.‘1

1:.

A”.

~
a2.}... ,
.. 1t
..

§ I
. Vw s
fibHMuVm

17:... Q

it; =3
$1":
2““.

 

lb“;

This is to certify that the
dissertation entitled

CATALYTIC ASYMMETRIC AZA-DlELS-ALDER REACTION:

REGULATION OF ORTHOGONAL FUNCTIONS IN A DUAL
CHIRAL/NON-CHIRAL CATALYST SYSTEM

presented by

CORY ALLAN NEWMAN

has been accepted towards fulfillment
of the requirements for the

Ph.D. degree in Chemistry

 

 

WbI/JLM/

Major Professor’s Signature
5 /Q/ 6 7

Date

 

MSU is an affirmative-action, equal-opportunity employer

 

LIBRARY
Michigan State
University

 

_~_o---o-o--¢-.-c—n-o—-o-.-u-o-o-o-u-c—n-n—o—.-—.—-—-—-—.¢-.--n-.—

n-.-a-n.¢-- -—a-------.--c--.-.- —- -

PLACE IN RETURN BOX to remove this checkout from your record.
To AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DAIEDUE

DATEDUE

DATEDUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

6/07 p:/CIRClDateDue.indd-p.1

CATALYTIC ASYMMETRIC AZA-DIELS—ALDER REACTION:
REGULATION OF ORTHOGONAL FUNCTIONS IN A DUAL
CHIRAL/NON-CHIRAL CATALYST SYSTEM

By

Cory Allan Newman

A DISSERTATION

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY
Department of Chemistry

' 2007

ABSTRACT
CATALYTIC ASYMMETRIC AZA-DIELS-ALDER REACTION:
REGULATION OF ORTHOGONAL FUNCTIONS IN A DUAL
CHIRAL/NON-CHIRAL CATALYST SYSTEM

By
Cory Allan Newman

The subject of this thesis is the catalytic asymmetric aza—DieIs-Alder reaction and
a new method using a dual catalyst system. The optimization studies as well as the
generality of the reaction will be discussed in detail. In addition, NMR titrations were
done to obtain binding constants to gain insight into the role of the non-chiral Lewis acid
B(OPh)3. Lastly, efforts toward the total synthesis of cylindricine C will be discussed
wherethe aza—DieIs-Alder was used as the key step. '

A catalytic asymmetric aza—Diels-Alder reaction has been developed using the
catalyst derived from the chiral C2 symmetric ligand VAPOL and B(OPh)3.
Optimization studies found that the catalyst prepared from 5 mol% VAPOL and 100
mol% B(OPh)3 was optimal. This method was found to be widely general for imines
generated from aromatic and secondary aliphatic aldehydes, however, attempts to expand
the scope to primary aliphatic substrates failed. Several other surrogate imines for the
primary aliphatic substrates were attempted and success was limited with these substrates
as well. A second class of substrate that was studied were imines containing a substituted

benzhydryl group on the nitrogen of the imine. The TMB, DAM, and BUDAM imines

were studied and with few exceptions, the substitution on the benzhydryl group proved to
be detrimental to the outcome of the reaction

During optimization ’studies, a unique observation was made with the non-chiral
Lewis acid B(OPh)3, where it was found that the asymmetric induction of the product
remained constant (90% ee) as the amount of B(OPh)3 was increased and the ee did not
diminish "until the ratio of B(OPh)3 to VAPOL was 100:1 (82% ee). To determine the
mechanistic role of B(OPh)3, an NMR titrationstudy was conducted to determine the
affinity for binding of the VAPOL-B(AOPh)3 catalyst and B(OPh)3 to both the imine 150
and the Diels-Alder cycloadduct 151. It was found that the excess triphenylborate was
competing for binding to the product of the reaction, resulting in chiral catalyst turnover,
and the rate of the achiral background reaction was determined to be about 9 times slower
than chiral reaction.

Also discussed are efforts towards the asymmetric total synthesis of cylindricine
C using an aza-Diels-Alder reaction with the bis—TMS diene 278 to install its core decalin
structure. Unfortunately, all attempts to afford the Diels-Alder adduct with this diene
using a zirconium catalyst with a variety of chiral ligands failed. A racemic synthesis
was then carried out using BF3'OEt2 as the catalyst and an optimized yield of 69% was
achieved. A 1,4—addition of a butenyl side-chain was proposed as the second step in the
' synthesis but all attempts to install that group failed. Another route was proposed where
the diene 227 containing the butenyl side chain could be employed. One attempt was
made with this diene andonly the Mannich type product 296 was observed. The details

o

of these studies as well as a discussion of the results will be the focus of this thesis.

 

T 0:
George L. Newman

(my grandfather)

iv

 

ACKNOWLEDGEMENTS

First and foremost I would like to thank professor William D. Wulff, my research
advisor. I thank him for allowing me to join his research group and giving me the
freedom to take any idea I had and explore it in the lab no matter how foolish or brilliant
it might have sounded. It has been a pleasure not only learning from himand interacting
on a professional level but also in a less formal setting as well. We have also shared
many great conversations over a bottle (or two) of wine and he has given me an
appreciation for the taste of wine, which I did not have before joining his group.

I would now like to send my greatest appreciation to a few people, without whom
I would not have ventured to graduate school. The first person to thank is Anne Wilson,
my undergraduate research advisor. I first was exposed to organic chemistry in 1998 in
the general organic lectures taught by Anne Wilson. She was a great professor and if I
had been taught by someone else, I might not have enjoyed the experience, and would not
have chosen organic as my major field of study in, graduate school. I also had the
opportunity to do research for Anne in my senior year, which was my first exposure to
academic research in organic chemistry. She again made this experience a great one and
without her advice, I probably would not have decided to apply to graduate school.

I owe it also to my parents (Tim and Diane) as well as my grandfather for helping
me decide to go to graduate school. I had many discussions with them and I have distinct

memories of them telling me “just go to graduate school...you have nothing holding you

back.” So, without them, I may have just received my 8.8. degree and took a job in
industry, which would have been much less fulfilling than my experience here at
Michigan State University.

I would like to thank my committee members (Rob Maleczka, Babak Borhan, and
Mitch Smith) for their support as well. I appreciate them for teaching me in my second
year oral how little I knew at that point in my research career. That experience, as tough
as it was, means a lot in my development as a chemist and a research scientist.

I owe the secretaries in the department an acknowledgement as well. Lisa
Dillingham (graduate secretary) was always helpful when I needed to take care of the
formalities or just little things that come up during the graduate school process. If I
needed something mailed or printed out in a timely fashion, Deanne Pierce (organic
secretary) would always take care of what I needed immediately. Nancy Lavrik was a
great joy to work with during the semesters when I had to teach and she would always be
there tohelp when it was needed. These are just a few secretaries that I have worked
closely with at times, but experiences with all other secretaries, as they came up have
been very pleasurable as well.

I want to thank Dan Holmes and Kermit Johnson for all their help with NMR
when I needed it. They were always willing to stop what they were doing and help no
matter what the problem was.

Dr. Lee Fielding, our collaborator, helped me to analyze NMR titration data and I
thank him for all the time he spent helping me. i

I also want to thank DanieI Whitehead for reading my entire thesis for

proofreading purposes, this was a great help to me. In addition to Dan, Stewart Hart also

vi

 

proofread a chapter for me as well. I thank them for that and for also being great friends
during my time here.

During my time here, I have met many people and developed many friendships.
In my second year I met Janelle Sec] who became my best friend and ultimately my wife.
I owe her a great deal of gratitude for her patience when dealing with all that graduate
. school throws at you. Her support throughout and the motivation she has given me
definitely played some role in my success here (the $80 bottle of port for getting my
cume points did not hurt), There have been some stressful times throughout graduate
school and being able to go home to someone as special as Janelle has gotten me through
it. I cannot say enough about her but I will stop here otherwise these acknowledgements
will be about 100 pages.

Now I will just mention some names of other friends and express my thanks to
them: Yiqian Lian, my lab mate for 3 years is awesome and he is a great colleague and
friend. It was great to share the good times and hardships in lab together. Hopefully one
of these years, I will be able to fill out the NCAA basketball pools better than him. I
have spent my entire graduate career in the same lab as Kieth Korthals and he too has
shared in the frustrations that occur in graduate school. Dr. Patwardan, I owe thanks to
for teaching me all the basics of organic research and for being patient when I first started
in the lab. Glenn Phillips was a great deal of help when things were not going well in the
lab and he was kind enough to share some good stories to make me feel better about
myself when I did stupid things in the lab. Vijaygopal Gopalsamuthirim was also a lab
I mate of mine for a few years and he was very knowledgeable and I took advantage of this

many times. Zhengzheng Ding, how can I say enough about this guy. First, I will say he

vii

 

is a little “dorky” (inside joke) but he is witty and is-always ready with a good joke. I
only worked in the same lab as him for about a year, but it was a fantastic year hopefully
our friendship will last a long time. Many good conversations have been had with Alex
Predus, Kostas Rabalakos, Aman Desai, and all the other Wulff group members (past and
present) and I thank all of them for that and I wish them continued successin their
careers.

I will end my acknowledgements by thanking again my parents and my wife who
have been my lifeline for the past several years. They have been great supporters, great

friends, and my stress relief when needed. Thank you all very very much.

viii

 

TABLE OF CONTENTS

LIST OF SCHEMES ............................................................................... xii
LIST OF FIGURES ................................................................................ xix
LIST OF TABLES ................................................................................. xxi
' KEY TO SYMBOLS AND ABREVIATIONS. . . . . . . . . . .................................... xxii
CHAPTER ONE: INTRODUCTION .............................................................. 1
1.1 DIEIS-ALDER REACTION ........................................................................ 1
1.2 HErEROATOM-DIELs-ALDER REACTION ...................................................... 9
1.3 PREVIOUS EXAMPLES LEWIS ACID/BRQNSTED ACID CATALYZEDAZA-DIELS-ALDER
REACTION ......................................................................................... 11
1.3.1 Stoichiometric Catalyst .................................................................. 12
1.3.2 Sub-Stoichiometric Catalyst ............................................................ 18
1.4 VAPOL AND VANOL LIGANDS .............................................................. 37
CHAPTER TWO. OPTIMIZATION OF THE AZA- DIELS- ALDER REACTION ....... 43
2.1 THE TBS VERSION OF DANISHEFSKY’ s DIENE .............................................. 46
2. 2 DANISHEFSKY’ s DIENE ......................................................................... 52
2.2.1 VAPOL vs. VANOL vs. BINOL ................... 52
2.2.2 Temperature Effects .......... ' ........................................................... 55
2.2.3 Solvent Effects ..................... 60
2.2.4 Different Lewis Acid Sources ......................................................... 65 -
2.2.5 Catalyst Loading: B(0Ph)3NAPOL Ratios .......................................... 69
2.2.6'Danishefsky’s Diene: Quality, Equivalents, and Addition Time
and Temperature ........................................................................ 81
2. 2. 7 Optimal Conditions ..................................................................... 93
2.3Appendix......., ................................................................................. 96
CHAPTER THREE: SUBSTRATE SCOPE .................................................... 97
3.1 AROMATIC SUBSTRATES.............. .......................................................... 98
3.2 a, B-UNSATURATED SUBSTRATES ........................................................... 107
3.3 TERTIARY ALIPHATIC SUBSTRATES ......................................................... 111
3.4 SECONDARY ALIPHATIC SUBSTRATES ................................... f ................... 112
ix

 

3.5 PRIMARY ALIPHATIC SUBSTRATES .......................................................... 115

3.6 MODIFIABLE SUBSTRATES .................................................................... 120
3.6.1 a-Alkoxy Substrates .................................................................. 121
3.6.2 Glyoxylate Ester Substrates .......................................................... 122
3.6.3 Silyl-Acetylene Substrate ............................................................. 125
3.6.4 a-Silyl-a,B-Unsaturated Substrate .................................................. 126

3.7 BENZI-IYDRYL DERIVATIVES .................................................................. 128
3.7.1 T etramethyl Benzhydryl (T MB) Substrates ........................................ 131
3.7.2 DAM Substrates .................................................... . ................... 133
3.7.3 BUDAM Substrates ................................................................... 134

CHAPTER FOUR: EFFECTS OF TRIPHENY LBORATE ................................. 139

4.1 EXPLORATION AND EXPLANATION OF THE VAPOL-B(OPh)3 CATALYST
SYSTEM ......................................................................................... 142

4.2 BINDING CONSTANT DATA ANALYSIS ...................................................... 150

CHAPTER 5: EFFORTS TOWARD THE SYNTHESIS OF CYLINDRICINE C ...... 155

5.1 CYLINDRICINE C HISTORY .................................................................... 155

5.2 PREVIOUS SYNTHESES OF CYLINDRICINE C ................................................ 157

. 5.3 RETROSYNTHETIC ANALYSIS ................................................................. 165

5.4 Efforts Toward Cylindricine C .............................................................. 168
5.4.1 Catalytic Asymmetric aza-Diels-Alder Reaction .................................. 168
5.4.2 Efforts Toward the Synthesis of (i)-Cylindricine C .............................. 173

5.4.2.1 Optimization of the aza-Diels-Alder Reaction .......................... 173
5.4.2.2 1,4-Addition to Incorporate the Butenyl Side Chain ................... 176
5.4.2.3 N-N Bond Cleavage and Protection with Cbz .......................... 178
5.4.2.4 Protection of the Amine with Cbz ........................................ 182
5.4.2.5 1,4-Addition of the Butenyl Side Chain Using the Cbz
Protected Vinylogous Amide ................................................ 183
5.4.3 Synthesis of Cylindricine C via Route B ........................................... 185
CHAPTER 6: CATALYST STRUCTURE, CONCLUSIONS AND FUTURE
WORK............... .............................................................. 188
6.1 STRUCTURE OF THE (S)-VAPOL-BORON CATALYST ...................................... 188
6.1.1 aza-Diels-Alder Reaction Using Alternate Catalyst Preparations ............... 192
6.2 CONCLUSIONS AND FUTURE WORK .......................................................... 194
6.2.1 Aza-Diels-Alder Optimization ............................... , ......................... 194
6.2.2 Substrate Screening ..................................................................... 196
6.2.2.1 Other Possible Substrates to Screen ............................ ....l99
6.2.2.2 Exploration of Other Boron Catalysts ....................................... 203
6.2.3 Triphenylborate Effects ............................................................... 207 4
6.2.4 Progress Towards Cylindricine C .................................................... 208

CHAPTER 7 EXPERIMENTAL SECTION ................................................... 211
7.1 EXPERIMENTAL PROCEDURES AND CHARACTERIZATIONS DATA FOR

CHAPTERTwo ................................................................................. 211
7.2 EXPERIMENTAL PROCEDURES AND CHARACTERIZATIONS DATA FOR

CHAPTER THREE ............................................................................... 221
7.3 EXPERIMENTAL PROCEDURES AND CHARACTERIZATIONS DATA FOR

CHAPTER FOUR ................................................................................ 281
7.4 EXPERIMENTAL PROCEDURES AND CHARACTERIZATIONS DATA FOR

CHAPTER FIVE ................................................................................. 295
7.5 EXPERIMENTAL PROCEDURES AND CHARACTERIZATIONS DATA FOR

CHAPTER SIx ................................................................................... 345

..

xi

LIST OF SCHEMES

Scheme 1.1 Diels—Alder [4+2] Cycloaddition .................................................... 1
Scheme 1.2 Stereospecificity of the Diels-Alder Reaction ...................................... 4
Scheme 1.3 Possible Regiochemical Outcomes of the Diels—Alder Reaction ................ 4
Scheme 1.4 Effect of EWG’s and EDG’s on Regiochemistry .................................. 5
Scheme 1.5 s—cis versus s-trans ..................................................................... 6
Scheme 1.6 Examples of the Diene Locked s-cis ................................................ 6
Scheme 1.7 Possible Ways to Induce Stereochemistry in the Diels-Alder
Reaction ................................................................................. 8
Scheme 1.8 Heteroatom Diels-Alder Reaction .................................................. 10
Scheme 1.9 Regioselectivity in the Heteroatom Diels-Alder Reaction ...................... 11
Scheme 1.10 Pioneering Work by Yamamoto ................................................... 12
Scheme 1.11 Substrate Scope for B(OAr)3/BINOL Catalyst ................................. 13
Scheme 1.12 Screening Different Imine Protecting Groups for the
B(OAr)3/BINOL Catalyst ......................................................... 14
Scheme 1.13 Furman’s use of Yamamoto’s Catalyst .......................................... 15
Scheme 1.14 Yamamoto’s Chiral Br¢nsted Acid Catalyst ......... ' ........................... 16
Scheme 1.15 aza-Diels-Alder Reaction Using the Br¢nsted Acid Catalyst ................. 16
Scheme 1.16 Use of Stoichiometric Catalyst Derived from EIQZII and BINOL ............ 17
Scheme 1.17 Catalyst Derived from Zr(OtBu)4 and 6,6’-dibromoBINOL .................. 18

xii

Scheme 1.18 Optimal Conditions Using the Naphthyl Derived Imine ....................... 18
Scheme 1.19 Screening of SIIbstrates with the Zr/Br-BINOL Catalyst ...................... 19
Scheme 1.20 Switched Enantioselectivity with Different BINOL Derivative ....... . ...... 21
Scheme 1.21 Isomerization of Aliphatic [mines to the Enamine .............................. 23
Scheme 1.22 aza-Diels-Alder Reaction of Hydrazine Imines ................................. 24
Scheme 1.23 aza-Diels-Alder Reaction Using. a Silver Catalyst .............................. 25
Scheme 1.24 Use of Silver Catalyst with No Solvent or Undistilled THF in Air ......... 27
Scheme 1.25 aza-Diels-Alder Reaction Using Silver Catalyst with a
- Recoverable Ligand ................................................................. 28
Scheme 1.26 Copper Phosphino Sulfenyl Ferrocene Catalyst ................................. 28
Scheme 1.27 Substrate Scope Using Copper Phosphino Sulfenyl Ferrocene
Catalyst .............................................................................. 29
Scheme 1.28 Ligands Screened by Jorgensen ................................................... 30
Scheme 1.29 Screening of Substrates using the Cu/BINAP Catalyst ........................ 31
Scheme 1.30 Whiting’s Combinatorial Approach .............................................. 32
Scheme 1.31 Whiting’s Substoicheometric Approach ......................................... 34
Scheme 1.32 aza-Diels-Alder using Catalytic Chiral Bransted Acid ........................ 35
Scheme 1.33 Substrate Scope using the Chiral Phosphoric Brpnsted Acid. . . ..' ............ 36
Scheme 1.34 Imino-Aldol Reaction ............................................................... 38
Scheme 1.35 Diels-Alder Reaction and Baeyer Villager ReactiOn ........................... 39

Scheme 1.36Aziridination Reaction ............................................................. 40

Scheme 1.37 Comparison of VAPOL and VANOL in the Aziridination

Reaction .............................................................................. 41
Scheme 2.1 aza-Diels-Alder in the Prescence of 4A Molecular Sieves. . . . . . . . .............48

Scheme 2.2 azq-Diels—Alder Reaction Without 4A Molecular Sieves ....................... 49

xiii

 

Scheme 2.3 Effect Diene Equivalents and Addition Time in the aza—Diels-

Alder Reaction ........................................................................ 50
Scheme 2.4 First Attempt Using Danishefsky’s Diene ........................................ 52
Scheme 2.5 Comparison of VANOL, VAPOL, and BINOL. . . . . . . . . . . . . . .................. 53
Scheme 2.6 Effects of the Purity of VAPOL .................................................... 54
scheme 2.7 Temperature Effect Using VAPOL ................. 56
Scheme 2.8 VANOL Temperature Study ........................................................ 58
Scheme 2.9 Toluene Versus CHzCl2 as Solvent ............................................... :60
Scheme 2.10 CCl4 and THF as the Solvent ...................................................... 61
Scheme 2.11 Effects of Solvent Combinations ................................... _ ............... 63
Scheme 2.12 Different Lewis Acid Sources ..................................................... 66
Scheme 2.13 Reactions Using B(OPh)3 .......................................................... 70
Scheme 2.14 Reactions Using B(OPh)3/VAPOL (3: 1) ............................... ' ..... ....71
Scheme 2.15 B(OPh)3 to VAPOL Ratio Effects ................................................ 73
Scheme 2.16 VAPOL Loading with 150 mol% lB(OPh)3 ...................................... 77
Scheme 2.17 VAPOL Loading with 100 mol% B(OPh)3 ...................................... 78
Scheme 2.18 Reactions Using 100 mol% VAPOL ............................................. 80
Scheme 2.19 Source and Equivalents of Diene ................................................. 82
Scheme 2.20 Danishefsky’s Diene Addition Times ............................................ 84 '
Scheme 2.21 Temperature of the Addition of Danishefsky’s Diene ......................... 86

I Scheme 2.22 In Situ Preparaton of the Imine ................................................... 91

Scheme 2.23 Heteroatom Diels-Alder Reactcion of Benzaldehyde .......................... 92
Scheme 2.24 Optimal Conditions ................................................................. 93
Scheme 2.25 Jon Antilla’s Results ........... 96

xiv

 

Scheme 3.1 Optimal Conditions for the aza-Diels-Alder ...................................... 97

Scheme 3.2 aza-Diels-Alder Reaction of 4—methoxybenzaldimine (154) ................... 99
Scheme 3.3 aza-Diels-Alder Reaction of 4—bromobenzaldimine (156) .................... 100
Scheme 3.4 aza-Diels-Alder Reaction of 4~nitrobenzaldimine (158) ...................... 101
Scheme 3.5 aza-Diels—Alder Reaction of 2-methylbenzaldimine (160) ................... 103
Scheme 3.6 aza-Diels-Alder Reaction of l-naphthaldimine (162) .......................... 104
Scheme 3.7 Preparation of 4—fluoro-2—methylbenzaldehyde (168) .......................... 105 '
Scheme 3.8 aza-Diels-Alder Reaction of 4~fluoro-2-methylbenzaldimine (169).........106
Scheme 3.9 Reduction of 4—fluoro-2-methylphenyl—Diels-Alder Adduct (164) ........... 106
Scheme 3.10 aza-Diels-Alder Reaction of Phthalaldimine (171) ............................ 107
Scheme 3.11 aza-Diels—Alder Reaction of trans-cinnamaldimine (173) ................... 108
Scheme 3.12 aza—Diels—Alder Reaction of 3-methylcrotanaldimine (175) ................ 109
Scheme 3.13 aza-Diels-Alder Reaction of Cyclohexenecarboxaldimine (177) ........... 110
Scheme 3.14 aza-Diels—Alder Reaction of tert—butyl—aldimine (179) ...................... 111
Scheme 3.15 aza-Diels—Alder Reaction of Cyclohexane Carboxaldimine (181) ......... 112
Scheme 3.16 aza-Diels-Alder Reaction of Isopropylaldimine (183) ........................ 114
Scheme 3.17-aza-Diels-Alder Reaction of n—heptylaldimine (185) ......................... 115

Scheme 3.18 aza-Diels-Alder Reaction of n-heptylaldimine (185) (more attempts). . ...117

Scheme 3.19 aza-Diels-Alder Reaction of n-heptylaldimine (185) (final attempts). ..I 18

Scheme 3.20 aza-Diels-Alder Reaction of n-propylaldimine (189) ........................ 119
Scheme 3.21 aza-Diels-Alder Reaction of a—benzyloxyaldimine (191) .................. 121
Scheme 3.22 Preparation of 2-tert-butyl-diphenylsilyloxyethanal (196) .................. 122

Scheme 3.23 aza~Diels~Alder Reaction of a-tert-butyl~diphenylsilyloxyaldimine
(197) .................................................................................. 122

XV

Scheme 3.24 aza-Diels-Alder Reaction of Ethyl glyoxaldimine (199) ..................... 123
Scheme 3.25 aza-Diels-Alder Reaction of Isopropylglyoxaldimine (201) ................ 12S
Scheme 3.26 Preparation of 3-triisopropylsilyl-2-propynal (205) .......................... 125
Scheme 3.27 aza-DielS-Alder Reaction of Triisopropylsilyl- acetyleneylaldimine

(206) ................................................................................. 126
Scheme 3.28 Preparation of Z-2—trimethylsilyl-2-ocetnal (211) ............................ 127
Scheme 3.29 aza-Diels-Alder Reaction of Z-a-trimethylsilyl-

heptenylaldimine (212) ........................................................... 128
Scheme 3.30 Relative Rates and Selectivity of the Azirdination Reactions of

Substituted Benzhydryl Protected Benzaldimines ............................ 130
Scheme 3.31 aza-Diels-Alder Reaction of TMB—benzaldimine (215) ..................... 131

Scheme 3.32 aza-Diels-Alder Reaction of TMB-4—bromobenzaldimine (217). . ..........132
Scheme 3.33 aza-Diels-Alder Reaction of TMB-cyclohexane carboxaldimine (219)...133
Scheme 3.34 aza-Diels-Alder Reaction of DAM-protected Imines (2213-d) ............. 134
Scheme 3.35 aza-Diels-Alder Reaction of BUDAM-benzaldimine (223) ....... _ .......... 135
Scheme 3.36 aza—Diels-Alder Reaction of BUDAM—protected Aldimines (22Sa-c)..... 136
Scheme 4.1 Turnover Induced by Triphenylborate ............................. ' .............. 141

Scheme 4.2 Comparison of Bn and Bh Using the BINOL-boron Catalyst Prepared
using Yamamoto’s conditions ..................................................... 143

Scheme 4.3 Comparison of Bn and Bh Using the VAPOL-Boron or BINOL-

Boron Catalyst Prepared Using my Optimal Conditions. ..................... 144
Scheme 4.4 Reactions with B(OPh)3 ............................................................ 146
Scheme 4.5 Possible Mechanism involving B(OPh)3 ......................................... 147
Scheme 5.1 Molander Total Synthesis of (—)—Cylindricine C. ._ ............................ 158
Scheme 5.2 Trost Total Synthesis of (+)—Cylindricine C .................................... 158
Scheme 5.3 Kibayashi’s First Total Synthesis of (+)-Cycindricine C ...................... 159

xvi

Scheme 5.4 Kibayashi’s Second Total Synthesis of (+)-Cycindricine C .................. 160

Scheme 5.5 Ciufolini’s Total Synthesis of (—)—Cylindricine C ............................. 161
Scheme 5.6 Hsung’s Total Synthesis of (+)-Cylindricine C from D~Pyroglutamic

Acid ................................................................................... 162

Scheme 5.7 Hsung’s Total Synthesis of (—)—Cylindricine C from L—Serine ....... I ....... 163

Scheme 5.8 Shibasaki’s Total Synthesis of (+)—Cylindricine C ............................ 1.164

Scheme 5.9 Aza-Diels-Alder Reaction of Hydrazine Imines ................................ 168

Scheme 5.10 Screening of Ligands for the aza-Diels-Alder Reaction

Using Kobayashi’s Conditions .................................................. 170

Scheme 5.11 Reaction using (S)-VAPOL/B(OPh)3 as the Catalyst. ........................ 171
Scheme 5.12 Catalytic Asymmetric aza-DielS—Alder Reactions using the bis- 1

TMS Diene 278 ..................................................................... 172

- Scheme 5.13 Optimization of the Racemic aza—Piels-Alder Reaction ..................... 174

Scheme 5.14 cis 1,4—Addition as Reported by Martin ........................................ 176

Scheme 5.15 1,4—Addition to the Vinylogous Hydrazine Using Cuprate 3 ................ 177

Scheme 5.16 Reductive Cleavage Using SrnI2 ................................................ 178

Scheme 5.17 Initial Attempts for the Reductive Cleavage Using Mg° and HgClz... .....179

Scheme 5.18 Determination of the Optimal Amount of HgCl2 ............................ ' ..181
Scheme 5.19 Optimal Reduction Condition and Scalability................................,182
Scheme 5.20 Cbz Protection of the Amine .............................. ‘ ....................... 183
Scheme 5.21 Screening of Cuprates 1,2 and 3 .................................................. 184
Scheme 5.22 Lewis Acid Promoted 1,4—Addition....................................... ...... 185 7
Scheme 5.23 Aza—Diels-Alder Attempt Using Alternate Diene 227 ........................ 186

Scheme 6.1 aza- -Diels- Alder Reaction Using the One (B1) and Two Boron (B2)
Catalyst .............................................................................. 193

xvii

Scheme 6.2 Optimal Reaction Conditions ...................................................... 194 a
Scheme 6.3 Kunz’s Reaction Using Danishefsky’s Diene ................................... 200
Scheme 6.4 Results Using BLAH-Catalyst .................................................... 205
Scheme 6.5 Progress Towards Cylindricine C ........................ — ........................ 208
Scheme 6.6 Alternate Diene aza-Diels-Alder Attempt ....................................... 209

xviii

 

LIST OF FIGURES

Figure 1.1 Molecular Orbital Diagram of a Normal [4+2] Cycloaddition .................... 2
Figure 1.2 Molecular Orbital Diagram of the Inverse Electron Demand [4+2]
Cycloaddition ............................................................................ 3
Figure 1.3 Activated Dienes ............................................................ . ............ 7
Figure 1.4 pKa of sp2 and Sp3 Oxygen and Nitrogen ........................................... 11
Figure 1.5 Other Zirconium/BINOL Catalysts .................................................. 22
Figure 1.6 0-2 Symmetric Biaryl Ligands ....................................................... 38
Figure 2.1 Danishefsky’s dienes .................................................................. 47
Figure 2.2 Summary of Temperature Effects ................................................... 59
Figure 2.3 Summary of B(OPh)3 Loading Effects .............................................. 75
Figure 2.4 Glassware with Cooling Addition Coil ............................................. 88
Figure 2.5 Reaction Setup ..................... I ..................................................... 89
Figure 3.1 Tachykinin Antatonist and Key lnterrnediate ..................................... 105
Figure 3.2 Isomerization of Primary Alkyl Imines to Enamines ............................ 116

Figure 3.3 Original Propsed VANOL-Boron Catalyst/Benzaldimine
Interaction............................................ ..................... _. ......... _..129

Figure 3.4 Tetramethylbenzhydrylbenzaldimine (T MB-benzaldimine) (215). . 129
Figure 4.1 Summary of B(OPh)3 Effects ................. _ ......... _. ............................. 139

Figure 5.1 Cylindricines A-K .................................................................... 155

xix

Figure 5.2130merization of Cylindricine A to B .............................................. 156

Figure 5.3 Fasicularin and Lepadiforrnine ...................................................... 156
Figure 5.4 Flat representation of Lepadiformine, Fasicularin, and

Cylindricine C ......................................................................... 157
Figure 5.5 Retrosynthetic Analysis .............................................................. 166
Figure 5.6 Isomerization of Primary Alkyl Imines to Enamines ............................ 167
Figure 5.7 BINOL Derivatives. .L ................................................................ 169
Figure 5.8 Danishefsky’s Diene Versus bis-TMS Diene ..................................... 173
Figure 5.9 Cuprates 1-3 ............................................................................ 183
Figure 5.10 Alternate Substrates for the 1,4—Addition ........................................ 187
Figure 6.1 Original Proposed Catalyst .......................................................... 189
Figure 6.2 Two Boron Catalyst Structure ....................................................... 189
Figure 6.3 Boroxazine (H‘) Catalyst ............................................................ 190
Figure 6.4 Crystal Structure of the Three Boron Catalyst. and MEDAM Imine ........... .191
Figure 6.5 Chiral Imines .......................................................................... 199
Figure 6.6 Sugar Derived Imine .................................................................. 200
Figure 6.7 Electron Donating and Withdrawing Bh Derivatives...........................‘.201
Figure 6.8 Ketal or Thioketal Imines ........................................................ ....202
Figure 6.9 Dienes to be Screened ............................................................... 202
Figure 6.10 Access to B-Amino Acids .......................................................... 203
Figure 6.11 B1 and B3 Catalysts ................................................................ 204
Figure 6.12 BLAH Catalyst ...................................................................... 204
Figure 6.13 VANOL-BLAH Catalyst ........................................................... 207

Images in this thesis are presented in color

XX

 

LIST OF TABLES

Table 1.1 Best Results from Combinatorial Study ............................................. 33

Table 4.1 Binding Constants for Complexes of the Imine and Product with
B(OPh)3 and the VAPOL-Boron Catalyst .......................................... 149

Table 4.2 Relative Rates of the Two Catalysts and ee Prediction ........................... 153

Table 6.1 Binding Constants for Complexes of the Imine and Product with B(OPh)3
and the VAPOL-Boron Catalyst .................................................... 207

xxi

Bh
BINAP
BINOL
Bn

Boc
BUDAM
Cbz
CSA
DAM
DBU
sen
DIB
DIBAL
DME
DMF
DMSO
EDA
BBC

66

KEY TO SYMBOLS AND ABREVIATIONS

benzhydryl
2,2'—bis(diphenylphosphino)-1,1'-bilnaphthyl
1,1’—Dinaphthalene-2,2’-diol

benzyl

tert—butoxycarbonyl
bis-(3,5—ditert-butyl—p-anisyl)methylamine
benzyloxycarbonyl

camphor sulfonic acid
p-dianisylmethylamine
1,8-diazabicyclo[5.4.0]undec-7-ene
dichlorOmethane

(diacetoxyiodo)benzene ‘

diisopropyl aluminum hydride
1,2-dimethoxy ethane
N,N—dimethylformamide
dimethylsulfoxide

ethyl diazoacetate

electron donating group

enantiomeric excess

xxii

EWG

HOMO -

KHMDS
LUMO
M.S.

‘ MeCN

MEDAM

NCS
NMI

N MO
TBAF
TBAT
TBDME
TBDPS
TBME
TBS
TFA
THF
TIPS
TLC .
TMB

TMS

electron withdrawing group

highest occupied molecular orbital
potassium bis(trimethylsilyl)amide
lowest unoccupied molecular orbital
molecular Sieves

acetonitrile
bis-(3,5-dimethyl-p-anisyl)methylamine
Methanesulfonyl

N-chlorosuccinimide
N—methyli-midazole

N—methyl morpholine
tetrabutylammonium floride
tetrabutylammonium triphenyldifluorosilicate
tert—butyl methyl ether
tert-butyldiphenylsilyl

tert-butyl methyl ether
tert-butyldimethylsilyl

trifluoroacetic acid

tetrahydrofuran

triisopropylsilyl

thin layer chromatography
bis(3,5-dimethylphenyl)methylamine -

trimethylsilyl

xxiii

TON turn over number

TPAP I tetrapropylammonium chloride
VAN OL 3,3 ’—Diphenyl-2,2’-binaphthalene-1,1’-diol
VAPOL 2,2’—Diphenyl-[3,3 ’—biphenanthrene]—4,4’-diol

xxiv

CHAPTER 1
INTRODUCTION

The main focus of this dissertation is the development of new methodology for
the heteroatOm Diels-Alder reaction of imino-dienophiles with Danishefsky’s diene.
Before getting into the details of this research, several items need to be explained and
defined before the methodology can be understood. Discussed herein are the basics of
the Diels-Alder reaction, incorporation of a heteroatom, and how the use of a chrial
Lewis acid or Bransted acid allows access to enantioselectivity. In addition, to
understanding how this reaction adds to or contributes to the heteroatom-Diels-Alder
reaction field, the methods already present in the literature will be described.
1.1 Diels-Alder Reaction
The Diels-Alder reaction is arguably the most useful organic transformation ever
discovered in organic chemistry (Scheme 1.1). When the phrase “Diels-Alder” is typed
Scheme 1.1 Diels-Alder [4+2] Cycloaddition
/ .
E + l -—> O
l 2 3

into the Scifindersearch engine, the search reveals over 29000 items Containing the
concept Diels-Alder. The reaction was discovered in 19281 by the German chemists Otto
Diels and Kurt Alder. There are many things about this reaction that make it so useful,

and all have contributed to its popularity. First, the reaction forms a six-membered ring,

1

which is the most common cyclic structure in nature. In addition, one or more six- '
membered rings make up the core structure of vast numbers of these important and
naturally occurring compounds. The reaction has another desirable characteristic that is
very important in synthesis in that it is 100% atom economic. Every atom of the two
reactants are incorporated into the product. For example, when butadiene is added to
ethylene the reaction gives cyclohexene (C4H,5 + C2H4 -) CgHw), thus all of the six
carbons and 10 hydrogens in the starting materials are present in the product. The
mechanism of the Diels-Alder reaction occurs with rare exception as a concerted process
especially when there are'no heteroatoms involved in the reaction. The mechanism2 is a
41te' + 23te' process (Figure 1.1) where normally the HOMO of the diene reacts with the
Figure 1.1 Molecular Orbital Diagram of a Normal [4+2] Cycloaddition

4n: + 2313 Cycloaddition

LUMO of the dienophile.

Other examples are known where an electron donating

substituent on the dienophile raises the energy of the HOMO high enough to interact with

the LUMO of the diene. This is called an inverse electron demand Diels-Alder reaction

(Figure 1.2). The concerted nature of the reaction allows for the ability to set the relative

Figure 1.2 Molecular Orbital Diagram of the Inverse Electron Demand 4+2

Cycloaddition

Inverse electron demand 4n + 2n Cycloaddition

LUMO ,

>3???

1
l
i

stereochemistry of 4 out of the 6 carbons of the Six-membered ring (Scheme 1.2).

Scheme 1.2 Stereospecificity of the Diels-Alder Reaction

R3

 

One issue that can present a problem for this reaction is the aspect of
regiochemistry. When the diene and dienophile are both unsymmetrical, then mixtures of
two regioisomers can be produced (Scheme 1.3). This issue can be resolved, however by

Scheme 1.3 Possible Regiochemical Outcomes of the Diels-Alder Reaction

R1 1‘21
R2 R2 '
+i ——-» U +
R2
10 11 12

placing an electron donating group (EDG) on the diene and an electron withdrawing

l

R
/

w \
9

group (EWG) on the dienophile or visa versa (Scheme 1.4). In the case where the EDG

Scheme 1.4 Effect of EWG’s and EDG’s on Regiochemistry

EDG

6+
/ + P“ ——» Um Um
\ 6+

 

6—
13 14
major minor
OR
/6" EWG
153-rib +
EDG EDG EDG EWG
major minor
OR
EWG EWG EWG
6.-
/+ $610...
\6+ 6-
20 21
major minor
OR
f iG’G MUGS MU,
EWG 5+ " DG
major minor

is on the diene (l3 and 17) and the EWG is on the dienophile (14), the normal Diels-
Alder reaction occurs. However, when thevEWG is on the diene (20 and 24) and the
EDG is on the dienophile (21), the inverse electron demand situation can be realized.

In order for the Diels-Alder reaction to occur, there has to be sufficient orbital

overlap (as seen above), but what was not mentioned previously concerns the orientation

 

of the two double‘bonds with respect to the rotation about the single bond of the diene.
For the orbital overlap to be favorable (Scheme 1.5) the double-bonds have to be pointing

Scheme 1.5 S-cis Versus s-trans

(I) + H —-—-—> No Reaction

S—trans

©©+u——————».fii

S-cis
in the same direction (S-cis as opposed to S-trans). If they are not, the reaction will not

occur. The reaction rates can thus be greatly increased by using dienes with the

conformation of the diene locked in the s-cis conformation as in cyclopentadiene (27) or

1,3-cyclohexadiene (29) (Scheme 1.6).

Scheme 1.6 Examples of the Diene Locked s-cis

©+H——»2b

27 2 28
0+ H ——»2$
29 2 30

Another way to increase the rate of the Diels-Alder reaction is to make the diene
or the dienophile very electron rich or poor. During the history of the Diels-Alder
reaction, several electron rich dienes have been developed to increase reactivity, which in

turn increased the regioselectivity (Figure 1.3). In 19743 Samuel Danishefsky published

Figure 1.3 Activated Dienes

MeO / /N / MeO /
OSiMeg OSiMe3 TMSO OMe

31 32 ‘ 33
Danishefsky's diene Rawal's Diene Brassard's diene

an article where he employed the trans-4—methoxy-2—trimethylsilyloxybutadiene (31) as a
useful diene in the Diels—Alder reaction. This diene was found to be much more reactive
due to its very electron rich nature. This enhanced electron density also aids in the
regiochemical outcome of reactions using this diene with dienophiles containing EWG’S.
Thirty-four years later, Rawal, et. al. published an article using an amino version of
Danishefksy’s diene" 5 (32). This diene was found to be about twenty-five times more
reactive4' 5 than Danishefsky’s diene. Brassard alsp developed an electron rich diene6
(33). The diene developed in his lab allowed access to B-methoxy cyclohexenone
derivatives. The increased electron density of this diene also gives increased reactivity as
well as regioselectivity compared to other dienes.

It is well known that many naturally occurring compounds exist as a single
enantiomer. In addition, pharmaceutical agents have different mechanisms of action
when one enantiomer is used versus another. AS mentioned before, the Diels-Alder
reaction is attractive due to its ability to set four stereocenters in a single transformation.
The diastereoselectivity of this reaction is very good, but due to the issues just mentioned,
the ability to obtain a single enantiomer would also be desirable. For these reasons, the _
development of methods using the Diels-Alder reaction in an enantioselective manner has

been actively pursued.

There are several ways to achieve asymmetric, induction for the Diels-Alder
reaction (Scheme 1.7). One is the use of substrate control, as in the case where a

Scheme 1.7 Possible Ways. to Induce Stereochemistry in the Diels-Alder Reaction

R1 R3 R1 R3
/ + I|J\ _, . Substrate Control
\ .
R2 35 R2
34 36

 

R1 O R1 O
/ (IL Re Re
\ + I ' - Chiral Auxiliary
R2 37 R2
34 R* = Chiral Auxiliary 33
R1 o R1 O
l .
/ H Chiral Catalyst H _
\ + I > Chiral Catalyst
R2 39 R2 1
34 40

stereocenter in either the diene ordienophile relays the stereochemistry into the product.
Another option would be to use a chiral auxiliary that could be incorporated in either
reagent and ultimately removed in a later step. Lastly, an external chiral catalyst that
could coordinate to the diene or dienophile could be used. The chiral catalyst, when
coordinated to either reagent theoretically only allows the carbon-carbon bond formation
to occur preferably from one face over the other. There are pros and cons of each
method, but iniorder to produce enantiopure substrates the use of one of these methods is

necessary.

Each of these options will now be discussed in greater detail to provide insight
into the advantages and disadvantages of each. Using substrate control requires the use
of a substrate with a stereocenter previously incorporated. Some enantiopure compounds
are commercially available, but if they are not, then one must prepare the substrate with
the desiredlstereochemistry prior to executing the Diels-Alder reaction. This means that
the same issues need to be faced to make that stereocenter before attempting the Diels-
Alder reaction. Although extra effort is required, if that original stereocenter is
incorporated into the target compOund or could be used later in a synthetic route, then the
substrate control method would be attractive.‘ When a chiral auxiliary or chiral catalyst is
used, the issues of the atom economy of the reactionbecome compromised. In the case
of the chiral auxiliary, the group is first put on and then removed later. The down side to
this method is that extra steps, time, and effort are needed to recover the chiral auxiliary.
When using a chiral catalyst, the recoverability of the catalyst is also an issue. Even
thorigh a covalent bond is not necessarily needed to promote the reaction, the catalyst is
not incorporated into the product. In many cases these chiral catalysts use expensive
metals and/or ligands. It would be undesirable if in the end the expensive catalyst was
lost during the workup. Despite these challenges, successful installation of up to four
chiral centers make the effort and/or expense worth it.

1.2 Heteroatom-Diels-Alder Reaction

The discussion tothis point has dealt with the general aspects of the Diels-Alder
reaction and its utility in organic synthesis. The discussion will now turn to the A
heteroatom Diels-Alder reaction, as this topic is the main focus of this thesis. As

discussed above, the Diels-Alder reaction produces a Six-membered cycloalkene. When

a heteroatom is involved, the resulting product contains a six—membered heterocycle
(Scheme 1.8). There are many types of heteroatom Diels-Alder reactions in

Scheme 1.8 Heteroatom Diels-Alder Reaction

/ o O
c + t ——. U
\ R R
41 42 43
O R

/O R
c + i —-———-—» U
\

44 _ 45 46

i i

/N X‘R lN x\R
\ + I

47 48 49

¢ N
.. +1 ———-» U
\ X’R x/R
50 51 52
c n U
\ kR R
53 54 55

12. 13

the literature” and probably the most common are ones in which an oxygen or

nitrogen'“5

is incorporated. The oxygen and nitrogen can be included in either the diene
or dienophile and depending on the heteroatom location, the reaction will proceed to give
different substitution patterns around the heterocyclic ring.

Regioselectivity is usually very predictable in the heteroatom Diels-Alder
reactions due to the polarized nature of the imine and aldehyde carbon-nitrogen or

carbon-oxygen double bond. The reactivity of these type of starting materials, especially

in the case of the more sterically hindered imine also becomes an issue. However if the

10

dienes such as those reported by Danishefsky, Rawal, or Brassard are used to react with
aldehyde or imino dienophiles (Scheme 1.9) the rate of the reaction can be increased.

Scheme 1.9 Regioselectivity in the Heteroatom Diels-Alder Reaction

 

 

OMe ,. OMC W
/ o - / 0
+ t -————> O ————->
\ o R
TMSO R LTMSO R
31 42 ' 56

 

 

OMe ‘ _ 2

,P OMe

/ + N p / N’P
)1 ___. N, _.

TMSO \ R I o R
31 54 gFMSO RJ 57
Another difficult challenge deals with the basicity of the starting materials and products
and arises when developing methodologies for the heteroatom Diels-Alder reaction.
When an imine or aldehyde dienophile is used, the product contains an Sp3 oxygen or

nitrogen that is more basic (Figure 1.4) than the sp2 oxygen or nitrogen in the aldehyde or

Figure 1.4 pKa of Sp2 and Sp3 Oxygen and Nitrogen

 

H\ + H R\+/H
I I I
0+ \ I
R/ ‘R R H R’ R R H
[pKa 6 —6 J 10 7 j

 

imine. When catalytic amounts of the chiral Lewis acid is used, turnover prOblems occur
as a result of this issue.
1.3 Previous Examples of Lewis acid/Brdnsted Acid Catalyzed aza-DieIs-Alder
Reaction
The main focus of the work in this thesis has been the development of a catalytic

asymmetric Diels-Alder reaction of imino dienophiles with Danishefsky’s diene. Before

11

this work can be discussed, it is necessary to review examples that already exist in the
literature utilizing chiral catalysts to facilitate this Class of reaction. There have been
many researchers contributing work to this field and their successes and limitations will
be discussed at this time. 1
1.3.1 Stoichiometric Catalyst

Yamamoto’s studies using catalysts prepared from triarylborates and BINOL”l9
(Scheme 1.10) represents the first example when a chiral catalyst was used in the aza-

Scheme 1.10 Pioneering Work by Yamamoto

CH2C12,4A MS.
RT, 3h

 

B(OAr)3 + (R)—BINOL
58 59

 

Ar = Phenyl, 2—tolyl, 2,3-xylyl
Proposed Catalyst (60)

Diels-Alder of imino dienophiles. In his studies, he found that a Stoichiometric amount
of catalyst 60 was needed to facilitate the reaction. He screened reactions of

Danishefsky’s diene with several imines (61) prepared from various aldehydes and

benzyl aminel9 (Scheme 1.11). When R=Ph (61a), 86% ee could be achieved when the

12

 

Scheme 1.11 Substrate Scope for B(OAr),/BINOL Catalyst

0

l/
O/Si\ B(OAr)3/BINOL (60, 1.0 equiv.L
Rv/‘vah + M CH2C12,4A MS. I
OMe

 

 

 

o R N

61: 1-0 CQUIV- 31, 1.5 equiv. 78 C, 5 h I\Ph
62

Entry R Ar (cat.) Yield 62 (%) ee 62 (%)
1 Ph (613) Phenyl (60a) 75 82
3 Ph (613) 2-tolyl (60b) 76 84
4 Ph (61a) ' 3,5-xylyl (60c) 75 86
5 3-pyridyl (61b) Phenyl (60a) 70 h 90
6 Cy (61c) Phenyl (60a) 45 76
7 Cy (61c) . 3,5-xylyl (60c) 49 72
8 3,5-dimethoxy—Ph (61d) Phenyl (60a) 89 74
9 2-naphthyl (61c) Phenyl (60a) 83 84

 

catalyst was prepared from tri-3,5-xylylborate (60c) and for the other substrate/catalyst
combinations, a range of ee’s from 72-90% could be reached with yields ranging from 45
to 89%. In a subsequent paper”, several other metals with BINOL (59) were screened for
the reaction of the imine prepared from benzaldehyde and benzylamine and
Danishefsky’s diene. This study indicated that all other metals screened (Al, Ti, Zn,
. other borates) were inferior to B(O-3,5—xyly1)3. In this publication, he also screened
several substrates as well as different solvents. He found that using. propionitrile, THF,
and toluene all gave results far inferior to those obtained with CHZCIZ.

In this study he also screened imines (63) prepared from benzaldehyde and other

amines (Scheme 1.12). Entry 3 is very relevant to the present work as the imine

13

Scheme 1.12 Screening Different Imine Protecting Groups for the B(OAr),/BINOL

Catalyst

0
|./ '
O,SI\ B(OPh)3/BINOL (603, 1.0 equivi
thf‘k P + M CH2C12, 4A MS. 7 m
OMe I?
P
64

 

Ph
. . —78°C, 5 h
63. 1-0 equrv. 31, 1.5 equiv.

 

 

Entry P Yield 64 (%) ee 64 (%)
1 3,4—dimethoxy-Bn (633) 73 85
2 —CH2-CH=CH2 (63b) 97 70
3 Bh (63c) 0 —
4 _ Ph (63d) 77 24
5 i-Pr (63c) 13 4

 

used (63c) is identical to that used in optimization studies discussed in the next chapter.
It is interesting to note that when Yamamoto screened this substrate using a
stoichiometric amount of catalyst 603, no reaction was observed.

Although this system required 1.0 equivalent of the chiral catalyst, it was

interesting to find that Furrnan used this system 13 years later20 (Scheme 1.13) in attempt

14

Scheme 1.13 Furman’s use of Yamamoto’s Catalyst

 

 

 

 

o o
\ . B(OPh)3/(R)-BINOL I TBAT (2.0 equiv.) _ H
R/\ N/\”/\TMS (60a. 1.0 equiv) > THF, 30°C, 1-2 h V R N
CH2C12,4A M.S. R N
65 ' —78°C, 125 h TMS
66 . 67 (indolizidine)
Entry R Yield 66 (%) ee 66 (%)
1 C.,H5 (653) 75 90
2 4—MeC6H4 (65b) 70 . 86
3 4-MeOC6H4 (65c) 68 60
4 4—CIC6H4 (65d) 75 95
5 2—pyridyl (65c) 57 62
6 CH3CH2CH2 (650 70 72
7 CH3(CH2)-,CH2 (65g) 70 64
8 Cy (65b) 30 71
9

 

(CH3)3C (650 0 —

to provide the Diels-Alder adduct (66) which could be further transformed into
indolizidines (67). The initial step in his syntheses of the indolizidines was the aza-
Diels-Alder reaction. All of the imines screened produced Diels-Alder product with the
exception of the substrate where R = t-butyl (651). For all the others, a range of 57-80%
yield and 62-95% ee could be achieved for this classgof imines. Ultimately each of these
substrates were successfully taken .on to do a fluoride mediated Hosomi-Sakurai
allylation to afford the trans-indolizidines (67) diastereoselectively21 (64-85% yield).

In 1994, Yamamoto reported an aza-Diels-Alder reaction using a chiral Br¢nsted
acid22 (Catalyst 68) prepared by simply adding 2 equivalents BINOL with one equivalent

of a trimethylborate (Scheme 1.14). When catalyst 68 was employed in the aza-Diels-

15

 

Scheme 1.14 Yamamoto’s Chiral Brdnsted Acid Catalyst

 

B oM R BINOL CH2C12,4A M.S. _
( 676% + ( )-59 reflux, 23 h
(Soxhlet Thimble)

 

Catalyst 68 '

Alder reaction of the imine 613 prepared from benzaldehyde and benzylamine with
Danishefsky’s diene, a 78% yield and 86% ee was realized for 623 (Scheme 1.15).

Scheme 1.15 aza-DieIs-Alder Reaction Using the Brensted Acid Catalyst

0

l/
,5! \ Catalyst 68
O > I
7 th NV Ph + /I\/\ CH2C12,4A M.S.
/
OMe

 

Ph N '
' ——-78°C, 5 h
613, 1-0 equiv. 31, 1.5 equiv. 623 I\Ph

75% yield
86% ee

It is worthy of mention that this work has been repeated by Bull and James23 for
probing the structure of the catalyst proposed by Yamamoto. They found non-linear
effects forthe catalyst preparedpfrom BINOL operate and that dynamic ligand exchange
occurs for these types of catalysts. Since the non-linear effects were observed using
scalemic BINOL, it was concluded that two equivalents of BINOL were included in the
active catalyst even for the case when only one equivalent of BINOL was used (Scheme
1.10). If only one BINOL was incorporated, then the resulting ee would drop linearly
with respect to the cc of BINOL used whereas the incorporation of two equivalents of

BINOL in the active catalyst structure allows for the observation of a nonlinear

relationship with the ee of BINOL used.

16

Another report in the literature where a stoichiometric amount of catalyst was
used in a heteroatom Diels-Alder reaction was that by Whiting using a zinc catalyst24 (70)
(Scheme 1.16). The catalyst he used also involved BINOL. In this work the catalyst was

Scheme 1.16 Use of Stoichiometric Catalyst Derived from Ean and BINOL

0 I
BINOUEt2Zn (70, 100 mol%) \/0

o
. / l
E o/Ik/q I 0/51 \ N
I + M CHZCIz’ temperature V O
0M6 OMe 2.5 h 0

 

 

 

69 31 V 71
OMe
Entry Temperature (°C) Yield 71 (%) ee 71 (%)
1 22 78 93
2 0 72 92
3 —40 66 88
4 —78 63 72

 

prepared using Ean and the more reactive imine (69) prepared from ethyl glyoxylate and
4-methoxy-aniline was the substrate. Whiting found that when the temperature of the
reaction was raised, an increase in yield and ee was observed.‘ When the reaction was
done at —78°C (entry 4) the reaction gave 63% yield and only 72% ee, however, when
done at room temperature (entry 1) the yield was 78% and the ee surprisingly was 93%.
This trend is opposite from that which is usually observed in asymmetric catalysis, but
has been observed in other cases before.” 26 This kind of trend indicatesthe possibility
for more than one catalytic species in the reaction that have different reactivity at
different temperatures. Whiting also did Several reactions using substoichiometric

amounts of catalyst and these results will be discussed in the next section.

17

 

 

 

1.3.2 Sub-Stoichiometric Catalyst

AS dichssed earlier, it is not desirable to use stoichiometric amounts of the
catalyst to accomplish these reactions. Since the initial reports by Yamamoto using a
stoichiometric chiral Lewis acid, many other researchers developed systems that were .
able to achieve turnover with asymmetric induction. One of the major contributors in this
area is Shu Kobayashi at the University of Tokyo, who mainly focused on catalysts that

”'3‘. In his first report30 he describes the use of a

are zirconium BINOL derivatives
catalyst prepared from Zr(OtBu)4 and 6,6’-dibromoBINOL (Catalyst 74) (Scheme 1.17).

Scheme 1.17 Catalyst Derived from Zr(OtBu)4 and 6,6’-dibr0moBINOL

ZKOIBU)‘; +
72

    

Catalyst 74

He also attempted to use Hf and Ti derivatives but they both gave results that were
inferior to Zr. His optimization studies were done using the imine 753 prepared from 1- ,
naphthaldehyde and 2-aminophenol (Scheme 1.18). The requirement for the phenol

Scheme 1.18 Optimal Conditions Using the Naphthyl Derived Imine

, ’ | / Catalyst 74(10 mol%)
Si
/ \ =
O \N O + AA L NMI(30mol%) >
/ toluene
0 OH 0M6 _450C

753 31

 

 

86% yield
82% ee

18

 

 

group on the imine was proposed to be a result of the fact that zirconium has binding sites
available that can accommodate a bidentate substrate. When the imine approaches the
metal it will coordinate through the imine nitrogen and the phenol oxygen, activating the
imine allowing it to react with the diene. He found that the optimal conditions were those
where the catalyst was prepared in toluene using N-methyl imidazole (NMI) as the
additive L. With the optimal results in hand, he then screened several imines with
varying R groups and found that this catalyst was quite good for most substrates (Scheme
1.19). When the catalyst loading was increased to 20 mol%, the ee

Scheme 1.19 Screening of Substrates with the Zr/Br-BINOL Catalyst

 

 

 

0 .
I Catalyst 74
O O,Si< L = NMI _ I
RAN + / toluene V R N
OH OMe —45°C on
75 31
76
Entry R Cat. 74 (mol%) Yield 76 (%) ee 76 (%)
1 1-naphthyl (753) 5 72 67
2 ' l-naphthyl (753) 10 86 82
3 l-naphthyl (753) 20 96 88
4 _ l-naphthyl (753) 30 98 89
5 1-naphthyl (753) 50 88 90
6 10 92 80
(75b)
7 o—tol (75c) 10 81 76
8 o-tol (75c) , 2O 83 82
9 2-thienyl (75d) 10 86 64

 

19

 

 

reached 88% for the l-naphthyl substrate (753) and could only be increased to 90% using
50 mol% catalyst. For the other substrates, moderate ee’s were also produced and the
yields were all high. The lowest observed ee was when R: 2-thienyl (75d) and only gave
64% ee. This was found when 10 mol% catalyst was used and the reaction with
increased catalystloading for this substrate was not reported.

About one year later, Kobayashi reported similar catalysts using different BINOL
derivatives.28 It was observed from this work that a switch in enantioselectivity could be
achieved by simply by putting an aryl group in the 3 and 3’-positions of the BINOL

Ii gand (73) (Scheme 1.20). This catalyst proved to be just as efficient and selective as the

20

Scheme 1.20 Switched Enantioselectivity with Different BINOL Derivative

INMI benzene

 

ZT(OIBU)4 +

 

 

 

72, 20 mol% 60 mol% 3A M.S., 23°C
Br
Catalyst 77
' 0
Catalyst 77 ‘ RWEIIj
R N benzene r
0M6 3A M.S., 23°C (III/OH
76

 

Yield 76 ee 76

 

Entry R (%) (%)
1 Ph (75c) 94 82
2 o-tol (75c) 93 91
3 l—naphthyl (753) 88 84
4 2—thienyl (75d) 74 86
5 cyclohexyl (751’) 64 81
6 67 85

(75b)

\l

O
< 0059 90 90
0

original catalyst and a range of ee’s from 81-91% was observed and yields from 64-94%

 

could be achieved for a range of substrates. It is worth mentioning that a major

difference between the two reaction conditions was that in the second report, the

21

reactions Could be run at ambient temperature whereas the first system needed much
colder conditions (—45°C) in order to achieve acceptable selectivities.

In 2000, Kobayashi reported many other variations of the zirconium/BINOL
based catalysts29 (Figure 1.5). In this report he discussed the use of catalysts where the

Figure 1.5 Other Zirconium/BINOL Catalysts

OR

 

Catalyst 78 Catalyst 79

BINOL had a variety of different groups in the 3,3’ and 6,6’ positions (Catalyst 78 and
79) including one where the BINOL was tethered to a solid support (Catalyst 78). In
addition, he also explored zirconium catalysts where R‘zOtBu or R4=CN. He screened
many combinations of R', R2, R3, and R4 and discovered that yields and
enantioselectivites around 90% could be achieved. When 20 mol% of the polymer
supported catalyst (78) where R2=H and R3=F was used, the reaction gave 99% yield of
76 and 91% ee was observed. The catalyst was easily recovered by simple filtration and
used in subsequent reactions. Using the recovered catalyst for the second run, the
reaction gave 97% yield and 90% ee, and the third gave 97% yield and 90% ee. This
appears to be the first literature example where a polymer-supported Lewis acid was used
to catalyze the aza-Diels-Alder reaction. This is a desirable system due to its ease in

recoverability and efficiency when reused.

22

In the previously described work by Kobayashi, it should be noted that he did not
report the use of any primary aliphatic aldimines. One issue that arises when attempting
reactions using this class of imines is. that in the presence of an acid, they are prone to
isomerization to the enamine (81) (Scheme 1.21). When this isomerization takes place,

Scheme 1.21 Isomerization of Aliphatic Imines to the Enamine
R/V N‘P JC—I‘L» R/VRP
80 81
the nitrogen becomes much more basic and will ultimately bind preferentially to the
catalyst over the imine eliminating the possiblility for the heteroatom Diels-Alder
reaction to occur. This problem was solved by Kobayashi and in a report published in
2005:“; he described the aza~Diels-Alder reaction of the C-N double bond of the
hydrazone 82 as a surrogate for an imine. The hydrazones were prepared from primary

aliphatic aldehydes and benzoylhydrazines (Scheme 1.22). The hydrazones are stable

23

Scheme 1.22 aza-Diels-Alder Reaction of Hydrazones

 

o
O in/ 240"?!)4-PTOH (84. 10 mol%)
R N + 0’ \ (R)-3.3',6.6'-I4-BINOL(85, 12 mol%)_ ' I
V \E P“ / o 'BuOMe/DME (471) ‘ ' R“ N
32 83 — 10°C, 48h HN\n/ Ph
86 O

 

Entry R Yield 86 (%) ee 86 (%)

1's
1 O/ (823) - 44 89
2 X25» (82b) 31 92

 

 

 

 

 

 

4 /\/H'1 (82d) 70 93

 

and it was found that the rate of the aza-Diels-Alder reaction exceeded that of the
isomerization of the imine to the enamine. For this reaction, he again used a zirconium
catalyst, but this time found it was best to use Zr(OnPr)4-PTOH 84 utilizing (R)-3,3’,6,6’-

tetraiodoBINOL 85 as the ligand. Using 20 mol% Zr and 24 mol% ligand, he optimized
this systemin toluene at 0°C using the imine prepared from dihydrocinnamaldehyde and
benzoylhydrazine (82c). Reaction of 82c with Danishefsky’s diene only gave 19% yield
and 90% ee. However, if the methoxy group on the diene was replaced with a t—butoxy
group (83), the yield could be increased to 35% with 93% ee observed. Using this diene
he screened several solvents and found that a 4:1 ratio of t—butyl methyl ether (T BME)
and dimethoxyethane (DME) was optimal. Finally, he screened other substrates and all

were found to be successful (see Scheme 1.22). The best result was observed when

 

 

R=propyl (82d) and 93% ee was achieved with good yield. The other substrates gave
ee’s around 90 but the yield dropped significantly when cycthexyl (823) and isobutyl
(82b) groups were on the hydrazone.

All examples discussed to this point used at least 10 mol% catalyst. Prior to a
report by Snapper and Hoveyda in 200332, ~10 turnovers was the most that was achieved
by any catalyst. The silver catalyst developed by Snapper and Hoveyda (Scheme 1.23)

, Scheme 1.23 aza-Diels-Alder Reaction Using a Silver Catalyst

 

 

 

Me Et
It"
\ N\ O
N R
. 0““ I/ (t 0 I19
Ar N st 2
V O + 0’ \ (88, 0.1 - 1.0 mol%) Ar N
/ OMe AgOAc (0.1 - 1.0 mol%) 7 f 0M“
87 ,
31 iPrOH (1.0 equiv.) O
THF, 4 C, 12 h 89
Li gand/AgO Yield 89 ee 89
Em Ac (mol%) A' ' R (%) (%)
1 '1 .0 Ph (873) o-OMeCéH4 (883) 94 93
2 0.5 Ph (873) o-OMeCfiH4 (883) 92 . 92
3 0.1 Ph (873) o-OMeC6H4 (88a) 78 88
4 1.0 l-naphth (87b) o-OMeC6H4 (883) 94 90
5 0.5 2-naphth (87c) o-OMeC6H4 (883) >98 95
6 1.0 p-OMeC6H4 (87d) o-OMeC6H4 (883) 86 91
7 1.0 p-ClCfiH4 (87c) o-OMeC6H4 (883) 98 90
- 8 1.0 o-Br C6H4 (87f) o-OMeC6H4 (883) 91 89
9 1.0 m-N02C6H4 (87g) o-OMeC6H4 (883) 92 91
10 1.0 p-NOsz,H4 (87h) o-OMeC6H4 (883) >98 92
11 1.0 2-furyl (87i) o-OMeC6H4 (883) 89 92
12 1.0 p-CIC6H4 (87c) p-OMeC6H4 (88b) >98% conv. 92
13 1.0 p-ClC6H4 (87c) p-CF3C6H4 (88c) 75% conv. 88
14 1.0 p-CIC6H4 (87c) 2,6-Me2C6H4 (88d) 53% conv. 28
15 1.0 p-CIC6H4 (87c) NHBu (88c) >98% conv. 80
16 1.0 p-ClCfiH4 (87c) » ,Bn (88f) >98% conv. 80
17 1.0 p-ClCéH4 (87c) NH(OMe) (88g) 52% conv. 20

 

25

using a chiral phosphoryl-aryl imine amide ligand (88) is capable of turning over the aza-
Diels-Alder reaction of the imine 87a prepared with benzaldehyde and 2-methoxyaniline
with Danishefsky’s diene 780 times (i.e. using 0.1 mol% catalyst, the yield was 78%).
The reaction was very general for aromatic imines when R on the ligand was ortho-
OMeCéH4 (883). Using 0.5 or 1.0 mol% of this catalyst/ligand combination, a range of
yields from 86% to nearly quantitative could be achieved with ee’s ranging from 89 to
95%. Reactions were thenstudied to test the effect of the R group on the amide nitrogen
of the ligand. The selectivity remained in the 80’s for each of the different R groups with
the exception of the bulky aryl group (2,6—Me2C6H4) (88d) and R = NH(OMe) (88g)
which dropped the ee into the 20’s. The results from the screening and catalyst loading
studies demonstrate that this reaction is an attractive one from the standpoint of turnovers
and asymmetric induction.

Two other reactions were also attempted which makes this system even more
attractive. One is the reaction run in air using undistilled THF and the other without the

use of any solvent (Scheme 1.24). The reaction using undistilled solvent still gave ee’s in

26

Scheme 1.24 Use of Silver Catalyst with N o Solvent or Undistilled THF in Air

,_< Erick“). 35,.

 

Ar N ,S.
V b 0 h2,(88b I .0 mol%)
+ M
87 OMe A gOAc (1.0 mol%)
31 H20 (1.5 equiv.)

air, undistilled THF, 4°C

Ar = 2-Naphth (87c) 82% yield, 94% ee (89c)
Ar = p-Cl—C6H4 (87c) 31% yield, 90% ee (89c)

 

Ar N ’81
V b 0 (288b, 1.0 mol%)
+
31 i'PrOH (1.0 equiv.)

4°C, no solvent
Ar = p-OMe-C6H4 (87d) 66% yield, 90% ee (89d)
Ar = p-NOz-C6H4 (87b) 88% yield, 80% ee (89h)

the 90’s with good yields. When no solvent was used, the yields were still good, and for
p-OMe-C6H4 (88b) 90% ee was achieved. Lastly, Hoveyda, et. al. attempted to use a
ligand tethered to solid support (88h) which could be easily recovered, similar to the

work done by Kobayashi in 2000 (Scheme 1.25). This catalyst was used for the reaction

27

Scheme 1.25 aza-Diels-Alder Reaction Using Silver Catalyst with a Recoverable

 

Ligand
-Wan
M Et g
H
\ Ill/N O O
N
OMe I/ I I PPM 0 ((II]
2-na hth I N S'
P yv (E o’ '\ (88h, 5.0 mol%) 2-naphthyl N
+N %
OMe
87c / 0M6 AgOAc (5.0 mol%)
31 iPrOH (.0 equiv.)

4°C, undistilled THF, air
89c

96% yield, 86% ee
after ligand recycled 5 times

and recycled five times and when used again, 96% yield and 86% ee was achieved.

Each of the remaining Lewis acid catalyzed aza-DielS-Alder reactions to be “
discussed will deal with the use of activated imines. The imines used contain either an
electron-withdrawing group on the nitrogen or an ester on the imine carbon, making the
imine carbon more electrophilic and thus more reactive. The work done by Carrretero,
et. al. employed a copper (1) complex with planar chiral phosphino sulfenyl ferrocene33
(90) (Scheme 1.26). Using the tosyl imine prepared from benzaldehyde (923) he

Scheme 1.26 Copper Phosphino Sulfenyl Ferrocene Catalyst

 

S-r-B
@( " ©<~ ug/
THF/MeOH=

Fe PR2

: RT, 10 min

90 ©Catalyst 91
X = C1 or Br

R = p-F-C6H4, 2-furyl, Cy, o-Tol, or l-naphthyl
discovered that the best catalyst was where X=Br and R=1-naphthyl (913). Using 5.1

mol% of this catalyst and 10 mol% AgClO4 as a halogen scavenger, a 93% ee for 953

28

could be obtained when the reaction was run at room temperature and 97% ee was
observed at -20°C. Using this catalyst, he then screened several different imines
(Scheme 1.27). The reaction was found to be fairly general for the imines screened and a

Scheme 1.27 Substrate Scope Using Copper Phosphino Sulfenyl Ferrocene Catalyst

’Bu
I
qs""‘Cti/Br

R2 2
© 0
,SOZRZ I./
JN O,SI\ (913, R = l-naphthyl, 5.1 mol%) I
+ > -
R1 M - AgClo4 (10 mol%)

 

 

 

OMe RI If
R2 = o-tol (92) 31 (“We “1 1'? “ 50,122
R2 = p-OMe-C6H4 (93) then TFA (5 equiv.) R2 ___ 040] (95)
R2 = p—NOz-C6H4 (94) R2 =P-0Me-C6H4(96)
R2 = p-NOz—C6H4 (97)
Entry R1 R2 Product Yield (%) ee (%)
, 1 Ph (923) p-tol 953 90 93
2 o-tol (92b) p-tol 95b 82 93
3 p—F—C6H4 (92c) p-tol 95c 78 88
4 p-OMe-C6H4 (92d) p-tol 95d 76 91
5 p-NMez-CGH4 (92c), p-tol 95c 39 93
6 2-naphthyl (92f) p-tol 9511’ 85 - 86
7 PhCH=CH (92g) p-tol 95g 66 83
8 n-Pr (92h) p-tol 95h ' 65 73 (—20°C)
9 Ph (933) p-OMe- C6H4 963 61 94
10 o-tol (93b) p-OMe— C,.,H4 96b 78 90
ll 2-naphthyl (93c) p-OMe- C6H4 . 96c 56 82 .
12 . PhCH=CH (93d) p-OMe- CfiH4 96c 58 76 .
13 Ph (943) p-NOZ- C6H4 973 58 90

 

range of yields from 39-90% was achieved with ee’s ranging from 73-94%. It was also
found that in some-cases, the lower ee’s could be increased by Simply running the

reaction at lower temperature (i.e. entry 7, at —20°C gave 96% ee).

29

Jorgensen,'et. al. also published catalytic versions of the aza-Diels-Alder reaction
with the extremely reactive imines derived from and bearing a tosyl group on the
nitrogen“35 (983). Using a Wide variety of chiral ligands (Scheme 1.28) he screened

Scheme 1.28 Ligands Screened by Jorgensen

 

 

N/TOS SIi/ Ligand / Lewis Acid
no I + 0’ \ (10 mol%) I
M T; BIO “.t
O OMe ‘

 

 

 

I
(34 J 54 J .._Q><FJ

 

102 103

Ar- — Ph (100)
Ar = p-Tol (101)

PAer PA AT2 PAl'zN
R
Ar .-; Ph, R = Ph (R) (105)

Ar = Ph, R = iPr (S) (106) Ar: Ph (109) Ar = Ph (112)
Ar = Ph, R = tBu (S) (107) Ar = p-tol (110) Ar = p-tol (113)
Ar = p-tol, R = tBu (S) (108) Ar = 2,4—xylene (111) . Ar = 2,4~xylene (114)

CuClO4-4 MeCN, 2 CuOTf°C6H6, Cu(OTf)2, A ngFG, A gOTf, AgClO4, Pd(SbF6)2,
Pd(ClO4)2, Pd(OTf)2, RquFé, and Zn(OTf)2 as Lewis acids for this reaction. The copper
(1) Lewis acids gave the best enantioselectivities (61-80%) followed by the silver Lewis
acids which gave ee’s from 30% to 34%. Zn, Pd, and Cu (11) all gave ee’s in the teens or
less. In his initial publication,34 he found that the best ligand and Lewis acid combination

30

was the BINAP ligand (101) where Ar: Tol with CuClO4'4 MeCN giving 68% yield and
80% ee. Ina subsequent publication”, he studied the effectiveness of the phosphino-
oxazoline ligands (105-114) with the Lewis acid CuClO4-4 MeCN. Using chiral ligands
107 and 108, the ee’s could be increased to 87%, which is slightly better than the 80%
produced using the BINAP ligand. He then used the BINAP ligand (101) where Ar = Tol
with CuClO4 to Screen different imines using THF or CHZCl2 as the solvent (Scheme
1.29). The effect of solvent proved to be purely substrate dependant as indicated by entry

Scheme 1.29 Screening of Substrates using the Cu/BINAP Catalyst

 

 

 

N’ 2 Q’Sli< BINAP (AmTol) / CuClO4 O
| + (10 mol%) ¥ (L5
RKI Mom le N I
R2 = Tos (98) 31 1.8
R2 = COOEt (115) 99
R2 = o-MeO~Ph (116)
R2 = Ph (117)
1 2 o Yield/ee 99 (%) Yield/ee 99 (%)
. Entry R R Temperature ( C) (THF) (CHZCIZ)
l COOEt (98a) Tos —78 80/79 89/26
2 COOEt (115a) COOEt -—78 10/79 23/77
3 COOEt (116a) o-MeO-Ph RT 25/60 23/58 _.
.4 COOEt (117a) Ph RT 61/86 78/91 .-
5 COOEt (116a) o-MeO-Ph RT 93/15 89/72
6 ' COOEt (116a) . o-MeO-Ph —7 8 to RT 82/16 75/78
7 Ph (98b) Tos RT 65/46 65/48

 

1 versus entry 6. In entry 1, the ee went from 79% to 26% by switching from THF to
CH2Cl2 and in entry 6, the opposite effect was observed as the ee went from 16% in THF
to 78% in CHZClz. 9

With respect to the yield, it changed from 82% to 75% upon switching from THF
to CH2CI2 for R1 = COOEt, R2 = o-MeO-Ph (1163, entry 6) but for R1 = COOEt, R2 = Tos

31

(98a, entry 1), the yield went from 80% in THF to 89% in CHZClz. Jorgensen’s work has
proven the effectiveness of copper (1) Lewis acids with a variety of different ligands for
the aza-Diels-Alder reaction of activated imines.

