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.

 

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

6/01 c:/CIRCIDateDuo.p65-p. 15

 

 

THE USE OF EXCITON COUPLED CIRCULAR DICHROISM FOR THE
ABSOLUTE STEREOCHEMICAL DETERMINATION OF CHIRAL CARBOXYLIC
ACIDS

By

Qifei Yang

A DISSERTATION

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

DOCTOR OF PI-HLOSOPHY
Department of Chemistry

2004

ABSTRACT

THE USE OF EXCITON COUPLED CIRCULAR DICHROISM FOR THE
ABSOLUTE STEREOCHEMICAL DETERMINATION OF CHIRAL CARBOXYLIC
ACIDS

By

Qifei Yang

The Exciton Coupled Circular Dichroic (ECCD) method, a non-empirical method
which allows for the determination of the absolute configuration of chiral compounds on
a micro-scale, requires the through space coupling of two chromophores on the chiral
molecule. This fact limits the use of this method, since the majority of chiral compounds
do not fulfill this requirement. To overcome this problem, chromophoric hosts are
introduced in the system. As reported before, for the chiral compounds with two sites of
attachment, a pentanediol linked bis metal porphyrin tweezer could serve as the
chromophoric host for the determination of the absolute stereochemistry. In the case of
chiral compounds with only one site of attachment, achiral “carrier” molecules have been
designed, which upon derivatization with chiral molecules are capable of binding with his
metal porphyrin to form a 1:1 host/guest complex. The research presented herein is
focused on the absolute stereochemical determination of a-chiral carboxylic acids. A
series of carrier molecules have been designed, synthesized and derivatized with a-chiral
carboxylic acids, and the chiral conjugates were added to zinc porphyrin tweezer in

methylcylcohexane to measure their CD. The derivatives with, 1,4-diaminobenzene,

exhibit consistent ECCD spectra, and a mnemonic has been proposed to correlate the
observed ECCD sign to the absolute stereochemistry of oc—chiral carboxylic acids.

When the same carrier molecule is derivatized with or-chiral halocarboxylic acids,
the complexes exhibit consistent ECCD spectra, however, opposite to the predicted sign
based on the previous mnemonic. Based on spectroscopic studies, we can explain this
behavior by suggesting a novel cooperative binding of the a—halogen and carbonyl
oxygen to zinc porphyrin, which only occurs in the case of a—haloamides. This bi-
dentate binding leads to a new paradigm of porphyrin binding. In order to prove the
above suggestion, a series of (Jr-chiral butyrolactams were synthesized in order to restrict
the free rotation of the Cchim-Cczo bond. The spectroscopic study of the new compounds
support our suggestion for the most part.

We were also engaged in the synthesis of mono-substituted porphyrin on
insoluble and soluble polymers. The synthesis of porphyrins on polymeric backbone
provides mono-substituted porphyrins with comparable yields as to the corresponding
reaction in solutions, without the troublesome contamination from multiple-substituted
products.

Finally, we were interested in designing artificial carbohydrate receptors, which
would show selectivity for a specific carbohydrate. For this purpose, we designed

boronic porphyrin tweezers and several attempts towards their syntheses were carried out.

Dedicated to my dearest parents and my husband for their love and support.

iv

ACKNOWLEDGEMENT

Although I have a great feeling of accomplishment upon finishing my thesis, there
is a stronger feeling of gratitude for all the people who help me reach this point and who
helped to make the experience such an intellectually fulfilling and enjoyable one.

I would first like to express my utmost gratitude to my research advisor, Professor
Babak Borhan. I have sincerely appreciated his encouragement and his strong support
over my graduate career. I will always be grateful for the excellent education I received
under his direction.

Not only I learned from Professor Babak Borhan during graduate school, but also
from his research group. I will miss the daily interaction with such an interesting and
helpful group of people. I want to give special thanks to Chrysoula Vasileiou for her help
and patience. She helped me a lot in writing my thesis and she deserves a medal for
correcting my thesis and offering great suggestions. A special mention to my good
friend, a former lab member Dr. Meenakshi Ramarathnam for being such a warm hearted
person and for showing incredible patience with me and the mess I was making in the lab
while helping me a lot with my research. I would also like to thank Courtney Olmsted
and Marina Tanasova for working together with me. Through our valuable discussions I
achieved a better understanding of my research. Thanks especially to the people who
helped me so much during my first two years: Dr. Gao Liu, Dr. Chun Wang, Dr. Baoyu
Mi, Dr. Chen—yu Yeh and Dr. Jian-yang Cho.

I also want to thank all the lab members that I have ever worked with and all the

friends that I made in Michigan State University. When I came to East Lansing six years

ago, this was a brand new place for me, without any family member and friends. Now
six years passed by, I have met a lot of people, and they all are very nice to me and have
made me feel very comfortable in studying and living in America. In my mind, Babak’s
lab is the best lab anyone can ever have. We cared about each other and helped each
other in every way, just like a big family, and I am very thankful for this. I will miss all
the other past and present group members: Jennifer Schomaker, Jun Yan, Montserrat
Rabago-Smith, Rachael Crist, Radha Narayan, Tao Zheng, Ben Travis and Kristina
Parkanzky. I would like to thank my dear friend Jie Yu and Dr. Shi Jin for their
friendship.

Dr. Rui-Huang deserves a lot of thanks for patiently teaching me to use mass
spectrometry, and taking a lot of solid probe MS for me. I also should thank the staff of
MSU MS facility for taking all the FAB-MS for me.

Finally, I would like to thank my family, my dearest parents and husband, for
their unconditional love and support. I would not made it to or through graduate school
without them. Thank you all for all you have done and all you have sacrificed for me

over the years.

vi

TABLE OF CONTENTS

List of Figures ........................................................................................ xii
List of Schemes ..................................................................................... xix
List of Tables ....................................................................................... xxv
List of Abbreviations .............................................................................. xxvi

Chapter 1. Introduction

1-1. The Exciton Coupled Circular Dichroism Method (ECCD) ............................ 1
l-l.1. Chirality ....................................................................................... 1
1-1.2. Optical Rotary Dispersion (ORD) and Circular Dichroism (CD) Spectroscopy. ...4
1-2. The Exciton Coupled Circular Dichroic (ECCD) method and its applications ..... 12
1—2. 1. The qualitative explanation of Exciton Coupled Circular Dichroic method
(ECCD) ...................................................................................... 13
1-2.2. The explanation of ECCD theory based on Quantum Mechanical theories ........ 18
l-2.3. The application of ECCD method to determine the absolute stereochemistry of
chiral molecules ............................................................................ 21
1-3. Use of chromophoric host such as porphyrins for ECCD studies ................... 23
l-3.l. Use of chromophoric host .................................................................. 24
1-3.2. Use of porphyrins as chromophoric host ................................................ 26

Chapter 2. Polymer-supported synthesis of mono-substituted porphyrins

2-1.

22.

2-2.1.

2—2.2.

The synthesis of the zinc porphyrin tweezer ............................................ 43

The synthesis of mono-substituted porphyrins on polymers ......................... 48

Synthesis of mono-substituted porphyrin on insoluble polymer ..................... 50

Synthesis of mono-substituted porphyrin with soluble polymer ..................... 52
vii

 

 

Experimental materials and general procedures ................................................. 58

Reference ............................................................................................. 67

Chapter 3. Absolute stereochemical determination of a-chiral carboxylic acids

3-1. Stereochemical determination of a-chiral carboxylic acids with carrier A... . ....:74
3-2. Stereochemical determination of a-chiral carboxylic acids with carrier B ......... 81
3-3. Carrier C for the stereochemical study of chiral carboxylic acids .................... 97
3-4. Carrier D for the stereochemical study of chiral carboxylic acids .................. 104
3-5. Carrier E and F for the stereochemical study of a—chiral carboxylic acids........ 108
3-6. Use of 2-aminopyrazine as the new carrier G ........................................ 116
3-7. Design of a new carrier H for chiral carboxylic acids ................................ 118
3-8. Conclusion ................................................................................. 1 19
3-9. Another approach towards the same goal .............................................. 121
Experimental materials and general procedures ............................................... 124
Reference ........................................................................................... 149

Chapter 4. The stereochemical determination of a-chiral halocarboxylic acids

4-1. Synthesis of tat-chiral halocarboxylic acids derivatives and the ECCD data of
carrier B derivatives ....................................................................... 153

4-2. Explanations for the unexpected ECCD of carrier B derivatized a-chiral

halocarboxylic acids ....................................................................... 159
4-2.1. Determining the rotation of COL-Cc=0 bond by NMR study .......................... 160
4-2.2. The study with carrier B derivative of 2-trifluoromethylpropionic acid .......... 162
4-2.3. The possibility of G*—1t interaction ..................................................... 169

4-2.4. The NMR study to determine the interaction between zinc porphyrin tweezer and
carrier B derivatized a—chiral halocarboxylic acids. ._ ................................ 173

viii

 

4-3. The study of the interaction between the zinc porphyrin tweezer and carrier B
derivatized tit-chiral halocarboxylic acids by forming five-membered 0t-

halolactams ................................................................................ 178
4-3.1. The synthesis of chiral a—substituted—y—butyrolactams ............................... 179
4-3.2. The synthesis of chiral a—alkyl-y—butyrolactams ...................................... 19]
Experimental materials and general procedures ............................................... 210
Reference ........................................................................................... 240

Chapter 5. The chiral aggregation of porphyrin acids
5-1. Determination of extinction coefficient (8) of zinc porphyrin acid ................ 245
5-1.1. Determination of extinction coefficient (8) by Lambert—Beer’s Law ............... 245

5-1.2. Determination of extinction coefficient (6) of zinc porphyrin acid based on that of

zinc porphyrin methyl ester .............................................................. 246
5-2. The zinc porphyrin acid aggregate with carrier F derivatized (S)-0-

acetylmandelic acid 9F ................................................................... 248
5-3. Other chiral substrates for the zinc porphyrin acid aggregates ...................... 254
5-4. Porphyrin aggregates with other porphyrin compounds ............................. 255
Conclusion .......................................................................................... 257
Experimental materials and general procedures ............................................... 259
Reference ........................................................................................... 261

Chapter 6. The attempted synthesis of boronic porphyrin tweezer

6-1. Synthesis of porphyrin tweezer l ........................................................ 266
6-2. Synthesis of boronic porphyrin tweezer 2 ........................................... 270
6-2.1. Synthesis of boronic porphyrin tweezer 2 by MacDonald “2+2” cyclization...270
6-2.2. Synthesis of boronic porphyrin tweezer 2 by Pd catalyzed coupling reaction....284

6—2.3. Hydrolysis of pinacolborane ............................................................. 298

ix

6-3. Transmetallation to introduce boronic acid ............................................ 304

6—4. Synthesis of water soluble boron containing porphyrin 45 .......................... 305

Experimental materials and general procedures ............................................... 310
Reference ........................................................................................... 332
APPENDIX ......................................................................................... 335
Change in proton chemical of compound 21 upon addition to ZnTPP ..................... 336
Change in proton chemical of compound 26 upon addition to ZnTPP ..................... 337
Change in proton chemical of compound 27 upon addition to ZnTPP ..................... 338
Change in proton chemical of compound 28 upon addition to ZnTPP ..................... 339
ECCD spectrum of compound B-l-Br .......................................................... 340
ECCD spectrum of compound B-2-Br .......................................................... 341

ECCD spectrum of compound B-3-Br .......................................................... 342
ECCD spectrum of compound B-4-Br .......................................................... 343
ECCD spectrum of compound B-S-Br .......................................................... 344
ECCD spectrum of compound B-l-Cl ........................................................... 345
ECCD spectrum of compound B-2-Cl ........................................................... 346
ECCD spectrum of compound B-3-C1 ........................................................... 347
ECCD spectrum of compound B-4-Cl ........................................................... 348
ECCD spectrum of compound B-S-Cl ........................................................... 349
ECCD spectrum of compound B-l-F ............................................................ 350
ECCD spectrum of compound B-2-F. ........................................................... 351

ECCD spectrum of compound B-3-F ............................................................ 352

ECCD spectrum of compound B-4-F ............................................................ 353

ECCD spectrum of compound B-S-F ............................................................ 354

ECCD spectrum of compound B-l-I ............................................................. 355
ECCD spectrum of compound 37 ................................................................. 356
ECCD spectrum of compound 38 ................................................................. 357
ECCD spectrum of compound 50 ................................................................. 358
ECCD spectrum of compound 51 ................................................................. 359
ECCD spectrum of compound 52 ................................................................. 360
ECCD spectrum of compound 53 ................................................................. 361
ECCD spectrum of compound 54 ................................................................. 362
ECCD spectrum of compound 55 ................................................................. 363
ECCD spectrum of compound 75 ................................................................. 364
ECCD spectrum of compound 77 ................................................................. 365
ECCD spectrum of compound 78 ................................................................. 366
Addition of L-lysine to zinc porphyrin alcohol 2 .............................................. 367

xi

LIST OF FIGURES

Note: Images in this dissertation are presented in color.

Figure 1-1.
Figure 1-2.
Figure 1-3.
Figure 1-4.
Figure 1-5.

Figure 1-6.

Figure 1-7.

Figure 1-8.

Figure 1-9.

Figure 1-10.

Figure 1-11.

Figure 1-12.

Figure 1-13.

Figure 1-14.

