‘..-~
‘1‘"

   

‘......u--

 

 

RARIES

W

 

 

 

mas..- W“ W\\\\\\\\\\\\‘\ H

31293

i i

This is to certify that the

dissertation entitled
SYNTHETIC AND MEGiANISTIC ASPECTS OF ZIEGLER-NA'ITA
POLYMERIZATION: REGIO- AND DIASTEREOSELECI' IVE
FORMATION OF FIVE- AND SIX-MEMBERED CARBOCYCLES

presented by

Jonathan R. Young

has been accepted towards fulfillment
of the requirements for

Ph. D. Chemistry

degree in

 

 

Dr. J. R. Stille

 

 

Major professor

Date 6/29/92

MSU is an Affirmative Aaron/"Equal Opportunity Institution 0-12771

.———

._____

 

 

_—» ”hi—F

 

 

LIESRAEW
Ratetgan State
University

 

 

 

PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before duo duo.

DATE DUE DATE DUE DATE DUE

--i
T
L__\

       
  

 

 

 

 

 

 

 

 

 

 

 

 

 

3
D

 

 

 

 

 

 

U

[,1
[—_—l F

J i J

usu lo An Affirmative AotioNEqual Opportunity Institution

 

 

 

 

 

 

 

C1

 

 

 

 

 

 

7:“‘7—"7

 

EWMHT

 

SYNTHETIC AND MECHANISTIC ASPECTS
OF ZlEGLER-NATTA POLYMERIZATION:
REGIO- AND DIASTEREOSELECTIVE FORMATION
OF F IVE- AND SIX-MEMBERED CARBOCYCLES

By

Jonathan Russell Young

A DISSERTATION

Submitted to _
Michigan State Universuty
in partial fulfillment of the requurements
for the degree of

DOCTOR OF PHILOSOPHY

Department of Chemistry

1992

{7% {W7

ABSTRACT

SYNTHETIC AND MECHANISTIC ASPECTS
OF ZIEGLER-NA'ITA POLYMERIZATION:
REGIO- AND DIASTEREOSELECTIVE FORMATION OF FIVE- AND SIX-
MEMBERED CARBOCYCLES

By

Jonathan Russell Young

An ongoing project in our group involves the development and application
of novel ring-forming reactions that feature intramolecular addition of reactive
carbon intermediates to unactivated olefins. Although ring-forming methods
utilizing this synthetic design are known, none of these methods were able to
combine high regio- and stereoselective qualities that spanned applicability to
synthesis of both five- and six-membered rings. Our objective was to develop a
general, complimentary ring-tanning method that exhibited high selectivities for a
variety of ring sizes. We focused our attention on the use of organometallic
intermediates as a means of directing both the regio- and stereochemistry. The
organometallic complexes chosen were those of alkyltitanocene chlorides, which
are well known for undergoing efficient and stereoselective tat-olefin
polymerization.

Our initial goal was to ascertain the regioselectivity of olefin insertion into
the carbon-titanium bond for the unsubstituted 5—hexen-1-yl- and 6-hepten—1-

yltitanocene chlorides. These alkyltitanocene derivatives were stable for weeks

and exhibited no tendency toward olefin insertion. However, activation with
EtAlCIz drastically altered the electronic environment surrounding the metal and
the reactivity of these complexes towards olefin insertion. As a consequence of
the Lewis acid additive, facile olefin insertion resulted with exclusive exo
specificity in each case, providing methylcyclopentane (5-exo-trig, 95% yield) and
methylcyclohexane (6-exo-trig, 75% yield) after hydrolysis.

We next turned our attention to the stereochemical consequence of the
carbon-carbon bond formation with respect to an alkyl substituent on the tether.
In the case of cyclopentane formation, selectivity ranged from 97:3 to >99:1 for 2-,
3-, and 4-methyl-5-hexen-1-yltitanocene chlorides. The stereocontrol exhibited for
the six-exo-trig cyclizations were more complex, due to the greater flexibility of the
alkyl tether. Cyclization of 4- and 5-methyI-6-hepten-1-yltitanocene chlorides
resulted in >98:2 diastereomeric ratio of dimethylcyclohexanes. However, when a
methyl group was [3 or y to the metal, cyclization was less successful (82:18, and
50:50 trans:cis, respectively). Yet with an isopropyl group in these positions,
enhanced olefin facial selectivity was observed (92:8, transzcis and 74:26,
cisztrans, respectively).

In the course of deveIOping this synthetic methodology, we noticed that
olefin insertion was more facile for a certain substitution pattern. The substrates
that appeared to be more reactive had a common structural feature, an alkyl
substituent on the carbon [3 to titanium. We have developed a series of
competitive rate experiments to determine if a- or B—hydrogen activation plays a
role in the cyclization process. At this stage of the project, optimization of the

experimental conditions is in progress.

To Mom, Dad, Nancy, and Karl

Mom and Dad, thank you for always being there when I needed you. You
two have been terrific parents and have always been my biggest supporters. '
From every hockey, baseball, or football games, to drama and band concerts,
you were always there. I can never adequately express what your love and
support has meant to me.

Nancy, thank you for your patience and allowing me to fulfill my dreams
during the past years. I know these years have been difficult for us, but let's be
happy we survived them, together. I am sorry that we didn't always have as
much time with each other as we wanted, but remember, brighter days and
opportunities await us in Madison. Let's have some fun as a family.

Karl, thank you for all the happiness you constantly bring into our lives. I

want you to never be afraid to chase your dreams.

ACKNOWLEDGMENTS

I would first of all like to thank the anonymous person in the Natural
Resources office that studied my application so thoroughly during the months of
February to late October of 1987. I made it here anyway!

I will be forever indebted to Dr. Stille (that is John, now) for all that he
taught me. I will always miss our long conversations in the morning over a cup of ‘
coffee, weightlifting and running at the crack of dawn, and the enthusiasm that
we shared together in pursuit of common goals. Chemistry aside, your friendship
is something very special to me. Thanks for all that you have done and given to
me. Please stay in touch in the years to come.

I want to wish the best to all of my friends in the Stille group. Greg and
Art, good luck at Stanford and say hi to Barry for me. Polymervanan, thanks for
the midnight brainstorming, it was a lot of fun.

Words cannot express the wonderful friend and person I met at MSU.
Nancy, you provided a wonderful spark to not only me, but to the whole group.
All the help and support you gave me over the last two years meant so much to
me. I clearly could not have survived this dissertation without your proofreading
skills or your daily positive attitude and kind words of encouragement. Thank you
for all you have done and most of all, for believing in me. My morning donut will
certainly not be the same anymore but it will remind be of countless special
mornings we shared. How will I cool my coffee without you? You are the best

friend I could ever have asked for and I will miss you very much.

TABLE OF CONTENTS

LIST OF TABLES ..................................................................................................... ix
LIST OF SCHEMES ................................................................................................ xi
LIST OF FIGURES ................................................................................................... xiii
LIST OF ABBREVIATIONS ..................................................................................... xiv
INTRODUCTION

1. Outline of the Synthetic Problem ............................................................ 1

2. Background on the Synthetic Aspects of Current Cyclization
Methodology .............................................................................................. 4
A. Carbomtionic Cyclizations ........................................................... 5
B. Free Radical Cyclizations .............................................................. 8
C. Lithium-Mediated Cyclizations ..................................................... 17
D. Magnesium-Mediated Cyclizations ............................................. 22
E. Aluminum-Mediated Cyclizations ................................................ 24
F. Palladium-Mediated Cyclizations ................................................ 25
3. Background for the Mechanism of Zregler-Natta Polymerization ..... 26

CHAPTER 1. SYNTHESIS OF CYCLIZATION PRECURSOR

SUBSTRATES AND STANDARDS

1. Introduction ................................................................................................. 34
2. Synthesis of Five-Membered Ring Precursor Substrates ................. 35
3. Synthesis of Six-Membered Ring Precursor Substrates ................... 37
4. Synthesis of Competitive Rate Substrates ........................................... 4O
5. Conclusion ................................................................................................. 43

vi

6. Experimental .............................................................................................. 44

CHAPTER 2. DIASTEREOSELECTIVE CYCLOPENTANE
FORMATION MEDIATED BY TITANOCENE
DICHLORIDE

1. Introduction ................................................................................................. 75

2. General Features and Limitations for the Synthesis of Methyl-
Substituted 5-Hexen-1-yltitanocene Chlorides ................................... 78

3. Synthesis and Cyclization of Methyl-Substituted 5-Hexen-1-
yltitanocene Chlorides ............................................................................. 80

4. Comparison of the Stereoselective Outcome of Various Methods. 82

5. Optimization and Cyclization of 133: Formation of

Bicyclo[4.3.0]nonane ................................................................................ 84
6. Conclusion ................................................................................................. 86
7. Experimental .............................................................................................. 87

CHAPTER 3. DIASTEREOSELECTIVE CYCLOHEXANE
FORMATION MEDIATED BY TITANOCENE

DICHLORIDE
1. Introduction ................................................................................................. 91
2. Optimization Studies for the Cyclization of 6-Hepten-1-yltitanocene
Chlorides .................................................................................................... 92
3. Potential Explanations for the Observed Stereoselectivities ............ 102
4. Conchsion ................................................................................................. 104
5. Experimental .............................................................................................. 106

CHAPTER 4. MECHANISTIC STUDIES OF ZlEGLER-NATI'A

POLYMERIZATION: PROBING FOR a- OR B-
HYDROGEN ACTIVATION

1. Introduction ................................................................................................. 113
2. Preliminary findings for the 2-Alkyl-5-hexen-1-yltitanocene
Chlorides .................................................................................................... 116

3. Optimization Studies for the Competitive Cyclization of 2-AlkyI-6-
hepten-l-yltitanocene Chlorides with MezAlCl, Mng, and MAO.... 118

4. Conclusion ................................................................................................. 133

vii

5. Experimental .............................................................................................. 134
CONCLUSION ...................................................................................................... 138
LIST OF REFERENCES .................................................................................... 139

viii

Table 1.
Table 2.

Table 3.
Table 4.
Table 5.

Table 6.
Table 7.
Table 8.
Table 9.

Table 10.

Table 1 1.
Table 12.
Table 13.

Table 14.

Table 15.

Table 16.

LIST OF TABLES

Regioselectivity of Atom Transfer Reaction ................................... 14
Diastereomer Ratios of Dimethylcyclopentane Products

via Trtanium and Free Radical Methods ......................................... 79
Comparison of Diastereoselectivities for Different Methods ........... 84
Optimization of the Cyclization Conditions for 133b ....................... 85
Diastereomer Ratios of Cyclohexane Products via

Titanium and Free RadicalMethods ............................................... 94
Optimization of the Cyclization Conditions for 155b ....................... 96
Optimization of the Cyclization Conditions for 164b ....................... 97
Optimization of the Cyclization Conditions for 160b ....................... 98
Optimization of the Cyclization Conditions for 137b ....................... 99
Diastereomer Ratios of Cyclohexane Products Performed

Under the Optimized Conditions .................................................... 101
Cyclization of Unpurified 175b Using Mng or MAC ...................... 119
Cyclization of Purified 168b and 175b Using M9X2 ....................... 121
Cyclization of Alkyltitanocene Chlorides Promoted with

MezAICI .......................................................................................... 123
Cyclization of Alkyltitanocene Chlorides Promoted with

MAO ............................................................................................... 124

Competitive Cyclization of 8-Hydrogen vs 8-Hydrogen
(168b vs 175b) Promoted Under the
Optimized Conditions for MAO ....................................................... 127

Competitive Cyclization of 8-Hydrogen vs B-Deuteron
(168b vs 178b) Promoted Underthe
Optimized Conditions for MAC ....................................................... 128

ix

Table 1 7. Competitive Cyclization of BDeuteron vs B-Hydrogen

(171b vs 175b) and or,or-Dideuteron vs B—Hydrogen
(184b vs 175b) Promoted Under the
Optimized Conditions for MAO ....................................................... 129

Scheme 1.
Scheme 2.
Scheme 3
Scheme 4.
Scheme 5.

Scheme 6.
Scheme 7.
Scheme 8.
Scheme 9.

Scheme 10.
Scheme 1 1.
Scheme 12.
Scheme 13.
Scheme 14.
Scheme 15.

Scheme 16.
Scheme 17.
Scheme 18.
Scheme 19.
Scheme 20.

LIST OF SCHEMES

Regioselectivity of Carbocationic Cyclizations ........................... 6
Stereoselectivity of Carbocationic Cyclizations .......................... 6
Regioselectivity of Free Radical Cyclizations ............................. 10
Stereoselectivity of Free Radical Cyclizations ........................... 11
A Possible Propagation Sequence for the Atom

Transfer Reaction ....................................................................... 12
Evidence for Proton Abstraction ................................................. 18
Regioselectivity of Alkyllithium Cyclizations ............................... 20
Stereoselectivity of Alkyllithium Cyclizations .............................. 20
Stereoselectivity of Vinyllithium Cyclizations .............................. 21
Stereoselectivity of Aluminum Cyclizations ................................ 24
The Cossee-Arlmann Mechanism for Chain Propagation .......... 28
The Green-Rooney Mechanism for Chain Prepagation ............. 29
Grubbs' Test for or-Hydrogen Activation ..................................... 32
Bercaw's Test for (l-HYdfOQSII Activation ................................... 33
Retrosynthetic Analysis of the T‘rtanocene Dichloride-Mediated
Carbocyclization ......................................................................... 34
Synthetic Route to 125 ............................................................... 35
Synthetic Route to 129 ............................................................... 36
Synthetic Route to 133 ............................................................... 37
Synthetic Route to 137 ............................................................... 38
Synthetic Route to 144 and 151 ................................................. 39

xi

Scheme 21.
Scheme 22.
Scheme 23.
Scheme 24.
SCherne 25.
Scheme 26.
Scheme 27.

SCheme 28.
Scheme 29.
Scheme 30.

Scheme 31.

Scheme 32.

Scheme 33.

Synthetic Route to 155 and 160 ................................................. 39

Synthetic Route to 164 ............................................................... 40
Synthetic Route to Competitive Rate Substrates ....................... 42
Synthetic Route to Cyclohexane Standards ............................... 43
Synthetic Route to 196 and 197 ................................................. 43
Preparation of 5-Hexen-1-yltitanocene Chloride (198) ............... 76
Potential Pathways of Cyclization and Decomposition

of198 ......................................................................................... 76
Proposed Transition States for Five-Membered Ring

Formation ................................................................................... 82

Comparison of Titanium-Mediated Cyclization to Free
Radical Cyclization for 133a ....................................................... 86

Possible Transition States for Six-Membered Ring
Formation ................................................................................... 104

Mechanistic Possibilities for Ziegler-Natta
Polymerization: or- and B—Hydrogen Interactions ........................ 115

Competitive Rate Experiments Designed to Test the
Mechanistic Possibilities ............................................................ 116

Cyclization of 5-Hexen-1-yltitanocene Chlorides
Performed Under the Transmetallation Conditions .................... 117

xii

LIST OF FIGURES

Figure 1. a-Agostic Interaction for 107 ............................................................ 30
Figure 2. B-Agostic Interaction for 108 ............................................................ 30
Figure 3. B-Agostic Interaction for 109 ............................................................ 30
Figure 4. Competitive Cyclization Between 168b and 175b ........................... 131
Figure 5. Competitive Cyclization Between 168b and 178b ........................... 131
Figure 6. Competitive Cyclization Between 171b and 175b ........................... 132
Figure 7. Competitive Cyclization Between 184b and 175b ........................... 132

xiii

AIBN
Bu
n-BuLi
tBuU
CeHa
CP
mCPBA

DiBAI
DCE
DIPA
DMAP
DMF
DMSO
Et
EtzO
LAD
LAH
LDA
MAO
Me

LIST OF ABBREVIATIONS

Azobisisobutyronitrile

Butyl

n-Butyllithium
fen-Butyllithium

Benzene

Cyclopentadienyl
meta-Chloroperbenzoic acid
Deuteron (2H)
Diisobutylaluminum hydride
1 ,2-Dichloroethane
Diisopropylamine
4-Dimethylaminopyridine
N,N-Dimethylformamide
Methyl sulfoxide

Ethyl

Diethyl ether

Lithium aluminum deuteride
Lithium aluminum hydride
Lithium diisopropylamide
Methaluminoxane (O-Al(Me))n
Methyl

xiv

Ms
NBS
Ni[COD]2
Ph
PMDTA
Pr

iPr
TBAF
TBDMS
TEA
THF
TMEDA
SET

Methanesulfonyl

N-Bromosuccinimide

Nickel cyclooctadiene

Phenyl

N, N, N', N", N"-Pentamethyldiethylenetriamine
Propyl

lsopropyl

Tetrabutylammonium fluoride
fert-Butyldimethylsilyl

Triethylamine

Tetrahydrofuran
N,N,N',N'-Tetramethylethylenediamine

Single Electron Transfer

INTRODUCTION

1 . Outline of the Synthetic Problem

Carbocycles of the quinane and steroid families have presented
chemists with a number of interesting synthetic challenges. The efforts exerted
toward the preparation of these medicinally useful compounds have benefited
chemists and non-chemists alike. Upon examination of the complex
architectural framework of the quinanes and steroids, a common theme woven
throughout these families of compounds was the repetition of five- and six-
membered ring subunits, fused together in a variety of regio- and
stereochemical patterns. To effectively launch an assault on quinane or steroid
targets, a strategic plan must be devised for the efficient construction of the
carbocyclic skeleton. This plan should address, not only the preparation of five-
and six-membered rings but to also incorporate the numerous stereoconters as
well. An inspired plan would be one that could achieve both of these goals
simultaneously, thus minimizing the number of manipulations. Although many
valuable ring-forming methods exist for achieving any one of these goals, a
single, general method applicable to five- and six-membered rings with good
control over regio- and stereochemistry would be useful. Our interest in
preparing compounds of the above variety has compelled us to develop a
general, yet selective, ring-forming method capable of fulfilling the

aforementioned criteria. Our efforts have concentrated on

2
developing organotransition metal reagents to direct selective, intramolecular

carbon-carbon bond construction.‘

The interest in organometallic methods has been building ever since
Grignard discovered the landmark reaction between alkyl halides and
magnesium. The crucial observation was the novel change in the polarity of
carbon (umpolung), and the useful reversal in the chemical reactivity of carbon.
Grignard's efforts blossomed into an exciting new approach to carbon-carbon
bond formation. Early in the development of organometallic chemistry, chemists
were laying the groundwork, reacting various metals together with different
organic substrates in an apparent random fashion. What resulted from this ”hit
or miss" approach was a firm foundation for understanding the interaction and
subsequent reaction of organic functional groups with metal complexes.2 As a
result, very predictable methods for engineering carbon-carbon bond
construction utilizing organometallic reagents have evolved. The sophistication
of of organometallic reagents for carbon-carbon bond formation lies in the
ability to fine tune the reactivity, and thus selectivity, of the metal complex by
varying the steric and electronic environment surrounding the metal. This can
be accomplished by judicious choice of the metal, its oxidation state, and the
surrounding ligands.

One of the most useful and exploited organometallic processes is the
reaction of alkyl-metal complexes with olefins. Certain electron deficient metal
complexes, which have a vacant coordination site available, can form a 1:-
complex with olefins and can often be induced to insert the olefin into the metal-
carbon bond. In many cases, the carbon—carbon bond could be formed in a
highly regio- and stereoselective manner. A good example is that of Ziegler-

Natta polymerization.3

3
In the early 1950's, Ziegler4 and Narta5 independently discovered that

titanium salts, in combination with alkyl aluminums, facilitated the
polymerization of ethylene. The active polymerization catalyst was suggested
to be an alkyltitanium species, formed by alkyl ligand exchange from aluminum
to titanium. This proposal was later substantiated after the combined action of
ethyltitanocene chloride6 and ethylaluminum dichloride was shown to
polymerize ethylene.7

Several reports have indicated that Ziegler-Natta chemistry should be
well suited to ring forming reactions. In 1958, Marvel reported that 1,5-
hexadiene was polymerized with TrCl4 and an alkyl aluminum co-catalyst (eqn.
1).821 The polymer contained repeating units of methylenecyclopentanes,
connected in a 1,3 relationship. Similarly, 1,6-heptadiene resulted in a polymer
of methylenecyclohexanes under the same conditions (eqn. 2). Thus, each
intermolecular olefin insertion was followed by an intramolecular insertion
resulting in carbocycle formation. At this point, the stereochemical details of this
reaction have not been determined. More recently, Waymouth has shown that a
zirconocene catalyst, together with methaluminoxane (MAO), polymerized 1,5-
hexadiene in the same fashion as in eqn. 1, yet polymerization of 1,6-
heptadiene yielded only low molecular weight oligomers.8b There have been a
number of paramount reports in which monomer formation was realized as
well. It has been demonstrated that 5-hexen-1-ylmetal complexes (Metal a Ti,
Zr, Sc, Y)9 could be induced to isomerize to methylcyclopentylmetal products.
Specifically, Grubbs described that EtAIClg successfully catalyzed the
intramolecular olefin insertion of 5-hexen-1-yltitanocene chloride to provide,
after hydrolysis, high yields of methylcyclopentane (Scheme 13).93 In a similar
fashion, Schwartz observed that when transmetallation of 5 to aluminum was

attempted, methylcyclopentane was produced (eqn. 3).9b These results provide

4
convincing evidence to the viability of cyclopentane and cyclohexane formation

via organometallic intermediates, yet leave many questions unanswered

concerning the stereochemical details of the newly formed asymmetric center.

T'Cl4('B )3Al Me R
W ' ~ . . .
e (1)

 

 

 

1 Rs Hor iBu
M “C“ “BMW.” (2)
3 n-R Hor iBu
Me
2) AICIa
AN 1 ) szszIC') _W ZGCch fl. (3)
1 5 6

2. Background on the Synthetic Aspects of Current Cyclization
Methodology

A number of ring-forming methods have been creatively designed in
which a reactive carbon intermediate added across an unactivated olefin. The
preparative utility of intramolecular addition of carbon centered radicals, anions,
and cations, as well as covalently bound transition metals, to distal double
bonds has been demonstrated. However, each of these methods had certain
inherent limitations that diminished their applicability to general situations. An
overview of various ring forming methods will be provided, addressing such
issues as regioselectivity, stereoselectivity, generality to different ring sizes, and
the ability to functionalize the final product. This discussion will begin by
examining both carbocationic and anionic, as well as free radical approaches to

carbocyclization. Lastly, organometallic carbocyclizations, in which themetal

5
was covalently bound to carbon, will be described for metals such as

magnesium, aluminum, and palladium.

A. Carbocationic Cyclizations

The field of carbocationic cyclization was very active during 1960-75
because of its resemblance to nature's way of producing complex ring systems,
such as steroids.10 Both the regio- and stereochemistry were well defined and
predictable. Generally, cyclization proceeded in an endo manner to form the
larger ring, and a more stable carbocation (Scheme 1).10 For example,
cyclization of 7 formed a cyclohexane (9) rather than methylcyclopentane.
However, five-membered rings have been formed when the substrate contained
an appropriately placed substituent that could direct exo cyclization by
producing a more stable carbocation. Thus, cyclization of 10 produced the
product of exo addition (11), possibly due to the steric restriction of the endo
pathway. The stereochemistry was found to be dependent on the
stereochemistry of the olefins (Scheme 2).10 Thus, a polyolefinated substrate
that possessed trans olefins situated in a 1,5 relationship (see 13) were found
to undergo cyclization to produce trans fused six-membered rings (14).
Similarily, cyclization of substrates with the corresponding cis arrangement (see

15). resulted cis ring junctions (16).

6
Scheme 1. Regioselectivity of Carbocationic Cycllzations

(+)
O endo addition 0 B—elimination
I » +) :7

 

 

7 8 9
M0 M0 M9 (+)MO M9 MO
I? ’) exo addition 6 B-elimination “°
1 O 1 1 1 2

Scheme 2. Stereoselectlvlty of Carbocationic Cycllzatlons

H (+) H

 

 

‘ \
H I
13 14
H H
16M

 

Two other important design elements were crucial to cationic cyclization.
Specifically, these two elements were a carbocation initiator and terminator.
Treatment of a polyolefinated compound with a protic acid was found to be a
poor method for selective carbocation formation due to the propensity of the
acid to protonate indiscriminately, and to isomerize olefins.1o However, the
advent of Lewis acids in combination with either an acetal, epoxide, or an allylic
alcohol, led to clean formation of the desired carbocation. The main function of
the terminator was to direct the size of the final ring formed. Common
terminators that have been used were alkyl, allyl, benzyl, or 8-silyl groups. Alkyl
groups proved useful for directing the size of the final ring, however, unselective

B—elimination resulted in a number of alkene isomers (see 12, Scheme 1). Allyl

7
and benzyl terminators alleviated this problem, but difficulty encountered in

attempting to remove these unnatural steroid entities overshadowed this small
success. Silyl groups were able to overcome both of these difficulties, owing to
their well-known propensity to stabilize B-carbocations and to undergo B-silyl
elimination. Thus, the strategic location of the silyl group served to selectively
direct ring closure to form a B-carbocation, followed by rapid B-silyl elimination

providing a single structural isomer.

Me
Me

Me SnCl4
END"). -80 °C

70% Yleld ’

 

OH

 

17

. Johnson and co-workers were the first to apply the above principles for
the complete formation of the steroid skeleton in a single tandem cyclization
(eqn. 4).") Thus, treatment of allylic alcohol 17 with tin chloride resulted in the
formation of three new rings in better than 70% yield (18). More importantly,
five new chiral centers were produced that possessed the correct relative
configuration of 16,17-dehydroprogesterone. More recently, Burke has used
this methodology, equipped with a dioxane initiator and a vinyl silane
terminator, as the key step in the synthesis of the decalin ring systems for
mevinic acid (eqn. 5),“3 nagilactones,11b (+)-dihydrocompactin and (+)-
compactin,11c (+)-fragolide,11d and (-)-perinporin B."d

Another attractive feature of carbocationic cyclization was the ability to
affect enantioselective bond formation.12 The simplicity with which chirality was

introduced into the cyclization precursor was a major advantage. Thus, an

8
achiral alkenal was made chiral by trivial transformation to an acetal using a

chiral diol (21b, eqn. 6). Cyclization of 21b resulted in the normal ratio of
diastereomeric isomers (see cyclization of 21a) but with an excellent
enantiomeric excess. The opposite absolute configuration was also available

by using the other enantiomer of the diol.

m R HO(CH2)3(_) n
= H

O O
5

  
  

 

 

 

 

M0 Me
Ter. -78 °C _
v , (5)
74% YIBId 5
ii
TMS
1 9 2 0
Me
4.
> z :
o ; fr
0),, Hoc,H,o HOCzH‘O
a ‘3"
21a; R -H 228;90(1:1) 23a;10(1:1)
21b; R . CH3 . 22b; 90 (93:7) 23: 10 (94:6)

8. Free Radical Cyclizations

Free radical cyclization has enjoyed extensive success in the application
to natural product synthesis.13 Far and away, the best attributes of this method
were the mild conditions, functional group tolerance, and wide array of radical
precursors. For example, remote functional groups such as alcohols, amines,
ketones, esters, ethers and chlorides were stable to the conditions. Precursors

to carbon radicals that have been exploited include bromides, iodides, alkynes,

9
alkyl and aryl sulfides, alkyl and aryl selenides, and nitro groups. In terms of

stereoselectivity, the most successful examples of free radical cyclization have
usually contained a structural feature, such as a ring, that restricted the freedom
of rotation of the alkene or the tether. This limitation was a result of the minimal
alkene facial selectivity, or lack of conformational control, exhibited by an alkyl
radical. Curran and Fraser-Reid have utilized this approach to solve quinane
structural problems. Curran has prepared a number of angular,”a linear,14b
and propellane14c triquinane natural products using a tandem cyclization
approach, as illustrated for an intermediate to crinnipelin (eqn. 7).143 In this
case, Curran took advantage of the kinetic propensity of the sequential alkyl
radicals to react from the same face of the cyclopentane ring, thus controlling
the relative stereochemistry of the ring fusion. Fraser-Reid has controlled the
absolute stereochemistry by performing similar serial cyclizations (eqn. 8),153 in
which the quinane precursor was secured to a chiral carbohydrate template. In

this way, enantiomerically pure quinanes were produced.

Me Buaan. AIBN
Benzene, 80 °C>

 

   

24:; x . CHzl 51% Yleld 25:; v . H, H
24b; X-COSOPII 62% Yleld 25b; v .. o

1) 803$nH
2) Silica Gel
73% Yleld

(8)

   

In the simplest case, radical cyclization was effective for selective carbon-

carbon bond formation (Scheme 3).13 For example, cyclization of 28 provided

10
a 98:2 ratio of exo:endo products. However, the regioselectivity of ring closure

was dramatically reduced when a geminally disubstituted olefin (30) was
employed. Thus, the important terpene element, the gem-dimethyl group, was
not accessible via this route. Moreover, extention to six-membered rings via
exo cyclization (33) has been a troublesome operation for entropic reasons,
providing only 11% cyclization contaminated by a trace of the endo product as
well.16 In this case, the lifetime of the radical was too short for the significantly
slower cyclization to occur prior to hydrogen-atom abstraction. For the sake of
clarity, the rates of 5-exo-trig cyclization, six-exo-trig cyclization, and hydrogen
atom transfer are, (k=2x105, 25 °C),17 (k=5x103, 25 °C),‘7 (k=6x106, 25 °C),18
respectively, which help explain the different levels of success for the two

reactions.

Scheme 3. Regioselectivity of Free Radical Cyclizations

War m"—”—~°" G O

23 D (98:2)29
M0 MD
M ”arr-W, 6 O
9' car.
,30 31 (40:60) 32

craftsmen Hm~+ 0+. M..
N\/\/

34(89) 32 (10) 35(1)

Substrates which possessed an alkyl group on the tether suffered a lack
of stereoselectivity for radical ring closure (Scheme 4).19 The
diastereoselectivities for bromides 36a-c ranged from 2:1 to 4:1, with the major

product arising by cyclization from the more stable chair conformation.20

10
a 98:2 ratio of exo:endo products. However. the regioselectivity of ring closure

was dramatically reduced when a geminally disubstituted olefin (30) was
employed. Thus, the important terpene element, the gem-dimethyl group, was
not accessible via this route. Moreover, extention to six-membered rings via
exo cyclization (33) has been a troublesome operation for entropic reasons,
providing only 11% cyclization contaminated by a trace of the endo product as
well.16 In this case, the lifetime of the radical was too short for the significantly
slower cyclization to occur prior to hydrogen-atom abstraction. For the sake of
clarity, the rates of 5-exo-trig cyclization, six-exo-trig cyclization, and hydrogen
atom transfer are, (k=2x105, 25 °C),17 (k=5x103, 25 °C),17 (k=6x105, 25 °C),18
respectively, which help explain the different levels of success for the two

reactions.

Scheme 3. Regioselectivity of Free Radical Cyclizatlons

A/2V\a.c,-.-.,—-> BEL-“B" b O

6 (98:2)29
Me Ms
M, W “‘5‘, g
.30 31 (40:60) 32

arBufirHAreN HmM+ 0’". M0.
W3\/\/

34 (89) 32 (10) 35 (1)

Substrates which possessed an alkyl group on the father suffered a lack
of stereoselectivity for radical ring closure (Scheme 4).19 The
diastereoselectivities for bromides 36a-c ranged from 2:1 to 4:1, with the major

product arising by cyclization from the more stable chair conformation.”

11
Scheme 4. Stereoselectlvlty of Free Radical Cyclizations

 

R2 R‘ 51
Bu3$nH,AIBN "3
We. 06H6,95% Yleld M“ n2 W ....in2
> +
at n3 .
0"
R3
36 37 F" 38
B‘ 32 32
e. Me H H 17 as
b: H Me H 75 25
c H H Me 35 65

In their current form, radical cyclizations have practical and
environmental limitations as well. In order to promote intramolecular events
over hydrogen abstraction or dimerization, concentrations of <0.01 M were
required. Industrial application of this technique would be costly and
cumbersome in terms of solvent requirements, recovery or disposal of solvent,
and the handling and disposal of the toxic tin byproducts.

The continued efforts of many researchers have increased the level of
sophistication and applicability of free radical processes. As a result,
”yesterday's limitations” are becoming obsolete. One major problem of the tin-
hydride method was that substituents surrounding the radical had little effect on
the rate of hydrogen-atom abstraction. Consequently, most alkyl radicals had
similar lifetimes with respect to hydrogen-atom abstraction. For relatively slow
cyclizations, such as 6-exo-trig cylizations, hydrogen abstraction becomes
competitive with cyclization. Although not recognized at the time, a major
synthetic breakthrough happened in 1948 when Kharasch discovered what is
called today the ”atom-transfer reaction".21 This reaction delivers a C-X
(X=halogen) bond across a double or a triple bond. The attractive feature of this
reaction was that it was devoid of a hydrogen atom donor. Consequently,
longer lifetimes of the incipient radicals were permitted without interception by a

hydrogen-atom donor, which in turn allowed for slower inter- or intramolecular

12
events to occur. Finally, predictable and efficient five-, six-, or seven-membered

ring formation at a reasonable concentration (0.3 M) was at last a viable option

through radical methodology.22

Scheme 5. A Possible Propagation Sequence for the Atom
Transfer Reaction

I TMS
I
COzMa BngSnSnBu,
+ Ir 0.3 M. cfil_la 002m
TMS 55% Yield
3 9 4 3

l l

 

F ms ms ms I -
- K,
I/ . 002Me 002m I
.vrl
_ 4 o 4 1 4 2 "‘5‘

 

Curran has developed a number of new strategies based on the atom
transfer principle. For example, a formal [3+2] annulation procedure was
designed that began by treating alkyliodide 39 with hexabutylditin and light in
the presence of an activated olefin (Scheme 5).23 The initial radical formed
(40) added to the olefin in a Michael fashion which produced 41, followed by
an intramolecular 5-exo-dig cyclization. The resulting vinyl radical 42, which
was less stable than the preceding ones, underwent iodine atom transfer from
the starting iodide which produced 43 and completed the chain process. The
key ingredients to the successful design of this procedure were the delicate
balance between the relative strength of the carbon-iodide bond, and the

reactivity of the carbon radicals along the way. An illustrative example was the

13
divergent behavior of 44 with 4-methyI-2-pentene (eqn. 9) compared to that

shown in Scheme 5. In this case, a two-step sequence was necessitated. In
the first step, the intermolecular atom transfer addition across the olefin could be
thermally promoted in the absence of the tin reagent, owing to the more labile
carbon-iodide bond, which provided a single regioisomer. This reaction
produced no trace of cyclized products, only the acyclic secondary alkyl iodide.
The greater reactivity of the 2° alkyl radical compared to the stabilized radical
41 coupled with the weaker carbon-iodide bond, resulted in premature iodide

atom transfer. Treatment of the 2° iodide with BU3SI‘IH yielded a single regio-

and stereoisomer, methylenecyclopentane 45.