Andrew Whiting has also contributed to the field studying the reactive imines
(118) prepared from the glyoxylate aldehydes and 4—methoxy anilineu'“. In his initial
report”, he screened Yb(OTf)3, Cu(OTf)2, Mglz, and Fe'Cl3 as possible Lewis acids in
conjunction with BINOL (59), a bis-oxazoline (102) and a chiral diamine (120). Using a
combinatorial approach, he explored different combinations of solvents, additives, Lewis
acids and ligands (Scheme 1.30). He found that the chiral diamine (120) was the best

Scheme 1.30 Whiting’s Combinatorial Approach

0
OMe
l Lewis Acid 10 mol‘7
/Si/ , (, a) MeO I
IN 0 \ Chiral Ligand N
+ =
MeO N Additive 0
ll) / OMe O

 

Solvent 119

OMe
Lewis acid = Yb(OTf)3, Cu(OTf)2, Mglz, or FeCl3

Additive = 2,6—Lutidine, 4A MS, or none
Solvent = Toluene, MeCN, or CH2C12

O O O O
OH or ’ l I) r
on N N ._ "<

0
ma tBu HZN NH2

 
  

Chiral Ligand =

 

59 102 ' 120

ligand in conjunction with Mglz, Yb(OTf)3, and Cu(OTf)2, whereas the bis-oxazoline
(102) was best for FeCl3. The effects of additives and solvents used for these

ligand/Lewis acid combinations are presented in the follovving table (Table 1.1) with

32

 

 

Table 1.1 Best Results from Combinatorial Study

 

 

Entry Lewis Acid Ligand Additive Solvent Yield 119 (%) ee 119 (%)
l Mgl2 120 2,6-Lutidine MeCN 64 97
2 Yb(OTf)3 120 2,6-Leutidine Toluene 6O 87
3 Cu(OTf)2 120 none MeCN 58 86
4 Fed, 102 4A M.S. CH2c1, 67 92

 

yields ranging from 58-67% and ee’s from 86—97%.

Interestingly, about 5 years later the same author published a corrigendum to this
paper37 indicating that the results in Table 1.1 were not reproducible. For MgI2 only a
range of ee’s from 0% to 55% could be obtained and the remainder of the entries when
repeated only gave racemic product.

In a subsequent report in 200424, Whiting attempted to find a metal Lewis acid
that would produce acceptable yields and/or ee’s when BINOL (59) was used as the
ligand for the aza—Diels-Alder reaction of the imine 121 derived from ethyl glyoxylate and
4-methoxyaniline and Danishefsky’s diene. Initially Whiting screened Et3Al, EtzAlCl,
EtzZn, Et3B, (iPrO)3B, (PhO)3B, and (iPrO)4Ti. He found that the only two Lewis acids
that gave any asymmetric induction were Et3Al and‘EtQZn. Et3Al gave 65% yield and
15% ee whereas EtZZn gave 57% yield and 36% ee. One entry relevant to the present
work was the attempt made using triphenylborate. This reaction failed to produce any
product. There was not a detailed experimental procedure outlined for this/reaction so it
was not clear at what temperature he ran this reaction.

I Due to the better selectivity with EtZZn, Whiting prepared a catalyst from Ean
and BINOL and optimized the reaction of imine 121 with Danishefsky’s diene using 10

mol% catalyst (Scheme 1.31). He found that the best cc of 123 that could be achieved '

33

 

Scheme 1.31 Whiting’s Substoicheometric Approach

OMe
l. / EtZZn (122, 10 mol%) l
SI EtO
N o’ \ BlNOL(59, 11 mol%) N
+ >
EtO ' M Toluene, RT, 15 h o
OMe O

O 121 31 52% yield
84% ee

 

OMe

was 84% with 52% isolated yield using toluene as the solvent and running the reaction at
room temperature for 15 hours. Comparing this work from Whiting to the work of
Jorgensen with the BINAP ligand“ 35 on this class of imine, it can be determined that
Jorgensen’s system is slightly better since he has examples where 290% ee could be
achieved.

This concludes the discussion of work previously doneusing chiral Lewis acid
catalysts. However, recent reports with organocatalysts also have described systems that
can produce good asymmetric induction with sub-stoichiometric amounts of catalyst.

Akiyama, et. al. have reported success with chiral Br¢nsted acids38 (Scheme 1.32) that are

34

 

 

Scheme 1.32 aza-Diels-Alder Using Catalytic Chiral Bransted Acid

0

Phosphoric acid catalyst

Me
Sli/ . I
\ O + O/ \ L N
N N (lOmol%)
/ O 0”
OH

 

toluene, —78°C

125a

Phosphoric acid catalyst =

 

126

 

phosphoric acid derivatives of a 3,3’- diarleINOLs (126-128). Upon initial screening of
the reactions using 10 mol% of each of these catalysts, it was found that catalyst 128 with
the bulky tri-isopropyl aryl groups in the 3,3’-positions gave much better selectivity (42%
versus <5% ee). This reaction was run at —78°C in toluene but only gave 32% yield. In
order to optimize the yield'as well as the ee, several additives and solvents were screened.
The optimization studies ultimately indicated that toluene was the best solvent for the
reaction using 10 mol% phosphoric acid (128), 1.2 equivalents acetic acid, and a reaction
temperature of —78°C. Using these optimal conditions, he then studied the substrate
scOpe for the aza-Diels-Alder reaction using the chiral phosphoric Br¢nsted acid 128

(Scheme 1.33). All the reactions produced significant quantities of product with yields

'35

Scheme 1.33 Substrate Scope using the Chiral Phosphoric Bransted Acid

 

0
Me
SI" / Phosphoric acid catalyst I
1 ,.
O 0’ \ (128, 10 mol%) R“ N
/&§ + . >
R N M CH3C02H (1.2 equrv.)

OMe OH
OH toluene, —78°C
124
Me

 

 

 

 

 

125
Proposed transition state
Entry R Reaction time (h)Yield 125 (%) ee 125 (%)
1 Ph (124a) 18 99 80
2 p-IC6H4 (124b) 24 86 84
3 p-BrC6H4 (124c) 13 100 84
4 p-C1C6H4 (124d) 35 72 84
5 p-FC6H4 (1246:) 13 77 78
6 p-CF3C6H4( 1241') 21 82 81
7 o—BrC6H4 (124g) 10 96 80
8 o-C1C6H4 (124h) 12 100 76
9 l—naphthyl (124i) 12 100 91

 

ranging from 72% to quantitative. The selectivities were also quite good for this catalyst

ranging from 76 to 91% ee. The first application of a chiral Br¢nsted acid catalyst to the

aza-Diels-Alder reaction proved to be successful and is a nice addition to the Lewis acid

based catalyst systems.

36

 

1.4’VAPOL and VANOL Ligands

As discussed above, there have been many catalysts used successfully in the aza-
Diels-Alder reaction with most derived from a range of Lewis acidic metals and a diverse
array of chiral ligands. However, there is still room for improvement via either continued
development of existing catalyst systems or the development of a new catalyst system.
As can be seen from the history of the catalytic asymmetric aza-Diels-Alder reaction,
summarized above, BINOL and its many derivatives have'been key to many successful
catalytic systems. -

The C2 symmetric ligand BINOL is very popular and has been shown to be very
successful in many areas of asymmetric catalysis” 4° beyond the aza-DieIs-Alder
reaction. The way BINOL ligand function is through the coordination or binding of the
diol to a metal center. The substrate then coordinates to the metal and the chiral ligand
should preferentially allow the approach of another reactant from one face. However, the
bulk of the space that is asymmetrically discriminated is on the opposite side of the chiral

axis of BINOL as the metal center. The Wulff group developed the vaulted biaryl ligands

VANOL and VAPOL4M6 (Figure 1.6) to surmount this limitation of BINOL.

37

Figure 1.6 02 Symmetric Biaryl Ligands

Linear Biagl Vaulted Biaryls

E E on Ph i on ' Ph E on
. OH Ph on Ph on
3,3, ‘0 ‘00

magma VANOL VAPOL
Vaulted Binaphthol Vaulted Biphenanthrol
129 130

The vaulted nature of VANOL and VAPOL was envisioned to create a chiral pocket
around the coordinated substrate that would aid in the facial selectivity of reactibn.
Before the present work on the aza-Diels-Alder reaction, the success of these ligands was
limited to the Diels-Alder reaction,4749 imino-aldol reactionf0 Baeyer Villager reaction,“
and the aziridination reaction.”57

For each of these reactions, a variety of Lewis acids were used including
zirconium, aluminum, and boron. In the imino-aldol reaction50 (Scheme 1.34), a catalyst

Scheme 1.34'Imino-Aldol Reaction

R3
Zr(OiPr)4/iPrOH (135, 0.5 to 20 mol%) OH
R2 E

R2
. O + > _ :OTMS (S)-VAPOL(130,1.1to44mol%)A
R1 AN R3 OMe NM1(1.2 equiv.) NH 0
on

_ 25°C ~ RIVLOW
‘ 85-100% 'ld
R1 = Aryl (131) 134 y”

93-99.8%
R2 = H or Me (132) 66 R1 = Aryl (136)

R3 = H or Me (133) R2 = H or Me (137)
‘ R3 = H or Me (138)

 

38

similar to those described for Kobayashi’s aza-Diels-Alder work was employed using
VAPOL. The utility of VANOL for this reaction has not been studied.
Scheme 1.35 Diels-Alder Reaction and Baeyer Villager Reaction
Diels-Alder Reaction

CH0 + Q CHZClz ‘ /
=< —79 to --82 °C 7 CH0
139 140

Catalyst

 

141
Catalyst: EtzAlCl + Ligand

 

Ligand . exo/endo % ee cxo 141 yield 141 (endo + exo)
(S)—VAPOL (130) 98:2 , 97.8 100
(S)-VANOL (129) 92:8 17 85
(R)-BINOL (59) ??? ??? ???

Baeyer Villager Reaction

 

o R2AlCl, VANOL (129) 0
Jj (20 mol% each) > O
. Cumene hydroperoxide (1.2 equiv),
R toluene, —30°C R
142 R = Ar, Bn, alkyl 143
up to 96% yield
up to 84% ee
14% ee with VAPOL (130)

The Diels-Alder reaction“9 and Baeyer VillagerSl reactions (Scheme 1.35) both
employed RzAlCl as the Lewis acid. For the Diels-Alder reaction, both VANOL (129)
and VAPOL (130) were screened and it was found that VAPOL was clearly the superior
ligand for this transformation. The reaction with VAPOL gave 98% ee with quantitative
yield whereas VANOL only gave 17% ee and a slightly worse exo to endo selectivity. _
Forthe Baeyer Villager reaction, upon initial screening, VAPOL gave only 14% ee
whereas VANOL gave 80% ee. With further optimization, the VANOL catalyst was able

to provide yields of up to 96% and the maximum ee observed was 84%.

39

The other reaction that was found to be very successful using VANOL and
VAPOL derived catalyst was the aziridination reactions“7 (Scheme 1.36). For this

Scheme 1.36 Aziridination Reaction

 

Ph Ph
~ Ph Ph
Ph
A O (S)~VAPOL-B I Ph/k NH Ph/l\ NH .
> - + +
t i Ph+ (“\OEt ( 5 H )\,/COZEI R JyCOZEt
R H N2 R COZEt R H
144 145 cis-l46 147 148

reaction, the ligand (was combined with a boron species to prepare the chiral Lewis acid.
In the initial report”, it was discovered that the benzhydryl protecting group on the
nitrogen in imine 144 was important for asymmetric induction to be realized in any
substantial amount. VAPOL was combined with BH3-THF to prepare the catalyst and the
reaction proved to be very successful. For this catalyst, the cis/trans selectivity was
found to be 3:1 to >50:1 for a range of substrates with ee’s all in the 90’s for the cis
isomer. The reaction of imines with ethyl diazoacetate are known to give enamine side
products, however the yield of the enamines was always less than or equal to 15% with
the VANOL and VAPOL catalysis.

It was later discovered that the BH3-THF slowly decomposes to produce
significant amounts of tributylborate. Subsequent work56 revealed that catalysts derived
from B(OPh)3 are superior to those from the BH3-THF complex.

Upon screening many different imines 144, that were prepared from aldehydes
and aminodiphenylmethane, it was surprising to find that for this reaction that there is not
much difference between the VAPOL and VANOL catalysts in terms of selectivity or

turnovers (Scheme 1.37)“.

Scheme 1.37 Comparison of VAPOL and VANOL in the Aziridination Reaction

 

 

 

Ph Ph
Ph Ph
Ph 0 . Y 1311* NH PhANH
N kPh (S)-Ligand (10 mol%)> N + + CO Et
/“\ + HLOEt B(OPh)3 (30 mol%) A H/ngOzEt R/Kl/ 2
R H N2 CHZCIZ, RT R COzEt . R H 7
144 145 _ cis-146 147 g 148
. Yield ee 146 . .
. reaction . . . Yield Yield
Entry R Ligand time (h) oral)? 0:736 01:3an 147 (%) 148 (%)
Ph (144:!) VAPOL 48 77 95 >50:1 4.4 1.7
' Ph (144a) VANOL 0.5 85 96 >50:1 ‘ 3.3 1
- p-BrC6H4 (144b) VAPOL 5 91 98 >50:1 3.4 2.2
p-BrC6H4(144b) VANOL ‘ 1 85 98 >50:1 8 3
o-OMeC6H4 (144e) VAPOL 14 69 94 40: l 1 1.1 7.1
o-OMeC6H4 (l44c) VANOL 16 65 91 >50:1 6 4
3,4—(OAC)2C6H3 (144d) VAPOL 20 85 >50:1 <1 <1

3,4~(OAc)2C6H3 (l44d) VANOL 5 83 >50:1 <1 <1
2-furyl (144e) VANOL 16 55 >50:1 <1 <1
l-naphthyl (1441’) VAPOL 12 87 >50: 1 3.8 0.3

96
97
93
92
n-propyl(l44g) VAPOL 20 54 91 >50:1 8.3 9.2
90
91
97
94

5:3\ooo\ioxv-AuN-—

n-propyl (144g) VANOL 5 60 >50:1 6.7 9.2
13 terr-butyl(144h) VAPOL 12 78 40:1 <1 <1
14 tert-butyl (l44h) VANOL 5 77 >50:1 <1 <1
15 cyclohexyl (1441) VAPOL 8 74 38:1 <1 <1
Although the structures of VAPOL and VANOL are very similar with the only
difference being' the extra fused benzene ring in the phenanthrene unit of VAPOL, their
utility as ligands has been shown to be quite different, and somewhat unpredictable. It
was found that VAPOL was significantly better in the Diels-Alder reaction but in the
Baeyer Villager reaction VANOL was much better. In the aziridination reaction, the two
catalysts were very similar in their effiency, however VANOL was able to produce
slightly better enantioselectivity in some cases.
The inspiration for the azaeDiels-Alder reaction stemmed from the results

observed in the aziridination. Success for the aziridination was observed when studying

41

the imines prepared from benzaldehyde and aminodiphenylmethane using boron as the
Lewis acid in conjunction with the chiral ligands VANOL and VAPOL. The study of the
aza-Diels—Alder reaction was then undertaken for this class of imines using the vaulted

ligands in conjunction with metal Lewis acids as the catalyst.

42

CHAPTER 2

OPTIMIZATION OF THE AZA-DIELS—ALDER REACTION

As discussed in the previous chapter, the aza-Diels-Alder reaction has gained

‘5' 58'“. Since the first example of the catalytic

significant attention in the past 15 years
asymmetric Lewis acid catalyzed example in 1992,19 many groups have contributed to the
field. The main problem. with Yamamoto’s work was that he was, unable to get the
reaction to turnover and hence needed a stoicheometric amount of the catalyst. As is the
case with many asymmetric reactions, the metals and/or ligands can be very expensive.
In this respect, it is desirable in any case, to be able to obtain the highest possible turn
over numbers (TONS) and since 1992, contributions have been made from groups all
around the world towards finding a solution to this problem. In addition to increasing
TONS, it is desirable to develop high yielding reactions, which produce high asymmetric
inductiOn. With the appropriate combination of protecting group on the nitrogen, and.
catalyst, it has been shown that high yields and high asymmetric induction can be
afforded with good TONS as well. The bar has been set high for this reaction class as
both yields and ee’s have been achieved near 100% and the best TONS achieved being
. 780.32 The TON of 780 achieved by Hoveyda was very high and uses an exceptional

Silver catalyst system, but most other examples produced TONS of 5 to 20 with catalyst

loadings around 5-20 mol%. In order to have a reaction that is on par with systems in the

43

literature, it is important to try and obtain numbers equal to or higher than those in
systems previously reported.

When beginning to think about developing a new reaction, several items need to
be considered in reaction optimization. For a catalytic asymmetric reaction, the first, and
probably most important item is which catalyst is going to be employed. The catalyst can
either be developed specifically for the type of substrate that is going to be studied or a
substrate has to be used which can be predicted to interact favorably with the catalyst. In
the work done by Kobayashi?“ for example, he used zirconium as the metal. In this
case, he used the phenolic protecting group on the nitrogen of the imine so that the
substrate could occupy two free sites on the metal. The opposite scenario was seen in the
Wulff group in the development of the aziridination.“ ”‘57 The catalyst was developed
first and studies then indicated that some version of a benzhydryl protecting group was
needed on the imine nitrogen to induce a favorable interaction of the imine with the
catalyst. In any event, the catalyst and substrate interaction is a very important aspect of
these reactions. It is this interaction that will ultimately activate the substrate making it
more reactive in asymmetric reactions. It is also important that the lowest energy (or
most favorable) substrate-catalyst complex allow the second reagent to approach the
coordinated substrate only from one face. One Other issue for reactions of imines is that
the resulting products are in many cases amines, which brings up the importance of
catalyst loading. As discussed in Chapter 1, the Sp3 nitrogen in amines is more basic
than the sp2 nitrogen of the imine, which oftentimes makes turnover difficult'to achieve
causing the need for higher catalyst loading. When using metals with several

coordination sights, it can also be valuable to think about different additives and how

they may help or hurt the reaction. AS mentioned before, the desirable situation is that
where the catalyst-substrate coordination is such that the reaction will happen
preferentially from one face over the other. Although there are exceptions, it is often
observed that reactions will give increased selectivity at lower temperatures. This is
presumably due to the fact that at colder temperatures, the rate of the reaction slows
down, leading to a greater differentiation between the opposing transition States. Another
variable, which can prove to be very important, is the solvent or solvent combination that
is used in the reaction. Solubility is probably the first issue to consider. If the reagents
were insoluble in the solvent, the reaction would only be able to occur at the interface of
the phases, which makes the reaction much less efficient. In addition tosolubility, the
type of solvent used can play an important role in the reaction. In cases where the metal
catalyst has sufficient coordination sites, coordinating solvents could potentially be
favorable for the reaction, but for lower coordination metals, the presence of a
coordinating solvent could compete with the substrate for binding to the metal center.
For boron or aluminum catalysts, there is only one binding site available so a
coordinating solvent such as THF, ether, acetonitrile, or ethanol, etc. would be expected
to be detrimental to the outcome of the reaction. Concentration can also influence the
outcome of a reaction. It is expected that when. the concentration is increased, the
reaction rates would increase and thus potentially produce a higher yield of the desired
product in less reaction time. However, at increased concentration, solubility could
present a problem especially when reactions are run at low temperatures. A few other
items to keep in mind when developing conditions for a new reaction are stoichiometry of

reagents, how the reagents are added, the order in which reagents are added, and lastly,

45

reaction time. All of these items need to be considered when developing a new system
for a reaction and each can effect the yield or selectivity either in a positive or negative
way depending on the variable.

The VAPOL (130) and VANOL (129) ligands were developed in the Wulff group

and have been previously used as ligands in several catalytic asymmetric reactions

r4749 52. 55-

including Diels-Aide , imino-aldol-reaction”, Baeyer Villager”, and aziridination
57 reactions. In addition to trying to add to the collection of catalytic asymmetric aza-
Diels-Alder reactions present in the literature, it was also desirable to find another
reaction where one or both of the ligands developed in our group could be utilized.
When this thesis work was begun in 2001, one of the goals was to determine if VAPOL
(130) and VANOL (129) could be used as efficient ligands for the catalytic aSymmetric
' aza-Diels-Alder reaction. .

Preliminary results, which were obtained by Jon Antilla just before he graduated
in 2000, indicated that it would be worthwhile to explore this reaction (see appendix).
2.1 The TBS Version of Danishefsky’s Diene

The preliminary studies done by Antilla, along with previous observations using-
Boron—VAPOL or boron-VANOL catalysts, indicated that the TBS version (149) of

Danishefsky’s diene (Figure 2.1) was going to produce better results than the original

diene with the TMS group group (31). One could expect that the bulkier diene Should be

Figure 2.1 Danishefsky’s dienes

I/l<
Si\

| /
MOMe MOMe
31 149
Danishefsky's diene TBS version of
Danishefsky's diene

more selective for one face of the imine Simply due to steric interactions between the
diene and the ligand. Jon Antilla only used a catalyst prepared from BH3-THF and
VAPOL for his initial screening of the aza-Diels-Alder reaction. However, since the
catalyst prepared using a 3:1 ratio of B(OPh)3 to (S)-VAPOL was found to be most
effective for the aziridination reaction than was the catalyst prepared from BH3-THF and
VAPOL, the former was used for most of the initial attempts of the aza-Diels-Alder
reaction. The imine (150) prepared from benzaldehyde (152) and aminodiphenylmethane
(153, Bh-NHZ) was used exclusively for the optimization of the aza-DielS-Alder reaction.

Several reactions were run using 2.0 equivalents of the TBS version (149) of

Danishefsky’s diene in the presence of 4A molecular sieves (Scheme 2.1) where the

47

Scheme 2.1 aza-Diels-Alder in the Prescence of 4A Molecular Sieves

 

   

\N/kph + L B(0Ph)3NAPOL
/ OMC— 4A M.S.,CH2Cl2

 

 

 

150 149 temperature A
1.0 equiv 2,0 equiv, reaction time Ph Ph
round bottom flask 151
Entry “(11:30“ B(OPh)3 VAPOL temperature reaction concentration conversion

time (h) (mol%) (mol%) ( C) time (h) (M) (%)

1 3 30 10 —50 to RT 41 0.22 44

2 3 30 10 0 22 0.20 10

3 4 30 10 ——40 36 0.20 20
4 10 30 10 ~20 36 0.14 17

5 3 10 10 —45 25 0.22 21

 

temperature, reaction time and rate of addition of the diene were varied. At this time, the
isolated yield and enantiomeric excess were not measured and only the conversion was
measured by looking at the crude 'H NMR. The ratio of the final product and the starting
imine was measured to calculate the conversion. When the diene was added over 3
hours and the temperature was allowed to warrnfrom -—50°C to room temperature and
stir for 41 hours (entry 1), the conversion was 44%. When the reaction was run at 0°C or
—40°C (entries 2 and 3) for the entire reaction, the conversion dropped to 10 and 20%
respectively. If the reaction was run at --20°C and the diene was added over 10 hours
(entry 4), the conversion was only 17%. One other reaction was performed in the
presence of molecular Sieves using the catalyst prepared from 10 mol% B(OPh)3 and 10

mol% VAPOL (entry 5). When this reaction was stirred for 25 hours at —45°C, only

48'

21% conversion was observed. All of these reactions were very slow and the best
conversion was achieved when the reaction was allowed to stir for 41 hours.

Next, a few reactions were accomplished where molecular Sieves were not put in
the reaction (Scheme 2.2). 25% conversion was achieved (entry 1) when the reaction was

Scheme 2.2 aza-Diels-Alder Reaction Without 4A Molecular Sieves

 

 

 

 

j: S',J< morn)3 (30 mol%)
\N Ph + O \ VAPOL (10 mol%) _
MOMC CH2C12
150 149 ' temperature
1.0 equiv 2.0 equiv. reaction time
‘ round bottom flask 151
En t 149 addition tern era ture (o C) reaction time concentration conversion
“y time (h) p (h) (M) (%)
1 3.5 0 to RT 43 0.25 25
2 2.5 —40 23 0.33 20
3 ' 3 ——40 19 0.2 40

 

allowed to warm from 0°C to room temperature immediately after the addition of the
diene was complete. If the reaction was allowed to stir at —-40°C, at slightly different
concentrations, for about one day (entries 2 and 3), it was found that the Slightly less
concentrated (0.2 M with respect to imine) reaction was better, resulting in 40%
conversion.

Since the reactions all seemed to be very sluggish at this point, a series of
experiments were performed where the reaction temperature was held constant at 0°C

(Scheme 2.3). In this set of experiments, the amount of diene, addition time of diene,

49

Scheme 2.3 Effect Diene Equivalents and Addition Time in the aza-Diels-Alder

 

Reaction
Ph Sli B(OPh)3 (30 mol%)
N Ph + VAPOL (10 mol%) _
Mom CH2C12,0°C .

150 149
1.0 equiv

reaction time
round bottom flask

Ph
151

Ph

 

149 149 addition reaction concentration yield 151 cc 151 conversion

 

Em” (equiv) time (h) time (h) (M) (%) (%) (%)
1 2 o 24 0.2 ND ND 21
2 1.5 0 6 0.5 ND ND 10
3 2 2.5 45 0.45 ND ND 20
4 1.53 2.5 21 0.2 ND ND 12
5 1.58. 3 24 0.2 ND ND 59
6 1 3 24 0.29 36 68 ND
7 1.5 4.5 24 0.5 ND ND 17
8 2 ‘ 12 17 0.2 29.8 66 30
9 2 21 21 0.29 ND ND 24
10 , 2 21 45 0.22 ND ND 33

 

concentration, and reaction time were all varied. When the diene was added all at once

(entries 1 and 2) it was observed that only 21% conversion could be achieved even when

2 equivalents of the diene were added and the reaction allowed to stir for one day.

Increasing the additiontime to 2.5 hours (entry 3 and 4) was not that helpful. Using 1.58

equivalents of the diene and stirring for 21 hours gave only about 12% conversion (entry

4) and doubling the cOncentratiori and allowing the reaction to stir for 45 hours still only

' gave 20% conversion (entry 3). Two reactions were done where the diene was added

over 3 hours, one using 1.58 equivalents of the diene and the other using 1 equivalent of

the diene (entries 5 and 6 respectively). After one day of stirring, the one with 1.58

50

equivalents Showed 59% conversion. For the reaction where 1.0 equivalents of the diene
was used, the conversion was not measured, but the product was isolated and the
selectivity measured. This reaction resulted in 36% yield and an enantiomeric excess of
68% was observed. A 17% conversion could be achieved if 1.5 equivalents of diene
were added over 4.5 hours and the reaction allowed to stir for 24 hours (entry 7). One
reaction was performed where the conversion, isolated yield, and ee was measured (entry
8). The isolated yield was 30% with 66% ee observed; The reaction was again slow, but
it was reassuring to see that the yield was corresponding nicely to the conversion. The
conversion in this reaction was 30%, so this seemed to indicate that the only reaction that
the imine was undergoing was the desired aza-DielS-Alder reaction. If the amount of
conversion was high and the isolated yield was much lower, then it would indicate that
some competing reaction was taking place, but this was not observed. Lastly, two other
experiments were done where the 2.0 equivalents of the diene were added over 21 hours.
The reaction was stopped immediately after the addition was complete (entry 9), and only
24% conversion was observed. If allowed to stir an additional 24 hours, the conversion
could only be increased to 33%.

It can'be concluded from these sets of experiments that the aza-Diels-Alder
reaction with the TBS version (149) of Danishefsky’s diene was very Slow. The highest
conversion observed was 59% and the rest of these reactions showed conversions that
were significantly lower.

Some time was spent trying to synthesize a diene that contained even more steric

bulk, thinking that the increase in size would possibly help facilitate this reaction.

51

Meanwhile, while attempting to prepare the bulkier diene, Danishefsky’s diene (31) was
also prepared and one reaction was attempted (Scheme 2.4). The approach of the diene

Scheme 2.4 First Attempt Using Danishefsky’s Diene

 

Ph S|i/ B(OPh)3 (30 mol%) 0
\ A o’ \
N Ph + (S)-VAPOL (10 mol%) _ I
Now CH2Cl2, —40°c N
150 31 20h )\
1.0 equiv 2.0 equiv. 77% yield Ph Ph
added over 3 hours 30% 06 151

31, due to its less sterically hindered nature, allowed for an easier approach t0wards the
dienophile and the reaction worked well! Using this diene, the optimization of the aza-
Diels-Alder reaction using the benzhydryl—imine and Danishefsky’s diene was begun.
2.2 Danishefsky’s Diene
2.2.1 VAPOL vs. VANOL vs. BINOL .

As mentioned before it was not clear which ligand would perform the best for the
aza-Diels-Alder reaction. The first item to be investigated was a comparison of VANOL
(129), VAPOL (130), and BINOL (59) ligands used in catalysts prepared from B(OPh)3

(Scheme 2.5). Two reactions were done in toluene (entries 1 and 2) where the catalyst

52

Scheme 2.5 Comparison of VAN OL, VAPOL, and BINOL

 

     

 

Ph SI / 0
/ l
\N * Ph + O \ B(Oph)3 / Ligand ‘
MOMe solvent 7
150 3] temperature 1
1.0 equiv 2.0 equiv. reaction time Ph Ph
added over 3.0 h 151
. Yield ee
B(OPh)3 -. temperature reaction
Entry (mol%) Ligand (mol%) (0 C) Solvent flask type time (h) (15:71) (1;)1)

 

la 30 (S)-VANOL(10) —48 Toluene rbf 24 <5 ND

2" 3O (S)—VAPOL(10) —45 Toluene rbf 48 78 85

3° 100 (S)-VANOL(10) -45 CH2???“ coiLd 24 41 56
4b 100 (S)-VAPOL (10) —45 CHE???“ COILd 24 94 90
5b 100 (R)-BINOL(10) —45 “£912?“ COILd 24 26 23

 

a) 0.25 M with'respect to imine
b) 0.2 M with respect to imine

c) 0.1 M with respect to imine

d) For discussion and picture of the COIL flask see Scheme 2.21 and Figure 2.4
was prepared from 10 mol% VANOL or VAPOL and 30 mol% B(OPh)3. The reaction
with the VANOL catalyst gave little or no product while the VAPOL catalyst produced a
78% yield. It is easy to see from these results that VAPOL was the superior ligand when
_ toluene was used as the solvent. These same two reactions were repeated in the same
way except 100 mol% B(OPh)3 was used to prepare the catalyst and a 1:1 mixture of
CH2Cl2 /toluene was used as the solvent. The results with VAPOL (entry 4) were 94%
yield and 90% ee. These results are significantly higher than those with VANOL (entry

3), which only gave 41% yield and 56% ee. Another reaction was attempted using

BINOL as the ligand and it proved to be even worse than VANOL resulting in 26% yield

53

and 23% ee: The results of this series of experiments showed clearly that VAPOL was in
fact the best ligand to use in pursuit of the optimal conditiOnS of the reaction. With the
exception of the temperature studies, all the reactions discussed in this chapter will
involve the use of VAPOL as the chiral ligand.

Another important variable that was found to be an important consideration for
this reaction was the purity of VAPOL used for the preparation of the catalyst. A few
reactions were carried out using VAPOL sources with different purities (Scheme 2.6),

Scheme 2.6 Effects of the Purity of VAPOL

 

‘ 0
Ph Sli/ B(OPh)3 (30 mol%)
\ A o’ \

N ph + VAPOL (10 mol%) _ |

MOMe toluene, ~45°C N
150 31 reaction time A

1.0 equiv 2.0 equiv. round bottom flask Ph Ph
added over 3.0 h 151

 

Entry VAPOL (condition) reaction time (b) y mm 151 cc 151

 

(%) (%)
1“ Yellow” 48 43 86
2a Light Yellow“ 48 54 87
3a Whited 20 35 86
4a White° 20 45 89

 

a) 0.2 M reactions with respect to imine -

b) VAPOL from an old vial that was yellow in color was used

c) Light yellow VAPOL that had been purified by running two columns was used
,d) Very white VAPOL that TLC analysis showed one extra minor spot was used
e) Very white VAPOL that TLC analysis showed only one spot was used ‘

the purity of VAPOL was indeed very important to the outcome of these reactions. In
order to determine this, a vial of VAPOL that had been sitting around for an
undetermined amount of time was used in the preparation of the catalyst (entry 1). When
the catalyst was prepared from 30 mol% B(OPh)3 and 10 mol% of that VAPOL and the

reaction allowed to run for two days in toluene, 43% yield and 86% ee was obtained.

54

Two columns were then run to purify the VAPOL to a light yellow color. The reaction
with this VAPOL (entry 2) resulted in 54% yield and 87% ee. The VAPOL was then
purified to whiteness, but two spots were still visible by TLC analysis. This reaction
after 20 hours only gave 35% yield and 86% ee was observed again. Finally the VAPOL
was further purified and pristine white VAPOL was obtained. The reaction using this
VAPOL (entry 4) gave the best results and 89% ee was obtained. This series of
experiments indicated that the purity of the VAPOL has an effect on the selectivity of the
reaction but it is unclear whether the yield is affected because the reaction times were not
the same. It is important to note at this time that later in the course of these studies,
several reactions were repeated at random times using identical conditions and analysis of
the data collected from those experiments showed that the variation of yield was :1:7 and a
variation of ee was :1 (using the optimal conditions the error was 83.5:7% yield and
89:1:1% ee). This indicates that the increase in ee for the experiments conducted in
Scheme 2.6 indeed exceeded experimental error and thus carefully purified, extremely
clean VAPOL was used for the optimization of this reaction.
2.2.2 Temperature Effects

In asymmetric catalysis, temperature can have a significant effect on the outcome
of the reaction. Two temperature studies were accomplished using both VAPOL and
VANOL, where the temperatures were varied from —-70°C to room temperature. The

temperature effect on the reaction using VAPOL was studied first (Scheme 2.7). From

‘55

Scheme 2.7 Temperature Effect Using VAPOL

 

0
Ph 4% 8(0Ph)3 (30 mol%)
\ A o’ \

N Ph + VAPOL (10 mol%) = I

MOMe toluene N
150 31 temperature A

1.0 equiv 2.0 equiv. 22 h Ph Ph
added over 3.0 h round bottom flask 151

 

Temperature yield 151 ee 151

 

(°C) (%) (%)
——70 J 21 91
——60 16 81
——50 27 88
——4o 51 89
——30 40 86
——20 34 8O
-—10 36 85
RJ‘ 21 58

 

the results of the series of temperatures, it was found that the yield reached a maximum at
—40°C. It makes perfect sense that the yield would drop going to lower temperatures
and this can be explained by a decrease in the rate of the reaction at lower temperatures.
If this same argument was used for warmer temperatures, then it would be expected that
warrning the temperature above —40°C would result in increased yields. This effect was
not observed, and in fact the yields dropped significantly as the temperature was warmed
to room temperature. One explanation for this observation could be that at warmer
temperatures some Side reaction could start to occur at a faster rate and effectively reduce
the amount of desired product produced. Another explanation could be that if at warmer
reaction temperatures the stability of the reagents under the reaction conditions were

compromised, then decomposition of the reagents could occur. If this happened, then the

56

yields would drop in this case as well. This explanation is probably the correct one as it
was observed during the course of the reactions at warmer temperatures that the solution
became dark red and material could be observed at the base line in TLC analysis. The
thought was that the reactive diene could be decomposed or possibly polymerize in the
presence of Lewis acid. One reaction was done where only the catalyst and diene were
added to the flask in the absence of imine and the reaction again turned very dark red.
TLC analysis showed only baseline material was present and when purification was
attempted using column chromatography, all attempts to retrieve the material, using any
solvent, were unsuccessful. The dark red material just stuck at the top of the column and
could not be eluted. In conclusion, the best compromise between the rate of reaction and
minimal decomposition of the diene was a temperature of ~40°Ciwhere the highest yield
was observed. The other factor to consider when selecting the optimal temperature is its
effect on the selectivity of the reaction. When the reaction was run at room temperature,
58% enantiomeric excess was observed. This is only a moderate selectivity and it was
desired to find a temperature where the ee would be as high as possible. It was found that
going to lower temperatures, the ee could be increased to 86% and 89% at -30°C and
-—-40°C respectively. Unfortunately, going to lower temperatures seemed to have little
effect on- the ee. From this study it can be concluded that the optimal reaction
temperature was ——40°C, however after running this reaction many times Over the past
few years, and the observation of temperature fluctuations above and below this mark,
the optimal reaction temperature was decided to be -45°C.

VANOL was also subjected to the same temperature study (Scheme 2.8). Based

57

Scheme 2.8 VANOL Temperature Study

 

Ph 3 l / B O
SI (OPh)3 (30 mol%)
\ A o’ \

N Ph + VANOL (10 mol%) A l

MOMe toluene N
150 31 temperature A

1.0 equiv 2.0 equiv. 21 11 Ph Ph
added over 3.0 h round bottom flask 151

 

temperature (°C) yield 151 (%) ee 151 (%)

 

—70 6 25
—60 8 0
——50 8 9
—4O 15 21
—20 15 54
—10 10 24

RT 6 13

 

on the results in Scheme 2.5, it was not unexpected to find that the results were not as
goodhas those from VAPOL. Nonetheless, it revealed interesting differences between
VAPOL and VANOL. This Study indicated that the best yield and ee could be achieved
at —20°C where 15% yield and 54% ee was observed. In a manner similar to that
observed for VAPOL, both higher and lower temperatures had a negative effect on yield
of the reaction. It was interesting to find that the same negative effect was not observed
for the ee. When the temperature was decreased to —40°C, 21% ee was observed and
A when the temperaure was increased to — 10, 24% ee was observed. It is not clear why
this effect is observed, nontheless, it seemed that there was a narrow window of
temperatures around —20°C where the reaction was most efficient. The effect of
temperature on the reactions using VANOL and VAPOL are expressed in the following

plots (Figure 2.2). The bold bar indicates the optimal temperature in each case. Again, it

58

Figure 2.2 Summary of Temperature Effects

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

VAPOL
100 ~
90 i ' I I I I
80 — I I
70 _
5Q; 60 1 I Oyield
:5 50 ~ 8 '
l ee
1? 40 d . o
o
30 a . .
20 . 9 .
10 .
O T I I I I I
-80 -60 -40 -20 0 20 40
temperature (°C)
VANOL
60 n I
50 1 A
A 40 '-
52; 0 yield
g 30 a I
. g I I CC
20 ~ '
10 O O I
O ' O
0 * I r r r r
-80 -60 -40‘ -20 0 20 40
temperature (°C)

 

 

 

was found that VAPOL was indeed the superiorligand and the optimal temperature for

VAPOL was determined to be —45°C.

59

2.2.3. Solvent Effects
AS expected, it was found that the choice of solvent or solvent combination also
played a major role in the outcome of the reaction. Several reactions using different
conditions Were performed for the purpose of providing a comparison of toluene and
.

CHZCl2 as the solvent (Scheme 2.9). The first set of reactions were done using 5 mol%

Scheme 2.9 Toluene Versus CHZCI2 as Solvent

 

Ph l O
\ 1 AK
N Ph + 0 B(OPh)3/VAPOL
MOMe solvent 7 I
, N
150 31 '45 C A .
1.0 equiv 2.0 equiv. ”3090091116 Ph Ph

added in 3mL solvent 151

 

E B(OPh)3 VAPOL 31 addition reaction time yield 151 ee 151
ntry solvent

 

 

(mol%) . (mol%) time (h) (h) (%) (%)
1 CH2C12 100 5 3a 24 95 76
2 Toluene 100 5 3a 24 60 90
3 CH,Ci2 100 . 1 3b 49 78 47
4 Toluene 100 . 1 3b ' 49 53 80
5 CH,Ci2 30 10 3b 47 81 73
6 Toluene 30 10 3b 48 78 85
J 7 CH,C12 30 10 0b 47 82 70
8 Toluene 30 10 0b 47 54 88
a) A cooling addition coil flask was used (For discussion and picture of the flask see Scheme 2.21 and
Figure 2.4). '

b) A traditional 25 mL round bottom flask was used.

VAPOL and 100 mol% B(OPh)3 as the catalyst (entries 1 and 2). When CH2C12 was
uSed, the reaction gave 95%. yield and 76% ee, whereas when toluene was used, the
reaction produced 60% yield and 90% ee. It seemed from this experiment that the
reaction was slower in toluene but the selectivity was much better. In another set of
experiments where 1 mOl%VAPOL and 100 mol% B(OPh)3 were used (entries 3 and 4),
the same effect was‘observed. With this low level of VAPOL loading (1 mol%), the

60

difference in the selectivity was even greater when switching from toluene to CHZCIZ, as
the ee dropped from 80% to 47%. Two additional sets of experiments were carried out
with 30 mol% B(OPh)3 and 10 mol% VAPOL (entries 5-8). These results also examine
the effect of the diene addition time, which will be discussed in much more detail later in
the chapter. When the diene was added over 3 hours and CHZCl2 was used as the solvent,
the reaction produced 81% yield and 73% ee in two days (entry 5). On the contrary, the
reaction in toluene gave 78% yield and 85% ee (entry 6). Again, the increase in ee when
switching to toluene was observed. Switching from CH2Cl2 to toluene, when the diene
was added all at one time Showed an increase from 70% cc to 88% cc and a decrease in
yield from 82% to 54%. In this Series of experiments, all data was in agreement that
toluene was a better solvent for creating the desired selectivity and CHZCl2 was the better
solvent for enhancing the rate of turnover of the reaction.

THF and CCl4 were also examined as solvents for this reaction (Scheme 2.10) and

Scheme 2.10 CCl4 and THF as the Solvent

 

Ph I /
\ A A
N Ph + O B(OPh)3NAPOL _
MOMe . solvent ' V
- 150 31 temperature
1.0 equiv 2,0 equiv. reaction time
added Over 3.0 h

 

in 3 mL solvent'

 

 

B(OPh)3 ,VAPOL ‘ temperature reaction yield 151 cc 151
Em" (mol%) (mol%) 30mm (°C) time (h) (%) (%)
1a .150 ‘ 2.5 CCI, -—-45 . 24 - 38 75
2b 30 10 THF —50 48 43 20

 

a) A cooling addition coil flask was used (For discussion and picture of the flask
see Scheme 2.21 and Figure 2.4).
b) A traditional 25 mL round bottom flask was used.

61

each proved to be a much poorer choice for solvent in this reaction. When CCI4 was
used, the reaction gave 38% yield and 75% ee (entry 1). It is not so surprising that this
reaction failed to produce good results. In fact this was a poorly designed experiment
because the melting point of CCl4 is ——23°C. When the reaction was attempted at
—-45°C, the solvent did in fact freeze which cannot be good for the reaction. Also, at the
end of 24 hours, the reaction was allowed to warm to temperature where the solvent
would melt and stirred about 30 minutes before the workup was accomplished. This
Wanner temperature cOuld also have affected the outcome of the reaction. When THF
was used, it was also not surprising to find that this solvent produced inferior results.
THF contains an oxygen which could, and probably does compete with the imine for
binding to the catalyst. This competition for binding would at the very least lead to the .
expectation that the reaction would slow down. This was indeed observed and the
reaction only produced 43% yield after 48 hours reaction time. In addition to the lower
yield, the reaction was much less selective when THF was used as the solvent and
produced only 20% ee.

Carbon tetrachloride is very expensive and very toxic and is therefore not a
conventional solvent that is used very often in chemical laboratories. During the course
of these studies, some issues occurred for the reproducibility of the aziridination reaction,
and some previous results could not be duplicated. In order to obtain ee’s that were as
high as those seen previously, it was discovered that CC]4 was helpful in obtaining higher
ee’ for this reaction. It was for these reasons, that the carbon tetrachloride reaction above
was attempted at all. The desire to determine the effect of CCl4 for the aza-Diels-Alder

reaction prompted the use of solvent combinations using CCl4 and CHZClz. In addition to

62

this, Since the reactions in toluene gave good selectivity and CHZCI2 gave good yields, it

was also desired to try CHzCl2 and toluene mixtures as well. The discussion will now

turn to the studies of different solvent combinations (Scheme 2.11). It was determined

Scheme 2.11 Effects of Solvent Combinations

 

 

 

Ph SI / o
/ l
\NX ph ., O \ B(OPh)3/VAPOL > I
/ OMe solvent N
150 31 —45°C A
1.0 equiv 2.0 equiv. reaction time Ph Ph
added over 3.0 h 151
Ent 3(0Ph)3VAPOL solvent reaction Yield 151 ee 151 Conc.

” (mol%) (mol%) time (h) (%) (%) (M)
1: 100 5 CHzglzs/ICC“ 24 67 90 0.11
2“ 100 5 $281,250“ 24 83 94 0.2
38° 100 10 CHE???“ 24.5 86 84 0.2
4‘:d 100 10 $23312? 0' 24.5 94 81 0.2
5'“ 100 10 CH???“ 22.5 90 86 0.2
6” 100 10 CH???“ 24 9o 94 0.2

 

a) A cooling addition coil flask was used (For discussion and picture of the flask
see Scheme 2.2] and Figure 2.4).
b) A traditional 25 mL round bottom flask was used.
c) The catalyst was transferred in 2 mL DCM and the diene added in 3 mL toluene.
d) The catalyst was transferred in 2 mL toluene and the
diene added in 3 mL DCM.
e) The catslyst was transferred and the diene added in DCM/toluene (1:1).

from this study that if 1:5 ratio of CC]4 to CHzCl2 was cooled to —45°C, that the solution
would not freeze. The aza-Diels-Alder reaction was then attempted with this solvent

(entry 1). For this solvent it was found that twice as much solvent was required to

63

maintain solubility of the reagents and catalyst. The outcome of this reaction was really
good, resulting in 67% yield and 90% ee. Some precipitate was observed in the flask at
that reaction temperature which indicated that solubility at -45°C was possibly still an
issue and could be the reason for the modest yield produced in that reaction. One
reaction was then attempted where an increased amount of CH2C12 was used (entry 2).
When the ratio of CHZCl2 to carbon tetrachloride was changed to 1:2 (i.e. ~1.75 mL
CHZCl2 and 3.25 mL CCl4 as opposed to ~08 mL CHZCI2 and 4.2 mL CCl4), the reaction
was more efficient giving 83% yield and 94% ee. This was the best selectivity seen for
this reaction throughout the course of these studies but was not chosen as the optimal
conditions Simply due to the toxic nature and expense of CC],.

Now, thinking back, to the interesting results from the toluene versus CH2C12
studies, it was thought that high yields and selectivity could be achieved if a mixture of
these solvents was used. In order to test this hypothesis, several reactions were done
; using a combination of CHZCl2 and toluene. One reaction was attempted where the
catalyst was transferred to the imine in 2 mL CHZCI2 and upon cooling to —45°C, the
diene was then added in 3 mL toluene (entry 3). This reaction gave 86% yield and 84%
ee. If the solvents were switched in this reaction where the catalyst was transferred in
' tOIuene and the diene added in CH2CI2 (entry 4) the reaction produced 94% yield and
81% ee. Again, a similar effect of the two solvents was observed. When more CHzCl2
was present,the yield went up and the ee down and when more toluene was present, the
yield went down and the ee went up. One attempt was then done where a 1:1 ratio of '
toluene and CHZCl2 was used for both the catalyst transfer and the addition of the diene.

When this was attempted, it was found that a good cc and yield could be achieved as the

64

reaction produced 90% yield and 86% ee. When that same reaction was performed using
special coil addition glassware (discussed later in the chapter) (entry 6), it was found that
the ee could be increased to 90% and 94% yield was obtained. With these results
observed, tOluene/CHZCI2 (1:1) was then decided to be the optimal solvent for this aza-
Diels-Alder reaction. .
2.2.4 Different Lewis Acid Sources -

Before moving on to discuss the effects of catalyst loading and various ratios of
B(OPh)3 to VAPOL, it is important to note that several other Lewis acid sources were

studied and their affects on the outcome of the reactions were recorded (Scheme 2.12).

65

Scheme 2.12 Different Lewis Acid Sources -

 

 

 

Ph OSI / 0
l
\ 0’ \ Le . .
N Ph wrs ACid/VAPOL
©/\ + M solvent —> I
150 31 temperature /I\
1.0 equiv 2.0 equiv. reaction time Ph Ph .
added over 3.0 h round bottom flask 151
. yield ee
. . VAPOL temperature reaction Conc.

Entry Lewrs acid (mol%) solvent 0 . 151 151

(mol%) ( C) time (h) (M) (%) (%)

l Al(Me)3 (30) 10 Toluene —45 19 0.2 0 ND

2 Al(OiPr)3 (30) 10 Toluene —45 46.5 0.2 0 ND

3 B(OnBu)3 (30) 10 CHZCI2 —40 71 0.2 41 ND

4 B(OPh)3 (30) 10 CH2C12 —40 47 0.2 81 73

5 B(OPh)3 (30) 10 Toluene —40 48 0.2 78 85

6 BIO‘Z’6I'§III‘IMC'PII)3 10 CH2C12 —40 26 0.2 69 75

7 B(O—4—F-Ph)3 (30) 10 CHZCl2 —45 20 0.2 45 58

8 B(O-4—F-Ph)3 (30) 10 CHZCI2 —78 22.5 0.2 28 52

9 B(O-4--Me-Ph)3 (30) 10 CHZCI2 —40 22 0.2 80 88

10 B(O-4—Me-Ph)3 (30) 10 Toluene —40 20 0.2 68 90

11 B(O—4—OMe-Ph)3 (30) . 10 Toluene —45 23.5 0.2 45 84
12 Yb(OTf)3 (20) . 10 CHEEIIZIIOI —45 22 0.2 76 10* '
l3 Yb(OTf)3 (10) 20 Toluene ——45 24 0.125 24 10 ‘

 

Two attempts were made with catalysts prepared from aluminum Lewis acids.

One catalyst was prepared in the normal fashion by heating AlMe3 (30 mol%) and

VAPOL (10 mol%) in CHZCl2 to 55°C for one hour and then heating under high vacuum

for 30 minutes. Another catalyst was prepared in the same way from Al(OiPr)3 (entries 1

and 2). Neither of these reactions with these catalysts produced any Diels-Alder adduct.

In the first case, the excess AlMe3 was probably not necessary to drive catalyst formation

66

 

. to completion because aluminum is so oxophilic and the methyl groups were likely
readily protonated by the acidic oxygens of VAPOL. AlMe3 is a strong enough Lewis
acid that with excess around, the diene may decompose at a faster rate than the desired
azaeDiels-Alder reaction. Although the triisopropoxy aluminum is not as strong of a
Lewis acid, a Similar situation could have occurred, explaining the failure to produce the
product. It is entirely possible if a 1:1 ratio of aluminumrto VAPOL had been used, a
more favorable result would have been observed, but this investigation never occurred. I

In addition to studying aluminum catalysts, a series of experiments were done
using four different borate species. The slightly less Lewis acidic tributyl borate
(purchased from Aldrich) was used to prepare a catalyst first‘(entry 3) and its reaction
turned out to be much slower compared to the B(OPh)3 catalysts. In CHZCI2 at —40°C
this reaction only gave 41% yield in 71 hours (entry 3). Due to the slow reaction, the es
was not determined and B(OnBu)3 was not considered further. Next, a more sterically
hindered triarylborate was prepared from 2,6-dimethylphenol and boric acid by
azeotropic distillation. This triarylboratelwas used for the preparation of the catalyst
(entry 6) and the reaction gave 69% yield of 151, but the selectivity was not very good
(75% ee). A triarylborate with an electron withdrawing fluorine in the para-position was
then prepared in similar fashion as the 2,6—dimethylphenol. An initial attempt was made
where the reaction was run at —45°C (entry 7) and this gave 45% yield and 58% ee.
Having an electron-withdrawing group on the aryl group should increase the Lewis acidic
nature of the boron and make the catalyst more active. This argument indicates that the
reaction might be faster and this could account for the drop in enantiomeric excess. On

the other hand, the increased Lewis acidity could also potentially Slow down the rate of

67

3

turn over if the boron was coordinating more strongly to the more basic aza-Diels-Alder
adduct, thus lowering the yield. If the Lewis acid was more acidic and the rate was
faster, then presumably cooling the reaction down would slow the reaction and more
selectivity would be observed. This was not the case as only 52% ee was observed for
the reaction at ——78°C (entry 8). It was interesting to find that by switching to an
electron—donating group (tris-4—methoxyphenylborate, entry 11), the selectivity went back
up and 84% ee was observed. The reaction was done in toluene so the moderate yield of
45% was not unexpected. Two reactions were then done using tris-4-
methylphenylborate, a triarylborate species that is only slightly more electron rich than
B(OPh)3 (entries 9 and 10). One reaction was done in toluene and one in CHZCl2 and
again, the same trend as seen before with these two solvents was observed. When the
reaction was done in CH2C12 the yield was 80% and in toluene the yield was 68% for
reactions that were allowed to stir for about one day. A similar trend was also observed
for the outcome of the ee as well, only with this catalyst, the ee was only 2% less in pure
CHZClz-than in toluene (88% and 90% ee respectively). This is probably the most
intriguing result from the study of different borates because, it would not have been
expected that a borate that similar to B(OPh)3would give such a dramatic increase in ee
using CHZCl2 as the solvent. When this same reaction was done using B(OPh)3, only
73% ee was. observed (entry 4) and with the para-methylphenyl group, the ee was
increased to 88% (entry 9).

The synthesis of racemic mixtures of the aza-Diels-Alder products for the
screening of substrates were accomplished using Yb(OTf)3. The yields of these

reactions were moderate to good so it was curious to find out what would happen if

68

VAPOL were attached to the Ytterbium and the aza-Diels-Alder reaction was attempted.
One reaction was attempted where Yb(OTf)3 (20 mol%) and VAPOL (10 mol%) was
used to prepare the catalyst in the same way as for B(OPh)3 (entry 12). The reaction was ’
quite efficient, giving 76% yield, but the background reaction predominated and only
10% ee was observed. This reaction, interestingly enough, produced the opposite
enantiomer from the VAPOL-B(OPh)3 catalyst. In order to be sure that there was no
Yb(OTf)3 around to facilitate the racemic reaction, a catalyst was prepared uSing 10
mol% Yb(OTf)3 and 20 mol% VAPOL (entry 13). This reaction was done in toluene
under slightly less concentrated conditions and the reaction only gave 24% yield and 10%
ee. The major enantiomer in this case was the same as that produced for thevVAPOL-
B(OPh)3 catalyst.

Although some interesting results were observed from this series of experiments,
the readily available B(OPh)3 was still the desired Lewis acid for this reaction. However,
more exploration of the para-methylphenylborate may be interesting to study for its
. effect on this and other substrates in the aza-Diels—Alder reaction.

2.2.5 Catalyst Loading: B(OPh),/VAPOL Ratios

As mentioned in the introduction to this chapter, catalyst loading is a very
important aspect to catalytic asymmetric reactions. In many cases the metal and/or
ligands used for these transformations can be very expensive or difficult to prepare. If a
lot of money or time is spent to achieve these catalysts, it is desirable to use as little of the ~
catalyst as possible. The discussion will now turn to the catalyst loading needed for this
particular reaction. In this section, not only will the catalyst/ligand loading be discussed

but also the effects of different ratios of VAPOL to B(OPh)3 used to prepare the catalyst.

'69

i . In almost all cases for this study, an excess amount of B(OPh)3 was used for the
preparation of the catalyst. This leads to inevitability that there will be at least some
opportunity for the reaction to be catalyzed by the B(OPh)3 to give racemic product. A
few reactions were attempted to determine how efficient B(OPh)3 alone would be at
facilitating this reaction (Scheme 2.13). If 30 mol% triphenylborate was used and the

Scheme 2.13 Reactions Using B(OPh);I

 

 

 

Ph Sl / 0
/ l
\NAPh + O \ B(OPh)3 > I
MOMe solvent N
150 31 temperatrue /|\
1.0 equiv 2.0 equiv. "33090" time Ph Ph
151
31' B(OPh) temperature flask reaction Yield 151
. . 3
Entry :II‘II‘IT'I‘I’I'; (mol%) ”Ive” (°C) type time (h) (%)
1 3 30b CH,CI2 RT rbf 16 30
2 0 100b CH???“ —45 rbf 24 46
3 3 100° CHEEIIZIIOI —45 COIL 24 25

 

a) All reactions 0.2 M with respect to the imine.

b) B(OPh)3 was purchased from Aldrich and used right from bottle ,

c) B(OPh)3 was heated to 55°C in CH2C12 for one hour and heated under vacuum for 0.5
hours.

reaction was done in CHzCl2 at room temperature the reaction produced 30% yield of the
desired product (entry 1). Two reactions were then done using one equivalent of B(OPh)3
with the, optimal solvent and temperature (DCM/toluene (1:1) and -45°C). In one
reaction, triphenylborate was used as purchased from Aldrich (entry 2). This reaction
gave 46% yield in one day. If the B(OPh)3 was taken through the normal catalyst

preparation procedure only without VAPOL, the reaction Slowed down and only gave

70

25% yield (entry 3’). The results prove that the possibility exists for significant
background reaction to occur, however it is not clear why the reaction using B(OPh)3
directly from the bottle gave better results. It is possible that the bottle may contain other
boron species such as phenylboronic acid or phenylboricacid that could also potentially
catalyze the reaction. By distilling the material or taking it through the catalyst
preparation cycle, these other potential catalysts may be eliminated thus the reaction has
to be catalyzed by B(OPh)3 resulting in lower yields.

The original plan was to use the same catalyst that was found to be successful for
the aziridination reaction.56 This catalyst was prepared using a 1:3 ratio of VAPOL to
B(OPh)3. A series of experiments was then-carried out using this 1:3 ratio of VAPOL to
B(OPh)3 in the catalyst preparation for a study directed at exploring the catalyst loading
(Scheme 2.14). The total catalyst loading of VAPOL was varied from 10 mol% to 50

Scheme 2.14 Reactions Using B(OPh),/VAPOL (3:1)

Ph I / 0
,Si
- \ 0 \ B(OPh)3 NAPOL(3:1) _

N _ Ph + I
MOMe solvent

N

150 I. 31 —40°C )\

 

 

 

1.0 equiv . ‘ 2.0 equiv. reaction time Ph Ph
added over 3.0 h round bottom flask .151 -
Entry B(OPh)3 VAPOL solvent reaction time concentration yield 151 ee_151

(mol%) , (mol%) (h) (M) (%) (%)

1 30 10 toluene 22.5 0.2 58 93
2 ‘ 60 20' toluene 20 0.14 84 91
3 90 30 toluene l7 0. 13 74 92
4 ' 30 10 CHZCI2 20 0.2 ~ 61 90
5 60 20 CHZCI2 15 0. 14 84 83
6 9O 30 7 CH2CI2 21 . 0. 13 85 89
7 150 50, CHZCI2 ° 5 0.08 95 58

 

71

mol% adjusting the amount of B(OPh)3 accordingly to maintain the proper ratio. Three
reactions were done in toluene using 10 mol%, 20 mol% and 30 mol% VAPOL
respectively. It wasobserved from these three experiments that increasing the amount of
catalyst had little effect on the outcome of the reaction. The enantiomeric excess of the
reactions ranged from 91% to 93% with the reaction using 10 mol% being the best. The
yield was effected Slightly going from 10 mol% to 20 mol% VAPOL as the yield
increased from 58% in 22 hours for 10 mol% VAPOL to 84% yield for 20 mol% VAPOL
in 20 hours. When 30 mOI% VAPOL was employed, the yield was 74% after 17 hours,
so not a drastic difference from 20 mol% VAPOL. A similar observation was observed
in the outcome of the reactions with respect to yield for the same series of experiments in
CH2C12 (entries 4-7). In this s01vent one additional experiment was included using 50
mol% VAPOL. The yield of the reaction increased from 61% to 84% going from 10
mol% VAPOL to 20 mol%. Increasing the loading to 30 mol% did not effect the yield
but when 50 mol% VAPOL was employed the yield was nearly quantitative. The data
for the selectivity of these reactions was not very consistent and should be taken lightly,
but the ee was 90% when 10 mol% VAPOL was used, went down to 83% when 20 mol%
was used and then back up to 89% when 30 mol% was used. It was interesting though to
find that the enantiomeric excess dropped to 58% when 50 mol% VAPOL was used. If
the formation of the. catalyst Was complete, and the predominant catalyst species in the
reaction was the active VAPOL-boron species, one would expect the cc to increase.
However, it can be reasoned that this drop in selectivity is in fact due to unformed
catalyst at this lOading. It is possible that the preparation of the catalyst might not be

going to completion using the standard catalyst preparation time and temperature. At this

72

catalyst loading, there is also significantly more B(OPh)3 present that if not reacted with
VAPOL to form the catalyst, could potentially catalyze the background reaction and thus
account for the observed lower enantiomeric excess.

The focus will now turn to the effect of the VAPOL to B(OPh)3 ratio. A set of
experiments was carried out where the amount of VAPOL was maintained at 5 mol% and
the amount of B(OPh)3 was varied from 5 mol% to 500 mol% (Scheme 2.15). As a

Scheme 2.15 B(OPh)3 to VAPOL Ratio Effects

 

Ph | / o
\ A ,51\ B(Ophh (N mol%)
N ph + O VAPOL (5 mol%) > I
MOMe CHzClz/toluene (1 :1) N

150 31 —45°C A

1.0 equiv 2.0 equiv, reaCtion time ph
added over 30 h coil addition flaskb 151

Ph

through cooling
addition coil

 

Entry“ B(OPh)3 (N mol%) reaction time (b) y reld 151 ee 151

 

(%) (%)
1 5 24 50 65 ‘
2 IO 24 52 84
3 15 24 59 89
4 30 24 68 87
5 6O 24 71 90
6 100 30 85 90
7 150 22.5 96 92
8 500 24 98 83

 

a) All reactions run at 0.2 M with respect to the imine ,
b) For discussion and picture of the coil addition flask see Scheme 2.21 and Figure 2.4.

starting point, a reaction was performed using a 1:1 ratio of B(OPh)3 and VAPOL (5
mol% each, entry 1). This reaction gave 50% yield and 65% ee which is obviously not a

very good result compared to what has been observed before. It can be see from the data

73

 

in Scheme 2.15 that as the amount of triphenyborate was increased, so did the yield. The
yield could be increased to nearly quantitative when 500 mol% triphenylborate was used
(entry 8). This effect was not surprising, as it would be expected that upon introducing
more catalyst to the system; the reaction would proceed at a more rapid rate. It was very
interesting, however, to find that the same increasing trend was observed for the
selectivity of these reactions. The ee increased from 65% to 90% when the amount of
B(OPh)3 was increased from 5 mol% to 60 mol%. The highest induction of 92% ee was
observed when 150 mol% B(OPh)3 was used (entry 7) . The asymmetric induction did
not drop off until a 100:1 ratio of B(OPh)3 to VAPOL was used (entry 8).

This series of experiments proved to be quite intriguing. It was significant that an
ee of 92% could be obtained with good yield, but further analysis of the results may
reveal other information about this reaction. The results of this series of experiments are

displayed in an easier to follow bar graph shown in Figure 2.3 below. It is not hard to

74

Figure 2.3 Summary of B(OPh)3 Loading Effects

 

Effect of Triphenylborate Loading With
5 mol% VAPOL

..s
O
O

 

\D
O

on
O

Iyield (%)
lee (%)

Result (%)
m \l
o 0

U]
D

 

 

 

A
O

5 10 15 3O 60 100 150 500
Triphenylborate (mol%)

 

 

 

surmise why the yield goes up upon addition of more B(OPh)3. By adding more of the
achiral catalyst, there is more Lewis acid in the reaction mixture, and the possibility for
the racemic background reaction increases significantly. However, if the increase in
yield were due to background racemic reaction, then it would be expected that the
enantiomeric excess would drop off as the yield was increased. This effect was not
observed and even when a 30:1 ratio of B(OPh)3 to VAPOL was used, 92% ee was still
observed. An even more amazing result comes from the reaction where a 100:1 ratio of
achiral catalyst to chiral catalyst was used and the reaction still produced an astonishing
83% ee. In order to rationalize these observations, a study was done to determine the
association constants each of the catalysts (chiral and achiral) to both the imine and the
product of the reaction. The results and discussion of these experiments will be discussed

in Chapter 4. This phenomenon, however, where an achiral catalyst plays an orthogonal

75

role in the reaction to. increase TONS without depleting the selectivity of the reaction, is
. one that has not been observed to date in the aza-Diels-Alder reaction or for any other
asymmetric catalytic reaction. It is worthy of mention that Yu and coworkers62 haveeused
B(OMe)3 to accelerate the catalytic asymmetric allylation of aldehydes with allyl
stannanes using a BINOL-titanium catalyst. This differs from the present work because
the B(OMe)3 is forming a covalent bond with the product rather than acting as a Lewis
acid. In addition, Shibasaki published a review in 199963 discussing the use of additives
in asymmetric catalysis. Many examples therein display increased selectivity and/or
yield, but again there were no examples where a non-chiral Lewis acid was used to
enhance the reactionefficiency.

The best result observed for the above series of experiments was where 150 mol%
B(OPh)3 was used with 100 mol% a close second. A series of experiments was then
performed to determine the minimum loading of VAPOL in the presence of 150 and 100
mol% B(OPh)3. As seen in scheme 2.16, the reaction using 5 mol% VAPOL in the

presence of 150 mol% B(OPh)3 produced 96% yield and 92% ee.

76

 

Scheme 2.16 VAPOL Loading with 150>mo]% B(OPh)3

 

 

Si B(OPh)3 (150 mol%)
\ A o’ \
N Ph + VAPOL (N mol%)
MOMe CHZClzltoluene (1:1)
150 31 —45°C
1.0 equiv 2.0 equiv. “33011011 time
added over 3.0 h cooling 001] flaSk‘I

through cooling

addition coil

 

Emryb VAPOL (N mol%) reaction time (h) erId 151 cc 151

 

(%) (%)
I 5 22.5 96 92
2 2.5 26.5 91 87
3 1 24 86 83
4 0.5 24 68 66
-5 0.2 23.5 56 7

 

a) For discussion and picture of the coil addition flask see Scheme 2.21 and Figure 2.4.
b) All reactions run at 0.2 M with respect to imine.

When half as much VAPOL was used (entry 2), the reaction was still quite efficient and
resulted in 91% yield and 87% ee. It was not until the loading was decreased to 0.5
mol% VAPOL (entry 4) that the ee started to Si gnificantly drop off. This reaction (300:1
ratio achiral catalyst/chiral catalyst) resulted in 68% yield with a measured enantiomeric
excess of 66%. It was not until the ratio of achiral to chiral catalyst reached 750:1 (0.2
mol% VAPOL) that the background reaction became dominant. In this caSe only 7% ee
was observed with 56% isolated yield.

A similar study on the minimum loading of VAPOL was carried out using 100

mol% B(OPh)3 (Scheme 2.17). The results are similar to those observed using 150 mol%

77

Scheme 2.17 VAPOL Loading with 100 mol% B(OPh)3

 

 

 

 

P“ in/ B(OPh)3 (100 mol%)
\Nx ph + O/ \ VAPOL (N mol%)
©/\ MOMC solvent
150 31 . —45°C
1.0 equiv 2.0 equiv. rcaCtion time
added over 3.0 h round bottom flask 151
Entry $113302) solvent reaction time (h) y 18:30:51 6:71?
1 10 CHzClzltoluene 24 94 90
2 7.5 CHzClzltoluene ' 24 90 9O
3 5 CHZClz/toluene 24 85 90
4 5 Toluene 24 70 88
5 2.5 CH2C12/toluene 24 78 85
6 l CH2C12/toluene 24 66 82
7 1 toluene 24 33 78
8 1 toluene 49 53 8O
9 . 1 CHzCl2 _ 49 78 47

 

a) All reactions were done at 0.2 M with respect to the imine.

B(OPh)3 (Scheme 2.16). It was not until the VAPOL loading was decreased to 2.5 mol%
that the ee dropped below 90%. This reaction gave 78% yield and 85% ee (entry 5).
Decreasing VAPOL loading to 1 mol% (entry 6), resulted in 66% yield and 82% ee. A
few other experiments were done using 1 mol% VAPOL in pure toluene or CHzClz.
When the reaction was done in pure toluene (entry 7 and 8), 33% yield and 78% ee was
observed for the reaction that was allowed to stir for one day and 53% yield and 80% ee
was observed for the two day reaction. Allowing the reaction to stir longer increased the
yield, which shows again that reactions in toluene are slower than those in CHzClz. The

two-day reaction was repeated in CH2C12 and the yield increased to 78%, but it was not

78

surprising that the ee dropped to 47% as this difference in solvent has been seen several
times before (i.e. Scheme 2.9).

The best reaction conditions from the catalyst loading experiments was found to
be 150 mol% B(OPh)3 and 5 mol% VAPOL which gave 96% yield and 92% ee. This
reaction was only attempted one time and similar results were obtained from reactions
using 100 mol% B(OPh)3. B(OPh)3 is readily available and very cheap so cost was not
taken into consideration when deciding the optimal catalyst loading for this reaction. It
was unclear how other types of substrates would behave in the presence of excess
B(OPh)3, so it was decided that 5 mol% VAPOL and 100 mol% B(OPh)3 would be used
as the optimal catalyst loading for this reaction.

Some information was desired about what the outcome of the reaction would be if
only the chiral catalyst was involved. In other words, what is the asymmetric induction
in the absence of any background reaction, is it 90% ee or is it higher? A set of

experiments was performed using a stoichiometric amount of VAPOL (Scheme 2.18) to

79

Scheme 2.18 Reactions Using 100 mol% VAPOL

I / B(OPh)3 (N mol%)

Ph
,Si
\N/kph + O \ (S)-VAPOL(100mol%)> I
Now DCM/toluene(1:l)
.151 31

 

—45°C 1
1.0 equiv 2.0 equiv, reaction time Ph Ph
added over 3.0 h 151

in 3 mL solvent

 

Entry B(OPh)3 (N mol%) flask type reaction time (h) y mm 151 cc 151

 

(%) (%)
1“ 100 COILd 24 65 11
2b 300 rbf 18 71 28
3b 1000 COILd 39 28 74

 

a) Reaction run at 0.03 M with respect to the imine.
1 b) Reaction run at 0.2 M with respect to the imine.
0) Reaction run at 0.02 M with respect to the imine.

d) For discussion and picture of the COIL flask see Scheme 2.21 and Figure 2.4.
find the answer and the results were quite interesting. The first experiment was done
using a catalyst prepared with 100 mol% VAPOL and 100 mol% iB(OPh)3 (entry 1). As
was seen for the reaction using 5 mol% VAPOL and 5 mol% B(OPh)3 in the preparation
of the catalyst (Scheme 2.15, entry 1), the reaction was not efficient at all and in this case
an even worse result was obtained (65% yield and 11% ee). The cause for this poor
results could be due to the incomplete formation of the catalyst and significant
background reaction With the non-chiral B(OPh)3 catalyst (see Scheme 2.13 for reactions
with B(OPh)3). It was found during concurrent studies by Gang Hu in the Wulff group
that the active catalyst contains three boron atoms (see discussion in Chapter 6). This
finding offers insight as to why the reactions using a catalyst prepared from a 1:1 ratio of
B(OPh)3 and VAPOL is inferior. In order to prepare the active catalyst, three equivalents

of boron are needed thus when less than three equivalents of B(OPh)3 is used, the

80

likelihood that the catalyst is formed is much less probable. Another reaction was
performed where a 3:1 ratio of B(OPh)3 to VAPOL was used (entry 2) and 71% yield and
28% ee was observed in this reaction. The concentration for this reaction was 0.2 M and
with the large amount of catalyst present, solubility became a major issue, which could
have been the cause for the inefficiency of that reaction. Lastly, one reaction was carried
out where a 10:1 ratio of B(OPh)3 to VAPOL (1000 mol% and 100 mol% respectively)
was used to prepare the catalyst (entry 3). The concentration in this reaction was 10
times more dilute, but still the reaction only gave 28% yield and 71% ee. Other reactions
using lower catalyst loadings were done during the course of these studies where the
concentration of the reaction was Studied over the range of 0.4 M to 0.024 M and little
effect on the selectivity of the reaction although the very dilute reactions were slightly
slower. Unfortunately, these experiments with stoichiometric amounts of VAPOL ligand
employed did not provide any information about whether the 89—91% ee represents the
maximum asymmetric induction for this catalyst in the absence of background reaction.
2.2.6 Danishefsky’s Diene: Quality, Equivalents, and Addition Time and
Temperature

Another aspect of the reaction that needed to be considered was the purity and
source of materials used as well as how many equivalents of each were used. Purity Was
found to be important for VAPOL as well as how much of it was used (catalyst loading).
Some exploration. was done looking at the purity of the diene, diene equivalency, and
addition method. The general experimental procedure for the reaction was to prepare the
catalyst, transfer it to the imine, cool the reaction to the desired temperature, and finally

add the diene. to the reaction mixture. This experimental protocol allowed for the

81

modification of how the diene was added and in this section, this and the other aspects
dealing with the diene will be discussed.

The discussion will first examine the effect of purity and source of the diene and
on how many equivalents were needed to produce the best results (Scheme 2.19). The

Scheme 2.19 Source and Equivalents of Diene'

 

Ph | O
< 1 ’51:
©/\ N Ph + L B(OPh)3NAPOL _ I
/ solvent 7
OMe \ N
150 31. tcmperature /l\
1.0 equiv added over 3.0 h Radio" time Ph Pb

in 3.0 mL solvent 151

 

31 B(OPh)3VAPOL temperature flask reaction yield 16561

 

Entry (equiv) (mol%) (mol%) SOIVC'" (°C) type time (h) 151 (%) (%)
1 2b 30 10 CiI,Ci2 —40 rbf 20 12 ND
2 1.1c 100 5 CHEEIIII‘OI —45 COIL“ 24 72 90
3 15° 100 5 CHEE'IZITOI —45 COIL“ 24 76 89
4 2° 100 5 CH2???“ ——45 COIL“ 24 85 9o

 

a) All reactions were done at 0.2 M with respect to the imine

b) Danishefsky’s diene was purchased from Aldrich and used without further purification.
c) Danishefsky’s diene was independently prepared and purified by distillation.
.d) For discussion and picture of the COIL flask see Scheme 2.21 and Figure 2.4.

diene used for all the optimization Studies discussed prior to this point was prepared by
treating trans-4-methoxy-butene-2-one with KHMDS and trapping the enolate with
TMSCl. The resulting diene (31) was purified by bulb-to—bulb distillation before use in
the reaction. When the reaction was performed with 2.0.equivalents of Danishefsky’s
diene purchased from Aldrich (entry 1), only a 12% yield could be obtained after 20
, hours. Since this reaction was terrible with respect to yield, the ee was not determined. It

was found that the material purchased from Aldrich could be purified and the reaction
82

would work just fine, but due to the expense of the diene ($108/5g), and since it would
have to be purified anyway, it was far better to independently prepare Danishefsky’s
diene. A series of experiments were then performed using 1.1, 1.5, and 2.0 equivalents
. of the diene (entries 2, 3, and 4 respectively). It was found that the amount of diene had
no effect on the selectivity of the reaction. This is consistent with a control experiment,
which revealed that in the absence of catalyst, there was no reaction between the diene
and imine. On the other hand, the amount of Danishefsky’s diene used did have a slight
effect on the yield. The reaction using 1.1 equivalents of the diene gave 72% yield and
increasing the amount to 1.5 equivalents resulted in 76% yield. The yield of the reaction
was increased further to 85% when 2.0 equivalents of diene were used. Even when the
diene was independently prepared, the starting materials needed were relatively
expensive so it was not desirable to use a large excess of this reagent. Thus, no reactions
for this series of experiments were done using more than 2.0 equivalents, which was
chosen as the optimized amount. '

The temperature studies described earlier showed that temperature played a very
important role in the outcome of the reaction (Scheme 2.7). These reactions for the
temperature studies were performed, as mentioned above, by adding the diene to a
premixed and precooled solution (—45°C) of the imine and catalyst. If the diene solution
were added at room temperature all at one time in 3 mL solvent, the-temperature in the
reaction vessel would rise significantly at least for a short period of time. In order to
alleviate this problem, one could add the diene solution slowly using a syringe pump.
Several reactions were done to determine if the slow addition would have a significant

effect on the outcome of the reaction (Scheme 2.20). When the diene solution was added

83

Scheme 2.20 Danishefsky’s Diene Addition Times”

 

     

o
\ i ,s|i< 8(OPh)3 (N mol%)
N ph + 0 (S)~VAPOL (10 mol%) >
MOMe solvent, —45°C N _
150 . 31 reaction time /I\
1.0 equiv 2.0 equiv. Ph Ph
151

added in 3mL solvent
31 addition B(OPh)3

 

temperature reaction Yield 151 cc 151

Entry solvent

 

time (h) (N mol%) (°C) time (h) (%) (%)
1 0 3O CHzCl2 —45 24 69 78
2 3 3O CHzCl2 —4O ‘ 2O 77 8O
3 18 3O CH2Cl2 ‘ —4O 20 23 ND
4 O 30 Toluene —45 96 7O 86
5 3 3O Toluene .—45 18 22 89
6 9 3O Toluene —45 48 63 86
7 12 30 Toluene —45 48 61 87
8 3b 100 Toluene —45 22.5 68 83

 

a) All reactions were done at 0.2 M with respect to the imine.
b) For this reaction, the catalyst was transferred to the imine in 2 mL solvent and the resulting
solution was added to the diene (solution in 3 ml. solvent) over 3 hours.

all at one time (entry 1), the reaction in CHzCl2 gave 69% yield and 78% ee. When the
diene was added dropwise over 3 hours under the same conditions (entry 2) both the yield
and ee increased slightly to 77% and 80% respectively. However, when diene was added
over 18 hours (entry 3) the reaction was very poor and only gave 23% yield. Since the
yield was so low, the ee was not determined. When adding the diene very slowly, the
diene solution has to be in a syringe for the entire addition time. The longer the diene is
in the syringe, the possibility becomes greater for the diene to decompose. Also, since
plastic syringes were used, it is possible that some contaminants from the syringe or
plunger could dissolve in the organic solvent and get added to the reaction also having a

negative effect on the outcome of the reaction.