Examples of some chiral molecules and their mirror images ................. 1
The structures of (S)- and (R)-Carvone ............................................ 2
The structures of (R)- and (S)—Thalidomide ...................................... 3
The linearly polarized light and circularly polarized light ..................... 4
(a) Positive and (b) negative CD Cotton effects ................................. 7

The “right hand rule” to determine the direction of magnetic transition
dipole moment m ..................................................................... 8
The transition moments for hexahelicene. Antiparallel moments for (M)
and parallel moments for (P) helicities ............................................ 9
The CD spectropolarimeter ....................................................... 10

Exciton Coupled Circular Dichroism (ECCD) of steroidal 2,3-

bisbenzoate ......................................................................... 12
The qualitative explanation of ECCD based on a dibenzoate ester... ......14
The ECCD couplet for the configuration shown in Figure 11-10b.... . . . . . 17

The screw sense of the chromophores and its corresponding ECCD
sign .................................................................................. 17
Splitting of the excited states of isolated chromophores i and j by exciton
interaction. The energy gap 2Vij is called Davydov splitting ............... 17

The stereochemical determination of abscisic acid by ECCD method. . ...22

xii

Figure 1-15.

Figure 1-16.
Figure 1-17.

Figure 1-18.

Figure 1.19.

Figure 1-20.

Figure 1-21.

Figure 1-22.

Figure 3-1.

Figure 3-2.

Figure 3-3.
Figure 3-4.
Figure 3-5.
Figure 3-6.
Figure 3-7.

Figure 3-8.

 

Examples of exciton Chirality CD method in absolute configurational

assignment .......................................................................... 23
The induced Chirality of biphenol upon binding to chiral diamine ......... 24
Achiral porphyrins and inherently achiral porphyrins ........................ 27
UV-vis spectra of porphyrin methyl ester and zinc porphyrin methyl
ester ................................................................................. 29
Structure of zinc porphyrin tweezer 4 ............................................ 30
Assignment of absolute configuration of acyclic chiral diamines by
Exciton Coupled Circular Dichroism (ECCD) ................................ 31
The chiral complex and its most stable conformation ........................ 34
Use of carrier 5 to determine the absolute stereochemistry of chiral
monoamine by Exciton Coupling Circular Dichroism ........................ 36
Stereochemical determination of chiral monoamine with carrier 1.........70
The absolute stereochemical assignment of (it-chiral carboxylic acids by
NMR ................................................................................. 71
Chiral auxiliary reagents .......................................................... 71
The derivatization of PGME with chiral acids ................................. 71
Two-step derivatization of a—hydroxy carboxylic acids ...................... 72
The structure of 1-carboxyethyl substituted monosaccharides .............. 73
The structures of carrier 1 and carrier A ........................................ 74
Comparison of chiral complexes of carrier 1 and carrier A: carrier 1

complexes are locked in a conformation, while carrier A complexes are

not so rigid ......................................................................... 76

xiii

Figure 3-9.

Figure 3-10.

Figure 3-11.

Figure 3-12.
Figure 3-13.

Figure 3-14.

Figure 3-15.

Figure 3-16.

Figure 3-17.

Figure 3-18.

Figure 3-19.
Figure 3-20.
Figure 3-21.
Figure 3-22.

Figure 3-23.

The ECCD spectra of carrier A derivatized (R)— and (S)-phenoxypropionic
acid with zinc porphyrin tweezer 2 ............................................... 77
Lowest energy conformations of carrier A derivatized complexes based on
modeling without binding to zinc porphyrin ................................... 79
Based on modeling, binding of the Zn porphyrin to carrier A conjugates
forces the rotation of the Caryl-Namide bond ....................................... 80
The possible binding of carrier B derivatives to zinc porphyrin tweezer..83
UV-vis spectra of adding two amides to ZnTPP .............................. 84
The secondary derivative of UV spectrum of ZnTPP with t-
butylformamide (200 eq.) .......................................................... 85
The binding study of the chiral conjugatesl4B with zinc porphyrin
tweezer 2 by NMR ................................................................ 86
The structure of N—methyl-1,4-phenylenediamine ............................. 87
The binding of carrier B derivatives and zinc porphyrin tweezer based on
UV-vis, IR and NMR study ...................................................... 88

Minimized structure of Zn-TPP bound to 9B (most of the hydrogen atoms

deleted for clarity) in which the chiral center has adopted a conformation

to minimize steric strain .......................................................... 89
Proposed mnemonic for carrier B derivatized chiral acids ................... 90
The CD spectra of (S)-methoxyphenylacetic acid 8B at 0 0C ................ 93
The ECCD spectra of compound 13B and Me-13B ........................... 95
The structure of carrier B and C ................................................. 98
The structure of 13B and 13C .................................................. 101

xiv

Figure 3-24.
Figure 3-25.
Figure 3-26.
Figure 3-27.
Figure 3-28.
Figure 3-29.

Figure 3-30.

Figure 3-31.

Figure 3-32.

Figure 3-33.

Figure 3-34.

Figure 3-35.

Figure 3-36.

Figure 3-37.

Figure 3-38.

Figure 4-1.

Figure 4-2.

Figure 4-3.

The structure of 13C and 13C’ ................................................. 102
The conformations and ECCD of 13B, 13C and 13C’ ...................... 103
Structure of carrier D ............................................................ 104
Structures of carrier B and E ................................................... 108

The ECCD of carrier E derivatized methylbutyric acid 14E ............... 109
Carrier B and E derivatized amides bind to zinc porphyrin tweezer ...... 110
Structure of carrier F and its derivatized binds to zinc porphyrin

tweezer ............................................................................ 111

The ECCD of compound 8F at different temperatures ...................... 114
Hydrogen bonding vs. hydrogen-1t interaction ............................... 115

The different interactions between the zinc porphyrin tweezer and carrier
B and carrier F derivatives ....................................................... 1 16
The binding between carrier G derivatives 12G and zinc porphyrin
tweezer ............................................................................. 1 17
Carrier A-H ....................................................................... 120
The stereochemical study of carrier B derivatized a-chiral carboxylic
acids with zinc porphyrin tweezer 2 ........................................... 121

The binding of chiral conjugate to zinc porphyrin tweezer ................. 122

Opposite ECCD spectra with the two carriers B and X123
The contradiction of the observed ECCD sign vs. the predicted sign based

on the mnemonic ................................................................. 153
The observed and predicted ECCD of B-2-Br ............................... 159

Possible conformations leading to the observed ECCD signs... 160

XV

 

 

Figure 4-4.

Figure 4-5.

Figure 4-6.

Figure 4-7.

Figure 4-8.

Figure 4-9.

Figure 4-10.

Figure 4-11.

Figure 4-12.

Figure 4-13.

Figure 4-14.

Figure 4-15.

fiymmm.

The conformation of the amide part in the derivatives ...................... 161

NMR of compounds 6R and 6S ................................................ 161
The structures of 2-trifluoromethylpropionic acid 9 and 2-
trifluoromethylbutyric acid 10 .................................................. 165

The two possible conformations for compound 18 and its ECCD
spectrum ........................................................................... 170
The conformations dictated by G*—1t interaction and hydrogen
bonding ........................................................................... 171

The conformations of 11B and Me-l 1B ...................................... 172
The change of chemical shift of bromoamide 28 upon addition of
ZnTPP ................... I .......................................................... 176
The interaction of carrier B derivatized a—chiral halocarboxylic acids with
zinc porphyrin tweezer ........................................................... 177

The carrier B derivatives versus the five-membered chiral a-lactams.... 178

The comparison of interaction of five-membered lactam and an 0t-

halolactam interacting with zinc porphyrin tweezer .......................... 179
The synthesis five-membered a-chiral lactams from
a—chiral-y—butyrolactones ...................................................... l 79

The conformation of (R)-0t-alkoxy-y-butyrolactams 50-53 binding to zinc
porphyrin tweezer ................................................................ 190
The conformation of (R)-0t-chloro-y-butyrolactam 37 binding to zinc

porphyrin tweezer ................................................................. 190

xvi

Figure 4-17.

Figure 4-18.

Figure 5-1.

Figure 5-2.

Figure 5.3.

Figure 5.4.

Figure 5-5.
Figure 5-6.
Figure 5-7.
Figure 5-8.

Figure 5-9.

Figure 5-10.

Figure 5-11.

Figure 5-12.

Figure 5-13.

The conformation of chiral a—alkoxy-y—butyrolactams 38, 54 and 55
binding to zinc porphyrin tweezer .............................................. 191
The observed ECCD sign is opposite to the predicted sign based on
model ............................................................................... 208
The self-assembly of porphyrins ............................................... 243
Porphyrin acid aggregate as a chromophoric host to determine the
stereochemistry of chiral substrates ............................................ 244
The structure of the carrier F derivatized (S)-0-acetylmandelic acid 9F
and zinc porphyrin acid .......................................................... 248

The addition of chiral substrate 9F to zinc porphyrin acid in

methylcyclohexane leads to a positive ECCD ............................... 249
The time study of zinc porphyrin acid aggregate with 9F ................... 249
The photo bleaching study ...................................................... 250
The titration of zinc porphyrin acid with 9F .................................. 251
The ECCD of 9F in zinc porphyrin methyl ester ............................. 252
The addition of sodium carbonate to zinc porphyrin tweezer .............. 253
The CD spectrum of 9F to pre-treated zinc porphyrin acid ................. 253

Adding diisopropylmethylamine (1eq.) to the zinc porphyrin acid with
9F .................................................................................. 254
Structures of (S)-methoxyphenylpropionic acid derivative 8F and L-lysine
methyl ester ........................................................................ 255
The time study of zinc porphyrin mono-alcohol with carrier F derivatized

(S)—O-phenylmandelic acid 9F .................................................. 257

xvii

 

 

 

Figure 6-1.
Figure 6-2.
Figure 6-3.

Figure 6-4.

A receptor for glucopyranose and its glucose complex ...................... 263
The boronic porphyrin and its binding to sugars ............................. 264
The structure of boronic porphyrin tweezer 1 and 2 ......................... 264
The structure of a new porphyrin tweezer 45 ................................. 306

xviii

Scheme 1-1.
Scheme 1-2.
Scheme 2-1.
Scheme 2-2.
Scheme 2-3.
Scheme 2-4.
Scheme 2-5.
Scheme 2-6.
Scheme 2-7.
Scheme 2-8.
Scheme 3-1.
Scheme 3-2.
Scheme 3-3.
Scheme 3-4.
Scheme 3-5.
Scheme 3-6.

Scheme 3-7.

Scheme 3-8.

Scheme 4-1.

LIST OF SCHEMES

The derivatization of binaphthalene 3 with chiral alcohols .................. 25
Converting the amino acids and amino alcohols to diamines ................ 32
Synthesis of zinc porphyrin tweezer 1 from porphyrin carboxylic acid. . .43
Synthesis of mono-substituted porphyrin methyl ester 2 ...................... 44
Hydrolysis of porphyrin methyl ester 2 to carboxylic acid 3...... ...........44
EDC coupling of porphyrin carboxylic acid 3 with 1,5-pentanediol ....... 46

The synthesis of mono-substituted porphyrin 2 in BF3'EI20 condition. . ..48

Synthesis of mono-substituted porphyrins on insoluble polymer............51
Copolymerization to synthesize N CPS polymer ................................ 54
Synthesis of mono-substituted porphyrin on soluble polymer .............. 56
The synthesis of Boc-protected carrier A ....................................... 75
The synthesis of chiral complexes from Boc-A ................................ 75
The synthesis of carrier B derivatized chiral complexes ..................... 81
The synthesis of methylated carrier B derivative Me-IBB ................... 94
The synthesis of carrier C derivatized chiral acids ............................ 98
The coupling of carrier C with acids to form amide bond .................. 102

The structure of zinc porphyrin tweezer 2’, in which 1,3-propanediol
serves as the linkage ............................................................... 118
The synthesis of chiral benzimidazole ........................................ 119
The derivitization of (R)-2-bromopropionic acid l-Br to carrier A and

B .................................................................................... 152

xix

Scheme 4-2.

Scheme 4-3.

Scheme 4-4.

Scheme 4-5.

Scheme 4-6.

Scheme 4-7.

Scheme 4-8.

Scheme 4-9.

Scheme 4-10.

Scheme 4-11.

Scheme 4-12.

Scheme 4-13.

Scheme 4-14.

Scheme 4-15.

Scheme 4-16.

Scheme 4-17.

Scheme 4-18.

Scheme 4-19.

Scheme 4-20.

Scheme 4-21.

Scheme 4-22.

The synthesis of oc-chiral halocarboxylic acids .............................. 153
The mechanism of the a—chiral halocarboxylic acids synthesis ........... 154
The attempted synthesis of carrier B derivatized iodocarboxylic acid

B-l-I ................................................................................ 157
The synthesis of carrier B derivatized (S)-2-iodopropionic acid B-l-I... 158

The proposed synthesis of chiral 2-trifluoromethyl carboxylic acids... .. 163

The synthesis of triethylborane 8 ............................................... 164
The synthesis of trifluoromethyl iodide 7 .................................... 164
The amidation with Evans’ chiral auxiliary ................................... 165
The introduction of trifluoromethyl group to the a position ............... 166
The synthesis of carrier B derivatized 2-methyltrifluoropropionic acid
18 .................................................................................... 167
The synthesis of methylated carrier B derivative Me-l 1B ................. 171
The synthesis of modified carrier B derivative MeB-l-Br ................ 173
The synthesis of chiral Ot—chloro—‘y—butyrolactone 32 ..................... 180
The synthesis of a—chloro-y—butyrolactone 32 .............................. 181
The synthesis of oc-phenoxy-y-butyrolactone 36 ............................. 181
Converting butyrolactone to butyrolactam .................................... 182
Second route for the synthesis of (S)-a-phenoxy-y-butyrolactam 38.....183
Third route for the synthesis of (S)-a-phenoxy-y-butyrolactam 38 ....... 184
The synthesis (R)-oz-methoxy-'y-butyrolactam 43 ............................ 185
The attempted synthesis of a-alkoxy-y—butyrolactones ..................... 185
The synthesis of chiral benzoxy and allyoxy lactones ....................... 187

XX

Scheme 4-23.

