 0.7 mL of EtjN (5.0

mmole) was added to the reaction flask and stirring

continued for one additional hour. The solvent was then

43

removed under vacuum, and the crude oily product was flash

column chromatographed over silica gel with petroleum ether

under nitrogen. After the filtrate was evaporated under

vacuumm, the volatile compounds were sublimed from the

desired tin complex at 90-100 °C. The pure product was

obtained from the above sublimation residue using flash

column chromatography over silica gel with petroleum ether

under nitrogen.

For diag-CpRe(CO)2(SnBU3)2 : IR (Cyclohexane) : 1943 (m),

1890 (vs) cm"1. 119Sn-NMR (CgDg) ; -5.6 ppm.

For lat-Cp*Re(CO)2(SnBu3)2 : IR (Cyclohexane) : 1967 (vs),

1896 (s) cm"1. 119Sn-NMR (CgDg) : 8.8 ppm.

17. CpRe(CO)2L (L - PPh3, P(0Ph)3, (CH3)3CNC)

To a THF solution (30 mL) of diag-[ (rj*-C,jHg)Re(CO)2Br2]"

prepared from the reaction of diag-CpRe(CO)gBrg (93.4 mg,

0.20 mmole) and 1.0 M LiEtjBH (0.24 mL, 0.24 mmole) was

added L [0.22 mmole, i.e., L = PPh3, P(OPh)3, Me3CNC] at -78

°C. The yellow solution of diag-[ (^-CgHg)Re(CO) 2Br2]" anion

reacted instantaneously with added L to give CpRefCO^L at

-78 °C as determined by low-temperature FT-IR measurements.

IR (THF) : For CpRe(CO)2PPh3 : 1926 (s), 1844 (s) cm"1.

44

For CpRe(CO)2P(OPh)3 : 1961 (s), 1893 (s) cm"*.

For CpRe(CO)2(Me3CNC) : 1902 (s), 1850 (s) cm"1.

18. BrRe(CO)g

Re2(C0)jQ (7.0 g, 10.7 mmole) was placed into a 500 mL

of round bottom Schlenk flask equipped with a magnetic stir

bar. Freshly distilled pentane (300 mL) was transferred

into the reaction flask and bromine (0.7 mL) was then added

dropwise to the solution using a dropping funnel under a

stream of argon. A white precipitate was observed

immediately upon stirring at room temperature. The reaction

mixture was stirred at room temperature overnight and then

monitored by IR. Excess bromine was eliminated from the

reaction vessel by stirring open in the hood until all of

the orange bromine color disappeared. The solvent was

removed under vacuum, and the white powder transferred to a

sublimator and sublimed at 85-110 °C under vacuum. Yield :

8.2 g (94 %).

IR (Cyclohexane) : 2044 (vs), 1984 (m) cm"1.

19. (d5-Cp)Tl

Freshly distilled cyclopentadiene (1.2 mL, 4.9 mmole)

was added to a Schlenk flask containing a sodium deuteroxide

solution, prepared from Na metal (3.7 g, 0.16 mole) and D£0

45

(40 mL), and the reaction flask was stirred magnetically at

0 °C for 5 days in the cold room. Thallium (I) sulfate (3.0

g/ 0.01 mole) was then added to the reaction flask and

stirring was continued for 3 additional days. The yellow

precipitate was filtered, washed with 10 mL of DgO, and

dried under vacuum for 2 hr. The solid was transferred to a

water-cooled sublimation apparatus and plug of glass wool

was placed on top of the crude product. The crude (dg-Cp)Tl

was sublimed at 100-110 °C under vacuum to give a yellow

solid. The yield was 1.97 g (72 %).

20. (dg-Cp)Re(C0)3

(dg-Cp)Tl (1.74 g, 6.35 mmole) and BrRe(CO)5 (2.34 g,

5.76 mmole) were dissolved in 250 mL of Schlenk round bottom

flask containing 50 mL of benzene. The reaction mixture was

magnetically stirred at 60-70 °C for 1 day with monitoring

by IR. The benzene solution was flash column

chromatographed over silica gel to give the desired product.

The yield was 1.67 g (71 %). IR (Benzene) : 2023 (s), 1927

(vs) cm"*.

46

E. Thermodynamic Measurements

The Ion Pairing Equilibrium in 2-Na, SSIP vs. CIP

2.5 x 10"3 M THF solutions of the [Cp*Re(CO)gBr] [Na]

anion were prepared from the reaction of lat-Cp*Re(CO)2Br2

(26.85 mg, 0.05 mmole) and 0.5 M sodium naphthalide (0.24

mL, 0.12 mmole) in THF (20 mL) at -78 °C. The measurement

of the amount of contact ion pair (CIP) and solvent-

separated ion pairs (SSIP) in solution over the temperature

range of -70 °C to room temperature was assessed by FT-IR

spectroscopy. From the area associated with each asymmetric

carbonyl stretching band, the equilibrium constants were

easily determined as defined by Eq. 12.

Keq

[Cp*Re (CO) 2®r] [Na] * [Cp*Re(CO)2Br]"//Na+

CIP SSIP (12)

[8SIP] Keq s

[CIP]

The thermodynamic parameters were obtained by plotting

In Keq + 1 vs. 1/T from which the slope and intercept

afforded AH and AS, respectively. The error limits

associated with the enthalpy and entropy for ion-pairing

47

were calculated by using the available least-squares

regression program®® and should not be taken to reflect

uncertainties in sample preparation, band area intensities,

or temperature control, but rather the deviation of the data

points about the least-squares line.

F. Kinetic Studies

1. Reaction of diag-CpRe(CO)2^2 with Et3SiH

diag-CpRe(CO)2Br2 (46.7 mg, 0.1 mmole) was placed in a

30 mL Schlenk flask containing 10 mL of toluene. This

toluene solution was shaken vigorously for a few minute

prior to placement in a thermostatted bath set at 40.0 °C.

The reaction was monitored over a three day period by

following the decrease in absorbance of the reactant's

intense of asymmetric stretching CO band (2003 cm"1). The

extent of the reaction between diag-CpRe(CO)2**2 a n d Et3siH

was measured from the above data.

2. Isomerization of lat-CpRe(CO)2^*2 t o

diag-CpRefCO^B^ in the Presence of EtjSiH

lat-CpRe(CO)2Br2 (5 mg, 0.011 mmole) was placed in a 30

mL Schlenk flask containing 5 mL of toluene. This toluene

solution was shaken vigorously minute prior to placement in

48

a thermostatted bath set at 40.0 °C. A 10-fold excess of

EtgSiH (0.16 mL, 1.0 mmole) then was added to the reaction

flask at 40.0 °C. This reaction was monitored over an eight

hour period by following the decrease in absorbance of the

reactant's intense of symmetric stretching CO band (2046

cm *). The extent of the isomerization to diag-CpRe(CO)2Br2

was qualitatively measured from the above data and was shown

to be faster than the rate of bromide/hydride exchange (vide

infra).

49

CHAPTER III

RESULTS AND DISCUSSION

A. Aryl Rhenium Tricarbonyl Derivatives, Cp *Re(CO)^

(Cp' = Cp, MeCp, Cp*, cgH7)

The aryl rhenium tricarbonyl derivatives were prepared

by a modification of the procedure reported previously^"

Dirhenium decacarbonyl, Re2(CO)10, suspended in the

appropriate cyclopentadiene dimer or indene was heated up to

210 C in the Fisher—Porter tube equipped with a pressure

gauge. This method gives excellent yields of the aryl

rhenium tricarbonyl derivatives that are used for the

preparation of the necessary aryl rhenium dicarbonyl

dibromide derivatives. Equation 13 outlines the general

procedure for the synthesis of CpRe(CO)3. The aryl rhenium

Re2(CO)10 + A » 2 CpRe(CO)3 + Hg + 4 CO (13)

tricarbonyl derivatives were characterized by IR and NMR

spectroscopy. These data are summarized in Tables 1-2. The

representative IR spectra of the aryl rhenium tricarbonyl

50

- Q

i H O

• H

<#> r ^ oo

a\ 00 <n

cn oo

> i C O

fl

<d o

• H

• H G a)

« g

I a

t c ° 8

X m

O

a) c «d X a>

£ o H O &

c • H

§ 8

CO <«

CO

«d H 3

o

co CO w N—» CO i n CN H

• • CN (N

• d co - P

<w w o CO i n 00 cn CN \ o r * • • • •

i n i n i n m

CO CO CO CO 10 <w V** w * w . w cn i n CM CO o n CO CN n

o \ as as o\ H H H H H

% * * %»

*>*%

S s s CO S I M ' >—<•

o CO 00 CO CN H CN O o O o CN CN <N CN

OO

O a a) #

&

OO

O a a) «

fi-<D s

CO

O O

0) tf *

&

OO

O O

5 «

HJ c n

O

J T u Q

a

+> G 0) >

>1 a o

a (d o (d o a)

T *

• H c 0)

•8

0) 43 •M

a o

i * 0 (0

o <d C0 CQ

ro - Q

5 1

G O

•B <d o

•H u

•H c 0)

CM

CQ

a

i ft

c •H U

V 0)

0)

! 6 I X

•H Cft

C •H U

*d (I)

<1)

! s i 0) >

•H J*

(d rH P

o o

CO X o

i n O

O

CO O

CM a

u

o (x»

OH VO

OS H

VO

CO

CM •

CO CO

CO CO as as H H

o

CO

O

a

«

&

V0 •

VO O H

CO

O a

a) «

& a) S3

in

co CM H

VO CM

CO o H

CO • • • H

<?* CO CO CO CO OH

CO

O O

a) « # a -

H OH

CO

O O w 0) « t « cr»

O

C (d

(0 (d

*d a) w 3

w <d *

CO

o

4J 0

rH 04

•H

-M

> i 4-> # H A w 8 G g« <D Mi

•P _

g °

& r *

a>

g 0) o c 0)

8 8 u

H •p (0 c c a) M > a)

h - p 0 5 CO -H flj

52

derivatives are shown in Figure 2. The FT-IR spectra of

cyclopentadienylrhenium tricarbonyl derivatives exhibit two

carbonyl stretching bands (Aj + E) expected for C3y

66

symmetry. The symmetric CO stretching mode (Aj) of

cyclopentadienylrhenium derivatives appears in the range of

2030-2013 cm * while the asymmetric CO stretching mode (E)

appears in the range of 1939-1922 cm"1. Successive

substitution of methyl groups on the cyclopentadienyl ring

in the series of CpRe(CO)3, MeCpRe(C0)3, and Cp*Re(CO)3

leads to decreasingly lower CO stretching frequencies as the

cyclopentadienyl ring moiety donates more electron density

to the rhenium metal center in the order of Cp*Re(C0)3 >

MeCpRe(C0)3 > CpRe(C0)3.

On the other hand, the IR spectrum of the r?5-indenyl

derivative, CgH7Re(CO)3 (Fig. 2b) exhibits a splitting of

the E mode owing to sufficient asymmetry of the r^-indenyl-

rhenium bond as a result of deviation from idealized c3y

67 68 1

symmetry. * The H NMR spectrum of the yellow—orange

product CgHyRe(CO)3 exhibits four sets of resonances with a

relative integral ratio of 2:2:2:1 (Fig. 3). The two of

these resonances of relative intensities 2 are observed at

7.10 and 7.50 ppm with the AA'BB' spin system which are

consistent with four protons on an uncomplexed six-membered

benzenoid ring. The other two sets of resonances are

53

2000 — i 1900

v, cm i

Figure 2. Infrared spectra of the carbonyl region for

(a) CgHyRe(CO)^ and (b) CpRe (CO) . Both spectra were recorded at 25° C in cyclohexane.

54

C B

Q > A

' T \ c K l C,

D

/

A

B

1 1 1 1 I 1 ' ' I j I ' I I | I I I I | I I I I | I I | |

PPM

| i i i i | i i i r | 6 i 6 5

Figure 3. *H NMR spectrum of CgH7Re(CO)3 at 25° C in CDC13.

55

observed at 5.78 and 5.65 ppm with a relative integral

ratio of 2:1 which is correspond to the three protons

attached to the five-membered ring in the rj®-indenyl ligand.

B. Cyclopentadienyl Rhenium Dicarbonyl Dibromide

Derivatives, Cp'Re(CO)2&r2 = MeCP' CP*)

Synthetic routes to the cyclopentadienylrhenium

dicarbonyl dihalide compounds, CpRe(CO)2X2 (where X = Br or

28-31

I), have been known for several years. However,

synthetic pathways to the corresponding pentamethy1eye1o-

pentadienylrhenium dicarbonyl dihalides, Cp*Re(CO)( w h e r e

Cp* » rj®-CgMeg), have only recently been described in

detail.«.«"«

The cyclopentadienylrhenium dicarbonyl dibromide

derivatives were prepared by a modification of the route

29

described by King and Reimann. The appropriate rhenium

tricarbonyl and a slight molar excess of pyridinium

hydrobromide perbromide was treated with TFA (Eq. 14). This yielded the crude isomeric dibromides which were

TFA diag-Cp'Re(CO)2Br2 Cp*Re(CO)3 + CgHeNHBr, + (14)

rt lat-Cp'Re(CO)

(Cp1 = Cp, MeCp, Cp*)

56

subsequently isolated by column chromatography over

silica gel using mixture of petroleum ether and CH^Clg.

Successive elution with petroleum ether/CHgClg (1:1 v/v)

gave first unreacted rhenium tricarbonyl as a colorless

band, then diagonal dibromides as a red band, and finally

lateral dibromides as a brown band was isolated by using

CHgClg. An increased amount of TFA led to an overall

increase in the yield of the isomeric dibromides, but use of

Br2 directly and/or longer reaction times led to an overall

decrease in the yield of the isomeric dibromides. The aryl

rhenium dicarbonyl dibromide derivatives were characterized

by IR and NMR spectroscopy. The spectral properties are

consistent with those reported by King et al.^® and Sutton

32

et al. These data are summarized in Tables 3-4. The

four-legged piano-stool complexes of the form CpMX2Y2 may

exist as two nonequivalent stereoisomers that are commonly

referred to as cis and trans isomers. We have adopted

King's nomenclature to describe the isomeric dibromides

discussed in this study. Here the descriptors lateral and

diagonal correspond to the cis and trans stereoisomers,

respectively. The diagonal designation is more acceptable

than the trans designation because no two ligands are

situated at 180° angles relative to each as would be

expected for trans isomers. The diagonal (A) and lateral

57

C o •a <d o

• H Q

Q)

co

A

a

- Q T f rH 0)

• H

i S 4

ro

CM i H U

CM

5 C

•H

o o

<*>

r o

s

£ » LO

0 1

co <

CO

<d H

I o f*4

CO H CO T • vo • as

CO H CO CM vo •

H CO CO

0 O CO CO CO CO a> 0 • • • •

CM CM H CM

•h y—S t0mSl W X cm M

w 04 0

SO <n B « • • i n i n m i n

i n i n r * H • • « X i n VO N CM

w r * CO

• • i n i n

CO n CO S «T 8* <W w w <w

r o CO 00 CO 0 ON O CO as r * CO i n O 0 \ o^ o \ o\ <}\ CM H H H H H

«—*» 8 CO S e CO e w"

w w « w s—' CO vo 0 0

vo i n vo i n i n CO 0 0 0 0 0 0 O CM CM cm CM CM CM CM

CM u co CM

O O "W Q)

a

?

t n <d

•H TJ

CM

PQ CM

O U w a) «

P4

a

<d

CM

CQ CM

O 0

0) tf & 0 £ 1 Cn <d

• H T3

CM

CQ CM

O 0

0) Pd a u Q) s 1

-M <d

rH

CM U m CM

O O

0)

•K a< a » t n (0

CM *4 PQ

CM

0 a a) «

*

8-1

4-> <a

CO rH o Q O

. p C

a) >

> . G O

• 8 <d o

• H

«P

§ fi

>1 C a)

• H * 0 <d

•P C (1) §•

rH O >1 o

a>

3 a o

* d a) CO

o <d CO a txs JQ

53

>i c o A u (d o

s

PQ

6

o o

17* C •H U

•d a) a) i a) a i <D > •H

CO B O

8

8

f o

<d H 3

0

VO VO in CO in • • • • • •

(N VO co vo VO H 00 as 00 as 00 O H H H H H ca

a* • co o\

VO

en

ca

n H

CO 0\

in

CO

0\ o H

CO on

ca

00

o o H

IT) O

00 • vo o

CM u CM

CM m u u CM PQ m CM CM o

o O o w o u s <D a» w a» w 0 a) « & as * Q* a) a< O s

1 u i

tP 1 tr> 1 cn. 1 (d •P <d -P (d •P •H <d •H ed •H (d •O iH 15 H T* rH

c (d

w <d

*d a) CO 3

W <d

oo iH U Q U

o

4J a)

-P

>1 4J "H J W § C Q) «P C •H §

Oi a

r*

-P <d

0) 43 Eh

a) o c a)

_ u CO ® H <W O o> Q M U

-P C 0) > H o CO

as

<d a u <u •p c •H

59

(B) isomers were identified by determination of the angles

69 between their two C-0 bonds through the relationship"*

<X>j

I I S / v - B ^ c C o °

B

tan20 = Ia/I- (where 20 is the angle between the two C-0 a S

bonds, I. is the area under the asymmetric v(CO) band, and

a

Is is the area under the symmetric v(CO) band with the ratio

-3 I./I. being extrapolated to infinite dilution (~ 10 M) . a S

20 m5

Symmetric Asymmetric

The above relationship is derived from the fact that an IR

band intensity is directly proportional to the dipole moment

change that occurs during the vibration. For a CO stretch-

ing vibration, a symmetric vibrational mode appears at

higher frequency than its asymmetric counterpart since more

60

energy is required to stretch two CO bands simultaneously

than to execute one CO stretch and one CO contraction.

Thus, the ratio Ia/Is in diagonal isomers is 2.55-2.73,

corresponding to an angle 29 of 116-118°. Similarly the

ratio Ia/I$ in lateral isomers is 0.71-0.80, corresponding

to an angle 29 of 80-83°. Typical IR spectra of diagonal

and lateral dibromides are shown in Figure 4. The diagonal

isomer possesses higher infrared v(C0) frequencies and a

higher field cyclopentadienyl proton NMR chemical shift than

the corresponding lateral isomers. These two isomers

exhibited differences in (A[i/(C0)]) by about 13 cm"*•for

[J/(CO)sym] and 20 cm"1 for C»/(CO)asym] in CH2C12. This can

be understood in terms of the increased competition for

rhenium d-electrons between the carbonyl groups when they

are mutually trans. They also exhibited cyclopentadienyl

proton NMR resonances differing by about 0.4 ppm. The 13c

NMR chemical shift for the CO groups is smaller by about 15

ppm in the diagonal isomers. The lateral isomers appeared

to be considerably less soluble in organic solvents than the

corresponding diagonal isomers. Both the diagonal and

lateral isomers are stable in the solid state at room

temperature with respect to isomerization. Further study of

isomerization will be discussed (vide infra).