84

A similar series of experiments were done using toluene as the solvent (Scheme
2.20, entries 47). A small increase in % ee from 86% to 89% was observed upon going
from addition of the diene all at one time to addition over three hours, which was similar
to that seen in CH2Cl2. The yield of these two reactions in toluene are dramatically
different but the reaction time for one is five times longer. Slowing the additiOn down
even more to 12 hours did not have a positive effect on the selectivity of the reaction as
the outcome was within the :1 experimental error. The effect of adding a solution of the
catalyst and imine slowly (three hours) to the diene was also examined and this gave a
good yield but a slightly depressed induction (83% ee) (entry 8). It can be concluded
from this set of experiments that a 3 hour addition of the diene to the catalyst/imine
solution was best for the efficiency of the reaction.

It was found that a slow addition of the diene was necessary and presumably this
slow addition of a room temperature solution of the diene could cause the temperature of
the reaction mixture to rise slightly and thus influence the selectivity of the reaction. For
the ease of physically performing this addition, it would be desirable to not have to use
the syringe pump. If it was indeed a temperature issue, then cooling down the solution of
the diene before it was added would fix the problem and a few reactions were performed '

to determine this effect (Scheme 2.21). First, the diene solution was cooled to the exact

85

Scheme 2.21 Temperature of the Addition of Danishefsky’s Diene'I

Ph | o
/ B(OPh)3 (100 mol%)

’ \
\NX P], + 0 (S)-VAPOL (N mol%) > I
MOMe DCM/toluene (1:1) N

150 . 31 — 5°C . /.\

 

 

 

1.0 equiv 2.0 equiv. reaction time ph ph
151

Ent 31. addition time (h) VAPOL flask e reaction time Yield 151 cc 151
/ temperature (°C) (N mol%) typ (h) (%) (%)
l 0/ —45b 5 rbf 24 84° 86c
2 0 —78d 5 rbf 23 85 g 88
3 0.08/ —45° 5 COILf 22 83 88
4 3/ —.45f . 5 COILf 3O 93 9O

 

a) All the reactions were run at 0.2 M with respect to the imine.

b) The diene solution was cooled to —45°C in a separate flask
and transferred quickly via syringe to the reaction vessel.

c) Average of two runs. ‘

d) The diene solution was cooled to ——78°C in a separate flask
and transferred quickly via syringe to the reaction vessel.

e) These reactions were done side by side and the room temperture
solution of the diene was added dropwise over 5 minutes.

f) The diene was added through the cooling coil.
reaction temperature (——45°C) and then transferred all at one time to the reaction vessel
via syringe (entry 1). This reaction gave 84% yield and 86% ee. Its possible that the
temperature of the diene solution increased slightly during the transfer process so another
reaction was done where the diene solution was cooled to —78°C before it was
transferred (entry 2). The yield remained about the same, but the ee increased two
percent to 88% ee. These results indicated that temperature was probably an issue. The
additional effort to cool the diene solution again raised the question of ease of operation
of the reaction. It. was thought that it would be nice if there were some way to add a
room temperture solution of the diene and have it be cold before it came into contact with
the rest of the reaction mixture. It was envisioned that one way to accomplish this would

86

be to submerge the reaction vessel very deep into the cooling bath and when adding the
diene solution to the flask, be sure that the tip of the needle touched the side of the flask,
so the solution runs down the side of the flask, and be cooled to the reaction temperature
before it reached the contents of the flask. This proved to be not very easy to do and also
leads to some variability in how the reaction was done each time. Inspired by an invited
seminar speaker at Michigan State University who discussed the use of a flask with a side
arm coil through which a solution could be added that would be cooled to the reaction
temperature before it transversed the length of the coil and reached the reaction flask.
This particular flask appeared to be very fragile and probably not very convenient to use.
In collaboration with the glass blower at Michigan State University (Scott Bankroff), a

modified, more robust, version of this flask was developed (Figure 2.4). The flask is

87

Figure 2.4 Glassware with Cooling Addition Coil

 

Image 1 Image 2

simply a 50 mL round bottom flask with an extended neck and wrapped around it is a
glass coil that is topped with a 14/20 ground glass joint and the bottom of which opens
into the top side of the round bottom flask. In addition, to make the flask slightly less

fragile, a glass support connects the top of the coil to the neck of the flask.

88

In order to use this glassware effectively, a deep cooling bath was needed such
that a major portion of the coil could be submerged allowing for the solution of the diene
to properly cool as it runs down the coil and into the flask. The entire setup is depicted in

Figure 2.5. With the flask and cooling bath in hand, two experiments were attempted,

Figure 2.5 Reaction Setup

 

one where the diene solution was added all at one time through the addition coil (table
2.20, entry 3) and the other where the diene was added over 3 hours (Table 2.20, entry 4).
The result for the addition all at once was about the same as for the ~78°C addition for

the traditional round bottom flask giving 83% yield and 88% ee. However, when the

89

diene Was added slowly over three hours via syringe pump through the cooling addition
coil, the ee increased to 90%. The difference in ee from the round bottom flask to the
coil flask is small, however, it was found that using the coil flask that the resu1ts were
slightly more reproducible and eliminated the need to cool the diene solution before
addition. This flask and reaction setup seen in Figure 2.5 was determined to be the
optimal setup for this aza-Diels-Alder reaction and was used later for testing the
generality of this reaction.

_ For every reaction described in this chapter up to this point, the imine used was
prepared by treating benzaldehyde (152) with Bh-NH2 (153) in the presence of
magnesium sulfate and the crude solid material was purified by recrystallization to ensure
high purity of the imine. To make this reaction even more attractive, it would be

desirable to find conditions where the imine could be prepared in situ and the reaction run

without further purification of the imine. Several attempts using a variety of drying

agents were made to determine if this would be feasible (Scheme 2.22). These reactions

 

 

Scheme 2.22 In Situ Preparaton of the Iminea

 

H I

Ph /Si< B(OPh)3 (100 mol%)

0 /J\ + O (S)-VAPOL (N mol%)

+ >
H2N Ph Mom DCM/toluene (1:1)
152 153 31 “mfg?”
1.0 e uiv 2 e uiv. —

q q 24h

added over 3 h
in 3 mL solvent
through cooling coil

 

 

Entry 152 (equiv) VAPOL (N mol%) Drying agent

yield 151 cc 151

 

(%) (%)
1 1.0 5 4A M.s.b 81 56
2 1.0 10 4A M.S. b 80 72
3 ~ 1.1 10 4A M.S. b 85 74
4 1.1 10 MgSO4" 81 70
5 1.1 10 Na,so,° 82 7o
6 1.1 10 Caso,‘Le 75 79
7 1.1 , 10 CaSOf‘ 79 83
8 1.1 10 Caso,‘Lg 82 77
9 1.1 .10 CaSOj" 92 78 .

 

a) All reactions were done at 0.14 M with respect to the amine.
b) 0.15 g of the drying agent was used.

c) 0.16 g of the drying agent was used.

(1) CaSO4 that contained blue indicator was used.

c) 0.22 g of the drying agent was used.

0 CaSO4 without the indicator was used.

g) 0.136 g of the drying agent was used.

performed by first allowing benzaldehyde and aminodiphenylmethane to stir with the

drying agent overnight. The catalyst was then transferred directly to the flask containing

the imine solution, cooled to —45°C and then the diene added. The first drying agent

attempted using this procedure for this reaction was 4A molecular sieves. When 5 mol%

VAPOL was employed (entry 1), the reaction only gave 81% yield and 56% ee. The

same reaction with 10 mol% VAPOL gave the same yield but an improved induction

(72% ee) (entry 2). In these reactions, the amine and benzaldehyde are both Lewis bases

91

that could potentially compete for binding to the catalyst if the formation of the imine
was not complete. The amine 153 would be the strongest Lewis base, thus the rest of the
reactions were run using 1.1 equivalents of benzaldehyde. Benzaldehyde also has the
potential to participate in the heteroatom Diels—Alder reaction with Danishefsky’s diene
and in fact, one reaction was attempted with the VAPOL-boron catalyst (Scheme 2.23).

Scheme 2.23 Heteroatom Diels-Alder Reaction of Benzaldehyde

 

H l O
Si/ B(OPh)3 (30 mol%)
0 + o \ . (S)-VAPOL (10 mol%)_ I
MOMe toluene, —45°C V

\\‘.- O
152 31 48" ©

2 equiv.

added over 3 h 151'?
in 3 mL solvent 67% yield
28% ee

through cooling coil

The reaction was run in toluene for 48 hours at —45°C and gave 67 % yield of 151-0 and
28% ee. Other than possibly using up 0.1 equivalent of the diene, this was not a problem
because if the oxo-Diels-Alder adduct was formed it could be separated at the end of the
reaction (no attempts were made to determine if this product was formed). When 1.]
equivalents of benzaldehyde was used for the reaction in the presence of 4A molecular
sieves, 85% yield and 74% ee was observed (Scheme 2.21, entry 3). Although this was
slightly better than with 1.0 equivalents of benzaldehyde (entry 2), it is not as good as
seen in the optimization studies using the preformed imine (Scheme 2.21, entry 4). For
this reason, magnesium sulfate, sodium sulfate, and calcium sulfate were examined as
other potential drying agents. Magnesium sulfate and sodium sulfate both gave 70% ee
with yields in the low 80’s (entries 4 and 5). The last drying agent attempted was
calcium sulfate. In order to test this drying agent, Dri-rite with the blue indicator on the

' 92

calcium sulfate was first examined. This reaction was attempted using 0.22 g of Dri-rite
(entry 6) and surprisingly this gave a significantly better result (79% ee) than the other
drying agents. Inspired by this result, the same reaction was repeated with calcium
sulfate without the indicator (entry 7), which gave 83% ee. The asymmetric induction for
all the drying agents was lower than for the preformed imine (Scheme 2.21, entry 4) and
it was thought that this could be due to the Lewis acidic nature of the drying agents. As a
result, two more reactions were attempted where half the amount (0.136 g) of the drying
agent was used with and without the indicator (entries 8 and 9 respectively). The yield in
each case was excellent, but the selectivity still remained unsatisfactory. Unfortunately,
this set of experiments lead to the conclusion that the preparation of the imine in situ was
not the optimal way to do this reaction.
2.2.7 Optimal Conditions

The discussion in this chapter has focused on the many variables that could have
an effect on the catalytic asymmetric aza—Diels-Alder reaction. There are almost infinite
possibilities for different combinations when so many variables exist for a catalytic
asymmetric reaction. The effects of many conditions as well as various combinations
thereof were explored and the optimal conditions found are expressed in Scheme 2.24.

Scheme 2.24 Optimal Conditions

 

‘ p. 4/ o
I
\Nkph + 0’ \ . B(OPh)3 (100 mol%) _ I
@ Mom (S)-VAPOL (5 mol%) 7 N
150 31 CH2C12/toluene (1:1) /I\

1.0 equiv 2.0 equiv. -—45°C, 24 h Ph Ph
added over 3.0 h coil addition flask 151
through cooling 85% yield '

addition coil 90% ee

93

It was found that using the catalyst prepared from 5 mol% VAPOL and 100
mol% B(OPh)3 was best for this reaction. Although the yield could be increased from
85% to 94% using 10 mol% VAPOL, the ee remained the same. In order to preserve the
expensive ligand, it was decided that 5 mol% VAPOL would be used as the optimal
loading for the screening of additional substrates. Very low temperatures resulted in low
reaction yields and reactions run below —40°C, failed ,to produce higher ee’s. In
addition, if the reaction temperature was increased above ——40°C, both the selectivity and
yield dr0p off significantly. It was then concluded from the temperature studies as well
as many of the other reactions run around this temperature that —4S°C was optimal for
the aza-Diels-Alder reaction using the VAPOL-B(OPh)3 catalyst. With respect to
solvent, it was found that using a mixture of CHZClz/CCI4 (1:2), resulted in a selectivity
of 94% ee. However, since 90% ee could still be achieved using CHZClzltoluene (1:1),
the less expensive and less toxic solvent combination, was chosen as the desired optimal
solvent for this reaction. As far as the diene is concerned, it was found that 2.0
equivalents of Danishefsky’s diene were necessary to produce yields in the middle 80s.
It was also found that the addition time and temperature was important to the outcome of
the reaction. A three-hour addition of the diene using a syringe pump was found to be
the optimal addition time for obtaining the best selectivity. The cooling coil addition
glassware was found to slightly enhance this selectivity (2% to give 90% ee), but more
importantly the cooling coil flask obviated the need to precool the diene solution and lead
to improved reproducibility. Satisfactory results could also be achieved using a
traditional round bottom flask, but for the optimal results and ease of reproducibility, it

was best to use the special glassware. Several reactions were performed using these

94

optimal conditions with the finding that the reaction gave 83.5:7% yield and 89:1% ee.
The stage was now set to test these reaction conditions by exploring a broad range of

substrates.

95

2.3 Appendix

1.0 equiv

Scheme 2.25 Jon Antilla’s Results

BH3'THF (30 mol%)
VAPOL (10 mol%) __

 

CHZCIZ

temperature

reaction time

   

 

PhAI’h

151

 

Entry Diene Diene addition temperature reaction time yield ee 151

Conversion (%)

 

 

(equiv) time (h) (°C) (h) 151 (%) (%)
1° 1.3 0 -— 20 18 ND ND 62
2° 1.3 0 —20 to RT° overnight ND ND 60
3° 2 0 —40 22 55 84 100
4a 2 O —40 17 46 82 89
5°g 2 O —40 24 43 80 80
6b 2 3 —4O 6 67 77 80
7b 2 3°l —4O 18 0 ND 0
8" 1.2 3 —4O 7 ND ND 35
9" 2 3 —4O 16 63 80 90
10" 2 1 —4O 18 ND ND 67
1 1" 2 l —75 18 31 ND 50
12" 2 4 ——4O 45 ND ND 30
13‘l 2 0 -—40 1 ND ND low
14" 2 3 +40 overnight ND ND 30 I
15° ‘ 1.13 O —40 to RT 18+24f ND ND 17%
16"" 2 3 —40 to RT 18 ND ND 0
a) R = Me and the reaction run at 0.5 M with respect to the imine.

b)
e)
d)

e)

R = Me and the reaction run at 0.33 M with respect to the imine.

Reaction run at 0.1 M with respect to the imine.
The imine was added over 3 hours to a previously prepared
solution of the catalyst and the diene.
The reaction was stirred at —20°C until the diene addition

was complete and then warmed immediately to room temperature.

The reaction was stirred at —40°C for 18 hours and then
24 hours at room temperature.

4A molecular sieves were added to the reaction vessel.

R = t—Bu.

‘ 96’

CHAPTER '3

SUBSTRATE SCOPE

It was seen in the previous chapter that the optimization of the catalytic
asymmetric aza-Diels-Alder reaction utilizing the catalyst derived from triphenylborate
and VAPOL proved to be quite tedious. The optimization for the said reaction was
accomplished using the imine (150) derived from aminodiphenylmethane (153, Bh-NHZ)
and benzaldehyde (152) (as described in chapter 2) and the optimal conditions are
outlined in Scheme 3.1. It is desired for any new methodology that the conditions be

Scheme 3.1 Optimal Conditions for the aza-Diels-Alder

 

CH2C|2 _ 0.5 mmHg

(S)-VAPOL + B(OPh)} fi
55 °C. “1 55 °C, 0.5 h

5 mol % 100 mol %

Pb | . \
Si/ -

/ \
\N A Ph + O VAPOL-boron catalyst _
I MOMe CH2C12/toluene (1:1)

 

> Catalyst

 

 

 

 

» 150 31 -—45°C, 24 h
1.0 equiv 2.0 equiv. coil addition flask
added over 3.0 h 35% yield 151
through cooling 90% cc

addition coil

 

sufficiently general to accommodate many functional groups and multiple substrate
classes using the optimized conditions. With the optimal conditions found, the stage was
set to screen a variety of substrates to test the generality of the reaction.

In an ideal circumstance, studying the substrate generality of the reaction should
be facile whereby using the optimal conditions would give satisfactory results without
further Optimization. It is also well known that, in the world of chemistry, rarely 'does the
ideal circumstance exist and this reaction proved to be no different.

The initial screening process was accomplished by first using the optimal
conditions (Scheme 3.1). Two runs were performed for each substrate and the average of
the two runs was taken to determine if these conditions gave satisfactory results. If the
two reactions gave dissimilar results, then a third run was performed. If the average
reaction yields or enantioselectivity for any particular substrate was not satisfactory or if _
it was thought that the results could be improved, substrate specific optimization was
undertaken. For most substrates it was found that by either doubling the loading ofl(S)-
VAPOL to 10 mol% or by increasing the reaction time, satisfactory results could be
achieved. This was not the case for all substrates and for those exceptions a broader
range of conditions were explored to enhance the efficiency Of the reaction. A discussion
of all substrates and the necessary optimization for each will be discussed in this chapter.
3.1 Aromatic Substrates

The first class of substrates studied was imines prepared from aromatic aldehydes.
The scope included examples to determine the effect of electron donating groups,
electron withdrawing groups on the aryl ring, and the effect of increasing the steric bulk

by placing a substituent in the ortho position of the arene. The first substrate to be

98

discussed is the imine (154) prepared from 4-methoxybenzaldehyde and Bh-NH2
(Scheme 3.2). For the initial screening using the optimal conditions (entries 1 and

Scheme 3.2 aza-Diels-Alder Reaction of 4-methoxybenzaldimine (154)

 

 

 

Ph Sli/ O
\prh + O/ \ B(OPh)3 / VAPOL g
g MOMe CHzClzltoluene (1:1)7 I
M60 154 31 —45°C 1
1.0 equiv 2.0 equiv reaction time MeO Ph Ph
Added over 3.0 h 155
in 3mL solvent
Entry Condition liggfg; "($52: reaction time (h) 1062;;55 6:755
1 A 100 5 24 45 82
2 A 100 5 24 41 71
3 A 100 - 5 24 43 84
4 B 100 . 5 47 58 77
5 C 100 10 24 . 60. 88
6 D 100 10 48 . 71 90

 

2), a yield. of 45 and 41% was Observed, but there was a 9% difference in the
enantiomeric excess. Due to this significant discrepancy in the enantiomeric excess, the
reaction was repeated (entry 3) and 43% yield and 84% ee was obtained. This result
more closely represented entry 1 so these results were considered to be the true values.
However, analysis of these results indicated that both the enantiomeric excess and yield
were unacceptable. Attempting to increase the yield, another reaction was carried out,
which was allowed to stir for 47 hours. The yield increased to 58%, but only a 77% ee ,
was observed. In an effort to increase the ee, the loading of (S)-VAPOL was increased to
10 mol% (entries 5 and 6). When the reaction was allowed to stir for 24 hours, the
reaction produced 60% yield and 88% ee whereas a two day reaction time gave only

slightly higher yield (71%) and enantiomeric excess (90%). Although the yield was only
99

moderately high, the enantiomeric excess reached an acceptable level so no further
optimization of this substrate was attempted.

The focus was then turned to electron withdrawing groups on the aryl imine.
First, the imine (156) prepared from Bh-NH2 and 4—bromobenzaldehyde was studied
(Scheme 3.3) followed by "the imine (159) prepared from 4—nitrobenzaldehyde

Scheme 3.3 aza-Diels-Alder Reaction of 4-bromobenzaldimine (156)

 

Ph I / O
Si
' \N/kPh + O] \ B(OPh)3 /VAPOL g
/©/\ M0346 solvent 7 I
Br 156 31 temperature 1
1.0 equiv 2.0 equiv. reaction time Br Ph Ph
Added over 3.0 h 157

in 3mL solvent

 

 

. yield ee
EntryCondition1(3r(n(())::l7‘))3 Z3531)” solvent tempeéature is“: {:32th 157 157
° 0 y" (%) (%)

I A 100 5 Cflzi'fltlg'me —45 COIL 24 82 9o
2 A 100 5 CHzcglzf ‘1‘;'“°“° ——45 COIL 24 86 88
3 B 30 10 CH2CI2 —4o rbf 26 69 68

 

. (Scheme 3.4). For the 4—bromo substrate 156, the two initial runs (entries 1 and 2) gave
similar and satisfactory results with and average of 84% yield and 89% ee. One other
reaction was accomplished much earlier~ while the optimization studies were ongoing
where 30 mol% B(OPh)3 and 10 mol% VAPOL was used to prepare the catalyst. CHZCI2
was used as the solvent and the reaction was carried out at —40°C .for 26 hours. The

reaction resulted in inferior results giving 69% yield and only 68% ee.

100

The 4—nitrobenzaldimine 158 was not a very good substrate and the best yield

obtained was 71% and the best ee was 75%. The initial two runs (entries land 2) using

the optimal conditions gave an average of 65% yield and 73% ee. Several attempts were

then made to try and increase both the yield and ee. First, the VAPOL loading was

increased to 10 mol% (entries 3 and 4). This did not make the reaction more efficient and

only gave 69% and 64% yield and both gave 73%

Scheme 3.4 aza-Diels-Alder Reaction of 4-nitrobenzaldimine (158)

 

 

 

Ph SI. / o
/ l
\N * Ph + 0 \ B(OPh)3 / VAPOL > '
O N MOMe solvent N
2 158 31 temperature
1.0 equiv 2.0 equiv. reaction time 02N Ph/k Ph
Added over 3.0 h coil addition flask 159
in 3mL solvent
through cooling
addition coil
. . B(OPh)3 VAPOL temperature reaction Yield 159 cc 159
EntryCondItron (mol%) (mol%) solvent (0 C) time (h) (%) (%)
1 A 100 5 CH2C('12{‘1‘;'“°“° —45 24 59 75
2 A 100 5 Cflzci‘fftl‘yum —45 24, 71 71
3 B 100 10 CH2C('12{t1‘;'“°“° —45 24 69 73
4 B 100 10 C1123? gm" —45 24 64 73
- CHZCIZ/toluene —78 to

5 C 100 10 (1:1) _45 42 64 75
6 D 30 10 Cflzci'fftslui’m —45 24 49 64
7 E 100 10 toluene ‘ —45 24 27 66

 

101

enantiomeric excess. It was thought that it might be possible to increase the asymmetric
induction by cooling the reaction to a colder temperature and thus the reaction was started
at —78°C and after 3 hours was allowed to warm slowly to —45°C. After a total of 42
hours reaction time, a 64% yield and 75% ee was obtained for the adduct 159. One
explanation for the low asymmetric induction could be that the background reaction may
be competing with this substrate. However, when the catalyst was prepared from 30
mol% B(OPh)3 and 10 mol% VAPOL (entry 6), the asymmetric induction fell to 64% ee.
During the optimization studies it was observed that reactions run in toluene, gave higher
enantiomeric excess although they were slower. Therefore, the reaction of imine 158 was
carried out in toluene (entry 7), and as expected, the reaction was much slowergiving
only 27% yield, but unfortunately the ee remained low giving only 66% ee. The reason
for the inefficient nature of the reaction for this substrate is unclear at this point. One
possible explanation could be that the oxygens of the nitro group compete with the imine
nitrogen for binding with the catalyst. Another possibility could be that the strong
electron Withdrawing nature of the nitro group effectively pulls enough electron density
out of the imine nitrogen causing it to be a much weaker Lewis base causing coordination
to the catalyst to be much weaker. This could result in increased conformational
flexibility of the bound imine and as a result, less facial discrimination of the imine.
Ultimately this effect would cause the reaction to be much less efficient.

It was also of interest to probe the effect of sterics by introducing substituents on
the aryl group in the .ortho-position. The first substrate studied was imine 160 with a

methyl group in the ortho position of the aromatic ring (Scheme 3.5). The imine was

102

Scheme 3.5 aza-Diels-Alder Reaction of 2-methylbenzaldimine (160)

 

Ph l /
,Si
\ 0 \
N Ph + B(OPh)3 / VAPOL _
MOMe CHZCl2/toluene (1:1)
160 31 -——45°C
1.0 equiv 2.0 equiv. 24 h.

 

Added over 3.0 h cooling coil flask
in 3mL solvent
through the cooling
addition coil

 

Entry Condition B(OPh)3 (mol%) VAPOL (mol%) yield 161 (%) ee 161 (%)

l A 100 5 83 94
2 A 100 5 82 93

 

 

. prepared in the usual fashion from 2-methylbenzaldehyde and Bh-NHZ. It was found that
when the optimal conditions were employed, the reaction gave better than satisfactory
results; 83% and 82% yield with 94% and 93% ee, respectively, was observed for two
separate runs. From this result, it seems that the increased steric bulk close to the imine
aided in the efficiency of the reaction. Another sterically bulky aryl substrate derivative
(162) studied was that prepared from l-rnaphthaldehyde and Bh-NH2 (Scheme 6). The

initial attempts using the optimal conditions gave reproducible results

103

 

 

Scheme 3.6 aza-Diels-Alder Reaction of l-naphthaldimine (162)

 

Ph l /
Si
0 \NA Ph + 0/ \ B(OPh)3 / VAPOL _
0 Mom CHzClzltoluene (1:1)7
162 31 —45°C
1.0 equiv 2.0 equiv. 24 h
Added over 3.0 h cooling coil flask
in 3mL solvent
through cooling
adidtion coil

 

 

Entry. Condition B(OPh)3 (mol%) VAPOL (mol%) yield 163 (%) ee 163 (%)

 

1 A 100 5 75 86
2 A 100 5 78 86
3 B 100 10 79 90

 

with an average yield of 76.5% yield and both reactions resulted in 86% enantiomeric
_ excess. After observing an increased asymmetric induction for the ortho-methyl (160)
substrate, it was difficult to discern why the ee would drop to 86% for this substrate.
Increasing the loading of VAPOL to'10 mol% gave 79% yield and 90% ee, which is a
slightly better result than with 5 mol% VAPOL.

The second to last aromatic substrate screened was imine 169, which contains a
methyl group in the ortho-position and fluorine in the para-position. This substrate was
selected in order to determine the absolute configuration of the aza-Diels-Alder product.
In 2004, Hayashi, at Kyoto University in Japan, synthesized a key intermediate in the

synthesis of a tachykinin antagonist“ (Figure 3.1). A simple hydrogenation of the aza-

104

 

Figure 3.1 Tachykinin Antatonist and Key Intermediate

0 NR2
Me Me
(3:1 [)1 C: N .
ph F H F 0% NR2
164 165

Key intermediate Tachykinin antagonist

 

Diels-Alder product 170 would produce the same key intermediate 164. Under the
hydrogenation conditions, the benzhydryl group would be removed and the double bond
reduced. All the previous imines were prepared from commercially available aldehydes,
were used to prepare the imine, however the 4—fluoro-2-methylbenzaldehyde 168 is not
commercially available so it had to be prepared (Scheme 3.7). A known procedure65 was

Scheme 3.7 Preparation of 4-fluoro-2-methylbenzaldehyde (168)

O H
MgCl
+ H /U\ "0 Ether/THF O
0°C to RT
F 1h F
166 167 168

78%

followed where l-formylpiperidine was treated with the commercially available 4-fluoro-
2-methyl-phenylmagnesiumchloride as a solution in THF to give the aldehyde 168. The
resulting aldehyde was then transformed into the imine 169 in the standard way by
condensing it with Bh-NH2 and it was then screened in the aza-Diels-Alder reaction

(Scheme 3.8). Using the optimal conditions (entry 1) it was found that the reaction

105

Scheme 3.8 aza-Diels-Alder Reaction of 4-fluoro-2-methylbenzaldimine (169)

/ OMe CH2C12/toluene (1:1)?
169 31 —45°C
1.0 equiv 2.0 equiv. 24 h
Added over 3.0 h cooling coil flask
in 3mL solvent
through cooling
adidtion coil

SI
’ \
F

 

 

B(OPh), VAPOL

Entry CondItIon (mol%) (mol%)

reaction time (h) yield 170 (%) ee 170 (%)

 

l A 100 5 24 68 9O
2 B 100 5 50 84 89
3 ' C 100 10- 24 78 91

 

produced 68% yield and 90% ee. The yield was a little low, so the same reaction was
repeated and allowed to stir for 50 hours (entry 2). The ee remained about the same, but
the yield showed an increase to 84%. One attempt was also made using 10 mol%
VAPOL (entry 3) and in this reaction the ee only increased one percent to 91%. With the
cycloadduct 170 in'hand, it was then converted to the key intermediate 164, which had
previously been prepared by Hayashi64 in a catalytic asymmetric "Michael reaction.
(Scheme 3.9). I

Scheme 3.9 Reduction of 4-fluoro-2-methylphenyI-DieIs-Alder Adduct (164)

O O

 

Me I 112 (latm) / Pd/C Me
. S) > . S)
\“' N MeOH \“‘ N
I
A ~ 22 h H
F P“ P“ 45% yield F .
- 170 164

106

The product was dissolved in methanol and treated with hydrogen gas in the presence of
palladium on carbon to give the desired deprotected and reduced product (S-l64). The
optical rotation of this product was then determined to be [012013 -—77° (c 0.18, DMSO).
The measured optical rotation of the Hayashi intermediate (R-enantiomer) was [002"D
+77° (c 0.18, DMSO). Therefore from this experiment it was determined that the (S)—
enantiomer was produced in the reaction when (S)—VAPOL was used for the preparation
of the catalyst.

The last aromatic substrate explored was the imine (171) prepared from
isophthalaldehyde and two equivalents of Bh—NH2 (Scheme 3.10). Only one reaction was

Scheme 3.10 aza-Diels-Alder Reaction of Phthalaldimine (171)

 

 

j: j: 51/ B(OPh)3 (100 mol%)
’ \
Ph N / \N Ph + O (S)-VAPOL (5 mol%)_
MOMe CHzClzltolueneflzl)
I71 31 345°C
1.0 equiv 2.0 equiv. _ h
Added over 3.0 h °°°"“g °°" “a“
in 3mL solvent
through cooling
adidtion coil

attempted for this reaction. It was expected that the reaction would produce the bis-aza-
Diels-Alder adduct, but none of that product was observed. What actually was observed
was that the mono-Diels-Alder adduct 172. Apparently, during the workup conditions,
the second imine was hydrolyzed to the aldehyde giving product 172. Employing the
optimal conditions using two equivalents of the diene per imine (4 total equivalents),
48% yield of the mono-Diels-Alder product 172 was observed with 74% ee.

3.2 a, fi-Unsaturated Substrates

In order to determine if the scope of the reaCtion could be expanded to imines

107

containing 0,13 -unsaturated substituents, three different compounds were examined. The

first was the imine 173 prepared from Bh-NH2 and trans-cinnamaldehyde (Scheme 3.11).

Scheme 3.11 aza-Diels-Alder Reaction of trans-cinnamaldimine (173)

0
Ph ’ \
X + O B(OPh)3/VAPOL ‘ l
\ >
Ph/VN ph MOMe solvent Ph \ N

173 31 —45°C /I\

1.0 equiv 2.0 equiv. 24 h 1;:
Added over 3.0 h
in 3mL solvent

 

Ph

 

reaction yield 174 ee 174
time (h) (%) (%)

B(OPh)3 VAPOL

(mol%) (mol%) solvent flask type

Entry Condition

 

~ 1 A 100 5 CH2%{:'“°"° COIL 28 11 o
2 B 30 10 toluene rbf 48 1 l 49

 

Using the optimal conditions, it was found that this reaction was very inefficient giving
an 11% of cycloadduct 174 with 0% ee (entry 1). However, when the reaction was run in
toluene using 10 mol% catalyst (3:1 B(OPh)3/VAPOL, entry 2), the reaction, albeit slow,

gave 49% ee.

The next (1,8 ~unsaturated imine 175 that was studied was that prepared from 3-

crotanaldehyde (Scheme 3.12). As can be seen in the table, none of the attempted

108

Scheme 3.12 aza-Diels-Alder Reaction of 3-methylcrotanaldimine (175)

 

I O
./
81
Ph 0’ \
/'V\ A + B(OPh)3 [VAPOL C I
\ r
\ N Ph MOMe CHzClz/toluene (1:1) \ N

175 31 —45°C A

1.0 equiv 2.0 equiv. 24 h 1;;
Added over 3.0 h cooling addition flask
in 3mL solvent
through the cooling
addition coil

Ph

 

Entry Condition B(OPh)3 (mol%) VAPOL (mol%) yield 176 (%) ee 176 (%)

 

1 A 100 5 0 ND.
2 A 100 5 0 ND.
3 A 100 5 0 ND.
4 B 100 10 0 ND.

 

reactions with'this substrate produced any cycloadduct at all. This included reactions
under the optimal Conditions and a reaction with the catalyst prepared using 10 mol%
VAPOL.

Lastly, for this class of substrate, the imine (177) prepared from cyclohexene
carboxaldehyde was studied (Scheme 3.13). This substrate is different than imines 173

and 175 in that there is a substituent in the cat—position of the a,B-unsaturated imine.

'109

Scheme 3.13 aza-Diels-Alder Reaction of Cyclohexenecarboxaldimine (177)

\NAPh + AA B(OPh)3/VAPOL>
/ OMe CHZClzltoluene(1:1)

177
1.0 equiv

 

24 h

 

2.0 equiv.
Added over 3 O h cooling addition flask
in 3mL solvent
through the cooling
addition coil

 

Entry Conditions

B(OPh)3 VAPOL reaction tIme yield 178 (%) ee 178 (%)

 

(mol%) (mol%) (h)
1 A 100 5 24 29 93
2 B 100 10 24 32 93 '
3 C 100 10 48 45 94

 

This caused the reaction to behave quite differently from those attempted with imines 173

and 175, and in fact proved to be similar to the secondary aliphatic substrates (discussed

later). The optimal conditions produced a low yield (29%), but the ee was outstanding

(93%) (entry 1). In attempt to increase the yield, 10 mol% VAPOL was employed (why

2) but the yield only increased to 32%. When the reaction was repeated following these

conditions, only allowing the reaction run for 48 hours, the yield increased to 45% and

again excellent asymmetric induction (94%) was observed. It is not clear why imine 177

gives high asymmetric induction and why imines 173 and 175 either give low

asymmetric induction or fail to react. It may perhaps be related to the fact that imine 177

has a substituents in the (It—position of the a,B-unsaturated imine. A clear answer will

have to await further studies on this class of substrates.

110

3.3 Tertiary Aliphatic Substrates
The next class of substrates studied was the aliphatic substituted imines. The first
and only tertiary aliphatic substrate studied was the imine 179 containing a tert-butyl

substituents, which was prepared from pivalaldehyde and Bh-NH2 (Scheme 3.14). As

Scheme 3.14 aza-Diels-Alder Reaction of tert-butyl-aldimine (179)

 

 

 

Ph sl/ O
x l
\N/I\Ph + O \ B(OPh)3 / VAPOL _
, M r I
OMe solvent
N
179 31 temperature A
1.0 equiv Added over 3.0 h reaction time Ph Pb
in 3mL solvent 180
. . 31 B(OPh)3VAPOL temperature flask reaction yield 180
E""5’C0"d't“”‘(equiv) (mol%) (mol%) S°'V°"t (°C) type time (h) (%)
l A 2 100 5 CHZCi'fftl‘ymm ——45 COIL 24 o
2 B ' 2+1 3O 10 . toluene —40 rbf 16 O
3 C 2 30 10 toluene —45 rbf 48 O
4 D O

2 30 10 toluene RT rbf 22

 

seen from the data in the table, all attempts to effect the reaction of imine 179 and
Danishefsky’s diene were unsuccessful. One possible explanation could be that the
bulkier terr-butyl group would limits the necessary interaction between the catalyst and
the imine. If this were the case, then the imine would not be activated and thus no
reaction would occur. Another explanation could be that even if the imine is coordinated
to the catalyst, the tert-butyl group could be too sterically bulky to allow the relatively
large nucleophile (the diene) to get close enough for the aza-Diels-Alder reaction to

occur.

111

3.4 Secondary Aliphatic Substrates
Two secondary aliphatic substrates were studied, the first being the imine 181

prepared from cyclohexanecarboxaldehyde and Bh-NH2 (Scheme 3.15). The two initial

Scheme 3.15 aza-DieIs-Alder Reaction of Cyclohexane Carboxaldimine (181)

 

 

 

l / 0
.. x 0 \ B(OPh)3/VAPOL
wN Ph MOMe solvent N l
181 ' 31 ‘ temperature A
1.0 equiv 2.0 equiv reaction time Ph Ph
Added over 3.0 h 182
in 3mL solvent
' . yield ee
EntryConditionl?r(‘l(:f;))3\(/r:§3§ solvent tempeéature flag: $3,318; 182 182
° 0 ”9 (%) (%)
1 A 100 5 CHZCE'fl‘SWM —45 COIL 24 56 74
2 A 100 5 CHZS'ffaglm‘i —45 COIL 24 57 . 78
3 B 100 10 CHZCZ‘fl‘SWM’ —45 COIL 24 54 91
4 C 100 10 CHici'lzftSWM _45 COIL 46 91 93
5 D 100 10 CH2C(‘12{;3'"°"° ——45 COIL 24 54 94
6 E 100 10 (ZOE/5:502 ——45 COIL 46 50 93
7 E 100 10 cat/£3120, ——45 COIL 48 46 90
8 F 30 10 Toluene —45 COIL 24 45 95
9 G 30 10 Toluene —45 rbf 43.5 76 93

10 H 30 10 Toluene o rbf 48 42 80

 

112

reactions were performed utilizing the optimal conditions developed for the imine from
benzaldehyde (entries 1 and 2). These reactions gave an average of 56.5% yield and 76%
enantiomeric excess. Due to the poor ee observed for this reaction, the VAPOL loading
was increased to 10 mol% (entry 3) and as a result the % ee increased to 91%. This
reaction was slow and increasing the reaction time from 24 to 46 hours increased the
yield from 54% to 91% (entries 3 and 4). It was seen in the optimization studies on the
phenyl imine 150 that reactions in toluene produced higher ee’s and lower yields than in
CHZCIZ. For the cyclohexyl substrate 181, using a smaller amount of CHZCI2 (1:4
CHzClz/toluene) (entry 5) increased the cc to 94% however the yield remained
unchanged. During the course of this study, work was being done simultaneously in our .
laboratory to study the details of the aziridination reaction using a similar catalysts". It
was found that the ee’s could be raised slightly using CCl4 as the solvent for that reaction.
Due to the fact that the melting point of carbon tetrachloride is ——23°C it would not be a
suitable solvent for the aza-Diels-Alder reaction at —45°C. It was found, however, that
if a ratio of 2:1 of CCl4/CH2C12 was used, the reaction would not freeze at -45°C. Using
this solvent ratio (entries 6 and 7), the reaction of 181 gave 182 in 48% yield and 91.5%
ee (average of two runs) and thus, this solvent system was not as" good as the 1:1 ratio of
CHZCIZ/toluene (entry 4). The cc of the product could be increased to 95% (entry 8)
when pure toluene was used as the solvent and-the catalyst was prepared using only 30
mol% triphenylborate and 10 mol% VAPOL. .As expected, the reaction was slow in pure
toluene and increasing the reaction time to 43.5 hours increased the yield to 76% (entry
9). I Finally an attempt was made to increase the yield by raising the temperature. Thus

repeating the reaction in entry 9 at 0°C (entry 10) actually lead to a decrease in yield and

113

in the asymmetric induction.

In addition to the cyclohexyl substrate, one more secondary aliphatic group was
tested. The imine 183 was prepared from isobutyraldehyde and Bh-NH2 and its reaction
with diene 31 was examined (Scheme 3.16). All prior imine substrates were crystalline

Scheme 3.16 aza-DieIs-Alder Reaction of Isopropylaldimine (183)

 

0
Ph /SI,<
\ + O B(OPh)3 /VAPOL
\p N Ph M > I
OMe solvent N
183 31 temperature /i\
1.0 equiv 2.0 equiv reaction time Ph Ph
Added over 3.0 h 184

in 3mL solvent

 

 

. . yield ee
EntryCondition?§1?)r;))3yr:§gli solvent tempféature flas: $338!; 184 184
" 0 WP (%) (%)

l A 100 5 CHzilffjilum ——45 COIL 24 61 9o

2 A 100 5 CH2C(112{ 3‘31”“ —45 COIL 24 54 88
B 100 10 CH2%{‘8‘“°"° —45 COIL 24 64 90

4 C 30 10 toluene —50 rbf ‘ 48 57 91

 

and could bepurified by recrystallization, however the isopropyl substrate 183 was
produced as an oil which was used in this form without further purification. The first two
runs with this substrate using the optimized conditions gave an average of 57.5% yield
and 89% ee (entries 1 and 2). In order to increase the yield, a third reaction was
performed using 10 mol% VAPOL (entry 3). This reaction gaVe a slightly higher yield of
64% but the ee remained about the same at 90%. The reaction of imine 183 was also

carried out in pure toluene to see if the asymmetric induction could be enhanced (entry

114

4). As had been seen before the reaction was slower in toluene and the increase in % ee
was within experimental error and perhaps not significant.
3.5 Primary Aliphatic Substrates

The next logical step was to then look at primary aliphatic substrates to see if the
optimized protocol was applicable to this substrate class. The first choice was the imine
prepared from octanal and Bh--NH2 (Schemes 3.17-3.19). Employing the optimal

Scheme 3.17 aza-Diels-Alder Reaction of n-heptylaldimine (185)

 

| 0
Ph ,sr<
\ A 1 O B(OPh)3 / VAPOL k I
W N P“ MOMe CH2Clz/toluene (1:1)

N

185 31 —45°C, 24 h 6 /|\

1.0 equiv 2.0 equiv cooling coil flask Ph Ph

Added over 3.0 h 136

in 3mL solvent
through cooling
addition coil

 

Entry Condition B(OPh)3 (mol%) VAPOL (mol%) yield 186 (%) ee 186 (%)

 

1 A ' 100 5 45 O
2 A 100 5 37 5
3 B 1003 O 31 NA.
4 C ' 100" 0 40 NA.

 

a) triphenylborate used directly from bottle without further purification

b) triphenylborate used after going through the catalyst preparation cycle (55°C, 1 h and 0.1 mmHg,
0.5 h)

conditions (entries 1 and 2), the cycloadduct 186' was obtained only in moderate yield
with an average of 41% yield over two runs, and even more disappointingly the ee was
less than or equal to 5% for both runs. One possible explanation for the lack of ee could

be that under these conditions, enamine formation (Figure 3.2) could be taking place to

115

Figure 3.21somerization of Primary Alkyl Imines to Enamines

H
N Ph

‘
r

/ N ph isomerization
R/\/ Y R N

Ph Ph
187 188

give a much stronger Lewis base which in turn would bind much stronger to the active
catalyst only allowing the background reaction to occur. To get a feel for whether or not
this was the case, two reactions were done using B(OPh)3 alone with no VAPOL. First,
one reaction was carried out using B(OPh)3 right out of the bottle (entry 3). This reaction
gave 31% yield of the product, which was lower than the yield when the VAPOL catalyst
was employed. During the preparation of the VAPOL-boron catalyst, any excess
B‘(OPh)3 would of course be taken through the catalyst preparation cycle. Therefore, to
insure the same boron species were present as would be during the attempted asymmetric
reaction, another reaction was performed using B(OPh)3 that had been treated to the same
catalyst preparation procedure (entry 4). This reaction gave 40% yield, which is about
the same as Observed when the VAPOL catalyst was used. This suggests that it is indeed
possible that the background reaction could be the only one taking place for this primary
substrate. It is also entirely possible that the VAPOL catalyst for these substrates does
not produce any selectivity.

X To decrease the possibility of any background reaction, the amount of.B(OPh)3 in.
the reaction was lowered and several reactions were attempted using 30 mol% B(OPh)3

and 10 mol% VAPOL for the preparation of the catalyst (Scheme 3.18). The first attempt

116

Scheme 3.18 aza-Diels-Alder Reaction of n-heptylaldimine (185) (more attempts)

 

l/ o
Ph O/SI\
\ k + N B(OPhh/VAPOLA l
WN ' Ph / OMe CHZClz N
185 31 —45°C . 6 /i\
1.0e uiv 2.0e uiv reacuont'mc Ph Ph
q q 186

Added over 3.0 h round bottom flask
in 3mL solvent

 

B(OPh)3 VAPOL

(m 01% ) (mol%) reactlon tlme (h) yleld 186 (%) ee 186 (%)

Entry Condition

 

1 A 30 10 24 9.4 ND.
2 Ba 30 10 24 15.5 o
3 Ba 30 10 24 21 0
4 C 30 10 22 24 4
5 D" 30 10 21 21 0
6 Bic 30 10 24.5 34 0
7 F""' 30 10 21.5 _ 47 4
8 G" 60 20 2o 30 0

 

a) Aminodiphenylmethane was redistilled just before used for the in situ preparatioin of the imine.

b) 1.1 equivalent of the aldehyde was used for the in situ preparation of the imine.

c) The aldehyde was distilled and aminodiphenylmethane was purified by first distillation, then
column chromatography, and distillation once more before the in situ preparation of the imine.

d) The aldehyde was distilled and aminodiphenylmethane was purified by column chromatography
before the in situ preparation of the imine.

(entry 1) was done using the imine 185, prepared in the usual fashion and used as the
crude oil without purification. The reaction was quite poor and only gave 9% yield. For
the rest of the entries, the imine was prepared in situ and the reactions were performed
without removal of the magnesium sulfate. Using this method, the yield could be
increased to 18% (average of two runs, entries 2 and 3) but still no asymmetric induction
was observed. An additiOnal attempt was made with the same conditions (entry 4) and
this time the reaction produced a 24% yield and 4% ee. Since the imine was being

prepared in situ with no purification, it was thought that the imine conversion might not

be 100% and that that the Bh-NH2 could be still present during the reaction and compete
117

for binding with the catalyst. To push the imine formation to completion, 1.1 equivalents
of the aldehyde was added and any excess aldehyde was removed under high vacuum
before the aza-Diels-Alder reaction was attempted. The first reaction where this was
attempted gave only 21% yield and no enantiomeric excess (entry 5). Two reactions
were attempted where careful purification of the aldehyde and BH-NH2 was
accomplished prior to the reaction (entries 6 and 7). Even when the aldehyde was
distilled and the amine purified three times before use, the reaction still only gave 47%
yield and 0% cc. Another reaction was attempted using double the catalyst loading.
where 20 mol% VAPOL and 60 mol% triphenylborate was used to prepare the catalyst.
No ee was observed again for this reaction and the yield (30%) was still unsatisfactory.

Imine 185 is the first substrate that failed to produce any asymmetric induction
with the VAPOL catalyst and thus other ligands (VANOL and BINOL) were examined.
In addition, another triarylborate was explored as well (Scheme 3.19). In this series of

Scheme 3.19 aza-Diels-Alder Reaction of n-heptylaldimine (185) (final attempts)

 

 

 

l/ 0
Ph /31\
/l\ + O B(OAr)3 /VAPOL
\ >
WN Ph MOMe toluene N I
185 31 “45°? 6 1
1.0 equiv 2.0 equiv - “mm“ “m" Ph Ph
Added over 3.0 h round "0mm ms" 186
in 3mL solvent
Entry Condition Ar (mol%) Ligand (mol%) reaction time (h) y ”(1%); 86 620/236
1 . A Ph (30) VAPOL (10) 20 . 17 20
2 B Ph (30) VANOL (10) 51 21 0
3 C Ph (30) BINOL (10) 51 15 18
4 D 0,0-di-Me (30) VAPOL (10) 48 23 16 ’

 

118

experiments, tOluene was used, as it was known to be the best solvent for achieving high
enantiomeric excess. The first attempt was done using 10 mol% VAPOL and 20% ee '
was observed, which is still low, but at least some selectivity was finally seen for this
substrate. When VANOL was used (entry 2), about the same yield was obtained, but no
enantioselectivity occurred using this ligand for the reaction. BINOL proved to be just
about as selective as VAPOL and gave 15% yield and 18% ee (entry 3). In the last entry
(entry 4), the catalyst was prepared using 30 mol% tris-(2,6-dimethylphenyl)borate and
10 mol% VAPOL. It was curious to find that this hindered triarylborate had little effect
on the reaction since it gave 23% yield and 16% ee.

The imine 189 prepared from butyraldehyde and Bh-NH2 was then looked at to
determine if a shorter alkyl chain would have any effect on the reaction (Scheme 3.20). It

Scheme 3.20 aza-Diels-Alder Reaction of n-propylaldimine (189)

 

l/ o
Ph o’s'\
\ k + B(OPh)3 / VAPOL g I
M N Ph / QMe toluene N
189 31 -50°C, 241) A
1.0 equiv 2.0 equiv round bottom flask Pb ph
Added over 3.0 h 190

in 3mL solvent

 

Entry Condition B(OPh)3 (mol%) VAPOL (mol%) yield 190 (%) ee 190 (%)
l A 30 10 35 22

 

was not so surprising to find that this substrate behaved similarly to the longer straight
chain imine 185. The yield of 190 was 35% and 22% ee was obtained for this substrate.
The study of the primary aliphatic substrates 185 and 189 further establishes the

importance of substitution at the position or to the imine.

119

The study of primary aliphatic substrates raises the question of why the
enantioselectivity drops off so drastically when there is no substitution at the or—position.
To be able to give a good explanation of this, it is necessary to know the structure of the
catalyst. Just as the studies in this thesis were coming to a close, studies within the group
were beginning to shed light on the structure of the catalyst (see Chapter 6 for discussion
of the catalyst). It could be that the imine is binding to the catalyst in two different
orientations, each of which results in a different stereochemical outcome. Furthermore,
it is possible that by increasing the Steric bulk at the Ot~position one of the orientations is
favored leading to a majority of one stereochemical outcome and the high ee’s that are
observed. These are all speculations and once the catalyst structure is determined,
hopefully a reasonable explanation for the drop in ee will be evident. As mentioned
earlier, it is also known that primary aliphatic imines are not particularly stable and are
prone to isomerize to an enamine (Figure 3.2). Lewis acids can influence the rate of
isomerization and if the imines are indeed isomerizing, it is also pOssible that the enamine
competes with the imine for binding to the catalyst thus inhibiting the chiral catalyst and
leaving predominantly a non-asymmetric background reaction.

3.6 Modifiable Substrates

To make this methodology more attractive, it would be desirable to at least have a
successful reaction using a substrate containing functionality that could be easily
modified to give a primary aliphatic side chain in separate operations. In addition,
expanding the generality to substrates containing heteroatoms would also be desirable

from the aspect that they provide an easily modified handle for further modification.

120

3.6.1 or-Alkoxy Substrates

Two substrates containing oxygen at the (It-position were studied, the first being
the imine 191 prepared from commercially available benzyloxyacetaldehyde and the
second being the imine 197 prepared from tert-butyl-diphenylsilyloxy acetaldehyde. The

benzyloxy substrate (191), when subjected to the optimal conditions (Scheme 3.21),

Scheme 3.21 aza-Diels-Alder Reaction of a-benzyloxyaldimine (191)

 

| 0
j: + O/Si< B(OPh)3 / VAPOL)
Ph VOWN Ph Mom CHZCIZ/toluene (1:1)7 Ph VO N l
191 31 >—45°C A
1.0 equiv 2.0 equiv. 24 h Ph Ph
Added over 3.0 h cooling coil flask 192
in 3mL solvent
through the cooling

addition coil

 

Entry Condition B(OPh)3 (mol%)VAPOL (mol%) yield 192 (%) ee 192 (%)

l A 100 5 37 (complex mixture) 0
2 A 100 5 51 (complex mixture) 0

 

 

gave a cOmplex mixture compounds which were one spot by TLC analysis. The average
yield of the mixture in the two runs (entries 1 and 2) was 44% and since pure product
could not be isolated, this imine was not explored further. The focus then turned to the
silyloxy substituents. The aldehyde 196 needed for preparation of the imine 197 is not

commercially available and its synthesis is outlined in Scheme 3.22. The bis protection

12]

Scheme 3.22 Preparation of 2-tert-butyl-diphenylsilyloxyethanal (196)

 

Ph Cl P
HO——\_—/'—OH + \Si/ imidazole > ./ O—\_/_0 \S / h
_ >i/Ph toluene, DMF >i/Ph 195 Ph\i<
193 194 0°C to RT
1)03
97‘?
o 2) Triphenylphosphine

O=(—10\S ./Ph
9616

of cis-2-butene-1,4-diol (193) was accomplished by treatment with lTBDPSCI (194) and
imidazole in DMF. In the second step, ozonolysis was carried out to afford the desired
aldehyde (196) in 97% yield. The imine 197 was prepared from 196 and benzhydryl
amine over MgSO4 and the resulting crude oil was used in the aza-Diels-Alder reaction '
without further purification (Scheme 3.23). Only one reaction wasattempted for this

Scheme 3.23 aza-Diels-Alder Reaction of ‘ a-tert-butyI-diphenylsilyloxyaldimine (197)

 

P“ Sli/ B(OPh)3 (100 mol%) 0
Ph ‘si’OwN Ph + 2:; VAPOL (5 mol%) = I
X ph / OMe CH2C12/toluene(1:1) Ph ‘sr’o N
197 31 —45°c X Il’h /l\
1.0 equiv 2.0 equiv. 24 h Ph . Ph
Added over 3.0 h cooling coil flask 198
in 3mL solvent 0% yield

through the cooling
addition coil

substrate following the optimal conditions. The reaction failed to produce any of the
desired product.
3.6.2 Glyoxylate Ester Substrates

It was envisioned that if the aza-Diels-Alder reaction was successful with the

122

imine prepared from ethylglyoxylate or isopropylglyoxylate that the ester could be
reduced and the resulting alcohol could be further functionalized in order to install other
functional groups. The first substrate to be investigated was the imine 199 prepared from
Bh-NH2 and ethyl glyoxylate and purified by recrystallization. This substrate was studied
in detail and several reaction conditions were attempted to try and obtain high
stereoselectivity (Scheme 24). For this substrate, the optimal conditions were attempted ,

Scheme 3.24 aza-Diels-Alder Reaction of Ethylglyoxaldimine (199)

 

Ph l / O
Si
"10va ph + 0’ \ B(OPh)3 / VAPOL _
0 / OMe , solvent 7 EtO N |
199 31 temperature 0 /l\
1.0 equiv 2.0 equiv. reaction time Ph Ph
Added over 3.0 h cooling coil flask 200

in 3mL solvent
through the cooling
addition coil

 

B(OPh)3 VAPOL temperature reaction yield 200 cc 200

Entry Condition solvent

 

(mol%) (mol%) (°C) time (h) (%) (%)
1 A 100 , 10 Cflzci'fftl‘yum —45 24 80 52
2 B 100 10 CHzCillzftSuene ——60 24 72 35
3 C 100 10 CH2C(112{ ‘I‘I‘m‘i —30 24 83 55
4 D 30 10 CHZCI'fi‘I‘I'Wm —45 24 88 53
5 E 100 10 CUE/zilfc'z —45 24 74 60
6 F 100 10 (ice/231202 —78 48 39 3o

 

with the exception that 10 mol% VAPOL was employed instead of 5 mol% (entry 1).

This reaction gave 80% yield and 52% ee. The ester functionality is potentially an

123

activating group due to its electron withdrawing nature, so it was thought that imine 199
might be more reactive than the corresponding imine of benzaldehyde and lower
temperatures might lead to higher selectivity. When this reaction was carried out at
—60°C (entry 2), in fact, the opposite was found and the ee dropped to 35%. After
observing this negative effect, an adjustment was made by varying the temperature in the
other direction. The reaction at —30°C was performed (entry 3) and it was surprising to
find that the yield and ee increased, albeit not significantly, to 83% and 55% respectively.
To decrease the Opportunity for background reaction, one attempt was made where the
catalyst was prepared using only 30 mol% B(OPh)3 (entry 4). The yield increased to
88%‘but the ee was still only 55%. Inspired by the increase in %ee observed for the
reaction of imine 150 when CCI,,/CH2CI2 (2:1) was used as the solvent (scheme 2.11),
two more reactions were attempted for imine 199 where this solvent mixture was uSed.
When the reaction was carried out at —45°C for 24 hours (entry 5), the yield dropped to
74% and the ee did increase slightly to 60%. The last effort to increase the asymmetric
induction involved a reaction at -—78°C for about 2 days (entry 6). These conditions
proved to be the worst for this substrate and only gave 39% yield and 30% ee.

To determine if steric bulk on the ester of the glyoxylate might have an impact on
the outcome of the reaction, the imine 201 prepared from isopropylglyoxylate and

purified by recrystallization was screened (Scheme 3.25). The optimal conditions were

124

Scheme 3.25 aza-Diels-Alder Reaction of Isopropylglyoxaldimine (201)

 

0
Ph Sli/ B(OPh)3 (100 mol%)
i-PrO \ A O’ \
\II/\N ph + VAPOL (10 mol%)_ I
o / OMe CH2C12/toluene(1:1) i-PrO N
201 31 —45°C -
. . o A
1.0 equrv 2.0 equrv. 24 h Ph Ph
' Added over 3.0 h cooling coil flask 202
in 3mL solvent 81% Yield
56% ee

through the cooling
addition coil

employed except that 10 mol% VAPOL and 100 mol% B(OPh)3 were used to prepare the
catalyst and the yield and ee were determined to be 81% and 56% respectively. This
result was not significantly different from the ethyl ester substrate (Scheme 3.24, entry 1),
so no other conditions were studied for this substrate.
3.6.3 Silyl-acetylene Substrate
The next substrate investigated was the imine prepared from an aldehyde
containing a TIPS-protected acetylene. This aldehyde was prepared (Scheme 3.26) by

Scheme 3.26 Preparation of 3-triisopropylsilyl-2-propynal (205)

 

 

, O
TIPS __ H. 1)EtMgBr(204) g
"" 2) DMF,THF, reflux 7 ¢ H
203 65% . “PS 205

treatment of TIPS acetylene'(203) with ethylmagnesium bromide (204) followed by DMF
in refluxing THF. This reaction gave 65% yield of the desired aldehyde 205. The imine
206 was then prepared in the usual fashion and used as the ~crude oil in the aza-Diels-.

Alder reaction attempted (Scheme 3.27). The optimal conditiOns were used except that

125

Scheme 3.27 aza-Diels-Alder Reaction of TriisopropylsinI-acetyleneylaldimine (206)

O

Sli< B(OPh)3 (100 mol%)

Ph
< /§N*ph + 0/ VAPOL(10mol%)_ I
Si / MOMe CHZCIZ/toluene (1:1) / N
\ff 206 31 —45°c < / X

 

Sr
1.0 equiv 2.0 equiv. 24 h \I/ )— Ph Ph
Added over 30 h cooling coil flask 207
in 3mL solvent 71% yield
through the coolIng 22% cc

addition coil

10 mol% VAPOL was employed rather than 5 mol%. Imine 206 proved to be an
inefficient substrate and only gave a 71% yield of cycloadduct 207 in 22% ee. These
results were also not very promising so no further optimization was attempted for this
, substrate.
3.6.4 a-Silyl-a,B-unsaturated Substrate

Of the three (LB-unsaturated imines that were discussed in section 3.2, the only
substrate that gave high asymmetric induction was imine 177 (Scheme 3.13). This was
the only substrate that had an a-substituent and this was suggestive that an (1,8—
unsaturated substrate containing a silyl-group in the (it-position might be successful. In
1 addition, this substrate would be useful because the silyl-group could be easily removed
allowing access to the aliphatic side chain, which could not be obtained directly from the
aza-Diels-Alder reaction of primary aliphatic imines. Preparation of the or—silylaldehyde

211 was acComplished in three steps from 1-heptyne(208) (Scheme 3.28). l-Heptyne

126

Scheme 3.28 Preparation of Z-2-trimethylsilyl-2-ocetnal (211)

\

 

H l) n-BuLi ether —78°C I
/ I 9 % S]
208 209

1) DIBAL (neat), ether, 40°C, lb
2) Bl'z, CH2C12, —78°C, 111

 

0
milk” < l) sec-BuLI,THF. —90 C, 0.5 h Br
8. 2) DMF, —90°C to RT, 3 h M
211 / |'\ Si

87 % Z isomer

was first silylated by treatment with n-BuLi followed by chlorotrimethylsilane giving
209. The regioselective bromination was then accomplished by treating l-tlimethylsilyl—
l-heptyne (209) with neat DIBAL and subsequent quenching with bromide to give 210.
A lithium halogen exchange was then carried out followed by trapping with DMF to
introduce the aldehyde functionality (211). It was important to ensure that the reaction
temperature was kept below' —85°C while adding the reagents because if not, the
stereomeric purity of the olefin would be depleted. As it tumed out, even when the
reaction was kept very cold, only an 87% retention of the geometry of the alkene was
observed. Imine 212 was then prepared from this aldehyde by treatment with Bh-NHZ.
The resulting imine was an oil so this had to be used without further purification for the

aza-Diels-Alder reaction (Scheme 3.29). The first attempt at the reaction of 212 with

127

Scheme 3.29 aza-Diels-Alder Reaction of Z-a-trimethylsilyl-heptenylaldimine (212)

 

 

 

4/ o
i ' + O] K B(OPh)3/ VAPOL
W14 ph MOW CHZClzltoluene (1:1)7 l
Si 31 —45°C \ N
Added over 3.0 h ”am" “"‘e // Ph Ph
212 . cooling coil flask
. In 3mL solvent . 213
1.0 equrv .
through the cooling
addition coil
. . 31 B(OPh)3 VAPOL reaction time yield 213 cc 213
Entry Conditions (equiv) (mol%) (mol%) (h) (%) (%)
1 A 2 100 10‘ 24 0 ND.
2 Ba 2.72 113 13.6 96 0 ND.
3 ' C" 1.87 100 10 96 0 ND.
4 D° 2 . 100 10 96 0 ND.

 

a) The imine was prepared in situ using magnesium sulfate as the drying agent.
b) The imine was prepared in situ using as the 4A molecular sieves as the drying agent.
c) The imine was prepared in situ using sodium sulfate as the drying agent.

Danishefsky’s diene (entry 1) was made using 10» mol% VAPOL and in 24 hours, the
reaction failed to give any product. Three reactions were performed where the imine was
prepared in situ using three different drying agents. Magnesium sulfate, 4A molecular
sieves, and sodium sulfate were all examined (entries 2—4 respectively), but neither of
these reactions gave the desired product even after 3 days reaction time. It was
unfortunate that this reaction did not work because this, would have been a great solution
to the problem of the failure of primary aliphatic imines to give significant'asymmetric
induction for this reaction. The solution to this problem remains unsolved and awaits
future investigations.
3.7 Benzhydryl Derivatives

As mentioned before, during the course of this study, efforts were being made in

the group to determine the structure of the catalyst prepared from VAPOL and B(OPh)3.

128

In conjunction with this, other studies were being conducted to learn about the interaction
of the imine with the catalyst. The originally proposed catalyst was a species that
Contained one boron atom where the two oxygen atoms of the ligand replaced two of the
phenoxy groups on triphenylborate. It was thought that if this structure was correct, the
benzhydryl group on the nitrogen of the imine 150 could have C—H edge/face pi

interactions with the pi system of the ligand (see Figure 3.3). In Order to test this

Figure 3.3 Original Propsed VANOL-Boron Catalyst/Benzaldimine Interactions

 

hypothesis, Yu Zhang (former Wulff group member) prepared imines from benzaldehyde
and benzhydryl amines with a wide variety of substituents. It was anticipated that the

methyl groups in the 3,3’,5,5;—tetramethylbenzhdryl substituted imine 215 (Figure 3.4),

 

Figure 3.4 Tetr ‘5“ H3-“ ' (TMB-benzaldimine)(215) .

 

129

would block the possibility of CH pi interactions and thus make the substrate/catalyst
interaction lessfavorable and ultimately cause the reaction to be much less efficient.
When this imine was used in the aziridination reaction, the opposite effect was observed.
The ee was raised from 93% to 98% and the rate was 10 times faster. This observation
prompted the synthesis of several other benzhydryl derivatives, which were also
_ evaluated in the aziridination reaction (Scheme 3.30). The relative rates and % ee for the
Scheme 3.30 Relative Rates and Selectivity of the Azirdination Reactions of

. Substituted Benzhydryl Protected Benzaldimines

O 0.2 equiv EDA
0.05 equiv VAPOL—borate:

¢\ amrruh

 

 

 

 

 

 

 

 

 

 

N m (N5 (”3
X : relative rate P11 (3023 ph C025!
1.0 equiv. (XX% cc) : product ec‘b
“‘0 0-11 12’! gmo/Oé MID?
(99% ee) (98% ee) (95% ee) (.94 7% ee)
(48%7 ee) ! l f OF/i 1%.:
(93% °°) (93.5% ee) (93% “9
ca
003 ..r" 0054, 06 o. 02 {Rare
(81 % ee) (90% ee) Fsc (61% e6) (90% 3ee)

aziridination with many of these substrates are shoWn in Scheme 3.30. Since the rates

were much faster and the reactions were more selective for some of these substituted

130

benzhydryl groups in the aziridination reaction, it was decided to determine if the same
positive effects would be observed in the aza-Diels-Alder reaction. The tetramethyl
(TMB), tetra-tert—butyl-dimethoxy (BUDAM), and dimethoxy (DAM) substituted
benzhydryl groups were selected for study in the ciza-Diels-Alder reaction. .
3.7.1 Tetramethyl Benzhydryl (TMB) Substrates

First, the TMB imine 215 was studied (Scheme 3.31). Initially, the optimal

Scheme 3.31 aza-Diels-Alder Reaction of TMB-benzaldimine (215)

0 0
O,Si\

+ B(OPh)3 / VAPOL _ l
\N MOMe solvent N
31 45°C 0
2.0 equiv. 24 h 0
215 Added over 3.0 h coming 0°" flaSk

1-0 equiv in 3mL solvent
through the cooling 216
addition coil

 

 

 

Entry Condition {3:121:53 2,35%]; solvent ‘ y 163316 6:92.36
1 A 100 5 CHzClzltoluene (1:1) 70 93
2 B 100 5 CCl4/ CH2CI2 (2: 1) 56 91
3 C 100 10 CHzClzltoluene (1:1) 84 93
4 D 100 . 10 CHZCI2 66 76
5 E 100 10 CCl4/ CHZCI2 (2:1) 75 95
6 F 100 10" CHZClz/toluene (1:1) 79 96
7 G 100 10' CCl4/ CHzCl2 (2:1) 63 93
8 H 100 10‘ toluene 27* 82

 

a) The catalyst was prepared in CO4

conditions were examined (entry 1) and the cycloadduct 216 was obtained with slightly
higher asymmetric induction (3% cc) than the corresponding benzhydryl imine 150

(Scheme 3.1) but in lower yield. The same reaction was repeated using CCl4/CH2Cl2

131

(2:1) as the solvent (entry 2) and this gave 56% yield and 91% ee. It was found that the
yield could be increased to 84% by doubling the amount of VAPOL and the induction
remained the same (entries 1 vs. 3). When the solvent was changed to CC]../CH2Cl2 (2:1),
the ee could be increased to 95% with 10 mol% VAPOL (entry 5). Interestingly, if the
catalyst was prepared in CC]4 at 80°C rather than in CHZCl2 at 55°C, and the reaction run
in DCM/toluene (1:1), a 79% yield of the Diels—Alder adduct was obtained in 96% ee
which is the highest achieved to date. Other solvents were explored and all were found to
be inferior (entries 4, 7, and 8).

Due to the enhanced asymmetric induction observed for imine 215, it was decided
to determine if this effect was general for other substrates containing the TMB group as
well. First, the 4—bromo-phenyl-TMB imine 217 was easmined (Scheme 3.32). The

Scheme 3.32 aza-Diels-Alder Reaction of TMB-4-bromobenzaldimine (217)

 

 

O
S!i/ B(OPh)3 (100 mol%)
0’ \
+ VAPOL (10 mol%) ‘ l
MOMe CCUCH2C|2(2:1) N
31 —45°C
B
2.0 equiv. . 24" r
Added over 3.0 h c°°'"‘8 °°" “ask
in 3mL solvent *
through the cooling 2l3
addition coil 73% Yield
86% cc

reaction of the 4—bromo substrate 156 with the benzhydryl imine gave 89% ee (Scheme
3.3), and the corresponding TMB imine 217 gave the Diels-Alder productwith 86% ee.
It was also desired to determine the effect of the TMB protecting group on the reaction of
aliphatic substrates. The imine 219‘ was then prepared from cycllohexane carboxaldehyde

and the results of the aza-Diels—Alder reaction of this substrate are shown in Scheme

132

3.33. When the optimal conditions with 10 mol% VAPOL) were employed (entry 1), the

Scheme 3.33 aza-Diels-Alder Reaction of TMB-cyclohexane carboxaldimine (219)

 

l/ O
/Si
O \
+ N B(OPh)3 / VAPOL _ |
/ OMe ecu/C11202 (2: 1) N
31 —45°c

 

2.0 equiv. 24" O 0
Added over 30 h cooling coal flask

in 3mL solvent
through the cooling
addition coil

1.0 equiv

220

 

Entry Condition B(OPh)3 (mol%) VAPOL (mol%) yield 220 (%) ee 220 (%)

1 A 100 10 3O 7O
2 B 100 10“ 28 80
a) The catalyst was prepared in CCI4 at 80°C.

 

reaction gave 30% yield of 220 with an asymmetric induction (70% ee) that was much
lower than that for the unsubstituted benzhydryl imine 181 (93% ee) under the same
conditions (Scheme 3.15, entry 6). The ee‘ could be increased to 80% when the catalyst
was prepared in CO4 at 80°C (entry 2). Although the reaction with the TMB imine 215
from benzaldehyde showed increased induction (Scheme 3.31), this proved to not be a
general effect as the reactions of the cyclohexyl and 4~bromophenyl derivatives both
produced poorer results.
3.7.2 DAM Substrates

. The next benzhydryl derivative studied was the DAM-imines (221) (Scheme

3.34). For each of the substrates, the yields were on par with those produced for the

133

Scheme 3.34 aza-Diels-Alder Reaction of DAM-protected Imines (221a-d)

OMe ' 0
|./
,Si\ I
+ O B(OPh)3 / VAPOL _ R N

 

/ OMe CHZClzltoluenefltl)’

A
. ” O 31 O o
OMe 2.0 equiv. 24 h

 

 

221 . Added over 3.0 h coming CO“ “33k MeO OMe
1.0 equiv _
in 3mL solvent 222
through the cooling
addition coil
Entry Condition R . B(OPh)3 (mol%) 2,35% ProductYliI;)222 6:732
1 A phenyl (221a) 100 10 2223 93 84
2 A 4—bromophenyl (221b) 100 10 222b 100 90
3 A 4—nitrophenyl (221C) 100 10 222C 68 81
4 A cyclohexyl (221d) 100- 10 222d 66 56

 

corresponding benzhydryl imines. The ee for the phenyl (221a) and cyclohexyl (221d)
derivatives were lower, however for the 4-bromophenyl (221b) substrate the ee increased
1% to 90% and for the 4-nitrophenyl (221C) substrate, the ee was increased from 75% to
81%. The DAM group, therefore is not generally a better choice for the aza-Diels-Alder
reaction but may be useful for specific substrates. 2
3.7.3 BUDAM Substrates

In the aziridination reaction, it was discovered that the best benzhydryl derivative
was BUDAM. The observed rate was 16 times faster than the unsubstituted benzhydryl
group and reactions with imines derived from aryl aldehydes generally gave 96-99% ee
with very high yields. The imine 223 prepared from BUDAM amine and benzaldehyde

was prepared and the results from its Diels-Alder reactions are shown in Scheme 3.35.

134

Scheme 3.35 aza-Diels-Alder Reaction of BUDAM-benzaldimine (223)

O

 

+ B(OPh)3/ VAPOL _ N
MOMe solvent
3‘ O 0
CM: 2.0equiv. 24h

   

 

 

223 Added over 3.0 h °°°““g °°" “33" M30 0M,
1-0 equiv in 3mL solvent
through the cooling 22“
addition coil
. . B(OPh)3 VAPOL temperature .
Entry Condition (mol%) (mol%) solvent (0 C) yield 224 (%) ee 224 (%)
1 A 100 5 CH2C('12{‘1°)'“6“6 ' —45 53 20
2 B 100 10 C1123? films —45 35 4o
3 C ‘30 10 CC|4 RT 9 29

 

When the optimal conditions were employed, the reaction gave 58% yield and 20% ee
(entry 1). Obviously, this is much worse than the reaction of imine 150 with the
benzhydryl group (Scheme 3.1). One attempt was made using 10 mol% VAPOL (entry
2) and although the ee increased to 40%, the reaction was still much less efficient than
the unsubstituted benzhydryl imine. The best conditions for the aziridination reaction
employed 30 mol% B(OPh)3 and 10 mol% VAPOL in the catalyst preparation and then
carrying out'the reactions in CCl4 at room temperature. When these conditions were
utilized for the aza-Diels-Alder reaction, the yield dropped to 9% and the cc to 29%. It is
- definitely clear from these results that the BUDAM group does not have the same
positive effect on the aza—Diels-Alder reaction as it does on the aziridination reaction.
' Nonetheless, substrates were screened with the BUDAM group to see if this effect is

general (Scheme 3.36). As indicated by the data in Scheme 3.36 the tetra-t-butyl

135

Scheme 3.36 aza-Diels-Alder Reaction of BUDAM-protected Aldimines (225a-c)
OMe O

I/ '
O/Sl\ |

+ B(OPh)3 / VAPOL; R N
/ OMe CHzClzltoluene (lzl)

3‘ O O
OMe 2.0equiv. 24h

 

   

 

 

225 . .
[.0 equiv Added over 3.0 h cooling corl flask MeO OMe
in 3mL solvent
through the cooling 226
addition coil
. . B(OPh), VAPOL yield 226 ee 226

Entry Condition R (mol %) (mol %) Product (%) (%)

l A 4—bromophenyl (225a) 100 10 226a 15 29
2 A 4-nitrophenyl (225b) 100 10 226b 0 ND.