 

I Me Me
Me 1) 80 °C- CGHB .-"“k
+ 2) Bu35nH . Me
| -» <9)
NC | 59% Yield NC
no M0 NC M0

44 45

14

 

 

 

 

 

 

 

 

 

Table 1. Regioselectivity of Atom Transfer Reactions
Products
Substrate EXO ENDO Yleid
(cis/francs) (cis/trans)
455; R . OM 47a (53/140) 483 (SI/4) 83
46b; R . Ph 478 (21/54) 48b(8/17) 68
o o o
i
a fi ,1 E l a
R I
498; R s H 508 (<3) 513 (>97) 56
498; R . Me 508 (25) 518 L75) 73
M020 1 MeOzc r M6020 a
G (5*; C1
I
5289 . H 532 (48/32) 543 (15/5) 68
5171: - Me 538(50/50) 548(0) 60
I E E
9’ n2
55 iii
R‘
5.0028 2 553(90 ) 57 (10) 86
553 R1 - H: R - H 8
55bR1-H;R2-Me £1300) “‘2’00 g:
55cH1-Me;n2-H ‘ 0°) 57“ I
e ,. e
R R‘
5.6028 555 (60) we (40) 66
532 R1 - H- R2 - H
1 '. 2 59b (10) 60b (0) 62

58c 111- Me; RZ-H

 

 

15
The results in Table 1 established guidelines for application of atom

transfer reactions to larger rings.22 Treatment of 463 under the atom transfer
conditions resulted in excellent regiocontrol (93:7, 47a:48a) favoring 5-exo
over 6-endo cyclization. A reversal in the mode of cyclization could be brought
about by placing the carbonyl on the interior of the forming ring (see 49). in this
case, endo cyclization was the dominant pathway (3:97, 50a:51a).
Calculations have shown that the product of this cyclization, like all other
intramolecular free radical reactions, was the kinetic product. The transition
state for this cyclization resembled that of the [3,3] reaction. Cyclization of 52a
resulted in more endo cyclization than for the cyclization of 46. This mode
could be suppressed by incorporation of a terminal alkyl group on the olefin
(52b). Curran has also developed atom-transfer cyclizations based on
iodomalonates. Cyclization of 553 and 58a demonstrated that for a terminal
alkene, irrespective of tether length, a mixture of exo and endo products were
formed. Each of these modes could be controlled and amplified by situating an
alkyl group on the interior (see 558 and 58b) or on the exterior (see 55c and
58c) of the olefin. Although these examples extend the applicability of radical
cyclization, control of simple diastereoselectivity remains a hurdle to overcome.
Boger25 has recently described a new modification for free radical
cyclization. which featured an acyi radical. Acyl radicals were shown to
effectively participate in tandem intramolecular reactions as well as
intramolecular-intermolecular and intermolecular-intramolecular sequences.
Treatment of 61 (eqn. 10) with the standard tin hydride conditions preceded
with initial 6-endo-trig cyclization, followed by a rapid 6-exo-dig cyclization. The
initial radical ring closure was consistent with Curran's results for a substrate
that possessed a carbonyl on the interior of the forming ring. However, Boger

maintained that endo cyclization was merely a result of the kinetic deceleration

16
of the 5-exo mode due to the flanking methyl group on the olefin. Although the

regiocontrol was excellent in this case, the minimal stereoselectivity at the ring
junction was a serious concern. Attempted epimerization of 62 resulted in only
a slight increase for the trans fusion. The 5-exo-trig cyclization of 63 (eqn. 11)
was very stereoselective, as expected for a rigid system. A more impressive
result was the efficiency and stereocontrol of the subsequent intermolecular
Michael addition of the resultant 2° radical; only a single stereoisomer of 64
was produced. Treatment of 65 (eqn. 12) under the radical conditions,
illustrated that the acyl radical bypassed the 4-exo and the 5-endo
intramolecular pathways to participate in a regioselective addition to methyl
acrylate. The stabilized radical that resulted underwent a 6-exo-trig cyclization

to produce 66 (1.1 :1) to complete this formal [4+2] annulation sequence.

Buaan AIBN

SePh

 

62; 58:42 (cisnrans)
0" Bu38nI-I, AIBN
. * II/ 53% Yield ’
'1 0 SePh
63

C02 Me
Bu38nI-I, AIBN
SOP" + olr 33% YIOId (12)

P" COzMe
Ph 65 66;1.1:1 (transzcis)

(11)

 

 

he

lo:

he:

17
A number of new and exciting topics in free radical chemistry include

macrocyclizations26 and "chiral radicals".27 Using their respective methods.
Boger, Curran, and Porter have have produced macrocycles with ring sizes as
large as 20 members in surprisingly high yields. Chiral auxiliaries have shown
promise for induction of enantioselective bond formation for both intramolecular
cyclizations, and for the acyclic stereocontrolled addition of radicals to electron
deficient olefins. All of these new developments in free radical chemistry have
greatly expanded the repertoire of bond forming reactions available to the

synthetic chemist.
C. Lithium-Mediated Cyclizations

Cyclization methods based on alkyllithiums have received considerable
attention on a number of fronts, including synthetic and mechanistic fronts.
Product analysis for the isomerization of 5-hexen-1-yllithium has been used as
a probe for determining whether cyclization proceeds via an anionic
mechanism28 or by single electron transfer.29 If nothing else, the debate which
ensued between Bailey (anionic) and Ashby (single electron transfer) has made
for some interesting reading. Suffice it to say that there is compelling evidence
to support each claim28 depending on the alkyl halide (1°, 2°, or 3°), the nature
of the halogen (I, Br, Cl), and the method used to generate the alkyllithium. A
number of features of this method are synthetically attractive. First, the requisite
alkyllithium could be generated by numerous methods, including (1) lithium-
halogen exchange.30a (2) lithium-tin exchange,3°b (3) Shapiro degradation of
tosylhydrazones,3°¢ and (4) electrochemical reduction of carbon-halogen”d or
carbon-carbon bonds.3°° Secondly, cyclization of a variety of substituted 5-

hexen-l-yl systems demonstrated complete regiospecificity,30a exo addition,

18
along with excellent and predictable stereoselectivities.31 Finally, the post-

cyclization alkyllithium could be functionalized with a variety of electrophiles,

providing useful functionality for further elaboration.28-3°a-3°°

Scheme 6. Evidence for Proton Abstraction

are—w 8 =8-

 

 

 

67 _ 68 J 69 70
e:X-H 100 0
eX-H 9 0
bX-D 91 0
eX-H 9 91

    

89% Yleld

Treatment of iodide 67 with 2.2 equiv of t-BuLi at -78 °C led to clean
generation of the alkyllithium 68 (Scheme 6).303 This species was stable
indefinitely at ~78 °C. Also, upon quenching the reaction mixture at -78 °C with
CH30H, no detectable amount of methylcyclopentane was observed. However,
quenching the reaction mixture with CH3OD, resulted in a 95% yield of 693-b
and only 91% deuterium incorporation (698). Thus, only an 87% yield of the
alkyllithium was actually present. These results outline the fundamental
limitation of this method, extreme reactivity of the alkyllithium, in which the most
obvious consequences were reduced cyclization yields and eventual
separation problems. Two possible explanations have been set forth for the low
yields of alkyllithium formation: (1) the organolithium generated abstracted a
proton from the t-butyl iodide formed in situ, competitively with the second
equivalent of t-BuLi and/or (2) proton abstraction from the diethyl ether

occurred.28 In either case, the extreme reactivity demonstrated by the

19
alkyllithium minimized the potential functional groups that could be present.

Diminishing the amount of diethyl ether with an equal increase in the pentane
concentration has been shown recently to be marginally effective in reducing
the amount of substrate reduction.32 Withstanding these drawbacks, warming
68 to ambient temperature caused rapid and selective methylcyclopentane
formation. The activation parameters for this isomerization have been
determined to be AH = 11.8 i 0.5 kcal/mol and AS = -30 i 2 eu, which reflects
the importance of temperature.303 Treatment of 72 under the same conditions,
produced only 6% methylcyclohexane (Scheme 7). However, with the use of a
complexant such as TMEDA, the extent of 6-exo—trig cyclization was increased
to 68%. The role of the Lewis base served to lengthen, and thus weaken, the
carbon-lithium bond. This, in turn, accelerated cyclization to the point of
becoming competitive with proton abstraction. Substrates which possessed
geminal substitution on the olefin (71) also required the use of Lewis bases to
accelerate the sluggish reaction. The result was the regioselective formation of
31. It is important to note the marked difference in selectivity between lithium
cyclization of 71, and the analogous cyclization of 30 by radical methods
(Scheme 3). Acyclic compounds with vicinal substitution on the olefin failed to

cyclize under any conditions.

20
Scheme 7. Regioselectivity of Alkyllithium Cyclizations

1)tBuLi 78°C 6
W 2m 0
' l
8996de

6 7 (100.0) 29
Me l)tBuLi, -78°C
M 2 TMEDA. 25 °c M6; gM.
35W '
' m Yield
7 1 31 (100: 0)
Me
/\/\/\/ I D—l'———>B”U 78°C
/ 2)TNEDA, 25 °c
72 34(32) 32 (68) 35 (0)

Cyclization of 73a-c provided the opportunity to probe the
stereochemical aspects of the reaction (Scheme 8).31 Cyclization in the
absence of any additives produced excellent diastereoselectivities in all cases.
If PMDTA was present in the reaction, a noticeable increase in stereoselectivity
was achieved. The enhanced selectivity was attributed to the lower

temperatures needed for cyclization in the presence of PMDTA.

Scheme 8. Stereoselectlvlty of Alkyllithium Cyclizations

(5....

32
1)iBuLiI 48$
W's) 38°C M8 H2 M. “1le
1
R R3 “en:
73 3 8
Adam— IIIIII 3‘ 3’ 6°
none/PMDTA 79/81 0: Me Me Me 8/8 92/97
none/PMDTA 98788 b: H H H 90/94 10/6
none/PMDTA 83173 C: H H H 9/3 91I97

Chamberlin has examined the Stereoselectivities of a related process, in
which a vinyllithium species was cyclized (Scheme 9).306 Treatment of

tosylhydrazones 74a-c with t-BuLi (2.2 equiv) served to generate vinyllithium

21
species 75a-c. Upon warming these compounds to ambient temperature, high

yields and stereoselectivities were noted. As with other 5-exo-trig cyclizations,
the relative configuration of the newly formed asymmetric center could be
predicted by invoking a chair transition state, wherein the alkyl groups occupied

pseudo equatorial positions.

Scheme 9. Stereoselectlvlty of Vinyllithium Cyclizations

 

- R3 _
Ah RR2—_> tBUU, ‘78 °c 21) 5°C Mane. «88R2 M“ . R2
TsHNN‘Ji 2) H' .
R‘ ”H1
7 4 7 5 7 6 7 7
5‘ 132 133 188

a: Mo H H 70 17 33

b: H Me H 6 5 91 9

C: H H Me 70 9 91

Bailey has performed calculations for the transition state of the lithium
cyclization.31 Molecular mechanics and ab initio calculations arrived at an
energy minimum where the acyclic alkenyllithium was in a pseudo chair
conformation, not a linear array of atoms, and that this conformation
represented an intermediate. These calculations revealed that contributing to
the stabilization of this intermediate was the coordination of the lithium to the
carbon-carbon 1: bond, which was first suggested by Oliver in 1966.33 This
stabilization would be maximized when the carbon-lithium bond was syn-
coplanar to the olefin. This interaction between lithium and the olefin could
contribute to the alkene facial selectivity observed for lithium mediated
cyclizations. These calculations were qualitatively consistent with experimental
facts and help to ascertain the discrepancy in stereocontrol with its radical
counterpart (a radical would not be expected to interact with a 1: bond).

Substrates 73b and 73c illustrate that cyclization was under kinetic control.

22
Both substrates gave rise to 1,3-dimethylcycl0pentanes, yet were highly

selective for different isomers. Further proof was evident in the tandem
cyclization of 78 (eqn. 13),34 which produced the thermodynamically less
stable trans-[3.3.0] bicyclooctane (the cis isomer is 6.4 kcal/mol more stable). In
contrast, radical cyclization of the corresponding bromide produced multiple

products in which the trans isomer was present as <4% of the mixture?5

I 1)tBuI.i, -78°C Me“.
2 TMED 25°C
' 3) CH30H H5. (13)
87% YIIId

78

Finally, cyclization of 80 (eqn. 14) produced in high yields the [4.3.3]
propellane skeleton.36 More importantly, it serves to further illustrate the
adaptability of this method to complex ring patterns and that the final alkyllithium
was sufficiently stable to take part in useful, high yielding intermolecular events

with C02, aldehydes, and activated alkyl halides.

I
1) IBULI. -78 “C E
3114504 0 °c .

 

 

e:E*-CHgOI-I e:E-H;81% Yield
hrs-cg b: E.co,H;6ox Yield
1: E‘- PhCHO c: E . PhCHzoi-i; 30% Yield

d: 5* - CHch-ICI-IzBr a: E . CHZCHCHZ; 02% Yield

D. Magnesium-Mediated Cyclizations

Grignard reagents have shown little tendency toward addition to olefins
in intermolecular reactions. However, when 28 was added to magnesium in

diethyl ether or THF and the products hydrolyzed, the resulting mixture

23
consisted of a 95:5 ratio of 82 and 83 respectively (eqn. 15).37 Formation of 83

was suggested to arise from radical intermediates generated by single electron
transfer (SET) during the formation of the Grignard reagent. When the reaction
mixture was heated in excess of 100 °C. complete conversion to 83 resulted}38
Richey studied the rates of reaction and the simple diastereoselectivity of some
substituted 5-hexen-1-yl and 6-hepten-1-ylmagnesium bromides. As expected,
five-membered ring formation was faster than that of the six-membered ring (by
a factor of 2800). The diastereoselectivity was modest, favoring cyclization from
the more stable chair conformation. Utimoto incorporated a trimethylsilyl group
on the terminal carbon of the olefin to accelerate the cyclization and reduce the
necessary reaction temperatures (eqn. 16).39 Thus, treatment of 84 with 1.5
equivalents of magnesium in THF at 67 °C to provided, upon quenching, 866
as the only product. Moreover, when cyclization was complete, trapping the

final Grignard intermediate could be achieved with suitable electrophiles.

 

 

- 8-
M9 r Me
Br MgBr
—.W Cr + ——+:;:::i°°° O (.5,
23 82 (95:5) 83 6(100)
ms ' D H ms Mo
_. ”3% _.2 E’ J
1 ,67°C
H
04 85 86

E*-Hzo e: R-H;81%Yleld
E‘ -¢\.Br 8: R - Aliyi;m Yield

24
E. Aluminum-Mediated Cyclizations

Alkyl aluminum hydrides have received limited attention as a method to
convert 1,5-dienes into methylcyclopentanes. Although, when 1,5-hexadiene
(1) was treated with 2.1 equivalents of iBuzAlH at 60 °C for 16 hours (eqn. 17),
methylcyclopentane (6) accounted for >98% of the products formed, after
hydrolysis, with the balance of material being n-hexane.4° However, when 1,6-
heptadiene was treated under the same conditions, less than 1% of the product

mixture was methylcyclohexane with the major component n-heptane (>99%).

< (1)iBuzAlH. 60°C: [GE](2)H*GO, O’Me (17)

1

 

Scheme 10. Stereoselectivity of Aluminum Cyclizations

Me

Me
(2er MUM

\ mswmo-c + as 30000) 67
MO \ AIEI: 33
88
Me Me Me

EtzAl Me

 

 

_9o _ i%:0
7512M - Me‘. . Me
(ZIH’ +
WM _. Q... U...
. 92 388 (8:92) 55

t.

i) (1)Et,Al-I,60'C
Me 9‘ iiwm

L. 93 _ 388 (97.3) ' 37

 

 

25
In an attempt to answer some of the questions about the stereoregulating

forces present in the Ziegler-Natta polymerization of olefins, Stefani
investigated the diastereoselectivity for the aluminum-mediated cyclizations of
88 and 91 (Scheme 10).“0 Reaction of 88 and 91 with Et2AlH produced two
structural isomers in each case (89.90 and 92,93 respectively), due to the
indiscriminate addition of the alkylaluminum hydride to the dienes. Cyclization
and analysis of the reaction mixtures revealed important selectivity information.
For all cases, only exo insertion was observed, even for the geminally
substituted olefin, 89. The most important feature of the reaction was the ability
of the alkylaluminum to control the relative configuration of the newly formed
asymmetric center. Selectivities for the cyclization of 90, 92, and 93, ranged
from 92:8 to 97:3. These results. which paralleled those of lithium cyclizations,
reflect the ability of both lithium and aluminum to coordinate the olefin and thus
control the facial selectivity of olefin insertion and the equatorial deployment of

methyl groups.
F. Palladium-Mediated Cyclizations

Transition metals are particularly well-suited for this type of bond
construction because of their ability to insert olefins into carbon-metal bonds.
Palladium remains a convenient metal to use (despite the expensive cost)
because of the variety of ways it can be introduced into the cyclization substrate
and because catalytic processes are possible. Palladium can be bound to
carbon by either transmetallation or through oxidative addition into a carbon-
heteroatom bond. Overman has recently reported tandem cyclizations using
palladium, generating polycyclic molecules with spiro (eqn. 13)41a or fused

rings (eqn. 19).41b The catalytic cycle for these cyclizations consisted of

26
oxidative addition of Pd(0) (generated in situ) into the carbon-triflate or the

carbon-iodide bond. sequential intramolecular olefin insertions, B-hydride
elimination, and reductive elimination of H-X (X=heteroatom) from palladium
which regenerated the Pd(0) catalyst. Olefin insertion usually followed in an
exo manner to yield the smaller ring, unless some dominating steric factor
directed endo insertion. The major limitation of this methodolgy was that the
initial carbon-palladium bond must either contain an aryl or a vinyl carbon to

prevent B—hydride elimination prior to cyclization.

OTI

n \ 0.1 equiv Pd(OAc)2

0.4 equiv PPha (1 8)

 

 

: m-1;n-1 72% Yield 0 9 5
I m-1; n-2 70% VIC“!
I [II-2; I'll-1 53% VI.“
I I'D-2; "-2 50% VI.“

94

GOOD

Me
0.1 equiv Pd(OAc)2
l 0.4 equiv Part,

87% Yield (1 9)

 

 

96 ' 97;H.a:8-1.3:1
3. Background for the Mechanism of Ziegler-Matte Polymerization

Ziegler-Nana polymerization remains one of the most important industrial
processes. Due to the vivid imaginations of a number of polymer chemists,
extensive knowlege exists on how to control catalyst activity and the tacticity
(stereochemistry) of the polymer, be it isotactic42 or syndiotactic.43 However,

very little is actually known about the fundamental steps that comprise the

27
mechanism. Central to the controversy was the interaction of the olefin with the

metal, mode of insertion, and how these factors influenced the stereochemistry
of each insertion. Of the plethora of mechanisms that have come and gone, two
have received serious consideration, the first by Cossee44 and the other by the
combined efforts of Green and Rooney.45 Proof of the credibility of these
proposals can be seen by the enduring efforts to substantiate and/or disprove
these mechanistic possibilities.

Cossee suggested, in 1960, a direct insertion mechanism, the details of
which are shown in Scheme 11.44 Compound 98 (shown only for the
monomer, yet the dashed titanium-chloride bond indicates the point of
attachment of another monomer) was suggested to result from the treatment of
TiCla with an alkyl aluminum compound. Reaction of 98 with ethylene
generated complex 99 wherein the olefin interacted with the available 0 orbitals
of the titanium to form a 1: complex. Cossee reasoned that 99 was a credible
structure based on a similar finding by Chatt and Duncanson,46 in which a

crystal structure revealed that a platinum complex bound an olefin in this

manner. The 6 bond between the olefin and titanium was formed through the
interaction of the it electrons of the olefin with the dx2-y2 orbitals of the
metal.““:“7 Further stabilization could be provided by interaction of the dxy
orbital and the antibonding orbitals of the olefin. Propagation was proposed to
occur via a four centered transition state, 100, which yielded 101 after

insertion.48

28

Scheme 11. The Cossee-Arlmann Mechanism for Chain
Propagation

p- - P -

 

 

 

 

R n a...
or 0H2 l CH i I: H
cruyL/ir || —_ cr-7lr/C | 2 —» 08711-151" 2
or | CH2 or | CH2 or | "CH2
or or Cl
9 8 _ 9 9 _ _ 1 00
R\TH2
CI-7 lie-CH2
bonding C.
or
_( 101

 

 

 

Green and Rooney discounted mechanisms such as Cossee's, which
involved direct insertion.45 They did so on the grounds that isolable alkyl-
metal-olefin complexes (metal=W or Mo) could not be induced to insert the alkyl
group. On the other hand, similar hydrogen-metal-olefin complexes readily
inserted the hydrogen into the olefin. It has also been shown that certain olefin
polymerization catalysts could participate in ring-opening metathesis reactions.
All of these facts were combined to give rise to their mechanistic proposal
(Scheme 12), which differed from all others by the presence of a
metallocyclobutane intermediate (105). Thus. a-hydrogen elimination of 102
(the net result being oxidation of the metal) produced metallocarbene 103
which was proposed to complex the olefin to give rise to 104. A [2+2]
reorganization of the it electrons could form the metallocyclobutane 105 as the
key intermediate. The propagation sequence was completed by a 1,2-hydride

shift resulting in 106 (and return to the initial oxidation state of the metal).

29

Scheme 12. The Green-Rooney Mechanism for Chain
Propagation

r 7 CH2 r -

 

 

 

 

 

H II H
K H CH2
‘7: . ‘1“ CH2
[Tl] H—[Ti] H—mljl

1 0 2 _ 1 0 3 .. _ 1 041 “Hz—

H r H '

$13 ‘__
[1"] HI |
1 06 _ 1 05 ]

 

 

The lack of conclusive experimental evidence for these mechanisms has
prompted a number of different investigations. One path toward obtaining such
evidence involved synthesizing isolable metal complexes that resembled
proposed intermediates, and the second, which will be addressed shortly,
involved monitoring the stereochemical details of a single olefin insertion to
probe the microstructure of these intermediates. There have recently been a
number of reports of alkyl-metal complexes in which hydrogens on the alkyl
chain were suggestively close to the metal center, implying a bonding
interaction to the metal. Representative examples of these "agostic hydrogen
interactions” are shown in Figures 1-3, in which the only difference was whether
an a- or a B-hydrogen participated. Green suggested that the tar-hydrogen
interaction in Figure 1 lent support and resembled his propagation
mechanism.49 in contrast. B-hydrogen interaction (Figures 250a and 350D) was
proposed to represent a termination intermediate. namely that of B-hydride
elimination. A similar claim has been set forth for a zirconocene polymerization

catalyst.42a

30

“32 ’CH3
< 5 or c2
\ s C1 I \ 1
e C
/+{‘~- H‘\ / [CH3
‘93P / l H :Zr ___P..0\\\CH3
H30 I Cl cr H706“);-

CH3 H706 CH3
Figure 1. a-Agostic Interaction "007° 2- 51'A92051i0 Interaction
for 107: Ti-C1-H - 70°; Ti-H . 2.03 A. for 108: Zr-C C - 85°: Zr-H - 2.16 A.

”39 CH3
I

Figure 3. B-Agostic interaction
for 109: 1"i-c‘-c2 . 86°; T-H - 2.29 A.

Two major problems were apparent with the foundation of the
mechanism of Green and Rooney. First, the fact that the complexes of tungsten
and molybdenum did not insert an alkyl group should have not been surprising
at all. These complexes were electron rich, 18 electron complexes, which is in
direct contrast to typical, active Ziegler-Nana catalysts (primarily 14-16 electron
complexes). Secondly, a more serious problem was illustrated in Figures 2 and
3. These figures demonstrated that when 01- and B-hydrogens were present, as
in the growing polymer chain, B-agostic interactions dominate over or-
interactions. Unless 01- and B-hydrogen interactions are in equilibrium in
solution, it is unclear, based on the experimental facts and their conjecture, how
polyolefins would be produced. In fact, B-hydrogen interactions have been
suggested to play a role in polymerization,43-509851 although the cursory role

has not explicitly been determined.

31
The role of B-agostic hydrogens is complex. Brookhart has shown that a

cationic ethylcobalt(lll) complex adopts a B-agostic structure.51 Treatment of
this cobalt complex with three equivalents of ethylene in an NMR tube resulted
in sequential oligomers of ethylene. However, the first insertion was markedly
slower than subsequent insertions. Brookhart attributed this rate difference to
the unfavorable disturbance of the B-hydrogen interaction (9.2 kcal/mol), which
was contributing electronic stabilization to the electron deficient metal center.
After the initial insertion, other agostic interactions were observed by NMR
analysis. Independent synthesis and NMR analysis of a terminally bound
cationic butylcobalt(lll) complex, with a distinguishable [3-hydrogen agostic
interaction, provided direct correlation to that observed after the first insertion of
ethylene. Based on these results, Brookhart contended that alkyl migration was
the rate limiting step, and the agostic interaction played an undetermined
subordinate role. Bercaw has observed similar results for a scandium
complex.48 For example, ethylene insertion was found to be faster for those
alkylscandium complexes that did not contain a B-agostic hydrogen than those
that did. Bercaw rationalized this phenomenon in which he proposed that B-
agostic interactions [represented ground state structures, and or-agostic
interactions represented transition states for chain propagation. Jordan has
suggested that the role of B-agostic interactions may be to influence the
conformation of the 8-carbon, and thus play a role in the stereoselectivity of or-
olefin insertion!”a

Grubbs designed a clever experiment to test the Green and Rooney
mechanism.9a To simplify the analysis, a system was choosen that would allow
for only a single insertion to take place (Scheme 13). Central to the design was
the the incorporation of an tar-deuteron atom and an tar-hydrogen (110). Grubbs

reasoned that if or-activation was important, there would be an unequal

32
partitioning between the diastereomeric transition states (111a and 111b)

based on the relative strengths of the CH and C-D bonds. Thus, if ot-activation
was pivotal in the insertion process, an excess of the trans product should be
produced. Experimentally, Grubbs determined that, upon activation of 110 with
EtAlClz, a 1:1 ratio of diastereomers was produced, which provided convincing
evidence against the Green and Rooney mechanism. Grubbs proceeded to
point out that "...polymerization is the chemistry of carbon-metal single bonds
and metathesis is the chemistry of carbon-metal double bonds."9a Furthermore,

Grubbs suggested that these findings supported the Cossee mechanism.

Scheme 13. Grubbs' Test for a-Hydrogen Activation

 

 

{o /LH
\ ‘/ \
092T ~ 0927' .
H D 1) EtAlCl2, ‘3?“ -—- ‘13,“?
W “W2Cl '1 00 C ‘s‘ 5 9‘ '4'
1 1 0 8,1426, Et’A'CCl
Oi Cl
1 1 1 a 1 1 1 b
trans (favored) cis (disfavored) J
1) olefin insertion
2) H”
Me
cis/trans - 1.00 3: 0.05 QC
1 1 2

In contrast to the findings of Grubbs, Bercaw recently has shown that a
scandium complex (isoeiectronic with the Grubbs' complex) did in fact show an
o-agostic assistance.9° Treatment of 113 with a scandium-hydride complex
resulted in a regioselective production of 114 (Scheme 14). Cyclization of this

complex provided an excess of the trans product. Moreover. the deuterium

33
isotope effect measured correlated in a normal enthalpic manner with respect to

temperature. In order to insure that these results were not a consequence of
stories, based on the olefin geometry and the size of a deuterium atom versus a
hydrogen atom, the cis olefins (116) were reacted under the same conditions.
Consistent with the trans olefins, a deuterium isotope effect of the same
magnitude was observed. Bercaw also noted or-activation for the cyclization of
6-hepten-1-ylscandium complexes generated from 117. These results,
coupled with Grubbs' findings, suggest that not all olefin polymerization

catalysts proceeded through the same mechanistic pathways.

Scheme 14. Bercaw's Test for a-Hydrogen Activation

 

 

 

- H H _
0. l D o. J H
o OpScI-I(PM63) , . = ‘ -_
WV‘ 0 Hz, 4 atm OpSc 095,.
1 1 3 H
1149 1148
L trans (favored) cis (disfavored) d
Me
Me Me CHzD
Me\ M. trans/cis - 1.26 1; 0.03 (-10 °C) 0
/Si m - Op trans/cis - 1.19 :l: 0.04 (25 °C)
Me trans/cis - 1.07 :I: 0.03 (120 °C)
Me Me 1 1 5
M0 .
0 CH2”
OpScHlPMOo) 0
M ”2 4 atm trans/cis . 1.20 i: 0.04 (25 °C)
>
D 116 115
D D OPSGHIPMOa) CHZD
WVV H2, 4 atm . o
,, D ois/trans - 1.21 i 0.05 (25 C)

 

117 118

CHAPTER 1.

SYNTHESIS OF CYCLIZATION PRECURSOR
SUBSTRATES AND STANDARDS

1 . Introduction

The following chapters discuss the regio- and stereoselective features,
along with mechanistic considerations for the transformation of 120 to 119
(Scheme 15). The antithetic sequence indicates alkenyltitanocene chlorides
120 as common intermediates in each reaction. Substrates analogous to 120.
differing in tether length and location of an alkyl group (R), were prepared by
existing methodology6 via Grignard reagent 122, which could be prepared from
alkyl bromides 121. The present chapter will describe the synthetic routes for
the preparation of the cyclization precursors (121) to be used in our various

studies.

Scheme 15. Retrosynthetlc Analysls of the Tltanocene
Dichloride-Medlated Carbocyclization

opzcrn
g: = Tlefil =>ifmm =gV/B'
R R R R
120 122 121

1 1 9
n-3.4
R-akyl

34

35
2. Synthesis of Five-Membered Ring Precursor Substrates

The synthetic sequence followed for the preparation of compound 125,
needed for our mechanistic study, is shown in Scheme 16. Beginning from the
known a-methyl ester,19 reaction with LDA at -78 °C generated the ester
enolate. In order to remove acidic protons from the reaction mixture (i.e. DIPA),
1.25 eq nBuLi was added. The reaction was quenched with an excess of 020,
which provided the a-methyl-a-deuterioester (123) in modest yield (64% yield).
Subsequent treatment of 123 with LAH provided alcohol 124 (91% yield),
which was converted to bromide 125 (84% yield) by the combined action of
NBS and PPh3.52

Scheme 16. Synthetic Route to 125

o o
b
MOE-5.1K" More: 1115* W011
M0 M9 D M. D
123 124
6
W / e
A/MX‘ ,

1 2 5
(a) LDA. THF. -73 '0; neuu. -70 °C; op. (b) LAH. Ego. NaOH. Hp. (c) NBS. Plan, cruel,

Compound 129, to be used in a competitive rate experiment with 125,
was prepared in four linear steps beginning from the commercially available
diethyl propylmalonate (Scheme 17). Enolate formation of the malonate
derivative was accomplished with NaH in DMF and then subsequently coupled
with 4-bromo-1-butene to provide 126 in 96% yield.19 Deethoxycarboxylation
was encouraged by the action of LiCl and H20 in refluxing DMSO. which
resulted in the formation of ester 127 (91% yield).19 Conversion of 127 into

129 was brought about by adjustment of the oxidation state of 127 with LAH to

36
the alcohol level (128, 94% yield), followed by functional group manipulation to

give bromide 129 (82% yield).

Scheme 17. Synthetic Route to 129

E E
a H b E
Y '56—’16 / W /
Pr We /\/\l':
5,0025, 126 127
new» /\/\/\ou%> We:
W Pr

128 129
(a) NaH, DMF; 4-Bromo-1-butene. (b) ucu. H20, DMSO, 135 °c. (c) LAH, Ego; NaOi-l,

H2044 Nesmpha. cw.»

An ongoing project in our group deals with the synthesis of multiple ring
compounds. Cyclization of 1 33 (Scheme 18) would provide the
bicyclo[4.3.0]nonane, which is common to many families of natural products.
Synthesis of 133 required six linear steps beginning with radical bromination of
cyclohexene, accomplished with N88 and peroxides (130; 60% yield).53
Numerous literature methods for allylic coupling were attempted to prepare
131.54 After many sophisticated organometallic routes failed to give sufficeint
quanities of 131, recourse to a simple Grignard coupling was pursued. Thus,
when allylmagnesium bromide55 was reacted with 130, the key carbon—carbon
bond formation was secured (131; 60% yield). Next, a chemo- and
regioselective oxidation of the terminal olefin was needed. Through the use of a
bulky boranefi‘"6 this turned out to be a trivial transformation. Addition of 131 to a
THF solution of dicyclohexylborane, followed by oxidation with peroxides,57
resulted in a single structural isomer (132). To facilitate the chromatographic
separation of 132 from the reagent byproduct cyclohexanol, the 1° alcohol was

selectively protected as the TBDMS ether.58 Deprotection of the silyl other with

37
TBAF, followed by chromatography, provided 132 in pure form (52% yield).

Conversion of 132 to 133 was best accomplished by transforming the alcohol to
the mesylate followed by displacement with LiBr. This sequence provided a
modest yield of 133 (70% yield).59

Scheme 18. Synthetic Route to 133

O a O b
55% 60%
Br

130 131
6,.0
52%
Br ‘ OH
cu mm Cu
133 132

(a) NBS, CCI4. (b) Allylmagneslum bromide, THF. (c) Dicyclohexylborohydride, THF;
H202, NaOH. (d) TBDMSCl, DMAP, CHZClz; chromatographic separation. (e) TBAF;
chromatographic separation. (f) MsCl, CHZClz; LiBr. THF.

3. Synthesis of Slx-Membered Ring Precursor Substrates

The synthetic route for the parent substrate (137) is illustrated in Scheme
19. Alkylation of diethyl malonate with 5—bromo-1-pentene furnished 134 (83%
yield), which was subsequently treated with LiCl and H20 in refluxing DMSO to
provide ester 135 (86% yield). Reduction of the ester followed by bromination of
the resulting alcohol (136, 92% yield), completed the synthesis of bromide 137
(89% yield).

38
Scheme 19. Synthetic Route to 137

E
E E -——> W —?——> c
sass / E sees ME 92%
134 135
5.0025:
d
AAA/OH __> MS!
/ sees /
136 137

(a) NaH, DMF; 5—Bromo-1-pentene. (b) UCI. H20, DMSO, 180 °C. (c) LAH, E120; NaOH, H20. (d) NBS, PPha, CH2C'2.

Scheme 20 illustrates the synthesis of the allylic- (144) and the
homoallyic (151) methyl-7-bromo-1-heptenes. The key bromides 140 and 147,
needed for coupling with diethylmalonate, were prepared from esters 138 and
145, which were accessible by employing the ortho-ester Claisen methodology.
For example, crotyl alcohol was heated with triethyl orthoacetate and an acid
catalyst to provide 138 in 65% yield.60 Treatment of 138 with LAH (139, 69%
yield) followed by bromination gave 140 (75% yield). In a similar fashion, allyl
alcohol was condensed with triethyl Orthopropionate and rearranged to ester
145 (79% yield), which was subsequently reduced (146, 78% yield) and
brominated (147, 64% yield). Following the usual sequence of reactions, 140
and 147 were converted to the key cyclization substrates, 144 and 151, in 56%
and 53% overall yield.

Moreover, Scheme 21 illustrates the synthesis of 155 and 160, prepared

in the usual fashion, which were synthesized in 57% and 55% overall yield,

respectively.