61

OC-;>ReCBr Br CO

B

2200 2000 v, cm

1800

Figure 4. Infrared spectra of the carbonyl region for (a) diag-CpRe(CO)2Br2 and (b) lat-CpRe(CO)2Br2 Both spectra were recorded at 25° C in CHgClg.

62

C. Bromorheniumpentacarbonyl, BrRe(CO)c

The pentacarbonylrhenium halides, XRe(CO)c (X = CI Br D ' '

I), first prepared by Hieber,70'71 are precursors for the

synthesis of many novel rhenium carbonyl compounds. The

bromorheniumpentacarbonyl, BrRe(CO)5, was prepared by the

reported procedure previously.72 Dirhenium decacarbonyl was

dissolved in a 500 mL Schlenk flask containing freshly

distilled pentane (300 mL) and THF solution of bromine was

added dropwise to the reaction mixture under argon

atmosphere. A white precipitate was formed immediately upon

stirring at room temperature. After the excess of bromine

and solvent was removed, the white powder was transferred to

a sublimator and sublimed at 85-100 °C under vacuum to give

a 94 % yield of BrRe(CO)g. The compound was characterized

by IR spectroscopy. The observed medium intense symmetric

stretching band at 2044 cm"* and the very strong, intense

asymmetric stretching band at 1944 cm"* (in cyclohexane)

are in excellent agreement with the literature values.72

63

D. Deuterated Cyclopentadienyl Rhenium Tricarbonyl,

(d5-Cp)Re(CO)3

Synthetic pathways to the deuterated cyclopentadienyl

rhenium tricarbonyl, (dg-Cp)Re(CO)3, have recently been

26

described in detail. Refluxing a benzene solution of

73

(dg-Cp)Tl and BrRe(CO)j overnight gives (dg-CpJRefCOJj

according to Equation 15.

BrRe(CO)5 +

(dg-Cp)Tl

BZ

60 °C (d5-Cp)Ra(CO)3 + BrTl + 2 CO (15)

The resulting white solid was obtained in a 71 % yield. The

isotopically labelled compound was used in the preparation

of (dg-Cp)Re(CO) 2Br2* T h e isotopic purity of (dg-CpJRefCOJj

was assayed by *H NMR spectroscopy using para-methoxy-

benzene as an internal reference. The NMR spectrum,

recorded on a Varian VXR-300 MHz NMR spectrometer, revealed

a very weak, broad peak at 5.5-5.6 ppm and the integral

ratio between the residual protio cyclopentadienyl ligand

and the internal standard indicates that incorporation of

deuterium on the cyclopentadienyl ring is more than 95 %.

This compound was used directly in the preparation of

(dg-Cp)Re(CO)2Br£.

64

E. Deuterated Cyclopentadienyl Rhenium Dicarbonyl

Dibromide, (d5-Cp)Re(CO)2Br2

The deuterated cyclopentadienylrhenium dicarbonyl

dibromide, (d^-Cp)Re(CO)2Br2' w a s prepared from the reaction

(d5-Cp)Re(CO)3 and CgHgNHBr^ in TFA at room temperature.

Column chromatography gave first unreacted (d5-Cp)Re(CO)3 as

a colorless band (47%), diag-(d5-Cp)Re(CO)2Br2 as a red band

(32%), and lat-(dg-Cp)Re(CO)2Br2 as a brown band (11 %).

The isotopic purity of diag-(d5-Cp)Re(CO)2Br2 was assayed by

1h NMR spectroscopy using para-methoxybenzene as an internal

standard. The H NMR spectrum exhibited a very weak singlet

at 5.7 ppm and the integral ratio of proton chemical shifts

indicated that incorporation of deuterium on the

cyclopentadienyl ring is more than 95 %. From this

observation we can conclude that no additional hydrogen is

incorporated into the cyclopentadienyl ring during the

bromination. These deuterated CpRe(CO)2Br2 isomers

exhibited no differences in the infrared CO stretching

frequencies compared to the protio CpRe(CO)2Br2 isomers.

65

F. Hydride Reduction of diag-CpRe(CO)2Br2: Synthesis

and Characterization of the Thermally Unstable

Ring-Attacked Product, diag- [ (r^-CgHg) Re (CO) 2Br2 ]"

The reaction of diag-CpRe(CO)2Br2 with LiEtjBH shows

that cyclopentadienyl ring attack occurs according to

Equation 16 in preference to either direct metathetical

replacement of the bromide moiety or hydride attack at the

carbonyl ligand to generate the corresponding formyl

species. The low-temperature FT-IR spectrum exhibited three

carbonyl stretching bands (Fig. 5a) at 1884 (vs), 1785 (w),

and 1758 (s) cm~* Such a low energy shifting of the CO

H

rH~N LiEljBH \==*J (16)

0-.-Re-..Or THF/-78*" „ J-®r CO Vco

stretching bands of anionic species relative to neutral

compound, diag-CpRe(CO)2Br2, is directly attributed to the

increased charge density at the rhenium metal center which

manifests itself by reducing the CO stretching force

constant for each of the CO ligands.66'74 These three

carbonyl stretching bands are replaced by two new carbonyl

bands upon addition of 5 equivalents of 15-Crown-5 (1,4,7,

10,13-pentaoxacyclopentadecane) as shown in Figure 5b.

66

B

2200 2000 1800 v, cm -i

Figure 5. Infrared Spectra of the carbonyl region for (a) [ (n4-C5H6)Re(CO)2Br2][Li] and (b) [(r?4-C5H6)Re(CO)2Br2] [Li] with 5 equiv. of 15-Crown-5. Both spectra were recorded at -70 °C in THF.

67

This latter spectrum is also obtained when a THF solution of

diag-CpRe(CO)2Br2 containing 5 equivalents of 15-Crown-5 is

treated with LiEt^BH at -78 °C. These results are consis-

tent with the formation of an anionic cyclopentadienyl-

rhenium complex that displays extensive carbonyl oxygen-

lithium ion pairing in the absence of 15-Crown-5.75 > 76

The identity of this new complex was characterized by

low-temperature *H NMR spectroscopy. When a dg-THF solution

of diag-CpRe(CO) gBrg was treated with LiEt^BH at -78 °C

three new hydrogen resonances appear at 6.53, 6.43, and 2.90

ppm with an integral ratio of 2:2:2. On the other hand,

the use of LiEt^BD instead of LiEt^BH led to a decrease in

the intensity of the resonance at 2.90 ppm as shown in

Figure 6; the low-field resonances associated with the

AA'BB' spin system were unaffected yielding an observed

13 1 integral ratio of 2:2:1. A C{ H} NMR spectrum reveals two

low-field resonances at 134.3 and 133.1 ppm along with a

single resonance at 42.6 ppm which suggests the presence of

two types of olefinic carbons and a methylene moiety that

are derived from the Cp ring of diag-CpRe(CO) 2&r2 a s s^ o w n

13 in Figure 7a. A C NMR spectrum using spin-echo

77-79

J-modulation spectroscopy (APT spectrum) reveals two

downward resonances at 134.3 and 133.1 ppm and two upward

resonances at 208.6 and 42.6 ppm. Under these conditions

H

d " "

L 0 C - - R e - - B r B r ^CO

J k

D

d " H

L OC-^Rew--Br Br ^CO

^ V.

68

'' ' I . ' ' ' ' I ' ' ' ' I . ' ' ' I ' ' 1 ' I ' 1 1 1 | 1 1 < i | 1 1 i i | i B - s S.5 9.0 4.S 4.0 3 .5 3 .0 PI

F i g u r e 6 . *H NMR spec t ra o f (a) [ (r^-CgHgJRefCO) ~ and (b) [ (r?4-C5H5D)Re(CO)2Br2]" a t -70°C i n dg-THF.

69

5 mixyjjrn* 5**. p

- O ( 0

- O CD

I — 0 1

A O -o

m a>

&

70

inverted resonances would be observed for methyl and methine

carbons while normal resonances would be observed for

methylene or quaternary carbons. Among these carbon

resonances, the low-field resonance at 208.6 ppm can be

easily assigned to carbonyl carbons as shown in Figure 7b.

Finally, a chemical shift correlation spectrum

(HETCOR) was recorded to provide an accurate proton/carbon

relationship®^-®^ as shown in Figure 8. The *H NMR

resonances at 6.53/ 6.43, and 2.90 ppm were shown to 11

correlate with the 1 C NMR resonances at 133.1, 134.3, and

42.6 ppm, respectively. Based on these results we propose

that the reduction with LiEt^BH proceeds via hydride attack on the cyclopentadienyl ring of diag-CpRe(CO)£Br2 to yield

4 -diag-[ (n -CgHg)Re(CO) 28^] anion.

Additional NMR experiments were conducted in an attempt

. 4 to unequivocally demonstrate that diag-[ (r? -CgHg)Re(CO) gBrg]

anion resulted from hydride attack on the cyclopentadienyl

ring. The easiest experiment would have involved the

observation of NOE between the methylene protons and the

adjacent olefinic protons (H and H^'). Unfortunately, no

NOE was observed in the olefinic protons when the methylene

13

resonance at 2.90 ppm was irradiated. A selective C NOE

experiment was then performed using the resonance at 2.90

ppm to provide information concerning the adjacent olefinic

71

66.43,134.3 — \ J -

^•xoy 62.90,42.6

Hendo

Ha '— 66.53,133.1

0C--^Re^Br B r ^CO

Ft (PPM)

5.5 -

5.0 -

4.5

4.0

3.5 H

3.0

2.5 H i -

>40 "l30 120 no loo 90 BO 70 60 50~ 40~

F2 (PPM)

Figure 8. Heteronuclear chemical shift correlation spectrum of diag-[ (r74-c5H6)Re(CO)2Br2]" at -70 °C in dg-THF.

72

carbons. The 13C resonance at 133.1 ppm displayed a 15 %

enhancement in signal intensity relative to the resonance at

134.3 ppm. This indicates that the low-field resonance of

the proton AA'BB' spin system (6.53 ppm) is adjacent to the

methylene moiety. The relationship between the methylene

group and the remaining olefinic carbons (134.3 ppm) was

established by long-range *H-13C chemical shift correlation

spectroscopy using magnetization transfer delays optimized

for three-bond coupling (10 Hz).84 Figure 9 shows the long-

range coupling between the methylene protons and the

olefinic resonance at 134.3 ppm. These data collectively

support the existence of the ring-attacked anionic species,

diag-[ (r^-CgHgJRefCO^B^]". At this point we cannot rule

out polyene fluxionality (i.e., rotation about the rhenium)

in this complex, but we do believe that the initial diagonal

stereochemistry is maintained based on FT-IR spectral

comparison with the ring-attacked product from

lat-CpRe(CO).

The stereochemistry of addition of hydride or deuteride

to metal-bound rings has been established well by Bird and

Churchill in 1967.®® The examination of (r^-CgMegH) Re (CO) 3

obtained by hydride attack on (hexamethylbenzene)rhenium

tricarbonyl cation, does not display a low-frequency (2790

cm *) endo C-H stretch. Thus, the absence of a

73

^B^rsss* 66.43,134.3 —> Y »

H«xo^

OC Br / V

62.90,42.6

Hendo

HA '— 66.53,133.1

Br CO

FI 7.5

7.0

6.5 -

6.0

5.5

5.0

4.5 -

4.0

3.5

3.0 H

2.5

THF

THF

140 130 120 IfO 100 90 00 70 60 50 40

F2 (PPM)

Figure 9. Long-range heteronuclear chemical shift

correlation of diag-[ (J7*-CgHg)Re(CO) 2Br2]" at

-70 °C in dg-THF.

74

low-frequency C-H stretch in IR is taken as physical proof

of an exo hydride attack sequence. Unfortunately, we

couldn't observe this low-frequency C-H stretch by IR

spectroscopy. Alternatively, we have demonstrated the exo

addition of hydride or deuteride to the cyclopentadienyl

ring by isolation and CH NMR examination of the CpRe(CO)3

that decomposed from a reaction using LiEt^BD (vide infra).

G. Reduction of Diagonal and Lateral CpRe(CO)2Br2:

Ring Attack vs. Metal-Halogen Exchange

The reaction of diag-CpRe(CO) 2Br2 in THF at -78 °C with

various RLi reagents (1.1 eq.; R = Me, Ph, t-Bu) affords a

yellow colored solution whose infrared spectrum did not

correspond to that of CpRe(CO)2Br(R), the product of R"/Br"

34

exchange . Furthermore, the IR spectrum was inconsistent

with the presence of a formyl species and the ring-attacked

anionic species, diag-[ (rj4-C5H5R)Re(CO)2Br2]~. The IR

spectrum of the reaction exhibited two carbonyl stretching

bands at 1881 and 1806 cm"* which are consistent with the

[CpRe(CO)2Br]", the product of metal-halogen exchange.

Equation 17 gives the reaction conditions used to prepare

this compound and Figure 10 shows the resulting low-

temperature IR spectrum.

75

OC Re

Br / % CO

1900 1800 v, cm -i

Figure 10. Infrared spectrum of the carbonyl region for [CpRe(CO)2Br]" at -70 °C in THF.

76

THF + RLi . | + RBr + Li+ (17)

>Re^-Br "7» °c 0C->R« Br* ^CO Br ^CO

The identity of this new complex was also ascertained

using NMR spectroscopy. A dg-THF solution of [CpRe(CO)2Br]"

displays a single *H NMR resonance at 4.87 ppm for the

13 1

cyclopentadienyl ring. A C{ H) NMR spectrum reveals two

resonances at 87.2 and 214.3 ppm which correspond to the

cyclopentadienyl carbons and carbonyl groups, respectively.

A THF solution of [CpRe (CO) 2 Br]~ was next treated with

15-Crown-5 and methanol solution of PPP-C1 to give crystals

of [CpRe(CO)2Br][Li/15-Crown-5] and [CpRe(CO)2Br][PPP],

respectively. The latter crystals were characterized by

combustion analysis and the former crystals were employed in

the X-ray structural characterization. Complete details on

the X-ray crystal structure of [CpRefCO^Br] [Li/15-Crown- 5]

will be discussed (vide infra).

Interesting enough, the reaction of lat-CpRe(CO)£Br2 in

THF at -78 °C with various reducing agents (1.1 mole

equivalent; LiEt^BH, RLi; R = Me, Ph) gives both the product

of metal-halogen exchange, [CpRefCOJ Br]" and that of ring

attack, diag-[ (i? -CgH5R)Re(CO)2Br2]". The lateral isomer

reacts exclusively with t-BuLi to afford the [CpRe(CO)2Br]"

77

anion. The relative amounts of these anionic species are

dependent upon the nature of the reducing agent as shown in

Table 5.

The mixture of products obtained when lat-CpRe(CO)gBrg

was used was easily established by IR spectroscopy. The

low-temperature FT-IR spectrum exhibited five carbonyl

stretching bands. The two carbonyl stretching bands at 1882

and 1806 cm"* are readily assigned to the [CpRe(CO)2Br]"

anion, while the other three carbonyl stretching bands at

1938, 1906, and 1833 cm"* are similar to

lat-[ (t^-CjHgJTtefCO^Brg]" and consistent with the

lat-[ (n^-CgHgRJRefCOJgB^]" anion (Fig. 11). These three

carbonyl stretching bands are replaced by two new carbonyl

bands upon addition of 5 equivalents of 15-Crown-5 (Fig.

12). These results support the formation of an anionic

cyclopentadienylrhenium complex that displays extensive

carbonyl oxygen-lithium ion pairing in the absence of

lS-Crown-S^®'^ The identity of ring-attacked anionic

species from the lateral isomer was also characterized by

low-temperature NMR spectroscopy. When a dg-THF solution of

lat-CpRe (CO) 2^rz w a s treated with LiEtjBH at -78 °C three

new resonances appeared at 6.44, 6.34, and 2.86 ppm with an

integral ratio of 2:2:2. In comparison to

4 C (n -C5H5) Re (CO) 2 Br 2] from the diagonal isomer, the low-

78

tr 1 c S3 •H <w» 0 3 <D *d tn 0) c « <d

,a A D •p X •H w !*

w c

01 a) 0 o> X 0 0) rH iH <d ft B 6 1 0 rH u <d

• P s 9 X •H c •

0) (0 s

>

H < >t •

c a) w •H »d M (0 0 • p <d c • P 0 - P a < A H 01 a C & •H

PS 0 • •

•H co C - P 0 c •H <D C Cn < «<

in M 9 CO < E-»

as W 0) •H O a> ft CA o •H c o •H

c o •H •P •H

O a

CO • P §

S S •H 0 3 *0 «

CO t * G 3 O

f o

>« «

a o 00 r* i

£ A ro

«P W •H J

•H A

CO

CM M m CM

o a <w a) «

8*

<d

in B vo

a

<*> < * > <K> o o O CO in Km* <w w s £ x

•h * **

< # > <#> <#> o O O r* VO in >w *w» < «< «« • • •

Pi Pi «

i4 « I

«P

~ J: *

as (XI

CO -p W •H •J

•H CO

e

CM M PQ

CM «*"•«%

O 0 a> Pi » 1

• p

•H -H ^ d in 3 E tt u> | o i

o o

0 1

-P <d

•P •H CO G •H *0 a) u 0 co <d CD s co (d !* co 0) •H O a) a CO 0 •H c •

0 >i •H ft C 0 <d 0 <d

CO <w 0 0 k -p 'tf 0 H 0 0) ft •H CO >i

Pi Pi -P H q 1 <D D a) N P4 X5

79

H

L O C - ^ R e ^ - B r

OCT Br

Me

xd" O C - ^ R e ^ * B r

OCT Br

Ph

O C - ^ R e C * B r

O r Br

2200 2000 1800

v$ cm •i

Figure 11. Infrared spectra of the carbonyl region for (a) lat-[(f7j-C5H6)Re(CO)2Br2]-(b) lat-[(n*-C5H5Me)Re(CO)2Br2]" (c) lat-[ (r?4-C5H5Ph)Re(CO)2Br2]" All spectra were recorded at -70 °C in THF.

80

H

d " "

+ 0C--Li •—«0C

.Re-.. Br Br

H

*5" 0C->Rew*Br OCT Br

2200 2000 1800

v, cm •i

Figure 12. Infrared spectra of the carbonyl region for (a) lat-[(n4-C5H6)Re(CO)2Br2] [Li] (b) lat-[ (n4C5Hg)Re(CO)2Br2] [Li] with 5 equiv. of 15-Crown-5. Both spectra were recorded at -70 0 in THF.