3 A cyclohexyl (225C) 100 10 226C 21 '30

 

dimethoxy benzhydryl (BUDAM) appears to be uniformly detrimental to the aza-Diels-
Alder reaction.

Obviously, the aza-Diels-Alder reaction is clearly a different reaction than the
aziridination reaction. Although the catalyst is the same and the imines are the same, the
major difference is the nucleophile that is involved in the carbon-carbon bond forming
reaction. Danishifsky’s diene and ethyldiazoacetate clearly have significantly different
effects on the outcome of these reactions. With the BUDAM imines, it seems that the
large steric demand of this group does not allow the diene to approach the imine in a
selective manner. The smaller ethyldiazoacetate, on the other hand, is able to approach
the bound substrate and the formation of the carbon-carbon. bond is more facile thus
giving the outstanding enantiomeric excesses observed.

In conclusion, the aza-Diels-Alder reaction with the VAPOL/B(OPh)3 catalyst

described in this chapter is general for aromatic substrates and secondary aliphatic

136

groups. Yields and ee’s in the 903 could be achieved for most of the substrates. The
imines prepared from trans-cinnamaldehyde (173) and 3-methyl crotonaldehyde (175)
gave poor results, however that prepared from cyclohexene carboxaldehyde (177) was
successful. Unfortunately the generality of the aza-Diels-Alder reaction could not be
expanded to imines bearing primary aliphatic groups. These substrates give only
moderate yields and they fail to produce any enantiomeric excess. Attempts were made
using other substrates that would allow access to primary alkyl side chains, but none of
these substrates proved to be viable. The best result in this regard came from the imines
199 and 201 prepared from ethyl glyoxylate .and isopropylglyoxylate respectively,
however, the ee’s for these substrates was only around 60%. The TIPS acetylene
substituted imine 206 was also examined, but its reaction gave only 22% ee. Lastly, one
attempt was made with imine 212 which had an anti-unsaturated group with a
trimethylsilyl group in the a-position, but this substrate failed to give any of the desired
Diels-Alder adduct.

Other benzhydryl derivatives were also surveyed and a much different outcome
was seen for these substrate in the aza-Diels-Alder reaction than in the aziridination
reaction. The TMB imine was better for the phenyl substituted substrate (215), but this
'was not the case for the cyclohexyl (219) or 4—bromophenyl (217) substrates. The DAM
: and BUDAM imines were also studied and it was found that these were also not useful
for the aza-Diels—Alder reaction. The DAM imines gave similar yields as the benzhydryl
imines, but with the exception of the electron withdrawing phenyl substrates, the ee’s
were not better. BUDAM imines were found to be very pobr substrates for this reaction

and all of the results with the BUDAM imines were much inferior to those with”

137

unsubstituted benzhydryl imines. The highest ee observed for a reaction with a BUDAM

imine was 40%.

138

CHAPTER 4

EFFECTS OF TRIPHENYLBORATE

Discussed in chapters two and three was the development and generality of the
catalytic asymmetric aza-Diels—Alder reaction of iminodienophiles with Danishefsky’s
diene using the catalyst prepared from VAPOL and B(OPh)3. An interesting discovery
during these studies was the use of excess B(OPh)3 (a non-chiral Lewis acid) which led to
increased yields without any loss of asymmetric induction. As indicated in Figure 4.1, an

Figure 4.1 Summary of B(OPh)3 Effects

 

Effect of Triphenylborate Loading With
5 mol% VAPOL

 

Iyield (%)
I ee (%)

Result (%)

 

 

 

Triphenylborate (mol%)

 

 

 

139

an increased rate of the reaction could be achieved without the loss of induction until a
ratio of 100:1 B(OPh)3 to VAPOL was used to prepare the catalyst. To our knowledge,
this type of system, which involves two Lewis acid catalysts that can serve in orthogonal
capacities, has not been demonstrated before. Several examples do exist, however, where

63' 66' 67 where an increase of

non—chiral additives have been used in asymmetric catalysis
ee’s and/or yield could be observed, but this is the first example of where an achiral
Lewis acid has had the observed effect of increasing yield without changing the ee’s. Yu
and coworkers have noted that B(OMe)3 will accelerate the catalytic asymmetric
allylation of aldehydes with allyl stannanes using a BINOL-titanium complex62 as the
chiral catalyst. The mechanism of this process is not known but is thought to involve an
‘ alkoxide exchange between titanium and boron on the product. So, for Yu’s allylation
reaction, there is a covalent bond formed between the product and either of two catalysts.
The idea they propose is that the boron-oxygen bond is much stronger than the titanium-
oxygen bond and in the presence of B(OMe)3, the product will preferentially bind to the
boron thus releasing the titanium species to catalyze the reaction. This is different,
however, than the system developed for the VAPOL/B(OPh)3 catalyst because both the
starting imine and the product of the aza-Diels-Alder reaction are neutral and thus only
Lewis acid/Lewis base interactions can take place. In order to account for the results
shown in Figure 4.1, when the product is formed, the chiral catalyst would have to

exchange with the non-chiral B(OPh)3 thus liberating the chiral catalyst to catalyze the

reaction, allowing turnover without loss of asymmetric induction (Scheme 4.1). This

140

Scheme 4.1 Turnover Induced by Triphenylborate

VAPOL-Boron
I / catalyst ‘ ‘ ~ .0
\ I
+ _" I
RANAPh Mow. N OR I

‘ R N
'. RNA VAPOL-Boron
VAPOL-Boron catalyst Ph
catalyst
B(OPh)3

-.

VAPOL-Boron I OR I
+
catalyst R Nu“ R N

mink-1

Ph Ph

system is even more interesting in that it has two boron catalysts that are competing for
interaction with the imine and product. To explain the results, there must be a very large
difference in the rate for the two catalysts in favor of the chiral catalyst, or there must be
a very large difference in the binding constantas for the two catalysts for the imine in
favor of the chiral catalyst, or a combination of the two. A possible source of these
effects could be due to a greater Lewis acidity for boron if it were to form a borate
complex with a VAPOL that contained a 7-membered ring. It has been found in recent
studies in the Wulff group that the catalyst is actually a Bronsted acid (see discussion in

Chapter 6). The Br¢nsted acid would presumably have a stronger interaction to the imine

141

or product than the B(OPh)3 Lewis acid, but this still offers no solution to the question
about the role of triphenylborate.
4.1 Exploration and Explanation of the VAPOL-B(OPh)3 Catalyst System

The question that arises from these results is why does the rate of the reaction
increase without the lowering of the ee when excess B(OPh)3 is used? It was seen earlier
(see Scheme 2.13). that the reaction can indeed be catalyzed by B(OPh)3 so a concern
when using the large excess of B(OPh)3 is that the ee’s would decrease. However, it can
be concluded, based on experimental results, that whatever role the B(OPh)3 is playing in
this reaction, the rate of the. background reaction where the B(OPh)3 is catalyzing the
reaction must be very small compared to the VAPOL—boron catalyst. However, before
studies were undertaken to develop an explanation of this effect, studies were done to
compare thiswork to Yamamoto’s work in order develop an understanding of what kind
of interactions were necessary to make the reaction successful. In order to do‘this,
several experiments were done using Yamamoto’s conditions as well as the conditions
developed in this work using both VAPOL and BINOL and with both the imines 613 and
150 prepared from benzylamine (Bn-NHZ) and benzhydrylamine (Bh-NHz) respectively.

First, a series ~of experiments were done using Yamamoto’s system where the
catalyst was prepared at room temperature, stirring in the presence of 4A molecular

sieves in CHZCl2 for three hours (Scheme 4.2). After the catalyst preparation was

142 .

Scheme 4.2 Comparison of Bn and Bh Using the BINOL-boron Catalyst Prepared

using Yamamoto’s Conditions

 

B(OPh)3
(R)-BINOL > (R)-BINOL—B catalyst

 

CHZCIZ, rt, 4A MS

 

 

 

o
I./ )
H R O/SK (R)-BINOL-B catalyst
A A + N = h
Ph / OMe CH2C|2,-78°,5h '

 

 

Ph N Ph N
61a R = H 31 R/‘\ Pb
= 1.2 ' .
150 R Ph ( equiv) 62 R=H
151 R = Ph
(R)-BINOL B(OPh)3 Yield ee
Entry R (mol %) (mol %) Product (%) (%) Reference
18 23 °
1 H 100 100 62 68-75 77-8 6 Yamamoto , Bull , this
work
2 H 10 100 62 41 15-50 Bull”, this work
3 H 10 10 62 <5 — . Bull”
4 Ph 100 1 100 151 O — Yamamoto”, this work
5 Ph 10 100 151 O — this work

 

complete, the solution was cooled to 0°C and imine was added. Stirring commenced for
1.0 minutes, then the resulting catalyst-imine solution was cooled to —78°C at which
time Danishefsky’s diene (31) was added and the reaction was allowed to stir 5 hours. In
‘ Yamamoto’s publication”, he reported 75% yield and 82% ee for the benzyl imine 61a
under these conditions with 100 mol% catalyst. This reaction was repeated by Bull, et.
al.23 and also in this work. Bull was able to obtain the product 62 that measured 77% ee

while when the reaction was repeated for the present study, the product 62 was obtained

143

with 86% ee. Using 10 mol% BINOL and 100 mol% B(OPh)3 to prepare the catalyst (as
a corollary to the optimal conditions discovered in this work), it was not surprising to find
that that the ee dropped. Under these conditions, Bull reported 15% ee, but did not
mention the yield. When this reaction was carried out for the present study, the product
62 could be Obtained in 41% yield with 50% ee. Lastly, two reactions were attempted
using Yamamoto’s catalyst preparation protocol for'the benzhydryl imine (150).
Whether 10 mol% BINOL, or 100 mol% BINOL was used with 100 mol% B(OPh)3, the
reaction failed to give any product.

The next series of experiments was done using the catalyst preparation method
determined to be optimal for the aza-Diels-Alder reaction of the imine 150 (see chapter
2) using the VAPOL-B(OPh)3 catalyst (Scheme 4.3). For this study, however, 10 mol%

Scheme 4.3 Comparison of Bn and Bh Using the VAPOL-Boron or BINOL-Boron

Catalyst Prepared Using my Optimal Conditions

 

B(OPh)3 0.1 mmHg

(S)-VAPOL Tr (S)-VAPOL-B catalyst

 

CHZClz, 55°, 1 h 55°, 0.5 h

) ‘
l

(S)-VAPOL-B catalyst

S
H R 0’ \ |
/k\ k + > e'
Ph N Ph / CH2C12/toluene (1:1) ph“ N
OMe
31

61a R=H 45224“ A

 

 

 

R Ph
150 R = Pb (2.0 equiv.) 62 R = H
151 R = Ph

 

 

Entry R (Siggg)? (Rgmillggfl‘ 2:33? Product Yield (%) ee (%)
1 H 10 — 100 62 ,50 36
2 . Ph 10 — 100 151 94 9o
3 Ph —- 10 100 151 26 23

 

144

of the ligand was used rather than 5 mol%. The first reaction was done using the imine
61a and after observing the results from the experiments following Yamamoto’s protocol,
it was not unexpected to find that using the VAPOL-boron catalyst that the reaction
produced poor asymmetric induction (36% ee). Interestingly, under the same conditions,
using the benzhydryl imine (150), 94% yield of 151 was obtained in 90% ee. Lasty, one
reaction was done with the benzhydryl imine (150) using (R)-BINOL instead of (S)-
VAPOL and this reaction only gave 26% yield and 23% ee. _

Clearly, it can be seen from these results, that VAPOL is far superior to BINOL
when the benzhydryl imine is used indicating that there is something special about the
catalyst imine interaction that leads to high asymmetric inductiOn. The cause for reduced
enantioselectivity in these reactions would be presumably due to achiral background
reaction. For these systems, it is the reaction catalyzed by B(OPh)3 that would lead to
racemic product so three reactions were performed to determine the effectiveness of
B(OPh)3 as a catalyst for this reaction (Scheme 4.4). The first reaction was carried out
using B(OPh)3 that had been taken through the catalyst preparation cycle, heating to 55°C

for an hour and then evacuating under high vacuum at 55°Cfor 30 minutes (entry 1).

145

Scheme 4.4 Reactions with B(OPh),-

 

o
OSiMe3
H Ph B(OPh)3(l.0equiv.)
\ + > I
pt. N pt. I CH2C12/toluene(l:1) 1)}, N
OMe '_ 0

150 31 45 ,24h Ph Pb

(2.0 equiv.) 151

 

Entry B(OPh)3 (source) Yield 151(%)

 

1 Cat Prep Cycle1 26
2 Aldrich2 46
3 Distilled3 25

 

1) B(OPh)3 was heated to 55°C in CHZCI2 for
one hour and then heated at 55°C for 30 minutes
under high vacuum. 2) B(OPh)3 was used directly
as purchased from Aldrich. 3) B(OPh)3 was
purified by distillation and stored in a glove box
immediately before used.

This reaction only gave 26% yield of the desired product 151. Performing the reaction
under the same conditions only using B(OPh)3 taken directly from the bottle purchased
from Aldrich, the reaction gave 46% yield of the desired product 151 (entry 2). Lastly, a
reaction was attempted using B(OPh)3 that had been purified by distillation and stored in
a glove box before use and the reaction gave a 25 %lyielfl or obvious from
these experiments that there is definitely the possibility for the background reaction to
occur indicating that there must be something special about this system to account for
enhanced rates without loss of induction observed when significant excess B(OPh)3 was
used. In any event, an explanation of the results in Figure 4.1 is still needed.

One explanation consistent with these results is that B(OPh)3 can compete with
the chiral catalyst for binding to the product and thus liberate the sequestered chiral
Lewis acid to turn over more starting material. This explanation still does not, however,

explain why the increasing amounts of B(OPh)3 does not lead to background reaction and

146

loss of asymmetric induction. These observations could be explained simply by an
increase in Lewis acidity that would be expected if the VAPOL-boron catalyst were a
cyclic borate ester, which has been demonstrated in other systems.68 It was mentioned
earlier and will be expanded upon in chapter 6 that the VAPOL-boron catalyst is actually
a Bronsted acid, which does in fact indicate that it would be the stronger catalyst.
According to this argument, this catalyst should also bind stronger to the more basic
product (once it is formed) and thus not allow the catalyst to turn over the reaction as
rapidlly. The question really is how is the B(OPh)3 able to aid in the rate of turnover
without catalyzing the reaction itself? In order to distinguish between the theories
described above, it was decided to measure the binding constants of the VAPOL/B(OPh)3
catalyst and also that of B(OPh)3 with both the imine starting material and the product.
Before discussing the results of the binding study, it is worth mentioning some
mechanistic considerations for this reaction. Whether the reaction occurs via a stepwise
or concerted mechanism, the intermediate of the aza-Diels-Alder reaction with
Danishefsky’s diene is a TMS protected enolate with a B-methoxy group (Scheme 4.5), .

Scheme 4.5 Possible Mechanism Involving B(OPh)3

—

\./ _
. 0/81\ 0 '
Ph 0/ J
A \7 I -
)NL Ph + \ l —> - / _. R N
-/ R N o
R H / O’SK /‘\ Ph/k Ph
Ph Ph
_ B(OPh)3

 

 

which upon hydrolysis leads to the elimination of the methoxy group to give the final

dihydropyridinone (Vinylogous amide) product. One theory for the role of B(OPh)3 is

147 .

that it could be coordinating to the methoxy oxygen thus aiding in the elimination of the
methoxy group giving the final product. Attempts were made to isolate this intermediate
so that it could be exposed to B(OPh)3 to see if the elimination of the methoxy group
could be facilitated. Unfortunately, all attempts to isolate the intermediate were
unsuccessful and this experiment could not be conducted. During the attempts to isolate
this intermediate, several NMR’s of crude material were taken and it was interesting to
find that during the course of the reaction, the major species that was observed in the
crude reaction mixture was the eliminated final product.

Another theory for the role of B(OPh)3 could be that it could be competing for
binding with the product thus liberating the VAPOL-boron catalyst to turnover the
reaction. In order to determine if there were any merits to this possibility, NMR titration
experiments were conducted to determine the binding constants for the VAPOL-boron
, catalyst with both the imine 150 and product 151 as well as binding constants for
B(OPh)3 with the imine 150 and product 151. The study was carried out in collaboration
2 with Lee Fielding at Organon Laboratories Ltd., Newhouse, Lanarkshire, Scotland, MLl
SSH. These experiments were conducted by maintaining a constant concentration of the
imine or product while adding different concentrations of B(OPh)3 or VAPOL-boron
catalyst. A non-linear least squares evaluation of the data obtained from the NMR
titrations by plotting the chemical induced shift versus the concentration allowed for the
determination of the binding constant and the results from this analysis are outlined in

Table 4.1. It was found that the complex of B(OPh)3 with the imine 150 had a binding

148

Table 4.1 Binding Constants for Complexes of the Imine and Product with B(OPh)3

and the VAPOL-Boron-Catalyst

 

Entry Complex Binding Constant (M'l)

 

l B(OPh)3 / imine 0.32 $0.12
2 B(OPh)3 / product 2.7 10.4
3 VAPOL-B / imine 2.1 10.4
4 VAPOL-B / product 4.9 10.8

 

constant of 0.32 10.12 M"1 and the complex of B(OPh)3 with the product 151 had a
binding constant of 2.7 10.4 M". The complexes of the VAPOL-boron catalyst with the
imine.150 and product 151 gave binding constants of 2.1 :04 M'I and 4.9 10.8 M‘1
respectively.

Interpretation of this data would have been straight forward if the binding of the
B(OPh)3 to the product 151 would have been much stronger than the VAPOL-boron
catalyst to the product 151, and the binding of the VAPOL-boron catalyst to the imine
150 much stronger than B(OPh)3. As can be seen from the data, although the later is true,
the former is not, and in fact the binding of the B(OPh)3 to the product 151 is about half
as strong VAPOL-boron catalyst. However, since there is no product at the beginning of
the reaction, the binding of the two catalyst species to the product cannot totally explain
this phenomenon anyhow. Due to this, the binding of the catalysts to the imine must be
significant at the beginning of the reaction. It was found that the binding of the VAPOL-
boron catalyst to the imine 150 was about seven times as strong as B(OPh)3. While the
VAPOL-B(OPh)3 catalyst binds to the imine 7 times stronger than B(OPh)3, it is also true
that there is at the very least 85 mol% B(OPh)3 and only 5 mol% VAPOL-boron catalyst

if three boron atoms are incorporated into the catalyst (a 17:1 ratio of nonchiral to chiral

149

catalyst). Another issue that must be considered is that it is not clear which catalyst,
when bound to the imine, would cause the reaction to happen at a faster rate.

Using the following analysis in conjunction with our experimental data, it was
found that the rate of the reaction of the VAPOL-boron catalyst must be at least 9 times
as fast to account for the trend in Figure 2.1 when a 20 to 30 fold excess of B(OPh)3 is
used.

4.2 Binding Constant Data Analysis

The ultimate goal of the following analysis is to develop an explanation for why
the use of excess B(OPh)3 does not result in lower asymmetric induction. As mentioned
earlier, either the rate of the reaction with the VAPOL-boron catalyst had to be
significantly greater, or the binding constant much larger, or both to account for the ‘
experimental observation. The binding constants were measured but no information
about the rates of these reactions was collected. Due to the fact that the catalyst is
prepared from VAPOL and triphenylborate, it becomes difficult to obtain the VAPOL-
boron catalyst without any triphenylborate still present and it was for this reason that the
rate constant could not be obtained for the reaction of the VAPOL-boron catalyst.
However, using the binding constants, a correlation was made between the constants and
the rates of the reaction. The following analysis will be made for conditions present at
the beginning of the reaction.

The first question to be answered is what would the expected enantiomeric excess
of the reaction be if the rates of the chiral and racemic reaction were the same? The
binding constants were measured (Table 4.1)and using these numbers and the following

analysis the answer to this question was elucidated.

150

For the following analysis, VAPOL-boron catalyst will be abbreviated “B*” and
B(OPh)3 will be abbreviated “B” (as in the following reactions). First, the following rate

equations were developed for this specific reaction:

k
B—imine + Danishefsky's Diene A» Product + B

kch
B*—imine + Danishefsky's Diene -——> Product + B*

Rateracemic = krac [B-imine]*[diene]
Ratechiral = kch [B*-imine]*[diene]
In order to determine what the ee would be if the rates of the reactions were the same,
then the assumption has to be made that kch=kw Having made this assumption, the
following correlation can be made between the rates of the reactions and [B*-imine] and
[B-imine] and thus the binding constants Kmm, and K,(mm,c):
Ratechim/Ratemmic on [B*—imine] / [B-imine]
[B*-imine] = Kmm) * [imine] * [B*]
[B-imine] = Kflmm) * [imine] * [B]
At the beginning of the reaction there is 17 times excess of B over 8* in the reaction so it
was decided to do this analysis on the situation where [B] is 17 times more than [B*].
Before the final analysis could be done, one must understand what the enantiomeric
excess of the reaction would be for each of the two catalysts. The reaction that occurs
when B(OPh)3 (B) is acting as the catalyst, the reaction gives 0% cc (50:50 ratio of SIR).
In order to determine what the maximum ee that could be achieved for the VAPOL-boron
catalyst (8*), attempts to run the reaction using 100 mol% VAPOL (Scheme 2.18) were

made. It is difficult to prepare the VAPOL—boron catalyst in abscense of B(OPh)3 and

151

due to this and other reasons that are unclear, the results of this study were quite
ambiguous and instead, the data from Figure 4.1 was used to determine the maximum ee
for this catalyst. It was found for the reactions using 15 mol%—150 mol% B(OPh)3 that a
constant cc of 90% was observed with an error on these measurements of :1. Based on
these experimental observations, the maximum possible ee produced by the chiral
catalyst was assumed to be 90%.
For the chiral reaction a ratio of 95% S and 5% R is produced.

So a Kmhim,=2.l M" gives 95% S and 5% R
For the racemic reaction, a ratio of 50% S and 50% R is produced.

So a K,(wmic)=0.32 M" gives 50% S and 50% R
When [B] and [B*] are equal, the production of each enantiomer would be a sum of the
binding constants and the asymmetric induction for each catalyst:

[S] (2.1 * 95) + (0.32 * 50) =--. 199.5 + 16 = 215.5 S enantiomer

[R] (2.1 * 5) + (0.32 * 50) = 10.5 + 16 = 26.5 R enantiomer
However, at the beginning of thereaction, there is minimum of a 17 fold excess of the
racemic catalyst. This leads to the weighted equations shown below. When this was
done, it was found that at the beginning of the reaction, based on the binding constants,
and the relative concentrations of the two catalyst species, the actual ee expected would
be 25.1% ee. I

[S] (2.1 * 95) + (17* 0.32 * 50) = 199.5 + 272 = 471.5 S enantiomer

[R] (2.1 * 5) + (17* 0.32 * 50) = 10.5 + 272 = 282.5 R enantiomer
Using the correlation of chhim) / Kfimmic) a Ratechim, / Ratemmic and a similar analysis

where 25.1% ee was used‘as a starting point, a table was constructed (Table 4.2) to i

152

Table 4.2 Relative Rates of the Two Catalysts and ee Prediction

 

 

 

X ee (%)
1 25.1
1.67 35
2.10 40
2.62 45
3.24 50
4.10 55
5.19 60
6.76 65
9.10 70
12.95 75
20.76 80
44.05 85
55.71 86
75.14 87
114.00 88
230.57 89
oo 90
a) x: Itch/km

determine the minimum difference in rate between the two catalysts that could. account
for the experimental observations. As mentioned before, the error for the ee in this study
was :l:1% and-because of this the calculation was extrapolated to the relative rate where
89% ee would be observed. The minimun amount faster (as seen in Table 4.2) that the
chiral reaction had to be in order to observe the asymmetric induction for the reaction
would have to be about 230 times faster. .

The discussion will now turn to what is happening as the product is formed.
Assuming the extreme case where only product is in the solution (100% conversion),
there would be a ratio of about 4.9 to 2.7 of B*-product complex to B-product complex
(1.8 times as much B*-product) based on the measured binding constants. As mentioned
before, there is 17 times as much B(OPh)3 in the solution during the reaction. Due to
this, the ratio would switch to 17 * 2.7 = 45.9 B-product to 4.9 B*-pr‘oduct (9.37 times as

153

much Beproduct complex). In addition to the comparison of binding of the two catalysts
to the product it is worthwhile to compare the binding of B(OPh)3 to the product and
imine itself. Looking at the binding constants for these two, it‘ can be seen that the ratio
of B-product to B—imine is 8.44 (K, B-prod / Ka B-imine = 2.7 / 0.32 = 8.44). It can be
concluded that the combination of the weak binding interaction of B(OPh)3 with the
imine and the presumed much slower rate (based on experiment and K, interpretation) of
background reaction indicates that there would not be significant background reaction to
lower the asymmetric induction.

This interpretation of the binding constants provides sufficient evidence and to
account for the experimental results present in Figure 4.1. It can be concluded that it is a
combination of. the smaller rate of the background reaction as well as the fact that the
excess B(OPh)3 coordinates to the product enough to not allow the chiral catalyst to be
Sequestered by the product. This in turn shows that the two catalysts are working
orthogonally to facilitate an efficient reaction giving enhanced rates without the loss of

enantiomeric excess.

154

CHAPTER 5: EFFORTS TOWARD THE SYNTHESIS OF

CYLINDRICINE C

5.1 Cylindricine C History

Cylindricine C was isolated in 1991 from the sea squirt (Clavalina Cylindrica) off
the coast of Tasmania and was fully characterized. in 1994 by Li and Blackman‘”.
Cylindricine C is only one of many cylindricine molecules that have been isolated from

the marine ascidiam.‘ Cylindricines A and B71 and D-K‘”72 (Figure 5.1) have a similar

Figure 5.1 Cylindricines A-K

O
O
N
Cl n-C4H9
N
n-C6H13
H NCS

R=Cl Cylindricine A (227) Cylindricne B (232) Cylindricine G (233)
R=OH Cylindricine C (228)

=OMe Cylindricine D (229)
R=OAc Cylindricine E (230)

=SCN Cylindricine F (231)

 

O
O
H
N
AGO N SCN "‘C6Hl3
n—C6H 13
H Cl

R=SCN Cylindricne H (234) Cylindricine J (236) Cylindricine K (237) .
R=NCS Cylindricine I (235) '

 

155

core tricyclic ring structure and contain minor structural differences. Cylindricine B and
J are ring expanded isomers of Cylindricine A and F respectively. It is thought that
Cylindricine B is in equalibrium with Cylindricine A (Figure 5.2) via the quaternary

Figure 5.2 Isomerization of Cylindricine A to B

 

 

0 0 o
n‘C6Hl3 N ' n'C6Hl3 ®N n C H N C]
' 6 13
Cl 06) H
Cylindricine A " — Cylindricine B
227 238 232

ammonium salt 238 formed by attack of the nitrogen on the carbon containing the
cthrine followed by subsequent nucleophilic ring opening by the chlorine to give the
ring expanded Cylindricine B (232). All species in the Cylindricine class of compounds
have the cis fused aza-decaline system which is structurally unique compared to the
related compounds, Fasicularin73 and Lepadifonnine,74 a class of compounds which
contain the trans fused aza-decalin system (Figure 5.3). In addition to the transfused

Figure 5.3 Fasicularin and Lepadiformine

"j W A “0%
fi a: '
n-C6Hl3 N n’C6H13

Fasicularin (239) upadiformine (240)

aza-decalin system, Fasicularin and Lepadiforrnine differ from Cylindricine C also in that

the carbonyl functionality on the 6-membered ring containing the alkyl side chain is fully

156

reduced to a CH2. A side-by-side comparison of these three compounds drawn in the flat
representation shows these differences nicely (Figure 5.4).

Figure 5.4 Flat representation of Lepadiformine, Fasicularin, and Cylindricine C

     

‘ ‘ /~ Ncs‘"
HO/ n-C6H13 HO n-C6Hl3 n-C6H13

Lepadiformine (239) Cylindricne C (228), Fasicularin (240)

Q
‘

There have not been significant studies reported on the the biological activity of
Cylindricine C but it has been shown that a 3:2 mixture of Cylindricine A and B have
some toxicity in the brine shrimp assay.75 However, Fasicularin has been shown to be an

A73, 76

alkylating agent and has the ability to damage DN and Lepadiformine has shown
moderate cyctotoxicity against several cancer cell lines and has been shown to block the
cardiac muscle Kir channel.” 77' 78 In any event,ithe Cylindricine family, due to unique
structural features, has gained significant attention in the literature and many syntheses
have been accomplished in recent years. The focus of this chapter will be Cylindricine C
and presented herein will be a brief summary of prior efforts towards Cylindricine C as
well as the discussion of a different approach and efforts toward the total synthesis of
Cylindricine C using said approach.
5.2 Previous Syntheses of Cylindricne C

Five years after its structure was determined, the first synthesis of Cylindricine C
was reported by Molander”. He was able to accomplish the synthesis with the longest

linear sequence being 15 steps achieving a 10.4% overall yield from (S)-1,2,4—butanetriol

241 (Scheme 5.1). The key step in Molander’s synthesis was the CrCl2 reduction of the

157

Scheme 5.1 Molander Total Synthesis of (—)-Cylindricine C

 

 

O
9" Steps _ l) 002, H3O" _ N
“CA/V0" ————> 2) TBAF, THF 7 -. "Cw“
(S)-butane-I ,2,4~triol 45% ‘\OH
241 (—)-Cylindricine C
228

10.4% overall yield

15 steps

azide 242 followed by Michael addition of the resulting amine to form the tricyclic core
of (—)-Cylindricine C.

It was not until 5 years after Molander’s synthesis that Barry Trost at Stanford
University contributed his synthesis of Cylindricine C80. Trost started with 1,7-octadiyine
(243) and was able to complete the synthesis in 11 steps achieving 11.7%‘overall yield
for the longest linear sequence (Scheme 5.2). The key step in his synthesis was the

Scheme 5.2 Trost Total Synthesis of (+)-Cylindricine C

 

O
: steps :
_. [CpRu(CH3CN)3]PF6_
’ 10% water / acetone . NHBoc
I : 60°C, 2h ' g
1,7-octadiyne NHBOC 90 % 245 \OTBDPS

\
\

OTBDPS

243 244

...psl l 1

O

"'C6Hl3 N
HO
(+)-Cylindricine C
228
11.7% overall yield
11 steps

158

ruthenium-catalyzed hydrative diyne cyclization of 244 to give 245 in 90% yield. The
enone 245 could then be further modified by a subsequent Michael addition to give the
tricyclic core, which upon deprotection gave Cylindricine C.

Concurrently in the past 3 years, Kibayashis"83, Ciufolinig" 85 and Hsungm88 have
also been working towards the total synthesis of Cyindricine C, each having success
taking very different synthetic approaches. In Kibayashi’s first total synthesis83 (Scheme
5.3), he prepared (S)-N-Boc-2-pyrrolidinone 247 from the amino diacid 246

Scheme 5.3 Kibayashi’s First Total Synthesis of (+)-Cycindricine C

steps steps I
HOZCVYCOZH 8110\"(k BnO\w"O\‘“W
NH2 _, I _, I
Boc 3°C CHO
246
247 248

pyrrolidine, 4A M. s.
60% toluene, reflux
then 50% AcOH, RT

 

0
steps BnO
n-C6HI3 N ‘ ‘— \w“ I?!“
HO 3°“ CHO
(+)-Cylindricine C 249
. 228 .

2.6% overall yield

20 steps

after which was modified to the advanced intermediate 248. The key step was then the
spirocyclization of 248, which was carried out in the presence of pyrrolidine and 4A
molecular sieves to give the desired spirocyclic compund 249 in 60% yield. Modification

of the aldehyde to give 250 introduced a substrate that was able to undergo a Michael
addition followed by deprotection to give (+)-Cylindricine C. A second synthesis by

Kibayashi82 (Scheme 5.4) was accomplished using the same starting material (246).

159

Scheme 5.4 Kibayashi’s Second Total Synthesis of (+)-Cycindricine C

/ / n 'C6H l 3 QB"

steps
HOZCVYCOJ! _, Bn0\‘"_.&o 251 Nl-lBoc

 

247

n-C6Hl3 N
HO
(+)-Cylindricine C

228
3.5% overall yield

18 steps

 

Grignard addition to (S)—N-Boc-2-pyrrolidinone (247) gave the ring opened advanced
intermediate 251, which underwent subsequent cycilzation by treatment with formic acid
to give the conjugate spirocyclization product 252. Epoxidation, followed by epoxide
Opening gave a diol, which was then mesylated selectively at the alcohol farthest away
from the spirocycle to give advanced intermediate 253. The mesylated oxygem atom was
then displaced by the nitrogen leading to the tricyclic core, and subsequent oxidation and
deprotection gave (+)-Cylindricine C in 3.5% overall yield over '18 steps.

Ciufolini used a different approach starting with D-Homotyrosine85 254 (Scheme

5.5). D—homotyrosine could be modified to the N-mesylated and TBDPSO-protected

160

Scheme 5.5 Ciufolini’s Total Synthesis of (—)-Cylindricine C

Ms
HO steps HO Hill steps MSN
0 ———> OH ———>
_____._—> ——>
OH —__> ——>
255

OTB DPS

 

 

HZN it 0 256
254
l l lsteps
O O
StePs OTBDPS ' "‘C6H13\( 0 o OTBDPS
N ‘— 1)DBU,DMF ,m “s”
"'C6HB ‘— ¢ . _ . ‘ ‘N
<—— "u. 2) Bls(p|nacol|y|)diboronate. H".
OH CuCI. KOAc. RT
_ - ' ' ' 86%
( )CyllndncmeC 258
228 257
l4.6% overall yield
18 steps

intermediate 255 and cyclization to 256 was accomplished by oxidation in the presence
of DIB in hexafluoro-Z—propanol. After protection of the alcohol, 256 was transformed
into 257 several steps including treatment with KHMDS at — 100°C first, followed by
treatment with PhSH and BF3-OEtQ and subsequent reduction with Raney Ni to perform
the desulferization. After desulferization, treatment with base and (:t)-l-octene oxide in
the presence of BFg-OEt2 led to a compound, which could finally be oxidized under Dess-
Martin conditions to give the advanced intermediate 257. The boronate 258 could then
be prepared from 257 by treatment with DBU followed by bis(pinacollyl)diboronate.
The last steps in Ciufolini’s synthesis involved an oxidation and deprotection to give
(—)-Cylindricine C in 14.6% overall yield (18 step longest linear sequence).

Richard Hsung has published two different approaches toward Cylindricine C. In
his first approach he starts from D-Pyroglutamic acid” 88 using an aza-Prins cyclization

and a Wharton rearrangement (Scheme 5.6).

161

Scheme 5.6 Hsung’s Total Synthesis of (+)-Cylindricine C from D-Pyroglutamic

Acid

1
1)
%n-C H OH
OH steps OTBDPS 261 6 B Y I \ "'C6Ht3
M ——> t-BuLi.THF NBoc
N O ' N t

 

 

0 H —————- o Boc 2) Hcozn : THF : toluene OTBDPS
D—Pyroglutamic Acid 260 — 10°C. 3 h 262
259 1
l 1 steps
0 O
l)TFA/CH2C12(1:1) ,
5 A —10°CtoRT,2h t 5
ncénl3 N *2) DBU. toluene, RT NB“ I
(epimerization at C5) OT BDPS n-C6H'3
"0 263
(+)-Cylindricine C
228
8.3% overall yield
12 steps

The nitrogen of D-Pyroglutamic acid was protected with a Boc group and reduction of
the carboxylic acid gave an alcohol, which was protected to give 260. Ring opening of
260 by addition of the lithiated diene 261 gave a compound that subsequently underwent
the aza-Prins cyclization resulting in theiadvanced intermediate 262. Derivatization of
262 alloWed for a Wharton rearrangement to give a more advanced intermediate 263.
The Boc group was then removed and the 1.4—addition took place readily to give the C5-
epimeric tricyclic core of Cylindricine C. Treatment with DBU in toluene allowed for
the epimerization at C5 to give (+)-Cylindricine C in 8.3% yield over 12 steps.

Hsung’s second approach used an aux-[3+3] annulation strategy86 to prepare (—)-

Cylindricine C (Scheme 5.7). This strategy proved to be a much longer process (22

162

Scheme 5.7 Hsung’s Total Synthesis of (—)-Cylindricine C from L-Serine

steps steps
NBoc ———>
HzN'“
0::3 ——> O\/\/—_—> ——->

 

L-serineH 265 TBDPSO
264
26l6lnBu—l steps
0
N A Na(OAc)3Bl-l, HOAC HO 0 steps WTBDPSO
n-CeHn‘ CDCl3, RT, 21 h
then reflux 40 min N \ 4——— AN \
OH n-C6Hl3
(—)-Cylindricine C 268
228
4.5% overall yield
22 steps

steps inthe longest linear sequence), but the overall yield was still 4.5%. This synethesis
started from L-serine (264), which was easily transformed to the vinyloxazoline 265.
Several organic transformations were then needed to prepare precursor 266 which was
setup for the key intramolecular [3+3]-aza-annulation step, which upon treatment with
0.5 equivalents of piperidinium acetate and heating to 150°C for 12 hours gave the
desired annulation product 267. Going through a chlorohydrin intermediate, 267 was
transformed into the desired ketone 268 by treatment with NCS and t-BuOI-I/HZO (1:1)
followed by TPAP/NMO in the presence of 4A molecular sieves. Finally, a facile
reduction of 268 gave (—)-Cylindricine C.

The most recent and shortest synthesis was published in 2006 and was carried out
in the laboratories of Shibasaki89 using a catalytic asymmetric Michael reaction with a

two-center organocatalyst (Scheme 5.8). Pimelic acid (269) was easily modified in two

163

Scheme 5.8 Shibasaki’s Total Synthesis of (+)-Cylindricine C

HO '
O 2 steps 0 O
5 C6Hl3

pimelic acid 270
269 (S,S)-TaDiAS (10 mol%) Pb NVCOan
C52C03, chlorobenzene /
—40°C, 66 h Ph 271
5
0 \ C«5}‘13

  
 

CSA, LiCl

 

   

 

L O
CICHZCH2CI. . fit
5 o .- /
H0”: 50 C BnOZC‘ N Ph
. . , C H 51% yield
(+)-Cylmdncrne C 273 6 ‘3 272 4
228 84% yield
6 steps overall 82% cc
13.2% yield
last 4 steps ph

M2 /—C6H4-4—Me

N+
O \—C6H4-4—Me ZBF‘
O ”I /_C6H4-4-MC 4
.3 I N+

I

Me’ ¥C6H4-4—Me
Ph
(S,S)-TaDIAS

 

 

 

steps giving the ene-dione 270, which was treated with the imine 271 in the presence of
10 mol% (S,S)-TaDiAS to give the asymmetric Michael addition product 272 in 84%
yield and 82% ee. Cyclization was then accomplished by treating 272 with CSA in the
presence of LiCl to give 51% yield of the desired isomer 273. Finally, deprotection and
reduction of the advanced intermediate 273 gave (+)-Cylindricine C (13.2% yield from

270).

164

As can be seen in the discussion above, Cylindricine C has definitely gained
significant attention in the past several years. Molander, Trost, Kibayashi, Hsung,
Ciufolini and Shibasaki have all published unique syntheses, but choosing which is the
best, is a matter of opinion. There are several aspects to be considered when determining
the attractiveness of a particular synthesis. The two most important aspects are probably
the length and overall yield of the synthesis. With respect to the number of steps
involved in a synthesis, of course the shortest synthesis will be the more desirable one.
However, if the materials are very expensive, a longer synthesis might become more
attractive if the synthesis is more efficient. Of the syntheses of Cylindricine C reported to
date, Ciufolini has accomplished the synthesis giving the best overall yield, as he was
able to complete the synthesis in 14.6% yield from from D-Homotyrosine. This synthesis
took 18 steps, which is third largest number of synthetic steps needed by any of the
scientists that have published a synthesis Cylindricine C. Interestingly, however, the
shortest synthesis (6 steps) that was done by Shibasaki gave a slightly lower overall yield
(13.2% in the last four steps). This was only 1.4% lower than the former, so of the two,
the latter would likely be the attractive with respect to length and overall yield. The latter
also used an organocatalyst that was not commercially available, which may take
significant time and money to prepare, which could make this synthesis slightly less
attractive. All things considered, an attractive synthesis should involve few synthetic
steps with high yields starting from cheap readily available materials.

5.3 Retrosynthetic Analysis
The goal of the present work was to develop a new synthesis of Cylindricine C

that was not only shorter, but also more efficient than the previously reported syntheses.

165

The aza-Diels-Alder reaction is an atom economic and highly desired reaction to utilize
because of its ability to form carbon-carbon and carbon-nitrogen bonds in one reaction.
Another attractive feature of the aza-Diels-Alder reaction is if a concerted pathway is
followed, a stereoselective reaction will occur allowing for relative stereochemistry to be
set. In the retrosynthetic analysis (Figure 5.5), it was envisoned that the aza—Diels-Alder

Figure 5.5 Retrosynthetic Analysis

 

 

O 0 ‘
l ,4-addition
N
"'C6H13 > .- I.
"'C6H130‘ N
HO 1',
Cylindricine C 275

228

H Route A

, p TMSO OTMS

 

reaction would allow for the formation of the fused aza-decaline system in one
transformation. Route A-employs the aza—Diels-Alder reaction of an iminodienophile
276 and the known bis-TMS diene9097 278. Using the appropriate organocuprate, a 1,4-
addition to the Vinylogous amide produced from this aza-Diels-Alder reaction could be
envisioned to add the butenyl side chain. After incorporaton of the butenyl side chain, a
deprotection of the nitrogen and treatment with I2 and water should afford Cylindricine C.
A shorter synthesis could be envisioned using an alternate diene is outlined in route B.

This route undergoes an aza-Diels-Alder reaction of an iminodienophile as well, but in

166

this route, the diene already contains the butenyl side chain. Deprotection would give the
cyclization precursor 274 as discussed immediately before. In this retrosynthetic
analysis, the stereochemistry could be set in the initial aza-Diels—Alder reaction, however,
in route A, the butenyl side chain needs to be added cis to the n-hexyl side chain.
Literature precedent exists98 for the addition of a cuprate to a Vinylogous amide
containing a Cbz-protected nitrogen, where the cis-addition of an alkyl side chain
preferentially occured. The reaction in that report deals with a di-substituted olefin and
our proposed synthesis deals with a tetrasubstituted olefin, which could potentially result
in reactivity and selectivity problems. Nonetheless, the end cyclization process is known
on similar systems99 and should not be a problem.

An issue that arises from the proposed synthesis is the imine needed is one that
has to be prepared from heptanal. It is known that primary imines are not very stable and
are prone to isomerization (Figure 5.6). In 2005”, Kobayashi reported that the imines

Figure 5.6 Isomerization of Primary Alkyl Imines to Enamines

 

isomerization - H
N

N\ _

prepared from primary aliphatic aldehydes and benzoyl hydrazine were stable, and no
isomerization occurred even under his zirconium Lewis acidic conditions. He showed
that these imines could be used effectively in the aza-Diels-Alder reactions with
Danishefsky’s type dienes with the catalyst prepared from Zr(OnPr)4 and

tetraiodoBINOL (Scheme 5.9). Not only did this report by Kobayashi show that this

167

Scheme 5.9 Aza-Diels-Alder Reaction of Hydrazine Imines

 

o
H "Zr" / tetraiodoBINOL (281) _

N , N T 1’“ OTMS catalyst (10-20 mol%) l

I + M b R N
R\/' 0 on' 10-78% yield I

279 R' = Me (31) 53-95% ee "N P“

R' = t-Bu (280) O
282

class of imines could be easily prepared. but also indicated that they might be useful in the
asymmetric Synthesis of Cylindricine C using his zirconium catalyst system.
5.4 Efforts Toward Cylindricine C
5.4.1 Catalytic Asymmetric aza-Diels-Alder Reaction

The zirconium catalyst utilized by Kobayashi seemed promising for facilitating
the aza-Diels-Alder reaction that could eventually lead to the asymmetric synthesis of
Cylindricine C. Before the zirconium catalyzed aza—Diels-Alder reaction could be
attempted, tetraiodoBINOL (281) was prepared along with several other BINOL

derivatives (Figure 5.7). In order to ensure tetraiodo-BINOL was properly prepared,

168

Figure 5.7 BINOL Derivatives

   

(S)—BINOL(59) (S)—6,6'dibromo-BINOL (282)

   

(S)-6,6'di(trimethylsilyl)-3,3'diiodo-BINOL (284) (S)-3,3',6,6'-tetraiodo-BINOL (281)

thereaction of imine 2793 with diene 31 was repeated with Kobayashi’s catalyst using
his conditions. In addition, having prepared all the above-mentioned BINOL derivatives,
they in addition to (S)-VAPOL (130), (S)-VANOL (129), and (S)-3,3’-dinitro-VANOL
(285) were all screened (Scheme 5.10) to determine if they were viable ligands for this

reaction.

169

Scheme 5.10 Screening of Ligands for the aza-Diels-Alder Reaction Using

 

 

 

Kobayashi’s Conditions
0
H Sli/ zr(on-l>r)4 (20 mol%) l
\N I NT Ph + 0’ \ Ligand (24 mol%.) > N
©/\/\ 0 Mom TBDit/III; / :1»: :41). “111th
279a 31 281a 0
Entry . - Ligand Yield (%) ee (%)
l (S)-VAPOL (130) 0 ND.
2 (S)—VANOL (129) 0 ND.
3 (S)-BINOL (59) 0 ND.
4 (S)-6,6’-dibromo-BINOL (282) 6.3 42.2 -
5 (S)—6,6'-di(trimethylsilyl)—BINOL (283) <5 ND.
6 (S)-6,6'-di(trimethylsilyl)-3,3'-diiodo-BINOL (284) 1 1.7 74.5
7 (S)—3,3',6,6'-tetraiodo-BINOL (281) 41 94
8 (S)—3,3'-dinitro-VANOL (285) 21 87

 

The electron neutral ligands, (S)-VAPOL, (S)-VANOL, and (S)-BINOL (entries
1-3 respetively) all failed to produce any of the desired Diels-Alder adduct 281a. Using
the ligand 283, with trimethylsilyl groups in the 6,6’-positions, the reaction only gave
trace amount of the desired product 281a (entry 5). When electron-withdrawing groups
were incorporated, the reaction was accelerated and produced enough material to
determine theee of the product from the reaction. The BINOL derivative 282 containing
bromine atoms in the 6,6’-positions (entry 4) gave 6.3% yield and 42% ee, which was
significantly inferior to the results published by Kobayashi for tetraiodoBINOL 281. The
selectivity could, however, be increased to 74.5% ee when the (S)-6,6'-di(trimethylsilyl)—
3,3'-diiodo-BINOL 284 derivative was used as the ligand (entry 6). This reaction was
reluctant to turn over and gave only abOut 12% yield. Repeating exactlythe conditions

170

that Kobayashi reported for tetraiodoBINOL (281), the reaction gave 46% yield and 92%
ee (entry 7, average of two runs), which is very close to what he achieved (70% yield and
91%ee using 10 mol% “Zr” and 12 mol% 284). Interestingly, it was found that when
3,3’-dinitroVANOL 285 was used, the reaction gave 21% yield and 87% ee. Although
the reaction did not turnover, the asymmetric induction was excellent when compared to
Kobayashi’s tetraiodoBlNOL ligand and it may be worth while to explore this ligand in
more detail to find conditions to improve the selectivity as well as induce turnover. One
last attempt was made for this reaction using the optimal conditions developed for the
aza-Diels-Alder reaction of 150 with Danishefsky’s diene (see chapter 2). The VAPOL-
B(OPh)3 catalyst (Scheme 5.11) prepared using lOO'mol%

Scheme 5.11 Reaction using (S)--VAPOL/B(0Ph)3 as the Catalyst

 

(S)—VAPOL (130) + B(OPh)3 - CH2C12 . 0-5 mmHg
10 mol% 100 mol% 55 °C, 1h 55 °C, 0.51)

3

I cat“

at*

 

 

 

 

 

5 Ph Si< DCM/t l (1 1)
\ ’ o uene :
N’ T + O o

M —.45 Q24}.

0 OMe \n/Ph
279a 31
- 2818 0
13.3% yield
9% cc

B(OPh)3 and 10 mol% (S)-VAPOL was used and the reaction only gave 13% yield of the
desired product 281a with 9% ee. Although this catalyst system Worked well for
benzhydryl imines, it is clear from this reaction that there is no utility in using this
catalyst when the dienophile is a benzoylhydrazone. ‘ i
From the series of experiments described above, it was decided that

’ tetraiodoBINOL was indeed the superior ligand for the reaction using the hydrazone 279a

171

prepared from dihydrocinnamaldehyde and benzoyl hydrazine with Danishefsky’s diene.
The aza-Diels-Alder reaction desired for the synthesis of Cylindricine C, however,
requires the use of the imine (279b) prepared from benzoyl hydrazine and heptanal and
the bis-TMS diene 278. Although tetraiodoBINOL proved to be better for Kobayashi’s
reaction, it was not clear that this would be the case for the desired system so attempts '
were made to accomplish the reaction asymmetrically using Zr(OnPr)4 with BINOL,
VAPOL, VANOL, and all the previously prepared BINOL derivatives (Scheme 5.12). It
Scheme 5.12 Catalytic Asymmetric aza-Diels-Alder Reactions using the bis-TMS

Diene 278

\l l/ o

/Sl\ ,Sl\
0 0 Zr(On-Pr)4 (286. 20 mol%)
H \ - '
W ,N Ph Ligand (24 mol%) ; .
N r + N
0

 

TBDME/ DME (4zl)

 

 

279b 278 — [0 °C, 53 h 287 "NT“!
0
Entry Ligand Yield 287 (%)

l (S)-VAPOL (130) O

2 (S)-VANOL (129) O

3 (S)-BINOL (59) 0

4 (S)-6,6'-dibromo-BINOL (282) O

5 . (S)-6,6'-di(trimethylsilyl)-BINOL (283) 0

6 (S)-6,6'-di(trimethylsilyl)-3,3'-diiodo-BINOL (284) O

7 (S)-3,3',6,6'—tetraiodo-BINOL (281) 0

 

was dissappointing to find none of these ligands in conjunction with Zr(OnPr)4 catalyzed
the reaction to give the Diels-Alder adduct 287 using the imine 279b and diene 278.
Knowing that tetraiodoBINOL was superior to the other ligands, two more reactions were

carried Out using this ligand at higher temperatures (see Experimental). Even heating the

172

reaction to reflux in toluene (see experimental section for details) for 25 hours failed to
produce any of the desired product. The diene 278 in this reaction is significantly bulkier
than Danishefsky’s diene (31) as it contains two trimethylsilyloxy groups. In addition to
containing more steric bulk, the bis—TMS diene has one of the trimethylsilyloxy groups
locked in the cis position (Figure 5.8), which makes the orbital overlap that is

Figure 5.8 Danishefsky’s Diene Versus bis-TMS Diene

§

5';
\ 0

§
I"

Q.
Q
0

\
(£—
)é}

>4». \ \i‘i'u“ >4» \ Le.»
x R E}

Danishefsky's Diene (31) bis-TMS Diene (278)
necessary for the reaction even more difficult. This effect is enhanced even more when a
sterically bulky catalyst is bound to the imine. The lack of success of these reactions is
probably due to the fact that when the 'imine is coordinated to the zirconium with these
relatively large ligands, the approach of the diene is too hindered and cannot get close
enough for sufficient orbital overlap to occur and hence the reaction cannot happen. The
outcome of these reactions was disappointing because the asymmetric synthesis would
have to await further investigation of other chiral catalysts, however the synthesis of (1:)-
Cylindricine C was still persued.
5.4.2 Efforts Toward the Synthesis of (:t)-Cylindricine C
5.4.2.1 Optimization '0: the aza-Diels-Alder Reaction

Kobayashi’s report in Tetrahedron“, described the use of BF3-0Et,z (288) to

perform the racemic aza-Diels-Alder reaction of the hydrazone 279a and Danishefsky’s

diene. Even for the racemic synthesis of Cylindricne C, the aza-Diels-Alder reaction is

173

the key step, so several reactions were carried out to try and find optimal conditions using
BF3'OEt2 (288) as the catalyst for the reaction of the bis-TMS diene 278 with imine 279b
(Scheme 5.13).

Scheme 5.13 Optimization of the Racemic aza-Diels-Alder Reaction

 

H TMSO OTMS 0
H \
W ,N Ph BF3‘OEt2 (288, 1.1 cq)
N T + :
O CHZCIZ N
279b I
278 temperature 287 HN Ph
reaction time 1?

 

Entry Condition Temperature (°C) Reaction time (h) Yield 287 (%)

 

1 AA‘ —453 2 0
2 BB’ —454 20 0
3 CC1 ' -—455 2 29
4 DD' —78 to —456 19 29 '
5 BE —78 to RT7 19 36
6 1:1:2 —78 to ——456 19 32
7 Go"8 —-78 to RT7 20 - 64
8 H11"8 —78 to RT9 _ 20 ' 58.5
9 11“"10 —78 to RT" 20 42
10 11"” -—78 to RT7 20 68
11 JJW —78 to RT7 20 63
12 KKmm —78 to —20 to RT” 19 >40

 

. 1) Reaction run at 0.1 M with respect to imine. 2) Reaction run at 0.025 M with respect to imine. 3) [mine
and BF, combined at room temperature then cooled to —45°C before the diene was added. 4) Imine
cooled to —45°C before BF3 was added followed by. the diene. 5) [mine was cooled to —45°C and then
the diene was added followed by BF, 6) [mine was cooled to —-78°C and then the diene was added
followed by BF, and then warmed to —45°C. 7) Imine was cooled to -78°C and the diene was added
followed by BF3 then warmed slowly to room temperature. 8) Diene redistilled immediately before use. 9)
imine was cooled to —78°C and the diene was added followed by BF3 then warmed immediately to room
temperature. 10) BF3 redistilled immediately before use. 11)0.45 g imine used. 12) 6.21 g imine used.
13) 16.3 g imine used. 14) The imine was cooled to —78°C and the diene was added followed by BF3 then
the reaction was allowe to warm slowly to ——20°C overlS hours and then to room temperature for 4 hours.

The first two reactions involved the combination of the imine and BFa-OEt2 before

the addition of the diene (entries 1 and 2) and neither reaction gave the Diels-Alder

174

product 287. It was found that when the solution of the imine was cooled to 945°C first
and then adding the diene and BF3'OEt2 in that order, the reaction produced 29% yield of
287 in only 2 hours (entry 3). A subsequent reaction was then attempted where the imine
solution was cooled to —78°C before the diene and BF3-0Et2 were added (entry 4) and
then warmed to —45°C and no improvement was observed (29% yield) even when the
reaction was allowed to run for 20 hours. This reaction was repeated except that the
reaction was allowed to warm to room temperature after all reagents were added (entry 5)
and 36% yield of 287 was produced. Another reaction was performed at lower
concentration (0.025 M, entry 6) using the conditions where the reaction warmed from
—78°C to -—45°C, but still, the reaction gave a low yield (32%) of the desired product
287. The highest yield obtained (36%) was not satisfactory since this was the key step in
the synthesis of Cylindricine C. In an attempt to increase the yield, a new batch of the
diene was prepared and distilled immediately before use. Repeating the reaction where it
was cooled to —-78°C before the diene and BF3-OEQ were added and then wartning to
room temperature, using the freshly prepared diene, the reaction gave 64% yield (entry
7). The yield dropped to 58.5% (entry 8) if the temperature was immediately allowed to
.warm to room temperature rather than slowly over several hours. The same conditions as
for entry 7 were followed only redistilled BF3-0Et2 was used (entry 9) and the reaction
only gave 42% yield of 287. Lastly, three reactions were then carried Out on larger scale
(0.45g, 6.21g, and 16.3 g of imine 279b), The reaction performed on a 0.45 g scale gave
68% yield (entry 10), 6.21 g scale gave 63% yield (entry 11), and 16.3 g scale gave at
least 40% yield (entry 12). The 40% yield in the last reaction was after one

recrystalization and more of the crude product was isolated from the mother liquor but

175

another recrystalization was not accomplished at that time. It is encouraging that the
reaction could be scaled up and this scalability makes this route even more attractive.
5.4.2.2 1,4-Addition to Incorporate the Butenyl Side Chain

With plenty of the Diels-Alder product 287 in hand, attempts were made to find
conditions to facilitate the 1,4-addition of the butenyl side chain. Work publiShed by
Martin98 showed that the reaction took place between an organomagnesiumcuprate
prepared from a mixture of vinylmagnesium bromide, MeLi, and CuCN (1:121) with the
Cbz-protected Vinylogous amide 289 (Scheme 5.14). Not only did the reaction work

Scheme 5.14 cis 1,4-Addition as Reported by Martin98

hngr
0 A + MeLi + CuCN 0
65 290 291 292 61] ,
RP" pg 1 : l : 1 44%: FPv‘ bl ””Téép
0%0 0&0

289 293

 

well, but the major isomer of the product was where cis-addition of the vinyl side chain
occurred. The Vinylogous amide (278) resulting from aza-Diels-Alder using the present
strategy contains hydrazine functionality rather than the Cbz carbamate functionality.
The hydrazine is not as electron withdrawing as the Cbz group, but in order to minimize
the amount of steps needed in the synthesis, several attempts were made to facilitate the

1,4-addition on the hydrazine adduct 287 (Scheme 5.15). Two reactions were first

176

Scheme 5.15 1,4-Addition to the Vinylogous Hydrazine Using Cuprate 1

 

 

 

 

O
I. + Mumps 30mm > ‘
temperature
5 If Cuprate 1 Lewis Acid
HN Ph
\n/ reaction time
0 O
287 288
Cuprate 1 o ' . . reaction .
Entry (equiv) solvent Temperature( C) LeWIs Acrd time (h) yield (%)
l 1.5 E90 —78 to —25 none 19 O
2 1.5 EtzO —78 to —25 none ' 16 O
3 2.5 E90 —78 to ——10 none 22 O
4 1.5 1393:3192 —78 to —50 to —25 Zr(OnPr)4 67 0

 

attempted using 1.5 equivalents of theicyanocuprate (cuprate 1) prepared in situ from
’ two equivalents of the butenyl lithium and one equivalent of copper cyanide. Both
reactions were allowed to warm from —78°C to —25°C and one was stirred for for 19
hours and one for 16 hours, but neither reaction produced the desired product 288. When .
2.5 equivalents of the cyanocuprate was used and the reaction allowed to warm to
— 10°C (entry 3) also failed to produce the desired product. The lack of reactivity codld
either be due to steric hindrance of the tetrasubstitutted double bond or the electron n'ch
nature of the double bond.‘ It was thought then that the benzoyl hydrazine in the product
287 could coordinate to Zr(OnPr)4 (as in the aza-Diels-Alder published by Kobayashi)

and pull electron density out of the Vinylogous amide’s carbon—carbon double bond.

177

When Zr(OnPr)4 was added to the reaction and the resulting mixture warmed slowly
from—78°C to —25°C over 67 hours (entry 4), no reaction was observed.
5.4.2.3 N-N Bond Cleavage and Protection with Cbz I

The 1,4—addition did not take place on the hydrazine moiety 287 but the work
done by Martin (Scheme 5.13) indicated that if the nitrogen was protected with a Cbz
group, the reaction can occur readily and give the desired cis stereochemistry as the
major product. Before the 1,4—addition could be attempted, the hydrazine N-N bond had
to be cleaved and the Cbz group incorporated. Initial efforts to reductively cleave the N-
N bond were done using SmI2 (Scheme 5.16). The first two attempts (entries 1 and 2)

Scheme 5.16 Reductive Cleavage Using Sml2

 

 

 

O O
I. Sm!2 (239, 2.2 equiv.) _ .
N solvent 7 I
Hill Ph room temperature I”!
287 T reaction time 290
0
Entry solvent reaction time (h) yield 290 (%)
1“ THF 0.67 12
2“ THF/MeOH (3.5:1) 0.67 7.6
3" THF/MeOH (4: 1) 7 38

 

a) Sml2 was prepared using Sm° and 12 refluxing in THF.
b) Sm‘I2 was prepared using Sm° and C11212 at room temperature in THF.

used SmI2 that had been prepared by diluting samarium metal and iodine in THF and
refluxed until the solution turned blue. When only THF was used as the solvent (entry 1),
the reaction produced 12% yield of the free amine 29b in 45 minutes, however when
THF/MeOH was used as the solvent, the reaction only gave 7.6% yield. The last attempt

was carried out using Sml2 that had been prepared by stirring samarium metal in the

' 178

presence of CH212 at room temperature for several hours in THF. The resulting blue
solution was used for the reductive cleavage of the N-N bond and after 7 hours the
reaction produced 38% of the desired amine 290. This series of experiments indicated
that SmI2 was not a good choice for this reaction and another method was needed to
perform the reductive cleavage that would give higher yields.

In 1993, Pak published a method to efficiently mediate desulfonylation reactions
in methanol“). His method used 10 equivalents of magnesium with catalytic amount of
mercuric chloride as the reducing agent and was able to obtain the reduced product in
high yields. It was thought that this reducing agent may be able to facilitate the N-N
bond cleavage desired for production of the free amine 290. The reaction conditions
mentioned in Pak’s publication were vague and only mentioned that he used catalytic
‘ amounts of HgClz. In order to determine if this method would be feasible for the
cleavage of the N—N bond in 287, a few preliminary reactions were attempted (Scheme
5.17). When 10 equivalents of Mg‘and a few spatula tips of HgCl2 were used (entry 1),

Scheme 5.17 Initial Attempts for the Reductive Cleavage Using Mg° and HgCl2

 

o
Mg° (291, x equiv.) O
I ch212 (292, v amt.) _
’7' THF/EtOH(l:3) V I.
287 HN\n/Ph temperature, time a
0 290

 

Entry Mg° (equiv) HgClfilmt.) temperatrue reaction time (h) Yield 290 (%)

 

l 10 3 spatula tips RT 3 18.2
2 10 3 spatula tips 0°C to RT 3 16.7

 

3 3 0.7 equiv. RT 6 69

179

an exothermic reaction occurred when the HgCl2 was added. After 3 hours, TLC analysis
showed the starting material had been consumed but only 18.2 % yield of the product 290
was isolated. During the previous reaction, it was exothermic enough that the solvent
started to boil, so another reaction was done cooling to the hydrazine and magnesium to
0°C before the HgCl2 was added (entry 2). After the the addition of HgClz, the reaction
was warmed to room temperature and stirred for 3 hours, and only 17% yield of 290 was
obtained. One possible explanation of the low yield could be due to the large excess of
the reducing agent degrading the product after the N-N bond cleavage took place. With
10 equivalents of the reducing agent, it is not unreasonable to believe that reduction of
the carbonyl and possibly the double bond could also occur. In addition to over
reduction, it is possible that the excess reducing agent could also facilitate sOme radical
type polymerization reaction. Finally, a reaction was carried out using only 3 equivalents
of magnesium and 0.7 equivalents of the HgCl2 (entry 3) and the reaction proceeded
smoothly giving 69% yield of" 290 in only 6 hours. i

In order to determine the optimal amount of HgCl2 needed to give good
conversions, a series of experiments was conducted using HgCl2 amounts ranging from

0.1 equivalent to 1.0 equivalent (Scheme 5.18). It was found that the reaction was quite

180

Scheme 5.18 Determination of the Optimal Amount of HgCI2

o
Mg° (291, 3 equiv.) O
I. HgCl2(292,Xequiv.) _
If ‘ THF/ EtOI-I(1:3) ' N I.

287 ”NY P“ RT, 18.5 H

O 290

 

 

 

Entry HgCl2 (equiv) Conversion

1 0.1 <5
2 0.2 <5
3 0.3 19
4 0.5 80
5 0.6 70
6 0.8 >95
7 1 >95

 

slow (19% conversion or less) when less than 0.5 equivalents of HgCl2 were used.
However when 0.8 and 1.0 equivalents of HgCl2 were used, only trace amounts of the
starting material were observed in the crude 1H NMR after 18.5. hours. Since only
conversions were measured for these reactions and the isolated yields were not
determined, a reaction was carried out using 1.0 equivalent of the HgCl2 to determine the

isolable yield from this reaction (Scheme 5.19). Using these conditions on a 0.355 g

181

Scheme 5.19 Optimal Reduction Condition and Scalability

0
O

Mg° (291, 3.0 equiv.)
I HgCl2 (292, 1.0 equiv.)
N > I
N

 

, THF/ EtOH (1:3)
287 “NY P“ 0°C to RT, 17 h

0

H
290

 

Entry Scale (g) Yield 290 (%)

 

. 1 0.355 80
2 1.0 62
3 1.2 62
4 9.0 60

 

scale, the reaction gave 80% yield of the desired free amime 290. To determine the
scalability of this reaction, three other reactions were carried out on larger scale. When
the scale was increased to 1.0 g the yield dropped slightly to 62% (entry 2), but scaling
up even more to 9.0 g (entry 4), showed no detrimental effects and the reaction still gave
60% yield of the desired product 290.
5.4.2.4 Protection of the Amine with Cbz

With a significant amount of the free amine 290 in hand, the stage was set for
protecting the amine with the Cbz protecting group and ultimately attempting the 1,4—
addition with the Cbz protected product 294. Literature precedent indicated that either

'0' or Hunig’s base102 could be used as an efficient base for deprotonation

sodium hydride
of the nitrogen. Subsequent treatment of the nitrogen anion with Cszl (293) should
have given the desired Cbz protected Vinylogous amide 294, however, using these

cOnditions, the desired product was not obtained (see experimental). Fortunately, it was

182

found that treatment with nBuLi followed by the addition of Cszl gave the desired
product (Scheme 5.20) in 80% yield. It is worthy to note that the reagents all need to be

Scheme 5.20 Cbz Protection of the Amine

o ' o

 

o
THF _
a 22 h j;
290 1.0 equiv. 293 80% yield 0 0
1.0 equiv. 1.02 equiv. Ph)

294

added slowly at ——78°C before the reaction could be warmed to room temperature or the
yield of the reaction was significantly decreased. An additional reaction was done where
the reaction was only allowed to stir for 30 minutes after reaching room temperature, and
the yield for this reaction was 98% based on recovered starting material.
5.4.2.5 1,4-Addition of the Butenyl Side Chain Using the Cbz Protected Vinylogous
Amide
With the Cbz protected Vinylogous amide 294 in hand, the 1,4—addition of the
, butenyl side chain could be attempted. The three different cuprates depicted in Figure 5.9

Figure 5.9 Cuprates 1-3
MCUUZCN MCuLngCN __
Cuprate 1 Cuprate 2

were explored for this reaction. The lithium cyanocuprate (Cuprate 1) is the most
reactive of the three and only contains the butenyl groups so the possibility only exists for
a butenyl group to be incorporated. A less reactiVe lithium cyanomagnesiumcuprate

(Cuprate 2), similar to the one used by Martin,98 contains a methyl group also, which

183

may compete with the aliphatic butenyl group for the 1,4 addition. To alleviate this
problem, a second lithium cyanomagnesium cuprate (Cuprate 3) was prepared that
contains the “dummy” thiophenyl ligand. This group takes up one site on the cuprate but
will transfer much slower, only allowing for the addition of the butenyl group. Each of
these cuprates was then tested in the 1,4-addition reaction (Scheme 5.21). All attempts

Scheme 5.21 Screening of Cuprates l, 2 and 3

THF

temperature

ll

+ Cuprate

reaction time

 

 

Entry Cuprate (equiv) temperature (°C) reaction time (h) Conversion (%)

 

 

1 Cuprate 2 (1.5) —78 20 <5
2 Cuprate 3 (1.5) ~78 to RT 17 0
3 Cuprate 3 (1.5) —78 to —‘20 to 10 42 0
4 Cuprate l (1.5) —78 to RT 2 0
5 Cuprate 1 (1.5) ——78 to —20 53 58‘
6 Cuprate l (1.5) —78 to —25 19 0
7 Cuprate 1 (2.45) —78 to -—10 48 ' 38"

a) none of the desired product was isolated

using Cuprates 2 and 3 (entries 1-3) failed to install the butenyl side chain even when the
reaction was increased to room temperature. When cuprate 1 was used and the reaction
was allowed to warm slowly from --78°C to —20°C (entry5), 58% conversion was
observed, but none of the desired product 295 was isolated. A second reaction was
performed using 2.45 equivalents of cuprate 1 (entry 7) and the conversion of the reaction
was still low (38%), and again, the desired product‘295 was notiisolated. The results

from this set of experiments indicated that the reaction using the tetrasubstituted double

184
)

bond was going to be a bigger challenge than originally anticipated. Three different
Lewis acids (BF3°OEt2, ZnClz, and TMSCl) were then used in an attempt to activate the
double bond by decreasing its electon density (Scheme 5.22). BF3,-OEt2 and ZnCl2

Scheme 5.22 Lewis Acid Promoted 1,4-Addition

+ MCuLiZCN + Lewis Acid ELZO

—78°C to —50°cr

 

2
Cuprate l 54 h
1.5 equiv.

   

 

Entry Lewis acid Conversion (%)

 

1 BF3-OEt2 0
2 2110, 0
3 TMSC] 68.5

 

a) none of the desired product was isolated

(entries 1 and 2 respectively) both failed to facilitate the reaction and only starting
material was observed in the crude 1H NMR for these reactions. When TMSCI was used
(entry 3), 68.5% of the starting material was consumed, but none of the 1,4—addition
product 295 was isolated. I
5.4.3 Synthesis of Cylindricine C via Route B

After problems occurred with the originally proposed l,4~addition following route
A for the synthesis of Cylindricine C, the alternative route B was inveStigated. The
proposed route B (Figure 5.5) involved the use of the diene 227, which already has the
butenyl side chain incorporated. It was envisioined that if the aza-Diels-Alder reaction
were to occur with this diene, then the only other steps would be the deprotection and

cyclization. Presumably, the cleavage of the hydrazine N-N bond should not pose a

185

problem, as that issue was previously solved for route A, and as mentiomed earlier, the
final cyclization step is known99 and has been carried out on a similar substrate. One
attempt was made to perform the aza-Diels—Alder reaction using the alternate diene 227
(Scheme 5.23) under the optimal conditions found for diene 278 (scheme 5.13, entry 10)

Scheme 5.23 Aza-Diels-Alder Attempt Using Alternate Diene 227

 

\
H H OTMS
M ,N Ph + \ 313051.201 eq) >
N \n/ 01,02
0 —78°C to RT
279b
1.0 equiv. 227 .
1.2 equ1v

 

296
23% yield

The reaction proceeded to give 23% yield of the non-cyclized Mannich type product 296,
which is of course not the desired aza-Diels-Alder adduct. It is possible that heating the
Mannich product or treating it with a weak acid or base could facilitate the cyclization to
give the core decalin system of Cylindricne C but this was not attempted. Another issue
that may present a problem using this diene is if the process is not concerted, the relative
stereochemistry at the fused six-membered rings might not be exclusively cis after the
cyclization step. However, if this reaction can be optimized and the issues of
stereochemistry can be solved, then this route to Cylindricine C seems very promising
and is shorter than the originally proposed route A.

Efforts toward the racemic synthesis of Cylindricine C unfortunately were halted
at this point. It was found that the 1,4-addition of the butenylside chain is a difficult
problem and could be due to the more hindered tetrasubstituted double bond causing the
addition of the butenyl group to be unfavorable. It could also be that even with the Cbz

group on the nitrogen, the tetrasubstituted olefin is not electrophilic enough to allow the

186

1,4—addition to take place. A solution to the latter issue would be to put a more electron—
withdrawing group on the nitrogen and attempt the 1,4- addition again. The compounds
with a benzoyl group on the nitrogen and a 4-nitro-benzoyl group on the nitrogen were
synthesized (Figure 5.10) but the 1,4- addition on these compounds was not attempted. If

Figure 5.10 Alternate Substrates for the 1,4-Addition

 

o O 0 O
. No2
294b 294C
the 1,4-addition could be carried out on one of these compunds, the synthesis should be
easily completed by simple deprotection of the nitrogen followed by cyclization to
incorporate the final 5-membered ring in the tricyclic core of Cylindricine C.
Alternatively, route B also seems promising and if the aza-Diels-Alder reaction can be
optimized or the cyclization to be induced, this route may provide an even shorter

synthesis of Cylindricine C (6 steps) including the preparation of the diene.

187

CHAPTER 6

CATALYST STRUCTURE, CONCLUSIONS AND FUTURE WORK

6.1 Structure of the (S)-VAPOL-Boron Catalyst

In order to obtain a clear idea of how a catalyst is functioning and to be able to
predict the stereochemical outcome of an asymmetric reaction, one must have a good
understanding of the actual catalyst structure. The VAPOL—boron catalyst utilized in the
aza-Diels-Alder reaction and previously in the aziridination reaction was prepared using
an excess triphenylborate (23:1 B(OPh)3/VAPOL). It was originally assumed that the
excess B(OPh)3 was simply driving the catalyst formation to completion and that the
catalyst was formed by the replacement of two phenoxy groups on triphenylborate with
the two oxygen atoms on VAPOL (Figure 6.1). A related catalyst structure was
originally proposed by Yamamoto in 199318 for his catalyst prepared from BINOL and

B(OPh)3.

188

Figure 6.1 Original Proposed Catalyst

 

Ph ! on B(OPh)3

——————>
Ph I OH

+ 2 PhOH

 

 

 

 

Original Proposed
Catalyst

During the time that the work presented in this thesis was going on, studies in the Wulff
group were being directed at determining the exact structure of the active catalyst
prepared from VAPOL and B(OPh)3. Towards the beginning of these studies, mass
spectral data was obtained that indicated that the active catalyst‘species contained two

boron atoms (Figure 6.2).

Figure 6.2 Two Boron Catalyst Structure

 

 

 

 

 

Proposed Catalyst 2

With the assumption that the active catalyst contained two boron atoms, a method was
developed to prepare this catalyst by using 1 equivalent of VAPOL, 2 equivalents of

BH3-THF, 2 equivalents of phenol, and 1 equivalent of H20. This method was developed
189

because it uses the appropriate number of atoms needed to prepare the proposed catalyst
2. In addition, a method was also developed to prepare the originally proposed one boron.
catalyst species using 1 equivalent of VAPOL, 1 equivalent of BH3-THF, and 1
equivalent of phenol. The two catalysts could be distinguished from each other by
observing different chemical shifts for the bay proton in the 1H NMR. Interestingly,
recent data has indicated that neither of the proposed catalysts 1 or 2 described above are
actually the active catalyst in the aziridination reaction. Based on crystallographic data, it
is now believed that the actual active catalyst species contains three boron atoms (Figure
6.3). This catalyst contains a boroxazine ring with one of the

Figure 6.3 Boroxazine (11*) Catalyst

 

 

 

 

 

Active Catalyst Species

borons attached to two oxygens in the boroxazine ring and two oxygen atoms from
VAPOL. It was found that this boroxazine catalyst forms readily in the presence of an
imine and by just placing the imine. Specifically, it was found that the boroxazine
catalyst spontaneously forms when the imine, VAPOL, and triphenylborate are simply
placed in a flasak and dissolved. This VAPOL-boroxazine catalyst is negatively charged

with a proton as the counter ion. Knowing that this is the actual catalyst structure, one

190

can imagine that the imine is either activated by the proton (Br¢nsted acid) or possibly it
could be one of the neutral boron atoms in the boroxazine ring (Lewis acid).