39
Scheme 20. Synthetic Route to 144 and 151

R2 R2 32
R‘ OH L b CH C B
W / 0025: —> / ——> / r
+ R‘ R‘ 31

W08):

130; uses; R'aMe, R’sH 1393914; R‘uMe, R241 140; 75%; R'-Me, az-H
145; m; R'sH, R2=Me 145; me; R'sH, Rz-Me 147; 04%; R'sH, stMe

R2 E R2 R2
d o 9 OH
R‘ R‘ R1

141; 95%: R‘IMO, RzzH 1‘2; 30%; R‘ aMe, R2311 143; 85%; R‘sm, Rat-H
14s; R‘af-l, stm 149; we (2 steps); R‘sH, R’sMe 150; 93%; 8‘41, RZ-Me

R2
——D- f
R1
144; uses; Hum. stH

151; sees; R‘aH, R2=Me

(a) (HachCO-fi, 145 °c. (b) LAH, Ego; NaOH, H20. (c) was, Pang CH20l2. (d) Diethyl malonate, NaH, DMF.
(e) ucr. Hp, 01490, 100 °C.

Scheme 21. Synthetic Route to 155 and 160

1
E E R
d
EVE - EYE . ,, W“, I We
nz R2

5.0025: R‘
R'uH L» 152; ms; R‘.H, Fla-Me 153; ms; nun, 1:11.149
“new; R1-iPr -°—> 157:95%;R‘-iPr,R2-H 150;m;n‘.ipr.n2.u

R‘ at .

OH f
-°-> W —>- WB'
n2 32
154; ease; R'sl-l, 93-149 155; sex; R'cl-f, 92.1w;
159; ease; R'a-iPr, nz-H 100; e494; R'siPr, 122.14

(a) NaH, DW; 2-Bnomopropene. (b) Nal-l, DMF; s-Bromo-Htexene. (c) NaH, DMF;5-Bromo-1-pentene. (d) UCl, H20.
01490, 185 ‘C. (e) LAH, Efgo; NaOH. H20. (0 N88. PPM, CHZClz.

40
Scheme 22. Synthetic Route to 164

CH O O
O CH
C: fi” ¢ %, E ) 36%, WW
iPr
iPr iPr iP
1 6 1 1 6 2 1 6 3
125:. WBr
iPr
1 64

(a) Jones' reagent. Ergo. (b) mCPBA, 0142c:2 (c) DiBAl, toluene, -73 (*0. (d) Ph3PCH2; Hgot.
(e) NBS, PPhg, cuzcr,

Jones' oxidation61 of 4-isopropylcyclohexanol (Scheme 22) provided
ketone 161 (92% yield), which was further oxidized using mCPBA62 to give
caprolactone 162 (82% yield). The reductive-olefination sequence, 162 to
163, proved to be troublesome, however, after a number of attempts and
chromatographic purification, 163 was in hand albeit in only 22% yield. Lactone
161 was reduced with DiBAI to the lactol at -78 °C, and olefinated with the
methyl Wittig reagent, which secured the alkenol.63 Conversion of 163 to
bromide 164 was performed with NBS and PPha without further difficulties (89%
yield).

4. Synthesis of Competitive Rate Substrates

Our interest in the elucidation of the mechanism of Ziegler-Natta
polymerization has compelled us to design an experiment to test whether on-
and/or B-activation contribute to the mechanism. Substrates 168, 171, 175,
178, 182, 184, and 186 (Scheme 23) were chosen to test our hypothesis.64

Upon examination of Scheme 23, three unique, yet similar, linear sequences

41
were envisioned in which three intermediates (166, 173, and 180) were

common to all. Reaction of the enolate from the appropriate malonate (R: ethyl,
propyl, or butyl) with 5-bromo-1-pentene in DMF provided 165, 172, and 179,
which were independently carried on to the respective esters. At this juncture,
the three sequences diverged. The aforementioned esters were reduced and
brominated to provide 168, 175, and 182, respectively. The second sequence
involved reduction of esters 166 and 173 with LAD followed by bromination,
which furnished a,or-dideuterio bromides 184 and 186. Finally, conversion of
esters 166 and 173 to the B-deuterio esters 169 and 176 was accomplished,
as previously described and carried on to bromides 171 and 178 in the usual
fashion.

Schemes 24 and 25 illustrate the sequences performed to prepare the
expected products of cyclization for the appr0priate substrates. In Scheme 24,
cyclohexenone was reacted at low temperatures with the appropriate Grignard
reagent in the presence of a catalytic amount of a copper salt.55v66 The resultant
B-alkyl ketones were methylenated via Corey's procedure.67 Finally,
hydrogenation of the exocylic olefins provided a mixture of the cis- and trans-
dialkylcyclohexanesfi.8 In a similar manner, ketone 161 was methylenated and

hydrogenated to provide 197.

42

Scheme 23. Synthetic Route to Competitive Rate Substrates

E E a We
Y -» / .
R 105; ease; RsEt

172; 99%; RsPr
E43025! 170; 04%; 11.80
R

W9

175; 83%; R-Pr

R R
Br
MO”; M
o o o o

183; 97%; R28 184; 88%; Flea
185; 73%; Rst 186; 91%; R-Pr

).

R R 0

b We 1.,W\><e

—>
180; 88%; R-Et 189; 78%; RsEt
173; 84%: RsPr 176; 61%; RzPr
180; 88%; R-Bu

l‘ . l; .

d W
+— / OH

107; 95%; R-Et 170; 92%; R-Et

174:0596; R-Pr 177; 91%; R-Pr
181; 90%; RsBu ‘d
R 0

MW

1 71:91 %; RsE!
173:77%; R-Pr

(a) NaH, DMF; 5-Bromo-1-pentene. (b) ucr, Hp, ouso, 100 °c. (c) LAH, Exp; NaOH, Hp. (d) NBS.
Pm, chi, (0) LAD, Exp; NaOf-l. Hp. (r) LDA, THF, -73 -c; nBuLi, .73 .c; op.

43
Scheme 24. Synthetic Route to Cyclohexane Standards

{heeded

187; 82%; RsEt 188; 71%; R28 189; R-Et
190; 57%; RsPr 191; 71%; R=Pr 192; R-Pr
193; 78%; R: iPr 194; 89%; R: iPr 195; R: iPr

(a) CulP(nBu)3, RMgBr, Et-p; Hp’. (b) Ph3PCH2, DMSO. (c) H2, P02

Scheme 25. Synthetic Route to 196 and 197

0 Me
a b
iPr iPr iPr
1 6 2 1 9 6 1 9 7

(a) Ph3PCl-12, DMSO. (b) H2. P02, MeOH.
5. Conclusion

The preparation of the substrates needed for the diastereoselective five-
membered ring formation (Chapter 2), diastereoselective six-membered ring
formation (Chapter3), and our mechanistic studies (Chapter 4) was described.

In all cases, recourse to established literature methods was utilized.

44

6. Experimental

General Methods. 6-Bromo-3-methyl-1-hexene (363), 6-bromo-4-methyl-
1-hexene (36b), 6-bromo-5-methyl-1-hexene (36c) and 6-bromo-6-methyl-1-
hexene (36d) were kindly provided by Dr. J. R. Stille. Diethyl 2-butyl-2-(4-
pentenyl)-propanedioate (179), 2-butyl-6-heptenoate (180), 2-butyl-6-hepten-1-ol
(181), and 7-bromo-6-butyl-1-heptene (182) were kindly donated by Nancy S.
Barta. 7-Bromo-6-methylhept-1-ene was provided by Steve Gabel. The following
cyclization standards were purchased from either Aldrich Chemical Co. or Wiley
Organics: cis— and trans-1,2- and 1,3-dimethylcyclopentane, methylcyclopentane,
cis- and trans-12, 1.3-, and 1,4-dimethylcyclohexane, cyclohexane,
methylcyclohexane, cycloheptane, methylcycloheptane, and bicyclo [4.3.0]
nonane. MezAlCl and EtAlClz were purchased from Aldrich Chemical Co. from
which stock solutions were prepared. Tetrahydrofuran (T HF), diethyl ether (Et20),
toluene, and benzene, when used as solvents, were distilled from
sodium/benzophenone prior to use. Hexane was stirred over sulfuric acid for five
days, then washed sequentially with water, saturated NaHCOa, dried over CaCl2,
and the supernatant was distilled from sodium/benzophenone/tetraglyme.
Dichloromethane (CH2Cl2) was distilled from CaH2 prior to use. Anhydrous
dimethylsulfoxide and N,N-dimethylformamide were purchased from Aldrich. All
reactions were conducted under nitrogen or argon atmospheres.

NMR spectra were obtained on a Varian Gemini 300 or a VXR-300
instrument with CDCl3 as solvent. Signals are reported in units of ppm relative to
the residual chloroform peak. Multiplicities are given as s (singlet), d (doublet), t
(triplet), q (quartet), quint (quintet), sext (sextet), sept (septet), oct (octet), rn

(multiplet), or b (broad). 13C NMR resonances for carbons containing a carbon-

45
deuterium were not observed due to the quadrapolar effect caused by the

deuterium atom. Infrared spectra (IR) were obtained on a Nicolet FT -IR 42
spectrometer using NaCl plates with thin films of organic compounds and reported
in cm'1. Analytical gas chromatography (GC) was performed on a Perkin Elmer
8500 instrument with a 50 m RSLZOO column (5% methyl phenyl silicone
equivalent to SE-54 or BBS). All GO yields described were calculated by
comparison to an internal standard and corrected for detector response.

Ethyl 2-methyl-2-deuterlo-S-hexenoate (123). A solution of LDA was
prepared by adding n-BuLi (2.29 M in hexanes, 10.1 mL, 23.1 mmol) to a 0 °C
solution of DIPA (2.55 g, 25.2 mmol) in THF (30 mL). Stirring was continued for
30 minutes and the mixture was cooled to -78 °C. Ethyl 2-methyl-5-hexenoate
(3.28 g, 21 mmol) was added dropwise to the LDA solution and stirred for 50
minutes. n-BuLi (2.29 M in hexanes, 13.80 mL, 31.5 mmol) was added to the
reaction mixture at -78 °C (to deprotonate DIPA), stirred for 30 minutes, quenched
with 020 (10 mL), and warmed to ambient temperature. The aqueous layer was
extracted with diethyl ether (4 x 20 mL). The combined diethyl extracts were
concentrated and redissolved in diethyl ether (60 mL). The diethyl other solution
was washed with 1M HCI (2 x 30 mL), saturated NaHCCa (2 x 60 mL), saturated
NaCl (60 mL), dried over MgSO4, filtered, and conentrated to an oil. The crude
residue was distilled under reduced pressure (bp22 70-80 °C) and 2.10 g of 123
were obtained (64% yield). 1H NMR (300 MHz) (CDCl3) 81.11 (s, 3H), 1.22 (t, J
. 7 Hz, 3H), 1.30-1.52 (m, 1H), 1.67-1.79 (m, 1H), 2.03 (tq, J- 1, 7 Hz, 2H), 4.10
(q, J- 7 Hz, 2H), 4.93 (ddt, J: 10, 2, 1 Hz, 1H), 4.98 (dq, J: 17, 2 Hz, 1H), 5.75
(ddt, J . 17, 10, 7 Hz, 1H); 130 NMR (75 MHz) (CDCI3) 8 14.22, 16.88, 31.32,
32.74, 60.11, 114.96, 137.92, 176.00.

General Procedure For The Reduction 01 Esters- 2-DeuterIo-2-methyl-
5-hexen-1-ol (124). A 100 mL flask, equipped with a dropping funnel, was loaded

46
with LAH (0.36 g, 9.36 mmol), suspended with diethyl ether (35 mL), and cooled

to 0 °C. A solution of 123 (2.10 g, 13.38 mmol) in diethyl ether (40 mL) was
added to the LAH suspension over 30 minutes. Stirring was continued at 0 °C for
1 hour, after which the dropping funnel was replaced with a reflux condenser and
the reaction was heated to reflux for 3 hours. The reaction mixture was cooled to
0 °C and quenched by the slow sequential addition of H20 (0.36 mL), 15% NaOH
(0.36 mL), H20 (1.08 mL) and stirred for 8 hours. The white precipitate was
removed by filtration through a pad of silica gel on top of a glass frit.
Concentration of the solution by rotary evaporation followed by distillation of the
final residue (bp13 70-80 °C) provided 1.40 g of 124 (91% yield); 1H NMR (300
MHz) (CDCI3) 5 0.88 (s, 3H), 1.08-1.22 (m, 1H), 1.40-1.58 (m, 1H), 1.74 (bs, 1H),
1.88-2.18 (m, 2H), 3.38 (d, J: 10 Hz, 1H), 3.46 (d, J= 10 Hz, 1H), 4.91 (dq, J=
10,2 Hz, 1H), 4.97 (dq, J: 17, 2 Hz, 1H), 5.77 (ddt, J: 17, 1o, 7 Hz, 1H);13C
NMR (75 MHz) (00013) a 16.26, 31.09, 32.15, 67.99, 114.33, 138.87.

General Procedure For The Bromlnatlon Of Alcohols— 6-Bromo-5-
deuterlo-S-methyI-1-hexene (125). Triphenylphosphine (3.15 g, 12.0 mmol) and
alcohol 124 (1.15 g, 10.0 mmol) were combined and diluted with CHzClz (20 mL),
and cooled to 0 °C. NBS (2.14 g, 12.0 mmol) was added over 2 hours by means
of a solid addition funnel, stirred for 1 hour at 0 °C, and an additional 3 hours at
ambient temperature. The reaction mixture was concentrated to a slurry by rotary
evaporation, and the resulting solids were dissolved in a minimum of CH2CI2. To
this solution was added 50 mL petroleum ether (bp 30-65 °C) at ambient
temperature, stirred for 10 minutes, and cooled to 0 °C for an additional 10
minutes. The supernatant was filtered via cannula, with a piece of filter paper
attached to the end of the cannula, using a positive pressure of argon. The solids
were washed sequentially with petroleum ether (2 x 25 mL) at 0 °C as before.

This filtration sequence was performed twice more. The organic fractions were

47
combined and concentrated by rotary evaporation to approximately 15 mL and

then filtered through a plug of basic alumina. The remaining solvent was removed
by rotary evaporation and the residue was purified by distillation (bp4o 65-75 °C)
to provide 1.51 g of bromide 125 (84% yield); IR (Film) 3079, 2965, 2930, 2260,
1642 cm"; 1H NMR (300 MHz) (00013) 51.00 (s, 3H), 1.30 (m, 1H), 1.50 (m, 1H),

1.95-2.15 (m, 2H), 3.31 (d, J= 10 Hz, 1H), 3.37 (d, J= 10 Hz, 1H), 4.95 (ddt, J:
10, 2, 1 Hz, 1H), 5.01 (dq, J=17, 2 Hz, 1H),5.77(ddt,J=17, 10, 7 Hz, 1H);13C
NMR (75 MHz) (CDCI3) 518.50, 31.01, 33.81, 41.18, 114.79, 138.29.

General Procedure For Alkylation Of Malonate Derivatives- Diethyl 2-
(3-butenyl)-2-(propyI)-propanedioate (126). To a suspension of NaH (1.83 g,
76.39 mmol) in DMF (70 mL) was added diethyl propylmalonate (14.45 g, 71.4
mmol) over 15 minutes by means of a pressure-equalized dropping funnel. After
the addition was complete, the reaction was stirred for 2 hours. 4-Bromo-1-
butene (10.79 g, 80.0 mmol) was added over 15 minutes, stirred for 30 minutes,
and then heated to 65 °C (generally 12-24 hours). When the alkylation was
complete, the reaction mixture was cooled to 0 °C, diluted with diethyl ether (80
mL), and quenched with H20 (100 mL). The aqueous layer was extracted with
diethyl ether (3 x 80 mL). The combined extracts were washed with H20 (2 x 50
mL), saturated NaCl (60 mL), and dried over M9804. The organic solution was
filtered through silica on a glass frit, eluted with diethyl ether, and the solvent was
evaporated. Distillation under reduced pressure (bpd 70-80 °C) provided 17.58 g
of 126 as a colorless oil (96% yield); IR (film) 3081 , 2967, 2876, 1732, 1644 cm";
1H NMR (300 MHz) (CDCI3) 5 0.88 (t, J =- 7 Hz, 3H), 1.07-1.20 (m, 2H), 1.20 (t, J
a: 7 Hz, 6H), 1.78-1.98 (m, 6H), 4.14 (q, J a 7 Hz, 4H), 4.88-5.03 (m, 2H), 5.65-
5.82 (m, 1H); 13C NMR (75 MHz) (CDCl3) 514.04, 14.34, 17.34, 28.36, 31.53.

34.53, 57.27, 60.95, 114.83, 137.70, 171.68.

48
General Procedure For Deethoxycarboxylation of Malonate

Derivatives- Ethyl 2-propyI-5-hexenoate (127). In a 250 mL flask were
combined the following: 126 (17.0 g, 66.4 mmol), LiCl (5.35 9, 126.2 mmol), H20
(1.20 g, 66.4 mmol), and DMSO (130 mL). The reaction mixture was heated to
185 °C until the malonate was consumed (generally 12-15 hours). After cooling to
0 °C, the reaction was diluted with H20 (125 mL) and diethyl ether (85 mL). The
aqueous phase was extracted with diethyl ether (5 x 75 mL) and the organic
fractions were combined and washed with H20 (75 mL), saturated NaHCOa (75
mL), and then saturated NaCl (75 mL). The organics were dried over M9804,
filtered through a pad of silica gel on top a glass frit, concentrated, and distilled
under reduced pressure (bp15 75-85 °C) which provided 11.10 g of 127 as a
colorless oil (91% yield); IR (Film) 3079, 2961, 2938, 2874, 1736, 1642 cm"; 1H
NMR (300 MHz) (CDCI3) 5 0.87 (t, J: 7 Hz, 3H), 1.23 (t, J- 7 Hz, 3H), 1.25-1.76
(m, 6H), 1.95206 (m, 2H), 2.33 (tt, J= 5, 9 Hz, 1H), 4.11 (q, J: 7 Hz, 2H), 4.84-
5.02 (m, 2H), 5.75 (ddt, J . 17, 10, 7 Hz, 1H); 13C NMR (75 MHz) (CDCI3) 5
13.96, 14.31, 20.53, 31.58, 34.60, 44.83, 59.99, 114.93, 137.98, 176.30. Two 130
NMR resonances coincided in the spectrum.

2-Propyl-5-hexen-1-ol (128). Compound 127 (5.52 g, 30 mmol) was
dissolved in diethyl ether (80 mL) and added dropwise to a 0 °C suspension of
LAH (0.80 g, 21 mmol) in diethyl ether (70 mL). Following the usual workup,
distillation of the crude residue (bpz7 85-95 °C) provided 4.01 g of 128 (94%
yield); IR (Film) 3341, 3079, 2959, 2930, 2874, 1642 cm";1H NMR (300 MHz)
(CDCI3) 50.87 (t, J- 7 Hz, 3H), 1.15-1.50 (m, 7H), 1.56 (bs, 1H), 2.05 (tq, J- 1, 7
Hz, 2H), 3.50 (m, 2H), 4.91 (ddt, J=10, 2, 1 Hz, 1H), 4.98 (dq, J: 17, 2 Hz, 1H),
5.78 (ddt, J . 17, 10, 7 Hz, 1H); 130 NMR (75 MHz) (CDCI3) 514.40, 19.91,
30.09, 31.06, 33.09, 39.66, 65.34, 114.32, 139.01.

49
6-Bromo-5-propyl-1-hexene (129). NBS (4.27 g, 24 mmol) was added

slowly to a 0 °C solution of 128 (2.84 g, 20 mmol) and PPh3 (6.29 g, 24 mmol) in
CH2CI2 (30 mL) according to the general procedure. After a typical workup and
distillation, 3.35 g of bromide 129 (82% yield) were recovered (bp19 75-80 °C); IR
(Film) 3079, 2959. 2932, 2874, 2859, 1642 cm'1;1H NMR (300 MHz) (coma) 5
0.89 (t, J a 7 Hz, 3H), 1.20-1.56 (rn, 6H), 1.58-1.70 (m, 1H), 1.96-2.16 (m, 2H),
3.42 (dd, J- 5, 11 Hz, 1H), 3.46 (dd, J = 5, 11 Hz, 1H), 4.94 (ddt, J: 10, 2, 1 Hz,
1H), 5.01 (dq, J: 17, 2 Hz, 1H), 5.78 (ddt, J = 17, 10, 7 Hz, 1H); 130 NMR (75
MHz) (CDCI3)514.17, 19.65, 30.74, 31.67, 34.67, 38.44, 39.27, 114.75, 138.39.
3-Bromo-1-cyclohexene (130). A 1-L flask was loaded with NBS (39.14
g, 220 mmol), cyclohexane (16.4 g, 200 mmol), and a catalytic amount of benzoyl
peroxide in CCI4 (500 mL) and the mixture was brought to reflux for 4 hours. The
reaction mixture was filtered, solvent evaporated, and the crude residue was
distilled ([5940 80 °C) to provide 18.13 g of 130 (60% yield); 1H NMR (300 MHz)
(CDCI3) 5 1.60-2.30 (m, 6H), 4.70-4.90 (m, 1H), 5.75-5.83 (m, 1H), 5.85-5.93 (m,
1H); 13c NMR (75 MHz) (CDCI3) 518.47, 24.61, 32.65, 48.97, 128.84, 131.03.
3-(2-propenyl)-1-cyclohexene (131). Allylmagnesium bromide was
prepared from allyl bromide (24.20 g, 200 mmol) and Mg (19.45 g, 800 mmol) in
B20 (180 mL) as previously reported.55 The resultant Grignard solution was
determined to have a concentration of 1.07 M by titration with sec-butyl alcohol.
The Grignard solution (100 mL, 107 mmol) was transferred to a 250 mL 3-necked
flask equipped with a dropping funnel and a reflux condenser. A solution of 130
(7.00 g, 43.5 mmol) in THF (50 mL) was added dropwise to the Grignard solution
over 60 minutes at ambient temperature and subsequently heated to reflux for 2-3
hours. After cooling to 0 °C, the solution was quenched with aqueous H2S04 (20
mL H2804 in 50 mL H20). The aqueous layer was extracted with Et20 (4 x 75
mL). The diethyl ether extracts were combined and washed with NaCl (2 x 100

50
mL), dried over Na2804, filtered through silica gel, and concentrated. The crude

residue was distilled under reduced pressure (bpgo 70-80 °C) to provide 3.10 g of
131 (60% yield); IR (Film) 3077, 3019, 2928, 1668, 1642 cm"; 1H NMR (300
MHz) (CDCI3) 51.14-1.28 (m, 1H), 1.42-1.58 (m, 1H), 1.64-1.80 (m, 2H), 1.90-
2.00 (m, 2H), 2.00-2.07 (m, 2H), 2.05-2.20 (m, 1H), 4.95-5.05 (m, 2H), 5.52-5.70
(m, 1H), 5.62-5.60 (m, 1H), 5.77 (ddt, J = 17, 10, 7 Hz, 1H); 13C NMR (75 MHz)
(CDCI3) 521.41, 25.27, 28.79, 34.99, 40.64, 115.71, 127.20, 131.38, 137.25.

3-(3-Hydroxypropyl)-1-cyclohexene (132). Dicyclohexyl-borohyride (3.94
g, 22.1 mmol) was loaded into a 250 mL flask, equipped with a dropping funnel,
under an atmosphere of nitrogen and diluted with THF (60 mL). Diene 131 (2.70
g, 22.1 mmol) was added dropwise to the solution, contained in a flask immersed
in an ambient temperature water bath, and stirred for 5 hours during which time
the diene was completely consumed. Oxidation of the trialkylborane was
accomplished at 0 °C by the slow sequential addition of 3 M NaOH (7.4 mL) and
30% H202 (7.5 mL). The mixture was heated to 67 °C for 2 hours to facilitate
complete oxidation. The solution was cooled to ambient temperature and
saturated with K2003. The aqueous layer was extracted with diethyl ether (2 x 25
mL) and the organic fractions were combined and washed with H20 (25 mL),
dried over M9804, filtered, and concentrated to an oil. The crude residue
contained the expected mixture of cylohexanol and 132.

Separation of this mixture was found to be a difficult task; however,
selective conversion of the 1° alcohol to the TBDMS ether allowed for trivial
separation. To this end, the mixture of alcohols was dissolved in CH2CI2 (75 mL)
followed by the addition of DMAP (0.14 g, 1.10 mmol) and TEA (3.35 g, 33.15
mmol). Successive amounts of TBDMSCI were added until complete
consumption of the 1° alcohol was achieved. The mixture was washed with H20

(20 mL), NH4CI (20 mL), dried over M9804, filtered, and concentrated to an oil.

51
The TBDMS ether was separated from cyclohexanol by washing the oil down a 6"

silica gel column eluted with low-boiling petroleum ether (35-60°C). Removal of
the solvent left behind an oil that contained the TBDMS ether of 132 and only a
trace amount of the TBDMS ether of cyclohexanol; 13C NMR (75 MHz) (CDCI3) 5
-5.25, 21.52, 25.37, 25.91, 25.98, 29.07, 30.26, 32.43, 34.96, 63.53, 126.83,
132.14.

Unmasking of the 1° alcohol was achieved by the addition of the silyl ethers
to a 250 mL flask which contained TBAF (60 mL, 1M) in THF (60 mL) at 0 °C.
The reaction mixture was diluted with diethyl other (150 mL) and 1M HCI (60 mL).
The aqueous layer was extracted with diethyl ether (3 x 100 mL). The combined
organic fractions were washed with H20 (25 mL), NaCl (2 x 50 mL), dried over
M9804, filtered, and concentrated to an oil. Alcohol 132 was separated from the
cyclohexyI-TBDMS ether by filtration of the oil down a 6" column of silica gel,
eluted first with petroleum ether (which removed the TBDMS ether) and finally
with diethyl ether (which removed the desired alcohol). After evaporation of the
diethyl ether, the crude residue was distilled under reduced pressure (bp13 100-
115 °C) which provided 1.60 g of 132 (52% yield for 4 steps); IR (Film) 3339,
3017, 2934, 2859, 1670 cm"; 1H NMR (300 MHz) (CDCla) 51.10-1.88 (m, 9H),
1.90-1.96 (m, 2H), 1.98-2.10 (m, 1H), 3.61 (t, J . 7 Hz, 2H), 5.50-5.58 (m, 1H),
5.60-5.67 (m, 1H); 13C NMR (75 MHz) (CDCI3) 5 21.43, 25.31, 29.00, 30.11,
32.30, 34.90, 63.20, 127.05, 131.77.

3-(3-Bromopropyl)-1-cyclohexene (133). Conversion of 132 to the
mesylate was achieved by the dropwise addition of methane sulfonyl chloride
(1.26 g, 11 mmol) to a -10 °C solution of 132 (1.40 g, 10 mmol) and TEA (1.52 g,
11 mmol) in CH2CI2 (50 mL). The mixture was allowed to stir for 45 minutes at
-10 °C and 90 minutes at ambient temperature. The mixture was diluted with

CH2CI2 (50 mL) and washed with H20 (20 mL), 10% HCI (30 mL), saturated

52
NaHC03 (30 mL), and saturated NaCl (30 mL). The organic solution was dried

over M9804, filtered, and concentrated to an oil. The mesylate was added to a
-10 °C solution of LiBr (1.74 9, 20 mmol) in THF (30 mL). Stirring was continued
for 30 minutes and the reaction was warmed to ambient temperatrure for 18
hours. After complete consumption of the mesylate, the mixture was filtered
through basic alumina, eluted with diethyl ether, and concentrated to an oil. The .
oil was redissolved in diethyl ether (60 mL) and sequentially washed with
saturated NaHC03 (20 mL), saturated NaCl (20 mL), dried over M9804, filtered,

and concentrated to an oil. The crude residue was distilled (bpio 90-100°C) and
1.40 g of 133 were recovered (70% yield); IR (Film) 3017, 2930, 2856, 1672.
1449, 1437 cm“; 1H NMR (300 MHz) (coma) 51.14-1.26 (m, 1H), 1.29-1.60 (m,

3H), 1.64-1.82 (m, 2H), 1.82-2.00 (m, 4H), 1.98-2.14 (m, 1H), 3.38 (t, J = 7 Hz,
2H), 5.47-5.56 (m, 1H), 5.61-5.70 (m, 1H); 130 NMR (75 MHz) (CDCI3) 5 21.37,

25.26, 28.89, 30.29, 34.14, 34.48, 34.78, 127.37, 131.31.

Diethyl 2-(pent-4-enyl)-propanedloate (134). The following reagents
were combined in the usual fashion: diethylmalonate (10.75 g, 67.2 mmol), NaH
(1.69 9, 70.6 mmol) in DMF (60 mL), and 5-bromo-1-pentene (12.0 9, 80.6 mmol).
Coupling was achieved by heating the mixture at 65 °C for 24 hours. After workup
and distillation under reduced pressure (bps 112-115 °C), 12.71 9 of 134 were
recovered (83% yield): IR (Film) 2982, 2938, 1755, 1734, 1642, 1460, 1446 cm";
1H NMR (300 MHz) (CDCI3) 51.23 (t, J =- 7 Hz, 6H), 1.40 (tt, J . 8, 7 Hz, 2H),
1.88 (q, J - 8 Hz, 2H), 2.05 (q. J a 7 Hz, 2H), 3.29 (t, J a 8 Hz, 1H), 4.16 (q, J a
7 Hz, 4H), 4.90-5.15 (m, 2H), 5.75 (ddt, J = 17, 10, 7 Hz, 1H); 130 NMR (75 MHz)
(CDCI3) 5 14.05, 26.54, 28.17, 33.24, 51.91, 61.23, 114.95, 137.92, 169.42.

Ethyl-6-heptenoate (135). Compound 134 (8.75 9, 38.4 mmol) was
combined with LiCl (3.09 9, 72.9 mmol) and H20 (0.69 g, 38.4 mmol) in DMSO
(80 mL) and heated to 185 °C for 18 hours. After the usual workup, 5.16 9 of

53
ester 135 were obtained (86% yield) after distillation (bp22 72-75 °C); IR (Film)

3079, 2980, 2936, 2865, 1738, 1642, 1462, 1446, 1420 cm“; 1H NMR (300 MHz)
(CDCI3) 5 1.22 (t, J: 7 Hz, 3H), 1.39 (quint, J: 7 Hz, 2H), 1.61 (quint, J: 7 Hz,
2H), 2.03 (q, J: 7 Hz, 2H), 2.27 (t, J = 7 Hz, 2H), 4.09 (q, J: 7 Hz, 2H), 4.88-5.02
(m, 2H), 5.76 (ddt, J = 17, 10, 7 Hz, 1H); 130 NMR (75 MHz) (CDCI3) 514.21,
24.40, 28.33, 33.32, 34.16, 60.13, 114.60, 138.39, 173.63.

6-Hepten-1-ol (136). Compound 135 (4.70 9, 30.1 mmol) was dissolved in
diethyl ether (50 mL) and added dropwise to a suspension of LAH (0.80 9, 21.1
mmol) in diethyl ether (100 mL) as described in the general procedure. Following
the usual workup, distillation of the crude residue (bp27 75-85°C) provided 3.15 g
of 136 (92% yield); IR (Film) 3337, 3079, 2998, 2978, 2936, 2861, 1642, 1462,
1437 cm"; 1H NMR (300 MHz) (CDCI3) 51.25-1.43 (m, 4H), 1.53 (quint, J: 7 Hz,
2H), 1.86-1.96 (bs, 1H), 2.02 (q, J : 7 Hz, 2H), 3.58 (t, J : 6 Hz, 2H), 4.82-5.01
(m, 2H), 5.76 (ddt, J : 17, 10, 7 Hz, 1H); 130 NMR (75 MHz) (CDCI3) 5 25.17,
28.62, 32.51, 33.64, 62.75, 114.29, 138.79.

7-Bromo-1-heptene (137). The following reagents were combined the
according to the general procedure: alcohol 136 (1.50 9, 13.2 mmol), PPh3 (4.14
9, 15.8 mmol), NBS (2.81 9, 15.8 mmol), and CH2CI2.(20 mL). After distillation
(bp44 77-81 °C), 2.08 9 of bromide 137 were recovered as a colorless oil (89%
yield); IR (Film) 3077, 3000, 2977, 2961, 2938, 2859, 1642, 1460, 1439 cm“; 1H
NMR (300 MHz) (CDCI3) 5 1.32-1.52 (m, 4H), 1.78-1.90 (m, 2H), 1.98-2.10 (m,
2H), 3.38 (t, J: 7 Hz, 2H), 4.90-5.02 (m, 2H), 5.78 (ddt, J: 17, 10, 7 Hz, 1H); 130
NMR (75 MHz) (cool-3) 8 27.62, 28.01, 32.66, 33.50, 33.75, 114.57, 138.55.

Ethyl 3-Methyl-4-Pentenoate (138). In a 100 mL 3-necked flask equipped
with a thermometer and a vigreux column with distillation head, were added crotyl
alcohol (10 g, 139 mmol), triethyl orthoacetate (67.65 9, 417 mmol), and a

catalytic amount of trimethyl acetic acid. This mixture was slowly heated to 145

54
°C which was maintained until a quanitatlve amount of ethanol was collected.

After cooling the reaction to ambient temperature, H20 (5.5 g, 306 mmol) was
added and stirred for 4-6 hours. Heating was resumed to 110 °C until all of the
ethanol and ethyl acetate had been collected, at which time the mixture was
heated to distill the ester (138; bp7so 152-154 °C) which provided 11.71 9 of 138
(63% yield); IR (Film) 2980, 2936, 2874, 1738, 1642, 1462, 1420 cm"; 1H NMR
(300 MHz) (CDCI3)51.03 (d, J : 7 Hz, 3H), 1.22 (t, J : 7 Hz, 3H), 2.22 (dd, J a
15, 7 Hz, 1H), 2.32 (dd, J . 15, 7 Hz, 1H), 2.65 (m, 1H), 4.10 (q, J = 7 Hz, 2H),
4.93 (dt, J: 10, 2 Hz, 1H), 4.99 (dt, J: 17, 2 Hz, 1H), 5.74 (ddd, J: 17, 10, 7 Hz,
1H); 13C NMR (75 MHz) (CDCl3) 514.25, 19.68, 34.42, 41.34, 60.22, 113.27,
142.49, 172.55.

3-Methyl-4-penten-1-ol (139). A solution of 138 (11.0 9, 77.5 mmol, 220
mL Et20) was added to a suspension of LAH (2.06 9, 54.2 mmol) in B20 (170
mL). After the usual workup, 139 was distilled (bp7so 146-148 °C) and 5.35 9
were collected (69% yield); IR (Film) 3339, 3081, 2963, 2932, 2876, 1640, 1454,
1420 cm"; 1H NMR (300 MHz) (CDCI3) 51.00 (d, J = 7 Hz, 3H), 1.42-1.48 (bs,
1H), 1.50-1.60 (m, 2H), 2.28 (sept, J = 7 Hz, 1H), 3.63 (t, J: 7 Hz, 2H), 4.92 (dt, J
a 10, 2 Hz, 1H), 4.98 (dt, J:17, 2 Hz, 1H), 5.69 (ddd, J: 17, 10, 7 Hz, 1H);13C
NMR (75 MHz) (CDCI3) 5 20.38, 34.89, 39.27, 61.21, 113.10, 144.23.