81

field resonances associated with the olefinic hydrogens are

0.09 ppm upfield while the high-field resonance associated

with a methylene moiety is 0.04 ppm upfield. Reaction of

the Grignard reagent MeMgBr with the CpRefCO^B^ (either

isomer) gives [CpRe(CO)2Br]" as a result of metal-halogen

exchange. A slight excess of MeMgBr gives 50 % conversion

of [CpRefCOJgBr]" at -78 °C as determined by low-temperature

FT-IR analysis. Warming the solution to room temperature

gave the complete conversion of [CpRe(CO)2Br]" which has

been characterized in situ by IR spectroscopy.

[CpRe(CO)2Br][MgBr] exhibited carbonyl bands at 1890 and

1757 cm"*. These frequencies are different from those

observed with [CpRe(CO)2Br][Li] (vide supra). In comparison

to [CpRe(CO)2Br] [Li], the symmetric CO stretching band of

[CpRe(CO)2Br][MgBr} is shifted 9 cm"* to higher frequency

while the asymmetric CO stretching band is shifted 49 cm"*

to lower frequency (Fig. 13). This behavior is typical of a

carbonyl oxygen-MgBr+ contact ion pair where jr-electron

density is polarized towards the carbonyl oxygen involved in

ion pairing with the MgBr+ gegencation as shown in Equation

18. This interaction is expected to lower the CO stretching

frequency of the CO group involved in ion pairing with

concomitant stretching of the remaining CO group.®®

82

1950 1750

v, cm

Figure 13. Infrared spectra of the carbonyl region for (a) [CpRe(CO)2Br][Li] and (b) [CpRe(CO)2Br][MgBr] Both spectra were recorded at -70 °C in THF.

83

MeMgBr | or J I (18)

O C - ^ - B r OC> R*CBr ™P/-7. °C J _

6 ^ C0 OC Br ° B V Vo...i!i9Br

H. X-Ray Crystallographic Structure of

[CpRe(CO)2Br][Li/15-Crown-5]

The structure of [CpRe(CO)2Br][Li/15-Crown-5] has been

confirmed by single-crystal X-ray diffraction analysis. A

THF solution of 15-Crown-5 was added to the solution of

[CpRefCOJgBr]" from the reaction of diag-CpRe(CO)2^2 a n d

t-BuLi at room temperature. The yellow crystals suitable

for X-ray diffraction analysis were grown from the reaction

solution that had been layered with n-heptane. The X-ray

data collection and processing parameters for

[CpRe(CO)2Br][Li/15-Crown-5] are given in Table 6 and the

final fractional coordinates are listed in Table 7. In the

solid state, [CpRe(CO)£Br][Li/15-Crown-5] exists as discrete

molecules which are held together by Van der Waals forces.

The ORTEP diagram in Figure 14 shows the molecular structure

and clearly establishes the six-coordinate geometry about

the rhenium, assuming the cyclopentadienyl ring functions as

a three-coordinate ligand, and an overall geometry that is

TABLE 6. X-Ray Crystallographic and Data Processing Parameters for [CpRe(CO)2Br][Li/15-Crown-5]

84

space group

cell constants

a , A

b , A

c, A

P, A

v , A3

mol. formula

fw

formula units per cell(z)

p, g cm'1

abs. coeff (/*), cm"1

radiation (X), A

data collection method

collection range, deg

total data collected

independent data, I > 3a(I)

total variables

R

Rw

Weights W

P2 j/monoclinic

10.860(4)

13.116(5)

7.417(3)

105.26(3)

1018.7(3)

CiyHggBrLiO^Re

614.44

2

2.003

80.20

Mo Ka - 0.71073

8 - 2$

3.0° <29 < 50.0°

3782

2884

219

0.0777

0.0913

"„2

[accounting + (0.04 fft )^]

85

TABLE 7. Table of Positional Parameters and Their Estimated Standard Deviations for [CpRe(CO)2Br][Li/15-Crown-5]

Atom B(A2)

Re 0.14669(9) 0.000 0.0824(1) 2.54(1)

Br 0.1415(4) 0.1473(3) 0.3157(5) 4.97(8)

01 0.430(2) 0.000(4) 0.119(3) 6.5(5)

02 -0.292(2) -0.001(4) 0.357(3) 7.9(6)

03 0.722(3) 0.145(3) 0.083(4) 8.9(9)

04 0.537(3) 0.096(4) -0.210(4) 12(1)

05 0.479(4) 0.391(3) 0.226(4) 18(1)

06 -0.315(3) -0.134(2) 0.099(3) 5.7(60

07 0.168(2) -0.154(1) 0.325(3) 2.0(4)*

CI 0.321(20 0.011(3) 0.111(3) 3.0(5)

C2 0.100(2) 0.055(3) -0.224(3) 4.3(8)*

C3 -0.007(3) 0.085(2) -0.134(3) 2.7(6)

C4 -0.067(2) -0.009(5) -0.088(3) 4.3(6)

C5 -0.006(4) -0.099(4) -0.140(6) 8(1)

C6 0.072(3) -0.061(2) -0.217(4) 4.0(6)*

C7 0.162(4) -0.114(3) 0.279(5) 5.4(8)*

C21 0.738(6) 0.101(5) 0.402(9) 11(2)*

C22 0.771(6) 0.163(4) 0.285(5) 14(2)

C31 0.676(4) 0.221(3) -0.050(6) 10(1)

86

Table 7. continued

C32 0.608(5) 0.182(4) -0.182(8) 10(1)*

C41 0.553(4) 0.532(4) 0.361(5) 11(1)

C42 0.538(6) 0.427(5) 0.352(7) 12(2)

C51 0.419(7) 0.324(3) 0.231(5) 9(2)

C52 0.331(6) 0.290(3) 0.027(6) 7(1)

C61 -0.229(4) -0.163(3) 0.270(5) 5.6(9)

C62 -0.180(4) -0.062(4) 0.378(5) 8(1)

Li 0.600(4) 0.012(4) 0.048(5) 3.5(9)

Standard atoms were refined isotropically. Anisotropically refined atoms are given in the forms of the isotropic equivalent thermal parameter defined as: (4/3) * [a^*B(l,l) + b^*B(2,2) + c^*B(3,3) + ab(cos gamma) *B(1,2) + ac(cos beta)*B(1,3) + be(cos alpha)*B(2,3)]

87

Figure 14. Perspective view (ORTEP plot) of [CpRe(CO) 2Br] [Li/15-Crown-5] showing the atom labeling (hydrogen atoms are omitted for clarity).

88

typical of three-legged piano-stool complexes.®^

Bond distances and angles are given in Tables 8 and 9,

respectively. The 15-Crown-5 macrocycle adopts an

approximately planar arrangement of oxygen atoms in which

the lithium ion is centrally located. The five oxygen atoms

alternate above and below the plane of the best fit. The

coordination geometry about the lithium ion is that of a

slightly distorted pentagonal pyramid with the oxygen atoms

of carbonyl ligand. The two Li-0 distances in the

macrocycle complex are 2.17 and 2.19 A. The Li-0 distances

to the carbonyl ligand is shorter by average 0.12 A than

those in the macrocycle complex. The top view (Fig. 15) of

this molecule reveals an approximate mirror plane of

symmetry that is defined by the plane formed by the Re-Br

and C(3)-H bonds. This plane bisects the two carbonyl

groups and C(5)-C(6) bond of the Cp ring. The Re-Br and Re-

C(7) bond lengths are unexceptional in comparison to other

cycjLopentadienylrhenium complexes.32,44,88 T h e C(7j _0(7)

bond length is explainably short due to the poor crystal

quality and high thermal motion associated with this

molecule. The cyclopentadienyl ring is planar and regularly

delpcalized. The C(5) and C(6) carbon atoms of the Cp rings

are most nearly perpendicular to the plane defined by the

bromo-carbon bond. The C-C bond lengths of the Cp ring are

in the range of 1.26(6)-1.56(6) A with an average length

TABLE 8. Bond Distances in Angstroms for [CpRe(CO)2Br][Li/15-Crown-5]

a

89

Atoml Atom2 Distance Atoml Atom2 Distance

Re Br 2.581(4) 04 Li 2.17(6)

Re CI 1.85(2) 05 C42 1.09(6)

Re C2 2.32(3) 05 C51 1.11(7)

Re C3 2.27(3) 06 C61 1.41(4)

Re C4 2.33(2) 07 C7 0.62(4)

Re C5 2.40(4) C2 C3 1.52(4)

Re C6 2.29(3) C2 C6 1.56(5)

Re C7 2.08(4) C3 C4 1.50(6)

01 CI 1.18(3) C4 C5 1.43(7)

01 Li 2.06(5) C5 C6 1.26(6)

02 C62 1.42(5) C21 C22 1.34(9)

03 C22 1.47(5) C31 C32 1.18(6)

03 C31 1.40(5) C41 C42 1.39(8)

03 Li 2.19(7) C51 C52 1.62(6)

04 C32 1.37(7) C61 C62 1.56(6)

aNumbers in parentheses are estimated standard deviations in the least significant digits.

TABLE 9. Bond Angles in Degrees for

[CpRe(CO)gBr][Li/15-Crown-5]

90

Atoml Atom2 Atom3 Angle Atoml Atom2 Atom3 Angle

Br Re CI 93(1) C31 03 Li 114(3)

Br Re C2 112.3(9) C32 04 Li 105(3)

Br Re C3 88.5(7) C42 05 C51 123(5)

Br Re C4 103(1) Re CI 01 168(4)

Br Re C5 137(1) Re C2 C3 69(1)

Br Re C6 146.8(8) Re C2 C6 69(2)

Br Re C7 95(1) C3 C2 C6 92(2)

CI Re C2 92(1) Re C3 C2 72(2)

CI Re C3 125(1) Re C3 C4 73(1)

CI Re C4 155.0(9) C2 C3 C4 109(3)

CI Re C5 130(1) Re C4 C3 69(2)

CI Re C6 104(1) Re C4 C5 75(2)

CI Re C7 96(2) C3 C4 C5 111(3)

C2 Re C3 39(1) Re C5 C4 70(2)

C2 Re C4 64(1) Re C5 C6 70(2)

C2 Re C5 65(1) C4 C5 C6 101(4)

C2 Re C6 40(1) Re C6 C2 71(1)

C2 Re C7 151(1) Re C6 C5 79(2)

C3 Re C4 38(2) C2 C6 C5 126(4)

C3 Re C5 62(1) Re C7 07 169(5)

91

Table 9. continued

C3 Re C6 58(1) 03 C22 C21 119(5)

C3 Re C7 139(1) 03 C31 C32 107(4)

C4 Re C5 35(2) 04 C32 C31 134(5)

C4 Re C6 54(1) 05 C42 C41 120(5)

C4 Re C7 102(2) 05 C51 C52 114(4)

C5 Re C6 31(1) 06 C61 C62 107(3)

C5 Re C7 90(2) 02 C62 C61 103(3)

C6 Re C7 112(1) 01 Li 03 127(3)

CI 01 Li 159(3) 01 Li 04 101(2)

C22 03 C31 125(4) 03 Li 04 74(2)

C22 03 Li 107(3)

92

°C C2-2.32

C3-2.27

/ C4— 2.33

Figure 15. Top view (ORTEP plot) of [CpRe(CO)2Br]" showing the Cp ring-rhenium bond lengths.

93

of 1.45(4) A. The C(5)-C(6) bond length is short in

comparison to the other four C-C bond lengths of the Cp ring

and to other cyclopentadienylrhenium complexes.®®

The Re-C(ring) bond lengths range from 2.27 A to 2.40 A

which are a little longer in comparison to other penta-

methylcyclopentadienylrhenium compounds,^ suggesting

less ff-back-bonding from the rhenium metal. The disparate

Re-C(ring) bond lengths indicate that the Cp ring is tilted

away from the Li/15—Crown-5 moiety and towards the bromine

and carbonyl (C7)0. The Re-C(5) bond is the longest at

2.40(4) A and is followed by the Re-C(4) and Re-C(2) bond

lengths of 2.33(2) and 2.32(3) A, respectively. The

shortest Re-C(ring) bond lengths of 2.27(3) and 2.29(3) A

belong to the Re-C(3) and Re-C(6), respectively. These

variations in the Re-C(ring) bond lengths are not entirely

unexpected since a bulky group of Li/15-Crown-5 moiety is

bonded to 01 carbonyl oxygen atom. However, in

[Cp*Re(C0)2Br] the Re-C(ring) bond that is pseudotrans

(formally opposite) to the bromo ligand is observed to be

longer than the corresponding Re-C(ring) bonds that are

opposite to the two carbonyl ligands. This result is

readily rationalized in terms of the w-acceptor properties

of the CO ligands. Therefore, the bond elongation of

Re-C(5) indicates that this is due to steric rather than

electronic factors associated with the Cp ring and the

94

rhenium atom.

I. Reactivity and Stability Studies of Anionic

Cyclopentadienyl Complexes

The functionalization of [CpRe(CO)2Br]" anion was

examined as a potential route to complexes of the form

CpRe(CO)2Br(R). Scheme 2 shows the reactivity of

(CpRe(CO)2Br] with alkylating reagents. Treatment of

[CpRe(CO)2Br] in THF with trifluoroacetic acid afforded two

new VCO bands at 2032 (s) and 1962 (vs) cm"1 as expected for

anion protonation. From the intensity pattern of the vCO

bands it may concluded that protonation proceeds to give

CpRe(CO)2Br(H) with diagonal stereochemistry. When a dg-THF

solution of [CpRe(CO) 2Br]" was treated with CF3C02H two new

*H NMR resonances appear at 6.06 and -9.03 ppm with an

integral ratio of 5:1. Furthermore, when the solution of

CpRe(CO)2Br(H) was treated with MeLi deprotonation proceeds

to regenerate [CpRe(CO)2Br] . Our experimental observations

are consistent with the theoretical predictions of Bursten

40

et al. [CpRe(CO)2Br]" also reacts with methyl triflate

and magic ethyl to yield the corresponding

diag-CpRe(CO)2Br(R) complexes based on in situ IR analysis

[VCO for diag-CpRe (CO) 2Br (Me) in THF s 2028, 1954 cm"1 and

95

"OS

ffl

as tf">

25 r\T m o <N

# o v

I /

g o

2 .o •u

\\ 8 ®

rsi vo <y>

O CM

<n 0 erf I

z ir>

1

K z

CI

4

43 o CO

i'

% 9

<N

96

VCO for diag-CpRe(CO)2Br(C2H5) in THF : 2032, 1959 cm"1].

All of these functionalized complexes decomposed in solution

over a period of days to give CpRefCO)^ as the only isolable

product (approximately 20-30 % isolated yield).

The reaction of [CpRe(CO)2Br]" with two equivalents of

Bu^SnH at 60 °C leads to the formation of the

bis(tributyltin)rhenium complex (Eq. 19), which has also

2 BUgSnH

Re " ~ 7 T Z _ Re „ + 2 B u 3 S n B r <19> OCyr 60 °C 0C-^P*^«SnBu3 Bf ^ C 0 Bu3Sn^ V o 3

been synthesized independently by deprotonation of

diag-CpRe(CO)2H2 with Et^N in the presence of Bu^SnBr at

room temperature (vide infra). IR spectrum shows two new

VCO bands at 1936 (s) and 1880 (vs) cm"1 in THF (Fig. 16) .

THF solutions of [CpRe(CO)2Br]~ were observed to be stable

when oxygen was rigorously excluded. The presence of oxygen

has been observed to lead to complex mixtures of rhenium

oxides in addition to CpRe(CO)^. For example, the

[CpRe(CO)2Br]" anion slowly decomposed in septum-capped

vessels on the benchtop to give the oxide ReO^" as the

major products. The i/Re=0 band observed was identical to

97

£ O

vo H

0) u

•H

98

that exhibited by pure KReO^. IR analysis (KBr pellet)

revealed an intense vRe=0 band at 913 cm"* attributed to

potassium perrhenate. Prismatic crystals of

[ReO^][Li/15-Crown-5] have been isolated from decomposed

solutions of [CpRe(CO)2Br][Li/15-Crown-5], and the structure

of [ReO^][Li/15-Crown-5] has been confirmed by single-

crystal X-ray diffraction analysis. The X-ray data

collection and processing parameters are given in Table 10

and the final fractional coordinates are listed in Table 11.

The ORTEP diagram in Figure 17 shows the molecular structure

and clearly establishes the slightly distorted tetrahedral

geometry about the rhenium. Selected bond distances and

angles are given in Tables 12-13. The 15-Crown-5 macrocycle

adopts an approximately planar arrangement of oxygen atoms

in which the lithium ion is centrally located. The five

oxygen atoms alternate above and below the plane of the best

fit. The Li-0(6) bond distance is 2.00(3) A which is

shorter than the value of the K-0 bond in KReO^.90 The Re-0

bond distances and O-Re-O bond angles are unexceptional in

comparison to potassium perrhenate®^ except Re-0(9) bond

distance, 1.61(3) A, and 0(7)-Re-0(8) angle, 105.7(8)°.

Finally, the exact pathways and other products involved in

this decomposition reaction remain as unanswered questions

for further study.