In order to determine how the imine is coordinating to the catalyst, crystals were
grown by Gang Hu in the group using the imine 325 prepared from benzaldehyde and
bis(4—methoxy—3,5-dimethylphenyl)methanamine (MEDAM). The crystals were grown
by first preparing the catalyst by adding B(OPh)3 (3.0 equiv.), VAPOL (1 equiv.) and
water (3.0 equiv.) to a Schlenk flask and heating to 80°C in THF. After 1 hour of
heating, the flask was placed under high vacuum and heated to 80°C to remove any
volatiles. To this active catalyst was then added the MEDAM imine and the crystals
were grown using a mixture of pentane, CHzClz, and CDCl3 as the solvent (see crystal
structure in Figure 6.4, solvent

Figure 6.4 Crystal Structure of the Three Boron Catalyst and MEDAM Imine

   

Diene Approach

molecules removed for clarity). As can be seen from the crystal structure, it is indeed the

proton that is coordinated to the imine. So, after many years of thinking that the catalyst

191

was a Lewis acid, it appears now that it is in fact a Br¢nsted acid that is facilitating these
reactions. The crystal structure shown was grown using the (S)-VAPOL ligand, which
was the ligand used in aza-Diels-Alder work discussed in this thesis. It was found that
the absolute stereochemistry of the product 170 was S (see Scheme 3.8 and 3.9) when
using (S)-VAPOL and this is precisely what the crystal structure would predict.
Although the catalyst is actually a “chiral proton” and it is a Br¢nsted acid catalyzed
reaction, it does not change the fact that a new successful method has been developed for
the aza-Diels—Alder reaction.
6.1.1 aza-Diels-Alder Reaction Using Alternate Catalyst Preparations

Previously, mention was made of studies carried out in the'Wulff research group
by Gang Hu to optimize conditions to prepare the VAPOL—boron catalyst containing one
boron atom (Figure 6.1) as well as the catalyst with two boron atoms (Figure 6.2). It was
decided to study these catalysts to determine the effectiveness of each for the aza-Diels-
Alder reaction (Scheme 6.1). All the reactions performed for this study were carried out
using the optimal conditions developed and described in Chapter 2 only 10 mol% of the

catalyst prepared in their respective methods.

192

Scheme 6.1 aza-Diels-Alder Reaction Using the One (B1) and Two Boron (B2)

 

 

Catalyst
Ph SI /
\N/l\ph + O’ l\ 10 mol% catalyst ¥
MOM CH2C12/toluene (1:1) ‘
e O
150 31 .—45 .C, 24h
1,0 equiv. 2.0 equiv. c011 add1t10n flask

added over 3 h

in 3 mL solvent

 

Entry Catalyst Yield 151 (%) ee 151 (%)

 

1 B 1a 35 30
2 B2b 82 71
3 B2c 31 75

 

a) catalyst prepared using 1.0 eq. BH3-SMe2, 1.0 eq. (S)-VAPOL, and 1.0 eq. phenol by
heating to 100°C in toluene for 1 hour and then heated to 100°C under high vacuum for
30 minutes.
b) catalyst prepared using 4.0 eq. of B(OPh)3, 1.0 eq. (S)-VAPOL, and 1.0 eq. of H20 by
heating to 80°C in toluene for 1 hour and then heated to 80°C under high vacuum for
30 minutes.
c) catalyst prepared using‘2.0 eq. BH3-SMe2, 1.0 eq. (S)-VAPOL, 3.0 eq. phenol, and 1.0 eq. H20
by heating to 100°C in toluene for 1 hour and then heated to 100°C under high vacuum for
30 minutes.

The first reaction was attempted using the Bl catalyst prepared using 1.0 equivalents of

BH3-SMe2, 1.0 equivalents of (S)-VAPOL, and 1.0 equivalent of phenol (entry 1). The

reagents were combined and heated to 100°C in toluene for 1 hour and then heated to

100°C under high vacuum for 30 minutes. When the reaction was attempted using this

catalyst, only 35% yield of the desired product 151 was obtained with 30% ee. The

second two reactions were attempted using the B2 catalyst. The reaction in entry 2 was

carried out using the catalyst prepared by heating 4.0 equivalents of B(OPh)3, 1.0

equivalents of (S)-VAPOL, and 1.0 equivalent of water to 80°C for one hour and placed

under high vacuum at 80°C for 0.5 hour. This reaction produced 82% yield of the

desired product 151 with 71% ee. The last reaction was attempted using another method

193

to prepare the B2 catalyst. For this reaction, 2.0 equivalents of BH3°SMe2, 1.0
equivalents of (S)-VAPOL, 3.0 equivalents of phenol, and 1.0 equivalent of water was
used for the preparation of the catalyst (entry 3). The reagents were combined and heated
to 100°C in toluene for 1 hour and then heated to 100°C under high vacuum for 30
minutes. When the reaction was attempted using this catalyst, only 31% yield of the
desired product 151 was obtained with 75% ee. It is clear from these experiments that the
B1 and 82 catalysts are not as effective in producing asymmetric induction as high as for
the optimal conditions.

The next few sections will contain a biref summary of the work that has been
described in this thesis. Conclusions can be drawn from the results observed during the
course of the study and along with those conclusions several questions arise as well.
Some of these questions will be discussed as well as the future work that needs to be
done to gain a better understanding of this catalyst and reaction system.

6.2 Conclusions and Future work
6.2.1 Aza-Diels-Alder Optimization

Scheme 6.2 Optimal Reaction Conditions

 

C5202 _ 0.5 mmHg

(S)-VAPOL + B(OPh)3 ,
55 °C, 1h 55 °c,o.5 h

5 mol % 100 mol %

 

 

> Catalyst '0

 

 

\ ’. \
N/kph + . O VAPOL-boron catalyst:
MOMe CH2C12/toluene (1:1)
150 31 —45°C, 241. 0%;1

1.0 equ1v 2.0 equ1v. cOil addition flask
added over 3.0 h 85% yield P151
through cooling 9()% Ce

addition coil

194

A significant amount of effort was spent on the exploration of the aza-Diels-Alder
reaction of benzhydryl (Bh) protected imino dienophiles during these studies. Several
1i gand/Lewis acid combinations were explored and it was found that the catalyst derived
from B(OPh)3 and (S)-VAPOL was optimal. Using this catalyst, a broad range of
conditions had to be explored to find those that would give the highest yield and
asymmetric induction. Originally the TBS version (149) of Danishefsky’s diene was
studied but later it was found that the less bulky original Danishefsky’s diene (31)
produced better results. Studies were then done to determine how other important
variables such as time, temperature, solvent, catalyst preparation, catalyst loading,
equivalents of reagents, order of addition of reagents, and how the reagents are added
would effect the outcome of the reaction. In order to determine the optimal conditions
for this reaction many different combinations of these variables were explored. This task
was found to be quite daunting, but in the end, the optimal results obtained were 85%
yield and 90% ee using only 5% loading of the chiral ligand. The catalyst was prepared
by heating 5 mol% (S)-VAPOL and 100 mol% B(OPh)3 in CHZCI2 to 55°C for one hour
followed by heating to 55°C under high vacuum for 30 minutes. The optimal
temperature for the reaction was determined to be —45°C and using a 1:1 mixture of
CHzClz/toluene as the solvent was optimal. An interesting effect was observed when
studying how the reagents were added and it was found that when the diene was added
over a period of three hours as a 0.33 M solution in CHzClz/toluene (1:1) was optimal. In
addition, the use of a home-made round bottom flask equipped with a coil that allowed

the solution of the diene to be cooled down to the reaction temperature as it was added

195

was able to increase the ee’s by about 2%. More importantly, however, it was observed
that the reproducibility of the reaction was slightly better When using this special flask.
6.2.2 Substrate Screening

The optimal conditions described above were used for the screening of a wide
variety of imines to test the generality of the reaction. Many substrates afforded good
results When using the optimal conditions, but some imines either needed increased
reaction time to achieve higher yields or increased catalyst loading (10 mol% VAPOL) to
aid in the asymmetric induction of the reaction. For most of the imines screened, when
one or both of these modified conditions were employed, the ee’s as well as the yield
could be increased. The increased reaction time or catalyst loading was only used when
the optimal conditions failed to provide satisfactory results.

The reaction was found to be general for aromatic substrates with yields from 71-
85% and ee’s from 89-95%. The only exception was the p-nitrobenzaldimine substrate
. 159 that gave only 69% yield and 73% ee even when 10 mol% (S)-VAPOL was used.
The anti-unsaturated imines proved to be interesting as it was found that when there was
no substituents in the a-position (i.e. trans-B-styryl 173 and B,B-dimethyl 175) the
reaction failed to produce any enantiomeric excess and low yields were obtained.
However, when the double bond was substituted at the a-position (i.e. cyclohexenyl
175), the reaction was very successful giving 45% yield and 93% ee using 10 mol% (S)-
VAPOL. It is still not clear whether it is the substituent at the a-position or the fact that
the double bond is contained in a ring that is causing the asymmetric induction. One
substrate that could be looked at to determine which is true would be that with just a

methyl group in the a-position. It was found when studying the aliphatic substrates that

196

secondary aliphatic groups worked quite well (64~90% yield and 90-93% ee), but the
primary aliphatic imines gave only racemic products. Due to the lack of success with the
primary substrates, several other imines could function as surrogates for the imines with a
primary aliphatic side chain were also examined and success was limited with these
compounds as well. The best results from these studies were found when imines 199 or
201 prepared from ethylglyolylate or isopropylglyoxylate were used and although yields
were around 80%, the reactions only gave around 60% ee. A substrate with a silyl-group
incorporated into the a-position (212) of an a,B-unsaturated substrate was also attempted,
but failed undergo the aza-Diels-Alder reaction and no product was formed. Lastly, a
substrate containing a silyl-substituted acetylene (207) was studied and the product 208
with only 22% ee was obtained from the reaction.

In addition to the benzhydryl imines, several other imines were studied that were
prepared using substituted benzhydryl amines. Those studied were the 3,3’5,5’-
tetramethyl benzhydryl (TMB), 4,4’—dimethoxy bezhydryl (DAM), and 3,3’,5,5’ tetra-t-
butyl-4,4’-dimethoxy benzhydryl (BUDAM) groups. For each of the different
benzhydryl amines, the corresponding imines were prepared using benzaldehyde, 4—
nitrobenzaldehyde, 4-bromobenzaldehyde, and cyclohane carboxaldehyde. 'The first
class of benzhydryl derivatives studied was the TMB imines. The TMB imine 215
(phenyl) was better than the original Bh group (150) for the phenyl substrate but this was
not observed for cyclohexyl (219) or 4—brom0phenyl (217) substrates. The other two
benzhydryl derivatives studied. were the di-4—methoxy benzhydryl group (DAM) and the
di-tetra-t-butyl-dimethoxy benzhydryl group (BUDAM) and they were also found to be

have an effect on the outcome in the aza-Diels-Alder reaction where the optimal

197

conditions were employed. The DAM imines gave similar yields as their corresponding
benzhydryl imines, but the effect on the assymetric induction varied. The phenyl DAM-
imine 221a and the cyclohexyl DAM imine 221d gave lower ee’s (84% and 56%
respectively). For the aromatic substrates with electron withdrawing groups, 4-
bromophenyl DAM imine 221b gave about the same ee (90%) but for the 4—nitrophenyl
DAM imine, the ee increased from 73% (Bh) to 81% (DAM). All of the BUDAM imines
that were studied were found to be inferior to the benzhydryl-substituted substrates and
40% ee was the highest asymmetric induction observed for this class of imines.

When developing a new methodology, it is important to compare results from that
study to those that have already been published in the literature. Arguably, the best
system in the literature was developed by Snapper and Hoyveda,32 and using a silver
. catalyst were able to achieve >98% yield and 95% ee. One detail that is impressive about
their system is that they were able to get these results using only 1 mol% catalyst, which
means that they were getting almost 100 turnovers. In addition they also developed a
recoverable catalyst, which could be used effectively several times in subsequent
reactions. The highest ee in the literature was 97 % and this was observed by Whiting”6
using 10 mol% of the catalyst prepared from MgI2 and the chiral diamine (R,R)-l,2-
diphenyl—ethy1ened_iamine (120). K

The work by Whiting and Hoyveda, just discussed, is the best that has been
achieved to date, however several other successful systems have been reported where
ee’s in the low 90’s with moderate to excellent yields have been achieved. The (S)-
VAPOL/B(OPh)3 system used for the studies described in this thesis gave similar results

to those in the literature, and yields and ee’s could both be achieved in the upper 80’s to

198

low 90’s. In addition, this catalyst system was able to achieve about 66 TONS with
minimal loss of asymmetric induction (1 mol% catalyst gave 66% yield and 82% ee).
6.2.2.1 Other Possible Substrates to Screen

When developing a new reaction or catalyst system, it is important to be as
exhaustive as possible when testing the generality of the reaction. As seen above, the ‘
method developed and described in this thesis has proven to be quite general with a few
exceptions. The substrate scope has primarily dealt with the modification of the aldehyde
used to prepare the imine using aminodiphenylmethane. Some other benzhydryl
derivatives were screened as well, but future work needs to‘be done to explore chiral
imine derivatives. Both imines derived from chiral amines (326 and 327) or chiral
aldehydes (328 and 329) should be examined as substrates for the aza-Diels-Alder
reaction (Figure 6.5). This study would indicate whether it was the catalyst or iminethat

Figure 6.5 Chiral Imines

R 328 R 329

had a greater influence on the asymmetric induction of the reaction. In addition, one may
discover a matched or mismatched case where one specific enantiomer combination

would be favorable.

199

In 1989 Kunz et al reported the use of a pivaloylated carbohydrate derivative
where the anomeric alcohol was replaced by an NH2 which in turn was used to to prepare
imines (330, Figure 6.6)

Figure 6.6 Sugar Derived Imine
PivO I

RA N O OPiv

OPiv OPW

330

for the aza-Diels-Alder reaction.103 In the decade thereafter, Kunz also published other
work using the carbohydrate as a template to incorporate chiral centers in other reactions
as well.""’”106 The aza-Diels-Alder reaction worked quite well when ZnCl2 was used as
the catalyst giving high yields and moderate diastereoselectivity. More specifically, the
reaction run with Danishefsky’s diene and the imine 330a prepared from 3- ,
pyridinecarboxaldehyde, gave 92% yield of 331 and a diastereomeric ratio of >20:l
(Scheme 6.3). The results of Kunz’s studies as well as the observations seen in the

Scheme 6.3 Kunz’s Reaction Using Danishefsky’s Diene”3

 

PivO I O
Si/ 1'. ZnC12 (2.0 equiv.)
N \ \N 0 0m, 0’ \ E120,—20°C,12h; (“j
. I / , _ 0m MOM, 2.1NHC1 N \vf" N
OPW 3. NaHCO3 I /
330a 31 PivO
O OPiv

92% yield PivO

>20:1 d.r. OPiv

331

aziridination reaction that the size of the .group on nitrogen may be important in the

reactions using the VAPOL-boron catalyst indicate that this class of imine should be

200

studied for the aziridination reaction as well as aza-Diels-Alder reaction VAPOL-boron
catalyst. In addition to exploring the double stereodifferentiation in these reactions, the
chiral group on the nitrogen may lead to selectivity in subsequent modification reactions
as well.

As mentioned before it has been shown in the aziridination reaction (unpublished
results) that the size of the group on nitrogen of the imine is important to the outcome of
the reaction. Although this was not particularly true for the aza-Diels-Alder reaction, it
would still be interesting to explore in more detail the effects of the size of Bh group and
the electron donating (332) or electron withdrawing (333) nature of the benzhydryl group
on the nitrogen (Figure 6.7).

Figure 6.7 Electron Donating and Withdrawing Bh Derivatives

   

332 333

A limitation of this catalyst system for the aza-Diels-Alder reaction was the
failure of imines prepared from primary aliphatic aldehydes to give any asymmetric
induction. In addition, all attempts to find suitable substrates that could be modified to
(primary alkyl groups failed. To make. this reaction more attractive for use in the
synthesis of more advanced compounds, it would be a great achievement to be able to
add this type of substrate to the list of successful imines for this reaction. One possible

class of imine that could provide access to the primary aliphatic side chain would be ones

201

containing a cyclic ketal (334) or thioketal (335) attached directly to the imine carbon
(Figure 6.8). These substrates contain a

Figure 6.8 Ketal or Thioketal Imines

m o m c

secondary a—carbon, which may be necessary for this reaction to be successful. Should
the reaction work well, the cyclic ketal or thioketal could be easily removed and the
resulting aldehydes modified to a linear carbon chain. Although incorporation of
heteroatoms may prove to be detrimental to the reaction, if this class of imines were
successful, it would be a nice addition to the scope of this reaction.

The other participant in the aza-Diels-Alder reaction is the diene and to date the
only dienes that have been studied using the VAPOL-boron catalyst were Danishefsky’s
diene (31) and its TBS derivative 149. It would be interesting to determine the
effectiveness of the VAPOL-boron catalyst using other dienes as well (Figure 6.9).

Figure 6.9 Dienes to be Screened I

O,Si\
‘3 OR 0 OR %NAO
. s's
cyclopentadiene cyclohexadiene / | \
336 337 2-aza-Diene
338

202

Especially of interest would be the use of unactivated dienes such as cyclopentadiene
(336) or 1,3-cyclohexadiene (337), which are locked in the s-cis conformation. It would
also be interesting to look into the possibility of using aza-dienes as well. Should the
reaction work with activated 2-aza-diene 338, one could potentially use the cycloadduct
to access B-amino acid derivatives 341 (Figure 6.10).

Figure 6.10 Access to B—Amino Acids

 

 

339

<u--

B—amino acid
341.
6.2.2.2 Exploration of Other Boron Catalysts

As discussed earlier, work has been going on in the Wulff group to determine the
structure of the active catalyst. During these studies, new methods of catalyst preparation
have beendeveloped where the catalyst containing only one boron atom (Bl-catalyst) can
be made selectively as well as the catalyst Containing three boron atoms (B3-Catalyst)

(Figure 6.11). Using these alternate methods, the catalysts can be made without the use of

203

Figure 6.11 B1 and B3 Catalysts

 

B 1 -Catalyst B3-Catal yst

excess triphenylborate, which would eliminate the possibility of background reaction.
Although the excess triphenylborate was found to have great utility in the aza-Diels-
Alder reaction described in this thesis, it would still be informative to determine the
effectiveness of these catalysts without the use of excess B(OPh)3. In addition to the
catalysts prepared using VAPOL, the same catalysts prepared using VANOL or BINOL
could also be explored. It may also be interesting to further explore the the BLAH
catalyst (Figure 6.12) utilized by Yamamoto22 for the aza-Diels-Alder reaction. Several

Figure 6.12 BLAH Catalyst

DO (‘9 00
H

0433/0 ,
I I o/ \o I I

' BLAH-Catalyst

preliminary reactions have been run to determine if this catalyst was worth exploring

(Scheme 6.4). The first imine that was studied was 613 in order to compare the reactions

204

Ph

Scheme 6.4 Results Using BLAH-Catalyst

 

 

(R)-BINOL B(OPh)3 (10 equiv.)

a BLAH-Catalyst

 

2.0 equiv. CHZCIZ, rt, 4A MS

 

 

) O

 

|./
i /I\ O/S'\ BLAH-Catalyst I
\ 'I' f
Ph MOMCCHZCb/tolueneazl) Ph N
613 R = H 31 —45°, 24 h R/k Ph
= 2.0 ' .
151) R Ph ( eqmv) 62 R=H
151 R = Ph

 

Entry R (

R)-BINOLB(OPh)3
(mol%) (mol%)

Product Yield (%) ee (%) Configuration

 

#9919)—

H
Ph

Ph
Ph

200
200

100 62 75
100 151 92
20 10 151 39
20 100" 151 54

~80-88a R
89 R
90 R
60 R

 

a) Approximation because baseline separation could not be achieved in HPLC
analysis. b) The imine was premixed with 20 mol% BINOL and 10 mol%
B(OPh)3, dissolved in solvent, cooled to —45°C and to it was added 90 mol%
B(OPh)3 immediately before the diene was added.

205

carried out using the optimal conditions found for the aza—Diels-Alder reaction (see
Chapter 2) and those reported by Yamamoto. For these reactions, BINOL, B(OPh)3'and
the imine were premixed in a glove box, taken out and dissolved in solvent. The
resulting mixture was stirred for about 10 minutes and then cooled to —45°C. Using
imine 61a (entry 1), the reaction gave 75% yield and ~80-88% ee (approximation, HPLC
not baseline resolved). This result is comparable to those achieved by Yamamoto22 (78%,
yield and 86% ee) only he used different conditions (1.2 equiv. diene, 4A M.S., CHZCIZ,
—78°C, and the reaction was only run for 12 hours). Yamamoto also reported that imine
150 prepared from benzaldehyde and aminodiphenylmethane failed to produce the

product 151 using the catalyst prepared from a 1:1 ratio of BINOL and B(OPh)3 but he

never attempted this reaction using the BLAH catalyst. To determine if this catalyst was
going to be worth exploring further, an attempt was made with imine 150 using a
stoichiometric amount of the BLAH catalyst (entry 2). The reaction worked quite well at
——45°C in CHzClz/toluene (1:1) and after the reaction was stirred for 24 hours, it gave
92% yield of the desired product 151 with 89% ee. The reaction was then attempted
using 10 mol% of the BLAH catalyst (entry 3) and surprisingly the reaction turned over
about 4 times giving 39% yield and 90% ee. Lastly, one reaction was attempted (entry 4)
to mimic the optimal conditions found for the aza-Diels-Alder using 5 mol% VAPOL
and 100 mol% B(OPh)3 to determine if the turnover could be increased without lowering
the induction. The imine was premixed with 10 mol% of the BLAH catalyst and stirred
after diluting with solvent for about 10 minutes. The reaction was cooled to —45°C and
solution containing 90 mol% more B(OPh)3 was added followed immediately by the slow
addition of the diene. The outcome of this reaction was inferior to those in entries 2 and
3 and only gave 54% yield and 60% ee indicating that the background reaction must be
competing with the BLAH catalyst under these conditions. Exploration of reaction
conditions for this catalyst may lead to more turnover numbers and another successful
use of BINOL. Although the VAPOL ligand might be too bulky toform this 2:1
ligand/boron (BLAH type) catalyst, it may be worthwhile to explore the slightly less

bulky VANOL ligand for this type of catalyst (V ANOL—BLAH-catalyst, Figure 6.13)

206

Figure 6.13 VANOL-BLAH Catalyst
0' 69 O
H
Ph o\<§3,o Ph
Ph 0/ \o Ph

VANOLBLAH—Catalyst

6.2.3 Triphenylborate Effects

The most interesting discovery during these studies was the fact that the success
of this reaction depended on the use of excess B(OPh)3 (a non-chiral Lewis acid) (see
Figure 6.1). As Figure 6.1 indicates, an increased rate of the reaction was observed
without a loss of ee until the ratio of 100 to 1 B(OPh)3:VAPOL was used to prepare the
catalyst. In effort to determine the role of B(OPh)3 in the reaction, attempts were made to
isolate intermediates formed during the course of the reaction. All of these attempts
failed so the role of B(OPh)3 had to be determined using a different method. Eventually,

' determination of the association constants (Table 6.1) for the VAPOL-boron catalyst and

Table 6.1 Binding Constants for Complexes of the Imine and Product with B(OPh)3

and the VAPOL-Boron Catalyst

 

 

Entry Comex Binding Constant (M“)
1 B(OPh)3 / imine 0.32 :0. 12
2 B(OPh)3 / product - 2.7 :04 '
3 VAPOL-B / imine / 2.1 :04
4 VAPOL—B / product 4.9 :08

 

'* B(OPh)3 with both the imine (150) and product (151) gave some indication of what was

actually going on during this reaction. After careful analysis of the data in table 6.1, it

207

was determined that the rate of the chiral reaction must be at least 9 times faster than the
background reaction. .

The experimental results are still very intriguing and to our knowledge, this is the
only catalytic asymmetric reaction where two similar catalysts, one chiral
(substoichiometric) and the other achiral (stoichiometric), are playing separate roles (in
acid-base interactions) making the reaction more efficient.

6.2.4 Progress Towards Cylindricine C

While studying the aza-Diels-Alder reaction, significant progress was also made

toward the total synthesis of (:)-Cylindricine C (Scheme 6.5). The key step in the

Scheme 6.5 Progress Towards Cylindricine C

 

 

 

0
WM TMSO BF3°OEtz (1.1 equiv.)
l + >
HN Ph \ OTMS CH2C|2 N
279b . I
, 65% yreld 287 HN ph
1.0 equrv. O \n/
278 Mg (3.0 equiv) O
. ' . H CI 0.7 '
l 5 equ1v g 2( equ) 87% yield
THF/EtOH (l :3)
room temparature 11
O
O
n-BuLi (1.1 equiv)
. . Cszl 1.] u'v
1,4Add1tron < ( eq 1 )
<- .............. N THF . If
k -—78°C to RT, 22h H
PhAO o 80% yield 290
294

synthesis was envisioned to be the aza-Diels-Alder reaction to produce two of the three
rings that make up the core structure of the Cylindricine C. Although all attempts to
discover an asymmetric version of this reaction failed, the racemic reaction catalyzed by
BF3,-OEt2 was optimized. It was found that this reaction could give up to 68% yield and

even on a 6.2 g scale, 63% yield of the desired product 287 could be obtained. The next
208

step was to install the butenyl side chain and all attempts to accomplish the conjugate
addition on the hydrazine moiety failed. Literature precedent98 indicated that the
incorporation of a Cbz protecting group would facilitate this reaction, so the hydrazine
needed to be cleaved and the Cbz group added. The hydrazine was easily reduced off by
the addition of magnesium and mercuric chloride to give the free amine 290 in up to 80%
yield and even on a 9.0 g scale, 60% yield could be obtained. The Cbz group was then
introduced by treatment of the secondary Vinylogous amide 290 with n-butyl lithium
followed by Cbz-Cl to give the Cbz-protected Vinylogous amide 294 in 80% yield.
Unfortunately, all attempts to facilitate the conjugate addition on this substrate failed as
well. This is where progress, using the original route A, halted and this problem still
needs to be solved before the synthesis can be completed.

An alternate route B has been envisioned where the butenyl side chain could be
incorporated into the diene (227), which would eliminate the need for the conjugate
addition. Also, if this route could be employed the total synthesis wOuld take only 6 steps
(including diene’preparation) as opposed to 7 steps (not including diene preparation) for
the original route. The aza-Diels-Alder reaction employing this diene was attempted
only one time (Scheme 6.6) and the reaction failed to give the Diels-Alder cyclo adduct,

Scheme 6.6 Alternate Diene aza-Diels-Alder Attempt

 

 

\ .
WJ H OTMS BF 05% (I 1 eq)
\ ,N Ph + 3 ' 5
N \[f \ CH2C12
0 -78°C to RT
279b
1.0 equiv. 227
1.2 equiv 296
23% yield

209

however, the Mannich product 296 was obtained in 23% yield. If this issue surrounding
the aza-Diels alder reaction of 227 and 279b in route B can be solved, and the synthesis
completed, it would match the shortest synthesis reported to date reported by Shibasaki,89

(only 6 steps longest linear sequence).

210

CHAPTER 7

EXPERIMENTAL SECTION

7 .1 Experimental Procedures and Characterizations Data for Chapter Two

Preparation of Danishefsky’s diene (31):

To a flame dried 500 mL three neck round bottom flask equipped with a magnetic
stir bar and a pressure equalizing liquid addition funnel was added KHMDS (80 mL, 40
mmol, 0.5M solution in toluene) and diluted with THF (50 ml). The solution was then
cooled to —78°C and the ketone (4.0 mL, 39.2 mmol) was added dropwise in 50 mL THF
over 15 min. through the liquid addition funnel. This was allowed to stir for 1.5 hours and
then warmed up to —30°C for 30 minutes. The reaction mixture was then cooled back to
—78°C at which time TMSCI (6.19 mL, 48.8 mmol) was added dropwise in 50mL THF
over 30 minutes through the liquid addition funnel. The reaction was then allowed to
warm up slowly to room temperature and stir for 1 hour. The solvent was then removed
under reduced pressure. The crude orange reaction mixture was diluted with ether and.
filtered through Celite. The Celite was washed several times with ether to ensure the
product was recovered. Finally the ether was removed under reduced pressure and the

product was purified by distillation (65-69°C, 13 mmHg) to give 4.87 g of the desired

211

diene 31 (72% yield). The spectral data matched perfectly to that in the literature.3 All
reactions involving the use of the diene 31 used diene prepared by this method unless

noted otherwise.

R R
| B(OH)3 (1.0 equiv.) |
/ \ on - / \ o}B

r

—- Toluene, reflux — 3

3.0 equiv.

Preparation of triarylborates:

To a round bottom flask equipped with a magnetic stir bar and an azeotropic
distillation apparatus (Dean-Stark trap) was added toluene (100 mL/163 mmol) the
phenol (3.0 equiv), and boric acid (1.0 equiv). The solution was heated to reflux and
allowed to stir overnight. The toluene was then removed by distillation to give the crude
triarylborate.

When R: 2,6—dimethyl, the product was purified by distillation (160-170°C, 0.05
mmHg) using an air condenser to give a solid which contained only 5% free 2,6-
dimethylphenol. This mixture was used. for the preparation of the VAPOL-boron
catalyst. .

When R= 4—fluoro, after removal of the toluene the crude. oil was heated under
high vacuum to remove any of the volatiles and the resulting oil was then used for the
preparation of the VAPOL-boron catalyst.

When R= 4—OMe, the crude solid was heated until melted and placed under high
vacuum to remove any of the volatiles and the resulting solid was then used for the

preparation of the VAPOL-boron catalyst.

212

When R: 4—Me, the crude solid was heated until melted and placed under high
vacuum to remove any of the volatiles and the resulting solid was then used for the

preparation of the VAPOL-boron catalyst.

 

\ A O’Si\ catalyst :
O N Ph + M solvent
OMe temperature
150 31

reaction time
flask

 

General reaction procedure:

To a flame dried, argon purged single-necked flask that had its 14/20 joint
replaced with a threaded Teflon high—vacuum T-shaped stop-cock equipped with a stir
bar was added B(OPh)3 and (S)-VAPOL (130). To this was added CHzCl2 (2 mL) and
then the flask was sealed with the stopcock and heated to 55 °C for one hour. After one
hour, the solvent was removed via high vacuum and left under high vacuum at 55 °C for
0.5 hours yielding the VAPOL-boron catalyst. After cooling the stopcock was removed
and replaced with a rubber septum. The catalyst was dissolved by the injection via
syringe of solvent. The catalyst solution was transferred by syringe to a flask containing
the imine prepared as immediately below.

To a flame dried, argon purged round bottom flask or homemade flask with a cold
addition coil (see Figure 2.4) equipped with a stir bar was added the imine (150) (1.0
mmol). The flask was topped with a rubber septum and the VAPOL-boron catalyst was
transfered in two 1.0 mL portions of solvent directly to the bottom of the flask by a

syringe equipped with a long needle. This Was allowed to stir for 5-10 minutes at room

213

temperature and then cooled to the reaction temperature. Meanwhile, in a separate flame
dried 5 or 10 mL round bottom flask purged with argon was added Danishefsky’s diene
(31) and solvent. The diene was taken up in a syringe and added to the solution of imine
and VAPOL-boron catalyst. After completion of the reaction, saturated sodium
bicarbonate(~20 mL) was added to the reaction flask. This was then transferred to a
separatory funnel and diluted with distilled water (25 mL) and extracted with three or
four 3040 mL portions of CHZCIZ. The combined organic layers were placed in a 2.50
mL round bottom flask and the solvent was then removed via rotary evaporation. The
flask was then equipped with a stir bar and cooled in an ice bath. To this was then added
a previotrsly cooled (0 °C) 20:1 mixture of THF and 1N HCl (50 mL) at which time the
flask was removed from the ice bath and allowed to stir (monitored by TLC) until the
undesired spots close to the desired product disappeared (usually less than one. hour).
This was then transferred to a separatory funnel containing water (75-100 mL) followed
by extraction of the crude product with four 50 mL portions of CHZCIZ. The combined
organic layers were then dried with magnesium sulfate, filtered, and solvent was removed
via rotary evaporation. Purification was accomplished using flash column

chromatography.

 

H
l/ B(OPh)3 (100 mol%)
Ph ’51 \
o A + 0 (S)-VAPOL (N mol%)
+ >
HZN Ph MOMe DCM/toluene (1: 1)
152 153 31 ”mist?“
1.0 equiv 2 equiv. —

 

added over 3 h 24b

in 3 mL solvent

through cooling coil

214

Reactions using in situ prepared imine 150:

To a flame dried, argon purged single—necked flask that had its 14/20 joint
replaced with a threaded Teflon high-vacuum T—shaped stop-cock equipped with a stir
bar was added B(OPh)3 (0.3125g, 1.0 mmol) and (S)-VAPOL (130) (27mg, 0.05 mmol). '
To this was added CHZCI2 (2 mL) and then the flask was sealed with the stopcock and
heated to 55 °C for one hour. After one hour, the- solvent was removed via high vacuum
and left under high vacuum at 55 °C for 0.5 hours yielding the VAPOL-boron catalyst.
After cooling the stopcock was removed and replaced with a rubber septum. The catalyst
was dissolved by the injection via syringe of 2.0 mL of a 1:1 mixture of toluene and
CH2C12 (in two portions). The catalyst solution was transferred by syringe to a‘solution
of the in situ prepared imine prepared as immediately. below.

To a flame dried, argon purged homemade flask with a cold addition coil (see
Figure 2.4) equipped with a stir bar was added the drying agent. The flask was topped
with two rubber septa and to it was added toluene/CH2C12 (1:1) (2 mL). To the flask was
then added benzaldehyde (152) (1.0 or 1.1 equivalents) followed immediately by
aminodiphenylmethane (153) (0.172 mL, 1.0 mmol). The resulting mixture was allowed
to stir at room temperature during the ~2.0 hour catalyst preparation and used without
further manipulation for the aza—Diels-Alder reaction. The IVAPOL-boron catalyst was
then transferred in two 1 mL portions of CHzClzltoluene (1:1) directly to the bottom of
the flask containing the imine by a syringe equipped with a long needle. This was
allowed to stir for 5-10 minutes at room temperature and then cooled to —45 °C.
Meanwhile, in a separate flame dried 5 or 10 mL round bottom flask purged with argon

was added Danishefsky’s diene (31) (0.38 mL, 2.0 mmol) and toluene/CH2C12 (1:1) (3.0

215

mL). The diene was taken up in a syringe and added over 3.0 hours via syringe pump
through the cold addition coil. The reaction was then allowed to stir at —45 °C for the
duration of the reaction (24 hours). After completion of the reaction, saturated sodium
bicarbonate (~20 mL) was added to the reaction flask at —45 °C. This was then
transferred to a separatory funnel and diluted with distilled water (25 mL) and extracted
with three or four 3040 mL portions of CH2C12. The combined organic layers were
placed in a 250 mL round bottom flask and the solvent was then removed via rotary
evaporation. The flask was then equipped with a stir bar and cooled in an ice bath. To
this was then added a previously cooled (0 °C) 20:1 mixture of THF and 1N HCl (50 mL)
at which time the flask was removed from the ice bath and allowed to stir (monitored by
TLC) until the undesired spots close to the desired product disappeared (usually less than
one hour). This was then transferred to a Separatory funnel containing water (75-100
mL) followed by extraction of the crude product with four 50 mL portions of CHzClz.
The combined organic layers were then dried with magnesium sulfate, filtered, and
solvent was removed via rotary evaporation. The crude mixture was then purified by

flash column chromatography.

Sli/ B(OPh)3 (30 mol%)
0 + 0 \ (S)-VAPOL(10moI%)_ ' I
MOMe toluene, —45°C 7 “v. 0
152 31 48" O

2 equiv.

 

added over 3 h 151-.0
in 3 mL solvent 67% we“
28% ee

through cooling coil

Procedures for Scheme 2.23 (heteroatom Diels-Alder reaction of 152):

216

To a flame dried, argon purged single-necked flask that had its 14/20 joint
replaced with a threaded Teflon high-vacuum T-shaped stop-cock equipped with a stir
bar was added B(OPh)3 (87 mg, 0.3 mmol) and (S)—VAPOL (130) (54 mg, 0.05 mmol).
To this was added CH2C12 (2 mL) and then the flask was sealed with the stopcock and .
heated-to 55 °C for one hour. After one hour, the solvent was removed via high vacuum
and left under high vacuum at 55 °C for 0.5 hours yielding the VAPOL-boron catalyst.
After cooling the stopcock was removed and replaced with a rubber septum. The catalyst
was dissolved by the injection via syringe of 2.0 mL of toluene (in two portions).

To a flame dried, argon purged round bottom flask‘equipped with a stir bar was
added the aldehyde (152) (0.101 mL, 1.0 mmol). The flask was topped with a rubber
septum and the VAPOL-boron catalyst was transfered in two 1.0 mL portions of toluene
directly to the bottom of the flask by a syringe equipped with a long needle. This was
allowed to stir for 5-10 minutes at room temperature and thencooled to —45°C.
Meanwhile, in a separate flame dried 5 or 10 mL round bottom flask purged with argon
was added Danishefsky’s diene (31) and toluene (3.0 mL). The diene was taken up in a
syringe and added dropwise to the solution of imine and VAPOL-boron catalyst over 3
hours via syringe pump. The resulting reaction mixture was allowed to stir for 48 hours
total reactiontime at which time trifluroroacetic acid (several drops) was added to the
reaction flask and allowed to stir for a couple hours. At this time, saturated sodium
bicarbonate (~20 mL) was added. This was then transferred to a separatory funnel
purged with ~25 mL distilled water. Extraction was accomplished using four 25 mL
washes of dichloromethane. The product was then isolated by flash column

chromatography (4:1 hexanes/ethyl acetate, rr 0.2) to give 118.7 mg (67% yield) of the

217

desired product 151-O. Spectral data matched perfectly to the literature107 and the

absolute configuration was determined by comparison of the HPLC retention times.

 

    

Ph | / 0
Si
\N /|\ Ph + O/ \ B(OPh)3 (100 mol%) _
Mom (S)-VAPOL (5 mol%) r
150 31 CHzClzltoluene (1:1)

1.0 equiv 2.0 equiv. —45°C, 24 h Ph Ph
added over 3.0 h coil addition flask ' 151
through cooling 85% yield

addition coil 90% ee

Optimal conditions:

To a flame dried, argon purged single-necked flask that had its 14/20 joint
replaced with a threaded Teflon high-vacuum T—shaped stop—cock equipped with a stir
bar was added B(OPh)3 (0.3125g, 1.0 mmol) and (S)-VAPOL (130) (27mg, 0.05 mmol).
To this was added CH2C12 (2 mL) and then the flask was sealed with the stopcock and
heated to 55 °C for one hour. After one hour, the solvent was removed via high vacuum
and left under high vacuum at 55 °C for 0.5 hours yielding the VAPOL-boron catalyst.
After cooling the stopcock was removed and replaced with a rubber septum. The catalyst
was dissolved by the injection via syringe of 2.0 mL of a 1:1 mixture of toluene and
CHZCI2 (in two portions). The catalyst solution was transferred by syringe to a solution
of the imine prepared as immediately below.

To a flame dried, argon purged homemade flask with a cold addition coil (see
Figure 2.4) equipped with a stir bar was added the imine (150) (1.0 mmol). The flask
was topped with two rubber septa and the VAPOL-boron catalyst was added in two 1.0

mL portions of toluene/CHzCl2 (1:1) directly to the bottom of the flask by a syringe

218

equipped with a long needle. This was allowed to stir for 5-10 minutes at room
temperature and then cooled to —45 °C. Meanwhile, in a separate flame dried 5 or 10 mL
round bottom flask purged with argon was added Danishefsky’s diene (31) (0.38 mL, 2.0
mmol) and toluene/CHzCl2 (1:1) (3.0 mL). The diene was taken up in a syringe and
added over 3.0 hours via syringe pump through the cold addition coil. The reactiOn was
then allowed to stir at —45 °C for the duration of the reaction 24 hours. After completion
of the reaction, saturated sodium bicarbonate (~20 mL) was added to the reaction flask at
—45 °C. This was then transferred to a separatory funnel and diluted with distilled water
(25 mL) and extracted with three or four 30-40 mL portions of CHzClz. The combined
organic layers were placed in a 250 mL round bottom flask and the solvent was then
removed via rotary evaporation. The flask was then equipped with a stir bar and cooled
in an ice bath. To this was then added a previously cooled (0 °C) 20:1 mixture of THF
and 1N HCI (50 mL) at which time the flask was removed from the ice bath and allowed
to stir (monitored by TLC) until the undesired spots close to the desired product
disappeared (usually less than one hour). This was then transferred to a separatory funnel
containing water (75-100 mL) followed by extraction of the crude product with four 50
mL portions of CH2C12. The combined organic layers were then dried with magnesium
sulfate, filtered, and solvent was removed via rotary evaporation. The crude mixture was
then purified by flash column chromatography (Rf 0.08, 2:] hexanes/ethyl acetate) to give
288.5 mg pure 151 (85% yield). The enantiomers could be separated by HPLC analysis
on Chiralcel OJ-H (75:25 hexane/isopropanol, 1 mL/min). Retention times: 8.55 and
17.61 min. The product 151 obtained from the reaction was. determined to be 90% ee

(major peak = 17.61 min).

219

Spectral Data for Compound 151 (C24H21NO):

IH NMR (CDCl;,) 5 2.77-2.96 (m, 2H), 4.59 (t, 1H, J=7.1 Hz), 5.06 (d, 1H, J=8
Hz), 5.46 (s, 1H), 7.46-7.06 (m, 16H); 13C NMR (CDC13) 6 43.51, 62.01, 67.67, 98.60,
127.04, 127.39, 128.01, 128.18, 128.32, 128.68, 128.82, 128.95, 129.36, 137.94, 138.11,
138.74, 151.30, 190.15; IR (CDCI3) 3031m, 2959m, 2926m, 1645vs, 1591vs, 1570m
cm"; mass spectrum m/z (% rel intensity) 339 M+ (5), 338 (13), 168 (14), 167 (100), 165
(31), 152 (17), 104 (12), 103 (10), 77 (8), 51 (5), 50 (3). White solid, mp 136—138 °C.

Optical rotation taken on 90% ee sample, [(112013 +106.2 (c 1.90, CHZCIZ).

220

7.2 Experimental Procedures and Characterizations Data for Chapter Three

 

Ph
Ph Ph CH Cl
EL + Y 2 2 7* NIJWh
NH
R H 2 MgSO4 RAH

Preparation of the aldirnines:

All imines were prepared using the following general protocol and spectral data
matched that published in the literature.” 57' 64‘ 103”" To a flame dried round bottom flask
equipped with a magnetic stir bar was added the MgSO4 (0.15g/mmol) and CHzCl2
(1.5mL/mL) followed by the aldehyde (1.0 equivalent). To this solution was then added
the desired benzhydrylamine (1.0 equivalent). The resulting solution was then allowed to
stir at room temperature overnight (18-25 hrs). At the end of the reaction time, the
solution. was filtered to remove the MgSO4, and the solvent removed under reduced
pressure. If the resulting imine was a solid, then it was purified by recrystallization
(hexane/CHZCIZ). If the resulting imine was an oil, it was simply used as the crude oil

without further purification.

 

o
l/
i 0,31 \ Yb(OTf)3 (10 mol%) I
\ + >
RA N Ph MOMe toluene, RT R N
24 h Ph/|\ Ph

General protocol for the preparation of racemic aza-DieIs-Alder products:

To a flame dried round bottom flask equipped with a magnetic stir bar was added
the imine (1 mmol) and ytterbium triflate (62 mg, 0.1 mmol). To the contents of the flask
were then added toluene (10 mL) followed by Danishefsky’s diene (31) (1.1 mmol). The

reaction was then allowed to stir for 24 hours at which time the reaction was quenched

221

with a mixture of THF and 1N HCl (20: 1) and stirred for one hour. The contents were
then transferred to a separatory funnel, diluted with distilled water (100 mL) and
extracted three times with CH2C12 (3 x 50 mL). The organic layers were then combined
and dried over magnesium sulfate, filtered, and the solvent was removed under reduced
pressure. The racemic compounds were then purified via flash column chromatography.
Protocol for the aza-Diels-Alder reaction optimal conditions used for screening of
substrates:

To a flame dried, argon purged single-necked flask that had its 14/20 joint
replaced with a threaded Teflon high-vacuum T-shaped stop-cock equipped with a stir
bar was added B(OPh)3 (0.3125g, 1.0 mmol) and (S)-VAPOL (27mg, 0.05 mmol). To
this was added CHZCI2 (2 mL) and then the flask was sealed with the stopcock and heated
to 55 °C for one hour. After one hour, the solvent was removed via high vacuum and left
under high vacuum at 55 °C for 0.5 hours yielding the VAPOL-boronlcatalyst. After
cooling the stopcock was removed and replaced with a rubber septum. The catalyst was
dissolved by the injection via syringe of 2.0 mL of a 1:1 mixture of toluene and CHzCl2
(in two portions). The catalyst solution was transferred by syringe to a flask containing
the imine prepared as immediately below. ‘

To a flame dried, argon purged homemade flask with a cold addition coil (see
Figure 2.4) equipped with a stir bar was added the imine (1.0 mmol). The flask was
topped with two rubber septa and the VAPOL-boron catalyst was added in two 1.0 mL
portions of toluene/CHZCI2 (1:1) directly to the bottom of the flask by a syringe equipped
with a long needle. This was allowed to stir for 5-10 minutes at room temperature and

then cooled to —45 °C. Meanwhile, in a separate flame dried 5 or 10 mL round bottom

222

flask purged with argon was added Danishefsky’s diene (31) (0.38 mL, 2.0 mmol) and
toluene/CHZCI2 (1:1) (3.0 mL). The diene was taken up in a syringe and added over 3.0
hours via syringe pump through the cold addition coil. The reaction was then allowed to
stir at —45 °C for the duration of the reaction (24 total hours). After completion of the
reaction, saturated sodium bicarbonate (~20 mL) was added to the reaction flask at -45
°C. This was then transferred to a separatory funnel and extracted with three or four 30-
40 mL portions of CHzClz. The combined organic layers were placed in a 250 mL round
bottom flask and the solvent was then removed via rotary evaporation. The flask was
then equipped with a stir bar and cooled in an ice bath. To this was then added a
previously cooled (0 °C) 20:1 mixture of THF and 1N HCl (50 mL) at which time the
flask was removed from the ice bath and allowed to stir (monitored by TLC) until the
undesired spots close to the desired product disappeared (usually less than one hour).
This was then transferred to a separatory funnel containing distilled water (75-100 mL)
followed by extraction of the crude product with four 50 mL portions of CHzClz. The
combined organic layers were then dried over magnesium sulfate, filtered, and solvent
was removed via rotary evaporation. The product was then purified via flash column
chromatography (36 cm x 2 cm). The enantiomeric excess was determined by chiral

HPLC analysis with the aid of an authentic sample of the racemic product.

223

Ph I/

 

 

Si
\Nk ph + O/ \ B(OPh)3 / VAPOL _
MOMe CHzClz/toluene (1 :1)'
M60 154 31 —45°c
1.0 equiv 2.0 equiv reaction time MeO
Added over 3.0 h

in 3mL solvent

Procedures for Scheme 3.2 (aza-Diels-Alder reaction of imine 154):

Condition A: The protocol for the optimal conditions was followed exactly where _
0.3014g (1.0 mmol) of imine 154 was used. The product (155) was then purified via flash
column chromatography (36 cm x 2 cm, 2:1 hexanes/ethyl acetate rf 0.04). The reaction
produced 162.5 mg (44% yield) of the desired product 155 and 83% enantiomeric excess
was measured (average of two runs).

Condition B: The protocol for the optimal conditions was followed exactly except
the reaction was allowed to stir for 47 hours. The reaction produced 214.3 mg (58%
yield) of the desired product 155 and 77 % enantiomeric excess was measured.

Condition C: The protocol for the optimal conditions was followed exactly except
the catalyst was prepared using 54 mg (S)-VAPOL (10 mol%). The reaction produced
221.7 mg (60% yield) of the desired product 155 and 88% enantiomeric excess was
measured.

Condition D: The protocol for the optimal conditions was followed exactly except
the catalyst was prepared using 54 mg (S)-VAPOL (10 mol%) and the reaction was
allowed to stir for 48 hours. The reaction produced 262.3 mg (71% yield) of the desired
product 155 and 90% enantiomeric excess was measured.

Spectral Data for Compound 155 (C25H23N02):

224

The enantiomers could be separated by HPLC using a Chiralcel OJ-H column
(90:10 hexane/isopropanol, 2 mL/min). Retention times: 17.38 min. (minor) and 33.63
min. (major). 1H NMR (CDC13) 5 1.83 (s, 3H), 2.63-2.79 (m, 2H), 4.77 (t, 1H, J=8.8 Hz),
5.05 (d, 1H, 128 Hz), 5.20 (s, 1H), 7.03-7.40 (m, 14H), 7.59 (d, 1H, J=7.7 Hz); 13C NMR
(CDCl3) 6 18.35, 42.83, 57.93, 67.13, 98.71, 126.33, 126.59, 127.46, 127.81, 127.93,
128.00, 128.55, 128.75, 129.26, 130.90, 135.55, 136.49, 137.55, 137.91, 152.06, 190.38;
1R (CDCI3) 3027s, 2953m, 2938m, 2909m, 1642vs, 1582vs cm"; mass spectrum m/z (%
rel intensity) M+ 353 (42), 281 (14), 267 (4), 249 (3), 225 (15), 209 (15), 207 (27), 182
(13), 168 (16), 167 (100), 166 (12), 165 (34), 152 (20), 133 (5), 117(7), 115(9), 104(7),
- 103 (7), 91 (10), 77 (9), 73 (10), 51 (5). Anal calcd for C25H23NO: C, 84.95; H, 6.56; N,
3.96. Found: C, 84.96; H, 6.47; N, 3.89. White solid, mp 174-175 °C. Optical rotation
was taken on 99.9% ee material (obtained by recrystallization of the 88% ee material in

CHzClzlhexanes), [or]2°D +91.5° (c 1.25, CHZCIZ).

 

Si -
\N/J\ ph + O/ \ B(OPh)3 IVAPOL _
MOMe solvent 7
Br 156 _ 31 temperature
1.0 equiv 2.0 equiv. reaction time
Added over 3.0 h

 

in 3mL solvent
Procedures for Scheme 3.3 (aza-Diels-Alder reaction of imine 156):
I Condition A: The protocol for the optimal conditions was followed exactly where
0.3503 g (1.0 mmol) of imine 156 was used. The product (157) was then purified via

flash column chromatography (36 cm x 2 cm, 2:1 hexanes/ethyl acetate rf 0.07). The

225

reaction produced 351.4 mg (84% yield) of the desired product 157 and 89%
enantiomeric excess was measured (average of two runs).

Condition B: The protocol for the optimal conditions was followed exactly except
the catalyst was prepared using 0.09 g B(OPh)3 (30 mol%) and 54mg (S)-VAPOL (10
mol%). The reaction was run using a standard 25 eround bottom flask, and
CHzClzwas used as the solvent to transfer the catalyst and add the diene rather than the
1:1 mixture of CHzClzltoluene. The reaction produced 288.6 mg (69% yield) of the
desired product 157 and 68% enantiomeric excess was measured.

Spectral Data for compound 157 (C24H20BrNO):

The enantiomers could be separated by HPLC using a Chiralcel OJ-H column
(80:20 hexanes/isopropanol, 1mL/min). Retention times: 16.36 min. (minor) and 28.10
min. (major). 1H NMR (CDC13) 6 2.62 (dd, 1H, J=l6.5, 8.4 Hz), 2.81 (dd, 1H, J=l6.5,
6.9 Hz), 4.47 (t, 1H, J=7.5 Hz), 4.96 (d, 1H, J=7.8 Hz), 5.34 (s, 1H), 6.98-7.52 (m, 15H);
13C NMR (CDCI3) 6 43.19, 61.20, 67.90, 98.76, 122.07, 127.24, 128.05, 128.22, 128.63,
128.69, 128.82, 129.23, 132.01, 137.53, 137.76, 137.84, 150.95, 189.51; IR (CDCI3)
3028m, l653vs, 1578vs, 1487s, 14498, 12218, 11408 cm"; mass spectrum m/z (% rel
intensity) 419 M“ (18, 81Br), 417 M+ (19, 79Br), 182 (12), 168 (15), 167 (100), 166 (9),
165 (31), 152 (15), 103 (7), 102 (7), 77 (6), 51 (4), 50 (4). Anal calcd for CuHmBrNO:
C, 68.91; H, 4.82; N, 3.35. Found: C, 68.97; H, 4.60; N, 3.26. White solid, mp 141-142

°C. Optical rotation was taken on 90% ee material, [(11200 +117.2° (c 1.285, CHzClz).

226

 

. 31
I \
\N /l\ ph + 0 B(OPh)3 /VAPOL _
MOMe solvent
ON
158 31 temperature
1.0 equiv 2.0 equiv. reaction time OZN

 

Added over 3.0 h coil addition flask
in 3mL solvent
through cooling

addition coil

Procedures for Scheme 3.4 (aza-Diels-Alder reaction of imine 158):

Condition A: The protocol for the Optimal conditions was followed exactly where
0.3164 g (1.0 mmol) of imine 158 was used. The product (159) was then purified via
flash column chromatography (36 cm x 2 cm, 2:1 hexanes/ethyl acetate rf 0.04). The
reaction produced 249.9 mg (65% yield) of the desired product 159 and 73%
enantiomeric excess was measured (average of two runs).

Condition B: The protocol for the optimal conditions was followed exactly except
the catalyst was prepared using 54 mg (S)-VAPOL (10 mol%). The reaction produced
257.6 mg (67% yield) of the desired product 159 and 73% enantiomeric excess was
Observed (average of two runs). I

Condition C: The protocol described in condition B was followed exactly except
the reaction was cooled to ~78°C while the diene was being added and then warmed to
—45°C for an additional 39 hours (42 h total reaction time). The reaction produced 246
mg (64% yield) of the desired product 159 and 75% enantiomeric excess was measured.

Condition D: The protocol for the optimal conditions was followed exactly except

the catalyst was prepared using 0.09 g (30 mol%) and 54 mg (S)-VAPOL (10 mol%).

‘ 227

The reaction produced 188.4 mg (49% yield) of the desired product 159 and 64%
enantiomeric excess was measured.

Condition E: The protocol described in condition B was followed exactly except
toluene was used as the solvent to transfer the catalyst to the imine and add the diene .
rather than the 1:1 mixture of CH2C12/toluene. The reaction produced 103.8 mg (27%
yield) of the desired product 159 and 66% enantiomeric excess was measured.

Spectral Data for compound 159 (C24H20N203):

The enantiomers could be separated by HPLC uaing a Chiralpak AD column
(75:25 hexane/isopropanol, 1 mL/min). Retention times: 6.31 min. (minor) and 12.46
min. (major). 1H NMR (CDC13) 6 2.61 (dd, 1H, J=16.5, 6.6 Hz), 2.98 (dd, 1H, J=l6.5,
7.1 Hz), 4.69 (t, 1H, J=6.9 Hz), 5.02 (d, 1H, J=8 Hz), 5.39 (s, 1H), 7.00-7.48 (m, 13H),
8.21 (d, 2H, J==2 Hz); l3C NMR (CDCI3) 6 42.65, 60.83, 68.86, 99.05, 124.07, 127.18,
127.68, 128.23, 128.44, 128.84, 128.90, 129.14, 137.08, 137.75, 146.09, 147.49, 150.66,
188.60; 1R (CDC13) 306lw, 2910w', 16448, 1590vs, 1578vs, 15208, l346vs, 1219m,
1138m cm"; mass spectrum m/z (% rel intensity) 286 M+2 (2), 285 M+1 (1), 384 M”
(2), 355 (20), 341 (11), 327 (9), 281 (54), 267 ( 17), 251 (7), 227 (18), 226(14), 225 (45),
224 (19), 223 (13), 211 (20), 210 (14), 209 (67), 208 (42), 207 (100), 194 (19), 191 (19),
177 (10), 149 (13), 147 (13), 135 (16), 133 (18), 119(8), 105 (9), 103 (9), 91 (13), 77
(16), 75 (17), 73 (43), 51 (6). Light yellow solid, mp 192-197°C. Optical rotation was
taken on a waxy material (99.9 % ee determined by HPLC) that deposited on the side of

the flask when attempting to recrystallize the material, [or]2°D +106.5° (c 3.835, CHZCIZ).

228

/ 0M6 CHzClz/tolueneflzl)’
160 31 ~45°C
1.0 equiv 2.0 equiv. 24 h
Added over 3.0 h cooling coil flask

in 3mL solvent

\ X O’S'\
O N Ph + B(OPh)3/VAPOL

 

through the cooling

addition coil

Procedure for Scheme 3.5 (aza-Diels-Alder reaction of imine 160):

Condition A: The protocol for the optimal conditions was followed exactly where
0.2854 g (1.0 mmol) of imine 160 was used. The product (161) was then purified via
flash column chromatography (36 cm x 2 cm, 2:1 hexanes/ethyl acetate r, 0.09). The
reaction produced 293.4 mg (65% yield) of the desired product 161 and 94%
enantiomeric excess was measured (average of two runs).

Spectral Data for compound 161 (CZSHBNO):

The enantiomers could be separated by HPLC using a Chiralcel OD column (98:2
hexane/isopropanol, 1 mL/min). Retention times: 47.09 min. (minor) and 51.34 min.
(major). 1H NMR (CDC13) 6 1.83 (s, 3H), 2.63-2.79 (m, 2H), 4.77 (t, 1H, J=8.8 Hz), 5.05
(d, 1H, J=8 Hz), 5.20 (s, 1H), 7.03-7.40 (m, 14H), 7.59 (d, 1H, J=7.7 Hz); 13C NMR
(CDCI3) 6 18.35, 42.83, 57.93, 67.13, 98.71, 126.33, 126.59, 127.46, 127.81, 127.93,
128.00, 128.55, 128.75, 129.26, 130.90, 135.55, 136.49, 137.55, 137.91, 152.06, 190.38;
IR (CDCI3) 3027s, 2953m, 2938m, 2909m, 1642vs, 1582vs cm"; mass spectrum m/z (%
rel intensity) M+ 353 (42), 281 (14), 267 (4), 249 (3), 225 (15), 209 (15), 207 (27), 182
(13), 168 (16), 167 (100), 166(12), 165 (34), 152 (20), 133 (5), 117(7), 115(9), 104(7),

103 (7), 91 (10), 77 (9), 73 (10), 51 (5). Anal calcd for C25H23NO: C, 84.95; H, 6.56; N,

229

3.96. Found: C, 84.96; H, 6.47; N, 3.89. White solid, mp 174-175 °C. Crystallization
from hexanes/CHZCI2 gave 161 that was 99.9% ee. Optical rotation was taken on 99.9%

ee material, [QI2OD +9l.5° (c 1.25, CHZCIZ).

 

0 Ph I /
Si
’ \
O \N/kPh AA B(OPh)3/VAPOL _
/

+
OMe CH2C12/toluene (1:1)
162 31 —45°C
1.0 equiv 2.0 equiv. 24 h

 

Added over 3.0 h cooling coil flask
in 3mL solvent
through cooling

adidtion coil

Procedures for Scheme 3.6 (aza-Diels-Alder reaction of imine 162):

Condition A: The protocol for the optimal conditions was followed exactly where
0.3214 g (1.0 mmol) of imine 162 was used. The product (163) was then purified via
flash column chromatography (36 cm x 2 cm, 2:1 hexanes/ethyl acetate r, 0.14). The
reaction produced 299.9 mg (77% yield) of the desired product 163 and 94%
enantiomeric excess was measured (average of two runs).

Condition B: The protocol for the optimal conditions was followed exactly except
the catalyst was prepared using 54 mg (S)-VAPOL (10 mol%). The reaction produced
307.7 mg (79% yield) of the desired product 163 and 90% enantiomeric excess was
measured.

Spectral data for compound 163 (CmHgNO):
The enantiomers could be separated by HPLC using a Chiralcel OJ-H column

(95:5 hexane/isopropanol, 2 mL/min). Retention times: 39.53 min. (minor) and 58.78

230

min. (major). 1H NMR (CDC13) 6 2.96 (br s, 2H), 5.15 (d, 1H, 128 Hz), 5.29 (br S, 1H),
5.45 (s, 1H), 7.02-7.52 (m, 15H), 7.84-7.93 (m, 3H); 13C NMR (CDC13) 6 42.09, 68.07,
98.30, 122.46, 125.12, 125.67, 126.17, 127.27, 127.94, 128.14, 128.63, 128.72, 128.97,
129.20, 129.43, 130.10, 132.95, 134.15, 137.62, 138.16, 151.30, 190.02 (two aromatic
carbons not located); 1R(CDC13) 30618, 30318, 2901m, 1649vs, 1593vs, 15108, 1449m,
1389s, 1240m, 1028m, 911m cm"; mass spectrum m/z (% rel intensity) M“ 389 (33), 308
(8), 281 (5), 248 (4), 222(6), 207 (12), 182 (12), 167 (100), 166 (11), 165 (37), 152 (32),
128 (4), 115 (6), 77 (5), 51 (4). Yellow solid, mp 73-81 °C. Optical rotation was taken
on 90% ee material, [or]2°D ——2..9° (c 2.8, CH2C12)

H

 

0
MgCl
+ H LN ether/T HF > O
0°C to RT
F 1 h F
166. 167 78% 168

Preparation of 4-fluoro-2-methylbenzaldehyde (168):

Following the literature protocol,”1

to a flame dried argon purged 250 mL round
bottom flask equipped with a magnetic stir bar was added 4-fluoro-2-methyl-
phenylmagnesium chloride (166, 40 mL, 20 mmol, 0.5 M solution in THF). This was
cooled to 0°C and piperadine-l-carbaldehyde (167, 2.22 mL, 20 mmol) was added in
ether (20 mL) over 2 minutes. The resulting mixture was then allowed to warm to room
temperature and stir for 1 hour. To the flask was then added 3N HCl until the reaction
was acidic (monitored with litmus paper). The contents of the flask were then transferred
to a separatory funnel, extracted 2 times with ether. The ether layers were then

combined, washed with water followed by saturated sodium bicarbonate and then brine.

The organic layer was then dried over magnesium sulfate, filtered, and the solvent was

231

removed under reduced pressure. Purification was then accomplished by bulb-to—bulb

distillation (63°C, 5 mmHg) to give 2.15 g (78% yield) of the desired aldehyde 168.

/ 0M6 CH2C12/toluene (1 :1)?
169 31 —45°C
1.0 equiv 2.0 equiv. 24 h
Added over 3.0 h cooling coil flask

in 3mL solvent

9 SI
/ \
O \NAPh + O B(OPh),/VAP0L_
F

 

through cooling

adidtion coil

Procedures for Scheme 3.8 (aza-Diels-Alder reaction of imine 169):

Condition A: The protocol for the optimal conditions was followed exactly where
0.3034 g (1.0 mmol) of imine 169 was used. The product (170) was then purified via
flash column chromatography (36 cm x 2 cm, 2:1 hexanes/ethyl acetate r, 0.08). The
reaction produced 252.6 mg (68% yield) of the desired product 170 and 94%
enantiomeric excess was measured (average of two runs).

Condition B: The protocol for the optimal conditions was followed exactly except
the reaction was allowed to stir for 50 hours. The reaction produced 312.0 mg of the
desired product 170 (84% yield) and'89% enantiomeric excess was measured.

Condition C: The protocol for the optimal conditions was followed exactly except
the catalyst was prepared using 54 mg (S)-VAPOL (10 mol%). The reaction produced
289.7 mg (78% yield) of the desired product 170 and 91% enantiomeric excess was
measured.

Spectral Data for compound 170 (CstnFNO);

232

The enantiomers could be separated by HPLC using a Chiralcel OD column (with
guard column) (98:2 hexane/isopropanol, 1 mL/min). Retention times: 55.03 min.
(minor) and 63.06 min. (major). 1H NMR (CDC13) 6 1.83 (s, 3H), 2.63-2.79 (m, 2H),
4.77 (t, 1H, J=8.7 Hz), 5.05 (d, 1H, J=7.7 Hz), 5.40 (s, 1H), 7.03-7.38 (m, 13H), 7.58 (d,
1H, J=7.4 Hz); 13C NMR (CDC13) 6 18.35, 42.82, 57 .93, 67.13, 98.70, 126.33, 126.58,
127.46, 127.81, 127.93, 128.00, 128.55, 128.75, 129.25, 130.89, 135.55, 136.47, 137.55,
137.91, 152.07, 190.35; IR (CDC13) 3027w, 1645vs, 1578vs, 1443m, 12388 cm"; mass
spectrum m/z (% rel intensity) 371 M” (91), 342(1), 294(1), 262(7), 248 (8), 206(9),
182 (25), 168 (13), 167 (100), 166 (14), 165 (40), 152(8), 133 (10), 115(9), 77(6), 51
(5). White solid, 55-59 °C. Optical rotation was taken on 88% ee material, [or]2°D +56.9°

(c 2.185, CHZCIZ).

 

o 0
Me I H2 (latm) / Pd/C Me
. S) > ,. 3)
\“’ N MeOH \“ N
1
. A 22h H
F Ph Ph 45% yield F

Preparation of cycloadduct 164:

To a flame dried argon purged round bottom flask equipped with a magnetic stir
bar was added 170 (89% ee as determined by HPLC) (50 mg, 0.135 mmol) and 10%
palladium on carbon (28.7 mg, 0.027 mmol) and methanol (2.5 mL). The reaction flask
was then flushed with hydrogen gas and kept under 1 atm hydrogen for 22 hours. The
reaction mixture was then filtered through Celite was flushed with ether (200 mL). The

solvent was then removed under reduced pressure. Purification was accomplished via

233

flash column chromatography (hexanes/ethyl acetate 1:1, rf = 0.08) and gave 12.7 mg
(45% yield) of the desired product 164. The lH-NMR of 164 matched perfectly to that
reported in the literaturef>4 The absolute configuration was then determined by
comparison of the optical rotation to the (R)-enantiomer (97 % ee, [or]2°D +77.4° (c 0.18,
DMSO)64 of 164 reported in the literature. The configuration of 164 obtained from the
reaction described above is thus assigned as the (S)-enantiomer based on the optical

rotation taken on 89% ee material: [or]2°D —77.0° (c 0.18, DMSO).

 

/l\ \ 0’ \
Ph N/ N Ph + N B(OPhh/VAPOL>
/ OM, CH2C12/toluene(1:1)
H

 

171 31 —45 C
1.0 equiv 2.0 equiv. . 2411
Added over 3.0 h 000““3 00" flask

in 3mL solvent
through cooling

adidtion coil

Procedure for Scheme 3.10 (aza-DieIs-Alder reaction of imine 171):

Condition A: The protocol for the optimal conditions was followed exactly where
0.4646 g (1.0 mmol) of imine 171 was used. The product (172) was then purified via
flash column chromatography (36 cm x 2 cm, 2:1 hexanes/ethyl acetate r, 0.07). The
reaction produced 176.4 mg (48% yield) of the mono-Diels-Alder product 172 and 74%
enantiomeric excess was measured.

Spectral data for compound 172 (C25H21N02):

The enantiomers could be separated by HPLC using a Chiralcel OJ-H column

(75:25 hexane/isopropanol, 1 mL/min). Retention times: 31.15 min. (minor) and 40.07

min. (major). 1H NMR (CDC13) 6263-271 (dd, 1H, J=7.2, 16.5 Hz), 2.91-2.99 (dd, 1H,
234 ‘

J=7.2, 16.5 Hz), 4.66 (t, 1H, J=7.2 Hz), 5.02 (d, 1H, J=7.5 Hz), 5.39 (s, 1H), 6.98-7.84
(br m, 15H), 9.96 (s, 1H); l3C NMR (CDC13) 6 42.91, 61.16, 68.48, 98.80, 127.20,
127.55, 128.11, 128.29, 128.75, 128.82, 129.17, 129.69, 132.68, 136.78, 137.37, 137.90,
139.92, 150.92, 189.21, 191.28 (one overlapping 8p2 carbon); IR (neat) 868.08w, 912.45
m, 1001.18w, 1030.12m, 1080.27w, 1138.15 8, 1180.598, 1223.028, 1217.558, 1381.218,
1410.14m, 1448.738, 1495.028, 1756.04vs, 1645.49vs, 1699.50vs, 2375.95w, 2849.22m,
2922.53m, 2971.59m, 3030.56m, 3063.35m; mass spectrum m/z (% rel intensity) M+
367 (2), 167 (100), 165 (15), 164 (15), 151 (15), 106 (2), 103 (2), 77 (3); Light yellow
solid, mp 65-68°C softens and 112-116°C melted. Optical rotation was taken on 75% ee

material, [a]2°D +66.7 (c 2.687, CHZCIZ).

 

|./ 0
Sr
Ph 0’ \
WM N Ph / OMe solvent Ph \ N
173 31 . —45°c A
1.0 equiv 2.0 equiv. 24 h Ph Ph
Added over 3.0 h 174

in 3mL solvent

Procedures for Scheme 3.11 (aza-Diels-Alder reaction of imine 173):

Condition A: The protocol for the optimal conditions was followed exactly where
0.2974 g (1.0 mmol) of “imine 173 was used. The product (174) was then purified via
flash column chromatography (36 cm x 2 cm, 2:1 hexanes/ethyl acetate r, 0.05). The
reaction produced 40.2 mg (11% yield) of the desired product 174 and 0% enantiomeric

excess was measured (average of two runs).

235

Condition B: The protocol for the optimal conditions was followed exactly except
the catalyst was prepared using 0.09 g B(OPh)3 (30 mol%) and 54 mg (S)-VAPOL (10
mol%) and toluene was used as the solvent to transfer the catalyst to the imine and add
the diene rather than the 1:1 mixture of CHZCIZ/toluene. The reaction was allowed to stir
for 48 hours and produced 40.2 mg (11% yield) of the desired product 174 and 0%
enantiomeric excess was observed. 1
Spectral Data for compound 174 (C26H23NO):

The enantiomers could be separated by HPLC using a Chiralcel OD column (with
guard Column) (75:25 hexane/isopropanol, lmL/min). Retention times: 10.82 and 19.13
min. 1H NMR (CDC13) 6 2.52 (dd, 1H, J=16.2, 6.0 Hz), 2.86 (dd, 1H, J=16.5, 6.6 Hz),
4.14-4.20 (m, 1H), 4.96 (d, 1H, J=7.7 Hz), 5.67 (s, 1H), 6.29-47 (m, 2H), 6.87 (d, 1H,
J=77 Hz), 7.09 (d, 2H, J=6.0 Hz), 7.24—7.45 (m, 13H); .13C NMR (CDCI3) 641.28, 60.60,
68.25, 97.89, 124.19, 126.38, 127.29, 127.93, 128.02, 128.15, 128.40, 128.70, 128.76,
129.28, 133.75, 135.49, 137.87, 138.63, 150.19, 190.11; IR (CDC13) 3031m, 2926m,
1642vs, 1576vs, 1449m, 1223m, 1140m; mass spectrum m/z (% rel intensity) 365 M“
(54), 363 (11), 288 (7), 207 (11), 198 (11), 168 (34), 167 (100), 166 (12), 365 (35), 152

(27), 115 (10), 102 (7), 101 (7), 91 (14), 76 (8), 64(6). White solid, mp 148-151 °C.

236

 

Ph 0’ \
M /l\ + B(OPh)3/VAPOL_ I
\ r
\ N p}, MOW CHzClz/tolueneflzl) \ N

175 31 —-45°C A

1.0 equiv 2.0 equiv. 24 h 171:!
Added over 3.0 h cooling addition flask

Pb

in 3mL solvent
through the cooling
addition coil
Procedures for Scheme 3.12 (aza-Diels-Alder reaction of imine 175):

Condition A: The protocol for the optimal conditions was followed exactly where
0.2494 g (1.0 mmol) of imine 175 was used. The reaction failed to produce any of the
desired product 176. ‘

Condition B: The protocol for the optimal conditions were followed exactly
except the catalyst was prepared using 54 mg (S)-VAPOL (10 mol%). The reaction
failed to produce any of the desired product 176. 1
Spectral data for compound 176 (CnHBNO):

The enantiomers could be separated by HPLC using a Chiralpak AS column
(90:10 hexane/isopropanol, lmL/min). Retention times: 93.12 min. and 109.44 min. 1H
NMR (CDC13) 6 1.28 (8', 3H), 1.72 (s, 3H), 2.34—2.42 (dd, 1H, J: 9.0, 16.5 Hz), 2.55-2.62
(dd, 1H, J=6.0, 16.2 Hz), 4.19-4.27 (m, 1H), 4.89 (d, 1H, 1:7.8 Hz), 5.42 (d, 1H, J=9.9
Hz), 5.68 (8, 1H), 6.84 (d, 1H, J=7.8 Hz), 7.05 (d, 2H, J=6.9 Hz), 7.18 (d, 2H, J=6.6 Hz),
7.26-7.40 (br m, 6H); 13C NMR (CDCl3) 6 17.36, 25.48, 41.83, 55.99, 67.43, 97.69,
120.88, 127.38, 127.70,) 127.90, 128.52, 128.61, 129.19, 137.50, 138.34, 138.46, 150.86,
190.88; IR (CDC13) 576.79 m, 607.65m, 621.16w, 700.258, 734.97m, 752.33m,
1030.12w, 1140.08m, 1199.88m, 1242.32m, 1275.11m, 1311.76w, 1377.35w, 1446.80m,
1495.02w, 1577.97vs, 1643.56vs, 2916.74w, 2987.33w, 3051.67w, 3092.40w; mass

237

spectrum m/z (% rel intensity) 318 M+1 (41), 317 M+ (100), 316 (16), 302(4), 274 (11),
262 (6), 206 (4), 194(4), 168 (34), 167 (12), 166 (25), 165 (8), 153 (7), 150(9), 104(4),

77 (4); White solid, mp 82-93°C.

 

 

Si
’ \
\N * Ph + O B(OPh)3 / VAPOL
MOMe CH2C12ltoluene (1:1)
—45°C
177 31
- 24 h
1.0 equiv 2,0 equ1v. . . .
Added over 3 0 h coolrng addrtron flask

in 3mL solvent
through the cooling
addition coil

Procedures for Scheme 3.13 (aza-Diels-Alder reaction of imine 177):

Condition A: The protocol for the optimal conditions was followed exactly where
0.2754 g (1.0 mmol) of imine 177 Was used. The product (178) was then purified via
flash column chromatography (36 cm x 2 cm, 2:1 hexanes/ethyl acetate r, 0.07). The
reaction produced 99.6 mg (29% yield) of the desired product 178 and 93% enantiomeric
excess was measured (average of two runs).