1-Bromo-3-methyl-4-pentene (140). The following reagents were
combined according to the general procedure: 139 (5.15 g, 51.5 mmol), NBS
(10.99 g, 61.8 mmol), PPh3 (16.19 9, 61.8 mmol), and CH2CI2 (65 mL). After
distillation (bp75 72-74 °C), 6.29 g of bromide 140 were recovered as a colorless
oil (75% yield); IR (Film) 3079, 2969, 2930, 2869, 1642, 1455, 1435, 1418 cm";
1H NMR (300 MHz) (CDCI3)51.01 (d, J : 7 Hz, 3H), 1.76-1.86 (m, 2H), 2.35
(sept, J: 7 Hz, 2H), 3.28-3.45 (m, 2H), 4.94-5.07 (m, 2H), 5.61 (ddd, J: 18, 10, 7

55
Hz, 1H); 13C NMR (75 MHz) (CDCI3)819.91, 31.92, 36.54, 39.21, 114.15,

142.59.

Dlethyl 2-(3-methyI-4-pentenyI)-propanedIoate (141). Diethyl malonate
(6.48 9, 40.5 mmol) was combined with NaH (1.01 9, 42.4 mmol) in DMF (32 mL)
and subsequently reacted with 140 (5.70 g, 34.6 mmol) as described in the
general procedure. After the usual workup, the residue was distilled under
reduced pressure (bpd 80-90 °C) which provided 8.02 g of 141 (95% yield); IR
(Film) 3079, 2980, 2965, 2938, 2911, 1752, 1734, 1642, 1464 cm"; 1H NMR (300
MHz) (coma) 8 0.97 (d, J .. 7 Hz, 3H), 1.23 (t, J =- 7 Hz, 6H), 1.19-1.35 (m, 2H),
1.75-1.95 (m, 2H), 2.00-2.20 (m, 1H), 3.26 (t, J = 7 Hz, 1H), 4.10-4.22 (m, 4H),
4.86-4.99 (m, 2H), 5.63 (ddd, J: 17, 10, 7 Hz, 1H); 130 NMR (75 MHz) (CDCI3) 5
14.05, 20.09, 26.52, 33.98, 37.57, 52.05, 61.25, 113.18, 143.76, 169.47, 169.50.

Ethyl 5-methyI-6-heptenoate (142). The following were combined and
reacted according to the usual procedure: 141 (7.40 g, 30.6 mmol), LiCl (2.53 g,
59.7 mmol), H20 (0.57 g, 31.4 mmol) and DMSO (65 mL). After workup and
distillation (bp15 80-85 °C), 4.12 9 of ester 142 were recovered as a colorless oil
(80% yield); IR (Film) 3079, 2963, 2934, 2911, 2872, 1740, 1642, 1456, 1420 cm'
1; 1H NMR (300 MHz) (CDCla) 8 0.96 (d, J = 7 Hz, 3H), 122 (t, J . 7 Hz, 3H),
1.22-1.36 (m, 2H), 1.50-1.70 (m, 2H), 2.10 (sept, J : 7 Hz, 1H), 2.25 (t, J : 7 Hz,
2H), 4.09 (q, J - 7 Hz, 2H), 4.84-4.99 (m, 2H), 5.65 (ddd, J a 17, 10, 7 Hz, 1H);
13C NMR (75 MHz) (CDCI3) 514.23, 20.13, 22.70, 34.39, 35.96, 37.56, 60.18,
112.80, 144.26, 173.77.

5-MethyI-6-hepten-1-ol (143). Following the usual procedure, ester 142
(3.73 g, 21.9 mmol) was reduced with LAH (0.63 9, 16.5 mmol) in 820 (120 mL).
After workup and distillation (bpao 90-100 °C), 2.41 9 of the alcohol were collected
(86% yield); IR (Film) 3337, 3079, 2934, 2865, 1640, 1458, 1437, 1420 cm‘1;1H
NMR (300 MHz) (00013) 8 0.95 (d, J :7Hz, 3H), 1.15-1.40 (m, 4H), 1.40-1.60 (m,

56
2H), 1.5-1.8 (bs, 1H), 2.10 (sept, J = 7 Hz, 1H), 3.59 (t, J = 7 Hz, 2H), 4.82-4.98

(m, 2H), 5.65 (ddd, J = 17, 10, 7 Hz, 1H); 130 NMR (75 MHz) (CDCl3) 8 20.17,
23.38, 32.82, 36.35, 37.77, 62.95, 112.50, 144.66.

7-Bromo-3-methyl-1-heptene (144). According to the usual bromination
procedure, alcohol 143 (2.0 g, 15.6 mmol), NBS (3.34 g, 18.8 mmol), and PPh3
(4.91 9, 18.8 mmol) were combined and reacted in CH2CI2 (20 mL). After a
typical workup, 2.54 g of bromide 144 were recovered (85% yield) after distillation
(0914 68-70 °C); IR (Film) 3079, 2963, 2936, 2865, 1642, 1456, 1439, 1431, 1420
cm"; 1H NMR (300 MHz) (CDCI3) 8 0.97 (d, J = 7 Hz, 3H), 1.20-1.50 (m, 4H),
1.83 (quint, J : 7 Hz, 2H), 2.10 (sept, J : 7 Hz, 1H), 3.38 (t, J : 7 Hz, 2H), 4.90
(ddd, J: 10, 2, 1 Hz, 1H), 4.94 (ddd, J: 17, 2, 1 Hz, 1H), 5.66 (ddd, J: 17, 10, 7
Hz, 1H); 13C NMR (75 MHz) (CDCI3) 5 20.18, 25.84, 32.88, 33.89, 35.65, 37.64,
112.72, 144.39.

Ethyl 2-methyI-4-pentenoate (145). The ortho ester Claisen
rearrangement was performed as described for 138. Allyl alcohol (9.28 9, 160
mmol), triethyl orthopropionate (84.61 9, 480 mmol), and trimethyl acetic acid
(1.64 9, 16 mmol) were combined and heated to 125 °C. After the reaction was
quenched as in the usual fashion, 17.85 g of ester 145 were collected by
distillation at atmospheric pressure (bp7eo 154 °C, 79% yield); IR (Film) 3079,
2980, 2940,2911, 1736, 1644, 1462 cm"; 1H NMR (300 MHz) (CDCI3) 51.12 (d,
J : 7 Hz, 3H), 1.22 (t, J : 7 Hz, 3H), 2.08-2.22 (m, 1H), 2.32-2.54 (m, 2H), 4.10
(q, J : 7 Hz, 2H), 4.95-5.04 (m, 2H), 5.72 (ddt, J : 17, 10, 7 Hz, 1H); 13C NMR
(75 MHz) (CDCI3)514.22, 16.51, 37.78, 39.21 , 60.20, 116.73, 135.49, 176.10.

2-Methyl-4-penten-1-ol (146). Ester 145 (14.97 9, 105.4 mmol) was
reduced with LAH (2.91 9, 76.7 mmol) in B20 (530 mL) by the usual procedure.
After atmospheric distillation (bp 146 °C), 8.24 9 of 145 was obtained as a
colorless oil (78% yield); IR (Film) 3341, 3079, 2974, 2959, 2919, 2876, 1642,

57
1458, 1441, 1416 cm"; 1H NMR (300 MHz) (CDCI3) 8 0.90 (d, J:7 Hz, 3H), 1.51

(bs, 1H), 1.71 (oct, J: 7 Hz, 1H), 1.86-1.98 (m, 1H), 2.09-2.20 (m, 1H), 3.43 (dd.
J: 6, 11 Hz, 1H), 3.49 (dd, J: 6, 11 Hz, 1H), 4.85-5.05 (m, 2H), 5.78 (ddt, J: 17,
10, 7 Hz, 1H); 130 NMR (75 MHz) (CDCI3) 5 16.34, 35.59, 37.80, 67.86, 116.04.
136.95.

5-Bromo-4-methyI-1-pentene (147). According to the usual bromination
procedure, alcohol 146 (7.59 9, 75.9 mmol) was reacted with NBS (16.65 9, 93.6
mmol) and PPh3 (24.52 g, 93.6 mmol) in CH2CI2 (100 mL). Distillation of the
crude residue at reduced pressure (bpyo 76-82 °C) furnished 7.85 g of 147 (64%
yield); IR (Film) 3079, 2965,2930, 2913, 2874, 1642, 1458, 1437 cm";1H NMR
(300 MHz) (CDCI3) 51.01 (d, J: 7 Hz, 3H), 1.87 (oct, J: 7 Hz, 1H), 1.96-2.07 (m,
1H), 2.13-2.25 (m, 1H), 3.31 (dd, J : 6, 10 Hz, 1H), 3.37 (dd, J : 6, 10 Hz, 1H),
5.00-5.10 (m, 2H), 5.72 (ddt, J: 17, 10, 7 Hz, 1H); 130 NMR (75 MHz) (CDCI3) 8
18.54, 35.01, 39.06, 40.62, 116.98, 135.76.

Dlethyl 2-(2-methyl-4-pentenyl)-propanedIoate (148). The sodium
enolate of diethyl malonate, prepared by reacting diethyl malonate (7.89 g, 49.3
mmol) and NaH (1.18 g, 49.3 mmol) in DMF (40 mL), was coupled with 147 (7.24
g, 44.8 mmol) in the usual manner. After workup and distillation (bpd 80-90 °C),
9.40 g of diester 148 was isolated (87% yield). This compound was carried on
without characterization.

Ethyl 4-methyl-6-heptenoate (149). By following the general procedure,
148 (9.40 9, 38.8 mmol) was deethoxycarboxylated by the combined action of LiCl
(3.13 g, 73.8 mmol) and H20 (0.73 g, 40.7 mmol) in DMSO (80 mL) at 185 °C.
After the usual workup, the crude residue was distilled under reduced pressure
(bp"; 80-85 °C) which provided 4.95 9 as a colorless oil (66% yield for 2 steps):
IR (Film) 3079, 2978, 2961, 2932, 2874, 1740, 1642, 1462, 1449 cm“; 1H NMR
(300 MHz) (00013) 8 0.87 (d, J : 7 Hz, 3H), 1.23 (t, J : 7 Hz, 3H), 1.37-1.57 (m,

58
2H), 1.61-1.74 (m, 1H), 1.82-1.95 (m, 1H), 1.98-2.10 (m, 1H), 2.19-2.37 (m; 2H),

4.10 (q, J : 7 Hz, 2H), 4.95-5.05 (m, 2H), 5.74 (ddt, J : 17, 10, 7 Hz, 1H); 130
NMR (75 MHz) (CDCI3)514.20, 19.02, 31.38, 32.11, 32.33, 41.00, 60.20, 115.94,
136.97, 173.95.

4-Methyl-6-hepten-1-ol (150). A solution of ester 149 (4.40 g, 25.9 mmol)
was added to a 0 °C suspension of LAH (0.69 g, 18.1 mmol) in diethyl other (130
mL). The reaction mixture was purified in the usual manner which yielded 3.08 g
(93 % yield ) of 150 after distillation (bpao 90-95 °C); IR (Film) 3335, 3077, 2955,
2934, 2872, 1642, 1458, 1441, 1416 cm"; 1H NMR (300 MHz) (CDCI3) 5 0.85 (d,
J: 7 Hz, 3H), 1.04-1.22 (m, 1H), 1.28-1.64 (m, 4H), 1.73 (bs, 1H), 1.80-1.92 (m,
1H), 1.98-2.09 (m, 1H), 3.58 (t, J: 6 Hz, 2H), 4.92-5.00 (m, 2H), 5.74 (ddt, J: 17,
10, 7 Hz, 1H); 13C NMR (75 MHz) (CDCI3) 5 19.36, 30.28, 32.38, 32.59, 41.31.
63.28, 1 15.65, 137.42.

7-Bromo-4-methyl-1-heptene (151). Alcohol 150 (1.46 g, 11.7 mmol),
NBS (2.50 9, 14.1 mmol), and PPh3 (3.68 g, 14.1 mmol) were combined in
CH2CI2 (15 mL) in the usual manner. After workup and distillation (bp14 68-70
°C), 1.92 g of bromide 23 (86% yield) was obtained; IR (film) 3079, 3002, 2961,
2913, 2874, 2855, 1642, 1460, 1439 cm"; 1H NMR (300 MHz) (CDCI3) 8 0.87 (d,
J: 7 Hz, 3 H), 1.14-1.32 (m, 1H), 1.36-1.60 (m, 2H), 1.72-1.96 (m, 3H), 1.97-2.12
(m, 1H), 3.37 (t, J : 7 Hz, 2H), 4.93-5.03 (m, 2H), 5.66-5.83 (m, 1H); 13C NMR
(75 MHz) (CDCI3) 519.34, 30.52, 32.20, 34.16, 34.95, 41.20, 115.89, 137.14.

Dlethyl 2-(1-methyI-4-pentenyl)-propanedIoate (152). The following
reagents were combined as described in the general procedure: diethyl malonate
(12.97 9, 81 mmol), NaH (1.97 9, 82.5 mmol), 5-bromo-1-hexene (12.00 9, 73.7
mmol) and DMF (50 mL). After a typical workup procedure, the crude residue
was distilled (bpd 80-90 °C) and 15.86 9 of 152 was recovered (81% yield); IR
(Film) 3079, 2982, 2938, 2878, 1736, 1642, 1466, 1449 cm"; 1H NMR (300 MHz)

59
(00013) 8 0.97 (d, J : 7 Hz, 3H), 1.18-1.35 (m, 1H), 1.24 (t, J : 7 Hz, 6H), 1.42-

1.58 (m, 1H), 1.92-2.35 (m, 3H), 3.22 (d, J : 8 Hz, 1H), 4.17 (q, J : 7 Hz, 4H),
4.93 (ddt, J: 10, 2, 1 Hz, 1H), 4.99 (dq, J: 17, 2 Hz, 1H), 5.76 (ddt, J: 17, 10, 7
Hz, 1H); 13C NMR (75 MHz) (CDCI3) 5 14.11, 16.76, 31.04, 32.81, 33.50, 57.66,
61.06, 61.12, 114.73, 138.21, 168.72, 168.86.

Ethyl 3-methyI-6-heptenoate (153). The following reagents were
combined and heated to 185 °C for 15 hours: 152 (14.75 9, 62 mmol), LiCl (4.99
9, 117.8 mmol), H20 (1.12 9, 62 mmol), and DMSO (115 mL). After the usual
workup, distillation of the crude residue under reduced pressure (bp12 86-90 °C)
furnished 8.54 g of 153 (81% yield); IR (Film) 3079, 2979, 2932, 2876, 2853,
1738, 1642, 1460, 1418 cm'1;‘H NMR (300 MHz) (CDCI3) 8 0.92 (d, J : 6 Hz,
3H), 1.18-1.48 (m, 2H), 1.24 (t, J : 7 Hz, 3H), 1.88-2.14 (m, 4H), 2.27 (dd, J : 6,
14 Hz, 1H), 4.10 (q, J : 7 Hz, 2H), 4.86-5.04 (m, 2H), 5.77 (ddt, J: 17, 10, 7 Hz.
1H); 13C NMR (75 MHz) (CDCI3) 514.26, 19.51, 29.86, 31.13, 35.82, 41.79,
60.07, 114.44, 138.60, 173.11.

3-MethyI-6-hepten-1-ol (154). Ester 153 (7.75 9, 46.5 mmol) was reduced
with LAH (1.23 g, 32.5 mmol) in 320 (235 mL) following the general procedure.
The alcohol was purified by distillation (bp25 90-95 °C) and 5.74 9 (98% yield) of
154 was obtained as a colorless oil; IR (Film) 3333, 3079, 2961, 2928, 2874,
1642, 1458 cm"; 1H NMR (300 MHz) (CDCI3) 8 0.87 (d, J: 7 Hz, 3H), 1.14-1.30
(m, 1H), 1.26-1.40 (m, 2H), 1.46-1.65 (m, 3H), 1.92-2.14 (m, 2H), 3.55-3.72 (m,
2H), 4.90 (ddt, J: 10, 2, 1 Hz, 1H), 4.97 (dq, J: 17,2 Hz, 1H), 5.77 (ddt, J: 17,
10, 6 Hz, 1H); 13C NMR (75 MHz) (CDCI3)519.41, 29.00, 31.18, 36.23, 39.79,
61.02, 114.17, 139.03.

7-Bromo-5-methyI-1-heptene (155). According to the general procedure,
NBS (5.00 g, 28.1 mmol) was combined with a solution of 154 (3.00 g, 23.4 mmol)
and PPh3 (7.37 g, 28.1 mmol) in CH2CI2 (30 mL). The usual workup was

60
performed, and the crude residue was purified by distillation (bp17 78-80 °C)

which provided 5.74 g of bromide 155 (88% yield); IR (Film) 3079, 2965, 2930,
2872, 1642, 1458 cm"; 1H NMR (300 MHz) (CDCI3) 5 0.89 (d, J: 6 Hz, 3H), 1.22
(dddd, 1H), 1.41 (dddd, 1H), 1.56-1.72 (m, 2H), 1.78-1.94 (m, 1H), 1.94-2.15 (m,
2H), 3.33-3.48 (m, 2H), 4.93 (ddt, J: 10, 2, 1 Hz, 1H), 4.99 (dq, J: 17, 2 Hz, 1H),
5.78 (ddt, J : 17, 10, 7 Hz, 1H); 13C NMR (75 MHz) (CDCI3) 518.76, 31.05.
31.18, 31.91, 35.64, 39.94, 114.42, 138.74.

Dlethyl 2-(1-methylethyI)-propanedloate (156). The following reagents
were combined and reacted as described in the general procedure; diethyl
malonate (10.40 g, 65 mmol), NaH (1.71 9, 71.5 mmol), 2-bromopropane (10.30
g, 84.50 mmol) and DMF (55 mL). After workup and distillation (bp<1 50-55°C).
12.46 9 of 156 was recovered (95% yield); IR (Film) 2971, 2942, 2909, 1736,
1468, 1448 cm"; 1H NMR (300 MHz) (CDCI3) 5 0.94 (d, J: 7 Hz, 6H), 1.21 (t, J:
7 Hz, 6H), 2.25-2.42 (m, 1H), 3.05 (d, J : 9 Hz, 1H), 4.13 (q, J: 7 Hz, 4H); 13C
NMR (75 MHz) (00013) 514.03, 20.28, 28.64, 59.02, 60.99, 168.75.

Dlethyl 2-(1-methylethyl)-2-(4-pentenyl)-propanedIoate (157). The
following reagents were combined and reacted as described in the general
procedure; 156 (13.00 9, 64.40 mmol), NaH (1.69 9, 70.8 mmol), 5-bromopentene
(11.99 g, 84.50 mmol) and DMF (60 mL). After workup and distillation (bpz1 80-
85 °C), 16.22 g of 29 was recovered (93% yield); 1H NMR (300 MHz) (CDCI3) 5
0.94 (d, J : 7 Hz, 6 H), 1.23 (t, J : 7 Hz, 6H), 1.18-1.35 (m, 2H), 1.78-1.90 (m,
2H), 2.02 (tq, J : 1, 7 Hz, 2H), 2.29 (sept, J: 7 Hz, 1H), 4.16 (q, J : 7 Hz, 4 H),
4.92 (ddt, J: 10, 2, 1 Hz, 1H), 4.97 (dq, J: 17,2 Hz, 1H), 5.74 (ddt, J: 17, 10, 7
Hz, 1H); 130 NMR (75 MHz) (CDCI3) 514.12, 18.55, 23.92, 31.87, 33.19, 33.97,
60.59, 61.57, 114.78, 138.19, 171.17.

Ethyl 2-(1-methylethyI)-6-heptenoate (158). Diester 157 (15.79 g, 58.5
mmol) was treated with LiCI (4.71 9, 111.2 mmol) and H20 (1.05 g, 58.5 mmol) in

61
DMSO (120 mL) and heated to 185 °C for 15 hours. After a typical workup, the

crude residue was distilled (bp2o 105-110 °C) and 8.08 9 of ester 158 was
recovered (70% yield); IR (Film) 2963, 2940, 2874, 1734, 1642, 1462, 1447 cm’1;
1H NMR (300 MHz) (CDCI3) 5 0.87 (d, J : 7 Hz, 3H), 0.89 (d, J: 7 Hz, 3H), 1.22
(t, J : 7 Hz, 3H), 1.25-1.39 (m, 2H), 1.40-1.64 (m, 2H), 1.81 (oct, J : 7 Hz, 1H),
1.952.10 (m, 3H), 4.10 (q, J: 7 Hz, 2H), 4.86-5.00 (m, 2H), 5.74 (ddt, J: 17, 10,
7 Hz, 1H); ‘30 NMR (75 MHz) (CDCI3) 5 14.34, 20.19, 20.47, 26.98, 29.09, 30.63.
33.63, 52.63, 59.80, 114.54, 138.51, 175.80.

2-(1-MethylethyI)-6-hepten-1-ol (159). Ester 158 (7.66 g, 39.9 mmol) was
reduced with LAH (1.06 9, 27.9 mmol) in B20 (200 mL) by the usual procedure.
The alcohol was purified by distillation (bp1o 90-105 °C) and 5.62 g (93% yield) of
159 was obtained as a colorless oil; IR (Film) 3343, 3079, 2959, 2932, 2874,
1642, 1464, 1443, 1416 cm'1;1H NMR (300 MHz) (CDCI3) 8 0.85 (d, J : 7 Hz,
3H), 0.87 (d, J : 7 Hz, 3H), 1.15-1.50 (m, 6H), 1.70-1.88 (m, 1H), 1.96-2.07 (m,
2H), 3.44-3.60 (m, 2H), 4.85-5.05 (m, 2H), 5.78 (ddt, J: 17, 10, 7 Hz, 1H); 13C
NMR (75 MHz) (c0013) 519.18, 19.75, 27.13, 27.18, 27.83, 34.11, 46.46, 63.65.
114.35, 138.86.

7-Bromo-6-(1-methylethyI)-1-heptene (160). According to the general
procedure, NBS (4.119, 23.1 mmol) was combined with a solution of 159 (3.00 g,
19.2 mmol) and PPh3 (6.05 9, 23.1 mmol) in CH2CI2 (25 mL). The usual workup
was performed and the ande residue was purified by distillation (bpg 90-95 °C)
which provided 3.94 9 of bromide 160 (94% yield); IR (film) 3079, 2963, 2934,
2872, 1642, 1462, 1441, 1416 cm"; 1H NMR (300 MHz) (CDCI3) 8 0.88 (d, J : 7
Hz, 3H), 0.89 (d, J : 7 Hz, 3H), 1.25-1.50 (m, 5H), 1.79 (oct, J : 7 Hz, 1H), 1.95-
2.10 (m, 2H), 3.44 (m, 2H), 4.94 (ddt, J : 10, 2, 1 Hz, 1H), 5.00 (dq, J : 17, 2 Hz,
1H), 5.79 (ddt, J: 17, 10,7 Hz, 1H); 13C NMR (75 MHz) (CDCla) 519.17, 19.76,
26.65, 28.78, 29.19, 33.85, 37.41, 45.98, 114.57, 138.62.

62
4-(1-MethylethyI)-cylohexanone (1 61). To a solution of 4-(1-methylethyl)-

cyclohexanol (12 9, 85.7 mmol) in diethyl ether (135 mL) was added 48 mL of a
chromic acid solution. The two layers were separated, and the aqueous phase
was extracted with diethyl ether (3 x 100 mL). The organic fractions were
combined and washed with saturated NaHC03 (100 mL), saturated NaCl (100
mL), dried over M9804, filtered through a pad of silica gel, eluted with diethyl
ether, and the solvent was removed by rotary evaporation. Distillation at reduced
pressure (bpzo 80-90 °C) provided 10.51 g of 161 (89% yield); IR (Film) 2961,
2932, 2872, 1721 cm";1H NMR (300 MHz) (00013) 8 0.87 (d, J : 7 Hz, 6H),
1.30-1.60 (m, 4H), 1.90-2.02 (m, 2H), 2.19-2.40 (m, 4H); 13C NMR (75 MHz)
(c0013) 819.89, 29.56, 31.69, 40.99, 42.44, 212.42.

4-(1-MethylethyI)-caprolactone (162). To an ambient temperature
solution of ketone 33 (9.0 9, 64.3 mmol), NaHC03 (9.44 9, 112.5 mmol) in CH2CI2
(425 mL) was added mCPBA (22.83 9, 112.5 mmol) over 2 hours, and the
resultant solution was stirred for an additional 48 hours. After complete
consumption of the ketone, the solution was washed with a 1N NaHC03 (270
mL), saturated NaCI (135 mL), and the solvent was removed by rotary
evaporation. Lactone 34 was distilled under reduced pressure (bp2o 145-150 °C)
using a Kugelrohr apparatus to provide 8.22 9 of 162 (82% yield); IR (Film) 3019,
2963, 2934, 2874, 1734, 1719 cm"; 1H NMR (300 MHz) (00013) 8 0.84 (d, J: 7
Hz, 6H), 1.28-1.64 (m, 4H), 1.75-1.95 (m, 2H), 2.48-2.70 (m, 2H), 4.11 (dd, J: 9.
2 Hz, 1H), 4.21-4.32 (m, 1H); 13C NMR (75 MHz) (CDCIa) 519.18, 19.29, 25.60,
31.93, 32.47, 33.31, 46.48, 68.42, 176.14.

3-(1-methylethyl)-6-hepten-1-ol (163). Methyltriphenylphosphonium
bromide (10.31 9, 28.85 mmol) was loaded into a Schlenk flask, which was then
evacuated, purged with argon, suspended in THF (135 mL), and cooled to -20 °C.
nBuLi (11.15 mL, 2.5 M solution in hexane) was added to the \Mttig salt, stirred for

63
1 hour, and cooled to -50 °C. In a separate flask, a solution of Iactone 162 (3.0 9,

19.23 mmol) in toluene (100 mL) was cooled to -78 °C followed by the addition of
a DiBAI solution (23.1 mL, 1.0 M in toluene) over 10 minutes and stirred for 1
hour. The lactol was warmed to 0 °C and transferred, via cannula, to the
preformed ylide. The reaction temperature was maintained for 10 minutes at -50
°C, warmed to 0 °C for 30 minutes, and finally allowed to warm to ambient
temperature. After 30 minutes at ambient temperature, an aqueous solution of
Na2804-H20 (6.3 g) in H20 (50 mL) was added to the reaction. The two layers
were separated and the aqueous phase was extracted with diethyl ether (4 x 80
mL). The organic fractions were combined and washed with saturated NaCl (2 x
50 mL) dried over M9804, and the organic solution was filtered through silica gel.
The solvent was removed by rotary evaporation and the residue was purified by
flash chromatography (silica gel 60 mesh, 8:2 diethyl etherzpetroleum ether;
R(:0.51). The solvent was removed and the residue was distilled (bp7 92-100° C)
to give 650 m9 163 (22% yield); IR (Film) 3337, 3079, 2959, 2932, 2872, 1642,
1466 cm"; 1H NMR (300 MHz) (CDCI3) 5 0.82 (d, J: 7 Hz, 3H), 0.83 (d, J: 7 Hz,
3H), 1.13-1.48 (m, 5H), 1.50-1.75 (m, 2H), 1.98-2.07 (m, 2H), 3.55-3.70 (m, 2H),
4.91 (ddt, J: 10, 2, 1 Hz, 1H), 4.98 (dq, J: 17, 2 Hz, 1H), 5.78 (ddt, J: 17, 10, 7
Hz, 1H); 13C NMR (75 MHz) (CDCI3) 518.69, 19.15, 29.28, 30.03, 31.80, 33.61,
39.60, 61.81, 114.27, 139.14.

7-Bromo-5-(1-methylethyl)-1-heptene (164). Alcohol 163 (500 mg, 3.2
mmol), NBS (684 mg, 3.85 mmol), PPh3 (1000 mg, 3.85 mmol), in CH2CI2 (5 mL)
were combined according to the general procedure. Following the usual workup,
the crude residue was distilled (bp11 88-92 °C) and 620 mg were collected (89%
yield); IR (Film) 3079, 2961, 2932, 2872, 1642, 1466, 1439, 1416 cm";1H NMR
(300 MHz) (CDCI3) 5 0.83 (d, J: 7 Hz, 6H), 1.14-1.44 (m, 4H), 1.54-1.92 (m, 4 H),
1.94-2.10 (m, 2 H), 3.39 (m, 2H), 4.94 (ddt, J: 10, 2, 1 Hz, 1H), 4.99 (dq, J: 17,

64
2 Hz, 1H), 5.78 (ddt, J: 17, 10, 7 Hz, 1H); 13C NMR (75 MHZ) (CDCI3) 5 18.77,

18.97, 28.80, 29.36, 31.62, 32.76, 34.07, 41 .96, 114.50, 138.81.

Dlethyl 2-(ethyI)-2-(4-pentenyl)-propanedIoate (165). The following
reagents were combined as outlined in the general procedure: diethyl
ethylmalonate (37.64 g, 200 mmol), NaH (5.03 g, 210 mmol) in DMF (200 mL),
and 5-bromo-1-pentene (32.16 g, 216 mmol). Coupling was achieved by heating
the mixture at 50 °C for 40 hours. After workup and distillation under reduced
pressure (bpd 90-100°C), 50.08 g of 165 were recovered (98% yield); IR (Film)
2980, 2942, 2884, 1732, 1642 cm"; 1H NMR (300 M2) (CDCI3) 8 0.78 (t, J : 7
Hz, 3H), 1.21 (t, J: 7 Hz, 6H), 1.14-1.28 (m, 2H), 1.76-1.95 (m, 4H), 2.03 (tq, J:
2, 7 Hz, 2H), 4.15 (q, J: 7 Hz, 4H), 4.88-5.01 (m, 2H), 5.75 (ddt, J: 17, 10, 7 Hz,
1H); 13C NMR (75 MHz) (CDCI3) 5 8.39, 14.09, 23.24, 25.17, 31 .06, 33.80, 57.82,
60.94, 114.89, 138.13, 171.82.

Ethyl 2-ethyl-6-heptenoate (166). Compound 165 (50.0 9, 195.3 mmol)
was combined with LiCl (15.73 9, 371.1 mmol) and H20 (3.51 9, 195.3 mmol) in
DMSO (400 mL) and heated to 180 °C for 18 hours. After the usual workup,
31.61 9 of ester 166 were obtained (88% yield) after distillation (bp1o 105-115 °C);
IR (Film) 3079, 2967, 2934, 1734, 1642 cm"; 1H NMR (300 MHz) (CDCI3) 8 0.86
(t, J: 7 Hz, 3H), 1.23 (t, J: 7 Hz, 3H), 1.27-1.68 (m, 6H), 2.02 (bq, J: 7 Hz, 2H),
2.23 (tt, J: 5, 9 Hz, 1H), 4.11 (q, J : 7 Hz, 2H), 4.87-5.02 (m, 2H), 5.75 (ddt, J :
17, 10, 7 Hz, 1H); 13C NMR (75 MHz) (CDCI3) 511.80, 14.34, 25.48, 26.65.
31.49, 33.61, 47.18, 59.97, 114.58, 138.50, 176.29.

2-EthyI-6-hepten-1-ol (167). Compound 166 (10.12 g, 55 mmol) was
dissolved in diethyl other (150 mL) and added dropwise to a 0 °C suspension of
LAH (1.46 g, 38.5 mmol) in diethyl other (125 mL). Following the usual workup,
distillation of the crude residue (bp22 90-100 °C) provided 7.39 g of 167 (95%
yield); IR (film) 3337, 3079, 2963, 2876, 1642 cm"; 1H NMR (300 MHz) (CDCI3)

65
8 0.87 (t, J : 7 Hz, 3H), 1.15-1 .45 (m, 8H), 2.02 (bq, J : 7 Hz, 2H), 3.51 (m, 2H),

4.88-5.02 (m, 2H), 5.78 (ddt, J : 17, 10, 7 Hz, 1H); 13C NMR (75 MHz) (CDCI3) 5
11.06, 23.28, 26.16, 29.88, 34.10, 41.86, 65.17, 114.36, 138.90.

7-Bromo-6-ethyl-1-heptene (168). NBS (4.27 g, 24 mmol) was added
slowly to a 0 °C solution of 167 (2.84 9, 20 mmol) and PPh3 (6.29 9, 24 mmol) in
CH2CI2 (25 mL). After a typical workup, 3.66 g of bromide 168 (89% yield) were
recovered after distillation (bp24 90-100 °C); IR (Film) 3079, 2965, 2934, 2876,
1642 cm";1H NMR (300 MHz) (00013) 8 0.87 (t, J : 7 Hz, 3H), 1.28-1.46 (m,
6H), 1.48-1.60 (m, 1H), 1.98-2.10 (m, 2H), 3.42 (dd, J: 5, 10-11 Hz, 1H), 3.44
(dd, J: 5, 10-11 Hz, 1H), 4.94 (ddt, J: 10, 2, 1 Hz, 1H), 4.99 (dq, J: 17, 2 Hz,
1H), 5.79 (ddt, J: 17, 10, 7 Hz, 1H);13C NMR (75 MHz) (CDCI3)510.85, 25.12,
25.88, 31.65, 33.81, 38.93, 40.95, 114.55, 138.64.

Ethyl 2-deuterlo-2-ethyI-6-heptenoate (169). According to the procedure
described for 123, ester 166 (6.44 g, 35 mmol) was added dropwise to a -78 °C
THF (70 mL) solution of LDA (38.5 mmol) and stirred for 50 minutes. The reaction
mixture was diluted with THF (20 mL) and n-BuLi (18.35 mL, 45.5 mmol) was
added, stirred for 30 minutes, and quenched with D20 (10 mL). After the usual
workup, the crude residue was distilled under reduced pressure (bp21 90-100°C)
and 4.91 g of 169 were recovered (78% yield); IR (Film) 3079, 2977, 2936, 1734,
1642 cm“; 1H NMR (300 MHz) (CDCI3) 8 0.86 (t, J: 7 Hz, 3H), 1.23 (t, J: 7 Hz,
3H), 1.26-1.68 (m, 6H), 2.01 (q, J : 7 Hz, 2H), 4.11 (q, J : 7 Hz, 2H), 4.92 (ddt, J
a 10, 2, 1 Hz,1H), 4.96 (dq, J: 17, 2 Hz, 1H), 5.75 (ddt, J: 17, 10, 7 Hz,1H);
13C NMR (75 MHz) (CDCI3)511.76, 14.33, 25.39, 26.62, 31.40, 33.60, 59.94.
114.57, 138.49, 176.29.

2-Deuterlo-2-ethyI-6-hepten-1-ol (170). Ester 169 (4.47 g, 24.2 mmol)
was dissolved in diethyl ether (75 mL) and added slowly to a 0 °C suspension of

LAH (0.65 g, 17.0 mmol) in diethyl ether (50 mL) according to the general

66
procedure. Following the usual workup, distillation of the crude residue (bpzz 90-

100 °C) provided 3.17 g of 170 (92% yield); IR (Film) 3343, 3079, 2963, 2861,
2130, 1642 cm"; 1H NMR (300 MHz) (00013) 8 0.87 (t, J : 7 Hz, 3H), 1.18-1.42
(m, 7H), 2.03 (bq, J : 7 Hz, 2H), 3.52 (m, 2H), 4.94-5.02 (m, 2H), 5.79 (ddt, J :
17, 10, 7 Hz, 1H); 130 NMR (75 MHz) (CDCI3) 511.05, 23.17, 26.14, 29.77.
34.12, 65.15, 114.39, 138.92.