99

TABLE 10. X-Ray Crystallographic and Data Processing Parameters for [Li/15-Crown-5][ReO^]

space group

a, A

b, A

c , A

crystal size, mm

V. A3

mol. formula

fw

formula units per cell(z)

p, g cm"*

abs. coeff (/*), cm"*

radiation (A), A

data collection method

collection range, deg

total data collected

independent data, I > 3a(I)

R

RW

Weights

P 2 | 2 j 2 j

9 . 2 6 9 ( 2 )

1 2 . 6 7 8 ( 3 )

1 3 . 5 7 4 ( 2 )

0 . 2 0 x 0 . 2 5 x 0 . 3 0

1 5 9 5 . 2 ( 2 )

C10H20LiO9Re

4 7 7 . 4 1

4

1 . 9 8 8

7 7 . 5

Mo Ka = 0 . 7 1 0 7 3

9-29

3 . 0 ° < 29 < 5 5 . 0 °

2108

1357

0 . 0 4 7

0 . 0 5 9

W = 4 FQ9 / 9c(Fq9)

100

TABLE 11. Table of Positional Parameters and Their Estimated Standard Deviations for [Li/15-Crown-5][ReO^]

Atoms X y z B(A2)

Re 0.15578(7) 0.14846(5) 0.11018(5) 5.03(1)

01 0.618(5) -0.063(3) 0.177(3) 28(2)*

02 0.556(3) -0.026(2) -0.006(1) 13.1(6)*

03 0.287(2) -0.046(1) -0.060(1) 11.3(5)*

04 0.207(3) -0.198(2) 0.059(2) 13.2(6)*

05 0.361(2) -0.166(1) 0.234(1) 11.0(5)*

06 0.286(1) 0.0616(9) 0.1469(8) 6.6(3)

07 0.216(2) 0.214(1) 0.010(1) 11.3(5)

08 0.136(2) 0.241(1) 0.2002(9) 9.3(4)

09 0.011(3) 0.084(2) 0.087(2) 17.1(9)*

Cll 0.667(5) -0.012(4) 0.138(3) 17(2)*

C12 0.683(4) 0.018(30 0.042(2) 14(1)*

C21 0.510(3) 0.058(2) -0.064(2) 10.8(8)*

C22 0.383(3) -0.005(2) -0.122(2) 8.8(6)*

C31 0.210(3) -0.122(2) -0.091(2) 9.0(6)*

C32 0.117(4) -0.160(2) -0.007(2) 12.5(9)*

C41 0.134(3) -0.217(2) 0.155(2) 10.5(7)*

C42 0.271(3) -0.257(2) 0.229(2) 9.8(7)*

C51 0.463(3) -0.194(2) 0.289(2) 10.5(8)*

C52 0.554(5) -0.094(3) 0.274(3) 17(1)*

101

Table 11. continued

Li 0.382(3) -0.073(2) 0.107(2) 6.1(6)*

Starred atoms were refined isotropically. Anisotropically refined atoms are given in the form of the isotropic equivalent displacement parameter defined as: (4/3) * [a2*B(l,l) + b2*B(2,2) + c**B(3,3) + ab(cos gamma) *B(1,2) + ac(cos beta)*B(1,3) + bc(cos alpha)*B(2,3)]

102

Figure 17. Perspective view (ORTEP plot) of [Li/15-Crown-5][ReO^] showing the atom labeling (hydrogen atoms are omitted for clarity).

TABLE 12. Bond Distances in Angstroms for [Li/15-Crown-5][ReO^]a

103

Atoml Atom2 distance Atoml Atom2 Distance

Re 06 1.71(1) 05 C42 1.42(3)

Re 07 1.69(2) 05 C51 1.25(3)

Re 08 1.71(1) 05 Li 2.10(4)

Re 09 1.61(3) 06 Li 2.00(3)

01 Cll 0.95(6) Cll C12 1.37(5)

01 C52 1.50(6) C21 C22 1.63(4)

02 C12 1.45(4) C31 C32 1.51(4)

02 C21 1.40(4) C41 C42 1.70(4)

02 Li 2.30(4) C51 C52 1.53(5)

03 C22 1.33(3) 01 Li 2.38(5)

03 C31 1.28(3) 03 Li 2.46(4)

04 C32 1.31(4) 04 Li 2.37(4)

04 C41 1.49(4)

aNumbers in parentheses are estimated standard deviations in the least significant digits.

104

Table 13. Bond Angles in Degrees for [Li/15-Crown-5][ReO^]

Atoml Atom2 Atom3 Angle Atoml Atom2 Atom3 Angle

06 Re 07 108.7(8) 05 C42 C41 103(2)

06 Re 08 108.2(7) 05 C51 C52 96(2)

06 Re 09 109(1) 01 Li 02 124(3)

07 Re 08 105.7(8) 01 Li 02 67(1)

07 Re 09 112(1) 01 Li 03 133(2)

08 Re 09 114(1) 01 Li 04 141(2)

Cll 01 C52 149(5) 01 Li 05 78(2)

C12 02 C21 102(2) 01 Li 06 105(2)

C22 03 C31 117(2) 02 Li 03 67(1)

C32 04 C41 112(2) 02 Li 04 118(2)

C42 05 C51 104(2) 02 Li 05 139(2)

Re 06 LI 142(1) 02 Li 06 106(1)

01 Cll C12 140(5) 03 Li 04 66(1)

02 C12 Cll 104(3) 03 Li 05 143(2)

02 C21 C22 96(2) 03 Li 06 88(1)

03 C22 C21 112(2) 04 Li 05 77(1)

03 C31 C32 108(2) 04 Li 06 110(1)

04 C32 C31 106(3) 05 Li 06 103(1)

04 C41 C42 103(2)

105

The ring-attacked product diag-[ (»7^-CgHg)Re(CO) '

(3) was next examined in the presence of donor ligands.

When THF solutions of 3, from the reaction of

diag-CpRefCO^B^ with LiEt^BH at -78 °C, were treated with

two electron donor ligands only CpRefCO^L (where L = PPhj,

P(OPh)3, Me^CNC) was observed (Eq. 20).

H

| THF

or-"R«~-Rr + L o 0C^-R

« • V B " 7 8 C csc+ Br ^C0 oc

• ^ L (20)

(L S PPh3, P(OPh)3, Me3CNC)

Reaction of 3 in THF at -78 °C with PPhj was instantaneous

and two vCO bands at 1926 (vs) and 1844 (vs) cm'* were

33 Q1 observed as expected for the known CpRe(CO)g(PPhj).

Complex 3 also react with P(OPh)3 and Me^CNC to yield the

corresponding CpRefCO^L complexes based on in situ IR

analysis [yCO for CpRe(CO)2[P(OPh)3] in THF : 1961, 1893

cm"1 and vCO for CpRe(CO)2(Me3CNC) in THF : 1902, 1850 cm"1]

Although the exact mechanism associated with this

substitution is unknown at this time, we suggest the

mechanism shown in Scheme 3.

SchttM 3

106

H(D) H

d " I -

0c--Re--Br Br ^co

m

dtH l_

oc'^R,eCBr sr I co

L 4

H(D)

0C 0C< ..-Re,

H(D)

0C->ReC:-L H ^CO

7 ox*

-Br

h(d; H

d ~ i

Oc- Re--L Br ^CO

-Br

H(D)

di* oc -Rv CO

C-H bond activation

107

The first step involves ligand attack at the rhenium center

with concomitant cyclopentadienyl ring slippage from r?4 -*• *?2

coordination. The resulting 18-electron complex 4 may then

lose bromide and reestablish the rj^-cyclopentadienyl ring.

Complex 5 is structurally similar to the aforementioned

r? -cyclopentadienyl complex 3. Further loss of bromide in 5

leads to the three-legged piano-stool complex 6 which is

coordinatively unsaturated (16-electron) and expected to be

quite reactive. We believe that an intramolecular C-H bond

activation occurs which regenerates, the ^-coordination of

cyclopentadienyl ring in complex 7. Such a C-H bond

activation scheme is supported by the fact that 6 is

isolobal to known organometallic compounds which exhibit

similar reactivity as expected for a carbene fragment (i.e.

6 19 d -MLg • q - CH2). Loss of a proton from 7 would give the

observed ligand substitution product CpRefCO^L. Supporting

evidence for this last step derives from the work of Bursten

et al. who have calculated that d* four-legged piano-stool

complexes, such as 7, should function as efficient hydronium

ion donors.*® Furthermore, Norton et al. have measured the

thermodynamic acidities of similar complexes and have shown

them to possess moderate acidities (pKa » 20-25).®*

Implicit in this activation scheme is the transfer of the

A endo C-H bond of the r? -CgHg ring to the rhenium center.

108

Assuming exo nucleophile attack on the r? -Cp ring, the exo

hydride should remain associated with resulting 77 -Cp in 7.

When Et^BD" was used to introduce an exo deuterium into 3,

followed by treatment with L, the isolated CpRe(CO)2L was

observed to contain one deuterium per Cp ring. If exo C-

H(D) activation occurs upon •+ ring conversion, then no

deuterium is expected in the Cp ring of the product.

In order to further define the reactivity of 3, 3 was

treated with t-BuLi in THF at -78 °C. Two complexes were

formed of which one was assigned to that of [CpRe(CO)2Br]".

The other complex possessed two uCO bands at 1842 (s) and

1715 (s) cm * and is tentatively assigned as the dianionic

species, [ (J7*-CgHg)Re(CO)2Br]We suggest that t-BuLi

deprotonation proceeds bimodality. Deprotonation of the

4

Tj -CjHg ring yields the dianionic species, 8. Facile loss

of bromide from dianion 8 leads to [CpRe(CO)2Br]' in 65 %

yield (by IR). Alternatively, t-BuLi reacts by metal-

halogen exchange to give [ (rj4-C5Hg)Re(CO) 2Br] "2 in 35% yield

as shown in Scheme 4.

The thermal stability of 3 was next examined. 3 was

observed to be stable for a period of six months when

maintained at -78 °C. However, decomposition of 3 is

noticeable upon warming to -30 °C, being complete by 0 °C to

afford a green solution. FT-IR analysis revealed the

109

ur> VD

O

1/ X o ®

00 I

- 8

"Vw

u

X o

I x> u. X

co X

*

m

rsT GO

* u

X <ni

O

<T

3 A CI CO

X

: ^ J

9 8

a#

°\V

g o

Of

110

22 93

presence of only CpRe(CO)^ ' which we have been able to

routinely Isolate in 25-35 % yield. While the exact

mechanism associated with this decomposition is unknown at

this time, the observation of CpRe(CO)^ (in part) provides

insight into the stereochemistry attendant upon hydride 4

attack. If we assume that the endo hydrogen of the n c5h6

ring is transferred to the rhenium metal during

94 decomposition, then the exo hydrogen must become

incorporated into the cyclopentadienyl ring of CpRe(CO)^

2 (vide supra). This was demonstrated by isolation and H NMR

examination of the CpRe(CO)^ that formed a reaction using

o

LiEtjBD. The H NMR spectrum exhibited a singlet at 5.48

ppm consistent with a dj-cyclopentadienyl moiety.

Integration against added dg-benzene or the natural

abundance solvent resonances (THF) supports the presence of

one deuterium per Cp ring and an initial exo attack of the

reducing agent. Based on these results we propose that

decomposition of 3 proceeds as in the Scheme 5.

A THF solution containing the anionic products from

lat-CpRe(CO)2Br2, lat-[ (r?4-C5H5R)Re(CO)2Br2]" and

[CpRe(CO)2Br]~, decompose similarly to afford R-CpRe(CO)^

which we have been able to routinely isolate in 25-35 %

yield as illustrated in Equation 21. It is seen that the

R-CpRe(CO)2 obtained must come from the anionic

Ill

B w

r . * ac.

a o

0) 0

1 0 01 m

92

" v g m

" 2 c o - 2

- a o

I S I U

U O

CD l

X i

9

X o

^ 3 - < " •. u GD

HI

§

4 3 O CO

© s

. *

112

ring-attacked complex lat-[ (rj*-C5H5R)Re(CO) 38^]". This

result suggests that [CpRe(CO)2Br]" may be the carbonyl

source required for R-CpRe(CO)j formation.

THF

OC Lc/ > Br Br

UEt3BH or RLi

N

d " H • L

OC- Re --Br OCT Br

decomposition

oc->R#-o<r

| RsMe.Ph.H + LiR«04

'CO

>C0

(21)

113

J. Reduction of Diagonal and Lateral, Cp*Re(CO)2Br2 with

RLi, RMgX, and Trialkylborohydrides.

The reduction behavior of diagonal and lateral

CP*Re(CO)2Br2 was investigated with RLi reagents (R = Me,

Ph, and t-Bu) in an attempt to prepare the corresponding

alkylbromide complex, Cp*Re(CO)2Br(R). Related reduction

studies are well documented in the isoelectronic Cp*Rh and

95

Cp*Ir systems. However, in all cases only the anionic

complex, [Cp*Re(C0)2Br] ,2-, was observed when one equiva-

lent of RLi was reacted with either isomer of Cp*Re(CO)2Br2

(Equation 22).

THP Cp*Re(CO)2Br2 + RLi [Cp*R«(CO)2Br] [Li] + RBr (22)

-78 C

(R •» Ni| Ph, t-Bu)

Red-brown solutions of Cp*Re(CO)2Br2 (either isomer) react

instantaneously with added RLi at -78 °C in THF to give a

yellow solution containing 2-; the essentially quantitative

conversion to 2— was easily confirmed by low—temperature

FT-IR measurements. The IR spectrum exhibited two carbonyl

stretching bands at 1863 and 1789 cm"1 which is consistent

with the ascription of 2- as an anionic three-legged piano-

114

stool complex. Such a low energy shifting of the CO

stretching bands of 2- relative to Cp*Re (CO) 2Br2 *-s directly

attributed to the increased charge density at the rhenium

center (Fig. 18). The higher frequency CO band at 1863 cm~*

is readily assigned to a symmetric CO stretching mode while

the lower frequency CO band at 1789 cm"* corresponds to the

asymmetric CO mode based on group theoretical considera-

96 tions. In such a molecule more energy is required to

stretch two CO bonds simultaneously than to execute one CO

stretch and one CO contraction. IR studies (in THF)

indicate that 2- exists as symmetrically solvated ion pairs

75 97

possessing idealized C$ symmetry. This is supported by

the spectral invariance of 2- over the temperature range

-78 °C to room temperature which serves to rule out a

dynamic equilibration between different ion pairs.®*'98

Furthermore, added Li+ (as CFjSOjLi), tetraphenylphospho-

nium chloride (PPP-C1), or 15-Crown-5 do not affect the

intensity or frequency of the initial IR spectrum of 2-,

reinforcing the existence of only solvent-separated ion

pairs in THF solution. 2- can be easily isolated in high

yield by treating [Cp*Re(CO) 2Br] [Li] with PPP-C1. The

resulting orange-yellow crystals of [Cp*Re(CO)2Br][PPP] gave

a satisfactory microanalysis and were used in the NMR

characterization of 2-. The *H-NMR spectrum of

115

oc Br

.-R

t r i 1

1900 1800 v, cm -1

Figure 18. Infrared spectrum of the carbonyl region for [Cp*Re(CO)2Br]" at -70 °C in THF.

116

2-PPP in CDgClg displayed a singlet at 1.94 ppm (15H) and a

multiplet centered at 7.75 ppm (20H) consistent with the

methyl groups of the Cp* ring and the aromatic protons of

the PPP gegenion, respectively. The 13C{1H} NMR spectrum

exhibited resonances at 11.0, 93.8, and 213.1 ppm for the

methyl, carbons of Cp ring, and the carbonyl carbons of 2-

PPP as shown in Figure 19. The metal carbonyl resonance is

shifted to downfield relative to either of the lateral and

diagonal isomers of Cp*Re(C0)2Br2 as a result of increased

charge density at the rhenium center.66,74,99 Reaction of

the Grignard reagents MeMgBr or t-BuMgCl with the isomeric

dibromides of Cp*Re(C0)2Br2 gives 2-MgBr as a result of

metal-halogen exchange like that observed with CpRe(CO)2Br2.

Use of MeMgBr (slight excess) affords 2-MgBr sluggishly at

-78 °C as determined by low-temperature FT-IR analysis (10%

conversion after 30 min). Warming the solution to room

temperature led to the rapid generation of 2-MgBr (Eq. 23)

which has been characterized in situ by IR spectroscopy

(Fig. 20b). 2-MgBr exhibits CO bands at 1872 and 1734 cm"1.

These frequencies are different from those observed with

2-Li (vide supra). In comparison to 2-Li, the symmetric CO

stretching band of 2-MgBr is shifted 9 cm"1 to higher

frequency while the asymmetric CO stretching band is shifted

55 cm"1 to lower frequency.

117

& CM r-4 o CM

Q a a •H

—g

a o in CM

04 04 04

U OQ CM <•—««»

O o w a) « * a a

0

1 .p o a) a w

a CO

© - © ru o\

H a) M 3 Cn

•H Cm

118

1950 1750

v, cm -i

Figure 20. Infrared spectra of the carbonyl region for (a) [Cp*Re(CO)2Br][Li] and (b) [Cp*Re(CO)2Br][MgBr], Both spectra were recorded at -70 °c in THF.

119

•RMgBr

,RBr OO^cO....;gBr (23)

(R:X s Me*Br, t-Bu:Cl)

This behavior is typical of a carbonyl oxygen-MgBr+ contact

ion pair where -electron density is polarized towards the

carbonyl oxygen involved in ion pairing with the MgBr+

gegencation as shown in Equation 23.7®>97 This interaction

is expected to lower the CO stretching frequency of the CO

group involved in ion pairing with concomitant strengthening

of the remaining CO group.®®

The reaction of lat-Cp*Re(CO)2Br2 with a stoichiomet-

ric amount (one equiv.) of t-BuMgCl is rapid at -78 °c and

gives 2- in near quantitative yield. The IR spectrum of

2 —MgCl is identical with that of 2-MgBr and suggests similar

carbonyl oxygen-MgX+ contact ion pairs. In this reaction

with t-BuMgCl the possibility exists for a bromide/chloride

exchange in the magnesium gegenion. The resulting MgBr*

cation would be expected to yield an IR spectrum of 2-MgBr

identical to that observed with MeMgBr. Unequivocal proof

for such carbonyl oxygen—MgX^ ion pairs was obtained by

examining the effect HMPA had on the solution IR spectrum of

120

2-MgX. Use of 10 mole equivalents of HMPA led to the

complete disruption of the contact ion pairs and products of

a two carbonyl band identical with 2-Li.

The reaction of lat-Cp*Re(C0)2Br2 with 1.0 mole equi-

valent of LiEtjBH or K-Selectride, [K(sec-Bu)3BH] at -78 °C

led to the immediate formation of 2- in 50 * yield as

determined by IR analysis. When the hydride concentration

was doubled, complete conversion to 2- was observed. At no

time was the corresponding bromohydride complex,

Cp*Re(CO)2Br(H), observed (vide infra). Based on these

observations we believe, in analogy to the RLi and RMgX

reactions, that the hydride reagents react with

lat-Cp*Re(CO)2Br2 to give equal molar amounts of 2- and HBr.

A subsequent, rapid reaction between the generated HBr and

excess R^BH would then account for the 1:2 rhenium/hydride

stoichiometry observed with this reaction. The hydrogen by-

product necessary for the proposed reaction was observed by

GC analysis; however, no attempt was made to quantify it.

An alternative path way involves the metathetical

replacement of a bromide ligand in lat-Cp*Re(CO)2Br2 to give

diag-Cp*Re(CO)2Br(H). Since most metal hydrides are known

to exhibit acidic character101, a fast follow-up reaction

involving metal-hydride deprotonation by additional H" would

also give 2- as outlined in Scheme 6. The IR spectra of 2-

121

m

CD

<N X

0

s A O co

m

CD •

• ** cr

O O

122

from these reactions are unexceptional when compared with

those obtained from the RLi reactions. This suggests that

the presence of R^B, an effective Lewis acid, in solution

does not perturb the local environment of the rhenium anion.