Condition B: The protocol for the optimal conditions was followed exactly except
the catalyst was prepared using 54 mg (S)-VAPOL ( 10 mol%). The reaction produced
109.9 mg (32% yield) of the desired product 178 and 93% enantiomeric excess was
measured.

Condition C: The protocol described in condition B was followed exactly except
the reaction was allowed to stir for 48 hours. The reaction produced 154.6 mg (45%

yield) of the desired product 178 and 94% enantiomeric excess was measured.

238

Spectral data for compound 178 (CMHZSNO):

The enantiomers could be separated by HPLC using a Chiralcel OJ-H column
(90:10 hexane/isopropanol, 1mL/min). Retention times: 11.77 min. (minor) and 15.97
min. (major). 1H NMR (CDCI3) 6 1.53-1.79 (m, 4H), 2.04—2.15 (m, 4H), 2.56-2.69 (m,
2H), 3.88 (t, 1H, J=7.7 Hz), 4.87 (d, 1H, J=7.7 Hz), 5.50 (s, 1H), 5.57 (s, 1H), 6.93 (d,
1H, J=7.7 Hz), 7.04-7.42 (m, 10H); 13C NMR (CDC13) 621.84, 22.81, 23.87, 24.80,
39.78, 64.26, 67.10, 97.45, 127.17, 127.34, 127.70, 127.96, 128.58, 129.25, 134.01,
138.25, 128.58, 151.19, 190.82 (one aromatic carbon not located); IR (CDCI3) 3028w,
2928m, 1643vs, 1589vs, 1578vs, 1449m, 1235m, 1219m, 1142m cm"; mass spectrum
m/z (% rel intensity) M+ 343 (42), 326 (2), 281 (10), 267 (3), 225 (10), 209 (13), 208
(10), 207 (22), 193 (11), 168 (15), 167 (100), 166 (11), 165 (34), 152 (21), 133(5), 115
(5), 91 (10), 77 (12), 73 (9), 51 (6). White solid, mp 189-191 °C. Crystallization from
CH2C12/hexanes gave 178 that was 98% ee. Optical rotation was taken on a 98% ee

sample, [or]2°D +137.2° (c 1.055, CHZCIZ).

 

Ph | / 0
,Si .
xii/k ph + O \ B(OPh)3 / VAPOL _
MOMe solvent 7 N I
179 31 temperature /I\
1.0 equiv Added over 3.0 h reaction time Ph Ph
' ' in 3mL solvent 13°

Procedures for. Scheme 3.14 (aza-Diels-Alder reaction of imine 179):
Condition A: The protocol for the optimal conditions was followed exactly where
0.2514 g (1.0 mmol) of imine 179 was used. Upon analysis of the crude lH-NMR, it was

determined that the reaction failed to give any of the desired product 180.

239

Condition B: The protocol for the optimal conditions were followed exactly
except the catalyst was prepared using 87 mg (30 mol%) B(OPh)3 and 54 mg (S)-VAPOL
(10 mol%). A standard 25 mL round bottom flask was used instead of the flask with the
cold addition coil and toluene was used for the transfer of the catalyst to the imine and for
the addition of the diene rather than a 1:1 mixture of CHzClz/toluene. The reaction failed
to produce any of the desired product 180.

Condition C: The protocol described in condition B was followed exactly except
3.0 mmol (3 equivalents) of the diene were used. The reaction was stirred for 16 hours
failed to produce any of the desired product 180.

Condition D: The protocol described in condition B was followed exactly except
the reaction was stirred at room temperature for 16 hours. The reaction failed to produce

any of the desired product 180.

 

S" / 0
Ph 0/ l\ B OPh IVAPOL
\ )\ + M ( )3 = I I
w N Ph 0 Me solvent N
181 31 temperature /|\
1.0 equiv 2.0 equiv reaction time Ph Ph
Added over 3.0 h 132

in 3mL solvent

Procedures for Scheme 3.15 (aza-Diels-Alder reaction of imine 181):

Condition A: The protocol for the optimal conditions was followed exactly where
0.2774 g (1.0 mmol) of imine 181 was used. The product (182) was then purified via
flash column chromatography (361cm x 2 cm, 2:1 hexanes/ethyl acetate rf 0.09). The
reaction produced 196.9 mg (57% yield) of the desired product 182 and 76%

enantiomeric excess was measured (average of two runs).

240

Condition B: The protocol for the optimal conditions was followed exactly except
the catalyst was prepared using 54 mg (S)-VAPOL (10 mol%). The reaction produced
186.6 mg (54% yield) of the desired product 182 and 91% enantiomeric excess was
measured. 1

Condition C: The protocol described for condition B was followed exactly except
the reaction was allowed to stir for 46 hours. The reaction produced 314.4 mg (91%
yield) of the desired product 182 and 93% enantiomeric excess was measured.

Condition D: The protocol described for condition B was followed exactly except
the ratio of CHzCl2 to toluene was changed from 1:1 to 1:4. The reaction produced 186.6
mg (54% yield) of the desired product 182 and 94% enantiomeric excess was measured.

Condition E: The protocol described for condition B was followed exactly except
a mixture of CCl4/CH2CI2 (2:1) was used to transfer the catalyst to the imine and add the
diene rather than a 1:1 mixture of CH2C12/toluene. The reaction was allowed to stir for 47
hours and produced 165.8 mg (48% yield) of the desired product 182 and 91.5%
enantiomeric excess was measured (average of two runs).

Condition F: The protocol for the optimal conditions was followed exactly except
the catalyst was prepared using 0.09 g B(OPh)3 (30 mol%) and 54 mg (S)-VAPOL (10
mol%). Toluene was used to transfer the catalyst to the imine and add the diene rather
thania 1:1 mixture of CHzCIZ/toluene. The reaction produced 155.5 mg (45% yield) of
the desired product 182 and 95% enantiomeric excess was measured.

Condition G: The protocol described in condition F was followed exactly except

the reaction was performed in a traditional 25 mL round bottom flask stirred for 44 hours.

241

The reaction produced 262.6 mg (76% yield) of the desired product 182 and 93%
enantiomeric excess was measured.

Condition H: The pr0tocol described in condition G was followed exactly except
the reaction was stirred at 0°C for 48 hours. The reaction produced 145.1 mg (42% yield)
of the desired product 182 and 80% enantiomeric excess was measured.

Spectral Data for compound 182 (C2,,H27NO):

The enantiomers could be separated by HPLC using a Chiralcel OJ-H column
(90:10 hexanes/isopropanol, 1mL/min). Retention times: 8.98 min. (minor) and 13.85
min. (major). 1H NMR (CDCl3) 6 0.96-1.20 (m, 5H), 1.59—1.95 (m, 6H), 2.35 (d, 1H,
J=16.8 Hz), 2.67 (dd, 1H, J=17.1, 8.1 Hz), 3.27 (br t, 1H, J=1.9 Hz), 4.76 (d, 1H, J=7.5
Hz), 5.68 (s, 1H), 6.82 (d, 1H, J=7.8 Hz), 7.04 (d, 2H, J=7.5 Hz), 7.19-7.37 (m, 8H); l3C
NMR (CDCl3) 6 25.76, 26.00, 26.06, 28.08, 29.43, 36.05, 40.51, 61.78, 69.56, 96.71,
126.99, 127.93, 128.17, 128.66, 128.69, 129.31, 137.61, 139.45, 150.21, 190.70; 1R
(CDCl3) 3063w, 3030w, 2928vs, 2853vs, 1638vs, 1576vs, 14498, 12238, 11438 cm";
mass spectrum m/z (% rel intensity) 345 M+ (63), 346 (19), 344 (9), 262 (12), 208 (4),
182 (16), 168 (15), 167 (100), 166(9), 165 (31), 152 (19), 115(5), 91 (4), 77(6), 51 (4).
Anal calcd for C24H27NO: C, 83.44; H, 7.88; N, 4.05. Found: C, 83.03; H, 8.11; N, 4.09.
White solid, mp 132-134 °C. Optical rotation was taken on 93% ee sample, [onlm’D

—127.2° (c 1.13, CHzClz, 93% ee).

242

 

\N /I\ Ph + 0’ \ . B(oph)3 / VAPOL > I
Y MOMe solvent N
183 31 temperature A
1.0 equiv 2.0 equiv reaction time Ph Ph
Added over 3.0 h 134

in 3mL solvent

Procedures for Scheme 3.16 (aza-Diels-Alder reaction of imine 183):

Condition A: The protocol for the optimal conditions was followed exactly where
0.2374 g (1.0 mmol) of imine 183 was used. The produCt (184) was then purified via
flash column chromatography (36 cm x 2 cm, 2:1 hexanes/ethyl acetate rf 0.06). The
reaction produced 177.1 mg (58% yield) of the desired product 184 and 89%
enantiomeric excess was measured (average of two runs).

Condition B: The protocol for the optimal conditions was followed exactly except
the catalyst was prepared using 54 mg (S)-VAPOL (10 mol%). The reaction produced
195.5 mg (64% yield) of the desired product 184 and 90% enantiomeric excess was
measured.

Condition C: The protocol for the optimal conditions was followed exactly except
the catalyst was prepared using 0.09 g B(OPh)3 (30 mol%) and 54 mg (S)-VAPOL (10
mol%). A traditional 25 mL round bottom flask was used rather than the flask with the
cooling addition coil and toluene was used to transfer the catalyst to the imine and add
the diene rather than a 1:1 mixture of CH2C12/toluene. The reaction produced 174.1 mg
(57 % yield) of the desired product 184 and 91% enantiomeric excess was measured.

' Spctral Data for compound 184 (C,,H,,N0):
The enantiomers could be separated by HPLC using a Chiralpak AS, column

(70:30 hexanes/isopropanol, lmL/min). Retention times: 14.82 min. (major) and 32.15
243

min. (minor). 1H NMR (CDC13) 6 0.98 (d, 6H, J=9.3 Hz), 2.34 (d, 1H, J=16.5 Hz), 2.34—
2.41 (m, 1H), 2.64 (dd, 1H, J=17.1, 8.1 Hz), 3.28-3.33 (br m, 1H), 4.78 (d, 1H, J=7.2
Hz), 5.70 (s, 1H), 6.83 (d, 1H, J=7.8 Hz), 7.05 (d, 2H, J=7.5 Hz), 7.24—7.43 (m, 8H); 13C
NMR (CDC13) 6 17.44, 19.32, 29.90, 35.23, 62.06, 69.18, 96.83, 126.99, 127.93, 128.15,
128.64, 128.70, 129.34, 137.66, 139.28, 150.24, 190.77; IR (CDCI3) 29658, 2932m,
2899m, 2876m, 1640vs, 1578vs, 1449s, 12278, 11448; mass spectrum rn/z (% rel
intensity) M+1 306 (19), 305 M“ (75), 304 (13), 262 (37), 228 (10), 191 (9), 168 (15),
167 (100), 166 (9), 165 (34), 152 (18), 105 (5), 77 (8), 50 (5). Light yellow solid, mp

139-142 °C. Optical rotation was taken a a 84% ee sample, [(1]2OD —155.2° (c 0.93,

 

CHZCIZ).
Ph 0,51: 0
\ + . B(OPh)3 / VAPOL > |
W” P“ MOMe CH2C12/toluene (1 : 1)
185 31 —45°C, 24 h , 6 i
1.0 equiv 2.0 equiv cooling coil flask Ph Ph
Added over 3.0 h 186

in 3mL solvent
through cooling
. addition coil

Procedures for Scheme 3.17 (aza-Diels-Alder reaction of imine 185):

Condition A: The protocol for the optimal conditions was followed exactly where
0.2935 g (1.0 mmol) of imine 185 was used. The product (186) was then purified via
flash column chromatography (36 cm x 2 cm, 2:1 hexanes/ethyl acetate rf 0.10). The
I reaction produced 148.2 mg (41% yield) of the desired product 186 and 0% enantiomeric

excess was measured (average of two runs).

244

 

Condition B: The protocol for the optimal conditions was followed exactly except
(S)-VAPOL was excluded from the preparation of the catalyst. The reaction produced
144.6 mg (40% yield) of the desired product 186.

Condition C: The protocol for the optimal conditions was followed exactly except
(S)-VAPOL was excluded from the reaction and B(OPh)3 was used directly from the
bottle with no catalyst preparation. The reaction produced 108.5 mg (30% yield) of the
desired product 186.

Spectral data for compound 186 (C5H31NO):

The enantiomers could be separated by HPLC using a Chiralpak AD column
(95:5 hexane/isopropanol, 1mL/min). Retention times: 13.25 min. and 16.32 min. 1H
NMR (CDC13) 6 0.85 (t, 3H, J=6.9 Hz), 1.25-1.39 (br m, 10H), 1.68-1.71 (br m, 1H),
1.84-1.94 (br m, 1H), 2.31 (d, 1H, J=16.5 Hz), 2.75 (dd, 1H, J=16.5, 6.9 Hz), 3.46 (br s,
1H), 4.83 (d, 1H, J=7.4 Hz), 5.63 (s, 1H), 6.72 (d, 1H, J=6.9 Hz), 7.06 (d, 2H, J=9.0 Hz),
7.24-7.43 (m, 8H); 13C NMR (CDC13) 6 13.75, 22.25, 25.16, 28.79, 29.16, 31.37, 38.44,
57.28, 69.21, 96.50, 126.85, 127.06, 127.91, 128.17, 128.46, 128.61.129.23, 137.53,
139.26, 149.22, 190.24; IR (neat) 3063w, 3030w, 2928vs, 2857vs, 1645vs, 1576vs,
14568, 11458 cm"; mass spectrum m/z (% rel intensity) 362 M+1 (54), 361 M+ (100), 304
(4), 276(3), 262 (7), 206(3), 182 ( 16), 167 (75), 165 (30), 152 (14), 77 (3). Light yellow

solid, mp 115-117 °C.

245

+ B(OPh)3 / VAPOL = I

\ A M
\(V)? N Ph OMe dichloromethane N
185 31 ”45°C 6 /I\

1.0 equiv 2.0 equiv reaction time Ph Ph
Added over 3 0 h round bottom flask 186

 

in 3mL solvent

Procedures for Scheme 3.18 (aza-Diels-Alder reaction of imine 185):

Condition A: The protocol for the optimal conditions was followed exactly except
the catalyst was prepared using 0.090 g B(OPh)3 (30 mol%) and 54 mg (S)-VAPOL (10
mol%). A- standard 25 mL round bottom flask was used instead of the flask with the cold
addition coil and CH2C12 was used for the transfer of the catalyst to the imine and for the
addition of the diene rather than a 1:1 mixture of CH2C12/toluene. The reaction produced
32.5 mg (9% yield) of the desired product 186 and the enantiomeric excess was not
measured.

Condition B: To a flame dried, argon purged single-necked flask that had its 14/20
joint replaced with a threaded Teflon high-vacuum T-shaped stop—cock equipped with a
stir bar was added B(OPh)3 (0.090 g, 0.30 mmol) and (S)-VAPOL (54. mg, 0.10 mmol).
To this was added CHZCI2 (2 mL) and then the flask was sealed with the stopcock and
heated to 55 °C for one hour. After one hour, the solvent was removed via high vacuum
and left under high vacuum at 55 °C for 0.5 hours yielding the. VAPOL-boron catalyst.
After cooling the stopcock was removed and replaced with a rubber septum. The catalyst
was dissolved by the injection via syringe of 2.0 mL of CHZCI2 (in two portions). The

catalyst solution was transferred by syringe to the flask containing the imine.

To a flame dried, argon purged 25 mL round bottom flask equipped with a stir bar
was added magnesium sulfate (0.17 g) and CHZCI2 (2 mL) followed by octanal (1.0
mmol) and aminodiphenylmethane (redistilled, 1.0 mmol). The reaction was stirred
overnight and then placed under high vacuum for 3 hours. The catalyst was added in two
1.0 mL portions of CHzClz. This was allowed to stir for 5-10 minutes at room
temperature and then cooled to —45 °C. Meanwhile, in a separate flame dried 5 or 10 mL
round bottom flask purged with argon was added Danishefsky’s diene (31) (0.38 mL, 2.0
mmol) and CH2C12(3.0 mL). Workup was done according to the optimal conditions and
the reaction produced 65.1 mg (18% yield) of the desired product 186 and 0%
enantiomeric excess was measured (average of two runs).

Condition C: The protocol described in condition B was followed exactly except -
the reaction was only allowed to stir for 22 hours. The reaction produced 86.8 mg (24%
yield) of the desired product 186 and 4% enantiomeric excess was measured.

Condition D: The protocol described in condition B was followed exactly except
1.1 equivalents of the aldehyde was used for the in situ preparation of the imine. The
reaction was only allowed to stir for 21 hours and produced 75.9 mg (21% yield) of the
desired product 186 and 0% enantiomeric excess was measured.

Condition E: The protocol described in cOndition D was followed exactly except
the aldehyde was redistilled immediately before use and aminodiphenylmethane was
purified by distillation, then running a column, and then distilling one more time before
_ use in the in situ preparation of the imine. ~The reaction was allowed to stir for 24.5 hours
and produced 122.9 mg (34% yield) of the desired product 186 and 0% enantiomeric

excess was measured.

247

Condition F: The protocol described in condition D was followed exactly except
the aldehyde was distilled and the aminodiphenylmethane purified by column
chromatography immediately before use in the in situ preparation of the imine. The
reaction Was allowed to stir for 21.5 hours produced 169.9 mg (47% yield) of the desired
product 186 and 4% enantiomeric excess was measured.

Condition C: The protocol described in condition D was followed exactly except
0.18 g B(OPh)3 (60 mol%) and 108 mg (S)-VAPOL (20 mol%) were used for the
preparation of the catalyst. The reaction was allowed to stir for 20 hours and produced

112.1 mg (30% yield) of the desired product 186 and 0% enantiomeric excess was

 

measured.
I. / 0
Ph 0,31 \

\ A + B(OPh)3 / VAPOL _ I

WN Ph MOMe toluene N
185 31 —45°c 6 /|\

1.0 equiv 2.0 equiv reaction time Ph Ph
Added over 3 o h round bottom flask 186

in 3mL solvent

Procedures for Scheme 3.19 (aza-DieIs-Alder reaction of imine 185):

Condition A: To a flame dried, argon purged single-necked flask that had its 14/20
joint replaced with a threaded Teflon hi gh-vacuum T-shaped stop-cock equipped with a
stir bar was added B(OPh)3 (0.090 g, 0.30 mmol) and (S)-VAPOL (54 mg, 0.10 mmol).
To this was added CH2C12 (2 mL) and then the flask was sealed with the stopcock and
heated to 55 °C for one hour. After one hour, the solvent was removed via high vacuum

and left under high vacuum at 55 °C for 0.5 hours yielding the VAPOL-boron catalyst.

248

After coOlin g the stopcock was removed and replaced with a rubber septum. The catalyst
was dissolved by the injection via syringe of 2.0 mL of toluene (in two portions). The
catalyst solution was transferred by syringe to the flask containing the imine as prepared
below.

To a flame dried, argon purged 25 mL round bottom flask equipped with a stir bar
was added magnesium sulfate (0.17 g) and CHZCI2 (2 mL) followed by octanal (1.0
mmol) and aminodiphenylmethane (redistilled, 1.0 mmol). The reaction was stirred
overnight and then placed under high vacuum for 3 hours. The VAPOL-boron catalyst
was added in two 1.0 mL portions of toluene. This was allowed to stir for 5-10 minutes
at room temperature and then cooled to —45 °C. Meanwhile, in a separate flame dried5
or 10 mL round bottom flask purged with argon was added Danishefsky’s diene (31)
(0.38 mL, 2.0 mmol) and toluene (3.0 mL). The diene was taken up in a syringe and
added over 3.0 hours via syringe pump. The reaction was then allowed to stir at -45 °C
for the duration of the reaction (20 hours). Workup was done according to the optimal
conditions and the reaction produced 61.5 mg (17 % yield) of the desired product 186 and
20% enantiomeric excess was measured.

Conditions B: The protocol described in condition A was followed exactly except
(S)-VANOL was used instead of (S)-VAPOL for the preparation of the catalyst. The
reaction was allowed to stir for 51 hours and produced 75.9 mg (21% yield) of the
desired product 186 and 0% enantiomeric excess was measured.

Conditions C: The protocol described in condition A was followed exactly except

(S)-BINOL was used instead of (S)-VAPOL for the preparation of the catalyst The

249

reaction was allowed to stir for 51 hours and produced 54.2 mg (15% yield) of the
desired product 186 and 18% enantiomeric excess was measured.

Conditions D: The protocol described in condition A was followed exactly except
tris-(2,6-dimethylphenyl)borate was used in place of B(OPh)3 for the preparation of the
catalyst. The reaction produced 83.1 mg (23% yield) of the desired product 186 and 16%

enantiomeric excess was measured.

 

|./ 0
Sr
B(OPh)3 / VAPOL
\ + >
M N Ph MOMe toluene N I
l 0 equiv 2 0 equiv round bottom flask Ph Ph
190
Added over 3.0 h

in 3mL solvent

Procedure for Scheme 3.20 (aza-Diels-Alder reaction of imine 189):

Condition A: To a flame dried, argon purged single-necked flask that had its 14/20
joint replaced with a threaded Teflon hi gh-vacuum T-shaped stop-cock equipped with a
stir bar was added B(OPh)3 (0.09g, 0.3 mmol) and (S)-VAPOL (54 mg, 0.10 mmol). To
this was added CH2C12(2 mL) and then the flask was sealed with the stopcock and heated
to 55 °C for one hour. After one hour, the solvent was removed via high vacuum and left
under high vacuum at 55 °C for 0.5 hours yielding the (S)-VAPOL catalyst. After
cooling the stopcock was removed and replaced with a rubber septum. The catalyst was
dissolved by the injection via syringe of 2.0 mL of a toluene (in two portions). The
catalyst solution was transferred by syringe to a flask containing the imine prepared as

prepared immediately below.

250

To a flame dried, argon purged 25 mL round bottom flask equipped with a stir bar
was added the imine (189) (0.2373 g, 1.0 mmol). The flask was topped with a rubber
septa and the catalyst was added in two 1.0 mL portions of toluene. This was allowed to
stir for 5-10 minutes at room temperature and then cooled to —50 °C. Meanwhile, in a
separate flame dried 5 or 10 mL round bottom flask purged with argon was added
Danishefsky’s diene (31) (0.38 mL, 2.0 mmol) and toluene (3.0 mL). The diene was
taken up in a syringe and added over 3.0 hours via syringe pump through the cold
addition coil. The reaction was then allowed to stir at —50 °C for the duration of the
reaction 24 hours. Workup was done according to the optimal conditions and the
reaction produced 106.9 mg (35% yield) of the desired product 190 and 22%
enantiomeric excess was measured (average of two runs).

Spectral data for compound 190 (C24H23NO):

The enantiomers could be separated by HPLC using a Chiralpak AS column
(90:10 hexane/isopropanol, 1mL/min). Retention times: 13.25 min. (major) and 16.32
min. (minor). 'H NMR (CDC13) 6 0.91 (t, 3H, J=7.2 Hz), 1.17-1.32 (br m, 1H), 1.36-1.51
(br m, 1H), 1.60-1.71 (br m, 1H), 1.85-1.98 (br m, 1H), 2.26-2.33 (dd, 1H, J=2.4, 16.5
Hz), 2.71-2.79 (dd, _1H, J=7.2, 16.8 Hz), 3.42-3.51 (m, lH),4.83 (d, 1H, J=7.5.Hz), 5.62
(s, 1H), 6.71 (d, 1H, J=7.5 Hz), 7.05-7.07 (m, 2H), 7.29-7.43 (br m, I8H); 13C ,NMR
(CDC13) 6 13.66, 18.443, 30.84, 38.51, 57.10, 62.25, 96.60, 127.06, 127.91, 128.17,
128.69, 129.47, 137.58, 139.34, 149.11, 190.23; IR (neat) 700.25m, 740.76m,
1030.12vw, 1080.27vw, 1124.64w, 1145.86m, 1184.44m, 1215.31m, 1240.39m,
1294.40w, 1319.48w, 1381.21vw, 1415.93vw, 1448.73m, 1495.02vw, 1576.04vs,

1635.84vs, 2860.35w, 2784.30w, 1898.65w, 2927.99m, 2959.18m, 2012.49vw,

251

3028.63w, 3046.88, 3059.78w, 3088.20w; mass spectrum m/z (% rel intensity) 306 M+1
(51), 305 M+ (66), 304 (22), 276 (2), 262 (2), 248 (3), 229 (2), 183 (1), 169 (5), 168
(100), 167 (32), 166 (43), 165 (11), 164(3), 152 3 (8), 152 (16), 139(2), 128(3), 116(4),

104(4), 77 (4), 51 (4), 50(4); light yellow solid, mp 157—159°C.

 

. Sli/ 0
j: + O, \ B(OPh)3 IVAPOL
Ph VOVEN Ph Mom CHzClz/toluene (1 :1); Ph Vo N I
191 31 -4S°C ‘ A
1.0 equiv 2.0 equiv. 24 h Ph Ph
Added over 3.0 h cooling coil flask 192

in 3mL solvent
through the cooling
addition coil
Procedure for Scheme 3.21 (aza-Diels-Alder reaction of imine 191):

Condition A: The protocol for the optimal conditions was followed exactly where
0.3154 g (1.0 mmol) of imine 191 was used. The product (192) was then purified via
flash column chromatography (36 cm x 2 cm, 2:1 hexanes/ethyl acetate rf 0.05). The ~
reaction produced 168.7 mg (one spot by TLC) of a complex mixture of products. The
1H NMR indicated that the desired product 192‘ may have been formed but could not be
isolated (average of two runs).

Spectral Data for compound 192:
' Even after two purifications, a single spot as observed by TLC analysis shows 1H

NMR and l3C-NMR that contain a complex mixture of peaks.

252

 

h Cl 0 O , Ph
P \S'/ imidazole Pb \Si/ U \SIi
1

HO OH + I _
I r-
U— X ph toluene, DMF Ph 195 Ph

193 194 0°C to RT
1) O3
2) Tri phenylphosphine

0 ,Ph
0:4(7 \Sij<

|
.H Ph

97%

196

Preparation of 2-tert-butyl-diphenylsilyloxyethana1 196:

To a flame dried argon purged 100 mL round bottom flask equipped with a
magnetic stir bar was added. imidazole (10. 2 g, 150 mmol) and DMF (20 mL). The
solution was cooled to 0 °C and the (Z)-2-butene-1,4—diol (193, 4.12 mL, 50 mmol) was
added followed immediately by the addition of a solution of TBDPSCI in 30 mL toluene
(194, 30.24 g, 110 mmol). The reaction mixture was then allowed to warm up to room
temperature and stir for 44 hours. After 44 hours, water (50 mL) was added, the solution
transferred to a separatory funnel and extracted three times with ether (3 x 50 mL). The
organic layers were combined and washed four times with water (4 x 50mL) and once
with brine (50 mL). The organic layer was dried over magnesium sulfate, filtered and the
solvent was removed under reduced pressure. Distillation was used to remove any
volatile compounds and 1H NMR indicated that the material left in the distillation flask
was 195. The unpurified 195 was then used for the ozonolysis step.

1,”2 to a flame dried argon purged 500 mL round

Following a literature protoco
bottom flask equipped with a magnetic stir bar was added the bis-protected diol (195,
6.56 g, 11.6 mmol) and a 1:1 mixture of CHZCIZ/methanol (220 mL). The solution was

cooled to —78 °C and while stirring, ozone was bubbled through the solution until the

253

color turned light blue (~30 min). To the blue solution was added PPh3 (4.68 g, 17.85
mmol), it was allowed to warm to room temperature, and stir for 2 hours. The solvent
was removed under reduced pressure and the product was purified by flash column
chromatography (85:15 hexanes/ethyl acetate) to give 6.7 g (97% yield) of the desired

aldehyde 196.

Si
’ \
P“ 3(0me 0 B(OPh)3 / VAPOL k

Ph +
>r ILII MOM, CH2C12/toluene(1:l) Ph\s.,0 N I
. I
197 31 —45°C >rl
Ph
- 24h A

1.0 equiv 2.0 equiv. Ph
Added over 3.0 h cooling coil flask 198

 

Pb

in 3mL solvent
through the cooling
addition coil
Procedure for Scheme 3.23 (aza-Diels-Alder reaction of imine 197):

Condition A: The protocol for the optimal conditions was followed exactly where

0.4637 g (1.0 mmol) of imine 197 was used and the reaction failed to produce any of the

 

desired product 198.
Ph 1 / 0
Si
’ \
E‘OYBN/kph + _ O B(OPh)3 / VAPOL _ I
0 . MOMe solvent EtO N
199 31 tern rature
1.0 equrv 2.0 equiv. reaction time Ph Ph
Added over 3.0 h cooling coil flask 200

in 3mL solvent
through the cooling

addition coil
Procedures for Scheme 3.24 (aza-Diels-Alder reaction of imine 199):

254

Condition A: The protocol for the optimal conditions was followed exactly except
54 mg of (S)-VAPOL (10 mol%) was used for the preparation of the catalyst. 0.2673 g
(1.0 mmol) of imine 199 was used for the reaction. The product (200) was purified via
flash column chromatography (36 cm'x 2 cm, 2:1 hexanes/ethyl acetate rf 0.07) to give
268.3 mg (80% yield) of the desired product 200 and 52% enantiomeric excess was
measured (average of two runs).

Condition B: The protocol described in condition A was followed exactly except
the reaction was allowed to take place at —60°C. The reaction produced 241.5 mg (72%
yield) of the desired product 200 and 35% enantiomeric excess was measured.

Condition C: The protocol described in condition A was followed exactly except
the reaction was allowed to take place at -—30°C. The reaction produced 278.4 mg (83%
yield) of the desired product 200 and 55% enantiomeric excess was measured.

Condition D: The protocol described in condition A was followed exactly except
the catalyst was prepared using 0.09 g B(OPh)3 (30 mol%) and 54 mg (S)—VAPOL (10
mol%). The reaction produced 295.2 mg (88% yield) of the desired product 200 and
53% enantiomeric excess was measured.

Condition E: The protocol described in condition A was followed exactly except
the catalyst was transferred to the imine and the diene was added in CCl4/CH2C12 (2:1)
rather than a 1:1 mixture CHZCIZ/toluene. The reaction produced 248.2 .mg (74% yield)
of the desired product 200 and 60% enantiomeric excess was measured.

Condition F: The protocol de8cribed in condition E was followed exactly except

the reaction was allowed to take place at —78°C for 48 hours. The reaction produced

255

130.8 mg (39% yield) of the desired product 200 and 30% enantiomeric excess was
measured.
Spectral data for compound 200 (CZIHZINO3):

The enantiomers could be separated by HPLC using a Chiralcel OJ-H column
(75:25 hexane/isopropanol, 1.0 mL/min). Retention times: 8.94 min. (major) and 18.61
9 min. (minor). 1H NMR (CDCI3) 6 1.15 (m‘, 3H), 2.74 (m, 2H), 4.10 (m, 3H), 4.78 (d, 1H,
J=7.1 Hz), 5.69 (s, 1H), 6.74 (d, 1H, J=7.7 Hz), 7.23 (m, 10H); ”C NMR (CDC13)
614.14, 37.78, 59.46, 62.06, 70.87, 98.72, 127.52, 128.39, 128.65, 129.03, 129.11,
129.85, 137.53, 139.11, 150.77, 170.23, 188.75; IR (CDC13) 3066.49, 3039.11, 2982.33,
2950.10, 2905.67, 1736.16, 1647.42, 1589.55, 1495.02, 1448.73, 1377.35, 1321.41,
1221.10, 184.44, 1140.08, 1032.05, 935.83 cm"; mass spectrum m/z (% rel intensity) 336
M+1 (29), 335 M-l- (44), 289 (10), 262 (5), 260 (4), 169 (4), 168 (32), 167 (100), 166
(28), 153 (12), 152 (4), 115 (4), 51 (3); light yellow wax, the optical rotation was taken

on 52% ee material, [(11200 —34.3° (c 1.782, CHZCIZ).

 

.- \ /
lPrO\n/\NJ\PII + k B(OPh)3/VAPOL I .
. / .

OMe CH2c12/toluene(1:1)r r—PrO

201 31 I V .—45°c 0 /l\

1-0 chIV 2.0 equiv. 24 h ph ph
Added over 3.0 n cooling coil flask 202
in 3mL solvent
through the cooling
addition coil

Procedures for Scheme 3.24 (aza-Diels-Alder reaction of imine 201):

256

 

Condition A: The protocol for the optimal conditions was followed exactly except
54 mg of (S)-VAPOL (10 mol%) was used for the preparation of the catalyst. 0.2814 g
(1.0 mmol) of imine 201 was used for the reaction. The product (202) was purified via
flash column chromatography (36 cm x 2 cm, 2:1 hexanes/ethyl acetate rf 0.07). The
reaction produced 283.0 mg (80% yield) of the desired product 202 and 52%
enantiomeric excess was measured (average of two runs).
Spectral data for compound 202 (C22H23NO3):

The enantiomers could be separated by HPLC using a Chiralcel OJ-H column
(75:25 hexane/isopropanol, 1.0 mL/min). Retention times: 7.00 min. (major) and .12.80
min. min. (minor). 1H NMR (CDCI3) 61.20 (m, 6H), 2.79 (m, 2H), 4.08 (m, 1H), 4.85 (d,
1H, J=7.7 Hz), 5.04 (m, 1H), 5.72 (s, 1H), 6.80 (d, 1H, 1:7.7 Hz) 7.30 (m, 10H); 13C
NMR (CDC13) 621.4, 37.55, 59.22, 69.74, 70.59, 98.48, 127.20, 128.06, 128.32, 128.70,
128.79, 129.49, 137.22, 138.78, 150.53, 169.44, 188.51; IR (CDC13) 3081.22, 3047.54,
2982.33, 2946.07, 1734.23, 1653.21, 1591.48, 1448.73, 1375.42, 1321.41, 1221.10,
1128.51, 1140.08, 1105.35, 1010.83, 740.76, 702.18cm"; mass spectrum m/z (% rel
intensity) 350 M+1 (16), 349 M+ (32), 289 (8), 262 (6), 168 (12), 167 (100), 166 (25),
165 (6), 153 (12); White solid, the optical rotation was taken on 56% ee material, [o't]2°D

—26.2° (c 2.963, CH,Cl,).

 

 

o
TIPS _ H 1)EtMgBr(204) _
_ 2) DMF, THF, reflux ’ // H
203 65% TIPS 205

Preparation of 3-triisopropylsilyl-2-propynal (205):113
To a flame dried argon purged 500 mL round bottom flask was added magnesium

turnings (1.07 g, 44.6 mmol) and THF (178 mL). To the flask was then added
257

bromoethane (3.33 mL, 44.6 mmol). The resulting mixture heated to reflux and allowed
to stir overnight. The resulting 0.25M solution of Grignard reagent 204 was then
transferred to a flame dried argon purged 1000 mL round bottom flask that had been
previously charged with TIPS—acetylene (203, 10 mL, 44.6 mmol) and THF (178 mL).
The contents were then heated to reflux for 10 minutes and then cooled back to room
temperature at which time the solution was transferred to flame dried argon purged 1000
mL round bottom flask that had previously been charged with DMF (21.1 mL, 73.09
mmol) and THF (178 mL). The resulting reaction mixture was then heated to reflux for
10 minutes. Upon cooling to room temperature, 1 N HCI (92 mL) was added and the
reaction mixture was then transferred to a 2 L separatory funnel and extracted three times
with EtzO. The organic layers were then combined and dried over magnesium sulfate,
filtered and the solvent Was removed under reduced pressure. The crude oil was then
purified by bulb-to-bulb distillation (93-95°C, 2 mmHg) to give 6.12 g (65% yield) of the

pure aldehyde 205. The NMR matched exactly to that in the literature."3

81
_.< /N/kPh + 0’ \ B(OPhh/VAPOLAV I/E/EB
Si MOMe CHZClz/toluene (1:1) _(S
Y >— 206 31 —45°C
1.0 equiv 2.0 equiv. 24h \FS' )_2II71>11/k Ph

Added over 3,0 h cooling c011. flask

in 3mL solvent

 

through the cooling

addition coil

Procedures for Scheme 3.27 (aza-DieIs-Alder reaction of imine 206):

258

Condition A: The protocol for the optimal conditions was followed exactly except
54 mg of (S)-VAPOL (10 mol%) was used for the preparation of the catalyst. 0.3756 g
(1.0 mmol) of imine 206 was used for the reaction. The product (207) was purified via
flash column chromatography (36 cm x 2 cm, 2:1 hexanes/ethyl acetate r, 0.23). The
reaction produced 315.0 mg (71%. yield) of the desired product 207 and 22%
enantiomeric excess was measured.
Spectral data for compound 207 (QOH37NO):
The enantiomers could be separated by HPLC using a Chiralcel OJ-H column
(98:2 hexane/isopropanol, 1.0 mL/min). Retention times: 10.79 min. (minor) and 14.38
min. (major). 1H NMR (CDCI3) 6 1.05 (s, 21H), 2.70 (dd, 2H, J=7.3, 15.9 Hz), 4.31 (m,
1H), 4.98 (d, 1H, J=8.0 Hz), 5.98 (s, 1H), 6.77 (d, 1H, J=7.9 Hz), 7.25 (m, 10H); 13C
NMR. (CDC13) 6 11.00, 18.49, 42.12, 50.81, 68.93, 87.64, 99.74, 103.15, 127.57, 128.15,
128.31, 128.89, 128.95, 129.36, 138.25, 138.41, 150.02, 189.86; IR (neat) 679.03m,
700.25m, 734.97w, 754.26w, 883.51m, 1142.00m, 1180.59w, 1219.17m, 1238.46m,
1275.11w, 1311.76w, 1454.51m, 1591.48vs, 1655.13vs, 2864.668, 2891.66m, 2943.758,
3030.56w, 3063.35w cm"; mass spectrum m/z (% rel intensity) 444 M+1 (54), 443 M+
(100), 415 (27), 400 (25), 373 (12), 276 (11), 235 (7), 168 (26), 167 (98), 166 (34), 165
(1'1). Colorless wax. Optical rotation was taken on 22% ee material, [011200 +60.2° (c

2.233, CDC13).

259

 

H 1) n-BuLi, ether, —78°C Sli/
/\/\/ = / \
2) TMSCI, RT, 2n /\/\/
208 209

l) DIBAL (neat), ether, 40°C, 1h
2) Brz, CH2C12, —78°C, ”'1

 

O .
MAL“ < 1) sec-BuLl, THF, —90 C, 0.5 h Br
SI 2) DMF, —90°c to RT, 3 h M
211 / I \ Si

87% Z isomer

Preparation of Z-2-trimethylsi1y1-2-ocetnal (21 l):114

Silylacetylene 209: To a flame dried 500 mL round bottom flask equipped with a
magnetic stir bar was added l-heptyne (208, 30 mL, 0.229 mmol) and 3,0 (125 mL).
The solution was then cooled to —78 °C at which time n-BuLi (146 mL, 0.233 mmol, 1.6
M solution in hexanes) was added followed by chlorotrimethylsilane (29.5 mL, 0.233
mmol). The reaction mixture was allowed to stir while warming to room temperature for
2 h. After 2 h, the reaction mixture was quenched with ice-cold water and extracted with
pentane. The organic layers were combined and washed with water followed by brine.
The organic extracts were then dried with MgSO4, filtered, and the solvent was removed
under reduced pressure. The crude oil was then purified via distillation (70°C, 10
mmHg) to give pure 209. The 1H NMR spectra matched perfectly with those in the

literature. 1 ‘4

260

(E)-I—bromohept-1—enyl)tfimethylsilane (210): To a flame dried 500 mL round
bottom flask equipped With a magnetic stir bar was added (hept-l-ynyl)trimethylsilane
(209, 8.14 g, 48.33 mmol) and Et,O (25 mL). The reaction flask was then placed in a
water bath (room temperature) and to the solution was added neat DIBAL-H (9.87 mL,
53.2 mmol). The reaction mixture was then heated to 40 °C and stirred for 1 h after
which it was cooled to 0 °C and added pyridine (7.73 mL, 101.5 mmol) was added. The
resulting solution was cooled to —78°C and Br, (3.24 mL, 62.8 mmol) was added in
CH,C1, (42 mL) and stirred for 1h. The —78°C solution was transferred to a vigorously
stirring mixture of NaOH (7.73 g), pentane (50 mL), and ice (60 g). The resulting
solution was then transferred to a separatory funnel and extracted three times with
pentane. The organic layers were combined and washed with 1N HCI, 20% solution of
CdCl3 (18 g) in H,O (87 mL), 1N HCl, and then saturated sodium bicarbonate. The
organic layers were combined and dried with MgSO4, filtered, and the solvent was
removed under reduced pressure. The resulting oil was then filtered through a plug of
SiO2 and flushed with hexanes (~400-500 mL) to give 10.11 g of the desired 100% E-

isomer product 210 (84% yield) as a pure colorless oil.

0
WWI-LII
211 /S|l\

261

2-(trimethylsilyl)oct-2-enal 211: Following a literature protocol”, to a flame
dried 100 mL three—neck round bottom flask equipped with a magnetic stir bar and
thermometer was added ((E)—1-bromohept-1-enyl)trimethylsilane (210, 3.19 g, 12.86
mmol) and THF (30 mL). The solution was then cooled to —90°C (acetone, N,). To the
reaction flask was then added sec-BuLi (10.1 mL, 14.15 mmol, 1.4 M solution in
cyclohexane) very slowly so as to maintain the temperature of the reaction mixture below
.—85°C. The resulting solution was then allowed to stir at —90°C for 30 minutes. To
the reaction mixture was then added DMF (1.99 mL, 25.72 mmol) very slowly so as to
again maintain the temperature below —85°C. After addition of DMF was complete, the
reaction was warmed slowly (~2 hrs) to room temperature and stir for an additional 1 h.
The reaction was quenched with water (pH 7 buffer) and extracted with ether. The
organic layer was then dried over MgSO4, filtered, and the solvent was removed under
reduced pressure. lH-NMR of the crude mixture indicated that there was only 87%
retention of the geometry of the olefin. Purification was done by eluting the product 211
with hexanes/EtOAc (2:1) through a plug of silicagel to give a mixture of the Z and E
isomers as a yellow oil.
Spectral data for compound 211 (C,,H,,SiO):

Z—Isomer: 1H NMR (CDC13) 6 0.20 (s, 9H), 0.89 (m, 3H, overlap with E-isomer),
1.28 (m, 4H, overlap with E—isomer), 1.48 (m, 2H, overlap with E-isomer), 2.41 (q, 2H,
J=7.5 Hz), 7.06 (t, 1H, J=7.2 Hz), 9.38 (8, 1H); 13C NMR (CDC13) 6; -1.85, 14.06, 25.53,
27.88, 31.51, 32.50, 142.23 (two overlapping peaks).

E—Isomer: 1H NMR (CDCl,) 6 0.10 (s, 9H), 0.89 (m, 3H, overlap with Z-isomer),

1.28 (m, 4H, overlap with Z—isomer), 1.48 (m, 2H, overlap with Z—isomer), 2.57 (q, 2H,

262

J=7.8 Hz), 6.81 (t, 1H, J=7.5 Hz), 10.27 (s, 1H); 13C NMR (CDC13) 6; minor isomer C-
peaks not detectable after 500 scans); IR (neat) 629.53w, 729.19w, 758.12w, 944.93vs,
972.25m, 1140.08m, 1250.038, 1379.28w, 1408.21w, 1466.09w, 1697.91vs, 1799.81w,
2731.55 w, 2860.80vs, 2928.32vs, 2957.25vs cm"; mass spectrum m/z (% rel intensity).
198 M+ (6), 184 (14), 183 (81), 182 (34), 181 (11), 155 (14), 142 (12), 141 (10), 139
(49), 128 (12), 127 (95), 126 (19), 125 (11), 113 (81), 112 (15), 111 (41), 103 (17), 85

(13), 77 (12), 75 (90), 74 (13), 73 (100), 67 (12), 61 (10), 59 (20), 55 (10), 47 (12), 45

 

(32).
i/ 0
Ph 0’ l\
/I\ + B(OPhb/VAPOL;
\ V
W” ph Mom CH,Cl,/toluene (1:1) I
s' ' —— 5°C \ N
212 Added over 3.0h ”°°“°“_“m° // Ph Ph
_ in 3mL solvent cooling c011 flask 213
1.0 equ1v

through the cooling

addition coil

Procedures for Scheme 3.29 (aza-Diels-Alder reaction of imine 212):
Condition A: To a flame dried, argon purged single—necked flask that had its 14/20
joint replaced with a threaded Teflon high-vacuum T-shaped stop—cock equipped with a
stir bar was added B(OPh)3 (0.3125 g, 1.0 mmol) and (S)-VAPOL (54 mg, 0.10 mmol).
To this was added CH,C1, (2 mL) and then the flask was sealed With the stopcock and
heated to 55 °C for one hour. After one hour, the sclvent was removed via high vacuum
and left under high vacuum at 55 °C for 0.5 hours yielding the VAPOL-boron catalyst.
After Cooling the stopcock was removed and replaced with a rubber septum. The Catalyst

was dissolved by the injection via syringe of 2.0 mL of toluene (in two portions). The

263

catalyst solution was transferred by syringe to the flask containing the imine as prepared
below.

To a flame dried, argon purged 25 mL round bottom flask equipped with a stir bar
was added magnesium sulfate (0.17 g) and CH,C1, (2 mL) followed by the aldehyde 211
(0.1984g, 1.0 mmol) and aminodiphenylmethane (153, 0.1833 g, 1.0 mmol). The reaction
I was stirred overnight and then placed under high vacuum for 3 hours. The VAPOL-
boron catalyst was added in two 1.0 mL portions of CH,Cl,/toluene (1:1). This was
allowed to stir for 5-10 minutes at room temperature and then cooled to -45 °C.
Meanwhile, in a separate flame dried 5 or 10 mL round bottom flask purged with argon
was added Danishefsky’s diene (31) (0.38 mL, 2.0 mmol) and CH,Cl,/toluene (3.0 mL).
The diene was taken up in a syringe and added over 3.0 hours via syringe. pump. The
reaction was then allowed to stir at —45 °C for the duration of the reaction (20 hours).
Workup was done according to the optimal conditions and the reaction failed to produce
any of the desired product 213.

Condition B: The protocol described in condition A was followed exactly except
the imine was prepared from 0.146 g (0.735 mmol) of the aldehyde 211 and 0.1347 g
aminodiphenylmethane (0.735 mmol) using magnesium Sulfate as the drying agent. 2.72
equivalents of the diene were used and the reaction was allowed to stir for 96 hours and
the reaction failed to produce any of the desired product 213.

Condition C: The protocol described in condition A was followed. exactly except
the imine was prepared‘using 4A molecular sieves rather than MgSO4 as the drying agent.
1.87 equivalents of the diene were used and the reaction was allowed to stir for 96 hours I

and the reaction failed to produce any of the desired product 213.

264

Condition D: The protocol described in condition A was followed exactly except
the imine was prepared using sodium sulfate as the drying agent and the reaction was
allowed to stir for 96 hours. The reaction failed to produce any of the desired product

213.

 

 

I O
,Si/
O \
+ B(OPh)3 / VAPOL _ I
\ / r
N OMe solvent
31 —45°C
2.0 equiv. . 24 h
215 Added over 3.0 h cooling corl flask

in 3mL solvent

 

through the cooling

addition coil

Procedures for Scheme 3.31 (aza-Diels-Alder reaction of imine 215):

Condition A: The protocol for the optimal conditions was followed exactly where
0.3275 g (1.0 mmol) of imine 215 was used. The product (216) was purified via flash
column chromatography (36 cm x 2 cm, 8:5 hexanes/ethyl acetate rf 0.17). The reaction
produced 280.8 mg (70% yield) of the desired product 216 and 93% enantiomeric excess
was measured.

Condition B: The protocol for the optimal conditions was followed exactly except
the catalyst was transferred to the imine and the diene was added in CCl4/CH,C1, (2:1)
rather than a 1:1 mixture of CH,Cl,/toluene. The reaction produced 221.5 mg (56%
yield) of the desired product 216 and 91 % enantiomeric excess was measured.

Condition C: The protocol for the Optimal conditions was followed exactly

except 54 mg (10 mol%) (S)-VAPOL was used for the preparation of the catalyst. The

265

reaction produced 332.3 mg (84% yield) of the desired product 216 and 93%
enantiomeric excess was measured.

Condition D: The protocol described in condition C was followed exactly except
the catalyst was transferred to the imine and the diene was added in CH,CI, rather than
the 1:1 mixture of CH,Cl,/toluene. The reaction produced 261.1 mg (66% yield) of the
desired product 216 and 76% enantiomeric excess was measured.

Condition E: The protocol described in condition C was followed exactly except
the catalyst was transferred to the imine and the diene added in CCl4/CH,C1, (2:1) rather
than a 1:1 mixture of CH,Cl,/toluene. The reaction produced 296.7 mg (75% yield) of
the desired product 216 and 95% enantiomeric excess was measured.

Condition F: The protocol for the optimal conditions was. followed exactly except
54 mg (S)-VAPOL (10 mol%) was used and the catalyst was prepared in CC14, heating to
80°C for one hour, and then heating under high vacuum at 80°C for 30 minutes. The
reaction produced 312.5 mg (79% yield) of the desired product 216 and 96%
enantiomericexcess was measured.

Condition G: The protocol described in condition F was followed exactly except
the catalyst was transferred to the imine and the diene added in CCl4/CH,CI, (2:1) rather
than a 1:1 mixture of CH,Cl,/toluene. The reaction produced 249.2 mg (63% yield) of
the desired product 216 and 93% enantiomeric excess was measured.

Condition H: The protocol described in condition F was followed exactly except
the catalyst was transferred to the imine and the diene added in toluene rather than a 1:1

mixture of CH,Cl,/toluene. The reaction produced 106.8 mg (27% yield) (some product

266

lost during the workup) of the desired product 216 and 82% enantiomeric excess was
measured.
Spectral data for compound 216 (CmH,9NO):

The enantiomers could be separated by HPLC using a Chiralcel OD column
(w/guard column) (95:5 hexane/isopropanol, 1.0 mL/min). Retention times: 15.02 min.
(major) and 17.02 min. (minor). 'H NMR (CDCl,) 6 2.21 (s, 6H), 2.28 (s, 6H), 2.81 (dd,
2H, J=7.1, 16.5 Hz), 2.97 (s, 1H), 4.56 (t, 1H, J=9.3 Hz), 5.02 (d, 1H, J=7.7 Hz), 6.63 (s,
2H), 6.74 (s, 2H), 6.87 (s, 1H), 6.94 (s, 1H), 7.11 (d, 1H, J=7.7 Hz), 7.32 (m, 5H); ”C
NMR (CDC13) 6 21.42, 21.54, 43.64, 62.03, 68.36, 98.27, 125.36, 127.32, 127.51,
128.45, 129.09, 129.83, 130.13, 138.12, 138.50, 138.54, 138.71, 139.12, 151.84, 190.20;
IR (CDCl3) 3024.77, 2918.67, 1653.21, 1576.04, 1495.02, 1454.51, 1379.28, 1284.75,
1221.10, 1155.51, 1142.00, 910.52 cm"; mass spectrum m/z (% rel intensity) 396 M"1
(22), 395 M+ (91), 394 (12), 393 (7), 392 (6), 304 (11), 262 (6), 239 (20), 225 (14), 224
(100), 223 (24), 209 (17), 208 (6), 194 (27), 193 (9), 209 (17), 208 (6), 116 (5), 115 (9),
104 (9), 103 (13), 102 (5), 91 (6), 77 (10, 51 (5); White solid, mp. 61-65°C. Optical

rOtation was taken on a 96% ee material, [(11200 +115.7° (c 1.575, CH,CI,).

 

 

I
,Si/
0 , \ ‘
+ N B(OPh)3 / VAPOL _
/ OM, CCl4/CH,C1, (2: 1)
31 —45°C
Br
2.0 equiv. . 24 h
Added over 3.0 h cooling corl flask I
1.0 equiv. in 3mL solvent

 

through the cooling

addition coil

Procedure for Scheme 3.32 (aza-Diels-Alder reaction of imine 217):
267

 

Condition A: The protocol for the Optimal conditions was followed exactly except
54 mg (S)-VAPOL (10 mol%) was used for the preparation of the catalyst and the
catalyst was prepared by heating to 80°C rather than 55°C for one hour and then heated
to 80°C rather than 55°C for 30 minutes under high vacuum. 0.4064 g (1.0 mmol) of
imine 217 was used, and the catalyst was transferred to the imine and the diene was
added in in CC14/CH,C1, (2:1) rather than a 1:1 mixture of CH,Cl,/toluene. The product
(218) was purified via flash column chromatography (36 cm x 2 cm, 8:5 hexanes/ethyl
acetate rf 0.17). The reaction produced 346.3 mg (70% yield) of the desired product 218
and 86% enantiomeric excess was measured.

Spectral data for compound 218 (C,3H,3BrNO):

The enantiomers could be separated by HPLC using a Chiralcel OD column
(w/guard column) (80/20 hexane/isopropanol, 1.0 mL/min). Retention times: 17.82 min
(major) and 24.66 min (minor). 1H NMR (CDCI3) 6 2.21 (s, 6H), 2.27 (s, 6H), 2.76 (dd,
2H, J=7.1, 16.5 Hz), 4.52 (t, 1H, J=7.2 Hz), 4.99 (d, 1H, J=8.0 Hz), 5.25 (s, 1H), 6.60 (s,
2H), 6.73 (s, 2H), 6.88 (s, 1H), 6.93 (s, 1H), 7.08 (d, 1H, J=7.7 Hz), 7.14 (d, 2H, J=8.2
. Hz), 7.45 (d, 2H, J=8.2 Hz); ”C NMR (CDC13) 6 21.04, 21.14, 42.92, 60.88, 68.33,
98.01,_121.90, 124.92, 127.05, 128.63, 129.57, 129.87, 131.83, 137.41, 137.82, 138.14,
138.22, 138.26, 151.38, 189.42; IR (CDCI3) 3017.05, 2918.67, 1643.56, 1581.83,
1487.31, 1464.16, 1406.29,. 1377.35, 1286.68, 1219.17, 1155.51, 1140.08, 1072.56,
1010.83, 910.52, 852.65, 821.78, 733.04, 646.24 cm"; mass spectrum m/z (% rel
intensity) 475 M+ (27), 473 (27), 304 (11), 283 (8), 281 (19), 276(8), 238 (18), 225 (26),
224 (58), 223 (100), 222 (12), 221 (9), 210 (29), 209 (63), 208 (61), 207 (20), 206 (8),

195(8), 194 (38), 193 (41), 192(8), 191 (11), 183 (11), 179 (11), 176(14), 134(9), 115

268

(16), 105 (10), 104(8), 103 (17), 102 (11), 77 (27), 75 (9), 73 (18), 51 (12); Anal Calcd
for Light yellow solid. Mp 82-88°C. Optical rotation was taken on 86% ee material,

lot)”.,+104.0° (c 1.455, CH,C1,).

I O
O \
+ N B(OPh)3 / VAPOLA I

/ OMe CC14/CH2C1,

\N
O 31 c
24 n

2.0 equiv.

219 _ _
Added over 3.0 n °°°"“8 001' flask

in 3mL solvent

 

 

through the cooling
addition coil

Procedures for Scheme 3.33 (aza-Diels-Alder reaction of imine 219):

Condition A: The protocol for the optimal conditions was followed exactly except
54 mg (S)-VAPOL (10 mol%) was used for the preparation of the catalyst and the
catalyst was prepared by heating to 80°C rather than 55°C for one hour and then heated
to 80°C rather than 55°C for 30 minutes under high vacuum. 0.3335 g (1.0 mmol) of
imine 219 was used, and the catalyst was transferred to the imine and the diene was
added in in CCl,,/CH,C12 (2:1) rather than a 1:1 mixture of CH,C1,/toluene. The product
(220) was then purified via flash column chromatography (36 cm x 2 cm, 8:5
hexanes/ethyl acetate rf 0.17). The reaction produced 120.5 mg (30% yield) of the
desired product 220 and 70% enantiomeric excess was measured.

Condition B: The protocol described in condition A was followed exactly except
the catalyst was prepared in CO4 rather than CH,C1,. The reaction produced 112.4 mg
(28% yield) of the desired product 220 and 80% enantiomeric excess was measured.

Specctral Data for compound 220 (CmH35NO):

269

The enantiomers could be separated by HPLC using a Chiralcel OD column
(w/guard column) (95:5 hexane/isopropanol, 1.0 mL/min). Retention times: 9.5 min.
(major) and 12.5 min. (minor). 1H NMR (300 MHz, CDCl,) 6 1.02-1.34 (hr in, 5H),
1.64-1.98 (br m, 6H), 2.04-2.50 (br m, 12H), 2.69-2.77 (dd, 1H, J=7.5, 16.8 Hz), 3.28 (t,
1H, J=5.7 Hz), 4.77 (d, 1H, J=7.5 Hz), 5.55 (s, 1H), 6.67 (s, 2H), 6.83-6.88 (m, 4H), 6.93
(s, 1H), 6.97 (s, 1H); ”C NMR (CDC13) 6 21.41, 21.47, 26.16, 26.37, 26.47, 28.52, 29.82,
36.41, 40.81, 62.02, 70.03, 96.59, 125.09, 127.52, 129.82, 130.15, 138.57, 139.97,
151.11, 191.17; IR (CDC13) 3013.19, 2926.39, 2853.08, 1635.84, 1577.97, 1450.65,
1224.95, 1157.44, 1037.83, 910.52, 852.65, 731.11cm"; mass spectrum m/z (% rel
intensity) 402 M+1 (27), 4011M+ (84), 281 (22), 238 (19), 225 (32), 224 (65), 223 (100),
210 (44), 209 (73), 208 (61), 207 (19), 194 (62), 193 (14), 192 (13’), 179 (13), 178 (‘15),
133 (12), 105 (10), 103 (12), 91 (17), 77 (20), 73(16), 67 (10), 55 (11); white solid, mp
57—59°C. Optical rotation was taken on 70% ee material, [(11201) ——65.8° (c 1.415,

CH,C1,).

O
IIN B(OPh)3 / VAPOL _ I
OMe CH2C12/toluene (1 : 1) N
OMe 2.0 equiv. 24 II

Added over 3.0 h 0 O

- in 3mL solvent MeO 2228 OMe

 

1.0 equiv

Procedure for Scheme 3.34 entry 1 (aza-Diels-Alder reaction of imine 221a):
Condition A: The protocol for the optimal conditions was followed exactly except
54 mg (S)-VAPOL (10 mol%) was used for the preparation of the catalyst. 0.3314 g (1.0

270

mmol) of imine 221a was used and the product (2223) was purified via flash column
chromatography (36 cm x 2 cm, 2:1 hexanes/ethyl acetate, rf 0.05). The reaction
produced 371.5 mg (93% yield) of the desired product 222a and 84% enantiomeric
excess was measured.
Spectral Data for compound 222a (C,6H,5NO3):

The enantiomers could be separated by HPLCusing a Chiralpak AD column
(95:5 hexane/isopropanol, 2.0 mL/min). Retention times: 18.8 min. (major) and 23.7
min. (minor). 'H NMR (300 MHz, CDCl,) 6 269-2.74 (dd, 1H, J=8.5, 16Hz), 2.82—2.86
(dd, 1H, J=7, 16.5 Hz), 3.76 (s, 3H), 3.82 (s, 3H), 4.52 (t, 1H, J=7.8 Hz), 4.99 (d, 1H,
J=7.5 Hz), 5.32 (s, 1H), 6.81-6.83 (m, 2H), 6.90-6.93 (m, 4H), 7.03-7.08 (m, 3H), 7.27-
7.29 (m, 2H), 7.33-7.38 (m, 3H); ”C-NMR (CDCI3) 6 43.97, 55.54, 55.58, 62.27, 67.22,
98.83, 114.47, 114.62, 127.48, 128.67, 128.91, 129.36, 130.67, 130.84, 130.96, 139.44,
151.59, 159.64, 159.71, 190.49; IR (neat) 731.118, 908.59m, 1032.058, 1136.228,
1174.808, 1250.03vs, 1304.058, 1454.51m, 1512.36vs, 1576.91vs, 1608.84vs, 1643.56vs, ,
2842.33w, 2898.23w, 2935.45w, 2951.33w, 3003.12w, 3025.67w, 3086.64w cm“; Mass
spectrum m/z (% rel intensity) 399 M+ (5),355 (7), 342 (6), 327 (5), 289 (14), 281 (18),
268 (7), 252 (8), 228 (6), 210 (6), 209 (17), 208 (100), 207 (27), 194 (9), 191 (14), 178 --
(8), 147 (6), 133 (12), 115 (5), 73 (20); White solid, mp 58-74 °C. Optical rotation was

taken on 93% ee material, [(1)200 +63.7° (c 1.78, CH,C1,).

271

OMe

o
I
o’ \
+ B(OPh)3 / VAPOLAV I

\ MOMe CH2C12/toluene (1:1) N
Br 221b OMe 2.0 equiv. 24 h Br
10 equiv Added over 3.0 h

in 3mL solvent MeO 222b OMe

 

Procedure for Scheme 3.34 entry 2 (aza-Diels-Alder reaction of imine 221b):

Condition A: The protocol for _the optimal conditions was followed exactly except
54 mg (S)-VAPOL (10 mol%) was used for the preparation of the catalyst. 0.4103 g (1.0
mmol) of imine 221b was used and the product (222b) was purified via flash column
chromatography (36 cm x 2 cm, 2:1 hexanes/ethyl acetate, rf 0.05). The reaction
produced 478.3 mg (100% yield) of the desired product 222b and 90% enantiomeric
excess was measured.

Spectral Data for compound 222b (C,6H,4BrNO3):

The enantiomers could be separated by HPLC analysis on Chiralpak AD (95:5
hexane/isopropano], 2.0 mL/min). Retention times: 20.2 min. (major) and 30.8 min.
(minor). 1H NMR (500 MHz, CDCl,) 6 2.61-2.66 (dd, 1H, J=8.5, 16.5 Hz), 2.81-2.86
(dd, 1H, J=7.0, 16.5 Hz), 3.76 (s, 3H), 3.82‘(s, 3H), 4.86 (t, 1H, J=7.5 Hz), 4.99 (d, 1H,
J=8.0 Hz), 5.28 (s, 1H), 6.81-6.84 (m, 2H), 6.90 (s, 2H),6.91 (s, 2H), 7.03-7.06 (m, 3H),
7.15 (br s, 1H), 7.16 (br s, 1H), 7.48 (br 8, 1H), 7.50 (br s, 1H); ”C-NMR (CDC13) 6
43.72, 55.56, 55.59, 61.59, 67.54, 99.05, 114.56, 114.69, 122.54, 128.86, 129.15, 130.38,
130.63, 130.91, 132.54, 138.54, 151.40, 159.72, 159.80, 190.04; IRi(neat); 731.11m,
821.78 m, 908.59w, 1033.988, 1136.228, 1174.808, 1250.03vs, 1304.05 8, 1344.56w,

1462.23m, 1512.36 vs, 1579.9vs, 1610.778, 1641.63vs, 2816.33w, 2891.56w, 2920.10w,

272

2963.23w, 3002.72w, 3017.39w cm"; mass spectrum m/z (% rel intensity) 479 M+2 (4),
477 M+ (4), 452 (11), 437 (5), 280 (5), 271 (19), 253 (28), 252 (49), 251 (100), 227
(100), 212 (18), 183 (17), 169 (12), 141 (16), 115 (10), 77 (8); White solid, mp 58-74 °C.

Optical rotation was taken on 93% ee material, [002°D +63.7° (c 1.78, CH,C1,).

OMe

I
O’ \
+ B(OPh)3 / VAPOL _
MOMe CHZCIZ/toluene (1: l)

\ .
r)“ o ..
OZN OMe 2.0 equiv. 24 h OZN

221c
. Added over 3.0 h
1.0 equw
in 3mL solvent MeO 222c ' OMe

 

 

Procedure for Scheme 3.34 entry 1 (aza-Diels-Alder reaction of imine 221c):

Condition A: The protocol for the optimal conditions was followed exactly except
54 mg (S)-VAPOL (10 mol%) was used for the preparation of the catalyst. 0.3764 g (1.0
mmol) of imine 221c was used and the product (222c) was purified via flash column
chromatography (36 cm x 2 cm, 2.5:] hexanes/ethyl acetate, rf 0.13). The reaction
produced 302.2 mg (68% yield) of the desired product 222c and 81% enantiomeric
excess was measured. I
Spectral Data for compound 222c (C,6H,4N,OS):

The enantiomers could be separated by HPLC using a Chiralpak AD column
(75:25 hexane/isopropanol, 1.0 mL/min). Retention times: 11.79 min. (major) and 23.07
min. (minor). 1H NMR (CDC13) 6 2.56 (dd, 1H, J=16.5, 6.3 Hz), 2.93 (dd, 1H, J=16.5,
7.5 Hz), 3.72 (s, 3H), 3.76 (s, 3H), 4.65 (t, 1H, J=6.9 Hz), 4.97 (d, 1H, J=7.8 Hz), 5.27 (s,

1H), 6.79-7.08 (br m, 9H), 7.43 (d, 2H, J=9 Hz), 8.17 (d, 2H, J=8.4 Hz); ”C NMR

273

(CDC13) 6 13.83, 42.57, 54.96, 60.56, 67.84, 98.67, 114.07, 114.16, 124.00, 127.64,
128.20, 129.26, 129.88, 130.26, 146.25, 147.41, 150.60, 159.22, 159.31, 188.56; IR
(pure) 1032m, 1136m, 1175m, 12528, 13488, 1462w, 1512vs, 1578vs, 16098, 16428,
2850w, 2932w, 2956w, 3002w, 3057w cm"; Mass spectrum m/z (% rel intensity) 444 M+
(9), 414 (5), 294 (32), 266 (17), 242 (33), 228 (64), 227 (100), 212 (32), 169 (59), 141
(60), 115 (39); Light yellow solid, mp 85-90 °C. The optical rotation was taken on an

81% ee material, [(11201) +69.2° (c 1.753, CH,C1,).

Si
+ B(OPh)3 / VAPOL _ I

' \ MOMe CH,Clzltoluene (1:1) N
on” 0 ..
2“ O o
Added over 3.0 h

1.0 equiv
in 3mL solvent MeO 222d OMe

 

Procedure for Scheme 3.34 entry 4 (aza-Diels-Alder reaction of imine 221d):
Condition A: The protocol for the optimal conditions was followed exactly except
54 mg (S)—VAPOL (10 mOl%) was used for the preparation of the catalyst. 0.3375 g (1.0
mmol) of imine 221d was used and the product (222d) was purified via flash column
chromatography (36 cm x 2 cm, 2.521 hexanes/ethyl acetate, rf 0.13). The reaction
produced 267.6 mg (66% yield) of the desired product 222d and 56% enantiomeric
excess was measured.
Spectral Data for compound 222d (CMH31NO3):
The enantiomers could be separated by HPLC using a Chiralpak AD column

(95:5 hexane/isopropanol, 2.0 mL/min). Retention times: 12.5 min. (major) and 17.2

274

min. (minor). 1H-NMR (500 MHz, CDCl,) 6 1029-126 (br m, 6H), 1.66 (br s, 2H, 1.75-
1.84 (br m, 3H), 1.97 (br s, 1H), 2.36 (d, 1H, J=17.0 Hz), 2.65-2.70 (dd, 1H, J=8.0 Hz,
J=16.5 Hz), 3.27 (br t, 1H, J=5.5 Hz), 3.77 (s, 3H), 3.81 (s, 3H), 4.78 (d, 1H, J=7.5 Hz),
5.60 (s, 1H), 6.84-6.86 (br m, 2H), 6.91 (br s, 1H), 6.92 (br s, 1H), 6.97 (br 8, 1H), 6.99
(br s, 1H), 7.19 (br s, 1H), 7.21 (br s, 1H); clean ”C NMR was not obtained; IR (neat)
1033.98m, 1142.00m, 1174.808, 1251.988, 1304.05m, 1462.23m, 1512.388, 1576.04vs,
1637.77vs, 2851.15m, 2926.39m, 3001.28w, 3025.69w, 3067.41w cm“; mass spectrum
m/z (% rel intensity) 406 M+1 (26), 405 M+ (100), 375 (9), 323 (5) 299 (10), 295 (9),
281 (7), 252 (19), 228 (31), 208 (10), 207 (9), 170 (4), 135 (6), 133 (5), 73 (7); White
solid, mp 121-125°C. Optical rotation was taken on 56% ee material, [camD —71.9° (c

1.615, CH,C1,).

 

+ B(OPh)3 I VAPOL _ N
MOMe solvent
3‘ WW O O
OMe 2.0 equiv. 24 h
MeO OMe

Added over 3.0 h cooling corl flask

in 3mL solvent

   

223
1.0 equiv

through the cooling 224

addition coil
Procedures for Scheme 3.35 (aza-Diels-Alder reaction of imine 223):
Condition A: The protocol for the optimal conditions was followed exactly where
, 0.5558 g (1.0 mmol) of imine 223 was used. The product (224) was purified via flaSh
columnchromatography (36 cm x 2 cm, 2:1 hexanes/ethyl acetate rf 0.34). The reaction

prOduced 361.9 mg (58% yield) of the desired product 224 and 81% enantiomeric excess

was measured.

275

Condition B: The protocol described in condition A was followed exactly except
54 mg (10 mol%) (S)-VAPOL was used for the preparation of the catalyst. The reaction
produced 218.4 mg (35% yield) of the desired prOduct 224 and 40% enantiomeric excess
was measured.

Condition C: The protocol described in condition A was followed exactly except
0.09 g B(OPh)3 (30 mol%) and 54 mg (S)-VAPOL (10 mol%) was used for the
preparation of the catalyst. The catalyst was transferred to the imine and the diene was
added in CCl4 rather than a 1:1 mixture CH,Cl,/toluene. The reaction was run at room
temperature and produced 56.2 mg (9% yield) of the desired product 224 and 29%
enantiomeric excess was measured.

Spectral data for compound 224 (C4,H5,NO3):

The enantiomers could be separated by HPLC using a Pirckle Covalent (R,R)
Whelk-O 1 column (75:25 hexane/isopropanol, 1.0 mL/min). Retention times: 8.74 min.
(minor) and 11.50 min. (major). lH-NMR (CDCI3) 6 1.28-1.38 (br m, 36 H), 2.71 (dd,
1H, J =16.5, 7.5 Hz), 2.9 (dd, 1H, J=16.5, 7.2 Hz), 3.62-3.67 (m, 6H), 4.54 (t, 1H, J=7.5
Hz), 5.03 (d, 1H, J=7.8 Hz), 5.27 (s, 1H), 6.79 (s, 2H), 6.97 (s, 2H), 7.06 (d, 1H, J=7.8
Hz), 7.20-7.36 (br m, 5H); ”C-NMR (CDC13) 6 29.73, 31.94, 32.06, 32.12, 35.82, 43.56,
61.88, 64.27, 64.37, 68.47, 98.27, 125.86, 127.31, 127.73, 128.38, 129.05, 132.24,
132.43, 143.93, 144.12, 151.60, 190.26; mass spectrum m/z (% relative intensity): 624
M+1 (13), 623 M+ (32), 568 (12), 567 (27), 454 (10), 453 (56), 452 (100), 451 (11), 438
( 19), 380 (8), 282 (8), 234 (4), 208 (35), 207 (10), 194 (6), 191 (7), 147 (5), 103 (6), 73

(9), 57 (13); IR (neat): cm"; 1013.72m, 1116.44m, 1225.928, 1414.008, 1450m, 1576.538,

276

1579.878, 1635.85m, 2962.56vs, 2892.93m, 3012.45m, 3087.43w; Light yellow wax, the

optical rotation was taken on 29% ee material [ot]2°D +3.3° (c 1.66,.CDC13).

 

OMe O
I /
Si .
O’ \
+ B(OPh)-J. / VAPOL N N
MOM, CH,Clzltoluene (1:1)

Br

” Q 0
0M6 2.0 equiv. 24 h

Added over 3.0 h °°°'"‘g °°" flask Mco OMe

   

1.0 equiv
in 3mL solvent

through the cooling

addition coil

Procedure for Scheme 3.36 entry 1 (aza-Diels-Alder reaction of imine 225a):

Condition A: The protocol for the optimal conditions was folloWed exactly except
54 mg of (S)-VAPOL (10 mol%) was used for the preparation of the catalyst. 0.6347 ‘ g
(1.0 mmol) of imine 225a was used for the reaction and the product (226a) was purified
via flash column chromatography (36 cm x 2 cm, 2:1 hexanes/ethyl acetate r, 0.27). The
reaction produced 105.4 mg (15% yield) of the desired product 226a and 29%
enantiomeric excess was measured.

Spectral Data for compound 226a (C4,H56BrNO3):

The enantiomers could be separated by HPLC using a Pirckle Covalent (R,R)
Whelk-O 1 column (75:25 hexane/isopropanol, 1.0 mL/min). Retention times: 8.20 min.
(minor) and 11.20 min. (major). 1H-NMR (500 MHz, CDC],) 6 1.34 (br s, 18H), 1.41 (br
8, 18H), 265-2.70 (dd, 1H, J=7.0, 16.5 Hz), 2.93-2.98 (dd, 1H, J=7.0, 16.5 Hz), 3.696 (s,
3H), 3.703 (s, 3H), 4.57 (t, 1H, J=7.0 Hz), 5.07 (d, 1H, J=8.0 Hz), 5.30 (s, 1H), 6.82 (s,
2H), 7.01 (s, 2H), 7.09 (d, 1H, J=8.0 Hz), 7.22 (d, 2H, J=8.5 Hz), 7.51 (d, 2H, J=8.5 Hz);

l3C-NMR (CDCI3) 6 32.16, 32.32, 36.04, 36.08, 43.52, 61.33, 64.50, 64.57, 69.19, 98.68,
277.

122.41, 126.11, 127.75, 129.13, 132.29, 132.37, 132.40, 138.79, 144.28, 144.44, 151.67,
159.48, 159.63, 189.96; IR (neat) 1116.938, 1143.93m, 1224.95vs, 1265.46m, 1361.92m,
1414.008, 1448.72m, 1487.31m, 1585.69vs, 1643.568, 2870.44m, 2908.238, 2964.97vs,
3009.76m cm"; Mass spectrum m/z (% rel intensity) 703 M+2 (14) 701 M+ (12), 647
(29), 646 (76), 644 (76), 643 (12), 518 (17), 497 (22), 495 (72), 494 (100), 452 (47), 451
(100), 450 (29), 421 (15), 379 (10), 218 (6), 183 (11), 57 (21); Light yellow solid, mp

166-174°C. Optical rotation was taken on 29% ee material, [cumD +3.6° (c 1.6, CH,C1,).

 

+ B(OPhH/VAPOL_ N
MoMe CH,C1,/toluene(1:l) O2N .

31 —45°C
OMe 2.0 equiv. 24 b O O
cooling coil flask
Added over 3.0 h McO OMe
in 3mL solvent

through the cooling

   

02N
1.0 equiv
2261)

addition coil

Procedure for Scheme 3.36 entry 2 (aza-Diels-Alder reaction of imine 225b):
Condition A: The protocol for the optimal conditions was followed exactly except
54 mg of (S)-VAPOL (10 mol%) was used for the preparation of the catalyst. 0.6008 g
(1.0 mmol) of imine 225b was used for the reaction and a complex mixture of
compounds as observed by TLC analysis was produced. None of the desired product

226b was observed by 1H—NMR.