Scheme 4-24.

Scheme 4-25.

Scheme 4-26.

Scheme 4-27.

Scheme 4-28.

Scheme 4-29.

Scheme 4-30.

Scheme 4-31.

Scheme 4-32.

Scheme 4-33.

Scheme 4-34.

Scheme 4-35.

Scheme 4-36.

Scheme 4-37.

The synthesis of (S)-0t-naphthyloxy-y-butyrolactone 48 and (S)-0t-p-tert-

butylphenoxy—y-butyrolactone 49 ............................................... 187

The hydrogenation of lactam 50 to 51 ......................................... 188
The synthesis of chiral or-alkyl-y-butyrolactams using cuprates. 192

The synthesis of (R)-0t-tosyl-y-butyrolactone 56 ............................. 192
Second route for the synthesis of (R)-0t-tosyl-y-butyrolactone 56 ......... 193
The first reaction of (R)-0t-tosyl-y—butyrolactone 56 with
dimethylcuprate .................................................................. 193
The second reaction of compound 56 with dimethylcuprate. . . . . . . . 194

The reaction of (R)-0t-tosyl-y-butyrolactone 56 with dibutylcuprate... .. 195
A palladium catalyzed coupling reaction of Reformatsky reagent with

allylic acetates ...................................................................... 196

The proposed synthesis of chiral a—allyl-y-butyrolactones by palladium

catalyzed Reformatsky reagent utilizing zinc on graphite .................. 196
The palladium coupling reaction of Reformatsky reagent with
allylacetate ........................................................................ 197

The proposed synthesis of chiral. a-alkyl-y—butyrolactones with chiral
auxiliary ........................................................................... 198
The synthesis of benzyl protected ‘y-hydroxy-carboxylic acid 58. . . .. 198
The attempted conversion of the acid 58 to acyl chloride. . . . . . 199

The unsuccessful reaction between chiral auxiliary 60 with butyrolactone

57 ................................................................................... 199

xxi

 

 

Scheme 4-38.

Scheme 4-39.

Scheme 4-40.

Scheme 4-41.

Scheme 4-42.

Scheme 4-43.

Scheme 4-44.

Scheme 4-45.

Scheme 4-46.

Scheme 5-1.

Scheme 6-1.

Scheme 6-2.

Scheme 6-3.

Scheme 6-4.

Scheme 6-5.
Scheme 6-6.

Scheme 6-7.

The coupling of chiral auxiliary 60 with acid 58 by generating mixture
anhydride ........................................................................... 200
The synthesis of chiral butyrolactones with chiral 2—methyloxazoline
62 ................................................................................... 201
Rationale for the absolute configuration of chiral butyrolactones... . . . . .202
The published synthesis of 2-(trimethylsilyloxy)ethyl iodide 64... .......202
The reactions for the synthesis of benzophenone ethylene acetal 69... . .203
The synthesis of 2-(OTBS)ethyl iodide 68 ................................... 204
The synthesis of (S)-2-allyl-y-butyrolactone 67 .............................. 204
The synthesis of (S)-2-allyl-y-butyrolactam 75 .............................. 205
The synthesis of (S)-2-propyl-y—butyrolactam 77 ............................ 207
The synthesis of porphyrin mono-alcohols 1 and 2 .......................... 256
Proposed synthesis of porphyrin tweezer 1 by MacDonald “2+2”

cyclization ......................................................................... 265
Proposed synthesis of porphyrin tweezer 2 by MacDonald “2+2”
cyclization ......................................................................... 266
Synthesis of dipyrromethane 5 .................................................. 267

The proposed synthesis of dipyrromethane-dialdehyde 6 from 2-pyrrole-

carboxylic acid 8 .................................................................. 268
The esterification of 2-pyrrol-carboxylic acid 8 ............................. 268
The condensation of ethyl ester 12 with aldehyde 9 ......................... 269

Unsuccessful synthesis of dipyrromethane-dialdehydes 6a and 6b from

pyrrole ............................................................................. 269

xxii

 

Scheme 6-8.

Scheme 6-9.

Scheme 6-10.

Scheme 6-11.

Scheme 6-12.

Scheme 6-13.

Scheme 6-14.

Scheme 6-15.

Scheme 6-16.

Scheme 6-17.

Scheme 6-18.

Scheme 6-19.

Scheme 6-20.

Scheme 6-21.

Scheme 6-22.

Scheme 6-23.

Scheme 6-24.

Scheme 6-25.

Scheme 6-26.

Scheme 6-27.

Scheme 6-28.

Scheme 6-29.

Scheme 6-30.

Proposed synthesis of boronic porphyrin acid 4 ............................. 271
Synthesis of dipyrromethane-dicarbinol 17 .................................. 272
Two pathways for the synthesis of porphyrin tweezer 2 .................... 274

The unsuccessful EDC coupling of compound 16 and 1,5-pentanediol..275

The unsuccessful hydrolysis of methyl ester 17 to acid 16. . . . . . . . .275
Alternate way to synthesize dimer 19 ......................................... 276
The synthesis of dimer 20 ....................................................... 276
The unsuccessful Freidel-Crafts reaction to synthesize dimer 19... ......277
The first attempt to synthesize compound 19 from methyl ester 17.. . . ...277

Directly synthesis of diester 19 from methyl ester 17 ....................... 278
The MacDonald “2+2” cyclization of compound 5 and 17 ................. 279
The attempted synthesis of boronic porphyrin tweezer 2 ................... 281
The synthesis of boronic methyl ester 24 ..................................... 282
The unsuccessful removal of methyl ester of compound 24 ................ 283
The attempted synthesis of porphyrin tweezer 2 ............................. 284
Pd catalyzed coupling of pinacolborane with aryl halides .................. 284
The synthesis of porphyrin tweezer 2 by Pd coupling reaction ............ 285
The synthesis of p-bromoporphyrin alcohol 29 .............................. 285
The attempted synthesis of dibromo tweezer 31 ............................. 286
The unsuccessful synthesis of bromoporphyrin carboxylic acid 32.. .....287

The synthesis of porphyrin methyl ester 33 .................................. 287
Hydrolysis of methyl ester 33 to acid 32 ..................................... 288
EDC coupling of acid 32 with 1,5-pentanediol .............................. 289

xxiii

 

Scheme 6-31.

Scheme 6-32.

Scheme 6-33.

Scheme 6-34.

Scheme 6-35.

Scheme 6-36.

Scheme 6-37.

Scheme 6-38.

Scheme 6-39.

Scheme 6-40.

Scheme 6-41.

Scheme 6-42.

Scheme 6-43.

Scheme 6-44.

Scheme 6-45.

Scheme 6-46.

Scheme 6-47.

Scheme 6-48.

Scheme 6-49.

Scheme 6-50.

The Pd catalyzed coupling of dibromo tweezer .............................. 290
The coupling reaction of porphyrin methyl ester 33 ......................... 291
The unsuccessful coupling with bis(pinacolato)diboron .................... 292
Synthesis of triflate porphyrin methyl ester 39 ................................ 293
The unsuccessful Pd coupling of triflate 39 ................................... 294
Synthesis of iodoporphyrin methyl ester 41 ................................... 295
The synthesis of diiodoporphyrin tweezer 42 ................................ 297
The Pd coupling of diiodoporphyrin tweezer 42 to synthesize porphyrin
tweezer pinacol-2 ................................................................ 298
The unsuccessful hydrolysis of boronic porphyrin 36 ....................... 299
The cleavage of pinanediol boronate to boronic acids ....................... 300
The attempted oxidation of porphyrin methyl ester 36 ...................... 301
The oxidation studies on model molecule 44 ................................ 302
The attempted reduction of methyl ester 36 .................................. 303
The transmetallation with bromobenzene ..................................... 304
The attempted transmetallation with bromoporphyrin 33 .................. 305
The synthesis of boronic porphyrin methyl ester 46 ......................... 306
The synthesis of pyridine porphyrin 47 ....................................... 307
The bromination of compound 48 .............................................. 307
The reaction between pyridine and primary bromide 48 .................... 308
The synthesis of water soluble porphyrin methyl ester 46. . . . . . . . . ......309

xxiv

Table 1-1.

Table 2-1.

Table 3-1.

Table 3-2.

Table 3-3.

Table 3-4.

Table 3-5.

Table 3-6.

Table 4-1.

Table 4-2.

Table 4-3.

Table 4-4.

Table 4-5.

Table 4-6.

Table 6-1.

LIST OF TABLES

Definition of exciton Chirality for a binary system ............................ 19
Synthesis of soluble polymer supported aldehyde 10 ......................... 55
CD predictions and amplitudes of carrier A derivatized chiral carboxylic

acids bound to porphyrin tweezer 2 ............................................. 78
CD predictions and amplitudes of carrier B derivatived chiral carboxylic

acids bound to porphyrin tweezer 2 ............................................. 91
CD amplitudes of carrier C derivatized chiral carboxylic acids bound to
zinc porphyrin tweezer 2 ......................................................... 100
The extinction coefficient 8 of Mg porphyrin tweezer in different
solvents ............................................................................ 106

The CD data of carrier D derivatized chiral acids with magnesium

porphyrin tweezer ................................................................. 107
Comparison of the CD data of carrier B and F derivatives ................ 1 12
The enatiomeric ratio for chiral a—halogen carboxylic acids ............... 155

The CD data of carrier B derivatives of a—chiral halocarboxylic acids... 156
Change in proton chemical shift upon addition to ZnTPP .................. 174

The change in chemical shifts of haloamides upon addition of ZnTPP.. 176

The CD data of chiral butyrolactarns ........................................... 189
The ECCD data of (S)-2-alkyl-'y-butyrolactams ............................. 207

The palladium catalyzed coupling of iodoporphyrin 41 with
pinacolborane ..................................................................... 295

XXV

 

 

[a]

9-AHA
AIBN

Ar
BF3rEtzO
Boc

BOP

BuzCuLi
Bu3P

CgK

CD

CE

CF31
(C0C12)2
p-chloranil
cm

d

DDQ

LIST OF ABBREVIATIONS

angle of rotation

specific rotation

CD amplitude
9-anthrylhydroxyacetic acid
2,2’-azobis(isobutyronitrile)
aromatic

boron trifluoride diethyl etherate
tert-butoxycarbonyl
benxotriaoyl-1-yloxytris(dimethylamino)phosphonium
hexafluorophosphate

lithium dibutylcuprate
tributylphosphine
potassium-graphite

circular dichroism

Cotton effect

trifluoromethyl iodide

oxalyl chloride

tetrachloro- l ,4-benzoquinone
centimeter

doublet

2,3-dichloro-5 ,6-dicyano- l ,4-Benzoquinone

xxvi

 

DIAD
DMAP
DMF
DMSO

EI3N

ECCD
EDC
EtOAc
eq
Et3B

FAB-MS

GPC

HPLC

LDA

diisopropylazodicarboxylate
4-dimethylaminopyridine

N ,N-dimethylformamide

dimethyl sulphoxide

triethylamine

molar absorption coefficient

exciton coupled circular dichroism
1-(3-dimethylaminopropyl)-3-ethylcarbodiimide
ethyl acetate

equivalents

triethylborane

fast atom bombardment mass spectroscopy
gram

gel permeation chromatography
high pressure liquid chromatography
hour

infrared

N MR coupling constant

potassium hydride

potassium iodide

lithium diisopropylamide

magnetic transition dipole moment

xxvii

mCPBA
MegCuLi
min

mg

MHz

MS

NaBH4

N aH

N aHMDS
NaIO4

NaNOz

NCPS
NIS

nm
NMR
ORD
PGME
PPh3
i-Peret
PDI

Pd(PPh3)4

3-chloroperoxybenzoic acid
lithium dimethylcuprate
minute

milligram

megahertz

mass spectroscopy
refractive index

sodium borohydrate

sodium hydride

sodium hexamethyldisilazane
sodium periodate

sodium nitrite

N -bromosuccinimide
non-cross-linked chloromethylated polystyrene
N-iodosuccinimide
nanometer

nuclear magnetic resonance
optical rotatory dispersion
phenylglycine methyl ester
triphenylphosphine
diisopropylethylamine
polydispersity index

tetrakis(triphenylphosphine)palladium

xxviii

Chapter 1
Introduction

1-1. The Exciton Coupled Circular Dichroic Method
1-1.l. Chirality

Chirality, based on the Greek word of “hand” (cheir) means “handness”, in
reference to that pair of non-superimposable mirror images constantly presented before
us: our two hands.1 In chemistry, Chirality is an important property of organic
compounds, which arises from asymmetric disposition of atoms. Molecules that are not
identical with their mirror image are defined to be “chira ”. The simplest example of a
chiral compound is a molecule containing a chiral center, which is a tetrahedral carbon
bound to four different substituents, such as 2-methylbutyric acid. Chemists use the
letters R and S to indicate the configuration (arrangement of groups) of the chiral center,
based on defined priorities.l

Molecules devoid of a chiral center can also be chiral, as long as they lack a plane
of symmetry. Examples of this type of Chirality are allenes,2 helicenes3 and

atropisomers.4 Figure 1-1 shows examples of chiral molecules and their mirror images.