K. X-Ray Crystallographic Structure of [Cp*Re(CO)2Br][PPP]

The structure of 2-PPP has been confirmed by single-

crystal X-ray diffraction analysis. A methanol solution of

PPP-C1 was added to the solution of 2-Li from the reaction

of lat-Cp*Re(CO)2Br2 and LiEtjBH. The orange yellow

crystals suitable for X-ray analysis were grown from the

reaction solution that had been layered with n-heptane.

2-ppp exists as discrete molecules in the unit cell with no

unusually short inter- or intramolecular contacts? the PPP+

gegenion is unexception and requires no further description.

The X-ray data collection and processing parameters for

2-PPP are given in Table 14 and the final fractional coordi-

nates are listed in Table 15. The ORTEP diagrams in Figures

21-22 show the molecular structure of 2-PPP and clearly

establishes the six-coordinate geometry about the rhenium,

assuming the Cp* ring functions as a three-coordinate

ligand, and an overall geometry that is typical of three-

87

legged piano-stool complexes. Bond distances and angles

are given in Tables 16-17. The top view (Fig. 23b) of 2-

123

TABLE 14. X-Ray Crystallographic and Data Processing Parameters for [Cp*Re(CO)2Br][Ph^P].

space group

cell constants

a, A

b , A

c, A

V, A3

mol. formula

fw

formula units per cell (Z)

p, g cm"*

crystal size, mm

abs coeff (/*), cm"*

radiation (X), A

data cillection method

collection range, deg

total data collected

independent data, I > 3a(I)

total variables

R

RW

Weights W

Pbca/orthorhombic

20.646 (5)

17.690 (5)

17.555 (3)

6411.6

c36H35Br02PRe

796.76

8

1.651

0.3 x 0.2 x 0.1

51.55

Mo Ka - 0.71073

9 - 20

2.0° £ 2B < 50.0°

6233

2995

370

0.0366

0.0432

4 fj

("'counting + <0-04 fo2>2l

124

TABLE 15. Positional Parameters for Non-Hydrogen Atoms for [Cp*Re(CO)2Br][Ph4P] with Estimated Standard Deviations in Parenthesis

Atom X y z B(A2)

Re 0.64733(2) 0.10035(3) 0.61146(3) 2.904(7)

Br 0.74495(6) 0.03683(7) 0.68025(8) 4.26(3)

P 0.0909(1) 0.6040(2) 0.0984(2) 2.65(5)

01 0.5710(5) 0.1711(6) 0.7393(5) 5.7(2)

02 0.7146(5) 0.2484(6) 0.5771(6) 6.6(2)

CI 0.6035(7) 0.1453(7) 0.6965(7) 4.1(3)

C2 0.6927(6) 0.1953(7) 0.5915(7) 4.3(3)

C3 0.5981(6) 0.1021(8) 0.4961(7) 4.3(3)

C3 • 0.595(1) 0.1694(9) 0.4441(9) 7.9(4)

C4 0.5531(6) 0.0824(8) 0.5528(9) 6.1(3)

C4' 0.4887(8) 0.122(1) 0.573(1) 13.3(6)

C5 0.5714(6) 0.0079(8) 0.5848(7) 5.1(3)

C5' 0.5359(8) -0.042(1) 0.644(1) 10.7(5)

C6' 0.6648(9) -0.0888(8) 0.556(1) 7.8(5)

C6 0.6272(6) -0.0155(7) 0.5457(7) 3.6(3)

C7' 0.6994(7) 0.036(1) 0.4315(8) 6.7(4)

C7 0.6446(6) 0.0419(7) 0.4926(6) 3.8(2)

Cll 0.1250(0) 0.5125(6) 0.0820(6) 2.9(2)

C12 0.1398(6) 0.4654(7) 0.1452(7) 3.6(2)

C13 0.1675(6) 0.3928(7) 0.1327(7) 4.3(3)

T a b l e 1 5 . c o n t i n u e d

125

C14 0 . 6 7 7 0 ( 6 ) 0 . 1 3 3 0 ( 8 ) 0 . 9 4 4 7 ( 8 ) 4 . 5 ( 3

C15 0 . 1 6 0 3 ( 5 ) 0 . 4 1 6 3 ( 7 ) - 0 . 0 0 5 5 ( 7 ) 4 . 2 ( 3

C16 0 . 1 3 4 5 ( 6 ) 0 . 4 8 7 8 ( 6 ) 0 . 0 0 6 3 ( 7 ) 3 . 4 ( 2

C21 0 . 0 1 4 9 ( 5 ) 0 . 5 9 2 9 ( 6 ) 0 . 1 4 7 9 ( 6 ) 2 . 7 ( 2

C22 - 0 . 0 1 3 0 ( 6 ) 0 . 5 2 0 6 ( 7 ) 0 . 1 4 8 9 ( 7 ) 3 . 6 ( 3

C23 - 0 . 0 7 4 6 ( 6 ) 0 . 5 1 4 3 ( 7 ) 0 . 1 8 5 7 ( 7 ) 3 . 6 ( 3

C24 - 0 . 1 0 3 1 ( 6 ) 0 . 5 7 6 2 ( 8 ) 0 . 2 1 7 4 ( 7 ) 4 . 3 ( 3

C25 0 . 5 7 4 1 ( 7 ) 0 . 3 5 0 9 ( 7 ) 0 . 7 1 5 8 ( 8 ) 4 . 5 ( 3

C26 - 0 . 0 1 4 3 ( 6 ) 0 . 6 5 6 4 ( 7 ) 0 . 1 7 9 4 ( 7 ) 4 . 0 ( 3

C31 0 . 1 4 4 9 ( 5 ) 0 . 6 6 4 5 ( 6 ) 0 . 1 5 1 3 ( 6 ) 2 . 7 ( 2

C32 0 . 1 3 9 5 ( 6 ) 0 . 7 4 3 6 ( 7 ) 0 . 1 3 9 0 ( 8 ) 4 . 4 ( 3

C33 0 . 1 8 0 2 ( 7 ) 0 . 7 9 2 5 ( 8 ) 0 . 1 8 1 6 ( 8 ) 5 . 0 ( 3

C34 0 . 2 2 3 4 ( 6 ) 0 . 7 6 3 2 ( 8 ) 0 . 2 3 4 2 ( 7 ) 4 . 1 ( 3

C35 0 . 2 2 7 0 ( 6 ) 0 . 6 8 5 2 ( 8 ) 0 . 2 4 7 0 ( 7 ) 4 . 4 ( 3

C36 0 . 1 8 7 9 ( 6 ) 0 . 6 3 5 6 ( 7 ) 0 . 2 0 4 8 ( 7 ) 3 . 4 ( 3

C41 0 . 0 7 6 1 ( 5 ) 0 . 6 4 6 7 ( 6 ) 0 . 0 0 6 9 ( 6 ) 2 . 8 ( 2

C42 0 . 0 1 4 7 ( 6 ) 0 . 6 4 9 2 ( 7 ) - 0 . 0 2 2 3 ( 7 ) 3 . 6 ( 3

C43 0 . 0 0 6 1 ( 6 ) 0 . 6 7 6 1 ( 7 ) - 0 . 0 9 7 3 ( 7 ) 4 . 1 ( 3

C44 0 . 0 5 9 6 ( 7 ) 0 . 7 0 3 3 ( 8 ) - 0 . 1 3 7 3 ( 7 ) 4 . 9 ( 3

C45 0 . 6 1 9 4 ( 7 ) - 0 . 2 0 2 5 ( 9 ) 1 . 1 0 6 0 ( 8 ) 5 . 5 ( 3

C46 0 . 6 3 1 2 ( 6 ) - 0 . 1 7 2 5 ( 8 ) 1 . 0 3 1 6 ( 7 ) 4 . 6 ( 3

126

Table 15. continued

Anisotropically refined atoms are given in the form of the isotropic equivalent thermal parameter defined as: (4/3)*[a *B(1,1) + b *B(2,2) + c2*B(3,3) + ab(cos gamma) *B(1/2) + ac(cos beta)*B(l,3) + be(cos alpha)*B(2,3)J.

127

Figure 21. Perspective view (ORTEP plot) of [Cp*Re(CO)2Br] showing the atom labeling (hydrogen atoms are omitted for clarity).

128

Figure 22. Perspective view (ORTEP plot) of PPP+ gegenion showing the atom labeling (hydrogen atoms are omitted for clarity).

TABLE 16. Bond Distances in Angstroms for [Cp*Re(CO)2Br][Ph4P]

a

129

Atoml Atom2 Distance Atoml Atom2 Distance

Re Br 2.604(1) Cll C12 1.42(2)

Re CI 1.919(12) Cll C16 1.41(2)

Re C2 1.955(13) C12 C13 1.42(2)

Re C3 2.266(12) C13 C14 1.45(2)

Re C4 2.224(13) C14 C15 1.42(2)

Re C5 2.313(14) C15 C16 1.39(2)

Re C6 2.388(12) C21 C22 1.40(2)

Re C7 2.330(11) C21 C26 1.39(2)

P Cll 1.789(11) C22 C23 1.42(2)

P C21 1.803(10) C23 C24 1.36(2)

P C31 1.803(11) C24 C25 1.42(2)

P C41 1.802(11) C25 C26 1.40(2)

01 CI 1.11(2) C31 C32 1.42(2)

02 C2 1.07(2) C31 C36 1.39(2)

C3 C3 • 1.50(2) C32 C33 1.42(2)

C3 C4 1.41(2) C33 C34 1.38(2)

C3 C7 1.44(2) C34 C35 1.40(2)

C4 C4 • 1.55(2) C35 C36 1.40(2)

C4 C5 1.48(2) C41 C42 1.37(2)

C5 C5' 1.55(2) C41 C46 1.40(2)

Table 16. continued

130

C5 C6 1.40(2) C42 C43 1.41(2)

C6 * C6 1.52(2) C43 C44 1.39(2)

C6 C7 1.42(2) C44 C45 1.35(2)

C7' C7 1.56(2) C45 C46 1.43(2)

aNumbers in parentheses are estimated standard deviations in the least significant digits.

TABLE 17• Bond Angles in Degrees for [Cp*Re(CO)2Br][Ph4P]

a

131

Atoml Atom2 Atom3 Angle Atoml Atom2 Atom3 Angle

Br Re CI 100.5(4) C6 Re C7 35.1(4)

Br Re C2 94.8(4) Cll P C21 108.7(5)

Br Re C3 140.1(3) Cll P C31 112.1(5)

Br Re C4 146.1(4) Cll P C41 107.6(5)

Br Re C5 108.3(3) C21 P C31 110.8(5)

Br Re C6 89.4(3) C21 P C41 109.1(5)

Br Re C7 104.0(3) C31 P C41 108.4(5)

CI Re C2 90.5(5) Re CI 01 170(1)

CI Re C3 118.6(5) Re C2 02 175(1)

CI Re C4 90.4(5) Re C3 C3' 125(1)

CI Re C5 97.5(5) Re C3 C4 70.1(7)

CI Re C6 130.5(5) Re C3 C7 74.3(7)

CI Re C7 150.4(5) C3' C3 C4 127(1)

C2 Re C3 92.4(5) C3' C3 C7 126(1)

C2 Re C4 117.3(5) C4 C3 C7 107(1)

C2 Re C5 153.6(5) Re C4 C3 73.4(7)

C2 Re C6 137.2(4) Re C4 C4' 126(1)

C2 Re C7 103.4(5) Re C4 C5 74.2(8)

C3 Re C4 36.5(5) C3 C4 C4' 128(1)

Table 17. continued

132

C3 Re C5 61.6 5) C3 C4 C5 109(1)

C3 Re C6 60.1 4) C4 * C4 C5 123(1)

C3 Re C7 36.4 5) Re C5 C4 67.7(7)

C4 Re C5 38.1 5) Re C5 C5' 126(1)

C4 Re C6 60.1 5) Re C5 C6 75.6(7)

C4 Re C7 60.1 5) C4 C5 C5' 130(1)

C5 Re C6 34.7 4) C4 C5 C6 107(1)

C5 Re C7 59.2 4) C5 • C5 C6 123(1)

Re C6 C5 69.7 7) C12 Cll C16 122(1)

Re C6 C6' 125.7 9) Cll C12 C13 120(1)

Re C6 C7 70.2 7) C12 C13 C14 119(1)

C5 C6 C6' 128(1 C13 C14 C15 119(1)

C5 C6 C7 109(1 C14 C15 C16 123(1)

C6' C6 C7 124(1 Cll C16 C15 118(1)

Re C7 C3 69.4 7) C22 C21 C26 123(1)

Re C7 C6 74.7 7) C21 C22 C23 117(1)

Re C7 C7' 128.8 9) C22 C23 C24 120(1)

C3 C7 C6 109(1 C23 C24 C25 123(1)

C3 C7 C7' 124(1 C24 C25 C26 118(1)

C6 C7 C7 • 126(1 C21 C26 C25 119(1)

P Cll C12 119.3 8) C32 C31 C36 121(1)

P Cll C16 119.1 8) C31 C32 C33 118(1)

P C21 C22 117.9 8) C32 C33 C34 120(1)

133

Table 17. continued

p C21 C26 118.8(9) C33 C34 C35 121(1

p C31 C32 117.2(8) C34 C35 C36 120(1

p C31 C36 121.7(8) C31 C36 C35 120(1

p C41 C42 120.4(9) C42 C41 C46 124(1

p C41 C46 115.4(8) C43 C44 C45 121(1

C44 C45 C46 122(1

C41 C46 C45 115(1

aNumbers in parentheses are estimated standard deviations in the least significant digits.

134

C5 — 2.37

I C 7 - 2 . 2 6

ft C 5 - 2 . 3 1

C 6 - 2 . 3 9

C 3 - 2 . 2 7

C 7 - 2 . 3 3

C 3 - 2 . 2 0

Figure 23. Top view of (a) lat-Cp*Re(CO)2I2 and (b) [Cp*Re(C0)2Br]" showing the Cp* ring-rhenium bond lengths (A). Distances for the former complex are taken from ref. 32.

135

reveals an approximate mirror plane of symmetry that is

defined by the plane formed by the Re-Br and C(6)-C(6')

bonds. This plane bisects the two carbonyl groups and the

C(3)-C(4) bond of the Cp ring. The Re-Br, Re-carbonyl, and

C-0 bond lengths are unexceptional in comparison to other

cyclopentadienylrhenium complexes.32'44,88 The C-c bonds of

the Cp* ring range from 1.40(2) to 1.48(2) A with an average

length of 1.43(2) A in agreement with other peralkylated

cyclopentadienylrhenium complexes.32,44,103 The disparate

Re-C(ring) bond lengths indicate that the Cp* ring is tilted

away from the bromide atom and towards the two carbonyl

groups. The Re-C(6) bond is the longest at 2.388(12) A and

is followed by the Re-C(7) and Re-C(5) lengths of 2.330(11)

and 2.313(14) A, respectively. The shortest Re-C(ring)

distances of 2.266(12) and 2.224(13) A belong to the Re-C(3)

and Re-C(4), respectively. These variations in the Re-C

(ring) bond lengths are not entirely unexpected when the

crystallographic results of lat-Cp*Re(CO)2I2 and the

theoretical studies of related CpML^ complexes are

considered.104 For example, in lat-Cp*Re(CO)2I2 the

Re-C(ring) bonds that are opposite the iodide ligands are

observed to be longer than the corresponding pseudotrans

Re-C(ring) bonds as shown in Figure 23a. The bond length

asymmetry in 2- (as in lat-Cp*Re(CO)2I2) is readily

rationalized in terms of the ir-acceptor properties of the CO

136

ligands. The CO ligands accept electron density (via the

w* manifold) from the Cp* ring at the expense of the

opposite Re-C(5,6,7) bonds; this results in an elongation

and a tilting of the Cp* ring towards the carbonyl ligands.

Alternatively, Cp* ring tilting may be viewed as arising

from the interaction between metal d-orbital (d ) and v* xy

orbital (e2) of Cp* ring.104 The Cp* ring is expected to

exhibit maximum tilting in d5-ML4 fragments that possess an

eclipsed conformation. Moreover, small deviations from an

eclipsed conformation are not expected to negate the

predicted trends in Cp* ring tilting.32 Based on this, the

isolobal relationship predicts that 2- should display a

similar tilting pattern as in lat-Cp*Re(CO)2I232 and

104 CpMo(CO)3Me since the [Re(CO)2Br]" fragment is formally a

d^-MLj species and is equivalent to its isolobal d^-ML^

19

counterparts. Here, the metal fragments are derived by

treating the Cp ring as a neutral five-electron donor. An

alternative procedure would involve treating the Cp ring as

an anionic six-electron donor with concomitant charge

adjustment for the metal fragment, i.e., Re(CO)2Br and

Re(CO)2I2+. Regardless of the exact factor(s) responsible

for Cp* ring tilting, the isolobal nature of these metal

fragments underscores the predicted and experimentally

observed similarity between 2-, lat-Cp*Re(CO)2I2, and

137

CpMo(CO)jMe.

L. Reactivity and Stability Studies of Anionic

Pentamethylcyclopentadienyl Rhenium Complex,

[Cp*Re(CO)2Br][Li]

The functionalization of [CpftRefCO^Br]', 2-, was

examined as a potential route to complexes of the form

Cp*Re(C0)2Br(R). Treatment of 2- in THF with trifluoro-

acetic acid afforded two new PCO bands at 2018 and 1945 cm"*

as expected for anion protonation. From the intensity

pattern of the vCO bands it was concluded that protonation

proceeds to give Cp*Re(C0)2Br(H) with diagonal stereo-

chemistry. The *H NMR spectrum of treatment of 2- with

CF3CO2H in dg-THF displayed a singlet at 2.01 ppm (15H) and

a singlet at -9.98 ppm (1H) consistent with the methyl

groups of the Cp* ring and Re-H bond, respectively. 2- also

reacts with methyl triflate and magic ethyl in CH2CI2 to

yield the corresponding diag-Cp*Re(CO)2Br(R) complexes based

on in situ IR analysis. The low-temperature FT-IR spectrum

exhibits two carbonyl bands at 2016 (s) and 1935 (s), and

2020 (s) and 1944 (vs) cm"* in CH2CI2 consistent with the

diag-Cp*Re(CO)2Br(Me) and diag-CpARefCO^Br^Hg),

respectively. All of these functionalized complexes

138

decomposed in solution over a period of days to give

Cp*Re(CO)j as the only isolable product (about 20-30 %

isolated yield).

Finally, THF solution of 2- were observed to be stable

for a period of weeks when oxygen was rigorously excluded.