278

 

+ B(OPh)3 / VAPOL y N
MOMe CHzClzltoluene (1:1)

0M6 2.0 equiv. 24 h

Added over 3.0 h °°°""g co" flask M60 OMe

  
 

\
Om
225c

1.0 equiv
in 3mL solvent

through the cooling 226C

addition coil

Procedure for Scheme 3.36 entry 3 (aza-Diels-Alder reaction of imine 225C):

Condition A: The protocol for the optimal conditions was followed exactly except
54 mg of (S)-VAPOL was used for the preparation of the catalyst. 0.5619 g (1.0 mmol)
of imine 225C was used and the product (2266) was purified via flash column
chromatography (36 cm x 2 cm, 2:1 hexanes/ethyl acetate rf 0.39). The reaction produced
132.3 mg (21% yield) of the desired product 226C and 30% enantiomeric excess was
measured.

Spectral Data for compound 226C (C42H63NO3):

The enantiomers could be separated by HPLC using a Pirckle Covalent (R,R)
Whelk-O 1 column (75 :25 hexane/isopropanol, 1.0 mL/min). Retention times: 5.80 min.
(minor) and 7.30 min. (major). 1H—NMR (500 MHz, CDCl3) 6 1.17-1.27 (br m, 5H), 1.33
(br s, 18H), 1.41 (br s, 18H), 1.69-1.71 (m, 2H), 1.81-1.85 (m, 2H), 1.92-1.94 (m, 1H),
2.02-2.04 (br s, 1H), 2.43 (d, 1H, J=16.5 Hz), 2.73-2.79 (dd, 1H, J=9.0, 17.0 Hz), 3.40
(br t, 1H, J=7.5 Hz), 3.70 (s, 3H), 3.71 (s, 3H), 4.86 (d, 1H, J=7.5 Hz), 5.57 (s, 1H), 6.86
(s, 2H), 6.88 (d, 1H, J=7.5 Hz), 7.16 (s, 2H); l3C-NMR (CDC13) 6 26.40, 26.73, 26.78,
28.76, 29.93, 32.14, 32.32, 36.01, 36.08, 36.67, 41.92, 62.15, 64.47, 64.55, 70.31, 96.73,

125.67, 127.94, 132.92, 133.92, 144.14, 144.34, 151.14, 159.38, 159.54, 191.25; IR

279

(neat) lll6.m 1147.79w, 1224.95vs, 1265.46m, 1361.92w, 1392.78m, 1414.008,
1448.72m, 1583.76 vs, 1637.778, 2856.43m, 2881.29w, 2924.69vs, 2963.04vs, 3006.18w
cm“; Mass spectrum m/z (% rel intensity) 631 M+l (13), 630 M+ (100), 629 (30), 575
(13), 574 (39), 573 (83), 453 (19), 452 (25), 411 (5); White solid, mp 93-95°C. Optical

rotation was taken on 30% ee material, [0.]200 —33.3° (c 1.505, CH2C12).

280

7 .3 Experimental Procedures and Characterizations Data for Chapter Four

 

Preparation of imine 61a:

To a flame dried round bottom flask equipped with a magnetic stir bar was added
benzaldehyde (5.23 g, 49.2 mmol) followed by MgSO4 (7.5 g) and CHZCI2 (100 mL). To
this solution was then added benzylamine (5.20 g, 49.2 mmol). The resulting solution
. was allowed to stir at roomltemperature for 5.5 hours. At the end of the reaction time, the
solution was filtered to remove the MgSO4, and the solvent was removed under reduced
pressure. The resulting imine was an oil which was purified by bulb-to-bulb distillation
(bp 133 °C, 3 mm Hg) to give 6.24 g of 61a (67% yield). The spectral data matched
those reported in the literature.‘ 1H-NMR (CDC13) 6 4.86 (s, 2H), 7.28-7.47 (m, 8H),
7.81-7.84 (m, 2H), 8.41 (s, 1H); l3C-NMR (CDC13) 5 64.76, 126.70, 127.70, 128.00,

128.22, 128.32, 130.48, 135.93, 139.07, 161.65.

281

 

B(oph)3
(R)-BINOL = (R)-BINOL-B catalyst
CHZCIZ, n, 4A MS

 

 

 

O

l/
i i O/S'\ (R) BlNOLBcatalyst NI
\ +
Ph N Ph M

OMe CHZCIZ, 78°, 5h

 

= 31
61a R H 12 . R/k Ph
= . e UIV.
150 R Ph C1 62 R=H
151 R=Ph

Procedures for Scheme 4.2 (aza-Diels-Alder reactions following Yamamoto’s
Conditions"):

Entry 1: To a flame dried round bottom flask equipped with a magnetic stir bar
was added 4 A molecular sieves (1.0 g), B(OPh)3 (101 mg, 0.35 mmol) and (R)-BINOL
(100 mg, 0.35 mmol). To this was added CHZCI2 (10 mL) and the reaction mixture was
allowed to stir for one hour at room temperature. The reaction was then cooled to 0 °C at
which time the imine 61a (68 mg, 0.35 mmol) was added in CHZCl2 (1.0 mL). This was
allowed to stir for 5 minutes and then the reaction mixture was cooled to - 78 °C and the
diene (31) (84 uL, 0.42 mmol) was added in CHZCI2 (1.0 mL) dropwise over about 3
minutes. The reaction was allowed to stir at —78 °C for an additional 5 hours. Suction
filtration was then used to remove the molecular sieves. The resulting mixture was then
washed with water (30 mL) and saturated sodium bicarbonate (50 mL). The organic
layers were combined, dried over magnesium sulfate, filtered, and the solvent was
removed under “reduced pressure. The product was purified via flash column
chromatography (36 cm x 2 cm), Rf = 0.06 (hexanes/ethyl acetate 2:1), to give 62.3 mg
(68% yield) of the desired product 62. Spectral data was collected and was found to be

identical to that in the literature.‘9 The enantiomers could be separated by HPLC using a

282

Chiralcel OJ-H column (80:20 hexane/isopropanol, 1mL/min). Retention times: 26.15
min. (major) and 31.81 min. (minor). The product 62 obtained from the reaction was
determined to be 85.5% ee.

Entry 2: The protocol as described for entry 1 was followed exactly except 10 mg
(R)-BlNOL (0.035 mmol, 10 mol%) was used for the preparation of the catalyst. This
reaction produced 38 mg of 62 (41% yield) and 40% ee was measured.

Entry 3: The protocol as described for entry 1 was followed exactly except 10.1
mg B(OPh)3 (0.035 mmol, 10 mol%) and 10 mg (R)-BlNOL (0.035 mmol, 10 mol%) was
used for the preparation of the catalyst. This reaction produced less than 10 mg of 62 (<5
% yield) and the ee was not measured.

Entry 4: The protocol as described for entry 1 was followed exactly except 94.9
mg (0.35 mmol) of the benzhydryl imine 150 was usedrather than imine 61a. None of
the desired product 151 was observed.

Entry 5: The protocol as described for entry 2 was followed exactly except 94.9
mg (0.35 mmol) of the benzhydryl imine 150 was used rather than imine 61a. None of

the desired product 151 was observed.

283

 

B(OPh)3 0.1 mmHg

(S)-VAPOL AV > (S)-VAPOL-B catalyst
CH2C12,55°,1h 55°,0.Sh

I / ) O

 

 

 

 

,Si - A L-
/}l{\\ i + .0 \ (S) V PO B catalyst: . I
Ph N Ph MOMe CH2C12/toluene (1 :1) 1311‘" N
_ -45°, 24 h
61a R — H 31 R/k Ph
150 R = Ph (2.0 equiv.) 62 R = H
151 R = Ph

Procedures for Scheme 4.3 (aza-Diels-Alder reactions using the optimal conditions
from chapter 2):

Entry 1: To a flame dried, argon purged single-necked flask that had its 14/20
joint replaced with a threaded Teflon high-vacuum T-shaped stop-cock equipped with a
stir bar was added B(OPh)3 (0.3 125 g, 1.0 mmol) and (S)-VAPOL (54 mg, 0.1 mmol). To
this was added CHzCl2 (2 mL) and then the flask was sealed with the stopcock and heated
to 55 °C for one hour. After one hour, the solvent was removed via high vacuum and left
under high vacuum at 55 °C for 0.5 hours yielding the (S')-VAPOL-boron catalyst. After
cooling the stopcock was removed and replaced with a rubber septum. The catalyst was
dissolved by the injection via syringe of 2.0'mL of a 1:1 mixture of toluene and CH2C12
(in two portions). The catalyst solution was transferred by syringe to a solution of the
imine prepared as immediately below.

To a flame dried, argon purged homemade flask with a cold addition coil (see
Figure 2.4) equipped with a stir bar was added the imine (0.195 mg, 1.0 mmol). The
flask was topped with two rubber septa and the (S)-VAPOL-boron catalyst was added in
two 1.0 mL portions of toluene/CH2C12 (1:1) directly to the bottom of the flask by a

syringe equipped with a long needle. This was allowed to stir for 5-10 minutes at room

284

temperature and then cooled to —45 °C. Meanwhile, in a separate flame dried 5 or 10 mL
round bottom flask purged with argon was added Danishefsky’s diene (31) (0.38 mL, 2.0
mmol) and toluene/CHZCI2 (1:1) (3.0 mL). The diene was taken up in a syringe and
added over 3.0 hours via syringe pump through the cold addition coil. The reaction was
then allowed to stir at —45 °C for the duration of the reaction (for reaction times see Table
3). After completion of the reaction, saturated sodium bicarbonate (~20 mL) was added
to the reaction flask at —45 °C. This was then transferred to a separatory funnel and
diluted with distilled water (25 mL) and extracted with three or four 3040 mL portions
of CHzClz. TLC analysis of the organic layer after extraction showed a small amount of
two compounds in addition to the dihydropyridinone. The crude 1H NMR confirmed that
only a small amount of other products were present. Isolation gave small amounts of
materials that had very complicated 1H NMR spectra with broad peaks and assignment of
structure was not made. These compounds were not observable by TLC after treatment
of the reaction mixture with 1N HCl diluted with THF. Therefore, the combined organic
layers were placed in a 250 mL round bottom flask and the solvent was then removed via
rotary evaporation. The flask was then equipped with a stir bar and cooled in an ice bath.
To this was then added a previously cooled (0 °C) 20:1 mixture of THF and 1N HCl (50
mL) at which time the flask was removed from the ice bath and allowed to stir
(monitored by TLC) until the undesired spots close to the desired product disappeared
(usually less than one hour). This was then transferred to a separatory funnel containing
water (75-100 mL) followed by extraction of the crude product with four 50 mL portions
of CHzClz. The combined organic layers were then dried with magnesium sulfate,

filtered, and solvent was removed via rotary evaporation. The product 62 was then

285

purified via flash column chromatography (36 cm x 2 cm) , Rr = 0.06 (hexanes/ethyl
acetate 2: l) yielding 131.7 mg of 62 (50% yield). Spectral data was collected and was
found to be identical to that in the literature.19 The enantiomers could be separated by
HPLC analysis on Chiralcel OJ-H (80:20 hexane/isopropanol, ImUmin), retention times:
26.15 min. (major) and 31.81 min. (minor). The product 62 obtained from the reaction
was determined to be 36% ee.
. Entry 2: The protocol as described for entry 1 was followed exactly except 0.271
g (1.0 mmol) of the benzhydryl imine 150 was used. The reaction produced 319.3 mg
(94% yield) of the desired product 150 and 90% enantiomeric excess was measured. .
Entry 3: The protocol as described for entry 1 was followed exactly except 0.271
g (1.0 mmol) of the benzhydryl imine 150 was used and 29.1 mg of (R)-BINOL (10
mol%) was used for the preparation of the catalyst. The reaction produced 88.2 mg (26%

yield) of the desired product 151 and 23% enantiomeric excess was measured.

 

O
|./
H Ph 0/5'\ B(OPh)3 (1.0 equiv.) l
/k\ k + M ;
Ph N Ph OMe CHZClz/toluene (1:1) ph N
150 31 45°, 24 h
. Ph Ph
. (2.0 equw.) 151

Procedures for Scheme 4.4 (aza-Diels-Alder reactions of imine 150 using B(OPh)3):
Entry I: To'a flame dried, argon purged single-necked flask that had. its 14/20

joint replaced with a threaded Teflon high-vacuum T-shaped stop-cock equipped with a'

stir bar was added B(OPh)3 (0.3125g, 1.0 mmol). To this was added CHZCl2 (2 mL) and

then the flask was sealed with the stopcock and heated to 55 °C for one hour. After one

286

hour, the solvent was removed via high vacuum and left under high vacuum at 55 °C for
0.5 hours. After cooling the stopcock was removed and replaced with a rubber septum.
The catalyst was dissolved by the injection via syringe of 2.0 mL of a 1:1 mixture of
toluene and CHzCl2 (in two portions). The catalyst solution was transferred by syringe to
a solution of the imine prepared as immediately below.

To a flame dried, argon purged homemade flask with a cold addition coil (see
Figure 2.4) equipped with a stir bar was added the imine (0.195 mg, 1.0 mmol). The
flask was topped with two rubber septa and the B(OPh)3 was added in two 1.0 mL
portions of t01uene/CH2C12 (1: 1) directly to the bottom of the flask by a syringe equipped
with a long needle. This was allowed to stir for 5-10 minutes at room temperature and
then cooled to —45 °C. Meanwhile, in a separate flame dried 5 or 10 mL round bottom
flask purged with argon was added Danishefsky’s diene (31) (0.38 mL, 2.0 mmol) and
toluene/CH2C12 (1:1) (3.0 mL). The diene was taken up in a syringe and added over 3.0
hours via syringe pump through the cold addition coil. The reaction was then allowed to
stir at —45 °C for the duration of the reaction (for reaction times see Table 3). After
completion of the reaction, saturated sodium bicarbonate (~20 mL) was added to the
reaction flask at —45 °C. This was then transferred to a separatory funnel and diluted
with distilled water (25 mL) and extracted with three or four 3040 mL portions bf
CHZCIZ. TLC analysis of the organic layer after extraction showed a small amount of two
compounds in addition to the dihydropyridinone. The crude lH—NMR confirmed that
only a small amount of other products were present. Isolation gave small amounts of
materials that had very complicated 1H NMR spectra with broad peaks and assignment of

structure was not made. These compounds were not observable by TLC after treatment

287

of the reaction mixture with 1N HCl diluted with THF. Therefore, the combined organic
layers were placed in a 250 mL round bottom flask and the solvent .was then removed via
rotary evaporation. The flask was then equipped with a stir bar and cooled in an ice bath.
To this was then added a previously cooled (0 °C) 20:1 mixture of THF and 1N HCl (50
mL) at which time the flask was removed from the ice bath and allowed to stir
(monitored by TLC) until the undesired spots close to the desired product disappeared
(usually less than one hour). This was then transferred to a separatory funnel containing
water (75—100 mL) followed by extraction of the crude product with four 50 mL portions
of CHzClz. The combined organic layers were then dried with magnesium sulfate,
filtered, and solvent was removed via rotary evaporation. The product 151 was then
purified via flash column chromatography (36 cm x 2 cm) to give 88 mg (26% yield) of
the desired product 151.

Entry 2: The protocol as described in entry 1 was followed exactly, except
B(OPh)3 was purchased from 'Aldrich and used directly without taking it through the
catalyst preparation cycle. This reaction produced 146 mg (46% yield) of the desired,
product 151.

Entry 3: The protocol as described in entry 2 was followed exactly, except the
reaction was done on a 0.369 mmol (of the imine 150) scale and the B(OPh)3 was
purified by distillation and stored in a glove box before use. This reaction produced 30.9

mg (25 % yield) of the desired product 151.

288

NMR Titration Experiment with (R)'-VAPOL-boron Catalyst and Imine 150:

To a flame dried, argon purged single—necked flask that had its 14/20 joint
replaced with a threaded Teflon high—vacuum T-shaped stop-cock equipped with a stir
bar was added (R)-VAPOL (0.0.957 g, 1.77 mmol) and phenol (0.333g, 3.54 mmol). To
this was added CHZCI2 (15 mL) followed by BH3-SMe2 (1.77 mL, 3.54 mmol, 2 M
solution in toluene) and water (0.031 mL, 1.77 mmol). The flask was sealed with the
stopcock and heated to 75 °C for one hour. After one hour, the solvent was removed via
high vacuum and heated to 100°C and left under high vacuum for 0.5 hours yielding the
(R)—VAPOL-boron catalyst. After cooling the stopcock was removed and replaced with a
rubber septum. The catalyst was dissolved by the injection via syringe of 5 mL of a
CDCl3 (stored over 4A MS.) to make a 0.35 M solution. Meanwhile, 50 mg (0.184
mmol) of imine 150 was added to 10 separate flame dried argon purged 2 mL volumetric
flasks. To each volumetric flask was then transferred via syringe a different amount of
the (R)-VAPOL-boron catalyst solution (see table below). After the addition of the
catalyst, the solutions were diluted to the 2 mL mark on the volumetric flask to ensure the
concentration of the imine was maintained. One separate flask was prepared using only
the imine with no catalyst. This was used as the reference sample. Approximately 0.8
mL of each volumetric flask was then transferred via syringe to a clean and dry NMR
tube and capped immediately to ensure no air was introduced to the system. The ‘H
NMR’s were then taken for each of the different catalyst/imine ratios. Recorded in the
table are the exact concentrations of imine 150 and (R)-VAPOL-boron catalyst, and the
observed chemical shifts-for the complexed species for the benzhydryl proton (H') and

the imine sp2 C-H (H2).

289

 

[imine 150] [(R)-VAPOL-boron catalyst] Obs Chem Shift Obs Chem Shift

 

NMR (M) (M) (H'. ppm) (H2, ppm)

1 0.092 0.0046 under aromatic under aromatic
2 0.092 0.00575 6.142 under aromatic
3 0.092 0.00775 6.035 8.163

4 0.092 0.0115 5.926 8.255

5 0.092 0.023 5.852 8.317

6 0.092 0.03075 5.791 8.37

7 0.092 0.046 5.735 8.405

8 0.092 0.069 5.713 8.428

9 0.092 0.092 5.7 8.44

10' 0.092 0.1505 - 5.694 _84447

 

NMR Titration Experiment with (S)-VAPOL-boron Catalyst and the Product 151.
To a flame dried, argon purged single—necked flask that had its 14/20 joint
replaced with a threaded Teflon high-vacuum T-shaped stop-cock equipped with a stir
bar was added (S)-VAPOL (0.0.957 g, 1.77 mmol) and phenol (0.333g, 3.54 mmol). To
this was added CH2C12 (15 mL) followed by BHz,-SMe2 (1.77 mL, 3.54 mmol, 2 M
solution in toluene) and water (0.031 mL, 1.77 mmol). The flask was sealed with the
stopcock and heated to 75 ,°C for one hour. After one hour, the solvent was removed via
high vacuum and heated to 100°C and left under high vacuum for 0.5 hours yielding (S)-
VAPOL-boron catalyst. After cooling the stopcock was removed and replaced with a
rubber septum. The catalyst was dissolved by the injection via syringe of 5 mL of a
CDCl3 (stored over 4A MS.) to make a 0.35 M solution. Meanwhile, 62.5 mg (0.184
mmol) of product 151 was added to 10 separate flame dried argon purged 2 mL
volumetric flasks. To each volumetric flask was then-transferred via syringe a different
amount of the (S)-VAPOL-boron catalyst solution (see table below). After the addition

of the catalyst, the solutions were diluted to the 2 mL mark on the volumetric flask to

290

ensure the concentration of the imine was maintained. One separate flask was prepared
using only the imine with no catalyst. This was u8ed as the reference sample.
Approximately 0.8 mL of each volumetric flask was then transferred via syringe to a
clean and dry N MR tube and capped immediately to ensure no air was introduced to the
system. The lH-NMR’s were then taken foreach of the different catalyst/imine ratios.
Recorded in the table are the exact concentrations of product 151 and (S)-VAPOL-boron
catalyst, and the observed chemical shifts for the complexed species for the vinylic

proton adjacent to the carbonyl.

 

NMR [product 3b] (M) [(S)-VAPOL-boron catalyst 8] Obs Chem Shift

 

(M) (ppm)
1 0.092 0.0046 5.338
2 0.092 . 0.00575 5.337
3 0.092 0.00775 5.335
4 0.092 0.0115 5.331
5 0.092 0.023 . 5.321
6 0.092 0.03075 5.314
7 0.092 0.046 5.301
8 0.092 . 0.069 5.292
9 0.092 0.092 5.288

 

NMR Titration Experiment with B(OPh)3 and Imine 150.

50 mg (0.184 mmol) of imine 150 was added to 10 separate flame dried argon
purged 2 mL volumetric flasks. 'To each volumetric flask was then transferred via
syringe a different amount of a 0.35 M solution of B(OPh)3 5 (see table below) prepared
using B(OPh)3 that had been distilled immediately before use. After the addition of the
B(OPh)3, the solutions were diluted to the 2 mL mark on the volumetric flask to ensure
(the. concentration of the imine was maintained. One separate flask was prepared using

only the imine with no catalyst. This was used as the reference sample. Approximately

291

0.8 mL of each volumetric flask was then transferred via syringe to a clean and dry NMR
tube and capped immediately to ensure no air was introduced to the system. The 1H-
NMR’s were then taken for each of the different catalyst/imine ratios. Recorded in the
table are the exact concentrations of imine 150 and B(OPh)3, and the observed chemical
shifts for the complexed species for the benzhydryl proton (H‘) and the imine 8p2 C-H

(H2).

 

NMR [imine 150] (M) [B(OPh)3] (M) Obs Chem Shift Obs Chem Shift

 

(H‘, ppm) (H2, 1mm)
1 0.092 0.0046 5.518 8.331
2 0.092 0.00575 5.517 8.329
3 0.092 0.00775 5.52 8.332
- 4 0.092 0.0115 5.522 8.334
5 0.092 0.023 5.526 8.338
6 0.092 0.03075 . 5.53 8.342
7 0.092 0.046 5.536 8.347
8 0.092 0.069 5.544 8.353
9 0.092 0.092 5.556 8.364
10 0.092 0.1505 5.59 , 8.392

 

NMR Titration Experiment with B(OPh)3 and the Product 151.

62.5 mg (0.184 mmol) of product 151 was added to 10 separate flame dried argon
purged 2 mL volumetric flasks. To each volumetric flask was then transferred via
syringe a different amount of a 0.35 M solution of B(OPh)3 (see table below) prepared
using B(OPh)3that had been distilled immediately before use. After the addition of the
B(OPh)3, the solutions were diluted to the 2 mL mark on the volumetric flask to ensure
the concentration of the imine was maintained. One separate flask was prepared using
only the imine with no catalyst. This was used as the reference sample. Approximately

0.8 mL of each volumetric flask was then transferred via syringe to a clean and dry NMR

292

tube and capped immediately to ensure no air was introduced to the system. The 1H-
NMR’s were then taken for each of the different catalyst/imine ratios. Recorded in the
table are the exact concentrations of product 151 and B(OPh)3, and the observed chemical

shifts for the complexed species for the vinylic proton adjacent to the carbonyl.

 

NMR [product 151] (M) [B(OPh)3] (M) Obs Chem Shift (ppm)

 

1 0.092 0.0046 5.344
2 0.092 0.00575 5.348
3 0.092 0.00775 5.348
4 0.092 0.0115 5.351
5 0.092 0.023 5.354
6 0.092 0.03075 3 5.359
7 0.092 , 0.046 5.365
8 0.092 0.069 5.374
9 0.092 0.092 5.383
10 0.092 0.1505 5.408

 

NMR Titration Analysis“‘

The stability constant comes from the NMR titration data by fitting the data to the
two site model. This assumes a 1:1 stoichiometery and fast exchange between the bound
and non bound forms of the NMR observedispecies. In the present case titrations were I
configured so that the chemical shifts of reporter protons on the starting imine or the
product amine (both termed ligand in the following discussion), were followed as a
function of varying catalyst concentration.

Any observed lH chemical shift is the mole fraction weighted average of the shifts
observed in the free and complexed molecule.

566s 7" XL 61. + XLCat BLCat (1)

293

where XL and XLCa; are the mole fractions of ligand that are free and bound to catalyst and

0L and 0mm are the chemical shifts of the reporter protons in the free and bound states.

For the formation of a 1:1 complex the following relationships describe the equilibrium

conditions.
[L] + [LCat] = [L]o (2)
[Cat] + [LCat] = [Cat]o ’ (3)
K, = [LCat]/[L] [Cat] (4)

[L]o and [Cat]o are the known solution compositions, and [L], [Cat] and [LCat] are the
equilibrium concentration of ligand, catalyst and complex respectively. The following

quadratic equation relates the equilibrium conditions to the known total concentrations

[LCat] = (a - b'”)/2K, (5)

where
a = K,[L]o + K,,[Cat]o + 1 (6)
b = (1611.1, - K.1Cat1.)2 + 2K.iLl.‘+ 2K.iCatl. + 1 (7)

This now allows solutions of equation (1) so that Sam can be calculated for any desired
solution composition and K,. I

K, is obtained from the NMR data by calculating a titration curve and matching it
to the experimental data by adjusting K, and 6m. This is accomplished within an Excel
spreadsheet, and using the ‘Solver’ tool to minimize the global error between the

experimental data and the calculated curve.

294

7.4 Experimental Procedures and Characterizations Data for Chapter Five

117-119

Preparation of BINOL Derivatives:

 

Preparation of 282:

To a flamed dried 500 mL round bottom flask equipped with a magnetic stir bar
was added (S)-BINOL (59, 9.5 g, 33.2 mmol) and CH2C12 (181 mL) and the reaction
mixture was then cooled to —78°C and Br2 (4.56 mL, 88.6 mmol) was added over 30
min. The reaction mixture was allowed to stir at —78°C for 30 min at which time the
cooling bath was removed and the reaction was warmed to room temperature over 2.5
hours. The reaction mixture was then transferred to a separatory funnel and washed with
10% sodium bisulfate, water, and sodium chloride. The organic extracts were combined,
dried over magnesium sulfate, filtered and the solvent removed under reduced pressure.
Purification was accomplished by recrystallization from benzene/cyclohexane to give
10.66 g of 282 (72% yield) with 1H NMR matching perfectly to the reference ."5
Spectroscopic data for 282 (ConlzBrzOz):

’H NMR (CDC13) 0 4.97 (s, 2H), 6.93 (d, 2H, J=9.0 Hz), 7.33-7.38 (m, 4H), 7.87

(d, 2H, J=9.0 Hz), 8.03 (d, 2H, J=1.8 Hz).

295

 

Preparation of 297‘”:

To a flamed dried 500 mL round bottom flask equipped with a magnetic stir bar
was added NaH (3.28 g, 82. mmol, 60% by wt. in mineral oil), diluted with DMF (7.7
mL) and cooled to 0°C. To the solution was then added (S)—6,6’-dibromo-BINOL (282,
5.122 g, 11.5 mmol) as a solution in DMF (76.7 mL). The reaction was allowed to stir at
0°C for 30 min at which time MOMCl (3.07 mL, 40.4 mmol) was added dropwise.
Stirring was continued at 0°C for 3 hours. The reaction was quenched with saturated
sodium bicarbonate and after workup yielded 6.25 g 297 (quantitative yield) with the 'H
N MR matching perfectly to the reference3 and was used in the next step without further
purification. I
Spectral data for 297 (CMHmBrZOQ:

1H NMR (CDC13) 5 3.13 (s, 6H), 5.01 (dd, 4H, J=33.3, 6.9 Hz), 6.95 (dd, 2H,
J=9.0, 0.6 Hz), 7.23 (m, 2H), 7.57 (d, 2H, J=9 Hz), 7.83 (d, 2H, J=9 Hz), 8.00 (d, 2H,

J=2.0 Hz).

296

 

Preparation of 298:”9

To a flamed dried 500 mL round bottom flask equipped with a magnetic stir bar
was added (5)—6,6’-dibromo-MOM-BINOL (297, 6.25 g, 11.5 mmol) and THF (98 mL)
and the reaction mixture was then cooled to —78°C. To.the cooled solution was added
n-BuLi (17.97 mL, 28.75 mmol, 1.6 M solution in hexanes) and was allowed to stir for
one hour at which time TMSCl (4.36 mL, 34.5 mmol) was added. The reaction mixture
was then allowed to stir for an additional 3 hours and then quenched with water and
diluted with ether. The water layer was extracted three times with ether and the organic
layers were combined and washed with water followed by brine. The organic extracts
were the combined, dried over magnesium sulfate, filtered and the solvent was removed
under reduced pressure. Purification was done by flash column chromatography and the
product was eluted with 3:1 hexanes/ethylacetate giving 7.75 g of 298 (91% yield) with
1H NMR matching the data from the reference.”9
Spectroscopic data for 298 (C30H3804Si2):

'H NMR(CDC13) 5 0.27 (s, 18H), 3.15 (s, 6H), 5.00 (dd, 4H, J=6.9, 21.3 Hz), 7.1
(d, 2H, J=8.4 Hz), 7.31 (dd, 2H, J=0.9, 8.1 Hz), 7.55 (d, 2H, J=9.3 Hz), 7.93 (d, 2H,

J=8.7 Hz), 8.00 (s, 2H).

297

 

Preparation of 299:”9

To a flamed dried 500 mL round bottom flask equipped with a magnetic stir bar
was added (5)-6,6’-ditrimethylsilyl-MOM-BINOL (298, 9.10 g, 17.54 mmol) and THF _
(90 mL) and the resulting solution was cooled to -—78°C. To the reaction mixture was
then added sec-BuLi (50.3 mL, 70.2 mmol, 1.4 M in cyclohexane) and was allowed to
stir at —78°C for 1.5 hours. To the reaction was then added 12 (26.7 g, 105.2 mmol) in
THF (50 mL). After stirring for 2 hours, the reaction was diluted with methanol and
transferred to a separatory funnel containing water and ethylacetate. The mixture was
extracted four times with ethylacetate. The organic layers were combined and washed
with 10% sodium bisulfate, water, sodium bicarbonate, and brine. The organic layers
were dried over magnesium sulfate, filtered, and the solvent was removed under reduced
pressure to give 12.66 g of crude 299 (93.7% yield) with lH NMR data matching exactly
that from the reference“. The crude 299 was used without purification for the next step.
Spectroscopic data for 299 (C30H361204Siz):

1H NMR (CDCl3) 5 0.28 (8, 18H), 2.62 (8, 6H), 4.71 (dd, 4H, J=5.7, 36.6 Hz),

7.10 (dd, 2H, J=0.6, 8.4 Hz), 7.38 (dd, 2H, J=1.2, 8.7), 7.88 (s, 2H), 8.52 (s, 2H).

298

 

Preparation of 281:”9

To a flamed dried 500 mL round bottom flask equipped with a magnetic stir bar
was added (S)—6,6’-ditrimethylsilyl—3,3’-diiodo-MOM-BINOL (299, 12.66 g, 16.4 mmol)
and CCl4 (170 mL) and the reaction mixture was then cooled to — 15°C. lCl (21g, 165.7
mmol) was then added in 20 ml CC14, The reaction was allowed to stir for 10 minutes at
which time the reaction mixture was quenched with 10% sodium bisulfate, The mixture
was then extracted three times with CH2C12. The organic extracts were combined and
washed with water, sodium bicarbonate, and brine. The solvent was then removed under
reduced pressure and the crude mixture was then diluted with CH2C12 (115 mL), cooled to
0°C, and treated with methanolic HCl (77 mL). This was allowed to stir at 0°C for 24
hours. The reaction mixture was transferred to a separatory funnel and diluted with water
and extracted three times with CH2C12. The organic layers were combined, dried over
magnesium sulfate and the solvent was removed under reduced pressure. Purification 3
was then accomplished by flash column chromatography (14:1 hexanes/ethyl acetate) to
give 281 with 1H NMR matching exactly to that from the reference.4
Spectroscopic data for 281 (C20HmI'402):

'H NMR (CDC13) 5 5.47 (br s, 2H), 6.7 (d, 2H, 9 Hz), 7.49 (dd, 2H, J=1.8, 9.3

Hz), 8.13 (d, 2H, J=1.5 Hz), 8.34 (s, 2H).

299

 

Preparation of 283:

To a flame dried 50 mL round bottom flask equipped with a magnetic stir bar was
added 6,6’-di-trimethylsilyl-MOM-BINOL (298, 0.48 g, 0.93 mmol) and CH2C12 (6.5
mL). This was then cooled to 0°C at which time saturated methanolic hydrogen chloride
(4.3 mL) was added.4 The reaction mixture was then allowed to stir overnight and
workup was accomplished as described for the preparation of 281. Purification was
accomplished using flash column chromatography with 14:1 hexanes/ethyl acetate used
as the eluent (rf 0.15 2.5:1 hexanes/ethylacetate) to give 253.4 mg (63% yield) of the
desired BINOL derivative 283 as a white solid.
Spectroscopic data for 283 (C26H3002Si2):

lH NMR (CDCl3) 5 0.31 (8, 1811), 4.99 (s, 2H), 7.11 (d, 2H, J=8.7 Hz), 7.33-7.41
(m, 4H), 7.96 (d, 2H, J=9.0 Hz), 8.02 (s, 2H); 13C NMR (CDC13) 5 — 1.43, 110.32,
117.33, 3122.98, 128.67, 131.30, 133.31, 133.89, 135.33, 152.72 (one overlapping sp2
carbon); IR (pure) 735m, 754m, 839vs, 899s, 1093m, 11348, 11558, 1201m, 1219s,
1250vs, 1383m, 1400m, l466vs, 1615vs, 29558, 3514br m cm"; Mass spectrum m/z (%
rel intensity) 432 M+2 (8), 431 M“1 (14), 430 M+ (33), 415 (40), 343 (5), 239(5), 200'

(100), 185 (9), 169(8), 73 (88); mp 97-103 °C.

300

 

Preparation of 284:

To a flame dried 50 mL round bottom flask equipped with a magnetic stir bar was
added 6,6’di-trimethylsilyl-3,3’-diiodo-MOM-BINOL (299, 0.917 g, 1.2 mmol) and g
CH2CI2 (8.4 mL). This was then cooled to 0°C at which time saturated methanolic
hydrogen chloride (5.6 mL) was added“. The reaction mixture was then allowed to stir
overnight and workup was accomplished as described for the preparation of 281.
Purification was accomplished via flash column chromatography using 535 mL 14.5:1
hexanes/CH2C12, then 270 mL 12.5:1 hexanes/ CH2C12, then 250 mL 14:1 hexanes/ethyl
acetate as the eluent (r,0.4 2.5:1 hexanes/ethyl acetate) to give 662.2 mg (81% yield) of
the desired BINOL derivative 284 as a white solid.
Spectroscopic data for 284 (C26H2812028i2):

1H NMR (CDC13) 5 0.28 (s, 18H), 5.43 (8, 2H), 7.06 (d, 2H, J=8.4 Hz), 7.41 (d,
2H, J=8.4 Hz), 7.93 (s, 2H), 8.52 (s, 2H); 13C NMR (CDCl;,) 5 -—1.49, 85.98, 112.11,
123.12, 129.92, 131.71, 132.72, 133.09, 136.47, 140.30, 150.13; IR 735m, 841vs, 8648,
9128, 1005w, 1070w, 1095m, 11528, 1190w, 1213w, 12508, 1379m, 14378, 1606w,
29538, 35208; Mass spectrum m/z, (% rel intensity) 685 M+3 (1), 684 M+2 (3), 683 MH (5),
682 'M+ (10), 670 (1), 669 (3), 668 (5), 667 (10), 610 (2), 595 (2), 556 (2), 541 (2), 326

(68), 262 (17), 199 (11), 128 (14), 73 (100); mp 128-134 °C.

301

Preparation of hydrazine imines‘z":

H H H
H
+ ,N Ph hexanes \ N Ph
Ph/\/KO H2N \g/ _’RT ”1M1“, \fl/
0
300 301 2“ 279a
Preparation of 2793:

To a 500 mL round bottom flask equipped with a magnetic stir bar was added the
dihydrocinnamaldehyde (300, 5 mL, 37.9 mmol), and hexanes (250 mL). To the
resulting suspension was added benzoylhydrazine (301, 3.45 g, 25.3 mmol). The resulting
' reaction mixture was allowed to stir for 2 hours. After 2 hours, the insoluble material
was collected and washed with hexanes. The powder was purified via recrystallization
from ether/MeOH resulting in 2.77g (44% yield) of the imine 279a as a crystalline solid.
The spectral data matched perfectly to that in the literature.‘20
Spectral data for compound 279a (CléHwNzO):

lH-NMR (CDC13) 5 2.62-2.83 (br m, 4H), 7.12-7.48 (br m, 8H), 7.61 (br s, 1H),
7.75-7.78 (m, 2H), 9.60 (br s, 1H); '3C—NMR (CDC13) 5 32.37, 33.53, 125.89, 127.03,

128.02, 128.20, 131.57, 132.80, 140.13, 151.31, 164.05 (one overlapping 8p2 carbon).

. ' H
M + N Ph hexanes V\/\/k\
H N ’ ——-> ,N Ph
0 2 \ll/ . RT N \n/
0 o

302 301 “might 279b

Preparation of 279b:
To a 500 mL round bottom flask equipped with a magnetic stir bar was added the

heptanal (302, 15 mL, 107.5 (mmol), and hexanes (350 mL). To the resulting suspension

302

was added benzoylhydrazine (301, 9.75g, 71.6 mmol). The resulting reaction mixture
was allowed to stir for 18 hours. After 18 hours, the insoluble material was collected and
washed with hexanes. The powder was purified via recrystallization from ether/MeOH
resulting in 15.5g (93% yield) of the imine 279b as a crystalline solid.
Spectral data for compound 279b (C , 4HwNzO):

lH-NMR (CDC13) 5 0.822 (t, 3H, J=6Hz), 1.22-1.40 (br m, 8H), 2.26 (q, 2H,
J=6.9 Hz), 7.29—7.45 (br m, 3H), 7.61-7.78 (br m, 3H), 9.96 (br s, 1H); l3C-NMR (CDC13) .
5 13.66, 22.13, 26.24, 28.58, 31.19, 32.15, 127.08, 128.14, 131.39, 132.83, 152.82,
164.00; mass spectrum m/z (% relative intensity); 233 M+1 (4), 232 M+ (17), 231 (5),
217 (2), 205 (13), 204(12), 190 (15), 188 (14), 176 (43), 174 (100), 161 (5), 147 (11),
120(9), 105 (50), 77 (50) 51 (5), 41 (8); IR (neat) 613.17m, 692.53m, 800.56m, 895.08w,
935.59w, 1051.34w, 1076.42w, 1142.00w, 1188.3w, 1268.68m, 1361.92m, 1466.09m,

1495.02m, 1577.978, 1653.218, 2581.018, 2922.538, 3063.358, 3227.328 cm"; mp:103-

 

 

105 °C.
I / Zr(On-Pr)4 (20 mol%)
H Si .
\ ,N Ph + 0’ \ Ligand (24 mol%) _
N T M TBDME/ DME (4: 1)
O 0M3 _10 °c 53 h W
27911 31 T

2813 0

Procedures for Scheme 5.9 (aza-Diels-Alder reaction of imine 279a using
Kobayashi’s conditions):

General Procedure: To a flame dried 5 mL round bottom flask equipped with a
magnetic stir bar was added the ligand (0.098 mmol, 24 mol%) and TBDME (0.4 mL)

followed by Zr(On-Pr)4 (0.032 mL, 0.082 mmol, 20 mol%). The reaction mixture was
303 '

allowed to stir at room temperature for 3 hours. Meanwhile, in a flame dried 10 mL
round bottom flask equipped with a magnetic stir bar was added'the imine 279a (103 mg,
0.41 mmol). The zirconium catalyst was then transferred from the 5 mL round bottom
flask to the flask containing 279a. The 5 mL flask was rinsed with 0.4 mL TBDME to
ensure all the catalyst was transferred. The reaction mixture was then cooled to 0°C at
which time the DME (0.2 mL) was added followed by the diene (31, 0.12 mL, 0.61
mmol). The reaction mixture was then allowed to stir for 53 hours after which, the
reaction mixture was transferred to a separatory funnel containing 25 mL saturated
sodium bicarbonate. The mixture was extracted three times with CH2C12, the organic
extracts. combined, dried over magnesium sulfate, filtered, and the solvent removed under
reduced pressure. Purification was done by flash column chromatography and the
enantiomeric excess Was mea8ured following the HPLC conditions from Kobayashi.”

Entry 1: The general procedure described above was followed exactly using (S)-
VAPOL (130, 53 mg, 0.098 mmol) and the reaction did not produce any of the desired
product 281a.

Entry 2: The general procedure described above was followed exactly using (S)-
VANOL (129, 43 mg, 0.098 mmol) and the reaction did not produce any of the desired
product 281a.

Entry 3: The general procedure described above was followed exactly using (S)-
BINOL (59, 28.1 mg, 0.098 mmol) and the reaction did not produce any of the desired

product 2813.

304

Entry 4: The general procedure described above was followed exactly using (S)-
6,6’-dibromo-BINOL (282, 43.5 mg, 0.098 mmol) and the reaction gave 8.3 mg (6.3%
yield) of the desired product 281a and 42% ee was measured.

Entry 5: The general procedure described above was followed exactly using (S)-
6,6’-ditrimethylsilyl-BINOL (283, 42.2 mg, 0.098 mmol) and the reaction gave less than
6 mg (<5% yield) of the desired product 281a and the ee for this reaction was not
measured.

Entry 6: The general procedure described above was followed exactly using (S)-
6,6’-ditrimethylsily1-3,3’-diiodo-BINOL (284, 67 mg, 0.098 mmol) and the reaction gave
15.4 mg (11.7% yield) of the desired product 2813 and 74.5% ee was measured.

Entry 7: The general procedure described above was followed exactly using (S)-
3,3’,6,6’-tetraiodo-BINOL (281, 77.4 mg, 0.098 mmol) and the reaction gave 53.9:mg
(41 % yield) of the desired product 281a and 94.2% ee was measured.

Entry 8: The general procedure described above was followed exactly using (S)-
3,3’-dinitroVANOL (285, 51.8 mg, 0.098 mmol) the reaction yielded 27.9 mg (21%
yield) of the desired product 281a and 87.1% ee was measured.

Spectral data for compound 281a (C20H20N202)3

1H NMR (CDCl;,) 5 1.89—2.02 (m, 2H), 2.47-2.70 (m, 4H), 4.05 (br s, 1H), 5.00

(d, 1H, J=7.8 Hz), 7.01-7.24 (br m, 5H), 7.38-7.55 (m, 4H), 7.72 (d, 2H, J=7.8 Hz), 9.04

(s, 1H).

305

 

(S)-VAPOL(130) + B(OPh)3 CH2C12: 0'5 mmHg:
10 mol% 100mo|% 55 °C, 1h 55 °C,0.5h

)

cat“

cat”

 

 

 

 

 

 

H Sl/
l
\ [N Ph ’ \ DCM/toluene(1:1)
N T + O . .
M ..45 c.2411
0 OMe TPh
279a ' 31
2813 O
13.3% yield
9%ee

Procedure for Scheme 5.9 (aza-Diels-Alder reaction of imine 279a using (S)-
VAPOL/B(OPh)3 as the Catalyst):

To a flame dried, argon purged single-necked flask that had its 14/20 joint
replaced with a threaded Teflon high-vacuum T-shaped stop-cock equipped with a stir
bar was added B(OPh)3 (0.128 g, 0.41 mmol) and (S)-VAPOL (130, 22 mg, 0.041 mmol).
To this was added CH2C12 (0.8 mL) and then the flask was sealed with the stopcock and
heated to 55 °C for one hour. After one hour, the solvent was removed via high vacuum
and left under high vacuum at 55 °C for 0.5 hours yielding catalyst. After cooling the
stopcock was removed and replaced with a rubber septum. The catalyst was dissolved by
the injection via syringe of 1.0 mL of a 1:1 mixture of toluene and CH2C12 (in two
portions). The catalyst solution was transferred by syringe to a solution of the imine
279a prepared as immediately below.

To a flame dried, argon purged homemade flask. with a. cold addition coil
equipped with a stir bar was added the imine (2793, 0.3764 g, 1.0 mmol). The flask was
topped with two rubber septa and the catalyst was added in two 0.5 mL portions of
toluene/ CH2C12 (1:1) directly to the bottom of the flask by a syringe equipped with a

long needle. This was allowed to stir for 5-10 minutes at room temperature and then

306

cooled to —45 °C. Meanwhile, in a separate flame dried 5 mL round bottom flask purged
with argon was added Danishefsky’s diene (31, 0.38 mL, 2.0 mmol) and toluene/CH2C12
(1:1) (1.5 mL). The diene was taken up in a syringe and added over 3.0 hours via syringe
pump through the cold addition coil. The reaction was then allowed to stir at —45 °C for
24 h. After completion of the reaction, saturated sodium bicarbonate (~20 mL) was
added to the reaction flask at —45 °C. This was then transferred to a separatory funnel
containing saturated sodium bicarbonate (25 mL) and extracted with three or four 30-40
mL portions of CH2C12. The combined organic layers were then dried with magnesium
sulfate, filtered, and solvent was removed via rotary evaporation. Purification was then
done using flash column chromatography to give 16.8 mg of 281a (13.3% yield) and 9%

ee was measured using the HPLC conditions form Kobayashi.31

 

 

o o OTMS 0 OTMS OTMS
1) LDA A \ A 1) LDA _ \
. 2) TMSCI . 2) TMSCI .
303 304 278

Preparation of the bisTMS diene (278)”:

To a flame dried argon purged 500 mL round bottom flask equipped with a
magnetic stir bar was added 2-acetyl-cyclohexanone (303, 4.95 mL, 37.5 mmol) and THF
(60 mL). The solution was cooled to —78°C and LDA (40.5 mL, 45 mmol, 1.11 M
solution in THF) was added over 5 minutes. The resulting solution was stirred for 30
minutes at which time TMSCl (7.11 mL, 56.25 mmol) was added over about 10 minutes.
After the addition of TMSCl, the reaction was immediately warmed to room temperature

to give 304. The solution containing 304 was cooled back to —78°C and LDA (40.5 mL,

307

45 mmol, 1.11 M solution in THF) was added over 5 minutes. The resulting solution was
stirred for 30 minutes at which time TMSCl (7.11 mL, 56.25 mmol) was added over
about 10 minutes. After the addition of TMSCI, the reaction was immediately allowed to
warm to room temperature. After the reaction was stirred for one hour at room
temperature, the solvent was removed under reduced pressure and the resulting gel was
dissolved in ether and filtered through Celite. The Celite was washed with several
portions of ether to ensure all the desired material was extracted from the Celite.
Purification was accomplished by distillation (94—95°C, 0.6 mmHg) to afford 9.5 g (89%
yield) of the desired diene 278.) The spectral data matched perfectly to that in the

literature.90

\ .
/SI\ ,51\
O 0 Zr(On-Pr)4 (286, 20 mol%)
H \ .
W ,N Ph Ligand (24 mol%) >
N T " N
O

TBDME/ DME. (4:1) I
— 10 °C, 53 h 287 ”N P“

 

Procedure for Scheme 5.11 (aza-Diels-Alder reaction of imine 279a and diene 278):
To a flame dried 5 mL round bottom flask equipped with a magnetic stir bar was
added the ligand (0.098 mmol, 24 mol%) and TBDME (0.4 mL) followed by Zr(On-Pr)4
(0.032 mL, 0.082 mmol, 20 mol%). The reaction mixture was allowed to stir at room
temperature for 3 hours. Meanwhile, in a flame dried 10 mL round bottom flask
equipped with a magnetic stir bar was added the imine (279b, 103 mg, 0.41 mmol). The
zirconium catalyst was then transferred from the 5 mL round bottom flask to the flask ‘
containing 279a. The 5 mL flask was rinsed with 0.4 mL TBDME to ensure all the
catalyst was transferred. The reaction mixture was then cooled to 0°C at which time
DME (0.2 mL) was added followed by the diene (278, 0.12 mL, 0.61 mmol). The

308

reaction mixture was then allowed to stir for 53 hours. After stirring for 53 hours, the
reaction mixture was transferred to a separatory funnel containing 25 mL saturated
sodium bicarbonate. The mixture was then extracted three times with CH2C12, the
organic extracts combined, dried over magnesium sulfate, filtered, and the solvent
removed under reduced pressure.

(S)-VAPOL (130), (5)-VANOL (129), (S)-BINOL (59), (S)—6,6'—dibromo-INOL
(282), (S)—6,6'—di(trimethylsilyl)-BINOL (283), (S)—6,6'-di(trimetliylsilyl)-3,3'-diiodo-
BINOL (284), and (S)—3,3',6,6'—tetraiodo-B1NOL (281) were all attempted as ligands for

this reaction and none of the reactions using these ligands gave any of the desired product

 

287.
o .
H TMSO OTMS
W E ph + \ Zr(OnPr)4/1‘1B1NOL_
N Y toluene
2791) 0 RT to reflux I?
0

Procedure for attempts at running reaction at warmer temperatures:

Reaction 1: To a flame dried argon purged round bottom flask equipped with a
magnetic stir bar was added (S)-tetraiodoB1NOL (281, 28.1 mg; 0.098 mmol) followed
by toluene (0.4 mL) and Zr(OnPr)4 (0.032 mL, 0.082 mmol). This was then allowed to
stir at room temperature for 3 hours. Meanwhile, to a flame dried 10 mL round bottom
flask was added the imine 279b (95.3 mg, 0.41 mmol). The zirconium catalyst was then
transferred to the flask containing the imine. The catalyst flask was rinsed with toluene
(0.4 mL) to ensure all of the catalyst was transferred. At room temperature, the diene
(278, 0.175 g, 0.61 mmol) was added in toluene (0.2 mL). The reaction was then allowed

309

to stir for 48 hours at room temperature. After 48 hours, more diene (0.12 mL, 0.41
mmol) was added to the reaction mixture and it was then heated to reflux for 24 hours.
The reaction mixture was then transferred to a separatory funnel containing saturated
sodium bicarbonate (25 mL). Extraction was done three times with CH2C12 and the
organic layers were combined, dried over magnesium sulfate, filtered, and the solvent
was removed under reduced pressure. Analysis of the crude NMR showed that none of
the desired product 287 had been produced.

Reaction 2: The conditions as described for reaction 1 were, followed exactly
except the reaction was diluted with toluene (3mL) and heated to reflux after addition of
the first aliquot of the diene 278. After 21.5 hours, the crude 1H NMR showed that only
starting material was present so another aliquot of the diene 278 (0.15 mL, 0.5 mmol)
was added. After refluxing for an additional 12 hours, the reaction was worked up as in

reaction 1 but still no product was observed in the crude 1H NMR.

\ ’ B . , u]

tern rature HN Ph
2.78 3° , 287 \ll/
[6861.101] time

 

Procedure for Scheme 5.12 (racemic aza-Diels-Alder reactions of imine 279b and
diene 278):

Condition AA: To a flame dried argon purged’round bottom flask equipped with a
magnetic stir bar was added the imine 279b (116.2 mg, 0.5 mmol) followed by CH2C12 (5
mL) and then BF3-OEt2 (288, 0.069 mL, 0.55 mmol). The flask was then cooled to

—45°C at which time the diene (278, 0.213 g, 0.75 mmol) was added. A white

310

precipitate formed before the reaction was cooled and upon addition of the diene, the
precipitate disappeared and the reaction turned slightly yellow. The reaction was then
allowed to stir at -—45°C for 2 hours at which time the reaction was transferred to a
separatory funnel. Saturated sodium bicarbonate was added and the extraction was done
four times with CH2C12 (30 mL). The organic layers were combined, dried over
magnesium sulfate, filtered and concentrated under reduced pressure. Several new spots
were observed by TLC analysis, but none of the desired product 287 was isolated.
Conditions BB: To a flame dried argon purged round bottom flask equipped with
a magnetic stir bar was added the imine 279b (116.2 mg, 0.5 mmol) followed by CH2C12
(5 mL). The reaction was then cooled to —45°C at which time BF3°OEt2 (288, 0.069 mL,
0.55 mmol) was added followed immediately by the diene (278, 0.213 g, 0.75 mmol).
No white precipitate was formed upon addition of BF3-OEtz. The reaction was allowed to
stir at -—45°C overnight and after 20 hours, the reaction was transferred to a separatory
funnel, treated with saturated sodium bicarbonate and extracted four times with CH2C12
(30 mL). The organic layers were combined, dried over magnesium sulfate, filtered and
concentrated under reduced pressure. None of the desired product 287 was isolated.
. Conditions CC: To a flame dried argon purged round bottom flask equipped with
a magnetic stir bar was added the imine 279b (116.2 mg, 0.5 mmol) followed by CH2C12
(5 mL). The reaction was then cooled to —-45°C and the diene (278, 0.213 g, 0.75 mmol)
'was added followed immediately by BF3-OEt2 (288, 0.069 mL, 0.55 mmol). The reaction
was allowed to stir at —45°C for 2 hours. At this time the reaction was transferred to a
separatory funnel, treated with saturated sodium bicarbonate and extracted four times

with CH2C12 (30 mL). The organic layers were combined, dried over magnesium sulfate,

311

filtered and concentrated under reduced pressure. Purification was accomplished via
flash column chromatography, Rf 0.05 (2.521 hexanes/ethyl acetate), by ramping the
solvent from 10:3 hexanes/EtOAc (500 mL) to 1:1 hexanes/EtOAc to give 51.6 mg (29%
yield) of the desired product 287 as a wax-like substance. 3

Conditions DD: To a flame dried argon purged round bottom flask equipped with
a magnetic stir bar was added the imine 279b (116.2 mg, 0.5 mmol) followed by CH2C12
(5 mL). The reaction was then cooled to -—78°C anad the diene (278, 0.213 g, 0.75
mmol) was added followed immediately by BF3-OEtQ (288, 0.069 mL, 0.55 mmol). The
reaction was then allowed to warm to —45°C and stirred for 19 hours. After 19 hours,
the reaction was transferred to a separatory funnel, treated with saturated sodium
bicarbonate and extracted four times with CH2C12 (30 mL). The organic layers were
combined, dried over magnesium sulfate, filtered and concentrated under reduced
pressure. Purification was accomplished via flash column chromatography, R, 0.05
(2.5:1 hexanes/ethyl acetate), by ramping the solvent from 10:3 hexanes/EtOAc (500 mL)
to 1:1 hexanes/EtOAc to give 51.7 mg (29% yield) of the desired product 287 as a'wax-
like substance.

Conditions EE: To a flame dried argon purged round bottom flask equipped with
a magnetic stir bar was added the imine 279b (116.2 mg, 0.5 mmol) followed by CH2C12
(5 mL). The reaction was then cooled to ——78°C and the diene (278, 0.213 g, 0.75 mmol)
was added followed immediately by BF3°OEt2 (288, 0.069 mL, 0.55 mmol). Immediately
after the addition of BF3-OEt2, the cooling bath was removed and the reaction was
allowed to warm to room temperature and stirred for 19 hours. After 19 hours, the

reaction was transferred to a separatory funnel, treated with saturated sodium bicarbonate

312

and extracted four times with CH2C12 (30 mL). The organic layers were combined, dried
over magnesium sulfate, filtered and concentrated under reduced pressure. Purification
was accomplished via flash column chromatography, R, 0.05 (2.5:1 hexanes/ethyl
acetate), by ramping the solvent from 10:3 hexanes/EtOAc (500 mL) to 1:1
hexanes/EtOAc to give 63.5 mg (36% yield) of the desired product 287 as a wax-like
substance.

Conditions FF: To a flame dried argon purged round bottom flask equipped with
a magnetic stir bar was added the imine 279b (116.2 mg, 0.5 mmol) followed by CH2C12
(20 mL). The reaction was then cooled to —78°C and the diene (278, 0.213 g, 0.75
mmol) was added followed immediately by BF;.,°OEt2 (288, 0.069 mL, 0.55 mmol). The
reaction was allowed to warm to —45°C and stirred for 19 hours. After 19 hours, the
reaction Was transferred to a separatory funnel, treated with saturated sodium bicarbonate
and extracted four times with CH2C12 (30 mL). The organic layers were combined, dried
over magnesium sulfate, filtered and concentrated under reduced pressure. Purification
was accomplished viaflash column chromatography, Rf 0.05 (2.5:1 hexanes/ethyl
acetate), by ramping the solvent from 10:3 hexanes/EtOAc (500 mL) to 1:1
hexanes/EtOAc to give 56.3 mg (32% yield) of the desired product 287 as a wax-like
substance. I

Conditions 00: To a flame dried argon purged round bottom flask equipped With
a magnetic stir bar was added the imine 279b (116.2 mg, 0.5 mmol) followed by CH2C12
(5 mL). The reaction was then cooled to —78°C and the diene 278 (redistilled
immediately before use) (0.213 g, 0.75 mmol) was added followed immediately by

BF3-OEt2 (288, 0.069 mL, 0.55 mmol). Immediately after the addition of BF3'OEt2, the

313

cooling bath was removed and the reaction was allowed to warm to room temperature
and stired for 20 hours. After 20 hours, the reaction was transferred to a separatory
funnel, treated with saturated sodium bicarbonate and extracted four times with CH2C12
(30 mL). The organic layers were combined, dried over magnesium sulfate, filtered and
concentrated under reduced pressure. Purification was accomplished via flash column
chromatography, Rf 0.05 (2.5 :1 hexanes/ethyl acetate), by ramping the solvent from 10:3
hexanes/EtOAc (500 mL) to 1:1 hexanes/EtOAc to give 114.1 mg (64% yield) of the
desired product 287 as a wax-like substance.

Conditions HH: To a flame dried argon purged round bottom flask equipped with
a magnetic stir bar was added the imine 279b (116.2 mg, 0.5 mmol) followed by CH2C12
(5 mL). The reaction was then cooled to —78°C and the diene 278 (redistilled
immediately before use) (0.213 g, 0.75 mmol) was added followed immediately by
BF3-Olit2 (288, redistilled immediately before use) (0.069 mL, 0.55 mmol). Immediately
after the addition of BF3-OEt2, the cooling bath was removed and the reaction was
allowed to warm to room temperature and stirred for 20 hours. After 20 hours, the
reaction was transferred to a separatory funnel, treated with saturated sodium bicarbonate
and extracted four times with CH2C12 (30 mL). The organic layers were combined, dried
over magnesium sulfate, filtered and concentrated under reduced pressure. Puurification
was accomplished via flash column chromatography, R, 0.05 (2.5:1 hexanes/ethyl
acetate), by ramping the solvent from 10:3 hexanes/EtOAc (500 mL) to 1:1
hexanes/EtOAc to give 103.7 mg (58.5% yield) of the desired product 287 as a wax-like

substance.

314

Conditions I]: The conditions as described for Conditions HH were followed
exactly, except 0.2323 g (1.0 mmol) of the imine 279b was used. The reaction gave
147.5 mg (42% yield) of the desired produt 287 as a wax-like substance.

Conditions JJ (0.4—46g): To a flame dried argon purged round bottom flask
equipped with a magnetic stir bar was added the imine 279b (446 mg, 1.92 mmol)
followed by CH2C12 (20 mL). The reaction was then cooled to —78°C and the diene 278
(redistilled immediately before use) (3 mL, 10.12 mmol) was added followed
immediately by BIfi-OEt2 (0.265 mL, 2.11 mmol). Immediately after the addition of
BF3'OEt2, the cooling bath was removed and the reaction was allowed to warm to room
temperature and stir for 20 hours. After 20 hours, the reaction was transferred to a
separatory funnel, treated with saturated sodium bicarbonate and extracted four times
with-CH2C12. The organic layers were combined, dried over magnesium sulfate, filtered
and concentrated under reduced pressure. Purification was accomplished via flash
column chromatography, Rf 0.05 (2.5:1 hexanes/ethyl acetate), by ramping the solvent
from 10:3 hexanes/EtOAc (500 mL) to 1:1 hexanes/EtOAc to give 466.2 _mg (68% yield)
of the desired product 287 as a wax-like substance.

Conditions JJ (6.21 g): To a flame dried argon purged round bottom flask
equipped with a magnetic stir bar was added the imine 279b (6.21g, 26.75 mmol)
followed by CH2C12 (275 mL). The reaction was then cooled to —78°C and the diene
278 (redistilled immediately before use) (11.4 g, 40.11 mmol) was added followed
immediately by BF3-OEt2 (288, 3.7 mL, 29.4 mmol). Immediately after the addition of
BF3'OEt2, the cooling bath was removed and the reaction was allowed to warm to room

temperature and stir for 20 hours. After 20 hours, the reaction was transferred to a

315

separatory funnel, treated with saturated sodium bicarbonate and extracted four times
with CH2C12. The organic layers were combined, dried over magnesium sulfate, filtered
and concentrated under reduced pressure. Purification was accomplished via flash
column chromatography, Rf 0.05 (2.521 hexanes/ethyl acetate), by ramping the solvent
from 10:3 hexanes/EtOAc (500 mL) to 1:1 hexanes/EtOAc to give 5.97 g (63% yield) of
the desired product 288 as a wax—like substance.

Conditions KK (16.3 g): To a flame dried argon purged round bottom flask
equipped with a magnetic stir bar was added the imine 279b (16.3 g, 70.3 mmol) followed
by CH2C12 (700 mL). The reaction was then cooled to -—78°C and the diene 278
(redistilled immediately before use) (30 g, 105.4 mmol) was added followed immediately
by BF3°OEt2 (288, 9.7 mL, 77.33 mmol). Immediately after the addition of BF3-OEt2, the
temperature was allowed to warm to —20° overnight. After 15 hours, the cooling bath
was removed and the reaction was allowed to warm to room temperature and stirred for
an additional 4 hours. After 19 hours total reaction time, the reaction mixture was
transferred to a separatory funnel, treated with saturated sodium bicarbonate and
extracted four times with CH2C12. The organic layers were combined, dried over
magnesium sulfate, filtered and concentrated under reduced pressure. Purification was
accomplished via flash column chromatography, R,'0.05 (2.5:1 hexanes/ethyl acetate), by
ramping the solvent from 10:3 hexanes/EtOAc (500 mL) to 1:1 hexanes/EtOAc to give
>95 g (>40% yield) of the desired product 287 as a wax-like substance.

Spectral data for compounds 287 (CnHwNZOZ):
The enantiomers could be separated by HPLC using a Chiralcel OD column (with

guard column) (90/10 hexanes/isopropanOl, 1 mL/min). Retention times: 14.12 min. and

316

16.99 min. 1H NMR (CDC13) 5 0.78 (t, 3H, J=6.9 Hz), 1.16-1.66 (br m, 13H), 2.07-2.40
(br m, 6H), 2.60 (dd, 1H, J=16.2, 4.5 Hz), 3.89 (s, 1H), 7.38-7.55 (m, 3H), 7.84 (d, 2H
8.4 Hz), 9.27 (br s, 1H); 13C NMR (CDC13) 5 13.62, 21.19, 21.37, 21.78, 22.14, 24.65,
26.12, 28.91, 30.51, 31.30, 40.95, 59.96, 108.77, 127.06, 128.43, 131.49, 132.13, 161.76,
166.61, 191.50; IR (pure) 1136w, 1262m, 15228, 1559vs, 1617s, 1653s, 1684s, 2859s,
2930vs, 3254br s cm"; Mass spectrum m/z (% rel intensity) 355 M+ (83), 269 (19), 234

(100), 163 (30), 105 (71). '

Preparation of Cuprates 1, 2, and 3:

Cuprate 1:

WI

 

W0“ + HNAN + PPh3 + 12 CH3CN/THF >
\_:-_/ 0°C to RT

305 306 - 307 308 309

Preparation of 4—iodo-1-butene (309):'21 To a flame dried argon purged 250 mL
round bottom flask equipped with a magnetic stir bar was added PPh3 (307, 23 g, 87.75 3
mmol),3-butene—1—ol (305,5 mL, 58.5 mmol), imidazole (30.6, 8 g, 117 mmol), and a
3.1:1 mixture of THF/CH3CN (111 mL). The resulting solution was then cooled to 0°C
and I2 (308, 25.4 g, 100 mmol) was added in three portions over ~5 minutes. After the
additionof I2 was completed, the ice bath was removed and the reaction mixture was
allowed to stir at room temperature for 30 minutes. The dark solution was then
transferred to a separatory funnel containing 10% sodium bisulfite (180 mL) and pentane
' (170 mL). The layers were separated and the organic layer was dried over magnesium

sulfate, filtered, and the solvent was removed under reduced pressure. Purification was

317

accomplished by flash column chromatography to give 3.82 g (36% yield) of 309 as a

light pink oil. The H—NMR matched perfectly to the literature.121

 

310 THF 311
—78°C to RT

Preparation of 3—butenyllithium (311): To a flame dried argon purged round
bottom flask equipped with a magnetic stir bar was added 4—iodo-1—butene (310, 1.0
equiv.) and THF (2.1 mL/mmol). The reaction mixture was then cooled to —78°C at‘
which time t—BuLi (1.0 equiv., 1.7 M solution in hexane-s) was added. The reaction
mixture was then allowed to warm slowly to room temperature. The yield was assumed
to be 90% and the entire solution of the resulting lithium species 311 was used directly

for the preparation of the cuprate 1.

 

\A/ U + CuCN 5‘20 > M '
311 312 —78 ac to 00C / 2CUL12CN
Cuprate 1

Preparation of cuprate 1: To a round bottom flask equipped with a magnetic stir
bar was added CuCN (312, 1.0 equiv.) The flask was then flame dried and purged with
argon and then EtzO was added (3 mL/mmol) and cooled to —78°C. To the shiny was
then transferred the previously prepared solution 3-butenyllithium 311. Ether (1.0
mL/mmol) was used to rinse any left over material from the butenyllithium flask. The
resulting solution was then warmed to 0°C for one minute to give a 0.13 M solution of
cuprate 1. The resulting solution of cuprate 1 was then cooled back to -—78°C and used

immediately for the 1,4—addition reactions.

318

Cuprate 2:

\A/ Br + Mg° ____> M MgBr
reflux
313 overnight 314

Preparation 0fButenylmagnesiumbromide (306): To a flame dried argon purged
round bottom flask equipped with a magnetic stir bar and a reflux condenser was added
magnesium (50 mesh, 1.05 equiv.), THF (1.25 mlJmmol), and 4—bromo-1—butene (313,
1.0 equiv). The resulting solution was then heated to reflux and stirred overnight. The

Grignard reagent 314 was then used immediately for the preparation. of the cuprate 2.

WMSB’ + MeLi + CuCN €246» MCuLngCN
314 315 312 " M6

Cuprate 2

Preparation of Cuprate 2: To a round bottom flask equipped with a magnetic stir
bar was added copper cyanide. (312, 1 equiv.).‘ The flask was then flame dried and
purged with argon and to it was added THF (1.5 mL/mmol) and cooled to —78°C. To
the slurry was added MeLi (315, 1.0 equiv., 1.6 M in ether) over two minutes. The
reaction was then placed in a 0°C ice bath for 2-3 minutes and then cooled back to
—78°C at which time the freshly prepared butenylmagnesium bromide (314, 1.0 equiv,
0.8M in THF) (as prepared above) was added dropwise over 4 minutes. Stirring was
allowed for another 10 minutes to give a 0.26 M solution _of cuprate 2. The resulting

solution of cuprate 2 was then used for the 1,4—addition reactions.

319

Cuprate 3:

S n-BuLi S Li
<\ /7 ———> if
THF
316 __780C 317

‘22 To a flame dried argon purged round

Preparation of thiophenyllithium (317):
bottom flask equipped with a magnetic stir bar was added thiophene (316, 1.0 equiv.) and
THF (0.9 mL/mmol). The reaction mixture was cooled to —78°C at whichtime n-BuLi
(1.0 equiv., 2.5 M solution in hexanes) was added. The temperature was then allowed to.
warm slowly to 0°C and was stirred for another 30 minutes to give a 0.676 M solution of

thiophenyllithium 317. The resulting solution of 317 was used immediately for the

preparation of cuprate 3.

M B 8 Li THF NCuLtMgCN
M g r.+ g + CuCN W _
314 317 ' 312 A;;:5

Cuprate 3

Preparation of Cuprate 3:'22 A round bottom flask was charged with CuCN (312,
1.0 equiv.) and a magnetic stir bar. The flask was then flame dried and purged with
argon and the contents were diluted with dry THF (1.33mL/mmol) and cooled to —78°C.
To the flask was added the freshly prepared thiophenylithium (317, 1.0 equiv., 0.676M ’
solution in T HP as prepared above) over 30 seconds after which, the reaction mixture
was then warmed to room temperature. The resulting amber solution was then cooled
back to —78°C and freshly prepared butenylmagnesiumbromide (314, 1.0 equiv. 0.8M

solution in THF as prepared above) was added over 1 minute. The reaction flask was

320

then transferred to a 0°C ice bath and remained there for 2 minutes to giving a 0.36 M
solution of cuprate 3. The resulting solution of cuprate 3 was then cooled back to

—78°C and used‘ immediately for the 1,4-addition reactions.

 

 

O
1 t
I. + MCuLiZCN so V6“ a
N 2 temperature
5 l Cuprate 1 - -
HN Ph LeWIs Ac1d
\n/ reaction time
0 O
288
287

Procedures for Scheme 5.14 (1,4-addition of cuprate 3 to hydrazine 287):

Entry I: To the previously prepared solution of cuprate 1 (3.7 mL, 0.48 mmol,
0.13 M solution in THF/hexanes/ether) was added the hydrazine 287 (0.1134 g, 0.32
mmol). The reaction was allowed to stir at —78°C for 3 hours and then warmed to
—25°C for 16 hours. After 19 hours total reaction time, a 9:1 solution of NH4Cl/NH4OH
(5.0 mL) was added. The cooling bath was removed and the reaction mixture was
allowed to stir for 1 hour. The solution was transferred to a separatory funnel, extracted
three times with ethyl acetate. The organic layers were then combined and washed with
brine, dried over magnesium sulfate, filtered and the solvent was removed under reduced
pressure. None of the desired product 288 was observed in the crude 1H NMR.

Entry 2: To the previously prepared Solution of cuprate 1 (2.75 mL, 0.689 mmol,
0.25 M solution in ether) was added the hydrazine 287 (0.627 g, 0.459 mmol) The
cooling bath was then packed with dry ice and the reaction was allowed to warm very

slowly to —25°C and stirred for a total of 16 hours. After 16 hours total reaction time, a

321

9:1 soluti0n of NH4Cl/NH4OH (5.0 mL) was added. The cooling bath was removed and
the reaction mixture was allowed to stir for 30 minutes. The solution was transferred to a
separatory funnel, extracted three times with ethyl acetate, The organic layers were then
combined and washed with brine, dried over magnesium sulfate, filtered and the solvent
was removed under reduced pressure. None of the desired product 288 was observed in
the crude 1H NMR.

Entry 3: To the previously prepared solution of cuprate 1 (2.84 mL, 0.71 mmol,
0.25 M solution in ether) was added the hydrazine 287 (0.1008 g, 0.284 mmol). The
reaction was allowed to stir at -—78°C for 3 hours, then warmed slowly to —— 10°C for an
additional 19 hours. After 22 hours total reaction time, a 9:1 solution of NH4Cl/NH4OH
(5.0 mL) was added. The cooling bath was removed and the reaction mixture was
allowed to stir for 30 minutes. The solution was transferred to a separatory funnel,-
extracted three times with ethyl acetate, The organic layers were then combined and
washed with brine, dried over magnesium sulfate, filtered and the solvent was removed
under reduced pressure. None of the desired product 288 was observed in the crude 1H
NMR. .

Entry 4: To the previously prepared sOlution of cuprate 1 (1.32 mL, 0.33 mmol,
0.25 M solution in ether) was added a previously prepared solution of the hydrazine 287
(78 mg, 0.22 mmol) and Zr(O-nPr)4 (0.0686 mL, 0.22 mmol) in a 4:3 mixture of .
ether/CH2C12 (3.5 mL) dropwise via syringe over 2-3 minutes. The resulting black
reaction mixture was allowed to stir at —78°C for 21 hours and the temperature was
raised to —50°C stirring for another 23 hours, and finally the reaction was warmed to

—25°C for an additional 24 hours. After a total of 67 hours, a 9:1 solution of

322

NH4Cl/NH4OH (5.0 mL) was added. The cooling bath was removed and the reaction
mixture was allowed to stir for 2 hours. The solution was transferred to a separatory
funnel, extracted three times with ethyl acetate, The organic layefs were then combined
and washed with brine, dried over magnesium sulfate, filtered and the solvent was
removed under reduced pressure. Only a few minor peaks were observed in the vinyl

region of the crude 1H NMR but none of the desired product 288 was isolated.

O

 

O
I. 8ml2 (289, 2.2 equiv.) _
N solvent f I
- N
HN Ph room temperature H
287 \n/ reaction time 290

O
Procedures for Scheme 5.15 (reductive cleavage using SmIz):

Entry I: To a flame dried argon purged 50 mL 3-neck round bottom flask
equipped with a magnetic stir bar and a reflux condenser was added Sm° (0.5 g, 3.33
mmol), 12 (0.767 g, 3.02 mmol) and THF (9 mL). The resulting solution was heated to
reflux for 1.5 hours. The dark blue 0.333 M solution of SmI2 (289) was stored in the dark
and used for the reductive cleavage as described immediately below.

To a flame dried argon purged 25 mL round bottom flask was added the
hydrazine 287 (0.2331 g, 0.658 mmol) and THF (9.5 mL). To the solution was then
added the previously prepared solution of Sm12(289, 4.35 mL, 1.45 mmol, 0.333 M
solution in THF). The reaction was allowed to stir for 45 minutes at which time the
solution was treated with 3N HCl (20 mL). The mixture was extracted three times with

ether. The aqueous layer was then treated with 10% aqueous sodium hydroxide and this

323

 

was then extracted three times with ether. These organic layers of the second extraction
were then combined, dried, filtered, and the solvent removed under reduced pressure.
Purification was accomplished by flash column chromatography (rf 0.13 5:2
hexanes/ethyl acetate) using 1:1 hexanes/ethyl acetate as the eluent to give 18 mg (12%
yield) of the desired product 290.

Entry 2: To a flame dried argon purged 50 mL round bottom flask equipped with
a magnetic stir bar and a reflux condenser was added Sm° (0.316 g, 2.1 mmol), and THF”
(30 mL) followed immediately by the addition of CHZI2 (0.16 mL, 2.0 mmol). The
resulting solution was stirred at room temperature until the solution turned dark blue (~3
hours). The resulting dark blue 0.067 M solution of Sml2 (289) was stored in the dark
and used for the reductive cleavage described immediately below.

To a flame dried argon purged 25 mL round bottom flask was added the
hydrazine 287 (0.1 g, 0.282 mmol) followed by the addition of the previously prepared
solution of SmI2 (289, 9.26 mL, 0.62 mmol, 0.067 M solution in THF). The solution
turned brown within 45 minutes but TLC analysis showed that the starting material (287)
was still present so more SmI2 (3.0 mL, 0.2 mmol) was added. The resulting mixture was
then allowed to stir one hour at which time MeOH (3 mL) was added and stirred for an
additional 5 hours. The reaction mixture was then treated with water (20 mL) and
extracted three times with EtQO. The organic extracts were combined and washed with
brine, dried over magnesium sulfate, filtered, and the solvent removed under reduced
pressure. Purification was accomplished by flash column chromatography (rf 0.13 5:2
hexanes/ethyl acetate) using 1:1 hexanes/ethyl aCetate as the eluent to give 27.4 mg (38%

yield) of the desired product 290.