ME: E we Me i Me
Et COzHi Et COZH Mfia=c=<H g H>=c=eMe
2-methylbutyric acid allene

.13) C]. cozHOzN NOZHOZC
©©©© ©©©©

N02 OZN

 

 

hexahelicene atropisomers

Figure 1-1. Examples of some chiral molecules and their mirror images.

Allenes are chiral because they lack a plane of symmetry. Since the two 1: bonds
in allenes are perpendicular to each other, the plane containing one H-C-Me group is
perpendicular to the plane containing the other H-C-Me group. Therefore, allenes are not
superimposable, and these compounds are chiral.2 The polycyclic hexahelicene is the
classic example of an inherently chiral chromophore.3 Because hexahelicene is not a
planar molecule but a helix, it lacks a plane of symmetry. (+)-Hexahelicene has right
handed helicity or the P configuration, and correspondingly, (-)-hexahelicene has left
handed helicity or the M configuration. Atropiosmerism is isomerism caused by
“freezing” the internal rotation about a single bond in a molecule. In the example shown
in Figure 1-1, because of the steric interactions of the substituents at ortho- position, the
biphenyls with o-substituents are chiral.5

The stereochemistry of natural compounds has a large impact on their chemical
and biological properties. For example, the (S) enantiomer of Carvone has the odor of

caraway seeds, while the (R) enantiomer has the odor of Spearmint (Figure 1-2).1

0 O
(Sl-Carvone (FD-Carvone
(in caraway seeds) (in spearmint oil)

Figure 1-2. The structures of (S)- and (R)-Carvone.

A dramatic example of how a simple change in stereochemistry can affect the
biological property of a molecule is found in the compound, Thalidomide, which was a
commercially available drug about half a century ago for morning sickness during
pregnancy. As shown in Figure 1-3, Thalidomide has a single stereogenic center and can

thus exist in two stereoisomeric forms. One of the enatiomers, (R)-Thalidomide, has

strong sedative properties, while other enantiomer, (S)-Thalidomide, is highly teratogenic.
Since the drug at that time was produced in racemic form, women who were given the

drug during the first three months of pregnancy had babies with a wide variety of birth

defects.

” O
(m-Thalidomide (S)-Thalidornide
Figure 1-3. The structures of (R)- and (3)-Thalidomide.

It is, therefore, of great importance to determine the absolute stereochemistry of
chiral compounds for natural product chemistry, pharmaceutics, and biochemistry.
However, despite the extensive research currently underway in this field, determination
of absolute stereochemistry is not a trivial task. The general techniques used include: (1)
X-ray crystallography; (2) chiroptical methods such as optical rotatory dispersion (ORD)
and circular dichroism (CD);6 (3) nuclear magnetic resonance (NMR) spectroscopy of

Mosher esters and related derivatives;7 and (4) chemical correlation with known chiral
(:()rnpounds.8
Up to date, the most dependable method for absolute stereochemical
determination of chiral molecules is X-ray crystallography. The crystal structure of any
compound can unambiguously determine the absolute stereochemistry of all chiral
centers in a single experiment, and so far, has been the method of choice for most
systems. However, there are significant drawbacks associated with this method. First of
all, not all compounds can easily crystallize to give single crystals of the quality required

for X—ray crystallography. This is part of the reason crystallography can be a very time

consuming technique that requires special training. At the same time, it requires
significant amount of material that might not always be available. That is why easier and
more accessible methods are needed and several spectroscopic methods have been
developed.
The goal of our research is to develop general and sensitive circular dichroic (CD)
methods for absolute stereochemical determination of chiral compounds.
1-1.2. Optical Rotatory Dispersion (ORD) and Circular Dichroism (CD)
Spectroscopy
Light, which is electromagnetic radiation, has an electric and a magnetic field.
'These two fields oscillate perpendicular to one another and to the direction of the
propagation of light. The polarization of light is defined by the direction of its electric
field vector.9
Light from ordinary light sources, such as sun and a light bulb, is unpolarized,
since it consists of light vibrating in all directions. However, if unpolarized light passes
through a polarizing filter, only the subset with parallel oscillation with the direction of
the filter survives, and is referred to as linearly polarized light. Linearly polarized light
can be decomposed into a pair of circularly polarized light beams: left and right circularly
polarized light. In a circularly polarized light beam, the electric field remains constant in
magnitude, but traces out a helix as a function of time. Figure 1-4 shows the linearly

polarized and circularly polarized light.10

E

A\
polarizing L R
filter 6"
light linearly left and right circularly
polarized light polarized light
x

I light propagation

arias. __..

right circularly polarized light

 

Figure 1-4. The linearly polarized light and circularly polarized light.

When circularly polarized light passes through an achiral compound or a racemic

Inixture, the velocity and absorbance of the left and right circularly polarized light is

affected similarly. On the other hand, if it passes through an optically active medium, the

velocity and absorbance of one circularly polarized component is altered to a greater

extent as compared to the other opposing circularly polarized component.ll This is

referred to as a Cotton effect (CE, in honor of French physicist Aime’ Cotton, who
observed both ORD and CD phenomena, beginning in 1896).

Anzm-nmtO (1-1)

In equation (1-1),9 rig and me are the refractive indices for left and right circularly

polarized light. Differences in refractive indices correspond to the differences in the

velocity of light. The difference in velocity of left and right circularly polarized light

results in Optical Rotatory Dispersion (ORD) spectroscopy. Since the left and right

circularly polarized light travels in the optically active medium with different speed, the

two components are no longer in phase and the resultant vector has been rotated by the

angle 0t relative to the original plane of polarization. Optical Rotatory Dispersion (ORD)
spectroscopy is the measurement of specific rotation, [or], as a function of wavelength.

The difference in refractive indices for left and right circularly polarized light is
related to the angle of rotation, or, as shown in equation 1-2:9

a = (m — nR)18001/xio (1-2)

where a is the angle of rotation, with the unit in degrees, 22,, and me are the
refractive indices for left and right circularly polarized light, I is the path length in
decimeters, and 20 is the wavelength in vacuum of the light beams in centimeters.

The specific rotation, [0t], can be calculated from the observed angle of rotation, at,

as expressed by equation 1-3:9

[a] = 3 (1-3)
cl

where ais the angle of rotation, with the unit in degrees, c is the concentration of
the sample in g mL'l, and I is the path length in decimeters.

Theoretically, ORD may be detected over all wavelengths, since it is based on the
unequal refractive indices for left and right circularly polarized light. All chiral
substances exhibit a molecular refraction at almost any wavelength of incident irradiation.
Optical rotation at a single wavelength, such as 589 nm (sodium D-line), has been used to

detect and quantitate optical activity.
When circularly polarized light passes through an optically active medium, not
only is there a difference in the velocity of the left and right circularly polarized light, but

also the absorbance of these two components is different (equation 1-4).9 The difference

in molar absorptivity, A8, is termed Circular Dichroism (CD). 61, and ER are the molar
absorption coefficients for left and right circularly polarized light.
A€=&-&¢0 (1—4)
CD is an absorptive process, namely the different absorption of left and right
circularly polarized light, and therefore, CD only occurs in the vicinity of an absorption
band. Thus, in order to observe a CD spectrum, the chiral substances need to contain a
chromophore. The shape and appearance of a CD curve closely resembles that of the
ordinary UV-vis absorption curve of the electronic transition to which it corresponds.
However, unlike the ordinary UV-vis absorption curves, CD curves may be positive or

negative, as in Figure 1-5. CD curves plot A8 vs. wavelength.9

 

 

 

 

 

(a) positive CD (b) negative CD
+ +
A8 A8 0 fl
0
Wavelength (nm) Wavelength (nm)

Figure 1-5. (a) Positive and (b) negative CD Cotton effects.

Circular Dichroism (CD) and Optical Rotatory Dispersion (ORD) are two
manifestations of one and the same phenomenon. The molar amplitude a of an ORD can
be related to the intensity of the CD curve, A8, by equation 1-5:12

a =- 40.28A£ (1-5)

In a molecule, the absorption of light promotes an electron from its ground state

10 an excited state, which creates a momentary dipole, commonly referred to as the

electric transition dipole moment and is denoted by the vector ,u. In an achiral molecule,
the net electron redistribution during the process is always planar, while in a chiral
molecule the electron rearrangement during the transition is always helical. In the latter
case, not only the charge displacement takes place during the electronic transition, which
generates the electric transition dipole moment [1, but also the rotation of electric charge
creates a magnetic field whose strength and direction may be described by the magnetic

transition dipole moment and which is denoted by the vector m.8

m & K m
Figure 1-6. The “right hand rule” to determine the direction of magnetic transition
dipole moment m.

As shown in Figure 1-6, the direction of magnetic transition dipole moment m can
be determined by the application of the “right hand rule” to the rotation of the charge
(circular current): If the current flows in the direction of the curved fingers of the right
hand, then the outstretched thumb points the direction of m.13

The rotational strength, R, which is a theoretical parameter representing the sign
and strength of a CD Cotton effect (CE), is given by the scalar product of the electric and

magnetic transition moments (equation 1-6):8
R=,uom=|,u||m|cosfl (1-6)
where ,u and m are the electric and magnetic transition dipole moments,
respectively, and ,6 is the angle between the two transition moments.
The sign of CE is positive when the angle is acute (0<,B<90°) or, in the limiting

case, parallel, and it is negative when the angle is obtuse (90°<fl<180°) or, in the limiting

case, antiparallel. Dextrorotation results when R>0, together with positive CD curve, and

levorotation is generated when R<0, together with negative CD curve. There is no CE

when the electric and magnetic transition dipole moments are perpendicular to each other.

direction of C]. direction of

electron flow © ©.©©© electron flow

a /U ’55 m (M)-Helix (©P)-©Hellx #/ 55 m

Figure 1-7. The transition moments for hexahelicene. Antiparallel moments for (M) and
parallel moments for (P) helicities.

The electron helical motion of chiral compounds during excitation can be easily
demonstrated with hexahelicene, as shown in Figure 1-7.14 The direction of electron flow
traces the helicity of the hexahelicene. The electric transition dipole moment ,1: is
identical for either enantiomer, whereas the magnetic transition dipole moment direction
m is reversed. According to equation (1-6), in (M)-helix, the antiparallel alignment of the
transition moments leads to a negative Cotton effect. So it is defined as being
levorotation, and has a negative CD spectrum. On the other hand, the (P)-helix, the
parallel alignment of the transition moments leads to a positive Cotton effect. So it is
defined as being dextrorotation, and has a positive CD spectrum.

The absorption of light promotes the electron from its ground state to an excited
state, and in a chiral molecule the electron rearrangement during the transition is always
helical. That the electron with helical movement interacts differently with the left and
right circularly polarized light is the cause of the Cotton effect (CE), which states that the
velocity and absorbance of one circularly polarized component is different from those for

the other one.ll

Both ORD and CD spectroscopy can be recorded by Circular Dichrometers
(spectropolarimeters). The essential features of a CD spectropolarimeter are a source of
monochromatic left and right circularly polarized light and a means of detecting the
difference in absorbance of the two polarized light.

Light source monochromator Modulator Sample cell

    
 

      

 

Linearly polarized light Circularly polarized light
CD spectrum
A5
wavelength Photomultiplier

 

 

 

Lock-in amplifier tube

Figure 1-8. The CD spectropolarimeter.

As shown in Figure 1-8,'5 the light typically from a Xenon lamp passes through
monochromators, which are crystal prisms, to achieve linearly polarized light. In the CD
spectropolarimeter, the optical system comprises of two monochromators, and is known
as double monochromator. The ability of a double monochromator in reducing stray light
makes it indispensable in CD measurement. The linearly polarized light is then
modulated into left and right circularly polarized light before it passes through the sample
chamber. The transmission of light through the sample is measured by a photomultiplier
tube, which produces a current whose magnitude depends on the number of incident

photons. This current is then detected by a lock—in amplifier and recorded.

Most CD spectropolarimeters measure differential absorbance, AA, which is
converted to A8 based on the Lambert-Beer’s Law (equation 1—7):8
AA = Ascl (1-7)
Since Optical Rotatory Dispersion (ORD) and Circular Dichroism (CD) are
manifestations of chiral substances and are not observed for achiral compounds or
racemic mixtures, they can be used to detect and quantitate optical activity. But it should
be emphasized that Optical Rotatory Dispersion (ORD) and Circular Dichroism (CD)
themselves do not allow the configuration of a given product to be defined. Attempts
have been made to find empirical rules16 and semi-empirical rules,17 such as sector rules,
which would permit a-priori calculation of the ORD or CD of a given structure.
However, these empirical and semi-empirical rules suffer from one significant
shortcoming, which is that they are not valid in all cases. These rules are to be applied
with great caution, and simple application of the rules does not permit the determination
of absolute configuration.
Recent developments on this field allow for a quantitatively reliable quantum

18. 19

chemical calculation of the rotation angle of chiral molecules. Accordingly, the

absolute stereochemistry of the chiral molecule can be assigned upon comparison of the
calculated rotation angle and the experimental result. The computation of the specific
rotation angle is based on the calculation of the electric-dipole magnetic—dipole

20. 21

polarization (G ’), a measure of the overlap of the electronic polarization induced by

the electric and the magnetic field of the incident light, which can be computed routinely

using the coupled-perturbed Hartree-Fock method22 on standard desktop workstation.” '9

ll

In the method developed by Peter Wipf and his coworkers, the individual
contributions of all the atoms and functional groups are calculated and summed up to
obtain the specific rotation angle of the chiral molecule.23’ 24 In this way, the structure of
the chiral compound is linked to the sign and magnitude of the optical rotation angle.25
This method is of special importance in assigning the absolute stereochemistry of large,
flexible molecules with multiple chiral centers, and it has been successfully applied to
determine the absolute stereochemistry of several natural products.”28 However, more
work remains to be done to establish a better understanding of this protocol and
determine the possible limitations of this fundamental approach for structural analysis.
1-2. The Exciton Coupled Circular Dichroic (ECCD) method and its
applicationsThe Exciton Coupled Circular Dichroic (ECCD) method29 is a non-empirical
method to establish the absolute configuration of chiral compounds on a microscale. The
ECCD method, developed from the coupled oscillator theory30 and group polarizability
theory,3| is based on the through space exciton coupling between two or more chirally
oriented chromophores, which are not conjugated to each other. The coupling of the
chromophores' electric transition dipole moment leads to the observed bisignate CD
spectrum and the non-empirical determination of their orientation. Figure 1-9 is an

example of exciton coupling between the two benzoates in steroidal 2,3-bisbenzoate.29

positive couplet
"""" ll

  

 

 

 

 

A
O H ______________ y
syn ___. nm

Figure 1-9. Exciton Coupled Circular Dichroism (ECCD) of steroidal 2,3—bisbenzoate.