A 7-Bromo-6-deuterlo-6-ethyl-1-heptene (171). NBS (3.43 g, 19.3 mmol)
was added slowly to a 0 °C solution of 170 (2.28 g, 16 mmol) and PPh3 (5.05 g,
19.3 mmol) in CH2CI2 (25 mL) according to the general procedure. After a typical
workup, 2.99 g of bromide 171 (91% yield) were recovered after distillation (bp19
80-90 °C); IR (Film) 3079, 2965, 2934, 2876, 2114, 1642 cm"; 1H NMR (300
MHz) (CDCI3) 5 0.87 (t, J : 7 Hz, 3H), 1.30-1.48 (m, 6H), 1.95-2.10 (m, 2H), 3.40
(d, J: 11 Hz, 1H), 3.44 (d, J: 11 Hz, 1H), 4.94 (ddt, J :10, 2, 1 Hz, 1H), 4.99
(dq, J : 17, 2 Hz, 1H), 5.79 (ddt, J : 17, 10, 7 Hz, 1H); 13C NMR (75 MHz)
(CDCI3) 510.80, 25.00, 25.85, 31.53, 33.82, 38.82, 114.55, 138.65.

Dlethyl 2-(4-pentenyl)-2-(propyl)-propanedloate (172). The following
reagents were combined as described in the general procedure: diethyl
propylmalonate (40.45 g, 200 mmol), NaH (5.03 g, 210 mmol) in DMF (200 mL),
and 5-bromo-1-pentene (32.16 g, 216 mmol). Coupling was achieved by heating
the mixture at 50 °C for 40 hours. After workup and distillation under reduced
pressure (bp<1 95-110 °C), 53.6 g of 172 were recovered (99 % yield); IR (Film)
3079, 2965, 2876, 1734, 1642 cm“; 1H NMR (300 MHz) (CDCI3) 8 0.89 (t, J : 7
Hz, 3H), 1.07-1.27 (m, 4H), 1.21 (t, J: 7 Hz, 6H), 1.77-1.89 (m, 4H), 2.02 (bq, J:
7 Hz, 2H), 4.14 (q, J : 7 Hz, 4H), 4.89-5.02 (m, 2H), 5.74 (ddt, J : 17, 10, 7 Hz,
1H); 130 NMR (75 MHz) (c0013) 8 14.09, 14.40, 17.36, 23.33, 31.71, 33.81,
34.46, 57.46, 60.94, 1 14.88, 138.14, 171.89.

67
Ethyl 2-propyI-6-heptenoate (173). Compound 172 (53.0 9, 196.3 mmol)

was combined with LiCl (15.81 g, 373 mmol) and H20 (3.53 9, 196.3 mmol) in
DMSO (400 mL) and heated to 180 °C for 18 hours. After the usual workup,
32.49 g of ester 173 were obtained (84% yield) after distillation (bpd 60-70 °C);
IR (Film) 3079, 2959, 2938, 1734, 1642 cm"; 1H NMR (300 MHz) (CDCI3) 8 0.87
(t, J : 7 Hz, 3H), 1.19-1.66 (m, 8H), 1.23 (t, J: 7 Hz, 3H), 2.02 (bq, J: 7 Hz, 2H),
2.31 (tt, J: 5, 9 Hz, 1H), 4.11 (q, J: 7 Hz, 2H), 4.88-5.02 (m, 2H), 5.76 (ddt, J:
17, 10, 7 Hz, 1H); 13C NMR (75 MHz) (CDCI3)514.00, 14.33, 20.60, 26.66,
31.90, 33.60, 34.67, 45.36, 59.98, 114.58, 138.51, 176.47.

2-PropyI-6-hepten-1-ol (174). Compound 173 (10.89 g, 55 mmol) was
dissolved in diethyl ether (150 mL) and added dropwise to a 0 °C suspension of
LAH (1.46 g, 38.5 mmol) in diethyl other (125 mL) as described in the general
procedure. Following the usual workup, distillation of the crude residue (bp2o 90-
100 °C) provided 8.16 g of 174 (95% yield); IR (Film) 3337, 3079, 2930, 2870,
1642 cm“; 1H NMR (300 MHz) (CDCI3) 5 0.87 (t, J : 7 Hz, 3H), 1.12-1.52 (m,
9H), 1.53 (bs, 1H), 1.94-2.06 (m, 2H), 3.44-3.54 (m, 2H), 4.86502 (m, 2H), 5.78
(ddt, J : 17, 10, 7 Hz, 1H);13C NMR (75 MHz) (00013) 514.40, 19.96, 26.14,
30.36, 33.16, 34.10, 40.14, 65.53, 114.34, 138.89.

7-Bromo-6-propyl-1-heptene (175). N88 (4.27 g, 24 mmol) was added
slowly to a 0 °C solution of 174 (3.12 g, 20 mmol) and PPh3 (6.29 g, 24 mmol) in
CH2CI2 (25 mL) according to the general procedure. After the typical workup,
3.63 9 of bromide 175 (83% yield) were recovered after distillation (bp21 100-110
°C); IR (film) 3079, 2959, 2860, 1642 cm"; 1H NMR (300 MHz) (CDCI3) 5 0.89 (t,
J: 7 Hz, 3H), 1.18-1.50 (m, 8H), 1.50-1.66 (m, 1H), 1.95-2.10 (m, 2H), 3.43 (d, J
: 5 Hz, 2H), 4.93 (ddt, J: 10, 2, 1 Hz, 1H), 4.99 (dq, J: 17, 2 Hz, 1H), 5.79 (ddt,
J: 17, 10, 7 Hz, 1H); 13c NMR (75 MHz) (00013) 514.17, 19.69, 25.83, 32.00,
33.82, 34.79, 39.13, 39.45, 114.54, 138.65.

68
Ethyl 2-deuterlo-2-propyl-6-heptenoate (176). According to the

procedure described for 123, ester 173 (6.88 g, 35 mmol) was added dropwise to
a -78 °C THF (70 mL) solution of LDA (38.5 mmol) and stirred for 50 minutes.
The reaction mixture was diluted with THF (20 mL), n-BuLi (18.35 mL, 45.5 mmol)
was added, stirred for 30 minutes, and quenched with D20 (10 mL). After a
typical workup, the crude residue was distilled under reduced pressure (bp21 95-
105 °C) which provided 4.19 g of 176 (61% yield); IR (Film) 3079, 2961, 2934,
2874, 2170, 1736, 1642 cm"; 1H NMR (300 MHz) (00013) 8 0.87 (t, J : 7 Hz,
3H), 1.18-1.48 (m, 8H), 1.23 (t, J: 7 Hz, 3H), 1.50-1.64 (m, 2H), 2.02 (bq, J : 7
Hz, 2H), 4.11 (q, J : 7 Hz, 2H), 4.87502 (m, 2H), 5.75 (ddt, J : 17, 10, 7 Hz,
1H); 13C NMR (75 MHz) (CDCI3) 513.98, 14.33, 20.56, 26.63, 31.80, 33.59.
34.57, 59.95, 114.57, 138.49, 176.46.

2-DeuterIo-2-propyI-6-hepten-1-ol (177). Compound 176 (3.87 9, 19.6
mmol) was dissolved in diethyl ether (60 mL) and added dropwise to a 0 °C
suspension of LAH (0.52 g, 13.72 mmol) in diethyl ether (40 mL) as described in
the general procedure. Following the usual workup, distillation of the crude
residue (bp20 90-100 °C) provided 2.78 g of 177 (91% yield); IR (Film) 3335,
3079, 2959. 2128, 1642 cm'1;1H NMR (300 MHz) (00013) 8 0.87 (t, J : 7 Hz,
3H), 1.15-1.44 (m, 9H), 2.02 (tq, J: 2, 7 Hz, 2H), 3.50 (s, 2H), 4.91 (ddt, J: 10,
2, 1 Hz, 1H), 4.97 (dq, J : 17, 2 Hz, 1H), 5.78 (ddt, 17, 1o, 7 Hz, 1H); 130 NMR
(75 MHz) (CDCI3)514.43, 19.94, 26.11, 30.24, 33.05, 34.12, 65.51, 114.36.
138.91.

7-Bromo-6-deuterlo-6-propyI-1-heptene (178). NBS (3.43 g, 19.3 mmol)
was slowly added to a 0 °C solution of 177 (2.28 g, 14.5 mmol) and PPha (5.06 g,
19.3 mmol) in CH2CI2 (25 mL) according to the usual procedure. After a typical
workup, 3.63 g of bromide 178 (77% yield) were recovered after distillation (bp21
95-105 °C); IR (Film) 3079, 2961, 2932, 2128, 1642 cm"; 1H NMR (300 MHz)

69
(CDCI3) 8 0.89 (t, J : 7 Hz, 3H). 1.20-1.45 (m, 8H), 1.98-2.08 (m, 2H), 3.42 (s,

2H), 4.94 (ddt, J: 10, 2, 1 Hz, 1H), 4.99 (dq, J: 17, 2 Hz, 1H), 5.79 (ddt, J: 17,
10, 7 Hz, 1H); 130 NMR (75 MHz) (CDCI3) 514.18, 19.65, 25.79, 31.88, 33.82,
34.68, 39.33, 114.54, 138.65.

1,1-Dldeuterlo-2-ethyl-6-hepten-1-oI (183). Compound 166 (2.76 g, 15.0
mmol) was dissolved in diethyl ether (25 mL) and added dropwise to a 0 °C
suspension of LAD (0.44 g, 10.5 mmol) in diethyl ether (50 mL) as described in
the general procedure. Following the usual workup, distillation of the crude
residue (bp2o 95-105 °C) provided 2.09 g of 183 (97% yield); IR (Film) 3345,
3079, 2965,2934, 2863, 2197, 2093, 1642 cm";1H NMR (300 MHz) (00013) 8
0.87 (t, J : 7 Hz, 3H), 1.20-1.50 (m, 8H), 2.03 (tq, J: 1, 7 Hz, 2H), 4.92 (ddt, J:
10, 2, 1 Hz, 1H), 4.98 (dq, J: 17, 2 Hz, 1H), 5.79 (ddt, J: 17, 10, 7 Hz, 1H);13C
NMR (75 MHz) (c0013) 8 11.07, 23.24, 26.16, 29.84, 34.11, 41.66, 114.37,
138.91.

7-Bromo-7,7-dldeuterio-S-ethyI-1-heptene (184). N88 (2.22 g, 12.5
mmol) was slowly added to a 0 °C solution of 183 (1 .50 g, 10.4 mmol) and PPh3
(3.28 g, 12.5 mmol) in CH2CI2 (15 mL) according to the usual procedure. After a
typical workup, 1.86 g of bromide 184 (88% yield) were recovered after distillation
(bp19 80-90 00); IR (Film) cm“; 1H NMR (300 MHz) (CDCI3) 8 0.87 (t, J : 7 Hz,
3H), 1.28-1.46 (m, 6H), 1.46-1.58 (m, 1H), 1.98-2.10 (m, 2H), 4.94 (ddt, J: 10, 2,
1 Hz, 1H), 4.99(dq, J: 17,2 Hz, 1H), 5.79 (ddt, J: 17, 10, 7 Hz, 1H);13C NMR
(75 MHz) (CDCIa) 510.86, 25.07, 25.88, 31.60, 33.82, 40.71 , 114.56, 138.65.

1,1-DIdeuterIo-2-propyl-6-hepten-1-oI (185). Compound 173 (2.97 g,
15.0 mmol) was dissolved in diethyl ether (25 mL) and added dropwise to a 0 °C
suspension of LAD (0.44 g, 10.5 mmol) in diethyl ether (50 mL) as described in
the general procedure. Following the usual workup, distillation of the crude
residue (bp22 105-115 °C) provided 1.72 g of 185 (73% yield); IR (Film) 3339,

70
3079. 2959, 2930. 2872. 2197, 2093. 1642, 1460, 1416 cm"; 1H NMR (300 MHz)

(CDCI3) 8 0.87 (t, J: 7 Hz, 3H). 1.17-1.49 (m, 10H), 2.02 (bq, J: 7 Hz. 2H), 4.87-
5.02 (m. 2H), 5.78 (ddt, J : 17, 10. 7 Hz, 1H); 13C NMR (75 MHz) (CDCI3) 5
14.39, 19.94, 26.10. 30.27, 33.08, 34.08, 39.92, 114.32. 138.88.
7-Bromo-6-deuterio-6-propyI-1-heptene (186). NBS (1.92 9, 10.8 mmol)
was slowly added to a 0 °c solution of 185 (1.42 g. 9.0 mmol) and PPh3'(2.83 g,
10.8 mmol) in CH2CI2 (15 mL) according to the usual procedure. After a typical
workup, 1.82 g of bromide 186 (91% yield) were recovered after distillation (bp22
100-110 °C); IR (Film) 3079, 2959, 2930. 2872, 2861, 1642. 1458, 1416 cm"; 1H
NMR (00013) 8 0.89 (t, J : 7 Hz, 3H), 1 .18-1.50 (m, 8H), 1.52-1.65 (m, 1H). 1.97-
2.01 (m, 2H), 4.90-5.04 (m. 2H), 5.79 (ddt, J : 17, 10, 7 Hz, 1H); 13c NMR
(CDCI3) 5 14.19, 19.70, 25.83, 31.95. 33.83, 34.74. 38.90. 11455, 138.66.
3-Ethylcyclohexanone (187). In a 100 mL Schlenk flask, equipped with a
dropping funnel, were loaded magnesium turnings (7.78 g, 320 mmol). THF (60
mL), and a small crystal of iodine. and the flask was heated to 45 °C. Ethyl
bromide (8.71 g, 80 mmol) was added dropwise over 2 hours and stirred an
additional 4 hours. A separate 250 mL flask was loaded with n-BuaPCul (1.99 g,
5.08 mmol). diluted with THF (10 mL). and cooled to -45 °C. The Grignard
solution was transferred via cannula to the copper salt over 5 minutes and stirred
for an additional 20 minutes. Cyclohexenone (3.70 g, x mmol) was added
dropwise over 15 minutes, stirred for 90 minutes, and the solution was allowed to
warm to -20 °C until all cyclohexenone was consumed. The reaction mixture was
quenched with saturated NH4CI (8 mL) and diluted with petroleum other (500 mL)
The organic layer was washed with NH40H (3 x 100 mL; 6 mL NH40H in 294 mL
H20), NaCI (40 mL), dried over M9804, filtered through silica gel, concentrated to
an oil, and distilled (bpzo 80-100 °C) which provided 3.96 g of 37 (82% yield); IR
(Fllm) 2963. 2934. 2876. 1715 cm"; 1H NMR (300 MHz) (CDCI3) 5 0.85 (t. J : 7

71
Hz, 3H), 1.18-1.42 (m, 3H), 1.48-1.70 (m. 2H). 1.78-2.05 (m. 3H), 2.15-2.42 (m,

3H); 130 NMR (75 MHz) (00013) 811.10, 25.22. 29.23, 30.82, 40.70, 41 .43,
47.78. 212.07.

3-Ethylmethylenecyclohexane (188). NaH (0.39 g. 16.5 mmol) was
loaded into a 250 mL flask under an atmosphere of nitrogen and suspended in
DMSO (45 mL). The suspension was heated to 75 °C for 45 minutes or until no
further H2 was evolved. A separate flask was loaded
methyltriphenylphosphonium bromide (5.90 g, 16.5 mmol). diluted with DMSO (30
mL), and heated to 50 °C until the solution was homogeneous. The Wittig salt
was added to the dimsyl anion at ambient temperature and stirred for 30 minutes,
during which time the solution became orange-red. Ketone 187 was added at
room temperature and the heated to 55 °C. After the ketone was consumed, the
reaction mixture was quenched with H20 (25 mL) and extracted with petroleum
ether (4 x 60 mL). The organic layers were combined and washed with (1:1)
DM80:H20 (65 mL). NaCI (130 mL), dried over M9804, filtered through neutral
alumina, and concentrated. The crude residue was distilled (bp51 60-65 °C)
which provided 1.14 g of 188 was recovered (71% yield); IR (Film) 3073, 2963.
2932. 2876, 2859, 1651 cm"; 1H NMR (300 MHz) (CDCI3) 5 0.86 (t. J: 7 Hz.
3H). 0.90-1.10 (m. 1H). 1.18-1.38 (m, 4H). 1.58-1.82 (m. 3H), 1.84-2.00 (m, 1H).
2.18-2.36 (m. 2H), 4.58 (bs, 2H); 13C NMR (75 MHz) (CDCI3)511.42, 27.13,
29.45, 32.21, 35.21, 40.95, 41.41, 106.66, 149.73.

1-EthyI-3-methylcyclohexane (189). In 50 mL flask were loaded Pt02 (25
mg). 188 (0.75 g. 6 mmol). and methanol (15 mL). The flask was filled and
purged with H2 three times and stirred under an atmosphere of H2. The
consumption of the olefin was monitored by NMR. GC analysis determined the
ratio of isomers to be 1.96:1 (trans:cis); 13C NMR (75 MHz) (CDCI3) 511.43,

72
11.97. 20.76, 20.87, 23.00, 26.37, 27.23. 27.28. 31.20. 31.30. 32.58, 32.73,

33.89, 34.23. 35.44, 38.93. 39.50. 41.99.

3-Propylcyclohexanone (190). The following reagents were combined
according to the procedure described for 187: propyl bromide (9.83 g. 80 mmol),
Mg (7.78 g, 320 mmol) and THF (60 mL); n-BU3PCuI (1.99 g, 5.08 mmol) in THF
(1 0 mL). and cyclohexenone (3.75 g. 39.1 mmol). The crude residue was distilled
(bP<1 50-60 °C) and 3.13 g 190 was obtained (57% yield); IR (Film) 2959, 2872,
1713 cm";1H NMR (300 MHz) (CDCla) 8 0.83 (1. J : 7 Hz, 3H). 1.16-1.38 (m,
5H), 1.50-1.90 (m, 3H), 1.92-2.06 (m. 2H), 2.14-2.42 (m, 3H); 13c NMR (75 MHz)
0003) 5 14.04. 19.67, 25.25, 31.23. 38.75, 38.77. 41 .45, 48.12. 212.09.

3-Propylmethylenecyclohexane (191). The following reagents were
combined as described for 188; NaH (0.39 g. 16.5 mmol) in DMSO (45 mL);
methyltriphenylphosphonium bromide (5.90 g, 16.5 mmol) in DMSO (30 mL). and
190 (1.93 g. 13.8 mmol). After workup and distillation (bps; 70-80 °C). 1.32 g 191
was recovered (71% yield); IR (Film) 3071. 2980. 1651 om";1H NMR (300 MHz)
(00%) 8 0.86 (1. J : 7 Hz. 3H), 0.95-1.12 (m. 1H), 1.10-1.42 (m, 6H), 1.50-1.82
(m. 3H), 1.85-2.02 (m. 1H). 2.18-2.35 (m. 2H). 4.60 (bs, 2H);13C NMR (75 MHz)
(CDCI3)514.33, 19.93. 27.16. 32.59. 35.21. 38.89, 39.12. 41 .79, 106.65. 149.76.

1-MethyI-3-propylcylohexane (192). In a 25 mL flask were loaded Pt02
(25 mg), 191 (0.82 g, 6 mmol). and diethyl ether (15 mL). The hydrogenation was
carried out as in 189. GO analysis determined the ratio of isomers to be 1.71:1
(trans:cis); 13C NMR (75 MHz) (CDCI3)514.33, 15.22, 19.89, 20.46, 20.75, 20.88.
22.95, 26.38. 27.22, 31.64, 32.05. 32.72, 32.98, 33.89, 35.42, 36.92, 37.40.
39.25. 39.97, 42.40.

3-(1-methylethyl)-cyclohexanone (193). The following reagents were
combined according to the procedure described for 187: 2-bromopropane (7.37

g. 60 mmol), Mg (5.83 g. 240 mmol) in THF (60 mL); n-BuaPCuI (1.99 g, 5.08

73
mmol) in THF (10 mL); and cyclohexenone (3.53 g, 36.8 mmol). The crude

residue was distilled (bp1o 80 °C) and 4.00 g 193 was obtained (78% yield); IR
(Film) 2961, 2872, 1715, 1466. 1449. 1426 cm";1H NMR (300 MHz) (CDCI3) 8
0.85 (d. J : 6 Hz, 3H), 0.86 (d, J : 6 Hz, 3H). 1.25-1.40 (m. 1H), 1.45-1.62 (m,
3H), 1.78-1.85 (m. 1H). 1.95210 (m, 2H), 2.15238 (m. 3H); 130 NMR (75 MHz)
(CDCI3) 5 19.25, 19.48, 25.47. 28.27. 32.42, 41 .43. 45.26. 45.35. 212.49.

3-(1-methylethyI)-methylenecyclohexane (194). The following reagents
were combined as described for 188; NaH (0.72 g, 30 mmol) in DMSO (75 mL);
methyltriphenylphosphonium iodide (12.12 g, 30 mmol) in DMSO (50 mL); and
193 (3.5 g, 25 mmol). After workup and distillation (bpss 60-70 °C), 3.08 g 194
were recovered (89% yield); IR (Film) 3073. 2959, 2932. 2874, 2839, 1651. 1464,
1447. 1433 cm“; 1H NMR (300 MHz) (00%) 8 0.861 (d, J: 7 Hz, 3H), 0.864 (d,
J: 7 Hz, 3H). 1.00-1.35 (m. 3H), 1.43 (oct. J: 7 Hz, 1H), 1.65-2.00 (m, 4H). 2.18-
2.35 (m, 2H). 4.56-4.60 (m, 2H); 130 NMR (75 MHz) (CDCI3) 519.72. 19.74,
27.50, 29.37. 32.59, 35.23, 38.72. 45.74, 106.65, 150.17.

I 3-(1-MethylethyI)-1-methylcyclohexane (195). A 25 mL flask was loaded
with Pt02 (50 mg). 1 94 (0.35 g, 2.5 mmol). and methanol (5 mL). The
hydrogenation was carried out as in 189. G0 analysis determined the ratio of
isomers to be 2.42:1 (trans:cis).

4-(1-MethylethyI)-methylenecyclohexane (196). The following reagents
were combined as described for 188; NaH (0.41 g, 17.1 mmol) in DMSO (17 mL);
methyltriphenylphosphonium iodide (6.92 9, 17.1 mmol) in DMSO (40 mL); and
161 (1.94 g, 14.3 mmol). After workup and distillation (bpsa 60-70 °C). 1.64 g 196
was recovered (86% yield); IR (Film) 3073, 2963, 2874. 1651 om";1H NMR (300
MHz) (CDCI3) 5 0.85 (d. J: 7 Hz. 6H), 0.98-1.25 (m, 3H). 1.42 (oct. J : 7 Hz, 1H).
1.70-1.84 (m, 2H). 1.90-2.05 (m. 2H). 2.22-2.35 (m, 2H). 4.56 (t, J : 2 Hz. 2H);
130 NMR (75 MHz) (CDCI3) 519.84, 31.10. 32.39. 34.94, 43.66, 106.24, 150.31.

74
4-(1-Methylethyl)-1-methylcyclohexane (197). A 25 mL flask was loaded

with Pt02 (50 m9). 196 (0.35 g. 2.5 mmol), and methanol (5 mL). The
hydrogenation was carried out as in 189. GO analysis determined the ratio of

isomers to be 2.23:1 (trans:cis).

CHAPTER 2.

DIASTEREOSELECTIVE CYCLOPENTANE FORMATION
MEDIATED BY TITANOCENE DICHLORIDE

1 . Introduction

Our initial study began with the cyclization of a substrate utilized by
Grubbs in an experiment previously described.9a 0ur intentions were to
optimize the different steps in the linear sequence (Scheme 15). more accurately
probe the regiocontrol of the olefin insertion, and determine the extent of B-
elimination. Preparation of hex-5-en-1-yltitanocene chloride (198) utilized
established procedures6 in which bromide 28 was metallated by treatment with
magnesium in THF (82. Scheme 26). Upon quenching an aliquot of 82 with an
HCI/diethyl ether solution, the product mixture was determined to consist of a
95:5 ratio of 1-hexene (69) and methylcyclopentane (6). Furthermore, this ratio
did not change upon continued heating. Transfer of the alkenyl ligand to
titanium was accomplished with TiCp20l2 in CH2Cl2, which provided 198 in
quanitatlve yield after purification.69 It is noteworthy that the ratio of 69 to 6 did

not change during the transmetallation process.

75

76
Scheme 26. Preparation of 5-Hexen-1-yltltanocene Chloride

(1 98)
X
M: “092012
Br 31". x + &. noppl
20 02;X:Mper 83;X:MgBr 190; 100% YIeId
69;X:H 6; X:H

Scheme 27 represents the potential pathways that complex 198 could
undergo. Treatment of 198 with a Lewis acid could result in simple destructive
B-elimination to 1,5-hexadiene if the Lewis acid was too strong or the reaction
temperature too high. 0n the other hand, if the conditions were too mild, 198
could be recovered unchanged. Upon fine tuning of the reaction conditions,
productive bond formation could proceed in either an exo (five-membered ring
formation) or an endo mode (six-membered ring formation). After cyclization. the
alkyltitanocene complexes could presumably undergo B-elimination or, if stable
to the reaction conditions, provide either methylcyclopentane or cyclohexane

upon hydrolysis.

Scheme 27. Potentlal Pathways of Cyclization and
Decomposition of 198

noppl H
9) - 4...... _..,, <5

1.4m...

 

Nv4—MEWW [mezc'lfi WH

196

 

1...: m8.

Trcppl H
0 8+1 Elirrlinulon 6 13’ G

77
Treatment of a toluene solution of 198 with EtAlCl2 (0.5-1.0 equiv) at -78

°C resulted in an immediate darkening of the mixture.73 This has been attributed
to the complexation of the aluminum catalyst to the chloride on the titanium. The
role of the aluminum was to either weaken or sever the titanium-chloride bond.
The net result was the development of substantial positive charge on titanium,
polarization the carbon-titanium bond. and to create an available coordination
site for olefin complexation. Hydrolysis of the reaction mixture after 30-60
minutes revealed that complete consumption of 198 had occurred with
concomitant formation of a single regioisomer. that of exo insertion. Yields of 85-
95% (from 198) were common for this insertion.1b Furthermore, no detectable
amount of B-elimination was observed under the reaction conditions.

We have since reported a thorough investigation on the regiospecificity of
5-exo-trig olefin insertions with respect to structural differences of the olefin.13 A
variety of vicinally and geminally substituted olefins. including systems where
the olefin was either endo- or exo-cyclic in a ring and simple acyclic systems,
were prepared and cyclized. In general. to promote complete cyclization, longer
reaction times, higher temperatures, and higher catalyst concentrations were
required. However, these factors had no effect on the yields or the
regiospecificity of the reaction, the product of exo insertion was exclusively
observed. A major consequence of the regiocontrol exhibited by this
organometallic approach was the ease with which quaternary centers were
produced. including geminal dimethyl and angular methyl groups, features
which have proved troublesome by other methods.

In the present chapter, we report our findings regarding the effect a methyl
group on the newly formed asymmetric center. In essence. could a respectable
level of diastereomeric excess be achieved? In order to probe this question. we

Prepared and cyclized substrates analogous to hex-5-en-1-yltitanocene

78
chlorides in which a methyl group was situated at the various positions on the

tether. The titanium metal and its ligands were anticipated to have a strong
influence on the stereochemistry of the product. First, as with other metals,
titanium should be able to complex the olefin, thus restricting all four atoms into a
syn-coplanar relationship, and thereby controlling the olefin geometry during
bond formation. Secondly, in contrast to other ionic or radical methods, the
ligands on the metal could have a pronounced effect on the geometrical
deployment of the olefin and/or the alkyl group. and therefore exert more

influence on the diastereomeric conformations in the transition state.

2. General Features and Limitations for the Synthesls of Methyl-
Substituted 5-Hexen-1-yltltanocene Chlorides

Scheme 26 illustrates our chosen method for the preparation of the
requisite alkenyltitanocene complexes. As already mentioned. reduction of the
alkyl bromide with magnesium produced a small amount of cyclopentylmethyl
magnesium bromide. This phenomena has been well documented by Ashby
and others and has been explained in terms of single electron transfer (SET).37
During this process. transient radical intermediates, thought to be formed in the
reaction, underwent a rapid 5-exo-trig cyclization. This process presented no
major problems for our model substrate since both radical cyclization and our
titanocene-mediated method produced the same compound. However, for the
substrates of interest in this chapter. SET was a menacing dilemma.
Compounds 368-d cyclized to a much greater extent (12-21%) than did 27
(T able 2, Method A), presumably due to the Thorpe-lngold substituent effect.70
In contrast to 28, 368-d produced a pair of diastereomers in a relatively
indiscriminant fashion. which paralleled those for radical cyclization (Table 2,

Method D). These shortcomings prompted a search for an alternative method of

79
Grignard formation. Rieke has developed a variety of methods to generate

“active metals” which have been shown to metallate allylic and benzylic halides
without dimerization.71 We had hoped that rapid generation of the Grignard
reagent at low temperatures with ”active magnesium" may diminish cyclization at
this stage. However, much to our dismay. we found only a marginal decrease in

cyclization and a substantial decrease in the yield (Table 2. Method E).

Table 2. Diastereomer Ratlos of Dlmethylcyclopentane .
Products via Titanium and Free Radical Methods

 

 

 

 

 

. Wins
Oyizann
Precursor Method' YIOIdb Uncyclized 1.2 (trans:cis) 1,3 (trans:cis)
A 100 m 10 2
War 8 100 87 10 3
M. c in o 92 8
(35.) D 90 O 79 21
E 76 91.5 6.5 2
J” A 98 83 3 14
B m 5 3 12
’ 3' c 93 0 3 97
‘3‘”) o 100 14 19 67
A 100 5 9 6
W9 8 64 77 14 9
Ma 0 94 0 92 8
(366) o m o s 41
MD A a 79 7 12°
N B 72 16 2 55°
9' - o as 17 19 61 ~
(3“) E 81 2 7°

 

 

 

‘MetItodA:(1)w,THF.45°C. MethodB:(1) MethodA; (2)1'ioppl,.c+lpl, MethodC:(1)
Method A; (2)MethodB; (3) 1.0 equiv. 614161,. reasons. -78 '0; (4) Hcl. Modo:AlBN,eu,SnH,
C.H.,w°c. MethodE: MgCI,K.THF.0°C. hReference”. °Belenceofmehndooneiehdofhree

pueblo p—hydrogon elimination products.

Ligand transfer from magnesium to titanium provided two interesting
results. Typically, no further cyclization occurred during the transmetallation
Process. However, an 8% increase was observed for 360 (Table 2. Method B).
When Grignard formation of 36c was performed in diethyl ether, as opposed to
THF, followed by transmetallation, 85-100% cyclization took place. Presumably

80
the less Lewis basic solvent, diethyl ether, did not complex the M9X2 as tightly

as THF. thereby allowing M9X2 to promote the olefin insertion. Interestingly,
M9X2 was not an effective catalyst for the other substrates under these
conditions. We are currently investigating the nature of this apparent rate
enhancement for substrate 366. It is noteworthy that the 01.8, and y carbons to
the titanium resemble the growing polyproylene chain in Ziegler-Natta
polymerization, a system in which M9X2 efficiently and stereoselectively
catalyzed the polymerization of propylene. Secondly, ligand transfer of 36d
from magnesium to titanium resulted in almost complete cyclization. even in THF
(Table 2. Method B). The similarity of the diastereomeric ratio formed to that
observed for radical methods, suggest that radical ring closure occurred. not
olefin insertion. It is not clear if the radical was produced by SET or by homolytic

cleavage of the sterically crowded 2° carbon-titanium bond.

3. Synthesis and Cycllzatlon of Methyl-Substltuted 5-Hexen-1-
yltitanocene Chlorides1b

Incorporation of a methyl group on the tether was found to have a
significant effect on the stereochemical outcome of carbon-carbon bond
formation. Treatment of the titanium complex of 36a (Method 0, Table 2) with
EtAlCl2, resulted in a 92:8 ratio (trans:cis) of 1.2-dimethylcyclopentanes. Taking
into account the amount of cis isomer already present, a 94:6 diastereoselectivity
during the Ziegler-Natta olefin insertion was realized. Similarily, 36b underwent
the Lewis acid assisted insertion which produced a 97:3 (cisztrans)
diastereomeric mixture of 1,3-dimethylcyclopentanes. Thus. the actual
diastereoselectivity was > 99:1. Finally. cyclization of the titanium complex of
366 resulted in a high trans selectivity (92:8) and > 99:1 in the EtAlCl2 step. The
fact that titanium complexes of 36b and 360 yielded the same product selectivity

81
for different isomers, negates the possibility for product equilibration under the

reaction conditions.

The stereoselectivity can be explained in terms of a psuedo-chair
transition state (Scheme 28). Treatment of the organometallic complex with
EtAlCl2, resulted in complexation of the Lewis acidic aluminum to the chloride
atom on the titanium.7a Although the extent of bond breaking of the titanium-
.chloride bond is unclear. the net result was distribution of substantial positive
charge on titanium, increasing the electrophilicity of the metal, and weakening of
the carbon-titanium bond. During olefin insertion. a four-centered transition state
has been proposed in which the carbon-carbon double bond was assumed to
be syn-coplanar with the carbon-titanium bond. Substantial stabilization would
result from such an orientation based on orbital considerations previously
described. Conformations 199-202 could accommodate such an interaction.
However. in order to minimize eclipsing interactions. 199 and 202 would be
strongly disfavored. Consequently, the diastereomeric ratio observed resulted
from an unequal partitioning between 200 (major) and 201 (minor).
Conformation 201 would be disfavored. with respect to 200. since the methyl
groups were axially deployed. Additional destabilization of an axial methyl
group could arise from steric interactions with the cyclopentadienyl ligands on

the metal.

82

Scheme 28. Proposed Transition States for Flve-Membered
Rlng Formation

\‘mppl at
+1" m ~—

H

 

 

 

4. Comparlson of the Stereoselective Outcome of Varlous Methods

With the variety of methods available to facilitate cyclization of the 5-
hexen-1-yl system (Table 3), each having different steric and electronic
characteristics at the activated carbon, some qualitative generalizations can be
made about the factors that contribute to high stereoselectivities. Factors that
deserve consideration include the nature of the reactive intermediate. ligands on
the intermediate, and the temperature at which the reaction was conducted. In
the case of radical cyclization (80 °C). selectivity drops off sharply as the methyl
group was further removed from the olefin. Beckwith17 and Houk2° agree.
based on independent calculations. that the major product was formed through a
chair transition state in which the methyl group was in a pseudo equatorial
position. As a point of divergency, Houk suggested that the conformation
leading to the minor product resembled a boat transition state with the methyl

group in an equatorial position. not a chair conformation with an axial methyl

83
group. In support of this. Houks calculations more accurately predict the relative

ratio of products. The difference in energy between the two transition states
(estimated to be 0.5 kcal/mol) was attributed to the closer proximity of the C3
methylene (relative to the radical center) to the olefin. Examination of lithium
cyclizations (~30 to 25 °C) revealed that excellent stereocontrol was realized for
each substrate. In order to show that this drastic difference was not merely
related to temperature, aluminum cyclizations are noted (67 °C). As in the case
of lithium. excellent control over the newly formed asymmetric center was
produced, despite the higher temperatures. Assuming Houk's model was an
accurate description of the minor isomer, incorporation of the 1: complexing
ability of the metal(s) into the transition state would increase the nonbonding
interactions with the 03 methylene and in turn, suppress the boat transition state.
It can be seen by comparison of aluminum- and lithium-mediated cyclizations.
that the ligands on aluminum had a subordinate role in the stereochemical
outcome of bond formation. The minimal effect that the alkyl ligands on
aluminum had on the stereoselectivity was likely a consequence of the sp2
hybridization of aluminum, which directs the ligands away from the alkyl tether.
Finally, cyclization mediated by titanium demonstrated the combined effects of it
complexation. low temperatures (-78 °C), and proper geometry of the ligands for
increased interactions with the alkyl tether. Thus. maximum stereocontrol was

achieved.