324

Entry 3: To a flame dried argon purged 50 mL 3-neck round bottom flask
equipped with a magnetic stir bar and a reflux condenser was added Sm° (0.5 g, 3.33
mmol), 12 (0.771 g, 3.02 mmol) and THF (15 mL). The resulting solution was heated to
reflux for 1.5 hours. The resulting dark blue 0.2 M solution of Sm12(289) was stored in
the dark and used for the reductive cleavage described immediately below.

To a flame dried argon purged 25 mL round bottom flask was added the
hydrazine 287 (0.2792 g, 0.787 mmol) and MeOH (2.5 mL). To the solution was then
added the previously prepared solution of SmI2 (289, 8.66 mL, 1.73 mmol, 0.2 M
solution in THF). The blue c010r disappeared immediately and after 45 minutes, TLC
analysis showed that the starting material 287 was still present so more Sm12(2.0 mL, 0.4
mmol) was added. The resulting mixture was then allowed to stir overnight after which
the solution was treated with 3N HCl (20 mL). The mixture was extracted three times
with EtzO. The aqueous layer was then treated with 10% aqueous sodium hydroxide and
this was then extracted three times with ether. These organic layers of the second
extraction were then combined, dried over magnesium sulfate, filtered, and the solvent
removed under reduced pressure. Purification was accomplished by flash column
chromatography (rf 0.13 5:2 hexanes/ethyl acetate) using 1:1 hexanes/ethyl acetate as the
eluent to give 14 mg (7 .6% yield) of the desired product 290 as a white solid.

Spectral data for compound 290 (ClstsNO):

1H NMR (CDC13) 5 0.853 (t, 3H, J=6.9 Hz), 1.22-1.27 (m, 10H), 1.46-1.72 (m,
5H), 2.09-2.25 (m, 3H), 2.30 (s, 1H), 2.39 (dd, 1H, J=3.6, 15.9 Hz), 3.44—3.55 (m, 1H),
4.24 (s, 1H); 13C NMR (CDC13) 5 13.63, 20.57,. 21.73, 22.14, 22.25, 24.89, 28.66, 28.75,

21.28, 34.35, 41.90, 52.68, 106.09, 159.02, 191.82; IR; 1147.79m, 1203.73m, 1219.17m,

325

1242.32m, 1265.46m, 1307.90m, 1344.56m, 1359.99m, 1412.07m, 1441.01m, 1468.028,
1502.748, 1535.538, 1549.048, 1566.408, 1610.778, 2853.08m, 2922.538, 3285.19br m;
Mass spectrum m/z (% rel intensity) 236 M+1 (46) 235 M+ (47), 164 (20), 150 (100), 146

(58), 122 (48); white solid, mp 95—97°C.

Mg° (291, X equiv.) O
I HgClz (292, Y amt.)_
If THF/ EtOH (1:3) 7 I
N

237 \n/ temperature, time H
O 290

 

Procedures for Scheme 5.16 (initial attempts for the reductive cleavage using Mg°
and HgClz):

Entry I : To a flame dried argon purged 5 mL round bottom flask equipped with a
magnetic stir hat was added the hydrazine 287 (100 mg, 0.282 mmol) and Mg° (mesh,
67.4 mg, 2.82 mmol) and a spatula tip of HgClz. The contents of the flask were then
diluted with methanol (1.07 mL) and THF (0.36 mL). The mixture was allowed to stir
for 18 hours and TLC analysis showed that only the starting material was present. At this
time, another two or three spatula tips fullof HgCl2 were added and immediately an
extremely exothermic reaction was observed .heating the solvent to boil. After 30 .
minutes the TLC analysis showed that the starting material had been consumed. The
reaction was quenched with water and extracted 3 times with E90. The ether extracts
were then washed with brine, dried over magnesium sulfate, filtered, and the solvent was

removed under reduced pressure. Purification was accomplished by flash column

326

chromatography (rf 0.13 5:2 hexanes/ethyl acetate) using 1:1 hexanes/ethyl acetate as the
eluent to give 13.2 mg (18.2% yield) of the desired product 290.

Entry 2: The protocol described for entry 1 was followed exactly except the
reaction was cooled to 0°C before the THF and methanol was added. TLC analysis after
20 minutes showed no reaction had taken place so more HgCl2 was added to initiate the
reaction. The same workup as described in entry 1 was followed to give 23.3 mg (16.7%
yield) of the desired product 290.

Entry 3: To a flame dried argon purged 5 mL round bottom flask equipped with a
magnetic stir bar was added the hydrazine 287 (100 mg, 0.282 mmol) and Mg° (mesh, .
67.4 mg, 2.82 mmol) and HgCl2 (54 mg, 0.199 mmol). The contents of the flask were
then diluted with methanol (1.07 mL) and THF (0.36 mL). The mixture was allowed to
stir for 6 hours and TLC analysis showed that the starting material was consumed. The
reaction was quenched with water and extracted 3 times with ether. The ether extracts
were then washed with brine, dried over magnesium sulfate, filtered, and the solvent was
removed under reduced pressure. Purificatidn was accomplished by flash column
chromatography using (r{ 0.13 5:2 hexanes/ethyl acetate) 1:1 hexanes/ethyl acetate as the

eluent to give 45.8 mg (69% yield) of the desired product 290.

 

| THF/ EtOH (1:3)

287 ”Kn/P" RT, 18.5 H

O 290

Mg° (291, 3 equiv.) O
I HgC12 (292, X equiv.)
N = I
N

Procedures for Scheme 5.17 (determination of the optimal amount of HgClz):

327

Entry I: The protocol as described for Scheme 5.16 entry 3 was followed exactly
where 7.7 mg (0.0282 mmol) of HgCl2 was added and the reaction was allowed to stir for
18.5 hours. Analysis of the crude 'H NMR showed that the reaction only went to <5% ,
conversion.

Entry 2: The protocol as described for Scheme 5.16 entry 3 was followed exactly
where 15.3 mg (0.0564 mmol) of HgCl2 was added and the reaction was allowed to stir
for 18.5 hours. Analysis of the crude 1H NMR showed that the reaction only went to
<5% conversion.

Entry 3: The protocol as described for Scheme 5.16 entry 3 was followed exactly
where 24.7 mg (0.0846 mmol) of HgCl2 was added and the reaction was allowed to stir
for 18.5 hours. Analysis of the crude 1H NMR showed that the reaction only went to
19% conversion.

Entry 4: The protocol as described for Scheme 5.16 entry 3 was followed exactly
where 38.3 mg (0.141 mmol) of HgCl2 was added and the reaction was allowed to stir for
18.5 hours. Analysis of the crude 1H NMR showed thatithe reaction went to 80%
conversion.

Entry 5: The protocol as described for Scheme 5.16 entry 3 was followed exactly
where 46 mg (0.1692 mmol) of HgCl2 was added and the reaction was allowed to stir for
18.5 hours. Analysis of the crude 1H NMR showed that the reaction went to 70%
conversion. 3

Entry 6: The protocol as described for Scheme 5.16 entry 3 was followed exactly

where 61.3 mg (0.2256 mmol) of HgCl2 was added and the reaction was allowed to stir

328

for 18.5 hours. Analysis of the crude 1H NMR showed that the reaction only that the
reaction went to >95% conversion.

' Entry 7: The protocol as described for Scheme 5.16 entry 3 was followed exactly
where 78.6 mg (0.282 mmol) of HgCl2 was added and the reaction was allowed to stir for
18.5 hours. Analysis of the crude 1H NMR showed that the reaction went to >95%

conversion.

0

Mg° (291, 3.0 equiv.)
I HgC12 (292, 1.0 equiv.)
N = I
N .

 

l THF/ EtOl-l (1:3)

287 “NY 0°C to RT, 17 h H
290

Procedures for Scheme 5.18 (optimization and scalability for reductive cleavage):
Entry I (0.355g): To a flame dried argon purged 10 mL round bottom flask
equipped with a magnetic stir bar was added the hydrazine 287 (0.355 g, 1.0 mmol) and
Mg° (mesh, 72 mg, 3.0 mmol) and HgCl2 (0.272 g, 1.0 mmol). The contents of the flask
were then cooled to 0°C and diluted with methanol (3.5 mL) and THF (1.2 mL). The
mixture was allowed to warm to room temperature and stir for 17 hours. TLC analysis
showed that the starting material was consumed. The reaction was quenched with cold
0.5 N HCl (20 mL) and extracted 3 times with CH2C12. The organic extracts were then
combined and washed with brine, dried over magnesium sulfate, filtered, and the solvent
was removed under reduced pressure. Purification was accomplished by flash column
chromatography (rf 0.13 5:2 hexanes/ethyl acetate) using 1:1 hexanes/ethyl acetate as the

eluent to giVe 202.2 mg (80% yield) of the desired product 290.

329

Entry 2 (1.0 g): To a flame dried argon purged 50 mL round bottom flask
equipped with a magnetic stir bar was added the hydrazine 287 (1.0 g, 2.82 mmol) and
Mg° (mesh, 203 mg, 8.46 mmol) and HgCl2 (0.766 g, 2.82 mmol). The contents of the
flask were then cooled to —20°C and diluted with methanol (10.0 mL) and THF (3.2
mL). The mixture was allowed to warm to room temp and stir for 17 hours. TLC
analysis showed that the starting material was consumed. The reaction was quenched
with cold 0.5 N HCl (67 mL) and extracted 3 times with CH2C12. The organic extracts
were then combined and washed with brine, dried over magnesium sulfate, filtered, and
the solvent was removed under reduced pressure. Purification was accomplished by flash
column chromatography (rf 0.13 5:2 hexanes/ethyl acetate) using 5:2 hexanes/ethyl
acetate ramping to 1:1 hexanes/ethyl acetate as the eluent to give 0.535 g (62% yield) of
the desired product 290.

Entry 3 (1.2g): To a flame dried argon purged 50 mL round bottom flask
equipped with a magnetic stir bar was added the hydrazine 287 (1.2 g, 3.376 mmol) and
Mg° (mesh, 243 mg, 10.13 mmol) and HgCl2 (0.916 g, 3.376 mmol). The contents of the
flask were then cooled to 0°C and then diluted with methanol (11.8 mL) and THF (3.8
mL). The mixture was allowed 'to warm to room temperature and stir for 17 hours. TLC
analysis showed that the starting material was consumed. The reaction was quenched
with cold 0.5 N HCl (80 mL) and extracted 3 times with CH2C12. The organic extracts
were then combined and washed with brine, dried over magnesium sulfate, filtered, and
the solvent was removed under reduced pressure. Purification was accomplished by flash

column chromatography (rf 0.13 5:2 hexanes/ethyl acetate) using 5:2 hexanes/ethyl

330

acetate ramping to 1:1 hexanes/ethyl acetate as the eluent to give 0.535 g (62% yield) of
the desired product 290.

. Entry 4 (9.0 g): To a flame dried argon purged 50 mL round bottom flask
equipped with a magnetic stir bar was added the hydrazine 287 (9.0 g, 25.39 mmol) and
Mg° (mesh, 1.83 g, 76.2 mmol) and HgCl2 (0.766 g, 25.39 mmol). The contents of the
flask were then cooled to 0°C and diluted with methanol (90 mL) and THF (30 mL). The
mixture was allowed to warm to room temp and stir for 17 hours. TLC analysis showed
that the starting material was consumed. The reaction was quenched with cold 0.5 N HCl
(550 mL) and extracted 3 times with CH2C12. The organic extracts were then combined
and washed with brine, dried over magnesium sulfate, filtered, and the solvent was
removed under reduced pressure. Purification. was accomplished by flash column
chromatography (rf 0.13 5:2 hexanes/ethyl acetate) using 5:2 hexanes/ethyl acetate

ramping to 1:1 hexanes/ethyl acetate as the eluent to give 3.6 g (60% yield) of the desired

 

product 290.
0 o
O THF
+ "3“” + PhAO/lkCl —78°C to RT 7
£11 22 h J;
290 1.0 equiv. 293 80% yield 0 O
1.0 equiv. 1.02 equiv.

”A
294

Procedures for Scheme 5.19 (preparation of Cbz protected amine 294):
Reaction 1 : To a flame dried argon purged 25 mL round bottom flask was added
the amine 290 (0.2401 g, 1.02 mmol) and THF (3.5 mL). The solution was then cooled

to —78°C and n-BuLi (0.37 mL, 1.03 mmol, 2.6 M solution in hexanes) was added over

331

1.5 hours vialsyringe pump. After the addition was complete, the resulting mixture was
stirred for 30 minutes and to it was then added Cszl (293, 0.15 mL, 1.04 mmol) in THF
(1.5 mL) over 3.0 minutes. This was allowed to stir for 30 minutesland the cooling bath
was removed warming the solution to room temperature in about 45 minutes. The
reaction was then transferred to a separatory funnel containing saturated ammonium
chloride and extracted with ethyl acetate three times, washed with brine, dried over
magnesium sulfate, filtered and the solvent was removed under reduced pressure.
Purification was accomplished using flash column chromatography (5:1 hexanes/ethyl
acetate, rr 0.35) go give 242.4 mg (71.9 % yield) of the desired product 294. In addition,
82.1 mg of the starting material was recovered (97.7% overall yield based on recovered
amine).

Reaction 2: The protocol as described for reaction 1 was followed exactly except
the reaction was allowed to stir for 22 hours at room temperature rather than 45 minutes.
The reaction gave 238.9 mg (80.1% yield) of the desired product 294.

Spectral data for compound 294 (CEH3,NO3):

1H NMR (CDCI3) 0.80 (t, 3H, J=6.9 Hz), 1.16—1.48 (br m, 11H), 1.62-1.77 (br m,
3H), 2.06—2.41 (br m, 4H), 2.71-2.78 (dd, 1H, J=6.0, 17.1Hz), 2.99-3.06 (br m, 1H), 4.63-
4.69 (m, 1H), 5.15 (s, 2H), 7.23-7.32 (m, 5H); 13C NMR (CDC13) 13.54, 13.84, 21.25,
22.11, 25.89, 28.55, 30.42, 31.22, 41.49, 41.83, 52.21, 54.90, 55.29, 55.94, 67.75, 120.10,

127.61, 128.06, 128.49, 135.32, 150.60, 153.33, 193.14.

332

 

0
W O THF
+ NaH + >
N PhAOJKCl RT, overnight
N

290 293 k0

O
294
n?

Unsuccessful Attempts for Cbz Protection using NaH: ’0‘

Attempt I: To a flame dried argon purged 25 mL round bottom flask equipped
with a magnetic stir bar was added NaH (16.6 mg, 0.356 mmol, 60% by wt. in mineral
oil) and THF (0.5 mL); To the reaction mixture was then added the amine 290 (69.9 mg,
0.297 mmol) and the resulting mixture was allowed to stir at room temperature for 1.5
hours. To the reaction mixture was then added Cszl°(293, 60.7 mg, 0.356 mmol). The
solution was then allowed to stir overnight. Purification was then accomplished by flash
column chromatography using 1:1 hexane/ethyl acetate as the eluent giving 53.3 mg
(~50% yield) of a mixture of material that contained only about 50% 294.

Attempt 2: The conditions as described for attempt 1 were followed exactly except
the reaction was cooled to —78°C before NaH and Cszl were added. After warming to
room temperature and working up the reaction, it was observed that none of the desired

product 294 was formed.

 

O .

' O
m 0
' + Hunigs Base + >
N , PhAO/lkCl RT. overnight '
H N
' 293
29° 294 J80

Unsuccessful Attempts for Cbz Protection using Hunig’s Base: '02

333

Following the protocol from the reference,”2 to a flame dried argon purged 25 mL
round bottom flask equipped with a magnetic stir bar was added the amine 290 (.2244g,
0.953 mmol) and CH2C12 (3.6 mL). To the reaction mixture was then added Hunigs base
(0.66 mL, 0.3.8 mmol) followed immediately by the addition of Cszl (293, 0.41 mL,
2.86 mmol). The resulting mixture was allowed to stir at room temperature for 19 hours.
The solution was then transferred to a separatory funnel containing 30 mL of water and
extracted three times with CH2C12, the organic layers combined, washed with 1N HCl (50
mL), and twice with saturated sodium bicarbonate (2 x 20 mL). None of the desired

product 294 was isolated from this reaction.

THF

temperature

 

V

+ Cuprate

reaction time

   

294 295

Procedures for Scheme 5.21 (addition of cuprates 1, 2, and 3 t0 the Cbz protected
amine 294):

Entry 1: To a previously prepared solution of cuprate 2 (2.3mL, 0.599 mmol, 0.25
M solution in THF/EtzO) (see preparaton of cuprates above) was added the Cbz protected
Vinylogous amide 294 (0.1474g, 0.399 mmol) in THF (0.5 mL). The reaction was then
allowed to stir at —78°C for 10.5 hours at which time the cooling bath was packed with
dry ice to ensure that the reaction mixture warmed slowly to room temperature. After 20

h, the reaction was cooled back to —78°C and to it was added 9:1 mixture of

334

NH4Cl/NH4OH (3 mL). The reaction was then allowed to warm to room temperature and
stirred for 3h. The resulting blue solution was then transferred to a separatory funnel and
extracted three times with ethyl acetate. The organic layers were then combined and
washed with brine, dried over magnesium sulfate, filtered and the solvent was removed
under reduced pressure. The crude lLH NMR indicated that mostly starting material was
present with only <5% other material observed in the baseline in the vinyl region.

Entry 2: To a previously prepared solution of cuprate 3122 (2.1 mL, 0.75 mmol,
0.36 M solution in THF/E90) (see preparaton of cuprates above) was added the
' Vinylogous amide 294 (0.77 mL, 0.5 mmol, 0.65M solution in THF) was then added all at
once. The acetone bath was then packed with dry ice to ensure that the reaction mixture
warmed very slowly to room temperature. After 17 h, the reaction was cooled back to
—78°C and to it was added 9:1 mixture of NH4Cl/NH4OH (2.5 mL). The reaction was
then allowed to warm to room temperature and stirred for 3h. The resulting blue solution
was then transferred to a separatory funnel and extracted three times with ethyl acetate.
The organic layers were then combined and washed with brine, dried-over magnesium
sulfate, filtered and the solvent was removed under reduced pressure. The crude 1H
NMR indicated that only starting material was present and none of the desired product
295 was formed. I 5

Entry 3: The protocol as described in entry 2 was followed exactly except after
the Vinylogous amide 294 was added, the reaction was immediately allowed to warm to
—45°C over one hour and stir for 30 minutes, then warmed to —20°C for 30 minutes,
and then to 0°C for 14 hours, and finally to 10°C for one more night. After 42 hours t0tal

reaction time, the reaction was cooled back to —78°Cto it was added 9:1 mixture of

335

NH4Cl/NH4OH (2.5 mL). The reaction was then allowed to warm to room temperature
and stirred for 3h. The resulting blue solution was then transferred to a separatory funnel
and extracted three times with ethyl acetate. The organic layers were then combined and
washed with brine, dried over magnesium sulfate, filtered and the solvent was removed
under reduced pressure. The crude 1H NMR indicated that only starting material was
present and none of the desired product 295 was formed.

Entry 4: To the previously prepared solution of cuprate 1 (4.03 mL, 0.525 mmol,
0.13 M solution in THF/hexanes/ether) (see preparaton of cuprates above) was added the
Vinylogous amide (0.129g, 0.35 mmol) dropwise in ether (1.0 mL). The reaction was
allowed to stir at —78°C for 3 hours and a 9:1 solution of NH4Cl/NH4OH (5.0 mL) was
added. The cooling bath was removed and the reaction mixture was allowed to stir for 2
hours. The solution was transferred to a separatory funnel, extracted three times with
ethyl acetate, The organic layers were then combined and washed with brine, dried over
magnesium sulfate, filtered and the solvent was removed under reduced pressure. None
of the desired product 295 was observed in the crude 1H NMR.

Entry 5: To the previously prepared solution of cuprate 1 (4.03 mL, 0.525 mmol,
0.13 M solution in THF/hexanes/ether) (see preparaton of cuprates, above) was added the
Vinylogous amide 294 (0.129g, 0.35 mmol) dropwise in ether (1.0 mL). The reaction was
allowed to stir at —78°Cfor 3 hours and then warmed to —20°C for 53 hours. After 53
hours, a 9:1 solution of NH4Cl/NH4OH (5.0 mL) was added. The cooling bath was
removed and the reaction mixture was allowed to stir for 2 hours. The solution was
transferred to a separatory funnel, extracted three times with ethyl acetate, The organic

layers were then combined and washed with brine, dried over magnesium sulfate, filtered

336

and the solvent was. removed under reduced pressure. The starting material was
consumed and purification was attempted via flash column chromatrgraphy using 10:3
hexanes/EtOAc to give 37.9 mg of a mixture of two spots by TLC and 36.9 mg of pure
uncharacterized material. The two isolated compounds were not the desired product 295
and further characterization. was not accomplished.

Entry 6: To the previously prepared solution of cuprate 1 (3.7 mL, 0.48 mmol,
0.13 M solution in THF/hexanes/ether) (see preparaton of cuprates above) was added the
Vinylogous amide 294 (0.118 g, 0.32 mmol). The reaction was allowed to stir at —78°C 3
for 3 hours and then warmed to —25°C for 16 hours. After 19 total reaction hours, a 9:1
solution of NH4Cl/NH4OH (5.0 mL) was added. The cooling bath was removed and the
reaction mixture was allowed to stir for 1 hour. The solution was transferred to a
separatory funnel, extracted three times with ethyl acetate, the organic layers were then
combined and washed with brine, dried over magnesium sulfate, filtered and the solvent
was removed under reduced pressure. None of the desired product 295 was observed in
the crude 1H NMR.

Entry 7: To the previously prepared solution of cuprate 1 (5.2 mL, 0.671 mmol,
0.13 M solution in THF/hexanes/ether) (see preparaton of cuprates above) was added the
Vinylogous amide 294 (101.2 mg, 0.274 mmol) dropwise in ether (1.0 mL). The reaction
was then allowed to warmed to —10°C and was stirred for 48 hours at which time a 9:1
solution of NH4Cl/NH4OH (5.0 mL) was added. The cooling bath was removed and the
reaction mixture was allowed to stir for 30 minutes. The solution was transferred to a
separatory funnel, extracted three times with ethyl acetate, The organic layers were then

combined and washed with brine, dried over magnesium sulfate, filtered and the solvent

337

was removed under reduced pressure. The starting material was consumed and
purification was accomplished via flash column chromatography using 5:1 mixture of
hexanes/EtOAc to give 38.8 mg of pure material that was not the desired product 295.

Further characterization was not accomplished.

 

 

O
+ MCuLiZCN + Lewis Acid ' B20 4,,
I N 2 —78°C to —50°c
5 A Cuprate 1 54 h
0 (I: 1.5 equiv.
294 P“

Procedures for Scheme 5.22 (Lewis Acid Promoted'm” addition of cuprate ~l to the
Cbz protected amine 294):

Entry I (BF3-OEtQ): To the previously prepared solution of cuprate 1 (2.23 mL,
0.29 mmol, 0.13 M solution in THF/hexanes/ether) (see preparaton of cuprates above) I
was added a previously prepared solution of the Vinylogous amide 294 (71.3 mg, 0.193
mmol) and BF3-OEt2‘ (0.0242 mL, 0.193 mmol) in ether (1.0 mL) dropwise via syringe
' over 2-3 minutes. The reaction was allowed to stir at —78°C for 21 hours and the
temperature was raised to —50°C and the reaction was stirred for another 23 hours.
After 44 hours total reaction time, a 9:1 solution of NH4Cl/NH4OH (5.0 mL) was added,
the cooling bath removed and the reaction mixture was allowed to stir for 2 hours. The
solution was transferred to a separatory funnel, extracted three times with ethyl acetate.
The organic layers were then combined and washed with brine, dried over magnesium
sulfate, filtered and the solvent was removed under reduced pressure. None of the

desired product 295 was observed in the crude 1H NMR.

338

Entry 2 (ZnClz): To the previously prepared solution of cuprate 1 (1.9 mL, 0.251
mmol, 0.13 M solution in THF/hexanes/ether) (see preparaton of cuprates above) was .
added a previously prepared solution of the vinylo‘gous amide 294 (61.7 mg, 0.167 mmol)
and ZnCl2 (0.2 mL, 0.167 mmol, 1.0 M solution in ether) in ether (0.87 mL) dropwise via
syringe over 2-3 minutes. The reaction was allowed to stir at —78°C for 21 hours and
the temperature was raised to —50°C and the reaction was stirred for another 23 hours
after which a 9:1 solution of NH4C1/NH4OH (5.0 mL) was added. The cooling bath was
removed and the reaction mixture was allowed to stir for 2 hours. The solution was
transferred to a separatory funnel, extracted three times with ethyl acetate. The organic
layers were then combined and washed with brine, dried over magnesium sulfate, filtered
and the solvent was removed under reduced pressure. None of the desired product 295
was observed in the crude 1H NMR.

Entry 3 (T MSCl): To the previously prepared solution of cuprate 1 (2.4 mL, 0.309
mmol, 0.13 M solution in THF/hexanes/ether) (see preparaton of cuprates above) was
added a previously prepared solution of the Vinylogous amide 294 (76.2mg, 0.206 mmol)
and'TMSCl (0.026 mL, 0.206 mmol), in ether (1.0 mL) dropwise via syringe over 2-3
minutes. The reaction was allowed to stir at —78‘?C for 21 hours and the temperature
was raised to —25°C over three hours and the reaction was stirred for another 22 hours.
After 56 hours total reaction time, a 9:1 solution of NH4C1/NH4OH (5.0 mL)'was added _ '
and the cooling bath was removed. The reaction mixture was allowed to stir for 2 hours
and the solution was transferred. to a separatory funnel, extracted three times with ethyl
acetate. The organic layers were then combined and washed with brine, dried over

magnesium sulfate, filtered and the solvent was removed under reduced pressure.

339

Purification was attempted via flash column chromatography using 10:1 hexanes/ethyl
acetate as the eluent and 24 mg of the starting material 294 was recovered. In addition,
17.8 mg of some other unknown material was isolated that was not the desired product
295.

Synthesis of the Alternate diene 227:

00 _orfo

 

I? I? THF \
. +LDA + F3C—fi—O—ISI—CF3 48°00“ > .
o o
318 319 320 - 321

Preparation of the triflate 321: To a flame dried argon purged 250 mL round
bottom flask equipped with a magnetic stir bar was added 2-acetyl-cyclohexanone 318‘
(7.0 mL, 50 mmol) and THF (65 mL). The solution was cooled to —7 8°C and LDA 319
(45 mL, 44 mmol, 1.0 M solution in THF) was added over 15 minutes. The resulting
solution was stirred at —78°C for 30 minutes at which time, trifluoromethyl sulfonic
. anhydride 320 (8.07 mL, 48 mmol) was added lover about 10 minutes. The reaction
mixture was then stirred for 30 minutes at —78°C and the cooling bath was removed
allowing the reaction to warm to room temperature. After 3 hours at room temperature,
the solvent was removed and purification was accomplished by flash column
chromatography using 15:1 hexanes/ethyl acetate ramping to 10:1 hexanes/ethylacetate to
to give the desired triflate 321 as a colorless oil.

Spectral data for compound 321 (C9H11F3O3S):

'H NMR (CDC13) 5 1.63-1.69 (br m, 2H), 1.73-1.79 (br m, 2H), 2.36 (s, 3H),

2.37-2.46 (br m, 4H). Final' structure determination was confirmed by further

modification to the diene 227.

340

OTf O

\ _
+ NMgBr + Fe(acac)3 THF/NMP(16.7:1) >
. —30°C
321 314

322 1"

 

 

323

Preparation of 323’ 28: To a flame dried argon purged 250 mL round bottom flask
equipped with a magnetic stir bar 'was added the triflate 321 (1.32 g, 4.85 mmol). The
triflate was then diluted with THF (70 mL) and Fe(acac)3 322 (0.171 g, 0.485 mmol) was
added followed immediately by N—methylpyrolidinone (NMP, 4.4 mL). The resulting
reaction mixture was then cooled to -—30°C and to it was added
butenylmagnesiumbromide 314 (11 mL, 5.35 mL, 0.5 M solution in THF). The reaction
was stirred for 30 minutes and TLC analysis showed that starting material was still
present so to the reaction was added more butenylmagnesiumbromide 314 (1.46 mL, 2.92
mmol). The reaction was stirred for an additional 15 minutes and then the reaction
mixture was transferred to a separatory funnel containing ammonium chloride.
Extraction was done three times with ether and the combined organic layers were dried
over magnesium sulfate, filtered and the solvent was removed under reduced pressure.
Purification was accomplished via flash column chromatography (r{ = 0.37 9:1
hexanes/ethyl acetate) using 15:1 hexanes/ethyl acetate as the eluent to give the desired
product323 as a colorless oil.

Spcetral data for compound 323 (C12H180):
‘H NMR (CDCl3) 5 1.50-1.56 (m, 3H), 2.05-2.35 (br m, 12H), 4.85-4.97 (in, 2H),

5.67-5.80 (m, 1H); 13C NMR (CDCl3) 5 21.83, 21.96, 26.50, 29.22, 30.01, 32.42, 33.92,

341

114.19, 133.47, 137.93, 142.5, 204.08; Mass spectrum m/z (% rel intensity) 179 M“1
(100), 178 M* (17), 163 (12), 150 (14), 149 (65), 136 (18), 135 (55), 121 (10), 107 (13).

Final structure determination was confirmed by further modification to the diene 227.

OTMS

1) LHMDS _ \
2) TMSCl .

227

 

 

Preparation of diene 227: To a flame dried argon purged 250 mL round bottom
flask equipped with a magnetic stir bar was added the ketone 324 (1.682 g, 9.435 mmol)
and THF (20 mL). The solution was cooled to -——78°C and LHMDS (9.9 mL, 9.91 mmol,
1.0 M solution in THF) was added over 15 minutes. The resulting solution was allowed
to warm slowly to —30°C and then cooled back to —78°C. To the reaction was then
added TMSCl (1.5 mL, 11.8 mmol) in THF (10 mL) over about 30 minutes. Immediately
after the addition was complete, the reaction was allowed to warm to room temperature.
The solvent was then removed under reduced pressure and the resulting gel was diluted
with ether and filtered through Celite. The Celite was washed with several portions of
ether to ensure all the desired material was extracted from the Celite. Purification was
accomplished by distillation (75°C, 0.1 mmHg) to afford 1.85 g (78.2% yield) of the
diene 227 as a colorless oil. ,

Spectral data for compound 227:
'H NMR (CDC13) 5 0.18 (s, 9H), 1.56 (br s, 4H), 1.99-2.26 (br m, 8H), 4.11 (s,

1H), 4.27 (s, 1H), 4.90-5.02 (m, 2H), 5.75—5.88 (br m, 1H); 13C NMR (CDCl3) 5 0.18,

342

22.76, 28.83, 28.46, 29.01, 32.73, 34.14, 92.80, 114.08, 130.77, 134.55, 138.89, 157.79;
IR (neat) 752.33w, 844.93vs, 864.238, 910.52m, 1010.838, 1035.91m, 1074.49w,
1089.92w, 1201.968, 1282.54 8, 1286.688, 1361.92vw, 1439.08w, 1448.73w, 1614.62m,
1641.63w, 1686.00vw, 2858.97m, 2930.248, 3076.85w cm"; mass spectrum m/z (% rel
intensity) 250 M+ (5), 236 (21), 235 (42), 222 (60), 221 (68), 205 (18), 179 (13), 145

(22), 132 (22), 120 (31), 117 (52), 75 (100), 73 (99),, 45 (72).

 

 

\
W\/|H\ H OTMS BF 'OEt (l leq)
\ ,N Ph + 3 2 ' >
N Y \ CH2C12
O —78°C to RT
279b
. 227
1.0 equ1v.
1.2 equiv
296
23% yield

Procedure for Scheme 5.23 (aza-Diels-Alder attempt using alternate diene 227):

To a flame dried argon purged 25 mL round bottom flask was added the imine
279b (0.2323 g, 1.0 mmol) and CH2C12 (5 mL). The contents of the flask were then
cooled to —78°C and to the flask was. added the diene 227 (0.3008 g, 1.20 mmol) in two
2.5 mL portions of CH2C12. After 5 minutes, BF3-OEt2 (0.138 mL, 1.10 mmol) was added
all at once and the cooling bath was packed with dry ice so the reaction would warm very
slowly to room temperature. After 22 hours total reaction time, the solution was
transferred to a separatory funnel containing saturated sodium bicarbonate, extracted
three times with CH2C12. The combined organic layers were dried, filtered and the
solvent was removed under reduced pressure. Purification was accomplished by flash

column chromatography using 6.5:] hexanes/ethyl acetate (500 mL) and then 3:1

343

hexanes/ethyl acetate to afford 100 mg (23% yield) of the uncyclized Mannich type
product 296.
Spectral data for compound 296:

'H NMR (CDC13) 5 0.83 (t, 3H, J=6.9 Hz), 1.23-1.59 (br m, 16H), 2.00-2.18 (br
m, 6H), 2.56-2.75 (m, 2H), 3.38 (br s, 1H), 4.86-4.98 .(br m, 2H), 5.12 (br s, 1H), 5.68-
5.81 (br m, 1H), 7.35-7.49 (br m, 3H), 7.72-7.76 (br m, 2H), 8.09 (br s, 1H); 13C NMR
13.68, 21.77, 21.96, 22.22, 25.67, 26.42, 29.00, 30.01, 31.34, 32.42, 33.00, 34.12, 45.35,
56.73, 114.21, 126.50, 128.23, 131.24, 132.56, 133.33, 137.94, 143.16, 166.08, 205.98;
IR (neat) 704.118, 794.77w, 908.598, 972.25 w, 995.40m, 1028.19w, 1074.49m,
1140.08m, 1178.66w, 1282.838, 1363.85m, 1475.738, 1506.608, 1579.90m, 1676.35vs,

2855.01vs, 2924.46vs, 3028.63m, 3067.218, 3289.05vs cm".

344

7.5 Experimental Procedures and Characterizations Data for Chapter Six

 

Ph |
. Si/
\N X Ph + 0’ \ 10 mol% catalyst _
IO CH2C12/toluene (1 : 1)
. MOMe

 

150 31 .—45 f3, 2411
1.0 equiv. 2.0 equiv. c011 addluon flask
added over 3 h

in 3 mL solvent

Procedure for Scheme 6.1 (aza-Diels-Alder Reaction Using the One (B1) and Two
Boron (B2) Catalyst)
Entry I: To a flame dried, argon purged single-necked flask that had its
14/20 joint replaced with a threaded Teflon high-vacuum T-shaped stop-cock equipped
with a stir bar was added BH3°SMe2 (0.1 mmol, 0.5 M solution in toluene), (S)-VAPOL
(54 mg, 0.1 mmol), and phenol (9.4 mg, 0.1 mmol). To this was added toluene (2 mL)
and then the flask was sealed with the stopcock and heated to 100°C for one hour. After
one hour, the solvent was removed ,via high vacuum and left under high vacuum at 100°C
for 0.5 hours yielding the B1 VAPOL-boron catalyst (~10:1 Bl/B2 as measured by 1H
NMR). After cooling the stopcock was removed and replaced with a rubber septum. The
catalyst was dissolved by the injection via syringe of 2.0 mL CH2C12/toluene (in two
portions). The catalyst solution was transferred by syringe to a flask containing the imine
prepared as immediately below.
To a flame dried, argon purged homemade flask with a cold addition coil (see
Figure 2.4) equipped with a stir bar Was added the imine 150 (0.271 g, 1.0 mmol). The

flask was topped with two rubber septa and the VAPOL—boron catalyst was added in two

345

1.0 mL portions of a 1:1 ratio of CH2C12/toluene directly to the bottom of the flask by a
syringe equipped with a long needle. This was allowed to stir for 5-10 minutes at room
temperature and then cooled to —45 °C. Meanwhile, in a separate flame dried 5 mL round
bottom flask purged with argon was added Danishefsky’s diene (31) (0.38 mL, 2.0 mmol)
and CH2C12/toluene (1:1) (3.0 mL). The diene was taken up in a syringe and added over
3.0 hours via syringe pump through the cold addition coil. The reaction was then allowed
to stir at —45 °C for the duration of the reaction (24 total hours). After completion of the
reaction, saturated sodium bicarbonate (~20 mL) was added to the reaction flask at —45
°C. This was then transferred to a separatory funnel and extracted with three or four 30-
40 mL portions of CH2C12. The combined organic layers were placed in a 250 mL round
bottom flask and the solvent was then removed via rotary evaporation. The flask was
then equipped with a stir bar and cooled in an ice bath. To this was then added a
previously cooled (0 °C) 20:1 mixture of THF and 1N HCl (50 mL) at which time the
flask was removed from the ice bath and allowed to stir (monitored by TLC) until the
undesired spots close to the desired product disappeared (usually less than one hour).
This was then transferred to a separatory funnel containing distilled water (75-100 mL)
followed by extraction of the crude product with four 50 mL portions of CH2C12. The
combined organic layers were then dried over magnesium sulfate, filtered, and solvent
was removed via rotary evaporation. The product was then purified via flash column
. chromatography (36 cm x 2 cm) to give 117.5 mg (35% yield) of the desired product 151
with 30% ee. I

Entry 2: The protocol as described in entry 1 was followed exactly, except the B2

catalyst was used and it was prepared in the following fashion. To a flame dried, argon

346

purged single-necked flask that had its 14/20 joint replaced with a threaded Teflon’high-
vacuum T—shaped stop-cock equipped with a stir bar was added B(OPh)3 (116 mg, 0.4
mmol), (S)-VAPOL (54 mg, 0.1 mmol), and distilled water (1.8 uL, 0.1 mmol). To this
was added toluene (2 mL) and then the flask was sealed with the stopcock and heated to
80°C for one hour. After one hour, the solvent was removed via high vacuum and left
under high vacuum at 80°C for 0.5 hours yielding the B2 VAPOL-boron catalyst (~19:1
B2/Bl by ). The workup was accomplished following the exact procedure from entry 1
and purification using flash column chromatography gave 277.7 mg (82% yield) of the
desired product 151 with 71% ee. _

Entry 3: The protocol as described in entry 1 was followed exactly, except the B2
catalyst was used and it was prepared in the following fashion. To a flame dried, argon
purged single-necked flask that had its 14/20 joint replaced with a threaded Teflon high-
vacuum T-shaped stop-cock equipped with a stir bar was added BH3-SMe2 (0.2 mmol, 0.5
M solution in toluene), (S)-VAPOL (54 mg, 0.1 mmol), phenol (9.4 mg, 0.1 mmol), and
distilled water (1.8 iiL, 0.1 mmol). To this was added toluene (2 mL) and then the flask
Was sealed with the stopcock and heated to 100°C for one hour. After one hour, the
solvent was removed via high vacuum and left under high vacuum at 100°C for 0.5 hours
yielding the B2 VAPOL-boron catalyst. The workup was accomplished fellowing the
exact procedure from entry 1 and purification using flash column chromatography gave

106 mg (31% yield) of the desired product 151 with 75% ee.

347

 

(R)-BINOL B(OPh)3 (1.0 equiv.)
2.0 equiv. CH2C12, '1’ 4A MS

 

> BLAH-Catalyst

 

 

 

 

0
1/ )
i i O/SK BLAH-Catalyst |
\ + =
Ph N Ph MQMCCHZCb/tolueneflzl) Ph N
613 R = H 31 —-45°,24h , RA Pb
= 2.0 ' .
150 R Ph ( equ1v) 62 R=H
151 R=Ph

Procedure for Scheme 6.4 (aza-Diels-Alder attempt using the BLAH-Catalyst):

Entry I : A 25 mL round bottom flask equipped with a magnetic stir bar was flame
dried and purged with argon. The flask was then put in a glove box and to it was added
B(OPh)3 (0.291 g, 1.0 mmol, purified by distillation), (R)-BINOL (0.573 g, 2.0 mmol),
and the imine 61a (0.195 g, 1.0 mmol). The flask was then taken out of the glove box
and to it was then added a 1:1 ratio of CH2C12/toluene (4.0 mL) and was allowed to stir
for 15 minutes at room temperature and then cooled to —45 °C. Meanwhile, in a separate
flame dried 5 or 10 mL round bottom flask purged with argon was added Danishefsky’s
diene (31) (0.38 mL, 2.0 mmol) and CH2C12/toluene (1:1) (3.0 mL). The diene was taken
up in a syringe and added over 3.0 hours via syringe pump. The reaction was allowed to
stir at -45 °C for an additional 21 hours. After completion of the reaction, saturated
sodium bicarbonate (~20 mL) was added to the reaction flask at —45 °C. This was then
transferred to a separatory funnel and extracted with three or four 3040 mL portions of
CH2C12. The combined organic layers were placed'in a 250 mL round bottom flask and
the solvent was then removed via rotary evaporation. The flask was then equipped with a
stir bar and cooled in an ice bath. To this was then added a previously cooled (0 °C) 20:1

mixture of T HF and 1N HCl (50 mL) at which time the flask was removed from the ice

348

bath and allowed to stir (monitored by TLC) until the undesired spots close to the desired
product disappeared (usually less than one hour). This was then transferred to a
separatory funnel containing distilled water (75-100 mL) followed by extraction of“ the
crude product with four 50 mL portions of CH2C12. The organic layers were combined,
dried over magnesium sulfate, filtered, and the solvent was removed under reduced
pressure. Purification was accomplished via flash column chromatography (36 cm x 2
cm) to give 197.3 mg of the desired product 62 (75% yield). The enantiomeric excess
was determined to be 80—88% (riot baseline separated) by chiral HPLC analysis with the
aid of an authentic sample of the racemic product.

Entry 2: The protocol as described for entry 1 was followed exactly except 0.271
g (1.0 mmol) of the benzhydryl imine 150 was used rather than the imine 61a. The
reaction gave 311.0 mg (92% yield) of the desired product 151 and 89% enantiomeric
excess was measured.

Entry 3: The protocol as described for entry 2 was followed exactly except 29.1
mg (0.1 mmol) of B(OPh)3 and 57.3 mg (0.2 mmol) (R)-BINOL (10 mol% BLAH-
catalyst) was premixed with the imine. The reaction gave 135 mg (39% yield) of the
desired product 151 and 90% enantiomeric excess was measured.

Entry 4: The protocol as described for entry 3 was followed exactly except after
the reaction was cooled to —45°C, a previously prepared solution of B(OPh)3 (0.262 g,
0.9 mmol) in CH2C12/toluene (1:1) (0.5 mL) was added to the reaction vessel immediately
before the diene was added. The reaction gave 186 mg (54% yield) of the desired

product 151 and 60% enantiomeric excess was measured.

349

10.

11.

REFERENCES

Diels, O.; Alder, K., Syntheses in the hydroaromatic series. 1. Addition of diene
hydrocarbons. Ann. 1928, 460, 98-122.

Seltzer, S., Mechanism of the Diels-Alder reaction. Advances in Alicyclic
Chemistry 1968, 2, 1-57.

Danishefsky, S.; Kitahara, T., Useful diene for the Diels-Alder reaction. Journal
of the American Chemical Society 1974, 96, (25), 7807.

Kozmin, S. A.; Janey, J. M.; Rawal, V. H., 1-Amino-3-8iloxy-1,3-butadienes:
Highly Reactive Dienes for the Diels-Alder Reaction. Journal of Organic
Chemistry 1999, 64, (9), 3039-3052.

Kozmin, S. A.; Rawal, V. H., Preparation and Diels-Alder Reactivity of l-Amino-

3-siloxy-1,3-butadienes. Journal of Organic Chemistry 1997, 62, (16), 5252-
5253. ’

Savard, 1.; Brassard, P., Regiospecific syntheses of quinones using vinylketene
acetals derived from unsaturated esters. Tetrahedron Letters 1979, (51), 4911-14.

Boger, D. L.; Weinreb, S. M., Hetero Diels-Alder Methodology in Organic
Synthesis. Academic: San Diego 1987.

Carruthers, W., Cycloaddition Reactions in Organic Synthesis. Pergamon: Oxford
1990.

Weinreb, S. M., Comprehensive Organic Synthesis. Pergamon: Oxford 1991, 5,
(Paquette, L. A. Ed.), 401.

Fringuelli, F .; Pierrnatti, O.; Pizzo, F ., Hetero Diels-Alder reactions in aqueous
medium. Targets in Heterocyclic Systems 1997, 1, 57-73.

Waldmann, H., Asymmetric hetero Diels-Alder reactions. Synthesis 1994, (6),
535-51.

350

12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

22.

23.

Jorgensen, K. A., Development and application of catalytic highly
enantioselective hetero-Diels-Alder reactions of aldehydes and ketones. Current
Trends in Organic Synthesis, [Proceedings of the International Conference on
Organic Synthesis], 12th, Venezia, June 28-July 2, 1998 1999, 207-212.

Maruoka, K.; Yamamoto, H., Asymmetric Diels-Alder reactions using chiral
Lewis acid catalysts. Yukagaku 1990, 39, (10), 852-7.

Buonora, P.; Olsen, J. C.; Oh, T., Recent developments in imino Diels-Alder
reactions. Tetrahedron 2001, 57, (29), 6099-6138.

Rowland, G. B.; Rowland, E. B.; Zhang, Q.; Antilla, J. C., Stereoselective Aza-
Diels-Alder reactions. Current Organic Chemistry 2006, 10, (9), 981-1005.

Bogdanowicz-Szwed, K.; Krasodomska, M., Synthesis of six-membered nitrogen
heterocycles in heterodiene cycloaddition reactions. Wiadomosci Chemiczne
1998, 52, (1 1-12), 821-842.

Hattori, K.; Yamamoto, H., Asymmetric aza-Diels-Alder reaction catalyzed by
boron reagent: effect of biphenol and binaphthol ligand. Synlett 1993, (2), 129-30.

Hattori, K.; Yamamoto, H., Asymmetric aza-Diels-Alder reaction: enantio- and
diastereoselective reaction of imine mediate by chiral Lewis acid. Tetrahedron
1993, 49, (9), 1749-60.

Hattori, K.; Yamamoto, H., Asymmetric aza-Diels-Alder reaction mediated by
chiral boron reagent. Journal of Organic Chemistry 1992, 57, (12), 3264-5.

Furman, B.; Frelek, J .; Dziedzic, M.; Kaminska, A., Asymmetric synthesis of
multi-substituted indolizidines by intramolecular addition of allylsilane to chiral
2,3-dihydro-4-pyridones. Polish Journal of Chemistry 2005, 79, (12), 1919-1928.

F urman, B.; Dziedzic, M., Tetrabutylammonium triphenyldifluorosilicate (TBAT)
initiated intramolecular addition of allylsilanes to 2,3-dihydro-4-pyridones. A
novel route for the stereoselective construction of indolizidine systems.
Tetrahedron Letters 2003, 44, (3 5), 6629-6632.

Ishihara, K.; Miyata, M.; Hattori, K.; Tada, T.; Yamamoto, H., A New Chiral
BLA Promoter for Asymmetric Aza Diels-Alder and Aldol-Type Reactions of
Imines. Journal of the American Chemical Society 1994, 116, (23), 10520-4.

Cros, J. P.; Perez-Fuertes, Y.; Thatcher, M. J .; Arimori, S.; Bull, S. D.; James, T.
D., Non-linear effects operate and dynamic ligand exchange occurs when chiral
BINOL-boron Lewis acids are used for asymmetric catalysis. Tetrahedron:
Asymmetry 2003, 14, (14), 1965-1968.

351

24.

25.

26.

27.

28.

29.

30.

31.

32.

33.

Guillarme, S.; Whiting, A., Unexpected temperature, time and solvent effects in
the catalytic asymmetric aza-Diels-Alder reaction of an ethyl glyoxylate-derived
N-aryl imine with Danishefsky's diene catalysed by a BINOL-zinc complex.
Synlett 2004, (4), 71 1-713.

Ostwald, R.; Chavant, P.-Y.; Stadtrnueller, H.; Knochel, P., Catalytic Asymmetric
Addition of Polyfunctional Dialkylzincs to b-Stannylated and b-Silylated
Unsaturated Aldehydes. Journal of Organic Chemistry 1994, 59, (15), 4143-53.

Fonseca, M. H.; Eibler, E.; Zabel, M.; Konig, B., Synthesis of novel nitrogen-
containing ligands for the enantioselective addition of diethylzinc to aldehydes.
Tetrahedron-Asymmetry 2003, 14, (14), 1989-1994.

Kobayashi, S.; Ueno, M.; Saito, S.; Mizuki, Y.; Ishitani, H.; Yamashita, Y., Air-
stable, storable, and highly efficient chiral zirconium catalysts for enantioselective
Mannich-type, aza Diels-Alder, aldol, and hetero Diels-Alder reactions.
Proceedings of the National Academy of Sciences of the United States of America

2004, 101, (15), 5476-81.

Kobayashi, S.; Kusakabe, K.-i.; Komiyama, S.; Ishitani, H., A Switch of
Enantiofacial Selectivities Using Designed Similar Chiral Ligands in Zirconium-
Catalyzed Asymmetric Aza Diels-Alder Reactions. Journal of Organic Chemistry
1999, 64, (12), 4220-4221.

Kobayashi, S.; Kusakabe, K.-i.; Ishitani, H., Chiral Catalyst Optimization Using
Both Solid-Phase and Liquid-Phase Methods in Asymmetric Aza Diels-Alder
Reactions. Organic Letters 2000, 2, (9), 1225-1227.

Kobayashi, S.; Komiyama, S.; Ishitani, H., The first enantioselective aza-Diels-
Alder reactions of imino dienophiles on use of a chiral zirconium catalyst.
Angewandte Chemie, International Edition 1998, 37, (7), 979-981.

Yamashita, Y.; Mizuki, Y.; Kobayashi, S., Catalytic asymmetric aza Diels-Alder
reactions of hydrazones using a chiral zirconium catalyst. Tetrahedron Letters
2005, 46, (11), 1803-1806.

Josephsohn, N. S.; Snapper, M. L.; Hoveyda, A. H., Efficient and Practical Ag-
Catalyzed Cycloadditions between Arylimines and the Danishefsky Diene.
Journal of the American Chemical Society 2003, 125, (14), 4018-4019.

Mancheno, O. G.; Arrayas, R. G.; Carretero, J. C., Chiral Copper Complexes of
Phosphino Sulfenyl Ferrocenes as Efficient Catalysts for Enantioselective Formal
Aza Diels-Alder Reactions of N-Sulfonyl Imines. Journal of the American

Chemical Society 2004, 126, (2), 456-457.

352

34.

35.

36.

37.

38.

39.

40.

41.

42.

43.

44.

Yao, S.; Johannsen, M.; Hazell, R. G.; Jorgensen, K. A., Catalytic
enantioselective aza Diels-Alder reactions of imino dienophiles. Angewandte
Chemie, International Edition 1998, 37, (22), 3121-3124.

Yao, S.; Saaby, S.; Hazell, R. G.; Jorgensen, K. A., Catalytic enantioselective aza-
Diels - Alder reactions of imines-an approach to optically active nonproteinogenic
a-arnino acids. Chemistry--A European Journal 2000, 6, (13), 2435-2448.

Bromidge, S.; Wilson, P. C.; Whiting, A., A parallel combinatorial approach to
locating homochiral Lewis acid catalysts for the asymmetric aza-Diels-Alder
reaction of an imino dienophile. Tetrahedron Letters 1998, 39, (48), 8905-8908.

Bundu, A.; Guillarme, S.; Harman, J.; Wan, H.; Whiting, A., A parallel
combinatorial approach to locating homochiral Lewis acid catalysts for the
asymmetric aza-Diels-Alder reaction of an imino dienophile. [Erratum to
document cited in CA130:095462]. Tetrahedron Letters 2003, 44, (42), 7849-
7850.

Akiyarna, T.; Tamura, Y.; Itoh, J .; Morita, H.; Fuchibe, K., Enantioselective aza-
Diels-Alder reaction catalyzed by a chiral Bronsted acid: Effect of the additive on
the enantioselectivity. Synlett 2006, (1), 141-143.

Brunel, J. M., BIN CL: A Versatile Chiral Reagent. [Erratum to document cited in
CA1421373277]. Chemical Reviews (Washington, DC, United States) 2005, 105,
(11), 4233.

Brunel, J. M., BINOL: A Versatile Chiral Reagent. Chemical Reviews
(Washington, DC, United States) 2005, 105, (3), 857-897.

Yu, S.; Rabalakos, C.; Mitchell, W. D.; Wulff, W. D., New synthesis of vaulted
biaryl ligands via the Snieckus phenol synthesis. Organic Letters 2005, 7, (3),
367-369.

Zhang, Y.; Yeung, S.-M.; Wu, H.; Heller, D. P.; Wu, C.; Wulff, W. D., Highly
Enantioselective Deracemization of Linear and Vaulted Biaryl Ligands. Organic
Letters 2003, 5, (11), 1813-1816.

Wulff, W. D.; Yu, S.; Antilla, J. Synthesis of VAPOL ligands. 2000-US6412
2000056691, 20000317., 2000.

Bao, J.; Wulff, W. D.; Dominy, J. B.; Fumo, M. J.; Grant, E. B.; Rob, A. C.;
Whitcomb, M. C.; Yeung, S.-M.; Ostrander, R. L.; Rheingold, A. L., Synthesis,
Resolution, and Determination of Absolute Configuration of a Vaulted 2,2'-
Binaphthol and a Vaulted 3,3'-Biphenanthrol (VAPOL). Journal of the American
Chemical Society 1996, 118, (14), 3392-405.

353

45.

46.

47.

48.

49.

50.

51.

52.

53.

54.

55.

Yu, S.; Rabalakos, C.; Mitchell William, D.; Wulff William, D., New synthesis of
vaulted biaryl ligands via the Snieckus phenol synthesis. Org Lett FIELD Full
Journal T itlesOrganic letters 2005, 7, (3), 367-9.

Zhang, Y.; Yeung, S.-M.; Wu, H.; Heller Douglas, P.; Wu, C.; Wulff William, D.,
Highly enantioselective deracemization of linear and vaulted biaryl ligands. Org
Lett FIELD Full Journal TitlesOrganic letters 2003, 5, (11), 1813-6.

Heller, D. P.; Goldberg, D. R.; Wu, H.; Wulff, W. D., An examination of
VANOL, VAPOL, and VAPOL derivatives as ligands for asymmetric catalytic
Diels-Alder reactions. Canadian Journal of Chemistry 2006, 84, (10), 1487-1503.

Heller, D. P.; Goldberg, D. R.; Wulff, W. D., Positive cooperativity of product
mimics in the asymmetric autoinduction of Diels-Alder reactions catalyzed by a
VAPOL-aluminum catalyst. Journal of the American Chemical Society 1997, 119,
(43), 10551-10552.

Bao, J.; Wulff, W. D.; Rheingold, A. L., Vaulted biaryls as chiral ligands for
asymmetric catalytic Diels-Alder reactions. Journal of the American Chemical
Society 1993, 115, (9), 3814-15.

Xue, S.; Yu, S.; Deng, Y.; Wulff, W. D., Active site design in a chemzyme:
development of a highly asymmetric and remarkably temperature-independent
catalyst for the imino aldol reaction. Angewandte Chemie, International Edition
2001, 40, (12), 2271-2274.

Bolm, C.; Frison, J.-C.; Zhang, Y.; Wulff, W. D., Vaulted biaryls: Efficient
ligands for the aluminum-catalyzed asymmetric Baeyer-Villiger reaction. Synlett
2004, (9), 1619-1621.

Loncaric, C.; Wulff, W. D., An efficient synthesis of (-)-chloramphenicol via
asymmetric catalytic aziridination: a comparison of catalysts prepared from
triphenylborate and various linear and vaulted biaryls. Organic Letters 2001, 3,
(23), 3675-3678.

Antilla, J .; Wulff, W. D. Catalytic asymmetric synthesis of chiral aziridines.
2000-528206625 8960, 20000317., 2001.

Wulff, W. D.; Xue, S. Transition metal catalyst and process for producing
optically active beta-amino acids and esters by reacting imines with heteroketene
acetals. 2000-US641 12000056448, 20000317., 2000.

Wulff, W. D.; Antilla, J. Catalytic asymmetric synthesis of chiral aziridines.
2000-US69782000056708, 20000317., 2000.

354

56.

57.

58.

59.

60.

61.

62.

63.

64.

65.

66.

67.

68.

Antilla, J. C.; Wulff, W. D., Catalytic asymmetric aziridination with arylborate
catalysts derived from VAPOL and VANOL ligands. Angewandte Chemie,
International Edition 2000, 39, (24), 4518-4521.

Antilla, J. C.; Wulff, W. D., Catalytic Asymmetric Aziridination with a Chiral
VAPOL-Boron Lewis Acid. Journal of the American Chemical Society 1999,
121, (21), 5099-5100.

Wang, S.-W.; Wang, Q.-M.; Huang, R.-Q., Catalysts for hetero Diels-Alder
reaction of imines. Youji Huaxue 2003, 23, (10), 1064-1075.

Kobayashi, S., Catalytic enantioselective aza Diels-Alder reactions. C ycloaddition
Reactions in Organic Synthesis 2002, 187-209.

Waldmann, H., Asymmetric Aza-Diels-Alder reactions. Organic Synthesis
Highlights 11 1995, 37-47.

Waldmann, H., Asymmetrical aza-Diels-Alder reactions. Nachrichten aus
Chemie, Technik und Laboratorium 1992, 40, (12), 1377-8, 1380—4.

Yu, C. M.; Choi, H. S.; Yoon, S. K.; Jung, W. H., Effects of subjoin Lewis acid
on the catalytic asymmetric allylic transfer reactions of aldehydes promoted by
BINOL-Ti(IV) complex. Synlett 1997, (8), 889-890.

Vogl, E. M.; Groger, H.; Shibasaki, M., Towards perfect asymmetric catalysis:
Additives and cocatalysts. Angewandte Chemie-International Edition 1999, 38,
(11), 1570-1577.

Shintani, R.; Tokunaga, N.; Doi, H.; Hayashi, T., A new entry of nucleophiles in
hodium-catalyzed asymmetric 1,4-addition reactions: addition of organozinc

reagents for the synthesis of 2-aryl-4-piperidones. Journal of the American
Chemical Society 2004, 126, (20), 6240-6241.

Olah, G. A.; Arvanaghi, M., Aldehydes by Forrnylation of Grignard or Organo-
Lithium Reagents with N-Forrnylpiperidine. Angewandte Chemie-International
Edition in English 1981, 20, (10), 878-879.

Fischer, G; Carreira, E. M., Mg12 an additive in lr(I)-catalyzed addition of
silylacetylenes to imines: Expeditions synthesis of propargylic amines. Synthesis-
Stuttgart 2004, (9), 1497-1503.

Sun, Y. H.; Wan, X. B.; Wang, J. P.; Meng, Q. H.; Zhang, H. W.; Jiang, L. J.;
Zhang, Z. G., Ru-catalyzed asymmetric hydrogenation of alpha-ketoesters with
CeCl3 center dot 7H(2)O as additive. Organic Letters 2005, 7, (24), 5425-5427.

James, T. D. S., S., Top. Curr. Chem. 2002, 218, 159.

355

69.

70.

71.

72.

73.

74.

75.

76.

77.

78.

79.

Li, C.; Blackman, A. J ., Cylindricines C-G, perhydropyrrolo[2,1-j]quinolin-7-one
alkaloids from the ascidian Clavelina cylindrica. Australian Journal of Chemistry
1994, 47, (7), 1355-61. ‘

Verbist, J. F., Ascidians, an example of the potential value of marine organisms as
sources of substances with pharmacological activity. Journal de Pharmacie de
Belgique 1995, 50, (2-3), 98-120.

Blackman, A. J.; Li, C.; Hockless, D. C. R.; Skelton, B. W.; White, A. H.,
Cylindricines A and B, novel alkaloids from the ascidian Clavelina cylindrica.
Tetrahedron 1993, 49, (38), 8645-56.

Li, C.; Blackman, A. J., Cylindricines H-K, novel alkaloids from the ascidian
Clavelina cylindrica. Australian Journal of Chemistry 1995, 48, (5), 955-65.

Patil, A. D.; Freyer, A. J.; Reichwein, R.; Carte, B.; Killmer, L. B.; Faucette, L.;
Johnson, R. K., F ascicularin, a novel tricyclic alkaloid from the ascidian Nephteis

fascicularis with selective activity against a DNA repair-deficient organism.
Tetrahedron Letters 1997, 38, (3), 363-364.

Biard, J. F .; Guyot, S.; Roussakis, C.; Verbist, J. F.; Vercauteren, J .; Weber, J. F.;
Boukef, K., Lepadiformine, a new marine cytotoxic alkaloid from Clavelina
lepadiformis Mueller. Tetrahedron Letters 1994, 35, (17), 2691-4.

Blackman, A. J.; Li, C. P.; Hockless, D. C. R.; Skelton, B. W.; White, A. H.,
Cylindricine-a and Cylindricine-B, Novel Alkaloids from the Ascidian Clavelina-
Cylindrica. Tetrahedron 1993, 49, (38), 8645-8656.

Dutta, S.; Abe, H.; Aoyagi, S.; Kibayashi, C.; Gates, K. 8., DNA Damage by
Fasicularin. Journal of the American Chemical Society 2005, 127, (43), 15004-
15005.

Sauviat, M.-P.; Vercauteren, J.; Grimaud, N.; Juge, M.; Nabil, M.; Petit, J.-Y.;
Biard, J .-F., Sensitivity of Cardiac Background Inward Rectifying K+ Outward
Current (1K1) to the Alkaloids Lepadiformines A, B, and C. Journal of Natural
Products 2006, 69, (4), 558-562.

Juge, M.; Grimaud, N.; Biard, J. F.; Sauviat, M. P.; Nabil, M.; Verbist, J. F.; Petit,
J. Y., Cardiovascular effects of lepadiforrnine, an alkaloid isolated fiom the
ascidians Clavelina lepadiformis (Muller) and C. moluccensis (Sluiter). T oxicon
FIELD Full Journal Title: Toxicon : oflicial journal of the International Society
on T oxinology 2001, 39, (8), 1231-7.

Molander, G. A.; Roenn, M., Total Synthesis of (-)-Cy1indricine C. Journal of
Organic Chemistry 1999, 64, (14), 5183-5187.

356

80.

81.

82.

83.

84.

85.

86.

87.

88.

89.

Trost, B. M.; Rudd, M. T., Chemoselectivity of the Ruthenium-Catalyzed
Hydrative Diyne Cyclization: Total Synthesis of (+)-Cylindricine C, D, and E.
Organic Letters 2003, 5, (24), 4599-4602.

Kibayashi, C., Development of new synthetic methods and its application to total
synthesis of nitrogen-containing bioactive natural products. Chemical &
Pharmaceutical Bulletin 2005, 53, (11), 1375-1386.

Abe, H.; Aoyagi, S.; Kibayashi, C., Total Synthesis of the Tricyclic Marine
Alkaloids (-)-Lepadiformine, (+)-Cylindricine C, and (-)-Fasicularin via a
Common Intermediate Formed by Forrnic Acid-Induced Intramolecular

Conjugate Azaspirocyclization. Journal of the American Chemical Society 2005,
127, (5), 1473-1480.

Arai, T.; Abe, H.; Aoyagi, S.; Kibayashi, C., Total synthesis of (+)-cylindricine C.
Tetrahedron Letters 2004, 45, (30), 5921-5924.

Ciufolini, M. A.; Canesi, S.; Ousmer, M.; Braun, N. A., Synthetic ventures
inspired by biosynthetic hypotheses: the evolution of a method for the oxidative
amidation of phenols. Tetrahedron 2006, 62, (22), 5318-5337.

Canesi, S.; Bouchu, D.; Ciufolini Marco, A., Fully stereocontrolled total
syntheses of (-)-cylindricine C and (-)-2-epicylindricine C: a departure in
sulfonamide chemistry. Angew Chem Int Ed Engl FIELD Full Journal
Title:Angewandte Chemie (International ed. in English) 2004, 43, (3 3), 4336-8.

Swidorski, J. J.; Wang, J.; Hsung, R. P., A Concise Total Synthesis of (-)-
Cylindricine C through a Stereoselective Intramolecular Aza-[3 + 3] Annulation
Strategy. Organic Letters 2006, 8, (4), 777-780.

Liu, J.; Swidorski, J. J.; Peters, S. D.; Hsung, R. P., An N-Acyliminium
Cyclization Approach to a Total Synthesis of (+)-Cylindricine C. Journal of
Organic Chemistry 2005, 70, (10), 3898-3902.

Liu, J.; Hsung, R. P.; Peters, S. D., Total Syntheses of (+)-Cylindricines C-E and
(-)-Lepadiformine through a Common Intermediate Derived from an aza-Prins
Cyclization and Wharton's Rearrangement. Organic Letters 2004, 6, (22), 3989-
3992.

Shibuguchi, T.; Mihara, H.; Kuramochi, A.; Sakuraba, S.; Ohshirna, T.;
Shibasaki, M., Short synthesis of (+)-cylindricine C by using a catalytic
asymmetric Michael reaction with a two-center organocatalyst. Angewandte
Chemie, International Edition 2006, 45, (28), 463 5-4637.

357

90.

91.

92.

93.

94.

95.

96.

97.

98.

99.

100.

Sestelo, J. P.; del Mar Real, M.; Sarandeses, L. A., Synthesis of Polycyclic
Structures by the Diels-Alder Reaction of Inner-Outer-Ring 1,3-
Bis(trimethy1silyloxy)dienes. Journal of Organic Chemistry 2001, 66, (4), 1395-
1402.

Langer, P.; Schneider, T.; Stoll, M., Domino reaction of 1,3-
bis(trimethylsilyloxy)-l,3-dienes with oxalyl chloride: general and stereoselective
synthesis of g-alkylidenebutenolides. ChemistrynA European Journal 2000, 6,
(17), 3204-3214.

Langer, P.; Schneider, T., Regioselective synthesis and acylation of cyclic
bis(trimethylsiloxy)-l,3-dienes - new and versatile 1,3-dianion synthons. Synlett
2000, (4), 497-500.

Sestelo, J. P.; Del Mar Real, M.; Mourino, A.; Sarandeses, L. A., Synthesis of
polycyclic structures by Diels-Alder reaction of inner-outer-ring 1,3-
bis[(trimethylsilyl)oxy]dienes. Tetrahedron Letters 1999, 40, (5), 985-988.

Bos, M. E.; Wulff, W. D.; Wilson, K. J., A new approach to hydrindenones by
tautomer-arrested annulations of Fischer carbene complexes. Chemical
Communications (Cambridge) 1996, (16), 1863-1864.

Hansson, L.; Carlson, R., Bis-siloxydienes from diketones. Acta Chemica
Scandinavica 1989, 43, (3), 304-6.

Babot, 0.; Cazeau, P.; Duboudin, F ., Use of iodotrimethylsilane, prepared in situ,
for the preparation of 1,3-bis(trimefl1ylsiloxy)-l,3-dienes. Journal of
Organometallic Chemistry 1987, 326, (2), C57-C60.

Kraegeloh, K.; Simchen, G.; Schweiker, K., Reactions of trialkylsilyl
trifluoromethanesulfonates. 111. Synthesis of 1,3-bis(tn'methylsiloxy)-l,3-dienes

and 3-trimethylsiloxy-2-butenoates silylated in position 4. Liebigs Annalen der
Chemie 1985, (12), 2352-62.

Neipp, C. E.; Martin, S. F ., Synthesis of Bridged Azabicyclic Structures via Ring-
Closing Olefin Metathesis. Journal of Organic Chemistry 2003, 68, (23), 8867-
8878.

Greshock, T. J .; Funk, R. L., Total Synthesis of (+-)-Lepadiformine via an
Amidoacrolein Cycloaddition. Organic Letters 2001, 3, (22), 3511-3514.

Lee, G. H.; Choi, E. B.; Lee, E.; Pak, C. S., An Efficient Desulfonylation Method

Mediated by Magnesium in Ethanol. Tetrahedron Letters 1993, 34, (28), 4541-
4542.

358

101.

102.

103.

104.

105.

106.

107.

108.

109.

110.

111.

112.

De Risi, C.; Pollini, G. P.; Veronese, A. C.; Bertolasi, V., A new simple route for
the synthesis of (+-)-2-azetidinones starting from b-enaminoketoesters.
Tetrahedron 2001, 57, (51), 10155-10161.

Chianelli, D.; Kim, Y.-C.; Lvovskiy, D.; Webb, T. R., Application of a novel
design paradigm to generate general nonpeptide combinatorial scaffolds

mimicking beta turns: synthesis of ligands for somatostatin receptors. Bioorganic
& Medicinal Chemistry 2003, 11, (23), 5059-5068.

Pfrengle, W.; Kunz, H., Hetero-Diels-Alder reactions on a carbohydrate template:
stereoselective synthesis of (S)-anabasin. Journal of Organic Chemistry 1989, 54,
(18), 4261-3.

Kunz, H.; Weymann, M.; Follmann, M.; Allef, P.; Oertel, K.; Schultz-Kukula,
M.; Hofmeister, A., Stereoselective syntheses of chiral heterocycles and alkaloids
using carbohydrate auxiliaries. Polish Journal of Chemistry 1999, 73, (1), 15-27.

Kunz, H.; Mueller, B.; Pfrengle, W.; Rueck, K.; Staehle, W., Carbohydrates as
chiral templates in stereoselective [4 + 2] cycloaddition reactions. ACS
Symposium Series 1992, 494, (Cycloaddit. React. Carbohydr. Chem), 131-46.

Kunz, H., Lewis acid—catalyzed stereoselection on carbohydrate templates. NATO
ASI Series, Series C: Mathematical and Physical Sciences 1989, 289, (Sel. Lewis
Acid Promoted React), 189-202.

Du, H. F.; Long, J .; Hu, J. Y.; Li, X.; Ding, K. L., 3,3 '-Br-2-BINOL-Zn complex:
A highly efficient catalyst for the enantioselective hetero-Diels-Alder reaction.
Organic Letters 2002, 4, (24), 4349-4352.

Armesto, D. H., W. M.; Perez-Ossorio, R.; Ramos, A. , Journal of Organic
Chemistry 1987, 52, (15), 3378-3381.

Dugat, D. J ., G.; Sahoo, 8., Canadian Journal of Chemistry 1987, 65, (1), 88-93.

Lautens, M.; Tayama, E.; Nguyen, D., Direct Vinylogous Mannich-type reactions
via ring opening and rearrangement of vinyloxiranes. Organic Letters 2004, 6,
(3), 345-347.

Olah, G. A. A., M., Aldehydes by Forrnylation of Grignaard ro Organolithium
Reagents with N-Forrnylpiperidine. Angewandte Chemie, International Edition
1981, 20, (10), 878-879.

Francesch, A.; Alvarez, R.; Lopez, S.; de Lera, A. R., Synthesis of Retinals

Fluorinated at Odd-Numbered Side-Chain Positions and of the Corresponding
Fluorobacteriorhodopsins. Journal of Organic Chemistry 1997, 62, (2), 310-319.

359

113.

114.

115.

116.

117.

118.

119.

120.

121.

122.

123.

124.

Plater, M. J.; Aiken, S.; Bourhill, G., Metallated porphyrins containing lead(II),
copper(II) or zinc(II). Tetrahedron 2002, 58, (12), 2415-2422.

Zweifel, G. L., W., Stereoselective Syntheses of ((E)- and (Z)—l-Halo-l-
alkenyl)silanes from Alkynes. Journal of Organic Chemistry 1978, 43, (14),
2739-2744.

Miller, B. R. M., G., Stereoselective Trisubstituted Olefin Synthesis via
Vinylsilanes. Journal of Organic Chemistry 1979, 44, (25), 4623-4633.

Fielding, L., Determination of Association Constants (Ka) from Solution NMR
Data. Tetrahedron 2000, 56, (34), 6151-6170.

Sogah, G. D. Y.; Cram, D. J., Host-guest complexation. 14. Host covalently
bound to polystyrene resin for chromatographic resolution of enantiomers of

amino acid and ester salts. Journal of the American Chemical Society 1979, 101,
(11), 3035-42.

Ishitani, H.; Ueno, M.; Kobayashi, S., Enantioselective Mannich-Type Reactions
Using a Novel Chiral Zirconium Catalyst for the Synthesis of Optically Active b-
Amino Acid Derivatives. Journal of the American Chemical Society 2000, 122,
(34), 8180-8186.

Yamashita, Y.; Ishitani, H.; Shimizu, H.; Kobayashi, 8., Highly anti-Selective
Asymmetric Aldol Reactions Using Chiral Zirconium Catalysts. Improvement of
Activities, Structure of the Novel Zirconium Complexes, and Effect of a Small
Amount of Water for the Preparation of the Catalysts. Journal of the American
Chemical Society 2002, 124, (13), 3292-3302.

Wu, P.-L.; Peng, S.-Y.; Magrath, J ., 1-Acyl-2-alkylhydrazines by the reduction of
acylhydrazones. Synthesis 1995, (4), 435-8.

Liu, J. F.; Heathcock, C. H., Total Synthesis of (+-)-Cylindricines A and B.
Journal of Organic Chemistry 1999, 64, (22), 8263-8266.

Lipshutz, B. H. K., J. A.; Parker, D. A.; Nguyen, S. L.; McCarthy, K. B, More
highly mixed, higher order cyanocuprates "Rt(2-thienyl)Cu(CN)Li2." Efficient
reagents which promote selective ligand transfer. Journal of Organometallic
Chemistry 1985, 285, 437-447.

Klegraf, E.; Follmann, M.; Schollmeyer, D.; Kunz, H., Stereoselective Synthesis
of Enantiomerically Pure Piperidine Derivatives by N-Galactosylation of

Pyridones. European Journal of Organic Chemistry 2004, 3346-3360.

Weymann, M.; Schultz-Kukula, M.; Kunz, H., Auxiliary-controlled
stereoselective enolate protonation: Enantioselective synthesis of cis and trans

360

125.

126.

127.

128.

annulated decahydroquinoline alkaloids. Tetrahedron Letters 1998, 39, (43),
7835-7838.

Matsuzawa, S.; Horiguchi, Y.; Nakamura, E.; Kuwajima, I., Chlorosilane-
Accelerated Conjugate Addition of Catalytic and Stoichiometric Organocopper
Reagents. Tetrahedron 1989, 45, (2), 349-362.

Lipshutz, B. H.; Ellsworth, E. L.; Siahaan, T. J., The Role of Bf3.Et20 in
Reactions of Lower Order (Gilman) Organocuprates. Journal of the American
Chemical Society 1989, 111, (4), 1351-1358.

Yamamoto, Y., Selective Synthesis by Use of Lewis-Acids in the Presence of
Organocopper and Related Reagents. Angewandte Chemie-International Edition
in English 1986, 25, (11), 947-959.

Scheiper, B.; Bonnekessel, M.; Krause, H.; Fuerstner, A., Selective Iron-
Catalyzed Cross-Coupling Reactions of Grignard Reagents with Enol Triflates,
Acid Chlorides, and Dichloroarenes. Journal of Organic Chemistry 2004, 69,
(11), 3943-3949.

361

 

 

 

   

111;"11‘1‘1Ijiilfllljlj'