12

 

 

 

As shown in Figure 1-9, a typical ECCD spectrum is a bisignate curve. The
extremum at longer wavelength is called the first Cotton effect, and correspondingly, the
extremum at shorter wavelength is called the second Cotton effect. If the sign of the first
Cotton effect is positive, the ECCD spectrum is defined as positive, and if the sign of the
first Cotton effect is negative, then the sign of the ECCD spectrum is negative. The
amplitude (A) of the ECCD spectrum is defined as: A = A81 - A82 where A81 and A82 are
the intensities of the first and second Cotton effect, respectively.

Based on the ECCD method, the sign of the ECCD spectrum is dictated by the
relative orientation of the chromophores in the chiral molecule. As the steroidal 2,3-
bisbenzoate system in Figure 1-9 indicates, the clockwise orientation of the two
chromophores results in a positive ECCD curve, and counterclockwise orientation will
result in a negative ECCD curve.29

What follows is an attempt to understand the Exciton Coupled Circular Dichroic
(ECCD) method based on qualitative explanation and Quantum Mechanical theories.
1-2.1. The qualitative explanation of Exciton Coupled Circular Dichroic method
(ECCD)

As it has been mentioned before, the absorption of light promotes the electron
from its ground state to an excited state. The redistribution of electron during the process
creates a momentary dipole, called the electric transition dipole moment (a), and the
rotation of electric charge creates a magnetic field whose strength and direction can be
described by the magnetic transition dipole moment (m).

When the molecule contains two identical chromophores, which are not

conjugated with each other, the most significant coupling takes place between the

13

transitions occurring at the same energy, E. The through space coupling of the electric
transition dipole moments of these two chromophores generates one of the two helices,
depending upon whether they are in-phase (symmetric) or out-of-phase (anti-symmetry)-
so we expect to see a CD spectrum with two peaks. This can be explained by examining
a dibenzoate ester in Figure 1-10.13

(a) ('3 O ('3)
C2

0: =0
a:

e(
I

_©,

 

(c) f; e

—e—»

(lime

Figure 1-10. The qualitative explanation of ECCD based on a dibenzoate ester. (a)

G)
\

6)
((1)/6‘9
é #

 

«mg

Orientation of the electric transition dipole moment in benzoates, Amax=230 nm; (b)
Preferred configuration of a vic-dibenzoate with negative dihedral angle; (c) and (d)
combination of the individual electric transition dipole moments and resultant magnetic

transition dipole moment. ' 3

In the benzoate esters, there is a strong UV absorption band at about 230 nm, and
it arise from the n—m’“ transition involving the benzene ring and COzR.32 The large
electric transition dipole moment ,u of each benzoate group is oriented approximately

collinearly with the long axis of the benzoate group and almost parallel to the direction of

the C-0 ester bond (Figure 1-10a).

l4

The conformations of benzoates are well known and are approximated in Figure
1-10b. In Figure 1-10b, the two benzoate groups are oriented in a counterclockwise
manner, and correspondingly, the dihedral angle a) between the C-0 bonds is negative.
As it has been mentioned above, since the individual electric transition dipole moment it
aligns almost parallel to the C-0 ester bonds, the dihedral angle a) between the C-0
bonds thus corresponds roughly to that between the two electric transition dipole moment
it.

The individual electric transition dipole moment it of an electronic transition
couples in-phase (symmetric) or out-of-phase (anti-symmetric) (Figure 1-10c and d,
respectively). In case c, where the two electric transition dipole moments couple in-
phase, the “total” electric transition dipole moments are oriented along the chromophoric
C2 axis, and in case d, where the two electric transition dipole moments couple out-of-
phase, the “total” electric transition dipole moments are oriented perpendicular to the
chromophoric C2 axis.

The charge rotation associated with the transition may be visualized by placing
the partial moments tangent to a cylinder with the latter aligned along the axis of the
“total” electric transition dipole moment ,u. In case c, the electric transition dipole

moment ,uand the resultant magnetic transition dipole moment m are arranged parallel,

and according to equation 1-6, R = ,u o m = lyllml cos ,6 , leads to a positive Cotton effect

and thus a positive CD band. In case d, the electric transition dipole moment ,u and the
resultant magnetic transition dipole moment m are arranged anti-parallel, and

correspondingly, that leads to a negative Cotton effect and a negative CD band.‘3

15

 

So the two resulting CD bands generated from the through space coupling of the
electric transition dipole moments of the dibenzoate have equal magnitude and opposite
sign, based on if the coupling of the electric transition dipole moments follows mode c or
d in Figure 1-10. Since the energies of the two CD signals are close to the localized
energy E, the observed CD spectrum is actually the overlapping of these two CD signals
with equal magnitude and opposite sign, and this leads to the adjacent bisignate CD
bands that are characteristic of Exciton Coupled Circular Dichroism (ECCD).

Repulsion between poles of the same sign in the c mode indicates that in-phase
coupling is of higher energy than the out-of—phase mode (1, where the poles of unlike sign
are closest together. Consequently, the positive CD band associated with the c mode is at
shorter wavelengths than the negative CD band resulting from the d mode; the energy gap
between these two CD bands is defined as Davydov splitting. Since the negative CD
band is at the longer wavelength than the positive CD band, for the absolute
configuration shown in Figure 1-10b, the bisignate couplet is negative overall as in
Figure 1-11.

As shown in Figure 1-11, in the case of the CD spectrum, the c and (1 modes
(Figure 1-10) have opposite signs and the summation peak is a bisignate couplet. On the
other hand, in the case of the UV spectrum, the c and d mode, which are real absorptions
observable in UV, are always positive, and therefore, summation of the two component

spectra always leads to an unsplit absorption band of double intensity.29

16

 

 

A1: Davydov splitting

Figure 1-11. The ECCD couplet for the configuration shown in Figure 11-10b. The
summation curves (the observed ECCD curve, shown in solid line) of two Cotton effects
(dashed line) of opposite signs separated by Davydov splitting AA.

In conclusion, the ECCD spectrum is a bisignate spectrum because of the through
space coupling of the electric transition dipole moments of the non-conjugated
chromophores, with the in-phase (symmetric) and out-of-phase (anti-symmetric) coupling
yielding opposite Cotton effect. Moreover, as shown in Figure 1-12, the sign of the
resulting ECCD spectrum is dictated by the orientation of the interacting chromophore:
the counterclockwise orientation of the chromophores, when the nearest chromophores
can be made to eclipse the farther chromophore by rotating along the shortest path in a
counterclockwise direction, leads to a negative ECCD spectrum, and correspondingly, the

clockwise orientation leads to a positive ECCD spectrum.33

69 6
0? TD
Figure 1-12. The screw sense of the chromophores and its corresponding ECCD sign.

l7

 

 

l-2.2. The explanation of ECCD theory based on Quantum Mechanical theories

The ECCD theory can also be explained by Quantum Mechanical theories. When
a molecule contains two identical non-conjugated chromophores, because of the through
space interactions, excitation is delocalized between the two chromophores and the
excited state (exciton)34 is split into two states, a and ,6, by resonance interaction of the

local excitations (Figure 1—13), just as in the case of electron delocalization of an ethylene
double bond.

2V“ ‘:;:»—— excited statea

energy

 

 

~ —— ground state 0

exciton system chromophore j
delocalized local excitation

excitation

chromophore i
local excitation

Figure 1-13. Splitting of the excited states of isolated chromophores i and j by exciton
interaction. The energy gap 2V” is called Davydov splitting.

Based on theoretical calculation on the binary system, the rotational strength, R,
which is a theoretical parameter representing the sign and strength of a CD Cotton effect,
can be approximated as equation 1-8:29

Ra, r = i-(1/2)7503Rr,- . (a... x gr...) (1-8)

where the positive and negative signs correspond to a— and ,3- state, respectively.

In equation 1-8, Rij is the distance vector from chromophore i to j, Moe and

ruler: are electric transition dipole moments for excitation 0—)a of group i and j,

respectively, and 60 is the excitation number of transition from ground state 0 to excited

18

 

state a, which can be expressed as: 0'0 = VO/ C = (Ea "‘ E0) /(hC) , h is the Planck
constant and c is the speed of light in vacuum. The equation 1-8 indicates that the Cotton
effect of a— and ,6- state have identical rotational strength of opposite sign. As shown in
Figure 1-11, this leads to a bisignate CD spectrum with two Cotton effect peaks of
opposite signs, which are separated by the energy gap AA (Davydov splitting).

The sign and amplitude of the bisignate ECCD spectrum is theoretically defined

in equation 1-9 as:29
Rij . (flea X [20(0th “-9)

Where Rij is the distance vector from chromophore i to j, [boa and fljoa are
electric transition dipole moments for excitation 0—)a of group i and j, respectively, and
Vij is interaction energy between the two group i and j.

If the equation 1-9 is positive, the observed ECCD spectrum is positive, and if it
is negative, then the observed ECCD spectrum is negative. From the equation above, the
sign of the bisignate curve completely depends on the spatial orientation of the two

chromophores.35 The definition of exciton Chirality for a binary system is shown in Table

 

 

l-l.33
Table 1-1. Definition of exciton Chirality for a binary system
TDSiiifiitgiie Quantitative Definition Cotton effects
to
. . '\ -__ ... _. ._ positive first and
2:32;; 6 R” ' (rum XIU’MW" > 0 negative second
Cotton effects
a

. f7 ‘.. ._ .. .. negative first and

héfig'iv; \d) R” . (tum X ”MW” < 0 positive second

Cotton effects

 

l9

If the electric transition dipole moments of the two chromophores from front to
back constitute a clockwise orientation, then according to the equation above, it will give
a positive ECCD spectrum, and vice versa.” 37 Therefore, as long as the orientation of
two chromophores in space is known, the sign of its ECCD spectrum can be predicted.
On the other hand, the sign of the bisignate ECCD spectrum enables one to determine the
relative orientation of the two chromophores in space in a non-empirical manner.

The amplitude (A) of the ECCD couplet, defined as the difference in A8 between
the first and second Cotton effect of the couplet, is affected by the following factors:38

(1) Molar absorptivity coefficient (8 value) of the chromophores: A is
proportional to 3. Therefore, chromophores with strong absorption are preferred for
higher sensitivities of the ECCD method. If highly sensitive chromophores are employed,
both the chromophoric derivatization and CD measurements can be performed at the
microgram or nanogram level;

(2) Interchromophore distance (R): A is sensitive to the distance between the
interacting chromophores, since it is inversely proportional to R2. To achieve high
amplitude, the coupling chromophores should be close to each other in space;

(3) Projection angle between the interacting chromophores: A value is
maximal when the projection angle is around 70°, while there is no exciton coupling
when the projection angle is 0° or 180°;

(4) Number of interacting chromophores (X, Y, Z): the A value of the whole

molecule is the summation of the A value of each pair of interacting chromophores,

20

i.e. Atom! = Axy + sz + Ayz . The additivity rule has been proven by experiments and
theoretical calculation of multichromophoric systems?“

Not only can exciton coupling take place between identical chromophores but

also it can occur between different chromophores, however it is less effective.4244 For
example, when different benzoates are used as the chromophores with 2mm between 200

nm to 300 nm, as long as the absorption maxima of the chromophores are not separated
by more than 80 nm, exciton coupling can happen between them.29

The interacting chromophores do not even have to be within one molecule, as
long as they are in close proximity to each other in space. For example, stacking of
anthocyanins45 and porphyrin containing brevetoxins46 give rise to exciton coupled CD
bands.
1-2.3. The application of ECCD method to determine the absolute stereochemistry
of chiral molecules

Since the Exciton Coupled Circular Dichroic (ECCD) method is based on sound
theoretical calculation, the absolute stereochemistry of organic compound exhibiting
typical bisignate ECCD spectrum is assignable in a non-empirical manner. The Exciton
Coupled Circular Dichroic (ECCD) method is a sensitive and general method requiring
only microgram scale samples. It has been successfully applied in determining absolute
stereochemistry of wide variety of compounds, including polyols,47 carbohydrates,48
quinuclidines49 and hydroxy acids,50 etc.5 '

Figure 1-14 is a simple example of determining the absolute stereochemistry of

(+)-abscisic acid by ECCD method.52‘ 53 Abscisic, a plant hormone, contains only one

chiral center in the molecule, and two non-conjugated chromophores: enone and dienoic

21

acid. The enone and dienoic acid chromophores interact with each other through space
resulting in a positive ECCD couplet. Based on the sign of the ECCD spectrum, the two
chromophores must be oriented in a clockwise manner in space, which directly indicates

the or- configuration of the tertiary hydroxyl group.