84

Table 3. Comparison of Dlastereoselectlvltles
for Different Methods

 

 

 

 

Cyclization Dimethycyclopentanes
Precursor M 1,2 (trans :cis) 1.3 (trans:cis)
WM Radical 79:21
Me U 97:3
AlEtz 97:3
TICpZCI 94:6
M. Radical 19:67
W Li 6:94
M AIEIZ 8192
T10p20l 1 :99
MM Radical 59:41
Me U 97:3
A1512 96:4
TiCp20I 99:1

 

 

 

 

 

5. Optimization and Cyclization of 133: Formation of
Bicyclo[4. 3. 0]nonane

Methodology directed toward the efficient formation of multiple ring
products is an important transformation. With the advent of new reaction
conditions to facilitate olefin insertion for our titanium approach, we were
compelled to reinvestigate the cyclization of 13313-72 In this substrate, 5-exo-
trig insertion into the vicinally substituted olefin presented new challenges
compared to the other substrates in this chapter. The primary obstacle to be
surmounted was bringing the sterically congested olefin into the titanocene
wedge.

Table 4 illustrates some represenative conditions attempted for cyclization
of 133b. As previously shown for other vicinal olefins.“) EtAlCl2 (entry 1) was
ineffective at -78 °C. Since EtAlCl2 was incompatible with 133b at higher
temperatures, recourse to other Lewis acids was necessary. Me2AlCl was
marginally successful in toluene (entry 2), providing a 54% recovery of material

(combined yield of acyclic and cyclic). By simply changing the reaction medium

85
to CH2CI2. reaction times were diminished and conversions were enhanced

(entry 3).

Table 4. Optimizatlon of the Cyclization Conditions for 133b

x a H
OJ -. UV)"
133.; x . Br Yield”

 

 

133b; X 3 “C920. H
1) EtAlCl2, (2 equiv.)
Toluene, -78 °C 93 86 1 4

2) Me2AICI (1.5 equiv.)
Toluene, ~30 to -10°C 54 23 77

3) MezAlCl (1.5 equiv.)
0120': 9° °0 59 10 90

O O o Int-g:

 

 

 

 

 

 

 

‘(1)133a, Mg, THF, 45 1'0; (2)1'le20I2, 50% yield. ”Reference 69.

Scheme 29 depicts the comparison of radical and titanium-mediated
cyclization of 133. Conversion of 133 to 133b was accomplished in the usual
fashion, which resulted in a 50% yield after purification. Further optimization for
the isolation of this complex is warranted. Treatment of 133b under the
optimized conditions resulted in the exclusive formation of the cis isomer in 82%
yield. Similarily, treatment of 133 under the tin-hydride conditions produced the
same stereoselectivity. Apparently, the highly rigid nature of this 5-hexen-1-yl

analog precludes attack from the opposite face.

86

Scheme 29. Comparison of Titanium-Mediated Cyclization to
Free Radical Cyclization for 1338

H H
X
8u3SnH. AIBN. Me
8.4.808 O\/l ..
99% YIOId I 5
H

8

133a; X:Br I:
1) Mg. THF 0
2) T160202.
CH
50% Yield
X MezAICl (2 equiv)
CHZCIZ, 9 91 0
-50 to -30 °C.
90% Yleld

 

1330; X-TiszCI

6. Conclusion

The results described in this chapter demonstrate a powerful new tool for
stereoselective cyclopentane formation. Several general features about the
efficiency and selectivity of the Ziegler-Natta insertion deserve mentioning.
Activation of the hex-5-en-1-yltitanocene complexes with 0.5-1.0 equivalents of
EtAICI2 facilitated complete cyclization within 30 minutes at -78 °C. Olefin
insertion was completely regiospecific (exo cyclization) and the conditions did
not produce detectable amounts of B-hydrogen elimination products. The
elegance of this methodology was inherent in the ability of the titanium and its
ligands to control the selectivity of carbon-carbon bond formation. In all cases.
>94:6 diastereoselectivity was observed for the intramolecular olefin insertion

with >90% yields achieved.

87
7. Experimental

Radical Cyclization of 6-Bromo-3-methyl-1-hexene (368). A Schlenk
flask was loaded with AIBN (10 mg), evacuated, purged with argon. and then
benzene (10 mL) was added. Bromide 368 (88.5 mg, 0.5 mmol) and Buaan
(174.6 mg. 0.6 mmol) were added to the flask and the reaction mixture was
heated to 80 °C for 8 hours. Analysis of the reaction mixture using gas
chromatographic techniques69 revealed that a 90% yield of cis- and trans-1,2-
dimethylcyclopentanes (21 :79, respectively) was obtained (see Table 2. Method
D).

Radical Cyclization of 6-Bromo-4-methyI-1-hexene (36b). A Schlenk
flask was loaded with AIBN (10 mg), evacuated. purged with argon, and then
benzene (10 mL) was added. Bromide 36b (88.5 mg, 0.5 mmol) and Bu38nH
(174.6 mg, 0.6 mmol) were added to the flask and the reaction mixture was
heated to 80 °C for 8 hours. Analysis of the reaction mixture (see Table 2,
Method D)69 showed the reaction resulted in a 100% yield of a mixture of 4-
methyl-t-hexene (14%) and cis- and trans-1,3-dimethylcyclopentanes (86%.
7822. respectively).

Radical Cyclization of 6-Bromo-5-methyI-1-hexene (360). A Schlenk
flask was loaded with AIBN (10 mg), evacuated, purged with argon. and then
benzene (10 mL) was added. Bromide 36c (88.5 mg, 0.5 mmol) and Buaan
(174.6 mg, 0.6 mmol) were added to the flask and the reaction mixture was
heated to 80 °C for 8 hours. Analysis of the reaction mixture using gas
chromatographic techniques69 revealed that an 88% yield of cis- and trans-1,3-
dimethylcyclopentanes (41 :59, respectively) was obtained (see Table 2. Method
D).

88
Radical Cyclization of 6-Bromo-6-methyI-1-hexene (36d). A Schlenk

flask was loaded with AIBN (10 mg), evacuated, purged with argon, and then
benzene (10 mL) was added. Bromide 36d (88.5 mg, 0.5 mmol) and Buaan
(174.6 mg, 0.6 mmol) were added to the flask and the reaction mixture was
heated to 80 °C for 8 hours. Analysis of the reaction mixture (see Table 2,
Method 0)69 showed the reaction resulted in a combined yield of 85% of the
following mixture: 1-heptene (17%) and cis- and trans-1.3-dimethylcyclopentanes
(83%, 77:23, respectively).

Grignard Formation of 6-Bromo-3-methyI-1-hexene (368) Using Rieke
Magnesium. Potassium metal (70.4 mg. 2.18 mmol) and anhydrous MgCl2
(95.21 mg, 1.00 mmol) were loaded into a Schlenk flask under an atmosphere of
nitrogen and THF (4 mL) was added. The mixture was heated to reflux for 3
hours, cooled to 0 °C, and 368 (88.45 mg, 0.5 mmol) was added. Analysis of the
final mixture (see Table 2, Method E). after quenching with 1.0 M HCI/E120 (3
mL). showed that a combined yield of 76% of the following mixture was obtained:
3-methyI-1-hexene (92%) and cis- and trans-1.2-dimethylcyclopentanes (8%. 2:6,
respectively).

Grignard Formation of 6-Bromo-6-methyI-1-hexene (36d) Uslng Rieke
Magnesium. Potassium metal (70.4 mg. 2.18 mmol) and anhydrous MgCl2
(95.21 mg. 1.00 mmol) were loaded into a Schlenk flask under an atmosphere of
nitrogen and THF (4 mL) was added. The mixture was heated to reflux for 3
hours, cooled to 0 °C, and 36d (88.45 mg. 0.5 mmol) was added. Analysis of the
final mixture (see Table 2. Method E). after quenching with 1.0 M HCL/Et20 (3
mL). showed that a combined yield of 76% of the following mixture was obtained:
1-heptene (81%), cis- and trans-1.2-dimethylcyclopentanes (9%, 2:7,
respectively), and a mixture of acyclic B-hydrogen elimination products (10%).

89
General Procedure for ZIegIer—Natta Cyclization: 6-Bromo-3-methyI-1-

hexene (368). A Schlenk flask was charged with M9 (195 mg. 8.0 mmol) and
heated under vacuum with a heatgun for 30 minutes. After cooling to ambient
temperature, the flask was purged with argon, THF (2 mL) was added. and the
flask was then placed in a 45 °C oil bath. Bromide 368 (353.8 mg. 2.0 mmol) was
added with a gas-tight syringe over 6 hours. followed by stirring for an additional 4
hours. A separate Schlenk flask was loaded with Tle2Cl2 (598 mg, 2.4 mmol),
evacuated and purged with argon, the Tle20l2 was suspended in CH2CI2 (8 mL),

and the flask was cooled to -45 °C. Transmetallation of the organic ligand was
accomplished by addition of the Grignard intermediate to the titanocene dichloride
solution via cannula (an additional 1 mL THF was used to rinse the Mg). stirred for
30 minutes at -45 °C. and warmed to ambient temperature for an additional 5
hours. After transmetallation. the mixture was concentrated to approximately 4
mL followed by purification of the organometallic complex by filtration to remove
the excess TiCp20l2 and the Lewis acidic Grignard salts. Purification was
accomplished by taking the solution up in toluene (5 mL) and hexane (5 mL),
stirring for 10 minutes at room temperature. and then cooling to 0 °C for 10
minutes. The entire mixture was transferred via cannula into another Schlenk
flask (the flask was previously evacuated and filled with argon as before) that was
equipped with a glass frit and the solution was pulled through the frit under a slight
vacuum. The solids were rinsed with additional toluene (3 x 10 mL) and filtered.
Concentration of the reaction mixture in vacuo provided the alkyltitanocene
complex as a red slurry. which was diluted with toluene (14 mL) and cooled to -78
°C. The precise composition of the reaction mixture. consisting of cyclic and
acyclic alkyltitanocene complexes, is illustrated in Table 2 (Method B). A 1.8 M
solution of EtAlCl2 (2 mmol) was added to the alkyltitanocene solution over 10

minutes. the reaction mixture was stirred for 60 minutes at -78 °C, and finally

90
quenched with 1.0 M HCI/Et20 (5 mL). Analysis of the final reaction mixture

(Table 2. Method C) showed that a 93% overall yield from the bromide of cis- and
trans-1.2-dimethylcyclopentanes (8:92. respectively) was obtained.

Ziegler-Natta Cyclization of 6-Bromo-4-methyI-1-hexene (36b).
Bromide 36b (353.8 mg. 2.00 mmol) was cyclized according to the general
procedure described for 368. The precise composition of the reaction mixture
prior to addition of EtAlCl2, consisting of cyclic and acyclic alkyltitanocene
complexes. is illustrated in Table 2 (Method 8). Analysis of the final reaction
mixture, after hydrolysis (Table 2. Method C), showed that a 78% overall yield
from 36b to cis- and trans-1,3-dimethylcyclopentanes (97:3. respectively) was
obtained.

ZIegler-Natta Cyclization of 6-Bromo-5-methyl-1-hexene (36c).
Bromide 36c (353.8 mg, 2.00 mmol) was cyclized according to the general
procedure described for 368. The precise composition of the reaction mixture
prior to addition of EtAlCl2, consisting of cyclic and acyclic alkyltitanocene
complexes. is illustrated in Table 2 (Method B). Analysis of the final reaction
mixture, after hydrolysis (Table 2. Method C). showed that a 61% overall yield
from 36c to cis- and trans-1,3-dimethylcyclopentanes (8:92, respectively) was
obtained.

Ziegler-Natta Cyclization of 6-Bromo-6-methyI-1-hexene (36d).
Bromide 36d (353.8 mg. 2.00 mmol) was converted to the titanocene derivative as
described in the general procedure. Analysis of the reaction mixture after
transmetallation (Table 2, Method 8) showed a complex mixture of products was
obtained in 72% overall yield from bromide 36d, which consisted of the following:
cis-1.2-dimethylcyclopentane (55%). trans-1.2-dimethylcyclopentane (22%). 1-
heptene (16%), 1.6-heptadiene (3%), and 1,5-heptadiene (4%, mixture of

isomers).

CHAPTER 3.

DIASTEREOSELECTIVE CYCLOHEXANE FORMATION
MEDIATED BY TITANOCENE DICHLORIDE

1 . Introduction

Probing the limits of our carbocycle forming methodology. compelled us to
investigate the regio- and stereochemical preferences for the cyclization of
substituted 6-hepten-1-yltitanocene chlorides, leading to dialkylcyclohexanes.
The potential impact of such a study would be significant, if successful. since few
methods exist for such a transformation. We were concerned that the entropic
factors that denied access to 6-exo—trig cyclizations for other methods may
diminish the success of our pursuits. Our primary concerns were the possible
repercussions posed by the increased tether length. The increased flexibility of
the alkyl tether could provide access to the endo insertion pathway. or at least
diminish the diastereomeric preference during carbon-carbon bond formation.
Based on the intramolecular-intermolecular polymerization of 1,6-heptadiene,
Marvel has provided convincing evidence that endo insertion was a minor
consideration.8a The polymer which was obtained consisted of 1.3-
dialkylcyclohexane monomer units, although the stereochemical details for this
polymerization are undetermined. In our intramolecular Ziegler-Natta
carbocyclization study. we found that the optimum conditions for cyclization were

highly dependent on the structure of the substrate. However. by judicious choice

91

92
of solvent, Lewis acid, temperature, and concentration, each substrate could be

efficiently cyclized to yield methylcyclohexane products.

2. Optimization Studies for the Cyclization of 6-Hepten-1-
yltitanocene Chlorides

Synthesis of the various 6-hepten-1-yltitanocene complexes paralleled
the method discussed previously with yields ranging from 75-100% (Method B,
Table 5). A number of differences between the 5-hexen-1-yl and 6-hepten-1-yl
systems are noted. First, although not surprising, no significant amount of
cyclization resulting from SET could be detected during Grignard formation. This
result was consistent with the fact that 6-exo-trig radical cyclizations were much
slower (k:5.4 x 103 s“, 25 °C) than the corresponding 5-exo-trig (k:2.3 x 105 5'
i. 25 °C) cyclization.17 This fact was emphasized by the low conversions (11-
30%) obtained for the 7-bromo-1-heptene substrates when treated under the tin-
hydride conditions (Table 5, Method A) used for the preparation of five-
membered rings (Chapter 2). In most cases. transfer of the alkenyl ligand from
magnesium to titanium did not cause further cyclization. However. in the case of
160, 30-35% cyclization was observed during transmetallation when performed
in the usual solvent (CH2CI2). The extent of ligand cyclization could be
controlled (10-20%) by use of toluene as a solvent. This cyclization phenomena
paralleled that of 366, as both compounds contained an alkyl group (3 to the
metal. These peculiarities will be the subject of Chapter 4. Finally, more forceful
conditions were required to promote these cyclizations, which were found to be
substrate dependent. For the sake of comparison. reaction times and Lewis acid
concentrations increased from 30 minutes/1.0 equiv EtAICI2 to promote five-
membered ring formation, to >150 minutes/>20 equiv EtAlCl2 required for six-

membered ring formation. What follows is a description of the various reaction

93
conditions used in order to achieve optimum conversion, yield,

diastereoselectivity, and minimization of B-hydrogen elimination for each
substrate.

0ur optimization study was initiated by attempting the cyclization of the 6-
hepten-1-yltitanocene chlorides using the conditions developed for
cyclopentane formation (Table 5). Much to our delight, the first substrate chosen
(203b) produced an 89% yield of cyclohexane products (Method C); no
evidence of methylcycloheptane was observed. The reaction mixture consisted
of a 76:17 (trans:cis) ratio of 1.3-dimethylcyclohexanes plus 5% 3-methyl-
methylenecyclohexane (the product of B-hydrogen elimination). The amount of
B-hydrogen elimination was dependent on the temperature at which the reaction
was quenched, but even at -78 °C, approximately 5% B-hydrogen elimination
was routinely observed. Treatment of 144b under the conditions of Method 0
resulted in complete conversion (85% yield) to 1,2-dimethylcyclohexanes with
excellent stereocontrol (>99:1. transzcis). In this case. no evidence of B-
elimination was seen. Similarly, activation of 151b with EtAICI2 (Method C)
produced a 79% yield of 1,3-dimethylcyclohexanes in >98:2 selectivity favoring
the cis isomer. At this juncture. the only modification required of the previous
cyclization conditions was an increase in reaction times. However. treatment of
155b under these conditions produced only 25% cyclization. Because the
uncyclized material coincided with the solvent peak by gas chromatographic
analysis. we could not determine if the low yield of cyclohexane products was
due to incomplete cyclization or some unknown side reaction. Moreover, low
conversions were also observed for the cyclizations of 160b and 137b (56%

and 27%, respectively).

94

Table 5. Diastereomer Ratios of Cyclohexane Products via
Titanium and Free Radical Methods

Me
R‘ | R‘ I H R‘ R1
Method C
—> + +
R2 R‘ R2 R‘ R2 R‘ R2 R‘
R3 R3 R3 R3

 

 

 

 

 

 

 

 

 

 

 

Cyclization 6 Methyl- Dialkylcvclohexane Products
Precursor Method‘ Yieldb Uncyclized 07mm 1.2 (to) 1.3 (to) 1.4 (to)
x .
W A 84‘ 84 8:5
144.; x . a. B 100 m 1:1
1448; x . rrcopl C m 0 99:1
Me
X A 90‘ 81 7:10
1818; x . Br 8 100 100 0.0
181b; x .. Troppl c 79 o 2:98
x
AM A 7541 a 20:"
M’ 8
1884: x - or 95 9 1 :1
1588; x .mppl C 25 e 7129
Me
NW" ‘ 9°“ 8° 79
9888; x . Br 3 100 m 22
2088; x . Troppl C 89' 2 75;; 7
iPr
x A a 70 1020
1808; x - 8r 3 E 64 1323
1888; x mom 0 95° 41 3128
x
1978; x . Br 8 73 m 1
1m: X “le9 O 51 a 52

 

 

 

'MethodA: AIBN, Bu,SnH. 0.11.. 8O 110. Mehod 8: (1) Mg. THF, 45 “c. (2) rrcppl, 011,01, Melted c:
(1) Method 8, (2) 1.2 equiv. EtAlCl2, Toluene, -78 -c. (3) Hovefher. °nelerence 89. “Numbers represent molar
percent d1141.81yleyeloheprane (314) was produced. ° 8-Hydropen eirninetion (51011.) was prowoed

'Cvdohom (1%) was produced

 

95
Entropic factors were certainly responsible for the slower cyclizations and

possibly the decreased yields for certain 6-exo-trig cyclization substrates. The
greater flexibility of the alkyl tether apparently diminished the ability of the olefin
to become properly aligned with the carbon-titanium bond. This being the case,
intermolecular insertions may supersede intramolecular insertions under these
conditions. Since the transition state geometry was unclear. the non-bonding
interactions experienced by the alkyl group may be more detrimental for the 6-
exo-trig cyclization, thus retarding cyclization. The following conditions were
systematically varied in order to attenuate intra- versus intermolecular events: (1)
concentration (0.01-0.06 M), (2) different solvents of greater polarity, and (3)
investigation of higher temperatures. Polar solvents were investigated in an
attempt to increase the solubility of the organometallic complex. and to aid in the
stabilization of polar intermediates.” Both of these factors could help to
increase the rate of cyclization. Higher temperatures were explored to surmount
these entr0pic and/or steric activation barriers.

Table 6 illustrates a number of conditions used to facilitate the cyclization
of 155b. When temperatures higher than -78 °C were used, Me2AlCI was the
Lewis acid employed since EtAlCl2 was shown to destroy the starting
organometallic complex under these conditions. Entry 1 demonstrated that. in
less than 140 minutes at -30 °C, three cyclohexane products were accounted for
in 93% yield . The excessive amount of B-hydrogen elimination (34%) made it
impossible to determine the selectivity. However. achieving complete
conversion for this substrate was gratifying enough at that moment. In an
attempt to decrease the amount of methylenecyclohexane produced. cyclization
was conducted -50 °C. However, the reaction was substantially slower at this
temperatute (26% cyclization. entry 3). When CH2CI2 was used as the solvent

under otherwise identical conditions (entry 2), a significant enhancement in the

96
conversion was observed (76% yield) with a concomitant decrease in the (3-

hydrogen elimination product (compare entries 1 and 3). Cyclization with
Me2AlCI at -78 °C could not be induced in CH2CI2. However. encouraging
results were realized with EtAlCl2 under these conditions (entry 4). Although the
yield of cyclized material was only 55%. 20% uncyclized material was still
present. An increase in reaction time or Lewis acid concentration may help drive
this reaction to completion. Under these conditions, selectivity was not

observed.

Table 6. Optimization of the Cyclization Conditions for 155b

 

 

Me Me
0leer I a 5
time b ; ¢ % Yield“
1 55b Me (min) uncyclized M9 M9 M9 (cisnrans)
1) MezAICI (1.5 equiv.). 0 0.98 0.02 0.00 0.00
Toluene, -30 °C,
Molarity . 994 M 140 d 0.31 0.28 0.34 m
2) MezAlCl (1.5 equiv.). o 0.98 0.02 0.00 0.00
CI-I20I2 -50 °C.
Molarity _ 0.04 M 130 d 0.36 0.33 0.08 76 (50:50)
3) MezAICl (1.5 equiv.), 0 0.98 0.02 0.00 0.00
Toluene, -50 °C.
m . 0,04 14 180 d 0.12 0.10 0.04 26 (5050)
4) EtAlClz (1.8 equiv.). 0 0.98 0.02 0.00 0.00
-78 '0.
m . 0.04 M 205 0.20 0.25 0.27 0.03 55 (52:48)

 

 

 

'(‘I‘fll- W-THF.45’C.(2)TICfiCl-2.OH20I,°Nunbenrepreeentrrillimolrratioeeceledb1for time-0nd
peroenhgflorlhefimliime.°Referenoes9.‘Cen‘tbedennninedbeceusethepeekformieprmooincidee
withlheeolventpeek.

Due to the lack of selectivity encountered for cyclization of 155b, we were
interested in seeing what effect a larger alkyl group would have on the
diastereoselectivity of bond formation. Table 7 illustrates the results for 164b.

As before. excellent conversion occurred at -30 °C with Me2AlCI (entry 1) yet 8-

97
hydrogen elimination (22%) was the fate of a substantial portion of the

cyclohexylmethyltitanocene intermediates under the reaction conditions. The
longer reaction time in this case was due to the concentration of the reaction.
Similar yields were obtained under the conditions of entry 2 with the usual
decrease in B-hydrogen elimination. As opposed to the substrate with a
corresponding methyl group in this position (155b), 164b underwent facile and
efficient cyclization in toluene at -78 °C (entry 4). More importantly, a 74:26

(cisnrans) ratio of 1-(1-methylethyl)-4-methylcyclohexanes was produced.

Table 7. Optimization of the Cyclization Conditions for 164b

 

 

I a A.“ M“
0mm 5
time $ 9" Yb”:
1 54b 5p, (min) uncyclized” 5p, 3p, iPr ( clearans )
1) MezAlCl (1.0 equiv.), 0 0.97 0.02 0.01 0.00
Toluene. -30 °C.
mm, _ 0.02 M 390 0.10 0.28 0.27 0.22 n
2) MGM. (1.5 mun). 0 0.94 0.04 0.02 0.00
9 & °C.
who}, _ 0.05 M 120 0.03 0.25 0.37 0.15 77
3) EWC'z “-5 “WW-)- 0 0.94 0.04 0.02 0.00
0112012, -78 °C.
Molan’ty - 0.05 M 150 0.14 0.18 0.50 0.02 69 (74:26)
4) EtAlCl2 (1.5 equiv.),
T 0100110, -78 .0. 0 0.97 0.02 0.01 0.00
Molar")! - 0-04 M 150 0.00 0.18 0.52 0.12 82 (7426)

 

 

 

‘(1)1848, w,THF.45°c.(2)T16ppl,.chl,.°Nm18enmpreeanmiundcmecescmdu1for fime-oend
peroentegeforthefineltlrne.°ReIefenoe89.

The conditions developed for 155b were applied to 160b with the results
shown in Table 8. Entry 1 shows that high conversions were possible with a
bulky isopropyl group near the metal center. A 79% yield of three cyclohexane
products was observed with 35% B-hydrogen elimination as before. Changing

98
solvents and decreasing the reaction temperature (entry 2) resulted in a 68%

yield and a marked decrease in B-hydrogen elimination. Finally. cyclization in
CH2CI2 at -78 °C (entry 3) resulted in low conversions (<55%). but the

diastereoselectivity was >99:1 during the Ziegler-Natta insertion reaction.

Table 8. Optimization of the Cyclization Conditions for 160b

a Me Me,“
CIC'iPzTl time ‘9 ¢p % Yieldc
' iPr iPr iPr

 

 

1 60b iPr (min) uncyclizedb (cisnrans )
1) MezAlCl (1.5 equiv.), 0 0.93 0.05 0.02 0.00
Toluene, -30 °C,
Molar", .. 0.07 M 180 0.04 0.15 0.30 0.34 79
2) MezAlCl (1.5 equiv.), 0 0.69 0.20 0.11 0.00
0112012. -50 °C.
Mommy .. 0.04 M 300 0.07 0.29 0.27 0.11 8

3 EtAICI 1.5 uiv.
)Tomnifqat‘é' ) 0 0.69 0.20 0.11 0.00

MolarIty - 0.04 M 150 0,23 0.20 0.31 0.05 84 (199)

 

 

 

'(1) 1808. My. THF. 45 °c. (2) noppl, Toluene. ”Numbers represenf milirnoler ratios scaled b 1 for time-0
and percentage Iorthefinel time. ‘Referenoe 69.

Finally, Table 9 contains the results for the cyclization of 137b. Since
alkyl substituents were absent on the tether. this substrate was expected to
cyclize even more slowly than the other 6-hepten-1-yl substrates. Under the
conditions for the 5-exo-trig cyclization, only 27% cyclization occurred along with
24% uncyclized material (Table 5). Once again, entry 1 showed that high yields
and conversions were possible at higher temperatures but a large amount of 8-
elimination rendered these conditions useless. A decrease in the concentration
(0.044 M vs 0.124 M) resulted in a marked enhancement in conversion. yet only
a 56% yield of methylcyclohexane was produced. An increase in the polarity of
"‘9 reaction medium (entries 4 and 5) revealed that moderate yields but good

°°nversions were possible at -78 °C. To combat the intermolecular insertions.

99
which were probably responsible for the low yields, substantially lower

concentrations were warranted.

Table 9. Optimization of the Cyclization Conditions for 137b

 

 

 

I a MC
Cl 11
°"’ 15 c 65
(min) uncyclizedb 1%. Yielo°
137b
1) Me c1(1.5equiv.),
Tolficlm’ .30 °C 0 0.99 0.01 0.00
WISH” I- 0.05 M .
125 0.04 0.71 0.19 90
2 EtAIC 1.5 uiv.
) “my”? I 0 0.99 0.01 0.00
Molarlty' . 0.12 M
210 0.24 0.27 0.00 41
3) EtAICI (2.0 equiv.),
Tom": .78 °c 0 0.98 0.02 0.00
Molarity . 0.04 M
120 0.00 0.56 0.00 56
4) EtAlClz (15 equiv.) -78 °C 0 0.98 0.02 0.00
Toluene:CH2Cl2 (1:1)
Molarity- 0.03 M 150 0.04 0.64 0.00 64
5 12 EtAI 12 uiv..
) 0.432, .78¢93( .q ) 0.98 0.02 0.00
Molarity . 0.04 M
105 0.07 0.52 0.00 52

 

 

 

°(1) 13711, My. THF, 45 -c (2)1'10ppl,. 011201, °Humbert: representmiliimolar rallossceled 81 IortIme-Oend
percentage for the final time. cReference 69. -

With optimal conditions obtained for each substrate, attention was
focused on performing these cyclizations on a synthetically more useful scale.“
Synthesis of the requisite 6-hepten-1yltitanocene complexes was accomplished
without incident (Table 10. Method A). Activation of 137b with EtAlCl2 under
high dilution conditions (0.01 M, CH20I2. -78 °C), resulted in exclusive exo
cIfolization to provide methylcyclohexane in modest yield (76% yield).
CYcIization of 144b and 151b were efficiently achieved under much more

c=0hcentrated conditions (0.12 M. toluene, -78 °C) and provided essentially a

100
single stereoisomer in each case, 99:1 and 3:97 (trans:cis), respectively.

Although the y-methyl group of 155b had no bearing on the selectivity of bond
formation (50:50). the increased steric demands of the isopropyl group of 164b
significantly enhanced the production of the cis diastereomer (trans:cis, 23:77),
with a paralleled increase in yield (91%). Cyclization of 203b and 160b were
unique compared to the others, as B-elimination seemed to be a more
accessible pathway (9% in each case). Compound 203b showed a useful
preference for the trans isomer (81 :19). Metallation of 1608 with magnesium
provided a high yield of the Grignard intermediate with no evidence of
cyclization. However, transfer of the alkenyl ligand to titanium resulted in 17%
cyclization, even with THF present. The most interesting observation was that
M9X2 was selective for the opposite isomer (trans:cis, 1:2) than produced with
an aluminum cocatalyst. Cyclization of 160b provided three cyclohexane
products (71 :18:9). If the extraordinary amount of cyclization that transpired
during transmetallation was factored out, a 92:8 preference for the trans isomer

was produced during the Ziegler-Natta olefin insertion process.

101

Table 10. Diastereomer Ratios of Cyclohexane Products
Performed Under the Optimized Conditions

Me
R‘ I Ft1 I H R1 R1
Memod A and B + +
———_—>
n2 n4 82 n‘ R2 R‘ 1:12 81
R3 R3 R3

 

 

 

 

 

 

 

 

 

R3
Cyclizan’on MEW- Dialkylcyolohexanes
Precursor MethodI Yield" Uncyclized‘= cyclohexane 1,2 (czt) 1.3 (on) 1.4 (on)
x
M
1379 x . Br A 73 m 1
1378 x . Trcppi e 76 3 97
NWT
Me
1445 x . Br A e m 29
1448 x .TICppl B G <1 99:1
Ms
AMX
1519 x . Br A 82 97 03
1518 x : Troppl B 91 <1 3:97
WK
Me
15511 x . Br A as 97 2,,
1558 x : Troppl 8" 63 <1 50:50
Wx
iPr
1548 x . Br 8' 8) 97 2:1
1848 x .Trcppl 8 91 3 23:77
Mi
IM"
/
2088 X - Br A 97 8 2:2
2088 x . Trcppi 8 72° 2 74:17
iPr '
M"
1809 x . Br A 74 a 6:11
108 x .‘ricppl 8° 88° 2 71 :18

 

 

 

 

 

 

 

‘MefhodA: (1) Bromide, Mg.THF,45°C. (2) TiCp-filz. 011,01, W8: (1)1.75-295 equiv. 5111101,,
Toluene.-78 -c.(3) HCIfether. °neference 69. °N1snoersrepresentrsficsscsletoloo. °ch12msusedss
aeohnnt'Coddnotbedebrmined. Findyieldoelwlabdfromthebromide. 'Toluenewnmedesasoivent.

°8-I-Iyd'ogen elimination (9%) w. observed.

102
These results illustrate for the first time a useful method to affect 6-exo-trig

cyclizations into unactivated alkenes with synthetically satisfying selectivities. In
addition, a number of useful pieces of information have been acquired about the
Ziegler-Natta process by examination of the alkene facial selectivity of a single
insertion. A methyl group was shown to have a pronounced effect during the
syn-coplanar insertion for 1448 and 1518. In contrast, the presence of the
methyl group yto the metal (155b) had no bearing at all on the stereogenicity of
the new asymmetric center. These results paralleled those of the polymerization
of racemic or optically active of-olefins.73 For example. 3-methyl-1-pentene and
4-methyl-1-hexene induced a high degree of stereoselection during
polymerization. yet 5-methyl-1-heptene had little effect on the facial selectivity.
The stereochemical outcome for the cyclization of 203b, provided evidence that
the polymerization of 1.6-heptadiene contained primarily a trans 1,3 relationship.
Finally, the stereoselective divergence between the different cocatalysts. M9X2
and EtAlCl2, suggested that the catalyst-cocatalyst interaction is more

sophisticated than simple generation of [Cp2TlR+].

3. Potential Explanations for the Observed Stereoselectivities

By examination of molecular models. three possible transition states for
Cyclization seemed reasonable. Scheme 30 illustrates these possibilities: chair
(204), boat (206). and twist-chair (208). Each transition state allowed for the
alkyl groups to be directed in psuedo-equatorial positions to minimize
unfavorable non-bonding interactions. The chair transition state was less likely
Since the desirable syn-coplanar alignment of the olefin and the carbon-titanium
bond was precluded. The boat and twist-chair conformations were able to

accommodate this transition state requirement. Severe eclipsing interactions

103
between the substituents at the B and 7 positions (relative to titanium) may tend

to disfavor the boat transition state. Examination of the products formed from the
substrates with various substitution patterns on the alkyl tether, however. did not
allow for differentiation between the boat and the twist-chair. Both conformations
accurately predicted the major product for substitution at the allylic and
homoallylic sites. The apparent similarity between the steric environment
surrounding the 8 and 7 positions suggested that a methyl substituent would
have similar effects in both cases. However, y-methyl substitution produced no
selectivity. Differences in selectivity between the B- and y-methyl compounds
was attributed to the increased steric interactions at the B-position due to the
closer proximity to the congested metal center. The steric environment created
by the cyclopentadienyl ligands on titanium and the Lewis acid-chloride-titanium
complex strongly influenced the axial or equatorial orientation of neighboring

alkyl groups.

104

Scheme 30. Possible Transition States for Slx-Membered
Ring Formation

H RI H
olezn H
EtAlClz H | Fl“ “2
n4 H H
205

   

 

H
R‘ “a 112
R1
EtAlCI2 H
—-—>
H H
ole21'1
TWIST-CHAIR 208 209

4. Conclusion

The results in this chapter describe a powerful new method for the
construction of six-membered rings. This method represents the most successful
procedure to affect 6-exo-trig cyclization from an sp3 carbon with an unactivated
alkene. The need for a syn-coplanar orientation of the alkene and the carbon-
titanium bond had a significant impact on the regio- and stereoselectivity. Olefin
insertion was found to proceed with complete exo specificity, which resulted in
the formation of a new asymmetric center. In most cases, an alkyl group induced
a high degree of alkene facial selectivity, providing synthetically useful

diastereomeric excesses. Although a single general procedure was not

105
realized. a variety of new conditions were developed along the way which will

provide additional avenues to affect the desired bond construction. This method
furnishes the synthetic chemist with another useful tool for the construction of

five- and six-membered rings with the high degree of stereocontrol demanded

by today's complex synthetic targets.