The presence of oxygen leads to complex mixtures of rhenium

oxides in addition to Cp*Re(C0)3. For example, 2-K (vide

infra) slowly decomposed in septum-capped vessels on the

benchtop to give the known oxides Cp*ReOg^ and ReO^" as the

major products. The yRe=0 band of ReO^" observed was

identical to that exihibited by pure KReO^. IR analysis

(KBr pellet) revealed intense pRe*0 bands at 909 and 877

cm"1 assignable to Cp*Re03 and an intense band at 913 cm"1

attributed to potassium perrhenate. The exact pathways and

other products involved in this decomposition reaction

remain as unanswered questions for future study.

M. Reduction Studies Using One-Electron Reducing Agents

About Cp'Re(CO)2Br2 (Cp* = Cp, Cp*)

the reduction of lateral and diagonal Cp'Re(CO)2&r2' 9'

was examined using known one-electron reducing agents in

order to probe for the intermediacy of the corresponding

19e" complex [Cp'RefCO^Brg]*, 10• Treatment of 9 (either

139

isomer) in THF at -78 °c with one equivalent of cobaltocene

or potassium naphthalide immediately gave 50 % conversion of

[Cp'RefCO^Br]" (11) which was assayed by low-temperature

FT-IR analysis. Use of two equivalents of reducing agent

afforded 11 in quantitative yield as expected for a two-

electron reduction process. The resulting two vCO bands of

the rhenium anion are identical in intensity and frequency

to 1-Li or 2-Li and suggest that negligible ion pairing is

present between 11 and gegencation (vide supra). The fact

that the 19e" radical anion is not observed indicates that

the lifetime of 10 is very short and is followed by faster

subsequent reactions to ultimately give 11. Loss of bromide

from 11 is anticipated to give the 17e" complex

[Cp'Re(CO)£Br], 12, which is then expected to accept a

second electron in a fast electron transfer step involving 106

the reducing agent. ° Examples of facile ligand loss upon

one-electron coupled with a second reduction step occurring

at less negative potential are quite common in

organometal1ic reduction reactions.106 Electron transfer

from 10 to 12 is also expected to yield 11 and regenerates

9. The importance of this pathway will depend on the

relative magnitude of the rates associated with bromide loss

and electron transfer from the 19e" complex 10. Facile loss

of bromide, compared to CO loss, is also predicted when one

140

considers the acidity of these two ligands. Here the weaker

jr-acceptor ligand preferentially dissociates so as to

minimize unfavorable metal-ligand orbital interactions in

the 19e" radical anion.^ Alternatively, # could undergo a

second electron accession to give a 20e" complex (assuming

no ring hapticity changes), followed by bromide loss to give

11 as shown in Scheme 7. We favor the former pathway based

on reports of dissociative halide loss in organic halides

and metal dimer fragmentation reaction upon one-electron

reduction*®® and note that these two schemes have precedence

in electrochemistry as they formally conform to an ECE and

109 EEC process, respectively.

Reaction of 9 in THF at -78 °C with sodium naphthalide

(2 mole equivalents) was observed to occur rapidly to give

9-Na. However, low-temperature FT-IR analysis revealed that

9- existed as a mixture of carbonyl oxygen-Na and solvent-

separated ion pairs. The existence of the former ion pairs

was easily confirmed through the addition of either

15-Crown-5 or HMPA (10 mole equivalents based on Re) which

immediately afforded an IR spectrum identical to

[Cp'RefCO^Br] [Li]. The vCO of 9-Na are readily assigned

based on their frequency and response to additives as was

done for 2-MgX. The two low frequency i/CO bands of 1-Na at

1808 and 1793 cm"* and 2-Na at 1791 and 1776 cm"* represent

Sch«n« 7

141

CP

OC>Rev'Br Or 9 Br

K or CP2C°

THF — 78 *C

2 Cp2Co Qf

s • K

C P '

-a Dp

OCy \-Br OC Br

- B r

OCv Br

Cp

,,-Re CO

CP

OC^ReC*Br OC Br

-to

Br

C p '

0C'yRe^C0 * 1 2

cp2co or

VI

142

the asymmetric carbonyl stretching mode associated with the

solvent-separated and carbonyl oxygen-sodium contact ion

pairs, respectively. In principle, two additional vCO bands

corresponding to a symmetric carbonyl stretching for each

type of ion pairs should be observed; however, only one band

is observed at 1882 cm ^ for 1-Na and 1863 cm~^ for 2-Na,

respectively. An analogous situation involving

[CpFe(CO)2] [Na]^®® and [CpMo(CO)^] [Na]^ has been observed

and presumably is the result of coupling and/or inadequate

band resolution.

N. Thermodynamic Study of the Ion Pairing Equilibrium for

[Cp*Re(CO)2Br][Na]

Solution FT-IR studies of 2-Na (in THF) at room

temperature display a 39:61 mixture of carbonyl oxygen-

sodium and solvent-separated ion pairs, respectively.

However, low-temperature FT-IR analysis at -70 °C reveals a

73:27 mixture of ion pairs, respectively. This behavior

indicates that these ion pairs in THF solution are involved

in a reversible temperature-dependent equilibrium. The

equilibrium constant for these ion pairs has been determined

by IR band shape analysis over the temperature range -70 °C

to room temperature and values of AH and As are obtained.

.3 2.5 x 10 M THF solutions of 2-Na were used to examine the

143

effect of temperature on the equilibrium between these ion

pairs in order to quantify the thermodynamics associated

with this system. Figure 24 shows selected IR spectra for

2-Na as a function of temperature. As the temperature is

raised from -70 °C to room temperature, the equilibrium

constant is gradually shifted to the solvent-separated ion

pairs in accordance with the equilibrium shown in Eq. 12.

The ion pairing exhibited by 2-Na represents another example

of the subtle balance between the electrostatic anion/cation

interaction and the solvation of the cation by THF in

determining the extent of ion pairing in organometallic

complexes. Lithium and potassium ion pairing in 2 is

negligible based on the formation of stable Li+»nTHF

solvates and an insufficient K+ electrostatic potential

which minimizes anion/cation interactions.^5>97 Extensive

sodium ion pairing results from Na+,s intermediate

oxophilicity and electrostatic properties which favor

contact ion pairs.75>97 Band area measurements on the two,

overlapping asymmetric carbonyl stretching bands were

performed with the assumption that the area under each

absorption may be taken to be proportional to the relative

amount of each ion pair. Table 18 gives the Keq values for

the equilibrium defined in Eq. 12 from which a Van't Hoff

plot showing the variation of In Keq as a function of

temperature is readily constructed as shown in Figure 25.

e ^

144

O O

o o CO

£ o

O o CT>

© «) 0) <a M 2 > tt

CM

<# u sj o>

145

TABLE 18. Equilibrium Parameters for the Conversion of 2-Na from CIP into SSIPa.

1000/T in Keqb

3.36 0.405

3.36 0.456c

3.53 0.306

3.60 0.261

3.66 0.148

3.80 0.040

3.80 -0.012°

3.95 -0.136

4.20 -0.439

4.39 -0.571

4.59 -0.717

4.69 -0.719

4.93 -0.974

AH » 1.8 ± 0.06 kcal/mole^ AS = 7.0 ± 0.2 eu^

From 2.5 x 10-3 M [Cp*Re(CO)2Br] [Na] in THF by following the changes in the area of the 1790 and 1776 cm"1 carbonyl bands. Defined as the [area of solvent-separated ions]/ [area of carbonyl oxygen-sodium contact ions]. Multiple determination from a second separate experiment. Error limits at 95 % confidence limits.

146

1.5-

- 1.0-• er XL c

0.5

0.0-

3.0

Cp*Re(CO)2Br_Na*

~r-3.5

Cp*Re(CO)2Br-||Na*

r~ 4.0

1000/T 0C')

Figure 25. Plot of In Keq + 1 vs. 1/T for the contact solvent-separated ion pair equilibrium for [Cp*Re(CO)2Br][Na] in THF,

147

Values of 1.8 ± 0.06 (kcal/mole) for AH and 7.0 ± 0.2 (eu)

for AS are computed from the slope of the straight line and

the intercept, respectively, it is immediately apparent

that the enthalpic and entropic contribution to the ion pair

equilibrium oppose each other. Low temperatures (< 263 K)

favor contact ion pairs as a result of an exothermic

electrostatic interaction between 2- and Na+ while higher

temperatures lead to solvent-separated ion pairs as the TAS

term dominates the free energy change associated with the

equilibrium. A similar temperature-dependent process has

also been reported for [Co(C0)4][Na] in THF by Edgell and

51a

Lyford. In that study the AH and AS values were observed

to be 3.7 (kcal/mole) and 4 (eu), respectively, for the

equilibrium defined analogously to our Eq. 12. Our

thermodynamic values compare well with those of Edgell and

Lyford, which indicate that the carbonyl oxygen-sodium

interaction in [Co(CO)4][Na] is stronger than in

[CpRe(CO)2Br][Na]. Finally, we note that the conversion of

contact ion pairs of 2-Na into solvent-separated ion pairs

displays a temperature response that is different from

several aromatic carbanions and complexes of macrocyclic

polyethers. For example, Jackman et al.^® have

demonstrated that sodium fluorenide exists primarily as

contact ions at room temperature (~ 95%) while potassium

fluorenide/18-Crown-6 solutions are predicted to exhibit

148

increased ion pairing as the temperature is raised above

room temperature. While an explanation concerning the

difference in AH for the conversion contact ions into

solvent-separated ions has been proposed®**, no current

theory exists that adequately explains such an inverse

temperature behavior between organic and organometallic

ions.

0. Reaction of Cp'Re(CO)2Br2 with BU3S11H and Et3SiH

(Cp' = Cp, Cp*)

The reaction of Cp'Re(CO)2Br2 with BU3S11H at room

temperature was discovered to afford the corresponding

dihydride in excellent yield and, thus, represents an

improved synthetic route for the synthesis of

Cp'Re(CO)2H2.49 Treatment of CpRe(CO)2Br2 (either isomer) in

benzene at room temperature with one equivalent of Bu^SnH

afforded four new i/CO bands at 2038 and 1970 cm"1; 2014 and

1939 cm"1 which are consistent with the diag-CpRe(CO)2Br(H)

(13) and diag-CpRe(CO)2H2 (14), respectively. Use of two

equivalents of Bu^SnH afforded complete conversion to 14

based on in situ IR analysis. These results indicate that

H"/Br"exchange occurs stepwise to produce the intermediate

complex 13, followed by a second H~/Br" exchange to give 14

149

complex (Eq. 24). The identity of 14 was also determined by

oc...r#...b I °bV \?O °£*>rV

Bu3SnH ^ 2 2 ? BujSnH J4 C°

VI I -v

oc->R#^Bf

(XT Br

OC->R*wH +" <24)

Br ^CO 2 BujSnBr

13

*H NMR spectroscopy. When a dg-benzene solution of

CpRe(CO)gBr£ (either isomer) was treated with two

equivalents of BujSnH at room temperature two new resonances

appeared at 4.35 and -9.68 ppm with an integral ratio of

5:2, respectively. Interesting enough, diag-CpRe(CO)2Br2

was observed to react faster with Bu3SnH than the

corresponding lateral isomer based on qualitative rate

measurements made using IR spectroscopy. Reaction of BugSnH

with lat-Cp*Re(CO)2Br2 was examined next. Use of one

equivalent of Bu^SnH led to the formation of two new vco

bands at 1999 and 1923 cm"1 which are readily to the

diagonal dihydride complex diag-Cp*Re(CO) 2H2 (15). The

yield of the dihydride was approximately 50 % with unreacted

dibromide accounting for the remaining material. The

intermediate complex diag-Cp*Re(CO)2Br(H) (16) was not

1 5 0

observed which is in direct contrast to the Cp analog where

the stepwise H /Br exchange was observed* This suggests

that the lifetime of the intermediate monohydride (16) is

very short and is followed by a faster subsequent H~/Br~

exchange reaction to ultimately give 15. The identity of 15

was also ascertained using NMR spectroscopy. Treatment

of lat-2 with two equivalents of Bu3SnH is dg-benzene gave

two resonances at 1.81 and -9.25 ppm with an integral ratio

of 15:2, respectively, consistent with the proposed

compound.

The dihydrides 14 and 15 could be converted directly to

the corresponding bis(tributyltin) complexes in a one-pot

synthesis using EtjN. Such a sequence undoubtedly involves

the deprotonation of the Re-H moiety to generate the

conjugate base, followed by anionic attack on the in situ

BujSnBr by-product present in solution. An analogous

reaction scheme using alkyl iodides has been examined by

37a

Bergman and Yang. The deprotonation/alkylation scheme is

shown in Equation 25.

Cp*R«(CO)2H2 +

2 Bu^SnBr

Cp* Re(CO)9(SnBu*)9 2 Bt*N . 4 C

3 , + (25) 2 Et3MHBr

151

The bis(tributyltin) derivatives were isolated by

chromatography, followed by sublimation in good to moderate

yield. IR analysis has been used to assign the stereo-

chemistry in both bis(tributyltin) compounds. Based on the

observed intensity pattern of the two vCO bands [1943 (m)

and 1890 (s) cm"1, ZOC-Re-CO - 118.4°] in the cyclopenta-

dienyl complex, diagonal stereochemistry is suggested for

CpRe(CO)2(SnBu3)2, 17 (Fig. 26a). This agrees with the

results of Graham et al. who have shown that reaction of

Me3SnCl with diag-CpRe(CO)2H2 gives diag-CpRe(CO)2(SnMe3)2.

41a

The same reaction using 15 leads to inversion of

stereochemistry and productive of the lat-Cp*Re(CO)2(SnBu3)2

(18). IR analysis reveals a reversed i/CO intensity pattern

[1967 (vs) and 1896 (s) cm"1, ZOC-Re-CO - 83°] indicative of lateral stereochemistry (Fig. 26b).

119 1

The Sn{ H) NMR spectra of both bis (tributyltin)

complexes were recorded in order to further characterize

these new complexes. The 119Sn chemical shifts of 17 and 18

were observed at -5.6 and 8.8 ppm, respectively, relative to

external Me4Sn (S - 0.0). Figure 27 displays the

representative 119Sn NMR spectrum.

Use of Et3SiH as a potential H donor was next examined

because silanes have been reported to function as reducing

agents.111 However, no reaction was observed between

152

1950 1850

v, cm -1

Figure 26. Infrared spectra of the carbonyl region for (a) diag-CpRe(CO)2(SnBu3)2 (b) lat-Cp*Re(CO)2(SnBu3)2 Both spectra were recorded at 25 °C in cyclohexane.

153

i i — i — i — i

100 0 -100

PPM

Figure 27. The Sn nmr spectrum of diag-CpRe (CO) 2 (SnBu3), at 25 °c in CDClj solution (0.1 M).

154

CpRe(CO)2Br2 (either isomer) and benzene at room

temperature. This observation stands in marked contrast to

the BUjSnH reactions which proceed readily at room

temperature. When the reaction between diag-CpRe(CO)2Br2

and EtjSiH (5 equiv.) was conducted at 60 °C the initial

products were diag-CpRe(CO)2Br(H) and CpRe(CO)3. Longer

reaction times led to an increase in the yield of the latter

compound at the expense of the former monohydride and

dibromide starting material. From IR analysis the solution

yield was 30-40 % and its formation is believed to arise

from the thermal decomposition of the monohydride (vide

supra). Equation 26 shows the proposed course of the

reaction.

CpRe(CO)2Br2 + EtgSiH diag-CpRe(CO)?Br(H) 60 °C

decomposition CpRe(CO)3

(26)

The reaction between lat-CpRe(CO)2Br2 and Et3SiH in

toluene was next examined at 40 °C where an initial fast

isomerization of the starting material to diag-CpRe(CO)2Br2

was observed, followed by a slower reaction to give

155

diag-CpRe (CO) 2Br(H) and CpRe(CO)g (vide supra). Evidence to

support an isomerization/H" exchange scheme was next probed

by examining this reaction using IR spectroscopy, if the

isomerization step does indeed precede (i.e., is faster) the

H exchange step, the rate of consumption of

lat-CpRe (CO) 2Br2 ®kould be faster than that observed with

diag-CpRe(CO)2Br2.

The isomerization reaction of lat-CpRe(CO)2Br2 was

studied in toluene solution both with and without added

silane by following the decrease in the absorbance of the

highest frequency vco band (symmetric vCO stretch) of the

starting material as a function of time. Changes in the

carbonyl absorbance (2046 cm"*) were negligible at 40 ° c in

the absence of added silane. On the other hand,

isomerization of lat-CpRe (CO) 2Br2 to diag-CpRe (CO) 2Br2 in

the presence of a measured excess of Et^SiH (10 equiv.)

proceeded rapidly at 40 °C and without any noticeable

Br /H" exchange. The isomerization reaction was complete

after eight hours at 40 °C (Fig. 2 8 ) .

Confirmatory proof for a slower H" exchange reaction

was obtained by qualitatively following the isomerization

reaction between diag-CpRe (CO) 2Br2 and Et3SiH. If the above

observations are correct, then the observed IR absorbance

changes for the consumption of diag-CpRe(CO)2Br2 should be

156

- o

o

00 E

- vo

- *4-

- CM

•sqv (00)A m

o o o

•P <d 0) c 0) 3 H 0 •p c •H

C\J CQ cm o o w 0) « a a •p •

(d E H •H

to M CO 0 4J W

m <W 0) 0 tn c a* (0 0 JS G 0

(0 a> 0 0 u c ai (0 JQ a)

£ 0 4J (0 3

g •H

oo CM 0) u §> •H

157

less than that observed with isomerization. The reaction

between diag-CpRe(CO)2Br2 and EtjSiH (10 equiv.) was studied

in toluene at 40 °C by monitoring the change in the

absorbance of the highest frequency vCO band (2000 cm"*).

Figure 29 shows these results. It is readily seen that the

consumption of diag-CpRe(CO)2Br2 is rather slow, requiring

three days for complete consumption. The consumption of

diag-CpRe(CO)2Br2 is clearly slower than the isomerization

reaction.

While these reactions with Bu^SnH and EtjSiH have not

been studied kinetically it is of interest to consider these

reactions on thermodynamic grounds in order to better

understand why the observed dihydride products formed. This

is readily done when the bond dissociation energies of the

participating reactants are examined. Equations 27 and 28

show the overall reactions under consideration.

I 2 Bu3snH + 2 Bu3snBr

0C> R'vJ3r 0 C > R , C - H U 7 )

Br ^ C 0 H ^ C 0

AH • 2 (BDE Re-H + BDE Sn-Br) - 2 (BDE Re-Br + BDE Sn-H)

AH » - 26 kcal/mole

158

-t 00

k.