6')
OH HOOC
\ \ E //
0 COOH o\— / l
\7 0H

Figure 1-14. The stereochemical determination of abscisic acid by ECCD method.

More examples of determining absolute stereochemistry of chiral compounds
with ECCD are shown in Figure l-15. (i) The configuration of the 15-glcNAc group in
pavoninin-4, a shark repellent, is determined from the sign of the ECCD between the
enone and bromobenzoate.54 (ii) The absolute configuration of the potent cockroach sex
excitant periplanone B is established by carbonyl reduction, benzoylation, and
measurement of the ECCD arising from interactions between the benzoate and diene
chromophores.55 (iii) The absolute configuration of the polyol side chain of ahopanoid
from Nostoc PCC 6720 is assigned from the bichromophoric excition coupled CD spectra.
This is a case of heterochromophoric coupling, which can provide fingerprint spectra for

the molecules.49

22

OAc

 

 

 

(ii) 0
O F O
I I10
i-Pr
(iii)

MeO

 

 

Figure 1-15. Examples of exciton Chirality CD method in absolute configurational
assignment.
1-3. Use of chromophoric host such as porphyrins for ECCD studies

As mentioned above, the Exciton Coupled Circular Dichroic (ECCD) method is a
strong tool for assigning the absolute stereochemistry of chiral compounds in a non-
empirical fashion. However, use of the ECCD method to determine the absolute
stereochemistry of any chiral compound requires the presence of at least two interacting
chromophores in the chiral molecule. This fact could limit the use of this method, since
majority of chiral compounds do not fulfill this requirement. In order to address the
above issue, chromophoric receptors have been designed to act as hosts for the non-

chromophoric chiral molecule. Upon binding of the chiral substrate, either covalently or

23

non-covalently, the host will adopt a preferred chiral conformation, induced by the chiral
guest, which can be observed as ECCD couplet.
l-3.1. Use of chromophoric hosts

Chiral induction, where a chiral molecule induces Chirality in an achiral molecule
by intermolecular forces, was apparently first demonstrated on dyes bound to
polypeptides in their helical conformations.56 It remains a tool for probing the

conformations of macromolecules.57

The process for transmission of configurational
information comprise of, at least, two elementary processes: (1) complex formation and
(2) some dynamic processes associated with it, such as conformational changes of
interacting molecules.58'60 If the interactions between two molecules are strong enough,
they will lead to efficient transmission of information.

The chromophoric hosts used for ECCD studies are achiral themselves, however,
upon binding to the chiral substrate, the chromophoric host will adopt an induced
Chirality dictated by the Chirality of the substrate through chiral induction. The induced
helicity of the chromophoric hosts could be observed and determined by ECCD, which in
turn will lead to the stereochemical determination of the bound chiral substrate. An

example of the induced Chirality in chromophoric host with chiral substrates is shown in

Figure 1-16.‘"
Br HOOH Br 1171 P H:
o o + OW =.
wN-CHZCHzBu‘ Br 00 Br
OZN Mo2 '

2
H
biphenol 1 chiral diamine 2

Figure 1-16. The induced Chirality of biphenol upon binding to chiral diamine.

24

In Figure 1-16, the biphenol compound 1 serves as the chromophoric host to bind
with the chiral diamine 2. The two phenyl rings in biphenol 1 are not in the same plane,
because of the steric interactions between the two hydroxyl group, thus they are not
conjugated with each other, and therefore, meet the criteria of chromophores that could
undergo exciton coupling.

Biphenol l and diamine 2 form a complex in solution through hydrogen bonding.
The hydrogen bonding between phenolic OH and aliphatic amines has been well
studied?“ Without chiral diamine 2, biphenol 1 has two chiral conformations, which
can rapidly convert to each other at room temperature. Biphenol 1 has no CD at room
temperature, but upon the addition of chiral diamine 2, an ECCD spectrum is observed.
The observed ECCD is due to the through space coupling of the two phenol groups and
indicates that Chirality has been induced in biphenol l by complexation with chiral
diamine 2. The Chirality adopted by biphenol 1 is dictated by the absolute
stereochemistry of the bound chiral diamine 2.

In a similar manner, the same has been applied to determine the absolute
stereochemistry of chiral secondary alcohols.65 As shown in Scheme 1-1, the
chromophoric reagent, 3-cyanocarbonyl-3’-methoxycarbonyl—2,2’binaphthalene 3 can be
esterified with chiral secondary alcohols and the resultant complex exhibits induced

ECCD as a result of favoring one atropisomer.

COOMe COOMe

0c COCN R'OH 0O coon*

0 DMAP, CH30N' O
3 O 0

Scheme 1-1. The derivatization of binaphthalene 3 with chiral alcohols.

 

25

The Chirality has been transmitted from the chiral alcohols to the binaphthylene
system by minimizing the steric interaction between the substituents on the chiral
alcohols with the methyl ester group in the binaphthylene. This conclusion has been
verified by molecular mechanics calculations, and it suggests that the transmission of the
Chirality occurred in a thermodynamically controlled fashion.65
1-3.2. Use of zinc porphyrin as chromophoric host

Porphyrins have received a great deal of attention as receptors for chiral
recognition due to several unique featuresz‘r’6

(1) Their planar structures provide a well-defined binding pocket that is
accentuated by substitutions on the ring. There are many sites that can be derivatized
such as the meso and ,B-positions, the central metal and the inner nitrogen atoms. By
varying substituents on the periphery of porphyrins the solubility of the porphyrin
containing compounds in non-polar and polar solvents can be easily modified;

(2) They are good chromophores, with absorption maximum around 418 nm for
the main absorption band (Soret band). The absorption of the porphyrins is far red-
shifted than most chromophores that likely preexist in the system, such as carbonyl and
alkenes. This prevents the unwanted interaction between the introduced chromophore
and the preexisting chromophores, which can complicate the analysis of the resulting
ECCD spectrum;

(3) They are highly chromophoric with 8 (molar absorption coefficient) higher
than 400,000. As it has been mentioned before, the amplitude (A) of the ECCD couplet is
proportional to the £2, so the sensitivity of CD will be greatly enhanced with the strong

absorbing chromophores. Therefore, porphyrins are ideal chromophores for detecting

26

 

subtle changes in their close environment such as binding of chiral ligands. These
intermolecular interactions can be easily followed by UV-vis, fluorescence, NMR, and
resonance Raman;(’6

(4) Metal incorporation into the porphyrin ring can be easily achieved.
Metalloporphyrins, such as zinc porphyrins and magnesium porphyrins, can provide extra
stereodifferentiation with their Lewis base binding sites.

There are two types of porphyrin chromophores: achiral porphyrins and inherently
chiral porphyrinsf'7 examples of each is shown in Figure 1-17. The achiral porphyrins

are planar molecules, while the chiral porphyrins are “saddle” shaped.

 

achiral porphyrin inherently chiral porphyrin side-view Of the chiral porphyrin

Figure 1-17. Achiral porphyrins and inherently achiral porphyrins.

Achiral porphyrins have been widely utilized in stereochemical studies by CD.

68, 69

These include absolute configurational assignments,“ stereochemical

7|

differentiations of sugars and amino acids,7°‘ and interactions with bio-

macromolecules.” 73
Porphyrins have been used broadly in UV-vis and CD study because of their far
red-shift absorption band and their high absorption. The porphyrin free base generally

has four absorption bands (Q band) in the visible region. The very intense band

27

appearing around 415 nm in the near ultraviolet is the B band, and more commonly
referred to as Soret band. Generally, the concentration of the porphyrin solution can be
determined by the intensity of the Soret band based on the Lambert-Beer’s Law: A = 8C1 .
There are also additional bands (N, L, M bands) at the ultraviolet regions but these bands
are usually much weaker. For porphyrins with different substituents on the periphery, the
positions of the Soret band and the other bands vary a little.

Upon formation of metalloporphyrins, such as zinc porphyrins, the four Q bands
in the visible region collapse into essentially two bands due to higher symmetry, while
the Soret band may remain in the usual range or shifted to higher or lower energy. The
molar absorption coefficient (8) of the Soret band changes with the insertion of metals in
the porphyrin ring. As shown in Figure 1-18, zinc porphyrin methyl ester (.E‘—--550,000)74
has higher molar absorption coefficient in the Soret band than the porphyrin free base

(lit-440,000),75 and the position of this band is also slightly red shifted.

28

 

porphyrin methyl ester zinc porphyrin methyl ester

0.4 ,

Soret Band
418 (nm) expand Q band

a... 57/ M

 

 

 

350 525 700
wavelength (nm)

0.4
Soret Band

419 (nm) expand Q band

 

550

 

 

 

 

 

350 525 700
wavelength (nm)

Figure 1-18. UV-vis spectra of porphyrin methyl ester and zinc porphyrin methyl ester.

29

Porphyrins, as the alkyl connected bis-metalloporphyrins 4 shown in Figure 1-19,
can be used as host systems for stereochemical determination of chiral compounds
because of their ability to bind to the chiral compounds and report the Chirality of bound

guests based on Exciton Coupled Circular Dichroic spectroscopy (ECCD).76'81

 

Figure 1-19. Structure of zinc porphyrin tweezer 4.

Zinc porphyrin tweezer 4, as shown in Figure 1-19, can undergo host-guest
complexation with chiral diamines and adopt a chirally oriented conformation (helicity)
leading to a bisignate ECCD spectrum. The helicity of the interacting chromophores can
be observed by CD and is exhibited as bisignate ECCD spectrum. The helicity of the

receptor in turn reflects the Chirality of the bound compound.

30

 

O zinc porphyrin \Zn~NH Q£H2W2n

“1902014” 5 fi tweezer A?
H2N NH2 HZN [\ng

 

O 0W0 O

 

 

 

 

 

 

 

 

 

 

 

 

 

 

100 b
(a) e 426 (Benzene) ( )
d 50 .. (+90)
L ‘ 0 T' T " “
O =>Ae [
' -50 -
_ __ 434
’ 100 ('117)
-150 1 I -
negative ECCD 350 virgSelen th 315:) 500 positive ECCD
couplet g couplet

Figure 1-20. Assignment of absolute configuration of acyclic chiral diamines by Exciton
Coupled Circular Dichroism (ECCD).

As shown in Figure 1-20, when zinc porphyrin tweezer 4 is added to the solution
of chiral diamine (S)-methyl-2,4-diaminobutyrate, the porphyrin tweezer binds to the
chiral diamine via Zn-N coordination to form a “sandwich” structure.8|

The porphyrin, which is far away from the chiral center, does not interact with the
chiral center. However, the porphyrin, coordinated to the end which is close to the chiral
center, will interact with the groups on the chiral center in a way to minimize the steric
interactions between the porphyrin ring and the chiral substituents. In order to minimize
the steric interaction, the porphyrin ring ‘slides away’ from the large group on the chiral
center and covers the small group (Figure 1-20a). This conformation is more stable than

the alternate (Figure l-20b), in which the porphyrin ring ‘slides away’ from the small

group and interacts with the large group instead. The more stable conformation (a), leads

31

to the two porphyrin planes adopting a counterclockwise orientation and a negative

ECCD couplet is predicted. This prediction matches the CD observation.

 

 

1. ”\xNHz 0
H COOH BOCHN : H NH2
NHBoc 2. Deprotection L NH;
H 1. /\/C02H O
NHBoc 2. Deprotection L NH2

Scheme 1-2. Converting the amino acids and amino alcohols to diamines.

This approach has been extended to determine the absolute stereochemistry of
chiral amino acids and amino alcohols, which have a primary amino group directly
attached to the stereogenic center.“ After simple chemical modifications as shown in
Scheme 1-2, the amino acid and amino alcohols can be converted to diamines, whose
stereochemistry can be determined by the method mentioned above.

Zinc porphyrin tweezer 4 cannot be used directly to determine the absolute
stereochemistry of chiral compounds that have only one site of attachment, since two
chromophores need to be bound and interact through space in order to observe ECCD.
To overcome this problem, carrier molecules have been designed to convert compounds
with only one site of attachment to molecules possessing two sites of attachment.
Carriers are achiral molecules, which can be derivatized with the chiral substrates to
provide the requisite two sites of attachment for binding to zinc porphyrin tweezer.

The carrier molecules need to possess the following characteristics:78
(1) They should have a functional group that can be derivatized with the chiral

compounds. For example, a carboxylic acid will serve well when a chiral amine is to be

analyzed;

32

(2) The derivatized carriers should be able to bind to the zinc porphyrin tweezer 4
or a comparable host at 1:1 ratio;

(3) The carrier derivatized conjugates should be rigid, since flexible structures
will give multiple conformations leading to weak or even complex CD spectra;

(4) The carriers should be achiral, since the introduction of the extra chiral center
into the system will result in diastereomers, which will complicate interpretation of the
ECCD spectrum.

Ultimately, the success of this carrier method relies on identifying the most
energetically favored conformation of the derivatized carrier. This is imperative because
the conformation of the derivatized carrier dictates the position of the substituents at the
chiral center in space relative to the porphyrins, and therefore, will dictate the spacial
orientation adopted by the two porphyrin planes. The ideal carrier, upon derivatization
with chiral compounds, exhibits a rigid and predictable conformation about the free
rotating single bonds. Rotation along any of the single bond can dramatically alter the
position of the large, medium, and small group with respect to the porphyrins.