106
5. Experimental

Radical Cyclization of 7-Bromo-1-heptene (1378). The following
reagents were combined and heated to 80 °C in a sealed tube for 10 hours:
bromide 1378 (88.5 mg. 0.5 mmol), AIBN (10 mg). Buaan (174.6 mg. 0.6 mmol),
and benzene (10 mL). Analysis of the reaction mixture (see Table 5. Method A)69
indicated that a combined yield of 96% was obtained which consisted of the
following compounds: 1-heptene (89%), methylcyclohexane (10%), and
cycloheptane (1%).

Radical Cyclization of 7-Bromo-3-methyI-1-heptene (1448). The
following reagents were combined and heated to 80 °C in a sealed tube for 10
hours: bromide 1448 (95.5 mg, 0.5 mmol), AIBN (10 mg), Buaan (174.6 mg, 0.6
mmol), and benzene (10 mL). Analysis of the reaction mixture (see Table 5,
Method ill)69 indicated that a combined yield of 84% was obtained which
consisted of the following compounds: 3-methyI-1-heptene (84%). cis- and trans-
1.2-dimethylcyclohexanes (13%. 5:8, respectively), and methylcycloheptane (3%).

Radical Cyclization of 7-Bromo-4-methyl-1-heptene (1518). The
following reagents were combined and heated to 80 °C in a sealed tube for 10
hours: bromide 1518 (95.5 mg, 0.5 mmol), AIBN (10 mg). Bu38nH (174.6 mg. 0.6
mmol). and benzene (10 mL). Analysis of the reaction mixture (see Table 5,
Method A)69 indicated that a combined yield of 90% was obtained which
consisted of the following compounds: 4-methyl-1-heptene (81%). cis- and trans-
1,3-dimethylcyclohexanes (17%. 10:7. respectively). and methylcycloheptane
(2%).

Radical Cyclization of 7-Bromo-5-methyI-1-heptene (1558). The
following reagents were combined and heated to 80 °C in a sealed tube for 10

hours: bromide 1558 (95.5 mg, 0.5 mmol), AIBN (10 m9). Bu38nH (174.6 mg, 0.6

107
mmol). and benzene (10 mL). Analysis of the reaction mixture (see Table 5,

Method A)69 indicated a combined yield of 75% was obtained which consisted of
the following compounds: 5-methyl-1-heptene (66%). cis- and trans-1,4-
dimethylcyclohexanes (31%, 11:20. respectively). and methylcycloheptane (3%).

Radical Cyclization of 7-Bromo-6-methyl-1-heptene (2038). The
following reagents were combined and heated to 80 °C in a sealed tube for 10
hours: bromide 2038 (95.5 mg. 0.5 mmol), AIBN (10 mg), Buaan (174.6 mg. 0.6
mmol), and benzene (10 mL). Analysis of the reaction mixture (see Table 5.
Method A)69 indicated a combined yield of 84% was obtained which consisted of
the following compounds: 6-methyl-1-heptene (83%), cis- and trans-1.3-
dimethylcyclohexanes (16%, 9:7, respectively), and methylcycloheptane (1%).

Radical Cyclization of 7-Bromo-6-(1-methylethyI)-1-heptene (1608).
The following reagents were combined and heated to 80 °C in a sealed tube for
10 hours: bromide 1608 (110 mg. 0.5 mmol). AIBN (10 mg). Buaan (174.6 mg,
0.6 mmol). and benzene (10 mL). Analysis of the reaction mixture (see Table 5,
Method A)69 indicated a combined yield of 88% was obtained which consisted of
the following compounds: 6,7-dimethyl-1-octene (70%), cis- and trans-t-(t-
methylethyI)-3-methylcyclohexanes (30%, 20:10. respectively).

General Procedure for ZIegler-Natta Cyclization for 7-Bromo-1-
heptene Substrates: A Schlenk flask was charged with M9 (195 mg, 8 mmol)
and flame-dried under vacuum. After cooling to ambient temperature. the flask
was purged with argon. THF (2 mL) was added. and the flask was then placed in
a 45 °C oil bath. The bromide (2 mmol) was added to the magnesium with a gas-
tight syringe over 6 hours, followed by stirring for an additional 4 hours. A
separate Schlenk flask was loaded with Tle2Cl2 (598 mg, 2.4 mmol). evacuated
and purged with argon, the Tle2Cl2 was suspended CH2CI2 (8 mL). and the flask

was cooled to -45 °C. Transmetallation was accomplished by adding the Grignard

108
reagent to the TiCp2Cl2 suspension via cannula (an additional 1 mL THF was

used to rinse the Mg). stirring for 30 minutes at -45 °C, and warming to ambient
temperature where the reaction mixture remained for an additional 5 hours. After
transmetallation was complete. the mixture was concentrated to approximately 4
mL, taken up in toluene (5 mL) and hexane (5 mL), stirred for 10 minutes at room
temperature, and then cooled to 0 °C for 10 minutes. The entire mixture was
transferred via cannula and filtered under argon through a glass frit. The solids
were rinsed with additional toluene (3 x 10 mL) and filtered. Removal of the
solvents in vacuo left behind the organotitanocene chloride as a red slurry.
Toluene was then added and the solution was cooled to -78 °C. In general.
EtAlCI2 was added to the alkyltitanocene chloride solution by three sequential
additions over a period of 90 minutes. When the reaction was complete. 1.0 M
HCI/Et20 (10 mL) was added at -78 °C to quench the reaction.

General Procedure for the Optimization of Cyclization Conditions.
Preparation of the alkyltitanocene chloride complex was the same as described in
the general procedure unless otherwise stated. Typically. 2 mL of the
alkyltitanocene chloride complex was added to a flame-dried flask, diluted to the
desired molarity, and cooled to the appropriate temperature. The Lewis acid to be
used was added over 10-15 minutes and the progress of the reaction was
monitored by gas chromatography. For optimization results for specific
substrates, see the following tables: Table 6 (155b). Table 7 (1648), Table 8
(1608), and Table 9 (137b).

Ziegler-Matte Cyclization of 7-Bromo-1-heptene (137a). Bromide 1378
(353.8 mg, 2.0 mmol) was converted to alkyltitanocene 137b in 73% yield
following the general procedure. The product mixture prior to addition of the
cocatalyst consisted of 1-heptene and methylcyclohexane (99:1), respectively
(see Table 10, Method A). The titanium residue, after filtration and evaporation of

109
solvents, was diluted with toluene (200 mL). cooled to -78 °C, and 1.0 M EtAlCl2

(2.25 eq) was added sequentially as follows: 1.0 equiv. (t:0 min). 0.75 equiv.
(t:45 min). and 0.50 equiv. (t:90 min). Analysis of the final reaction mixture after
hydrolysis (Table 10, Method B) showed that a 76% yield for the Lewis acid
assisted insertion (56% yield from the bromide) was obtained which consisted of
the following compounds: methylcyclohexane (97%). heptene (3%). and
methlyenecyclohexane (0%).

Ziegler-Matte Cyclization of 7-Bromo-3-methyI-1-heptene (1448).
Bromide 1448 (381.8 mg. 2.0 mmol) was converted to alkyltitanocene 144b
according to the general procedure. The product mixture prior to addition of the
cocatalyst consisted of 3-methyl-1-heptene and cis- and trans-1.2-
dimethyloyclohexanes (98:2:0), respectively (see Table 10. Method A). The
titanium residue, after filtration and evaporation of solvents. was diluted with
toluene (15 mL), cooled to -78 °C, and 1.0 M EtAICI2 (2.25 eq) was added
sequentially as follows: 1.0 equiv. (t:0 min). 0.75 equiv. (t:45 min). and 0.50
equiv. (t:90 min). Analysis of the final reaction mixture (Table 10, Method B),
after hydrolysis, showed that an 89% yield overall from the bromide was obtained
which consisted of the following compounds: trans-1.2-dimethylcyclohexane
(98%), cis-1,2-dimethylcyclohexane (1 %). and 2-methyl-1-methlyenecyclohexane
(<1%).

ZIegIer-Natta Cyclization of 7-Bromo-4-methyI-1-heptene (151a).
Bromide 1518 (353.8 mg, 2.0 mmol) was converted to alkyltitanocene 151b in
82% yield following the general procedure. The product mixture prior to addition
of the cocatalyst consisted of 4-methyl-1-heptene and cis- and trans-1,3-
dimethylcyclohexanes (97:0:3), respectively (see Table 10. Method A). The
titanium residue. after filtration and evaporation of solvents, was diluted diluted

toluene (15 mL), cooled to -78 °C. and 1.0 M EtAICI2 (2.25 eq) was added

110
sequentially as follows: 1.0 equiv. (1:0 min), 0.75 equiv. (1:45 min), and 0.50

equiv. (t:90 min). Analysis of the final reaction mixture after hydrolysis (Table 10.
Method B) showed that a 91% yield for the Lewis acid assisted insertion (75%
yield from the bromide) was obtained which consisted of the following
compounds: trans-1.3-dimethylcyclohexane (3%), cis-1,3-dimethylcyclohexane
(97%), and 3-methyl-1-methlyenecyclohexane (<1%).

Ziegler-Natta Cyclization of 7-Bromo-5-methyI-1-heptene (1558).
Bromide 1558 (353.8 mg. 2.0 mmol) was converted to alkyltitanocene 155b in
65% yield following the general procedure. The product mixture prior to addition
of the cocatalyst consisted of 5-methyl-1-heptene and cis- and trans-1,4-
dimethylcyclohexanes (97:2:1), respectively (see Table 10. Method A). The
titanium residue, after filtration and evaporation of solvents, was diluted with
CH2CI2 (45 mL). cooled to -78 °C, and 1.0 M EtAlCI2 (2.25 eq) was added
sequentially as follows: 1.0 equiv. (i=0 min), 0.75 equiv. (t:45 min). and 0.50
equiv. (1:90 min). Analysis of the final reaction mixture after hydrolysis (Table 10,
Method B) showed that a 63% yield for the Lewis acid assisted insertion (41%
yield from the bromide) was obtained which consisted of the following
compounds: trans -1,4-dimethylcyclohexane (50%). and cis-1 .4-
dimethylcyclohexane (50%). No detectable amount of 4-methyI-1-
methlyenecyclohexane or 5-methylhept-1-ene was observed.

Ziegler-Nana Cyclization of 7-Bromo-5-(1-methylethyI)-1-heptene
(1648). Bromide 1648 (218.9 mg, 1.0 mmol) was converted to alkyltitanocene
164b in 50% yield following the general procedure. The product mixture prior to
addition of the cocatalyst consisted of 5-ethyl-6-methyl-1-heptene and cis- and
trans-1,4-isopropylcyclohexanes (97:2:1), respectively (see Table 10, Mehtod A).
The titanium residue, after filtration and evaporation of solvents, was diluted with

toluene (12 mL). cooled to -78 °C, and 1.0 M EtAlCl2 (2.25 eq) was added

111
sequentially as follows: 1.0 equiv. (t:0 min). 0.75 equiv. (t:45 min), and 0.50

equiv. (t:90 min). Analysis of the final reaction mixture after hydrolysis (Table 10,
Method B) showed that a 91% yield for the Lewis acid assisted insertion (47%
yield from the bromide) was obtained which consisted of the following
compounds: cis-1-(1-methylethyl)-4-methylcyclohexane (74%). trans-1-(1-
methylethyl)-4-methylcyclohexane (22%), and 5-ethyl-6-methyl-1-heptene (3%).
No detectable amount of 4-(1-methylethyl)-1-methlyenecyclohexane was
observed.

ZIegIer-Natta Cyclization of 7-Bromo-6-methyI-1-heptene (203a).
Bromide 2038 (381.8 mg, 2.0 mmol) was converted to alkyltitanocene 2038 in
97% yield following the general procedure. The product mixture prior to addition
of the cocatalyst consisted of 6-methyl-1-heptene and cis- and trans-1,3-
dimethylcyclohexanes (96:2:2), respectively (see Table 10, Method A). The
titanium residue, after filtration and evaporation of solvents. was diluted with
toluene (12 mL). cooled to -78 °C. and 1.0 M EtAlCl2 (2.25 eq) was added
sequentially as follows: 1.0 equiv. (t:0 min), 0.75 equiv. (t:45 min), and 0.50
equiv. (1:90 min). Analysis of the final reaction mixture after hydrolysis (Table 10,
Method B) showed that a 72% yield for the Lewis acid assisted insertion (63%
yield from the bromide) was obtained which consisted of the following
compounds: trans -1,3-dimethylcyclohexane (74%). cis-1.3-dimethylcyclohexane
(17%). 3-methyI-1-methlyenecyclohexane (7%), and 6-methylhept-1-ene (2%).

ZIegIor-Natta Cyclization of 7-Bromo-6-(1-methylethyI)-1-heptene
(1608). Bromide 1608 (437.8 mg. 2.0 mmol) was converted to alkyltitanocene
160b in 74% yield following the general procedure except that the transmetallation
was performed in toluene. The product mixture prior to addition of the cocatalyst
consisted of 6,7-dimethyl-1-octene and cis- and trans-1-(1-methylethy)-3-

methylcyclohexanes (83:11z6), respectively (see Table 10. Method A). The

112
titanium residue after filtration and evaporation was diluted CH2CI2 (30 mL),

cooled to -78 °C, and 1.0 M EtAlCl2 (2.25 eq) was added sequentially as follows:
1.0 equiv. (i=0 min). 0.50 equiv. (t:120 min). and 0.25 equiv. (1:240 min).
Analysis of the final reaction mixture after hydrolysis (Table 10. Method B)
showed that a 88% yield for the Lewis acid assisted insertion (63% yield from the
bromide) was obtained which consisted of the following compounds: trans-1-(1-
methylethy)-3-methylcyclohexane (70%). cis-1-(1-methylethy)-3-
methylcyclohexane (18%), 3-(1-methylethyl)-1-methylenecyclohexane (9%). and
6,7-dimethyl-1-octene (2%).

CHAPTER 4.

MECHANISTIC STUDIES OF ZIEGLER-NATTA
POLYMERIZATION:

PROBING FOR a- OR B-HYDROGEN ACTIVATION

1. Introduction

During the development of our titanocene mediated ring-forming
methodology. we observed peculiar behavior for both the 5-hexen-1-yl- and 6-
hepten-1-yltitanocene chloride complexes when an alkyl group was in the )3
position with respect to the metal. For instance, reaction of a Grignard solution of
368 in diethyl other with titanocene dichloride resulted in nearly complete
cyclization in less than one hour (eqn. 20). To fulfill our synthetic objectives at the
time, we found that olefin insertion during transmetallation could be suppressed if
the reaction was run in the presence of THF. This result was presumably due to
the higher basicity of THF. which allowed for stronger complexation with M9X2 in
comparison to the corresponding diethyl ether complex. Substrates 368-b did not
cyclize to any extent under these conditions. In an analogous situation, the M9X2
salts catalyzed the cyclization of 160b to an extent of 35%, even with THF present
(eqn. 21). The most interesting aspect of this latter example was that these
conditions promoted olefin insertion with an opposite stereochemical preference
(cisrtrans, 2:1) compared to that promoted with EtAlCl2 (cisrtrans. 8:92). These
results have launched us into an exploration of the mechanism of Ziegler-Natta

polymerization.

113

114

 

 

231985881. M
I p; . o uone M e
W Br (3) HCI 2 W H + 90,

Me > (20)

364: Me
15:85 (trans:cis, 75:10)
iPr (1) M9. 5120 II" M" iPr
/\/\)\/B (2)1‘leZCl2. Tom“ NVk/ U
/ I (3) HCI ’ / H + (21)
160 65:35 (trans:cisJ 1 :24)

The apparent rate enhancement, observed only with an alkyl group present
in the position 8 to the metal. suggested the possible intervention of an electronic
effect. Possible explanations for these observations included either 01- or 8-
hydrogen activation (Scheme 31 ).74 During both a-hydrogen activation (210) and
B-hydrogen activation (211). the cocatalyst presumably complexes the chloride
atom on titanium, causing an increase in the positive charge on titanium and a
simultaneous polarization, and thus weakening, of the carbon-titanium bond.
Based on Figures 1-3 (see Introduction). there are precedents that either 01- or 8-
hydrogens could interact with the metal center. The nature of this interaction is
unclear, but assuming the interaction would be to stabilize the metal center, a
partial negative charge on hydrogen would be necessary. Also. the possible
interaction of the cocatalyst with either the 01- or B-hydrogen cannot be ruled out.
The net result would be the development of positive charge on either the tit-carbon
(210) or the 8-carbon (211) depending on the mechanism. The role of the B—alkyl
group could be to stabilize the development of the positive charge by either
hyperconjugation (210) or through induction (211). Both 01- or B-hydrogen

activation could reasonably account for an increased rate of cyclization.

115

Scheme 31 . Mechanistic Possibilities for ZIegIer-Natta
Polymerization: 01- and [ii-Hydrogen Interactions

a”’
5-
5+ 5+ R
CPzTI ‘ 4"
:= := ‘/ c“j4

CI I
e I d
‘ M1...

210 211

 

 

To test these potential pathways, we designed a series of competitive rate
experiments (Scheme 32). Our idea resembled aspects of independent

9° in which the relative strengths

experiments reported by Grubbsga and Bercaw,
of carbon-hydrogen and carbon-deuterium bonds were central to the design. The
major difference in our approach was how the deuterium isotope effect was to be
measured. Both Grubbs and Bercaw used the stereochemistry of the cyclized
products as an instrument for measuring an isotope effect. Our intentions were to
compare the relative rates of olefin insertion of two similar alkenyltitanocene
complexes in which one contained a hydrogen atom (01 or 8) and the other a
corresponding deuterium atom (01 or 8). Ideally, the most informative reaction
would be to directly monitor the relative rates of cyclization between 2128 and
2138 where R1:R2. However, we intended to monitor the cyclization progress
using gas chromatographic techniques which would make this approach
impossible. Alternatively, an indirect method for obtaining the same information is
outlined in Scheme 32 in which R1:R2. First, cyclization of a mixture of 2125 and
2138 (R‘s-R2) will determine the minor role of the B-alkyl substituents on the
relative rates of olefin insertion. By graphing the conversion versus time for each
substrate, the relative rate of cyclization can be obtained from the ratio of the
slopes [km-slope(2128/213a)]. Secondly. 212b and 2138 will be treated under

the identical conditions and the resultant conversions graphed in a similar fashion.

116
In this way, kH/ko can be obtained by the following ratio:

[slope(212a/213a)]/[slope(2128/213a)]. Finally, the competitive cyclization of
2128 and 213b provides a control for the second cyclization. A similar series of
competitive reactions will also be performed in which a,01-dihydrogen substrates
will be compared to or,0:-dideuteron substrates. These reactions outlined above
could provide insight into the fundamental processes of Ziegler-Natta

polymerization.

Scheme 32. Competitive Rate Experiments Designed to Test the
Mechanistic Possibilities

ClC 11E: ) ClC nu (”Mead Mega, Mag
92 H R‘+ 92 H R2—> (-—>2)H. H2

2128 213a
ClezTif: ) ni- ClezTif: P 82,—. (1)Lewisacid M°§_RH NIKE—Ra
D H (2)H
212b 2138
CIC Tin CIC 1; : >2 (1)LewisacidMe‘21q:—W mxé—
H D
2128 2138 2152

2. Preliminary Findings for the 2-AIkyl-5-hexen-1-yltltanocene Chlorides

Our initial investigation involved competitive rate experiments in which
cyclization was to be facilitated under the transmetallation conditions (Scheme
33). Two major problems were associated with this system. First of all,
quenching an aliquot of the Grignard intermediates revealed that 40% of the
reaction mixture from 129 were cyclopentane products compared to 20% for 36c.
Secondly. transmetallation of these Grignard species to 'I”le2Cl2 resulted in rapid
(<20 minutes) cyclization upon warming to ambient temperature. When colder

temperatures were maintained throughout the transmetallation process.

117
conversion did not occur to as great an extent. One major problem, SET

cyclization during Grignard formation. could be avoided if THF was used as a
solvent. but this also inhibited the effectiveness of M9X2 as a Lewis acid and
decreased the amount of cyclization. The problems associated with this system
suggested that extraction of the pertinent rate information would be difficult.

Specifically, the time window for monitoring the reaction progress was too small.

Scheme 33. Cyclization of 5-Hexen-1-yltltanocene Chlorides
Performed Under the Transmetallation Conditions

+ ——> + + 9'
R1 Me Pr M0 Me Pr Pr

36c : RI-H 129 (1) Mg, Etp 80 20 62 38
(2) TICQCIZ. Toluene 25 75 13 87
-45 to 25 °C
125 :R‘:D 129 (1) M9. 5120 80 20 62 38
(2)1'iCmCI2, Toluene 80 20 62 38
-45 °C

In order to avoid the rapid and uncontrollable cyclization discussed above.
we opted to pursue 7-bromo-1-heptene substrates. We have demonstrated that
these substrates cyclized to a lesser extent during Grignard formation, and the
corresponding titanocene derivatives undement 6-axo-trig insertion more slowly
than the corresponding 6-bromo-1-hexene compounds. Therefore, this
modification has the potential for overcoming both of the problems associated with
the cyclization of 6-bromo-1-hexene substrates. which should increase the
window of observation with a simultaneous increase in the accuracy and precision

of the data.

118

3. Optimizatlon Studies for the Competitive Cyclization of 2-AIkyI-6-hepten-
1-yltltanocene Chlorides with MezAICI, M9X2, and MAO

Our first task was to develop reproducible cyclization conditions using the
aforementioned Lewis acids that would provide high yields of cyclic products.
concurrent with a large window of time (90-180 minutes) to view the linear
conversion to cyclic products. In order to graph conversion versus time and
interpret the results with a high degree of certainty, we needed to accumulate
enough data points (5-8) to minimize the error. These considerations are
legitimate since an isotope effect in a similar system was reported to be
kH/ko:1.2.9c The variables that were systematically changed include the Lewis
acid. equivalents of Lewis acid. solvent, concentration. and temperature. We
chose to investigate three Lewis acids for these competitive rate experiments.
Me2AlCI. M9X2, and MAO. The interesting reversal in stereochemistry using
M9X2 as Opposed to Me2AlCI or EtAlCl2, suggesting a different mechanism,
coupled with the propensity of this catalyst to provide isotactic polypropylene
made M9X2 ideal for our investigation. Although the structure of this catalyst is
unclear. MAO also furnished isotactic polypropylene which compelled us to study

its behavior.

119
Table 11. Cyclization of Unpurifled 175b Using M9X2 or MAO

 

 

 

 

 

a Me
I noppl I H
“.11 H
Pr time I'I pr p
H (min) P' (cis/trans) r % Yieldc

1) ngz (x.c1 or Br, 0 96 371 o

1-0 equiv-l. Toluene. 420 41 39/8 4 d

45 °c.

Molarity - 0.02 M
2) MgCI2(8.0 equiv.) + o 96 371 0

)lflolxz (X-Sls or 82'" c 210 44 24720 5 93

o uone, to 5 ° .

Molar” _ 0.02 M. 590 14 33721 14 82
3) MgCl2(8.0 equiv.) + 0 96 3/1 0

M9X2 (X-CI or Br). 150 10 37716 23 86

T°'U°_fl°- ~45 I0 50 °C- 410 10 27715 24 76

Molarity . 0.02 M,
4) MAO (2.0 equiv.), + 0 96 371 0

M9X2 “'31:,” 322°C 80 46 25724 5 100

O UORO, 10 .
Molamy _ 0.02 M. 330 14 46/19 18 97

 

 

 

 

 

 

I(1) 1758 . Mg. THF. 45 °C, (2)11CQCI2, Toluene. I’Numbers represent millimolar ratios scaled to
100 for time-0 and percentage for the final time. °Reference 69. 6No internal standard was used in
this reaction.

Our initial efforts for promoting the cyclization of 175b saw a return to
magnesium dihalides, either with the M9X2 formed in situ during transmetallation
or with an excess of the salts (Table 11). The alkyltitanocene chloride 175b
illustrated in Table 11 was prepared by the addition of a diethyl other solution of
the Grignard reagent to TiCp2Cl2 in toluene. After transmetallation was complete,
the alkyltitanocene complex was not purified from the magnesium salts, but
rather, the diethyl ether was removed in vacuo and the mixture diluted with
CH2CI2. Entry 1 illustrates the effect of one equivalent of M9X2 on the cyclization
process. After 4 hours at 25 °C, no cyclization was detected. However, heating
this mixture to 45 °C led to 51% cyclization after 7 hours, at which time the

reaction ceased. These conditions most notably produced a 5:1 ratio of

120
diastereomers selective for the cis isomer. In an attempt to increase the amount

of conversion, excess MgCl2 (8 equiv.) was employed at 25 °C (entry 2) and 50
°C (entry 3). As expected. higher temperatures facilitated conversion, yet product
loss resulted after prolonged exposure to the reaction conditions. Interestingly,
the use of excess M9X2 caused a dramatic decrease in selectivity and also an
increase in B-hydrogen elimination. Finally. entry 4 demonstrates the behavior of
MAO (2 equiv.) in combination with M9X2 (1 equiv.). The desirable features of
this reaction were the high yield, near complete conversion, and the extended
period of time required for conversion. Although these conditions appeared ideal
for our study. the slow steady conversion proved difficult to repeat at this
temperature. A limitation imposed by avoiding purification of the alkyltitanocene
complexes before introduction of the cyclization catalyst was that occasionally the
effect of concentrating the alkyltitanocene complexes in the presence of M9X2
caused nearly complete cyclization. In order to alleviate this problem. all
reactions henceforth were performed with a "magnesium free“ alkyltitanocene
solution, which was combined with a solution (or suspension in the case of M902

and MgBr2) of the Lewis acid.

Table 12. Cyclization of Purified 1688 and 175b Uslng M9X2

121

 

a; rcppi (I/F:Cp:01

8:2-O-

.I'E R‘ Eta“

 

 

 

 

 

 

 

 

 

 

 

 

time "p % Yeld°
(min) 8' (cis/trans)R (cis/trans) Et/Pr
1) MgCl2(8.0 equiv.). 0 87 875 89 774
1.2-DOE. 0 °C. 490 74 1 878 74 1576 100/95
Molarity I 0.02 M
RI-H; Fri-H
2) MgCl2(8.0 equiv.). 0 87 875 89 774
1,2-DCE. 45 °C. 120 35 38720 44 33715 93788
Molarity . 0.02 M,
R‘.H; #:H
3) MgCl2(8.0 equiv.). 0 94 472 95 372
Toluene, 25 °C, 180 70 976 75 1076 85791
Molarity - 0.02 M, 255 52 16710 54 16710 78780
R‘IH; #IH
4) MgCI2(10.0 equiv.). o 84 1076 83 1077
Toluene:CI-§CI2 (3:1 ). 490 50 22713 46 18712 85776
25 “C. Molarity I 0.02 M,
RI-H; 1?an
I5) MgCI2(10.0 equiv.). 0 82 1177 83 1077
Toluene:CI-I2CI2:Et20 60 65 1479 66 14710 88790
(251.005). 25 °C, 420 45 20713 46 18713 78777
Molarity I 0.02 M.
R‘IH; RZIH
I6) MgBr2 (2.0 equiv.). 0 92 573 92 573
ToluenezEt20 (>955). 420 62 1379 60 14710 84784
25 °C, Molarity - 0.02 M.
H‘sH; RZ-H
7) MgCI2(2.0 equiv.). 0 92 573 92 573
MgBr2 (2.0 equiv.). 90 78 1176 75 976 95790
Tdugng;thC|2 230 46 22,20 51 16/14 88/81
(15:19). 25 11c, 350 22 27729 31 23714 78768
Molarity - 0.02 M,
R‘IH; RQIH

 

‘(1)1888 8nd1788 . Mg, THF. 45 °C. (21'IszCI,.Tquene."Number8 represent millimoler r8808 scaled to 100 for

time-Omdperoentageforihefinel IIme.°Referen0869.

 

 

122
Using the ”magnesium free" alkyltitanocene solution, we were persistent in

our quest to optimize M9X2 promoted cyclizations. The solvent 1,2-
dichloroethane has been shown to be an effective solvent for olefin
polymerization. Entries 1 and 2 (Table 12) illustrate the combined effects of
MgCI2 (8 equiv.) in 1.2-dichloroethane at either 0 °C or 45 °C. Although no
cyclization was observed at 0 °C, promising results were achieved at higher
temperatures. yet the reaction ceased after two hours. Entries 3-7 represented
systematic variations in the reaction conditions in an attempt to reproduce the
results obtained in Table 11, in which the organometallic complex was not
isolated. During those particular cyclizations, toluene and CH2CI2 were present,
however, the exact amount of diethyl ether and the precise composition of M9X2
(X:Br or Cl) was not clear. Cyclization of the mixture of alkyltitanocene
complexes with MgCI2 (8 equiv.) in toluene (entry 3) or in a toluene:CH2Cl2
mixture (3:1, entry 4) produced the same overall conversion and comparable
yields. Since MgCI2 was virtually insoluble in the previous solvent systems, the
need for a small amount of diethyl ether seemed necessary to solubilize the
catalyst. To explore this avenue, entries 5 and 6 utilized varying amounts of
diethyl other but, to our dismay, no significant improvements were observed.
Slightly improved conversion was finally realized when a 1:1 mixture of
MgCl2zMgBr2 (entry 7) was used. However, the irreproducibility of the M9X2
catalysis prompted us to pursue the other Lewis acids.

We have found in previous investigations that Me2AlCI was an efficient
catalyst for cyclohexane formation. During the optimization of 1648. we observed
a linear conversion to 197 when Me2AlCI (1.0 equiv.) in toluene (0.01 M) was
reacted with 1648 at -35 °C. Since these conditions appeared to provide results
that fulfilled our initial criteria. we treated a mixture of the B-ethyl- and B-propyl-

titanocene complexes under these conditions (T able 13). In entry 1, a slight

 

123
variation in the previous conditions were used which resulted in complete

cyclization in under 30 minutes. Using the precise conditions as in 164b, only a
marginal improvement in time was realized. We determined that accurate
analysis of the reaction progress under more dilute conditions (<0.01 M) was not
reliable. Although an extensive study of lesser equivalents of Me2AlCI was not

completed, we decided to concentrate all efforts on MAO.

Table 13. Cyclization of Alkyltitanocene Chlorides Promoted with

 

 

 

Me2AlCIa
drdr 8120- d: 0
time "I :1 0792 98 Y'Ieldc
(min) 8' (cis/trans) (cis/trans) EI/Pr
1) Me2AlCI (1.5 equiv.). 0 83 1176 83 1176
Touene, -50 to -35 °C, 30 8 28758 7 29/55 94791
Molarity . 0.015 M
RI-H; RZ-D
2) Me2AlCI (1.0 equiv.). 0 75 18” 83 1077
Toluene, ~50 to -35 °C, 20 35 21735 52 19729 917100
mlarity . 0.01 M, 55 7 28753 12 32156 887100
RI-H; 8°.H

 

 

 

 

 

 

'(1)1608 8nd 1788 mm end175e ,Mg. THF. 45 °c, (2)1'le20l2, Toluene. I’lvurnbers represent millimolsr
fatioeoceledb 100 fortime-Oendperoentage forthefinal time. °Referlena69.

Tables 14 and 15 represent preliminary results using MAO. The
information contained in these tables was separated based on the order of
addition of the reagents. and the form (solid or solution) in which MAO was
handled. In Table 14, MAO was weighed under an inert atmosphere prior to each
reaction, diluted with toluene and cooled to -78 °C. followed by addition of a
solution of the alkyltitanocene complexes. In contrast. the results in Table 15
were obtained by the addition of a 0.5 M solution of MAO (in toluene) to a -78 °C
solution of the alkyltitanocene complexes. In general, the latter approach was
more reliable since the precise amount of catalyst was more easily measured (in

the former case, 10-15 mg of MAO were required for each reaction).

124
Table 14. Cyclization of Alkyltitanocene Chlorides Promoted with

 

 

 

 

 

MAO
l l a I I“ l I“ "
fiche; Tle-fii H H
GB + dFPr - dpEt GE: (I/PPr OH - 6
time 'R' "IR-2 % Yield
8‘ 9’ (min) RI (cis/trans) R2 (cisnrans) EUP'
1) MAO (1.5 equiv.). 25 .c 0 79 1279 79 1279
CH-plz. -78 to 25 °c, 360 70 1279 79 1279 917100
Molarity - 0.035 M 660 36 27718 40 29716 81785
R‘ID; R2.”
2) MAO (1.5 equiv.). 0 88 874 88 775
Toluene, :7810 25 °C, 100 11 28750 15 27745 89787
Molarity - 0.01 M,
R‘IH; #IH
3) MAO (0.75 equiv.). 0 94 472 95 372
Toluene, -78 to 25 °C. 25 5 26764 7 25759 95791
Molarity I 0.02 M.
R‘II‘I; FfIH
4) MAO (3.0 equiv.). 0 83 1176 83 1176
CH2CI2, -78 to -55 °C. 65 6 28754 7 29751 88787
Molarity - 0.02 M.
RI-H; 82.0

 

 

 

 

 

 

°(1)188e snd175e or1888 md1788 «1718 «M758, Mg. THF,45°C, (2)110th Toluene. °Numbers
moment millimolarratiosscaledb 100 fortime-Oandperoenugefortheflnal tine.°Referonoe 69.

Although the experimental technique used to obtain the results in Table 14
was suspect. qualitative information about the necessary catalyst concentration
and reaction temperature was available. In entry 1, no cyclization was observed
at -78 °C after 6 hours with MAO (1.5 equiv.). However. allowing the reaction
mixture to warm to ambient temperature overnight resulted in approximately 50%
conversion. More importantly, these conditions showed a preference for the cis
isomer (2:1) which paralleled the results for M9X2. If the reagents were added in
the same manner as entry 1. but warmed to ambient temperature immediately
(entries 2 and 3), cyclization was very rapid and the product mixture was trans
selective (2:1). Entry 4 showed that with higher MAO concentrations (3 equiv.).

cyclization could be promoted at -55 °C. and trans selectivity (64:36) was again

125
observed. The observed selectivity dependence on the temperature is not without

precedent. Erker has recently reported that the tacticity of propylene
polymerization was drastically reversed at different temperatures with MAO
(isotactic at -50 °C and syndiotactic at -10 °C).75 Apparently, based on the
stereochemical outcome for entries 2 and 3, cyclization must have been complete
before ambient temperature was reached. The stereochemical results in entry 1
have not been duplicated. During the initial six hours of sampling the reaction, a
large portion of the catalyst was probably hydrolyzed, which prevented significant
cyclization from occurring prior to ambient temperature being reached.
Exhaustive attempts to reproduce these results have utilized less MAO, more
dilute conditions, inverse addition of reagents. and mixing of reagents at ambient
temperatures. All of these attempts have been met without reward.