CSI -<=

£ o ^

00 -4-

<£> CO

-<r cn

— CM

A •P • •H <D * C

0) eg 5 & H CQ o CNJ.p

1-5

« o° & 4.3 <G jj •h (d v bi w" 0 (Q <H 0) 3 -<D 0) tJ> c T3 2 H Xi o o (H Q> O H i <0 •9 w M -H S a„ 5 w

<n oa Q)

•sqv (oo) /i a i

•H

159

Table 19 lists the bond dissociation energies (BDE) needed

to evaluate the reactions shown in Equations 27 and 28.

While accurate measurements for the Sn-H (74 kcal/mole),

and Si-Br (96 kcal/mole) bond exist, no actual

2 Et38iH C C l T

+ 2 Et3SiBr

°bc;>%BO V%O <28>

- AH = 2 (BDE Re-H + BDE Si-Br) - 2 (BDE Re-Br + BDE Si-H)

AH - - 30 kcal/mole

measurements are available for the strengths of the Re-H and

Re-Br bonds. However, we may readily estimate the values

associated with these bonds by using similar model

compounds. For example, Cp*Ir(PMe3)(Br)-Br114 and

115

Cp2W(Br)-Br have bond dissociation energies of 76 and 72

kcal/mole, respectively. Based on these values for the M-Br

bond, a BDE value of 70 kcal/mole has been assumed for the

Re-Br bond in diag-CpRe(CO)2Br2. A similar situation exists

with respect to the BDE of the Re-H bond in

diag-CpRe(CO)2H2. Based on the reported M-H BDE values

given in Table 19 the Re-H BDE in diag-CpRe(CO) 2H2 may be

160

0) 0 G 0) k 0) «

ra H H

n in vo

w <D •H Cn o c w c o •H -P (d •H O 0 W CO •H Q

TJ C o A •g 0)

O <D H 0) CO

OH H

3 cq $

& a) c w c o •H <d •H o o (0 m

*0 c <§

0) r-4

S

(0 o M

in oo CO

vo Ol r*

vo r*

n CM oo oo

U CQ

S l l ^ ^ u « CQ

TJ a 3 o & o a

CO 0) CO

a)

M

u U £ s i w CQ « cq P* p« S. M 1 1 1 i W s CQ a c •H •H u «W <w CO CO CO CO H H s £ CO CO CO CO * * -K * a) se S 1 S3 & a o 04 o

U CQ

161

confidently estimated at -70 kcal/mole. This Re-H BDE value

is acceptable since it is known that M-H BDE's are not

particularly dependent upon the metal or the nature of the

ancillary ligands.3*'*^

The reaction in Equation 27 is calculated to proceed to

dihydride and BujSnBr based on the favorable enthalpy value

(AH » -26 kcal/mole). The driving forces in this transfor-

mation are the formation of two strong Re-H and Sn-Br bonds.

Reaction with silane (Equation 28) is also predicted to

proceed to diag-CpRe(CO) 2H2 and Et3SiBr based on an

exothermic enthalpy value of -30 kcal/mole. However, no

dihydride is observed presumably due to the fast and

irreversible decomposition reaction associated the

intermediate monohydride complex diag-CpRe(CO)2(Br)(H).

The thermodynamics associated with the bis(tributyltin)

complex diag-CpRe (CO)2(SnBuj) 2 are of interest because the

bis(tributyltin) complex is predicted to be unstable

according to Equation 29. The BDE values used for the Re-H

^7

+ 2 HBr (29) 2 Bu^SnBr |

0C-> R*CH 0C->R,^SnBu3 H ^ C 0 BuaSrr CO

AH - 2 (BDE Re-Sn + BDE H-Br) - 2 (BDE Re-H + BDE Sn-Br)

AH « 54 kcal/mole

162

and Sn-Br bonds have already been described while the BDE

value for HBr is given in Table 19. The only BDE value that

is unknown belongs to the Re-Sn bond in the product. Based

on the fact that metal-alkyl bond strengths are weaker than

the corresponding metal-hydride bonds,3a coupled with the

fact that many metal-alkyl BDE's are on the order of 20-40

kcal/mole, we have assured that the BDE of the Re-Sn bond in

diag—CpRe(CO)2(SnBUj)2 is ~40 kcal/mole. Using these values

the reaction as written in Equation 29 is observed to be

endothermic by 54 kcal/mole consistent with the observation

that no reaction occurs without the addition of EtjN (vide

supra). The base serves to drive the reaction to be right

by removing the generated HBr as [EtjNH][Br].

P. Hydride Reduction of Indeny Rhenium Tricarbonyl:

Synthesis and Characterization of [ (r?5-CgH7)Re(CO) 2H]"

The reaction of (f?5-CgH7)Re(CO)3, 19, with 1.0 mole

equiv. of LiEt^BH at room temperature led to the immediate

formation of [ (n5-C9H7)Re(CO)2H]", 20, in 50 % yield as

determined by IR analysis. When the hydride concentration

was doubled, complete conversion to 20 was observed. 20

exhibits two vCO bands at 1991 and 1906 cm"1. At no time

were any possible intermediates obtained in this reaction.

163

The identity of this new complex was characterized by NMR

spectroscopy. *H NMR spectrum of the yellow solution

obtained from the treatment of 19 with LiEtjBH in dg-THF is

shown in Figure 30. Two sets of resonances associated with

the AA'BB' spin system are observed at 7.24 and 6.43 ppm

with a relative integral ratio of 2:2 which are consistent

with four protons on an uncomplexed six-membered benzenoid

ring. The other two sets of resonances appearing at 6.55

and 5.64 ppm with a relative integral ratio of 1:2

correspond to the three protons attached to the five-

membered ring in the fj -indenyl ligand. The terminal metal

hydride resonance appeared at -5.5 ppm as expected for a

singlet with a relative intensity of one.

The reduction of (rj -CgHjJRefCOJj requires two equiva-

lents of LiEtjBH (vide supra). It is believed that one

equivalent of the hydride reagent reacts with

(T75-C9H7)Re(CO)3 to give [ (n5-CgH7)Re(CO)2H]" and Et^B as

shown in Scheme 8. A subsequent, rapid reaction between the

generated EtjB and excess R3BH" would then yield the dimer

[EtjB-H-BEtj]" and account for the 1:2 rhenium/hydride

stoichiometry observed with this reaction.107b»118 An

5 1

alternative pathway involves r) -* r? ring slippage of

indenyl ring in 19 upon hydride attack, followed by a fast

loss of CO to give 20. Since the presence of the fused

164

ift IO I

1

= 1

in

© - k

B I

*0° c •H o o in ca •P <d

K C\l

O u w 0) «

a as 0 1

IO c

o cn a) u & •H

165

Schema 8

o c / \

OCT CO

+ 2Et3BH

19

(Et3BJ2H"

R® H ocy \ oc f CO

21

benzene ring in t -CgHy complexes results in more a facile

5 3 .

r\ -*• r\ ring slippage for indenyl ligands than for

cyclopentadienyl ligands, coupled with the observation of

other slipped indenyl rings,47b we cannot rule out this

166

pathway based on the existing experimental data.

Finally, THF solutions of [ (rj5-CgH7)Re(CO) 2H]" were

observed to be stable when oxygen was rigorously excluded.

The functionalization of [ (»?5-C9H7)Re(CO)2H]" remains to be

explored.

167

REFERENCES

1. (a) Halpern, J. Adv. Chem. Ser, 1966, 7Q., 1. (b) Halpern, J. Discuss. Faradav Sac. 1968, 7, 46. (c) Parshall, G. W. Acc. Chem. Rea. 1975, 5, 114. (d) Parshall, G. W. Catalysis (Chemical Society Specialist Periodical Report) 1977, 1, 325. (e) Shilov, A. E.; Steinman, A. A. Coord. Chem. Rev. 1977, Zk, 97. (f) Webster, D. E. Adv. Oraanomet. che^T 1977, 147.

2. Chatt, J.; Davidson, J. M. J. Chem. Soc. 1965, 843.

3. (a) Halpern, J. Inorcr. Chim. Acta. 1985, 100. 41. (b) Crabtree, R. H. Chem. Rev. 1985, ££, 245. (c) Shilov, A. E. "Activation of Saturated Hydrocarbons by Transition Metal Complexes", D. Reidel Publishing Company: Boston, 1984.

4. (a) Walling, C. "Free Radicals in Solution", Wiley: New ¥ork, 1957. (b) Huyser, E. S. "Free Radical Chain Reactions", Wiley: New York, 1970. (c) Pryor, W. A. "Free Radicals". McGraw-Hill: New York, 1966.

5. (a) Cohen, Z.; Keinan, E.? Mazur, Y.? Haim, V. T.

Qrqt Chefflt 1975, 2141. (b) Deno, N. C.J Jedziniak, E. J.? Messer, L. A.; Meyer, M. D.; Stroud, S. G.; Tomezsko, E. S. Tetrahedron 1977, 12, 2503. (c) Olah, G. A.? Parker, D. G.; Yoneda, N. Ancrew. Chem. Int. Ed. Engl. 1978, H , 909. (d) Rozen, S.;

Gal, C.; Faust, Y. J. Am. Chem. Soc. 1980, 102. 6861.

6. (a) Bruce, M. I. Anqew. Chem. Int. Ed. Rnci. 1977, l£,

73. (b) Foley, P.; Discosimo, R.; Whitesides, G. M. Am. Chem. Soc. 1980, 102. 6713. (c) Whitesides, G.

M- Pyre AFPlt Chemt 19«1, 52, 287. (d) Reamey, R. H.; Whitesides, G. M. J. Am. Chem. Soc. 1984, lag, 81. (e) Calabrese, J. c . ; Colton, M. C.; Herskovitz, T.; Klabunde, U.; Parshall, G. W.; Thorn, D. L.; Tulip, T. H- Annt y, Y, A<?<Pdt ggit 1983, U S , 302. (f) Parshall, G- W. CHEM. TECH. 1984, 628. (g) Anderson, R. A.;

168

Jones, R. A.; Wilkinson, G. J. Chem. Soc.. Dalton Transt 1978, 446. (h) Werner, H.; Werner, R. J. Oraanomet. Chem. 1981, 209. C60.

7. Shilov, A . E. Pure APPI. Chem. 1 9 7 8 , 725.

8. (a) Crabtree, R. H.? Mellea, M. F.? Mihelcic, J. M.; Quirk, J. M. J. Am. Chem. Soc. 1982, 1 M , 107. (b) Crabtree, R. H. CHEM. TECH. 1982, 506. (c) Crabtree, R. H.; Demou, P. C.; Eden, D.; Mihelcic, J. M.; Parnell, C. A.; Quirk, J. M.? Morris, G. E. J. Am. Chem. Soc. 1982, 1M/ 6994. (d) Crabtree, R. H.; Parnell, C. P.; Uriarte, R. J. Oraanometallics 1984, 2, 816.

9. (a) Baudry, D.; Ephritikhine, M.; Felkin, H. J. chem. Soc.. Chem. Commun. 1980, 1243; 1982, 606; 1982, 1235. (b) Baudry, D.; Ephritikhine, M.; Holmes-Smith, R. J. Chem. Soc.. Chem. Commun. 1983, 788.

10. (a) Berry, M.; Elmitt, K.; Green, M. L. H. J. Chem. Soc.. Dalton Trans. 1979, 1950. (b) Bradly, M. G.; Roberts, D. A.; Geoffroy, G. L. J. Am. Chem. Soc. 1981, 103. 379. (c) Ittel, S. D.; Tolman, C. T.; English, A. 0.; Jesson, J. P. J. Am. Chem. Soc. 1978, 100. 7577.

11. Hoyano, J. K.; Graham, W. A. G. J. Am. Chem. Soc. 1982, IQi, 3723; 1983, lfi£, 7190.

12. (a) Janowicz, A. H.; Bergman, R. G. J. Am. Chem. Soc. 1983, 105. 3020. (b) Wax, M. J.; Stryker, J. M.; Buchanan, J. M.; Kovac, C. A.; Bergman, R. G. J. Am. Chem. Soc. 1984, 1££, 1121.

13. (a) Jones, W. D.; Feher, F. J. J. Am. Chem. Soc. 1984, JJ2£, 1650. (b) Jones, W. D.; Feher, F. J. Oraanometallics 1983, 2, 562.

14. For reviews on C02 activation by organometallic complexes and the manifestations of the "Greenhouse Effect", see: (a) Behr, A. in "Catalysis in Cj Chemistry", Keim, W.; Ed.; D. Reidel Publishing Co.: Boston, 1983; CH. 6. (b) Volpin, M. E.; Kolomnikov, I. S. Pure APPI. Chem. 1 9 7 3 , 22, 567. (c) Woodwell, G. M.

169

Sclent. Amer. 1978, 233., 34. (d) Volpin, M. E. ? Kolomnikov, I. s. in "Organometallic Reactions", Becker, E. S.; Tsutsui, M., Ed.? Wiley: New York,

Vplt—1/ 313—386. (e) Sneeden, R. P. A. in "Comprehensive Organometallic Chemistry", Wilkinson, G.; Stone, F. G.; Abel, E. W., Ed., Pergamon Press: London, 1982; CH 50. (f) Evans, G. 0.; Walter, W. F.; Mills, D. R.; Streit, C. A. J. Orcranomet. Cham. 1978, Ml/ C34. (g) Forshner, T.; Menard, K.; Culter, A. J. Chem. Soc.. Chem. Commun. 1984, 121. (h) Felkin, H.; Knowles, P. J. J. Oraanomet. Chem. 1972, 21, C14. (i) Felkin, H.; Knowles, P. J.; Meunier, B. J, QrqanPPett 1978, 146, 151. (j) Davies, S. G. ; Green, M. L. H. J. Chem. Soc.. Dalton Trans. 1978, 1510.

15. Lowry, R. P.; Auguilo, A. Hydrocarbon Proc. 1974, 103.

16. Roth, J. F.; Craddock, J. H.? Hershman, A.; Paulik, F. E- CHEM. TECH. 1971, 1, 600.

17. Foster, D. Adv. Oraanomet. Chem. 1979, 12, 255.

18. Broich, F.; Hofermann, H.; Hunsmann, W.; Simmrock, H. Erdol Kohle-Erdaas Petrochemie. 1963, 1&, 272.

19. (a) Hoffmann, R. Science (Washington. D. c.\ 1981, 211, 995. (b) Hoffmann, R. Anaew. Chem. Int. Ed. Enqlt 1982, 21, 711.

20. For other analogous processes employing iridium complexes and reviews on C-H bond activation, see: (a) Newman, L. J.; Bergman, R. G. J. Am. Chem. Soc. 1985, 5314. (b) Faller, J. W.; Felkin, H. Oraanometa11ics 1985, ±, 1488. (c) Burk, M. J.; Crabtree, R. H.; Parnell, C. P.; Uriarte, R. J. Qrqan<?ffletain<?? 1984, 2, 8I6. (d) Bergman, R. G. ggj^n<r? 1984, 221, 902. (e) Ephritikhine, M.

NQUVt Chlfflt 1986, lfl, 9. (f) Janowicz, A. H.; Periana, R. A.; Buchanan, J. M.? Kovac, C. A.; Stryker, J. M.; Wax, M. J.? Bergamn, R. G. Pure APPI. Chem. 1984, 108. 1537.

170

21. Fagan, P. J.; Manriquez, M. J.; Maries, T. J. in "0rganometalli.es of Elements"; D. Reidal: Pordrecht, 1979, p. 113.

22. Green, H. L. H.; Wilkinson, G. J. chem. Soc. 1958, 4314-17.

23. Nesmeyanov, A. V.; Anisimov, K. N.? Koloboba, N. E.; Baryshikov, L. I. Izv. Akad. Nauk SSSP. ntd. Khim. Nauk 1963, 193.

24. Zdanovich, V. I.? Lobanova, I. v.; Kolobova, N. E. IZYf Akad. Nauk SSSR. Ser. KMm. 1984, H i , 2149.

25. Fitzpatrick, P. J.; Le Page, Y.; Sedman, J.; Butler, I. s* Inora. Chem. 1981, 2852.

26. Heah, P. C.; Poatton, A. T.; Gladysz, J. A. J. Am. Chem. Soc. 1986, 108. 1185.

27. (a) Bercaw, J. E.; Marvich, R. H.; Bell, L. G.; Brintzinger, H. H. J. Am. Chem. Soc. 1972, M , 1219. (b) Bercaw, J. E. J. Am. Chem. Soc. 1974, 9£, 5087. (c) Thomas, J. L.; Brintzinger, H. H. J. Am. chem.

1972, 91, 1386. (d) Thomas, J. L. J. Am. Chem. Soc. 1973, 25, 1838.

28. Nesmeyanov, A. N.? Kolobova, N. E.; Makarov, Y. V.; Anisimov, K. N. Izv. Akad. Nauk sssr. ser. Khim. 1969, 1826.

29. (a) King, R. B.; Reimann, R. H.; Darensbourg, D. J.

1975, 21, C23. (b) King, R. B.; Reimann, R. H. Inora. chem. 1976, 15, 179.

30. Kolobova, N. E.; Valueva, Z. P.; Kazimirchuk, E. I. IZVf Akad. Nauk SSSR. Ser. Khim. 1981, 408.

31. Dong, D. F.; Hoyno, J. K.? Grasham, W. A. G.; Can. J. Chem. 1981, ££, 1455.

32. Einstein, F. W. B.; Klahn-Oliva, A. H.; Sutton, D.; Tyers, K. G. Oraanometallica 1986, 5, 53.

171

33. Lokshin, B. V.; Rusach, E. B.; Valueva, Z. P.; Ginzburg, A. G.; Kolobova, N. E. J. Oraanomet. Chem. 1975, U22, 535.

34. Nesmeyanov, A. N.? Kolobova, N. E.; Makarov, Y. V. ; Anisimov, K. N. Zh. Obshch. Khim. 1974, M , 2222.

35. Aleksandrov, G. G.; Struchkov, Y. T.; Makarou, Y. V. Zht KhiWt 1973, n , 86.

36. Nesmeyanov, A. N.; Kolovoba, N. E.; Makarov, Y. V.; Kazimirchuk, E. I.; Anisimov, K. N. Izv. Akad. Nauk SSSR. Ser. Khim. 1976, 159.

37. (a) Yang, G. K.; Bergman, R. G. J. Am. Chem. Soc. 1983, 1££, 6500. (b) Yang, G. K.; Bergman, R. G. Oraanometallics 1985, ±, 129.