Following these guidelines, carrier 5 has been designed and synthesized in order
to determine the absolute stereochemistry of chiral monoamine in conjunction with zinc
porphyrin tweezer.78 It can be derivatized with chiral monoamines by the carboxylic acid
functionality to form an amide bond. After the Boo protecting group is removed, it
provides two binding sites to coordinate with zinc porphyrin tweezer. This conjugate can
bind to the zinc porphyrin tweezer to form a complex, the stereochemistry of which can

be assigned based on the Exciton Coupled Circular Dichroic method.

33

As Figure 1-21 shows, derivatization of (R)-1-cyclohexylethylamine with carrier
5, followed by deprotection, leads to conjugate 6 with two nitrogen atoms, the pyridine
nitrogen and the free amino nitrogen, serving as the binding handle for both zinc sites of
porphyrin tweezer 4. The complexation leads to a unique helical arrangement of the two
porphyrin planes with respect to each other, and can be observed by ECCD spectrum.
The correlation between the sign of the ECCD spectrum and the absolute stereochemistry

of the chiral conjugate 6 can be deduced by the following reasoning.

 

H i i
Ligcoali HZN-iCHS quiche
NHBoc NH2
5 6

In the most stable conformation:

S.~CH3 o The carbonyl oxygen is anti to the pyridine ring due the
”N H electrostatic repulsion.
IN‘ 0
/ o The carbonyl oxygen is syn to the small group (H) on the
NH2 chiral center to reduce the pseudo A13 strain.

Figure 1-21. The chiral complex and its most stable conformation.
Molecular modeling of the conjugate 6 suggests that it is a rigid molecule and it
has a predictable energetically favored conformation.78 In this most favorable

conformation, the free rotations around the single bonds: CAr-Cc=0, Cc=o-NH, and NH-
Cchim. are restricted. Molecular modeling of the conjugate has shown that the conjugate 6

favors a coplanar orientation of the carbonyl group to the pyridine ring. Moreover, due to
electrostatic repulsion, the anti carbonyl oxygen/pyridine nitrogen disposition is preferred

by ~30 kJ/mol. Finally, it has been well understood that in an amide, due to the pseudo

34

A13 interaction, the small group on the chiral center (hydrogen) is syn to the carbonyl
oxygen.

Since the spacial arrangement of the substituents of the chiral center can be
deduced from this favored conformation, the orientation of the porphyrin planes can be
predicted based on the interaction between the chiral compound and the porphyrin planes.

The three substituent groups at the chiral center are assigned S (small), M
(medium), and L (large) according to their size,8 with hydrogen being the S group,
methyl being M group and cyclohexyl being L group. As shown in Figure 1-22, upon
binding to zinc porphyrin tweezer 4, the chiral substrate will adopt a conformation such
that L points out of the plane formed by the amide and the pyridine ring, while M points
into the plane. The zinc porphyrin’s approach of the pyridine nitrogen will bias the side
of M group in order to avoid the unfavorable steric interaction with L. Thus, the two
porphyrin planes of the tweezer orient in a clockwise way, and dictate a positive ECCD

spectrum. This prediction matches with the experimental observation.

35

 
 
   

small steric large steric
influence 7 influence

(7 6)
18H i O H i ‘ 8. coordination p
COZH HzN N\ N’i' Syni / site
[87 | / H CH3 :{> .
NHBoc NH2
5 6 coordination e
site
M
OH

L
2“ Z" '2 \1 ”MEL
W—N 22/“

 

 

 

 

 

 

 

 

 

6
o o’\/4\/‘o o
O HOWO 0
20° 432 C?“
. (+135)
A5 L
0 <2:
V...
i (-108)
-200 . . . .
410 420 430 440

 

wavelength (nm) preferred conformation
of the complex

Figure 1-22. Use of carrier 5 to determine the absolute stereochemistry of chiral
monoamine by Exciton Coupling Circular Dichroism.

The goal of our research is to further the above studies by determining the
absolute stereochemistry of chiral compounds with different functional group, namely 0t—
chiral carboxylic acids, by using the stated Exciton Coupled Circular Dichroic (ECCD)
method. In order to achieve that goal, we have studied the synthesis of zinc porphyrin
tweezers, the design and synthesis of different carriers for a-chiral carboxylic acids, and

their spectroscopic studies.

36

Reference:

1.

10.

11.

12.

l3.

14.

15.

16.

Bruice, P. Y. Organic chemistry, 3rd ed.; Prentice-Hall, Inc.: Upper Saddle River,
New Jersey, 2001.

Rossi, R.; Diversi, L. Synthesis 1973, 25.
Laarhoven, W. H.; Prinsen, W. J. C. Top. Curr. Chem. 1984, 125, 63.

Eliel, E. L. Stereochemistry of Carbon Compounds; McGraw-Hill: New York,
1962.

Oki, M. Recent Advances in Atropisomerism. In Topics in Stereochemistry;
Allinger, N. L., Eliel, E. L., Wilen, S. W., Eds.; An Interscience Publication: New
York, 1983.

Humpf, H. -U.; Berova, N.; Nakanishi, K.; Jarstfer, M. B.; Poulter, C. D. J. Org.
Chem. 1995, 60, 3539.

Dale, J. A.; Moser, H. S. J. Am. Chem. Soc. 1973, 95, 512.

Eliel, E. L.; Wilen, S. W.; Mander, L. N. Stereochemistry of Organic Compounds;
Wiley: New York, 1994.

Lambert, J. B.; Shurvell, J. F.; Lightner, D. A.; Cooks, R. G. Organic Structural
Spectroscopy; Prentice Hall: Upper Saddle River, 1998.

Solomons, T. W. G. Organic Chemistry; John Wiley & Sons. Inc.: New York,
1978.

Velluz, L.; LeGrand, M.; Grosjean, M. Optical Circular Dichroism; Academic
Press Inc.: New York, 1965.

Mislow, K.; Glass, M. A. W.; O’Brien, R. E.; Rutkin, P.; Steinberg, D. H.; Weiss,
J.; Djerassi, C. J. Am. Chem. Soc. 1962, 84, 1455.

Snatzke, G. Angew. Chem. Int. Ed. Engl. 1979, 18, 363.
Nogrady, M. Stereochemistry; Pergamon Press Ltd.: Oxford, 1981.

Berova, N.; Nakanishi, K.; Woody, R. W. Circular Dichroism: Principles and
Applications; Wiley-VCH: New York, 2000.

Moffitt, W.; Woodward, W. B.; Moscowitz, A.; Klyne, W.; Djerassi, C., J. Am.
Chem. Soc. 1961, 83.

37

17.

18.

19.

20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

30.

31.

32.

33.

34.

35.

36.

37.

Snatzke, G. Optical Activity and Chiral Discrimination; Reidel, Dordrecht, and
Neatherlands: 1979.

Polavarapu, P. L. Molec. Phys. 1997, 91, 551.

Polavarapu, P. L.; Chakraborty, D. K. J. Am. Chem. Soc. 1998, 120, 6160.
Amos, R. D. Chem. Phys. Lett. 1982, 87, 23.

Kondru, R. K.; Wipf, P.; Beratan, D. N. J. Am. Chem. Soc. 1998, 120, 2204.
Amos, T. D.; Rice, J. E.; deell S. M. J. Chem. Phys. 1990, 93, 8828.
Kondru, R. K.; Wipf, P.; Beratan, D. N. Science 1998, 282, 2247.

Kondru, R. K.; Lim, S.; Wipf, P.; Beratan, D. N. Chirality 1997, 9, 469.
Kondru, R. K.; Wipf, P.; Beratan, D. N. J. Phys. Chem. A 1999, 103, 6603.

Ribe, S.; Kondru, R. K.; Beratan, D. N.; Wipf, P. J. Am. Chem. Soc. 2000, 122,
4608.

Kondru, R. K.; Beratan, D. N .; Friestad, G. K.; Smith, 1., A. B.; Wipf, P. Org. Lett.
2000, 2, 1509.

Kondru, R. K.; Chen, C. H.; Curran, I). P.; Beratan, D. N.; Wipf, P. Tetrahedron:
Asymmetry 1999, 10, 4143.

Harada, N.; Nakanishi, K. Circular Dichroic Spectroscopy: Exciton Coupling in
Orgainc Stereochemistry; University Science Books: Mill Valley, 1983.

Kuhn, W. Trans. Faraday Soc. 1930, 26, 293.

Kirkwood, J. G. J. Chem. Soc. 1937, 5, 479.

Wiesler, W. T.; Vazquez, J. T.; Nakanishi, K. J. Am. Chem. Soc. 1987, 109, 5586.
Harada, N .; Chen, S. -M. L.; Nakanishi, K. J. Am. Chem. Soc. 1975, 97, 5345.
Chen, S. -M. L.; Harada, N.; Nakanishi, K., J. Am. Chem. Soc. 1974, 96, 7352.
Harada, N .; Nakanishi, K. Acc. Chem. Res. 1972, 5, 257.

Wagniere, G.; Hug, W, Tetrahedron Lett. 1970, I I, 4765.

Hug, W.; Wagniere, G. Tetrahedron 1972, 28, 1972.

38

38.

39.

40.

41-

42-

43-

44-

45-
46-
47-
48-

49-

SO-

51.

52.
53.
54.

55.

56.

57.

Weckerle, B.; Schreiber, P.; Humpf, H. -U. J. Org. Chem. 2001, 68, 8160.
Lichtenthaler, F. W.; Sakakibara, T. Carbohydr. Res. 1977, 59, 47.
Liu, H. W.; Nakanishi, K. J. Am. Chem. Soc. 1981, 103, 7005.

Dong, J. -G.; Akritopoulou-Aanae, I.; Guo, J.; Berova, N.; Nakanishi, K.; Harada,
N. Enantiomer 1997, 2, 397.

Harada, N.; Iwabuchi, J.; Yokota, Y.; Uda, H.; Nakanishi, K. J. Am. Chem. Soc.
1981, 103, 5590.

Gonnella, N. C.; Nakanishi, K.; Martin, V. S.; Sharpless, K. B., J. Am. Chem. Soc.
1974, 104, 3775.

Mori, Y.; Kohchi, Y.; Suzuki, M.; Furukawa, H. J. Am. Chem. Soc. 1992, 114,
3557.

Goto, T.; Kondo, T. Angew. Chem. Int. Ed. Engl. 1991, 30, 17.

Matile, S.; Berova, N.; Nakanishi, K. Chem. & Biol. 1996, 3, 379.

Rele, D.; Zhao, N.; Nakanishi, K.; Berova, N. Tetrahedron 1996, 52, 2795.
Wiesler, W. T.; Nakanishi, K. J. Am. Chem. Soc. 1989, 111, 9205.

Zhao, N.; Berova, N.; Nakanishi, K.; Rohmer, M.; Mougenot, P.; Jurgnes, U. J.
Tetrahedron 1996, 52, 2777.

Gimple, O.; Schreiber, P.; Humpf, H. -U. Tetrahedron: Asymmetry 1997, 8, l 1.

Nakanishi, K.; Berova, N. Circular Dichroism. Principles and Applications; VCH
Publishers: New York, 1994.

Harada, N. J. Am. Chem. Soc. 1973, 95, 240.
Koreeda, M.; Weiss, 6.; Nakanishi, K. J. Am. Chem. Soc. 1973, 95, 239.
Tachibana, K.; Sakaitani, M.; Nakanishi, K. Science 1973, 226, 703.

Adams, M.; Nakanishi, K.; Still, W. C.; Arnold, E. V.; Clardy, J .; Presoons, C. J.
J. Am. Chem. Soc. 1979, 101, 2495.

Blout, E. R.; Stryer, L. Proc. Natl. Acad. USA 1959, 45, 1591.

Stryer, L.; Blout, E. R. J. Am. Chem. Soc. 1961, 83, 1411.

39

58.

59.

60-

61-

62-

63-
54-
65-
66-
67.
68-
69-

70-

71.

72.

73.

74.

75.

76.

Beer, P. D.; Rothin, A. S. J. Chem. Soc., Chem. Commun. 1988, 52.

Mizutani, T.; Yagi, S.; Honmaru, A.; Ogoshi, H. J. Am. Chem. Soc. 1996, 118,
5318.

Hayashi, T.; Asai, T.; HoKazono, H.; Ogoshi, H. J. Am. Chem. Soc. 1993, 115,
12210.

Mizutani, T.; Takagi, H.; Hara, 0. Tetrahedron Lett. 1997, 38, 1991.

Hanessian, S.; Gomtsyan, A.; Simard, M.; Roelens, S. J. Am. Chem. Soc. 1995,
117, 7630.

Kramer, R.; Zundel, G. J. Chem. Soc., Faraday Trans. 1990, 86, 301.
Bell, C. L.; Barrow, G. M. J. Chem. Phys. 1959, 36, 5.

Hosoi, S.; Kamiya, M.; Ohta, T. Org. Lett. 2001, 3, 3659.

Ogoshi, H.; Mizutani, T. Acc. Chem. Res. 1998, 31, 81.

Mizutani, T.; Ema, T. J. Am. Chem. Soc. 1994, 116, 4240.

Matile, S.; Berova, N.; Nakanishi, K. J. Am. Chem. Soc. 1995, 117, 7021.
Arimori, S.; Takeuchi, M.; Ahinkai, S. J. Am. Chem. Soc. 1996, 118, 245.

Crossley, M. J .; Mackay, L. 6.; Try, A. C., J. Chem. Soc., Chem. Commun. 1995,
1925.