Addition of a stock solution of MAO to the alkyltitanocene complexes
provided only minor improvements for controlling the cyclization rate and the
reproducibility of any set of reaction conditions (Tables 15-17). The results in
Tables 15-17 demonstrated the necessary ratio of alkyltitanocene
complexeszMAO to be between 1.0:0.5 to 1.0:O.65. However, comparison of the
results in entries 2 and 3 (T able 15), and entries 4 and 5 (Table 15), demonstrated
that under identical conditions, a marked difference in cyclization rate (or no
cyclization at all, entry 3) was observed. These divergent results could be
attributed to the age of the alkyltitanocene complexes or to the technique with
which MAO was added. An apparent trend was that 12-24 hours after the
alkyltitanocene complexes were purified. a slight increase in the MAO
concentration was needed to initiate cyclization. This suggests that adventitious
moisture or oxygen may play a role in retarding cyclization. Another possible
scenario for the divergent results was that the requisite amount of MAO needed

for cyclization never reached the solution. In order to suppress rapid cyclization,

126
MAO solution was precooled by slow addition down the side of the flask

(approximately 0.25-0.50 inches above the solution) to a -78 °C solution of the
alkyltitanocene complexes, and rinsed by vigorous stirring. What may have been
occuring was that MAO was crystallizing on the side of the flask and failed to
redissolve.

Another level of frustration encountered during this study was an
uncontrollable, exponential conversion to cyclic products in the initial minutes after
MAO was added (see select entries in Tables 16 and 17). Moreover, during the
initial exponential rise, the substrate with the B-ethyl group with respect to titanium
cyclized to a far greater exent than the B-propyl substrate. Although we can
merely speculate as to the cause of the difference in rate in the initial time period,
we can say that these results were not isotope dependent. The net effect was to

diminish the useful window of observation.

127

Table 15. Competitive Cyclization of B-l-lydrogen vs B-Hydrogen (168b vs
175b) Promoted Under the Optimized Conditions tor MAO

 

 

 

 

 

 

 

 

 

 

I I a 1 M" l M“ b
nopzc1 TIszCl H f H d
a3 + a” time GB '47:: ELF” 07:; % Yieldc
R‘ 82 (min) 9' (cis/trans) "2 (cis/trans) EUPI'
1) MAO (0.5 equiv.). 0 97 271 96 272
Toluene, ~78 to 25 °C, 20 <1 33764 9 31758 100797
Molarity II 0.02 M
R‘ul-l; Rz-H
2) MAO (0.5 equiv.). 0 91 873 90 874
Toluene. ~78 to ~50 °C, 30 80 1179 84 977 100/100
Mohmy - 002 M, 150 55 1 9726 67 1371 8 10071 00
R1_H; Rag” 420 21 29750 48 20732 1007100
3) MAC (0.5 + 0.25 equiv.). 0 95 372 95 372
Toluene, ~78 to ~50 °C, 55 87 974 91 473 100798
Molarity . 0.02 M. 130
ELM; 112-H 025" 230 <1 33764 4 29759 98792
4) MAO (0.65 equiv.). 0 91 873 93 473
Toluene, ~78 to ~50 °C. 120 43 19734 59 15723 96/97
Molarity . 0,02 M, 240 5 32756 25 24744 93793
31.”; Re.” 300 3 32762 21 26750 97797
15) MAO (0.65 equiv.). 0 91 673 90 874
Toluene, ~78 to ~50 'C, 60 7 ? 48 20732 1007100
Molarity - 0.02 M. 120 <1 31789 7 28764 100799
R‘IH; R2."

 

‘ (1) 188. endt‘lSa . Mg, THF. 45 °C. (2)1sz02. Toluene. ° Numbers represent millimolar ratios scaled b 100
lor time-0 end percenme for the final time. °Relerence 69.

 

128

Table 16. Competitve Cyclization of B-Hydrogen vs B-Deuteron (168b
vs 178b) Promoted Under the M”Optimized Conditlobns tor MAO

 

 

 

 

 

 

 

 

 

a
@2092“ fifiszcl g/pw Ea: :Pr
9' time “E": R. w: 1%. Yield°
F" (min) 9' (cis/trans? (cis/trans) Et/Pr
1) MAO (0.65 equiv.). 0 96 371 96 371
Toluene. ~78 to -45 °C, 30 23 26751 45 20735 1007100
Molarity . 0.02 M 120 <1 33766 28 24748 1007100
R‘-H; nz-o
2) MAO (0.55 equiv.). 0 95 372 94 472
Toluene, ~78 to ~50 °C, 1 0 54 19727 70 12718 1 007100
Molarity . 0,02 M, 25 40 21739 59 15726 1007100
Ru”; 82.0 40 23 26751 48 19733 1007100
60 13 28757 23 29745 98797
3) MAO (0.50 equiv.). 0 93 473 93 473
Toluene. ~78 to ~55 °C, 10 61 15723 70 1271 8 997100
Molarity . 0.02 M. 40 25 25749 45 20/35 997100
an”, R24) 90 10 29780 21 27751 99799
4) MAO (0.4 equiv.). 0 93 473 93 473
Toluene, ~78 to 25 °C, 10 84 878 84 878 1007100
Mommy _ 0.02 M. 25 -c 105 53 25722 80 10710 1007100
an”; 92.0 145 4 38758 19 31747 100797

 

“(1) 1680 andt'ne . Mg. THF. 45 “C. (2"GC20l2. Toluene. E’Numbere represent millimolar ratios scaled to 100 f0
time-0 and percentage for the final time. °Fleierenoe 69.

 

129

Table 17. Competitive Cyclization of B—Deuteron vs B-i-iydrogen

(171b vs 175b) and B,B~Dideuteron vs ma-Dlhydrogen (184b
vs 175b) Promoted Under the Mt,Optlmlzed Conditions for MAO

 

 

 

 

 

 

 

 

time ":1 % Yieldc
(min) 8' E(cis7trans) Pr(ms/trans.) Et/Pr
1) MAC (0.50 + 0.05 equiv.). 0 97 271 97 271
Toluene. ~78 to ~55 “C. 20 81 8711 85 679 1007100
Molarity - 0.02 M 95 73 10717 80 7713 100/100
R‘.0; rte-11:18.11, H 120 64 13722 75 8716 99798
0.05 OWN: 140 47 20738 49 1 7732 99798
1 80 4 30/65 9 30761 99/100
2) MAO (0.55 equiv. ), 0 92 573 90 674
Toluene. ~78 to ~55 °C, 10 45 22733 60 16724 1007100
mlarity .. 0,02 M, 25 30 26743 47 20732 99799
R1-H; R2...“ 59.0. D 55 21 28749 35 23740 98/98
70 13 29755 20 27750 97797
3) MAO (0.55 equiv.). 0 d 7/6 92 573
Toluene. ~78 to ~55 °C. 10 d 16721 62 13720 d795
Molarity _ 0.02 M, 25 (I 23731 60 14722 (”96
an“; R’.H;a’~0. 0 55 0 33744 37 21738 d796

 

 

 

'(1)17ta and1750 «184a and175a . Mg. THF. 45 °C, (2TiCp20l2.Toiuene.°Numbersrepreeentmiliimolar
retioesceledthOlor fine-Omdperoentageiorthefinaltime.‘neference69.‘CoUdnotbe
dehmimdaoanblymebpoubanimmsohfimhmmofmeoudwmawicm.

130
Figures 4-7 illustrate a graphic view of the relative rate data for various

competitive experiments. If the initial growth period is ignored and only the linear
portion considered for each graph, the 01- or B- deuterium atoms had no
measurable effect on the rate. However, with the minimum number of data points
obtained in each linear portion of the graph, the error in these slopes was
considerable. in spite of these shortcomings, an amazing correlation was
observed. For the competitive cyclization of 168b and 175b (Figure 4), a relative
rate of k1sablk175b=122 was determined, which demonstrated the contribution of
the alkyl group to the rate of insertion. Figure 5 suggests that B-hydrogen
activation does not enter into the mechanism under the conditions that employ
MAO, as the relative rate was comparable to Figure 4, k153uk17ab=1.24. The
best graph of the relative rate data for the control experiment, Figure 6, was not
conclusive, as the erroneous graph indicates. Finally, the test for a-hydrogen
activation is illustrated in Figure 7 and the relative rate calculated to be

k13457k175b=1.30. These results indicate that a-hydrogen activation is likewise,

not important to the mechanism for this system.

131

 

lOO

—'_ 168b
—._ 1750

% Conversion

 

 

 

 

I ' I f
200 300 400
Time (min)

V

' r
O 100

Figure 4. Competitive Cyclization Between 168b and 175b (see Table 15, entry 4). Slopes were
calculated using a staple line program and omitting the exponential portions of the
graph. The relative rate was calculated to be: krel=k168b7k175b=122~

 

100

168!)
178!)

ll

% Conversion

 

 

 

80

fime (min)

Figure 5. Competitive Cyclization Between 168b and 178b (see Table 16, entry 2). Slopes were
calculated using a sinple line program and omitting the exponential portions of the
graph. The relative rate was calculated to be: krel-4‘168b7kt78b-1~24-

 

132

 

lOO

 

 

 

 

 

60 -1

C

.0

g 40 7 —9— 171b
.\° 20 -

o - , -
0 '00 0.05 mmol MAO 20°
added at time-125

 

Time (min)

Figure 6. Competitive Cyclization Between 171b and 175b (see Table 17, entry 1). Slopes were
calculated using a simple line program and omitting the exponential portions of the
graph. Relative rate calculated to be: krel=k171b7k175b-1-60-

 

lOC

 

 

 

 

’C
7% -—9— 184b
32
. . r r . . .
0 20 4o 60 80
Time (min)

Figure 7. Competitive Cyclization Between 184b and 175b (see Table 17, entry 2). Slopes were
calculated using a slrmle line program and omitting the exponential portions of the
graph. The relative rate was calculated to be: kreI-k184b7k175b-1-30.

 

133
Athough the above results suggest that a- and B-hydrogen activation do

not participate in the mechanism, the limited data points for each graph do not
provide confidence for any firm conclusion. in order to more accurately assess
the situation, we have recently prepared 6-butyl-7-br0mo-1~heptene (182) and
plan to compare it to 175. Preliminary results demonstrated that 182b underwent
cyclization more slowly than the B-propyl substrate (175b) in the initial time after
MAO was added. A trend has been clearly established in which the longer B-alkyl
chains gave rise to decreased cyclization in the early stages of the reaction. This

new substrate should allow for the rapid completion of this study.

4. Conclusion

Although these results were not conclusive as to the existence of 01- or [3-
hydrogen activation, conditions have been worked out using MAO that should
allow this project to be completed soon. Although, preliminary results discussed
here indicate that olefin insertion promoted by MAO did not show an a- or (3-
hydrogen effect. A number of interesting stereochemical observations were made
based on the Lewis acid and the temperature of the reaction. it was shown that
both MAO and Mng were selective for the cis isomer when cyclization was
carried out at 25 °C or higher. Yet, when lower temperatures were employed with
MAO. the trans isomer was preferentially obtained. The future of this mechanistic
project will be directed towards the investigation of the magnesium catalysts in an
attempt to understand the nature of stereochemistry obtained. For now, the

secrets of Ziegler-Natta polymerization are still secure.

1 34
5. Experimental

Methaluminoxane (MAO). A Schlenk flask was loaded with
Al2(SO4)3(H20)(14-13) followed by the removal of oxygen under vacuum and
purging with argon. The aluminum sulfate was suspended in toluene (80 mL)
and the flask immersed in a 25 °C water bath. MeaAl (20 mL, 209 mmol) was
diluted with toluene (20 mL) and the solution was added rapidly, via cannula,
under argon pressure to the aluminum sulfate suspension. after which the
reaction mixture was heated to 40 °C for 15—20 hours. The MAO was purified by
filtration under argon using a Schenk filter. To accomplish this purification, the
slurry was transferred to the filter via cannula. The solvent was removed in
vacuo and the MAO was stored under nitrogen. For the reactions in Table 14,
MAO (MW=130 g/mol) was weighed under nitrogen prior to each reaction, and
diluted with the appropriate solvent. For the all other reactions employing MAO.
a 0.5 M MAO solution (toluene) was used (3.25 g MAO. 25 mmol, 50 mL
toluene).

Magnesium leromlde. Magnesium turnings (1.46 g, 60 mmol) were
loaded in a 3~necked flask equipped with a reflux condenser and a dropping
funnel, and diethyl ether (30 mL) was added. An diethyl other solution of 1,2-
dibromoethane (11.27 g. 60 mmol, 10 mL diethyl ether) was added slowly to the
magnesium metal over 2 hours followed by heating to reflux for an additional 2
hours. Diethyl ether was removed under reduced pressure and the white
precipitate was heated 150 °C under vacuum (<0.001 mm Hg) for 8-10 hours.
The white solid was stored under nitrogen.

General Procedure for Metal Mediated Cyclizations Using Unpurlfled
Alkyltitanocene Chlorides. A Schlenk flask was charged with magnesium (583

mg, 24 mmol), flame dried under vacuum (<0.001 mm Hg), purged with argon

135

prior to the addition of diethyl ether (4 mL). Alkyl bromide 175 (438 mg, 2 mmol)
was slowly added to the magnesium over 6 hours using a gas tight syringe. After
the addition was complete, the reaction mixture was allowed to stir for an
additional 4 hours. In a separate flask was loaded Tiszclz (548 mg. 2.2 mmol),
and the flask was subsequently evacuated under reduced pressure. and purged
with argon. The TICp2Cl2 was suspended in toluene (8 mL) and the flask cooled
to ~50 °C. The alkylmagnesium bromide was transferred to the Tiszclz
suspension via cannula and rinsed with additional diethyl ether (1 mL). The
reaction temperature was maintained at ~50 °C for 30 minutes and then warmed
to ambient temperature for an additional 5 hours.

(A) Magnesium Dihallde Promoted Cyclization Uslng In slfu M9X2.
After transmetallation was complete, the reaction mixture was brought to 45 °C
(Table 11; entry 1). Cyclization was promoted using the Mng (X: Cl or Br)
which was formed in situ by the reaction of the Grignard reagent and Tiszclz.

(B) Magnesium Dihallde Promoted Cyclization Uslng Excess ngz or
MAC. After transmetallation was complete, the diethyl ether was removed under
vacuum and the reaction mixture was diluted with CH20I2 (5 mL). In a separate
flask was loaded either MgCI2 (189 mg, 1.98 mmol, Table 11: entries 2 and 3) or
MAO (86 mg, 0.66 mmol, Table 11; entry 4) under and atmosphere of nitrogen
and diluted with toluene (8 mL). A solution of 175b (2.0 mL, 0.33 mmol) was
added to a ~45 °C solution (or suspension) of the respective Lewis acids. via
syringe, and that temperature maintained for 10 minutes followed by warming to
the desired temperature (see Table 11).

General Procedure for Metal Promoted Cyclizations Uslng Purified
(M9X2 Free Solutions) Alkyltitanocene Chlorides. A Schlenk flask was
charged with magnesium (780 mg, 32 mmol), flame dried under vacuum, and

purged with argon. THF (4 mL) was added to the flask and heated to 45 °C. in a

136

typical competitive rate experiment. a mixture of alkyl bromides (1 mmol each)
was simultaneously added to the magnesium over 6 hours using a gas tight
syringe and after the addition was complete, the reaction temperature was
maintained for an additional 4 hours. In a separate Schlenk flask was loaded
Tiszclz (548 mg, 2.2 mmol) and the flask was evacuated and purged with
argon. The TiCp2CI2 was suspended in toluene (8 mL) and cooled to ~78 °C.
The Grignard reagents were transferred via cannula under argon pressure to the
cold suspension of Tiszclz and rinsed with additional THF (2 mL). The
transmetallation reaction was maintained at ~78 °C for 8 hours and then allowed
to warm to ambient temperature for an additional 90 minutes, at which time the
reaction mixture was diluted with hexane (25 mL) and cooled to 0 °C. The
combined action of the hexane and cold temperatures caused most of the M9X2
and excess TlezClz salts to fall out of solution. The cold reaction mixture was
transferred via cannula under argon pressure to a Schlenk filter and sequentially
washed with cold hexane (3 x 7 mL). The alkyltitanocene chlorides were
evaporated to dryness under vacuum (<0.001 mm Hg) to leave behind a dark red
oil (or slurry). Toluene (30 mL) and an internal standard were added to the
alkyltitanocene chlorides and the concentration of each substrate determined by
gas chromatographic techniques. All subsequent reactions utilized this stock
solution of known molarity.

(A) Competitive Bate Cyclizations Promoted by ngz or Crystalline
MAO. MgCI2, MgBrz (or both. see Table 12; entry 3) or MAO were loaded in a
Schlenk flask under nitrogen, suspended or dissolved in the solvent employed,
and cooled to low temperature (see Table 12 and 13 for more details on each
reaction). A calculated volume of the alkyltitanocene chlorides was added over
one minute to the Lewis acid, maintained at that temperature for 10 minutes, and

finally warmed to the appropriate temperature.

137

(B) Competitive Rate Cyclizations Promoted by Me2AlCI or MAC.
Solutions of Me2AlCI and MAO were prepared by dissolving the Lewis acids in an
appropriate volume of toluene to prepare a 0.5 M solution of each. A calculated
volume of the alkyltitanocene chloride solution was diluted to the appropriate
concentration (see Table 14, 15, 16, or 17) and cooled to either ~50 °C if MegAlCl
was to be used, or ~78 °C if MAO was to be used. The appropriate Lewis acid
was taken up in a gas tight syringe and slowly added down the side of the flask
about 0.5 inches above the solution, and the flask was vigourously swirled to
rinse the sides of the flask. The initial temperature employed was maintained for
10 minutes and then warmed to the described temperature (see Table 14. 15, 16,

or 17 for more details).

 

CONCLUSION

The development of a novel ring-forming method which featured an
intramolecular monomeric analog of Ziegler-Nana polymerization was discussed.
The ease with which unactivated olefins reacted with the carbon-titanium bond
was demonstrated. The metal and its surrounding ligands played a significant
role during the carbocyclization, controlling both the regio~ and the
stereochemistry of bond formation for both 5~hexen~1~yl~ and 6~hepten~1~
yltitanocene chlorides. Exo insertion was the sole pathway observed in both
cases. When an alkyl group was located on the alkyl tether. excellent and
predictable diastereomeric excesses were achieved during carbon-carbon bond
formation. in most cases. This ring-forming method represents the first procedure
that did not result in a loss of selectivity or efficiency when extended to six~
membered ring synthesis. This general and versatile procedure will certainly
enhance the repertoire of carbon-carbon bond forming reactions available to the

synthetic chemist.

138

 

LIST OF REFERENCES

LIST OF REFERENCES

(a) Rigollier, P.; Young, J. R.; Fowley, L. A.; Stille, J. R. J. Am. Chem. Soc.
1990, 112, 9441. (D) Young. J. R.; Stille. J.R. Organometallics 1990. 9,
3022. (c) Young, J. FL; Stille, J. R. J. Am. Chem. Soc. 1992, 114, 4936.

For an excellent summary of modern organometallic chemistry, see:
Collman, J. P.; Hegedus, L. 5.; Norton, J. Ft; Finke. Ft. 6.; ln Principles and
Applications of Organotransition Metal Chemistry: University Science Books:
California, 1987.

For reviews of Ziegler-Natta polymerization, see: (a) Boor, J., Jr. In Ziegler-
Natta Catalysts and Polymerization; Academic Press: New York, 1979. (b)
Pino, P.; Mulhaupt, R. Angew. Chem, Int. Ed. Eng/.1980, 19, 857. (c)
Kaminsky. W.; Sinn. H. Transition Metals and Organometallics as Catalysts
for Olefin Polymerization; Springer-Verlag: Berlin, Heidelberg, 1988, 361.
Ziegler, K. Angew. Chem. 1952, 64, 323.

Natta, 6.; Pine, P.; Corradin, P.; Danusso, F.; Mantica, E.; Mazzanti, G.;
Moraglio. G. J. Am. Chem. Soc. 1955, 77, 1708.

Waters. J. A.; Mortimer, G. A. J. Organomet. Chem. 1970, 22, 417.

(a) Breslow, D. 5.; Newburg, N. R. J. Am. Chem. Soc.1959. 81, 81. (b)
Ziegler, K.; Gellert, H. G.; Zosel, K.; Holzkamp, E.; Schneider, J.; Soil, M.;
Kroll, W. R. Ann. Chem. 1960, 121, 629.

(a) Marvel. C. S.; Stille, J. K. J. Am. Chem. Soc. 1958. 80. 1740. (b)
Waymouth, B. M.; Resconi, L. J. Am. Chem. Soc. 1990, 112, 4953.

139

10.

11.

12.

13.

14.

15.

140
For cyclization of 5~hexen~1~yl metal systems using titanium (a) and

zirconium (0) catalysts, see: (a) Clawson, L.; Soto, J.: Buchwald. S. L.;
Steigerwald, M. L.; Grubbs, R. H. J. Am. Chem. Soc.1985, 107, 3377. (b)
Carr, D. B.; Schwartz, J. J. Am. Chem. Soc. 1979, 101, 3521. For related
examples using scandium (c) and yttrium (d). see: (c) Piers, W. E.; Bercaw, J.
E. J. Am. Chem. Soc. 1990, 112, 9406. (d) Molander, G. A.; Hoberg, J. O. J.
Am. Chem. Soc. 1992, 114, 3123.

For an excellent review in the area of cationic cyclization see: Johnson, W.
S. Biorg. Chem. 1976, 5, 51.

(a) Burke, S. D.; Takeuchi, K.; Murtiashaw, C. W.; Liang, D. W. M.
Tetrahedron Lett.1989, 30. 6299. (b) Burke. S. D.; Strickland, S. M. 5.;
Organ, H. M.; Silks, L. A. III Tetrahedron Lett. 1989, 30. 6303. (c) Burke, 8.
D.; Deaton, D. N. Tetrahedron Lett.1991, 32, 4651. (d) Burke. 8. D.;
Shankaran. K.; Helber, M. J. Tetrahedron Left. 1989. 30, 4655.

For a review on the use of chiral acetals in organic synthesis, see: Alexakis,
A.; Mangeney, P. Tetrahedron 1990, 46. 477.

For a comprehensive perspective on the development of free-radical
processes, see: Giese, B. in Radicals in Organic Synthesis: Formation of
Carbon-Carbon Bonds, Volume 5; Oxford: 1986.

(a) Schwartz. 0. E.; Curran, D. P. J. Am. Chem. Soc. 1990. 112, 9272. For
other examples of tandem cyclizations leading to linear and propellane
triquinane natural products, respectively, see: (b) Curran, D. P.; Fiakiewicz,
D. M. J. Am. Chem. Soc. 1985. 107, 1448. (c) Curran, D. P.; Jasperse. C. P.
J. Am. Chem. Soc. 1990, 112, 5601.

(a) Fraser-Reid, B.; Pak, H.; Dickson, J. K., Jr. J. Org. Chem. 1989, 54, 5357.
For other examples of tandem cyclizations using carbohydrate templates,

see: (b) Fraser-Field, B.; Dickson, J. K., Jr.; Tsang, H.; Llera, M. J. Org. Chem.

16.

17.
18.

19.

20.

21.

22.
23.
24.
25.
26.

27.

141
1989, 54, 5350. (c) Fraser-Reid, B.; Pak, H.; Canalda. I. l. J. Org. Chem.

1990. 55. 3009.

For examples of free radical cyclization of 7-bromo-1-heptene compounds,
see the experimental in Chapter 3.

Beckwith. A. L. J.; Schiesser, C. H. Tetrahedron 1985. 41, 3925.
Chatgilialoglu, C.; lngold, K. U.; Scaiano. J. C. J. Am. Chem. 8001981.
103, 7739.

Beckwith. A. L. J.; Easton, C. J.; Lawrence, T.; Serelis, A. K. Aust. J. Chem.
1 983. 36, 545. For comparison of these literature results. see the
experimental in Chapter 2.

For a theoretical discussion on the regio- and stereochemical details for 5-
exo-trig radical ring closures. see reference 17 and: Spellmeyer, D. 0.;
Houk, K. N. J. Org. Chem. 1987, 52, 959.

Kharasch, M. S.; Skell, P. 8.; Fisher, P. J. J. Am. Chem. 800.1948, 70.
1055.

Curran, D. P.; Chang, C. J. Org. Chem. 1989, 54, 3140.

Curran, D. P.; Chen, M. J. Am. Chem. Soc. 1987, 109. 6558.

Curran, D. P.; Seong, C. M.. J. Am. Chem. Soc. 1990, 112, 9401.

Boger, D. L.; Mathvink, R. J. J. Am. Chem. Soc.1990, 112, 4003.

(a) Porter, N. A.; Chang, V. H. T.; Magnin, D. H.; Wright, B.T. J. Am. Chem.
Soc. 1988, 110, 3554, and references therein. (b) Boger, D. L.; Mathvink, R.
J. J. Am. Chem. Soc. 1990, 112, 4008, see reference 24 also.

(a) Curran, D. P.; Shen, W.; Zhang, J.; Heffner, T.A. J. Am. Chem. 800.1990,
112, 6738. (b) Porter, N. A.; Swann, E.; Nally, J.; McPhail, A. T. J. Am. Chem.
Soc. 1990, 112, 6741. (c) Giese, B.; Zehnder. M.; Roth, M.; Zeitz, H. J. Am.
Chem. Soc. 1990, 112, 6741.

28.

29.

30.

31.

32.
33.
34.
35.

36.

37.

38.

39.

40.

142
Bailey. W. F.; Punzalan, E. R. J. Org. Chem. 1990. 55. 5404, and references

therein.

(a) Ashby, E. 0.; Pham, T. N. J. Org. Chem. 1987, 52, 1291. (b) Ashby. E. C.;
Pham, T. N.; Park. B. Tetrahedron Lett.1985. 26, 4691, and references
therein.

(a) Bailey. W. F.; Normi, T. T.; Patricia, J. J.; Wang, W. J. Am. Chem. Soc.
1987, 109. 2442. (b) Broka. C. A.; Lee, W. J.; Shen, T. J. J. Org. Chem.
1988, 53, 1336. (c) Chamberlin, A. J.; Bloom, S. H.; Cervin, L. A.; Fotsch, C.
H. J. Am. Chem. Soc. 1988, 110, 4788. (d) Koppang, M. 0.; Ross. G. A.;
Woolsey. N. F.; Bartak, D. E. J. Am. Chem. Soc.1986, 108, 1441. (e)
Paquette, L. A.; Gilday, J. P.; Maynard, G. D. J. Org. Chem. 1989. 54, 5044.
Bailey, W. F.; Khanolkar, A. D.; Gavaskar, K.; Ovaska, T. V.; Rossi, K.; Thiel,
Y.; Wiberg. K. B. J. Am. Chem. Soc. 1991, 113, 5720.

Bailey, W. F.; Khanolkar, A. D. J. Org. Chem. 1990. 55, 6058.

Oliver, J. P.; Smart. J. B.; Emerson, M. T. J. Am. Chem. Soc. 1966, 88, 4101.
Bailey, W. F.; Khanolkar, A. D. Tetrahedron Left. 1990, 31, 5993.

Beckwith. A. L. J.; Phillipou, G. Serelis, A. K. Tetrahedron Lett. 1981, 22,
2811.

Bailey, W. F.; Rossi, K. J. Am. Chem. 800.1989, 111, 765.

Ashby. E. C.; Oswald, J. J. Org. Chem. 1988. 53. 6068.

Kossa, W. 0.. Jr.; Richey, H. 6., Jr.; Rees, T. C. Tetrahedron Left. 1971,
3455.

Utimoto. K.; Imi, K.; Shiragami, H.; Fujikura. S.; Nozaki, H. Tetrahedron Lett.
1985, 26, 2101.

Stefani, A. Helv. Chim. Acta. 1 974, 57, 1346.

41.

42.

43.

44.
45.

46.

47.

48.

49.

50.

143
(a) Carpenter, N. E.; Kucera, D. J.; Overman, L. E. J. Org. Chem. 1989, 54,

5846. (b) Ableman, M. M.; Overman, L. E. J. Am. Chem. Soc. 1988, 110,
2328.

(a) Roll, W.; Brintzinger, H.; Rieger, B.; Zolk, R. Angew. Chem, Int. Ed. Engl.
1990, 29. 279. (b) Mullin, D. T.; Rausch. M. 0.; Lin, Y.; Dong, S.; Chien, J. C.
W. J. Am. Chem. Soc. 1990, 112, 2030. (c) Soga, K.; Park, J. R.; Uchino, H.;
Uozumi, T.; Shiono, T. Macromolecules 1989, 22, 3824. (d) Ewen, J. A. J.
Am. Chem. Soc. 1984, 106, 6355.

(a) Ewen, J. A.; Jones, R. L.; Razqui, A. J. Am. Chem. Soc. 1988, 110. 6255.
(b) lshihara, N.; Kuramoto, M.; Uoi, M. Macromolecules 1988. 21, 3356.
Cossee. P. Tetrahedron Lett. 1960. 17, 12.

lvin, K. J.; Rooney, J. J.; Stewart. C. D.; Green, M. L. H.; Mahtab, R. J. Chem.
Soc., Chem. Commun. 1978. 604.

Chatt. J.; Duncanson, L. A. J. Chem. Soc. 1953, 2939.

For a molecular orbital description of bis(cyclopentadienyl)~MLn complexes
and their chemistry. see: Hoffman, R.; Lauher, J. W. J. Am. Chem. Soc.
1976, 98. 1729.

For a description of the propagation and termination steps for the
polymerization of ethylene with a scandium complex, see: Burger, B. J.;
Thompson, M. E.; Cotter, W. D.; Bercaw, J. E. J. Am. Chem. Soc.1990, 112,
1556.

Dawoodi, 2.; Green, M. L. H.; Mtetwa, V. S. B.; Prout, K. J. Chem. Soc,
Chem. Commun. 1982, 1410. For a theoretical study of this structure, see:
Koga. N.; Obara, S.; Morokuma, K. J. Am. Chem. Soc. 1984, 106. 4625.

(a) Jordan, R. F.; Bradley, P. K.; Baenziger, N. C.; LaPointe, R. E. J. Am.
Chem. 800.1990, 112, 1289. (b) Dawoodi, 2.; Green, M. L. H.; Mtetwa, V. S.
B.; Prout, K. J. Chem. 800., Chem. Commun. 1982, 802.

51.
52.
53.
54.

55.
56.
57.
58.
59.

60.

61.

62.

63.
64.

65.

144
Schmidt, G. F. ; Brookhart, M. J. Am. Chem. Soc. 1985, 107, 1443.

Bose. A. K.; Lal. B. Tetrahedron Lett. 1973. 3937.

Hartman, G. D.; Halczenko. W.; Phillips, B. J. Org. Chem 1986, 51, 147.
The following reactions were mutually unsuccessful: allylic coupling of 131,
an alkyl halide, and Ni(COD)2, and Grignard or cuprate addition of 131 to
ethylacrylate. For the use of Ni(COD)2 in organic synthesis, see: (a)
Hegedus, L. S.; Varaprath, S. Organometallics 1982, 259. For the addition
of Grignards or cuprates to acrylate derivatives, see: (b) Liu, S. H. J. Org.
Chem. 1977, 42, 3209. Finally, it was found that allyl metal reagents of Mg
or Cu could be successfully reacted with 131, see (c) Corey, E. J.; Posner,
G. H. J. Am. Chem. Soc. 1967, 89. 3911.

Org. Synthesis, Coll. Vol. 4, 1 963. 749.

Brown, H. C.; Mandal, A. K.; Kulkarni J. Org. Chem. 1977. 42, 1392.

Brown. H. C.; Prasad, J. V. N. V. J. Org. Chem. 1985. 50. 3002.

Chaudhary, S. K.; Hernandez, O. Tetrahedron Left. 1979, 99.

Howden, M. E. H.; Maercker, A.; Burdon, J.; Roberts, J. D. J. Am. Chem. Soc.
1986. 88. 1732.

Gadwood, R. C.; Lett. R. M.; Wissinger, J. E. J. Am. Chem. Soc. 1986, 108.
6343.

Brown, H. C.; Garg, C. F.; Liu. K. J. Org. Chem. 1971, 36, 387.

Whitesell, J. K.; Wang. P. K. S.; Aguilar, D. A. J. Org. Chem. 1983, 48. 2511.
Boland, W.; Ney, P.; Jaenicke, L. Synthesis 1980, 1015.

The linear sequence from and including 179-182 was kindly. proficiently,
and expediently performed by Nancy S. Barta. The proper gratitude cannot
be adequately expressed for her efforts when l was in need.

lnorg. Syn. 1963, 7, 9.

66.

67.

68.

69.

70.

71.

72.

73.

145
Danishefsky, S.; Vaughan, K.; Gadwood, R.; Tsuzuki, K. J. Am. Chem. Soc.

1981, 103, 4136.

Greenwald, R.; Chaykovsky, M.; Corey, E. J. J. Org. Chem. 1963, 28, 1128.
Sauvage, J. F.; Baker, R. H.; Hussey. A. S. J. Am. Chem. Soc. 1960, 82.
6090.

All yields reported here for free radical or titanium-mediated cyclizations
were obtained by gas chromatographic techniques compared to an internal
standard and corrected for detector response.

(a) Beesley, R. M.; lngold, C. K.; Thorpe. J. F. J. Chem. 800.1915, 107,
1080. (b) lngold, C. K. J. Chem. Soc. 1921, 113, 305. (c) Hammond, G. S. In
Steric Effects in Organic Chemistry; Newman, M. 8., Ed.; John Wiley and
Sons: New York. 1956; p. 463.

Rieke, R. D.; Li, P. T. J.; Burns, T. P.; Uhm, S.T. J. Org. Chem. 1981. 46.
4323.

Upon close examination of the mass spectrum fragmentation pattern of the
compound reported in reference 1a, we determined that it was not 133, but
rather a structural isomer. The actual compound in reference 1a was 4~(3~
bromopropyl)~1~cyclohexene.

(a) Pino, P. Adv. Polym. Sci. 1965, 4, 393. (b) Pino, P.; Ciardelli. F.;
Zandomeneghi, M. Ann. Rev. Phys. Chem. 1970, 21. 561. (c) lfissin, Y. V. In
lsospecitic Polymerization of Olefins with Heterogeneous Ziegler-Natta
Catalysts; Springer-Verlag: New York, 1985; p. 295-306. (d) Ciardelli. F.;
Carlini, C.; Altomore, A.; Menconi, F.; Chien, J. C. W. In Transition Metal
Catalyzed Polymerizations; Quirk, R. P., Ed.; Cambridge. University Press:
Cambridge, 1988; p. 25. (e) Vizzini, J.; Ciardelli. F.; Chien, J. C. W.
Macromolecules 1 992, 25, 108.

146
74. The term a-hydrogen activation is somewhat misleading and controversial.

Grubbs used this term when referring to the Green-Rooney mechanism (see
reference 9a). Bercaw chose the term ”or-agostic assistance” rather than "a-
hydrogen activation" (see reference 9c). The use of ”hydrogen-activation” is
often used to denote breaking of the carbon-hydrogen bond. When
hydrogen-activation was used in this thesis. only agostic stabilization and
acceleration of olefin insertion was implied.

75. Erker, G.; Fritze, C. Angew. Chem, Int. Ed. Engl. 1992, 31, 199.