38. Casey, C. P.; Rutter, E. W., Jr.? Haller, K. J. J. Am. Chem. Soc. 1987, 1££, 6886.

39. Kristjansdottir, S. S.? Moody, A. E.; Weberg, R. T.; Norton, J. R. Oraanometallics 1988, 7, 1983.

40. Bursten, B. E.? Gatter, M. G. Oraanometallics 1984, 3, 941-943.

41. For examples of other four-legged piano-stool complexes based on the CpRe fragment, see: (a) Hoyano, J. K.; Graham, W. A. G. Oraanometallics 1982, 1, 783. (b) Fischer, E. O. ? Frank, A. Chem. Ber. 1978, 3740. (c) Bergman, R. G.? Seidler, P. F.? Wenzel, T. T. J. Am. Chem. Soc. 1985, 1£I,4359. (d) Wenzel, T. T.; Bergman, R. G. J. Am. Chem. Soc. 1986, I M , 4856. (e) Graham, W. A. G. J. Oraanomet. Chem. 1986, 300. 81. (f) Pasman, P.? Snel, J. J. M. J. Oraanomet. Chem. 1986, 2Q1, 329. (g) Jones, W. D.? Maguire, J. A. Qrqangffigtalligg 1986, 5, 590. (h) Baudry, D. H.; Ephretikhine, M. J. Chem. Soc.. Chem. Commun. 1980, 249. (i) Goldberg, K. I.; Bergman, R. G. Oraanometallics 1987, £, 430. (j) ibid, J. Am. Chem.

1989, H I , 1285.

42. Hoyano, J. K.? Graham, W. A. G. J. Chem. Soc.. Chem. Commun. 1982, 27.

172

43. (a) Klahn-Oliva, A. H.? Singer, R. D.; Sutton, D. J. Afflt Cheffit g9<rt 1986, 1M, 3107. (b) For related chemistry, see: Klahn, A. H.; Sutton, D. Oraanometallics 1989, &, 198.

44. Barrientos-Penna, C. F.? Klahn-Oliva, A. H.; Sutton, D. Oraanometallics 1985, 1, 367.

45. Einstein, F. W. B.; Jones, R. H.; Klahn-Oliva, A. H.; Sutton, D. Oraanometallics 1986, 5, 2476.

46. King, R. B.; Efraty, A. J. Oraanomet. Chem. 1970, 22, 527.

47. (a) Casey, C. P.; O'Connor, J. M. Oraanometallics 1985, 1, 384. (b) O'Connor, J. M.; Casey, C. P. Chem. Rev. 1987, £7, 307.

48. Yezernitzkaya, M. G.? Lokshin, B. V.; Zdanovich, V.I.; Lobanova, I.A. J. Oraanomet. Chem. 1985, 222,363.

49. Lee, S. W.; Richmond, M. 6. Unpublished results.

50. King, R. B. Adv. Oraanomet. Chem. 1964, 2, 157.

51. (a) Edgell, W. F.; Lyford, J., IV. J. Am. Chem. Soc. 1971, 21, 6407. (b) Edgell, W. F.; Hegde, S.; Barbetta, A. J. Am. Chem. Soc. 1978, lfifl, 1406. (c) Edgell, W. F.? Chanjamsri, S. J. Am. Chem. Soc. 1980, 102. 147.

52. Pribula, C. D.? Brown, T. L. J. Oraanomet. Chem. 1974, 21/ 415.

53. (a) Darensbourg, M. Y•j Darensbourg, D. J.y Burns, D* r Drew, D. A. J. Am. Chem. Soc. 1976, M , 3127. (b) Darensbourg, M. Y.; Darensbourg, D. J.; Barros, H. L. C. Inoro. Chem. 1978, 12, 297. (c) Darensbourg, M. Y.; Hanckel, J. M. J. Oraanomet. £h£BL. 1981, 212, C9; Qrqanpffigtalligg 1982, 1, 82.

54. Darensbourg, H. Y.; Jimenez, P.? Sackett, J. R.? Hanckel, J. M.; Kump, R. L. J. Am. Chem. Soc. 1982, 1M» 1521.

173

55. Schriver, D. F. "The Manipulation of Air-sensitive Compounds", McGraw-Hill: Yew York, 1969.

56. Fieser, L. F.; Fieser, M. "Reagnets for Organic Synthesis", Wiley: New York, 1967: Vol.1, p. 967.

57. Eisch, J. J.; King, R. B. "Organometallic Syntheses", Academic Press: New York, 1965: pp. 70-71.

58. Weber, W. P.; Gokel, G. W. ? Ugi, I. K. Ancrew. Chem. I n t * E a , E n g l , 1 9 7 2 , £ .

59. Kuivila, H. G. and Beumel, O. F. Jr., J. Am. Chem., Soc. 1 9 6 1 , SI, 1246.

60. Freeman, R.? Morris, G. A. J. Chem. Soc.. Chem. CQPPVmt 1 9 7 8 , 6 8 4 .

61. Bax, A.; Morris, G. A. J. Maan. Reson. 1981, 42, 501.

62. Germain, G.: Main, P.; Wolfson, M. N. Acta Crvstlloqr.. Sect. A: Crvst. Phvs. Diffr. Theor. Gen. Crystalloar. 1971, hZL, 368.

63. International Tables for X-Ray Crystallography; Kynoch: Birmingham, England, 1974, Vol. IV.

64. Program STEPT, Chander, J. P., Oklahoma State Univ. Quantum Chemistry Exchange Program, #307.

65. Gordon, A. J.; Ford, R. A. "The Chemist's Companion. A Handbook of Practical Data, Techniques, and References", Wiley: New York, 1976.

66. Cotton, F. A.? Kraihanzel, C. S.; J. Am. Chem. Soc.. 1962, M # 4432.

67. King, R. B. ? Efraty A. J. Oraanomet. Chem. 1969, 2SL, 264.

68. Fayer, M. D.; Harris, C. B. Inora. Chem. 1969, fi, 2792.

69. (a) Brown, T. L.? Darensbourg, D. J. Inora. Chem. 1967, £, 971. (b) Beck, W.; Melnikoff, A.; Stahl, R. Anaew. Chem. 1965, 21, 719. (c) Mannig, A. R. J. Chem. Soc. 1967, &, 1984.

174

70. Hieber, W.; Schulten, H. Z. Anoq. Alia. Chen. 1939, Ml, 164.

71. Hieber, W.? Schuh, R.; Fuchs, H. Z. Anora. Alia. Chem. 1 9 4 1 , 21£, 243.

72. Schmit, S. P.; Troger, W. C.; Basolo, F. Inora. Syn. 1 9 8 5 , Vol. 21, 41.

73. Anderson, G. K.? Cross, R. J.; Phillips, I. G. J. Chem. Soc.. Chem. Commun. 1978, 709.

74. Cotton, F. A.; Wilkinson, G. "Advanced Inorganic Chemistry", 3rd Ed.; Wiley: New York, 1972; Chapter 22.

75. Darensbourg, M. Y. Proa. Inora. Chem. 1985, 21, 221.

76. Pannel, K. H.; Raghu Veer, K. S.; DelBene, J. E.; Nathan, F. J. Am. Chem. Soc. 1987, 109. 4890.

77. Brown, D. W.; Nakashima, T. T.; Robenstein, D. L., J. Maan. Reson. 1981, 41, 302.

78. Jakobsen, H. J.; Sorensen, 0. W.; Brey, W. S.; Kanyha, P. J. Maan. Reson. 1982, Ifi, 328.

79. Robenstein, D. L.; Nakashima, T. T. Anal. Chem. 1979, 11, 1465A.

80. Maudsley, A. A.; Huller, L.; Ernst, R. R. J. Maan. Reson. 1977, 21, 463.

81. Bodenhausen, G.; Freeman, R. J. Maan. Reson. 1977, 21, 471.

82. Ibid. J. Am. Chem. Soc. 1978, 100. 320.

83. Benn, R.; Gunther, H. Anaew. Chem. Int. Ed. Enal. 1 9 8 3 , 21, 350.

84. Martin, G. E.; Zekter, A. S. Maan. Reson. Chem. 1988, 21, 631.

85. Bird, P. H.; Churchill, M. R. Chem. Commun. 1967, 777.

86. Cotton, F. A.; Wing, R. M. Inora. Chem. 1965, 4, 1328.

175

87. (a) Elian, M.; Hoffmann, R. Inora. Chen. 1975, 1±, 1058. (b) Elian, M; Chem, M. M. L.; Hingos, D. M. P.; Hoffmann, R. InS£SU_eh61IU. 1976, 15, 1148. (c) Schilling, B. E. R.; Hoffmann, R.; Lichtenberger, D. L. J. Am. Chem. Soe. 1979, 101. 585.

88. (a) Smith, R. A.; Bennett, M. J. Acta Crvstallor.

B/ 1977, fill, 1113. (b) Harrison, W.; Trotter, J« Jt Chem. SQC.. Dalton Trans. 1972, 678. (c) Kreissel, F. R.; Friedrich, P. Anaew. Chem. Int. Erf. Engl* 1977, l£, 543. (d) Einstein, F. W. B.; Tyers, K. G.; Sutton, D. Oraanometa11 i as 1985, 489. (e) Barrientos-penna, C. F.; Einstein, F. W. B.; Jones, T.; Sutton, D. Inora. Chem. 1982, 21, 2578.

89. Fitzpatrick, P. J.? Le Page, Y.; Butler, I. S. Acta ClYSt- 1981, fi21, 1052.

90. (a) Krebs, B.; Hasse, K. D. Acta Crvst. 1976, B32. 1334. (b) Lock, C. J. L.; Turner, G. Acta Crvst. 1975, fill/ 1764.

91. (a) Lokshin, B. v.; Rusach, E. B.; Kolobova, N. E.; Makarov, Y. v.? Ustynuk, N. A.; Zdanovich, V. I.; Zhakaeva, A. Z.; Setkina, V. N. J. Qraanomet. chem. 1976, JLQSL# 353. (b) Lokshin, B. V.; Pasinsky, A. A.; Kolobova, N. E.; Anisimov, K. N.; Makarov, Y. V. J. Qraanomet. Chem. ia73r 315.

92. Kristjansdottir, S. S.; Moody, A. E.; Weberg, R. T.y Norton, J. R. Orqanometallles. 1988, 7, 1783.

93. (a) Nesmeyanov, A. N.? Anisimov, K. N.? Kolobova, N. E.; Varyshikov, L. I. Izv. Akad. Nauk SSSR. Ser. Khim. 1963, 193. (b) Tam, W.; Lin, G. Y.; Wong, W. K.; Kiel, W. A.; Wong, W. K.; Gladysz, J. A. J. Am. chem. §<?g. 1982, 141.

94. (a) Reger, D. L.? Belmore, K. A.; Atwood, J. L.; Hunter, W. E. J. Am. Chem. Soc. 1983, 105. 5710. (b) Bullock, R. M.; Headford, C. E. L.? Kegley, S. E.; Norton, J. R. J. Am. Chem. Soc. 1985, 107. 727. (c) McNally, J. p.; Gluek, D.? Cooper, N. J. J. Am. chem.

1988, Hfl, 4838. (d) Jones, W. D.; Maguire, J. A. Oraanometa11les 1987, 1301.

176

95. (a) Jones, W. D.? Feher, F. I. J. Am. Chem. Soc. 1982, 101, 4240? 1985, lfll, 620? 1986, 10&, 4814. (b) Janowicz, A. H. ? Bergman, R. G. J. Am. Chem. soc. 1983, 105, 3929. (c) Periana, R. A.? Bergman, R. G. J. Am. Chem. Soc. 1986, 1£&, 7332? 7346. (d) Jones, W. D.? Feher, F. J. Inora. Chem. 1984, 21, 2376. (e) Periana, R. A.? Bergman, R. G. Oraanometalllcs 1984, 2, 508.

96. Lukehart, C. M. "Fundamental Transition Metal Organometal1ic Chemistry1", Brooks/Cole Publishing Co.: Monterey, CA. 1985.

97. Marcus, Y. "Ion Solvation", Wily: New York, 1985.

98. (a) Pannell, K. H.? Jackson, D. J. Am. Chem. Soc. 1976, 4443. (b) Hogen-Esch, T. E. ? Smid, J. J. Am. Chem. Soc. 1966, M , 307? 318. (c) Takaki, U. ? Hogen-Esch, T. E.? Smid, J. J. Am. Chem. Soc. 1971, 22, 6760. (d) Chan, L. L.? Wong, K. H.? Smid, J. J. Am. Chem. Soc. 1970, 22, 1955.

99. Mann, B. E. ? Taylor, B. F. "13C NMR Data for Organometallic Compounds", Academic Press: New York, 1981.

100. (a) Coates, G. E.? Wade, K. "Organometallic Compounds ; The Main Group Elements", 3rd Ed.? Methuen & Co.: London, 1967, Vol. 1. (b) Ashby, E. C. Quart. Rev. 1967, 21, 259. (c) Ashby, E. C. Pure APPI. Chem. 1980 , 52., 545.

101. (a) Schunn, R. A. in "Transition Metal Hydrides". Ed. by Muetterties, E. L.? Marcel Dekker: New York, 1971? CH. 5. (b) Edidin, R. T.? Sullivan, J. M.? Norton, J.

J. Am. Chem. Soc. 1987, 102, 3945. (c) Jordan, R. F.? Norton, J. R. ACS Svmp. Ser. 1982, No. 198, 403, and references therein, (d) Moore, E. J.? Sullivan, J. M. ? Norton, J. R. J. Am. Chem. Soc. 1986, IM, 2257. (e) Walker, H. W.? Kresge, C. T.? Pearson, R. G.? Ford, P* C. J. Am. Chem. Soc. 1979, 101, 7428. (f) Walker, H. W.? Pearson, R. G.? Ford, P. C. J. Am. chem. Soc. 1983, 10£, 1179.

177

102. Muller, A.; Krickemeyer, E.; Boggle, H. A. Anora. Allgt Ch lRt 1987. $§4, 61.

103. (a) Okuda, J.; Herdtweck, E.? Herrmann, W. A. Inora. Chem. 1988, 21, 1254. (b) Herrmann, W. A.; Floel, M.; Herdtweck, E. J. Oraanomet. Chem. 1988, 2 M , 321. (c) Hermann, W. A.; Fischer, R. A.; Herdtweck, E. Anaew. Chem. Int. Ed. Enal. 1988, 21, 1509? 1987, 23., 1263.

104. Kubacek, P.; Hoffamnn, R.? Havlas, Z. Oraanometallics 1982, 1, 180.

105. (a) Hermann, W. A.; Serrano, R.; Bock, H. Anaew. Chem. Int. Ed. Enal. 1984, 21, 383. (b) Hermann, W. A.; Voss, E.; Floel, M. J. Oraanomet. Chem. 1985, 231, C5.

106. (a) Amatore, C.; Verpeaux, J, N.; Krusic, P. J. Oraanometallics 1988, 2, 2426. (b) Kelly, R. S.; Geiger, W. E. Oraanometallics 1987, £, 1432. (c) Reike, R. D.? Henry, W. P.? Arney, J. S. Inora. Chem. 1987, 23., 420. (d) Cyr, J. C.; DeGray, J.A.; Gosser, D. K.? Lee, E. S.; Reiger, P. H. Oraanometallics 1985.

950. (e) Leong, V. S.; Cooper, N. J. Oraanometallics 1987, 3, 2000; 1988, I, 2080. (f) Maher, J. M.; Beatty, R. P.; Cooper, N. J. PrqanQmetalligg 1985, i, 1354.

107. (a) Albright, T. A.; Burdett, J. K.; Whango, M. H. "Orbital Interactions in Chemistry," Wiley: New York, 1985; pp. 282-288. (b) Purcell, K. F.; Kotz, J. C. "Inorganic Chemistry", W. B. Saunders Co.: Philadelphia, 1977; CH. 9.

108. (a) Saveant, J. M. &<?<?, Cheit 1980, 12, 323. (b) Andrieux, C. P.; Merz, A.; Seveant, J. M. J. Am. Chem. Soc. 1985, 107. 6097. (c) Andriex, C. P.; Gallardo, I.; Seveant, J. M.; Su, K. B. J. Am. Chem. Soc. 1986, 108. 638. (d) Kadish, K. M.; Lacombe, D. A.; Anderson, J. E. Inora. Chem. 1986, 25. 2074; 2246. (e) Moulton, R.; Weidman, T. W.? Vollhardt, K. P. C.; Bard, A. J. In<?rq. chemt i98«, 25., 1846. (f) Dessy, r. e.; Weissman, P. H.; Pohl, R. L. J. Am. Chem. Soc. 1966, 88. 5117. (g) Connelly, N. G.; Geiger, W. E. Adv. Qrqan<?roett chgmt 1984, 21, i*

178

109. (a) Bard, A. J.? Faulkner, L. R. "Electrochemical Methods'*, Wiley: New York, 1980; CH. 11. (b) Nicholson, R. S.; Shain, I. Anal. Cham. 1964, 2£, 706; 1965, 12, 178.

110. (a) Grutzner, J. B.; Lawlor, J. M.; Jackman, L. M. J, Afflt Chgffit ?<?<?, 1972, 21, 2306. (b) Ashby, E. C. ; Dobbs, F. R.; Hopkins, H. P., Jr. J. Am. Chem. Soc. 1973, 25, 2823.

111. (a) Hudlicky, M. "Reductions in Organic Chemistry", Ellis Horwood: Chichester, 1984. (b) Kogure, T.; Ojima, I. J. Orqanomet. Chem. 1982, 221, 249. (c) Ojima, I.; Kogure, T. "Rev. Silicon, Germanium, Tin, and Lead Compounds", 1981, £, 6. (d) Ojima, I.; Yamamoto, K.; Kumada, M. "Aspects of Homogeneous Catalysis Vol. 3", D. Reidel Pub. Co.: Holland, 1977; CH. 3.

112. Jackson, R. A. J. Orqanomet. Chem. 1979, 1££, 17.

113. Colvin, E. W. Butterworths Monographs in Chemistry and Chemical Engineering "Silicon in Organic Synthesis", 1981, p 4.

114. Nolan, S. P.; Hoff, C. D.; Stoutland, P. O.; Newman, L. J.; Buchanan, J. M.; Bergman, R. G.; Peters, K. S. Jt Afflt Chem, §Q9t 1987, 1 £ 2 , 3143.

115. Calado, J. C. G.; Dias, A. R.; Salem, M. S.; Marthinho Simoes, J. A. Jt Chgffit , PaltOfl Iran? 1981, 1174.

116. Morrison, R. T.; Boyd, R. N. "Organic Chemistry", 3rd Ed.; Allyn and Bacon: Boston, 1974; CH. 1.

117. Connor, J . A. "Topics in Current Chemistry", 71. Springer-Verlag: New York, 1977; pp71-110.

118. (a) James, B. D.; Wallbridge, M. G. H. Proa. Inora. Chem. 1970, H , 99. (b) Brown, H. C.; Khuri, A.; Krishnamurthy, S. J. Am. Chem. Soc. 1977, 92, 6237.