Crossley, M. J .; Hambley, T. W.; Mackay, L. 6.; Try, A. C.; Walton, R. J., Chem.
Soc., Chem. Commun. 1995, 1077.

Suenaga, H.; Arimori, S.; Shinkai, S. J. Chem. Soc., Perkin Trans. 2 1996, 607.

Foster, N.; Singhal, A. K; Smith, M. W.; Marcos, N. G.; Schra, K. J. Biochim.
Biophys. Acta. 1988, 950, 118.

Matile, S.; Berova, N.; Nakanishi, K.; Fleischhuer, J .; Woody, R. W. J. Am. Chem.
Soc. 1996, 118, 5198.

Rickman, B. H.; Matile, S.; Nakanishi, K.; Berova, N. Tetrahedron 1998, 54,
5041.

Huang, X.; Fujioka, N.; Pescitelli, G.; Koehn, F. E.; Williamson, R. T.; Nakanishi,
K.; Berova, N. J. Am. Chem. Soc. 2002, 124, 10320.

40

77.

78.

79-

80-

81-

Huang, X. F.; Nakanishi, K.; Berova, N. Chirality 2000, 12, 237.

Huang, X.; Borhan, B.; Rickman, B. H.; Nakanishi, K.; Berova, N. Chem. Eur. J.
2000, 6, 216.

Kurtan, T.; Nesnas, N.; Li, Y. Q.; Huang, X. F.; Nakanishi, K.; Berova, N. J. Am.
Chem. Soc. 2001, 123, 5962.

Proni, G.; Pescitelli, G.; Huang, X. F.; Quraishi, N. Q.; Nakanishi, K.; Berova, N.
Chem. Commun. 2002, 1590.

Huang, X. F.; Rickman, B. H.; Borhan, B.; Berova, N.; Nakanishi, K. J. Am.
Chem. Soc. 1998, 120, 6185.

41

Chapter 2
Polymer-supported synthesis of mono-substituted porphyrins

Porphyrins have received a great deal of attention as receptors for chiral

recognition due to several unique features:1

(1) Their planar structures provide a well-defined binding pocket that is
accentuated by substitutions on the ring. There are many sites that can be derivatized
such as the mesa and ,B-positions, the central metal and the inner nitrogen atoms. By
val-ying substituents on the periphery of porphyrins, the solubility of the porphyrin
containing compounds in non-polar and polar solvents can be easily modified;

(2) They are good chromophores, with absorption maximum around 418 nm for
the main absorption band (Soret band).2 The absorption of the porphyrins is far red-
shifted than most chromophores that likely preexist in the system, such as carbonyl and
alkenes. This prevents the unwanted interaction between the introduced chromophore
and the preexisting chromophores, which can complicate the analysis of the resulting
ECCD spectrum;

(3) They are highly chromophoric with a (molar absorptivity coefficient) higher
than 400,000.3 As it has been mentioned before, since the amplitude (A) of the ECCD
couplet is proportional to the 52, the sensitivity of CD will be greatly enhanced when a
strong absorbing chromophore is used. So porphyrins are ideal chromophores for
detecting subtle changes in their close environment such as binding of chiral ligands.
“358 intermolecular interactions can be easily followed by UV-vis, fluorescence, NMR,

and resOnance Raman;l

42

(4) Metal incorporation into the porphyrin ring can be easily achieved.
Metalloporphyrins, such as zinc porphyrins and magnesium porphyrins, can provide extra
stereodifferentiation with their Lewis base binding sites.

2- l. The synthesis of the zinc porphyrin tweezer
The zinc porphyrin tweezer has been demonstrated to be successful in
determining the stereochemistry of chiral compounds, such as chiral diamines, amino
acids, and amino alcohols through host-guest complexes.4 The porphyrin tweezer can be
easily synthesized from mono-substituted porphyrins, such as porphyrin carboxylic acids
and alcohols. The carboxylic acid or alcohol functionality at the mesa position of the
mono-substituted porphyrin is used to connect with an alkyl chain to form the porphyrin

tweezer (Scheme 2—1).

  
 

1. 1,5-pentanediol
EDC, DMAP

2. Zn(OAc)2

 

Scheme 2-1. Synthesis of zinc porphyrin tweezer 1 from porphyrin carboxylic acid.
The synthesis of the zinc porphyrin tweezer 1 depends heavily on the synthesis of
monO—substituted porphyrins. As shown in Scheme 2-2, the mono—substituted porphyrin
methyl ester 2 could be synthesized by condensing pyrrole, benzaldehyde and 4-
carbox)lniethylbenzaldehyde in refluxed propionic acid at 423:] ratio.4 Zinc acetate was

used ‘0 serve as a template to help the assembly of the porphyrin rings. After the reaction

43

mixture was refluxed in propionic acid for several hours, DDQ was added to the reaction
mixture as an oxidation reagent to finish the synthesis of porphyrin compounds. The zinc

was removed from the porphyrins by washing it with 6 N hydrochloric acid.

CHO 1. ZD(OAC)2
Cl-lO propionic acid

is r: r.
reux
cu . +©+ -

non-substituted, di,
tri, and tretra-

   

 

7 substituted
002Me 2- DDQBCHzclz porphyrin products
4 eq. 3eq. 1 eq. 7 °
0 OMe
mono-substituted
porphyrin 2

Scheme 2-2. Synthesis of mono-substituted porphyrin methyl ester 2.
This synthesis leads to the formation of the desired mono-substituted porphyrin 2
along with a mixture of tetraphenylporphyrin, and di-, tri-, and tetra-substituted
porphyrins. The separation of the desired mono-substituted porphyrin from all the

byproduct is tedious, however, it can be achieved by multiple column chromatographies.

NaOH (2 N) /
THF, reflux,
overnight, 98%;

 

 

Scheme 2-3. Hydrolysis of porphyrin methyl ester 2 to carboxylic acid 3.
After the separation of the mono-substituted porphyrin methyl ester 2, it was

SUPSequently hydrolyzed under basic condition to yield the corresponding porphyrin

44

carboxylic acid 3 (Scheme 2-3). Since porphyrin methyl ester 2 does not dissolve in
water, THF was used in the hydrolysis step as a co-solvent for better solubility. After the
reaction mixture was refluxed overnight, it was acidified with 6 N hydrochloric acid and
extracted with CHzClz. The porphyrins are purple red in basic solutions, and when the
solution is acidified, a color change from purple red to green can be clearly observed.

The extraction of porphyrin acid 3 with CHzClz is tricky. Before extracting, THF
has to be removed. The porphyrin carboxylic acid 3 dissolves much better in THF than
in CHzClz because of its high polarity. Since THF is partially soluble in water, its
presence would make the extraction of porphyrin acid 3 harder. Small amount of
methanol needs to be added into CHzClz to help the extraction of the porphyrin acid 3,

however, since the porphyrin acid does not dissolve in MeOH, too much methanol will
reduce the solubility of porphyrin acid. The optimal ratio of methanol and CHzClz for
porphyrin acid extraction is about 1:100.

After the extraction of the acidified porphyrin with the mixture of methanol and
CHZClz, the organic layer was neutralized. Instead of using a basic solution, water was
used to wash away the acid, since the base was likely to form the carboxylate salt of the
Porphyrin carboxylic acid 3. The porphyrin acid 3 separated in this way is pure enough

for further reactions.

45

1 ,5-pentanediol

EDC, DMAP
CHZCIZ

   

Scheme 2-4. EDC coupling of porphyrin carboxylic acid 3 with 1,5-pentanediol.
After securing porphyrin carboxylic acid 3, a simple esterification with 1,5-
pen tanediol, as shown in Scheme 2-4, could provide the porphyrin tweezer 4. However,
the reaction under standard ECD coupling condition did not proceed as expected. TLC
analysis indicated the presence of many porphyrin spots with very close polarity, and it
was impossible to isolate any pure product by column chromatography. The reaction had
been repeated many times and each time the same result was observed. Finally, it was
realized that the reaction did not work because the porphyrin carboxylic acid 3 was not
pure. It contained both the free-base porphyrin acid and zinc porphyrin acid. Not only
could the zinc porphyrin acid interact with the coupling reagent EDC to reduce the yield
of the coupling product, but also the presence of both the free-base porphyrin acid and
Zinc POI‘phyrin acid in the EDC coupling reaction resulted in three porphyrin tweezer
PFOducts: porphyrin tweezer free base 4, di-zinc porphyrin tweezer 1 and mono-zinc
p Orphyr in tweezer. The polarities of the three tweezers were very close to each other,
and it was impossible to isolate them by column chromatography.

The procedure was altered to remove zinc from the first step of synthesizing

p Orphy r in methyl ester 2. Zinc acetate was used to aid in the synthesis of porphyrins, and

46

was to be removed by washing with 6 N hydrochloric acid. In our case, because of the
scale of the reaction, not all the zinc was removed from the porphyrin. In order not to
introduce zinc, the porphyrin methyl ester 2 was synthesized again by the same propionic
acid method but without the addition of zinc acetate. The separated yield of the desired
mono-substituted porphyrin methyl ester 2 was 7%, comparable to that of the reactions
with the addition of zinc acetate.

After the isolation of the pure porphyrin methyl ester 2, it was hydrolyzed to
porphyrin carboxylic acid 3 as described above. The EDC coupling of porphyrin acid 3
with 1,5-pentanediol proceeded very smoothly, and provided the desired porphyrin
tweezer 4 in good yield. Depending on the ratio between the porphyrin acid 3 and 1,5-
pentanediol, the synthesis of porphyrin tweezer 4 can be finished in a single step or step-
wise, i.e., synthesize the monoester, isolate and submit the product to another round of
esterification for yield the diester, or formation of the diester in one pot. The yield of the

step-wi se synthesis was higher than the single step synthesis.

Excess zinc acetate was then added to the porphyrin tweezer free—base 4 to yield
the zinc porphyrin tweezer 1. The excess zinc acetate could be easily removed by simple
filtration. However, even starting with a pure porphyrin tweezer free base 4, flash
column chromatography was required to secure the pure zinc porphyrin tweezer 1 due to

the decomposition of porphyrins.
A much milder condition was also used to synthesize porphyrin methyl ester 2.
As shown in Scheme 2-5, the porphyrin was synthesized in CHzClz with boron trifluoride
diethyl etherate (BF3-Et20) as the acidic catalyst.5' 6 Under these conditions, the reaction

c . . .
an be Carried out at room temperature instead of refluxrng at 160 0C.

47

CH0 1. BFa-EIZO,
CHO CH20|2

i\ /7 + if: + 2. p-chloranil,
002Me 25%
4 eq. 3 eq. 1 eq.

non-substituted, di,
tri, and tretra-
substituted
porphyrin products

 

mono-substituted
porphyrin 2

Scheme 2-5. The synthesis of mono-substituted porphyrin 2 in BF3-Et20 condition.

Pyrrole, benzaldehyde and 4-carboxymethylbenzaldehyde were dissolved in
CHzClz at 423:] ratio for the optimal yield of the mono-substituted product. Catalytic
amount of BF3-Et20 was added into the reaction, and after two-hour stirring at room
temperature, p-chloranil was added to finish the synthesis. The isolated yield of the
desired mono—substituted porphyrin 2 was improves from 7% (propionic acid condition)
to 25%.

Using BF3-Et20 as the catalyst, one has to be very careful about removing all the
air from the reaction. BF3-EtzO is very sensitive to moisture, and since only catalytic
amount of BF3-EtzO is used, small amounts of moisture could derail the reaction.
Therefore, the reaction solution was purged with nitrogen for at least 15 minutes before
BF3-EtzO was added.

2-2. The synthesis of mono-substituted porphyrins on polymers

As shown above, the synthesis of zinc porphyrin tweezer 1 depends heavily on

procuring mono-substituted porphyrins efficiently. The synthesis of structurally complex

7-

porphyrins has seen tremendous progress in the last 30 years, '0 and as such the synthesis

of mono-substituted porphyrins methyl ester 2 can be achieved by utilizing many of the

48

 

 

methods that have been developed.H However, the most common and direct
methodology to synthesize these mono-substituted porphyrins is to condense a mixture of
pyrrole, benzaldehyde, and the appropriately substituted benzaldehyde in a 4:3:1 ratio in
acidic condition. This leads to the synthesis of the desired mono-substituted porphyrin
together with a mixture of tetraphenylporphyrin, and di-, tri-, and tetra-substituted
compounds, which makes the chromatographic separation of these compounds tedious
and time consuming. In most cases, multiple flash column chromatographies are required
to isolate the pure desired porphyrin product.

As Scheme 2-2 and 2-5 shows, propionic acid and BF3'E120 are two commonly
used catalysts for the synthesis of mono-substituted porphyrins.4’ 6 The BF3-EtzO is much
milder than using propionic acid and the yield has also been improved, but it still
generates the unwanted byproducts. The propionic acid condition yield ratio between the
non-, mono-, and di-substituted compounds of 1:3: 1, and for the BF3-EtzO condition, the
ratio between non-, mono- and di-substituted compounds is 1:26: 1.3. The generation of
porphyrin byproducts not only leads to tedious and time consuming chromatographic
separations, but also, depending on the substituted benzaldehyde utilized in the reaction,
wasting of this compound by synthesizing undesired porphyrin can be synthetically
limiting. In order to simplify the isolation of desired mono-substituted porphyrin and
also improve the yield, we have modified the synthesis of mono-substituted porphyrins
by using insoluble and soluble polymers.

The method of utilizing insoluble polymer for the synthesis of mono-substituted
porphyrins was first developed by Leznoff and Svirskaya.12 There are several advantages

in synthesizing mono-substituted porphyrins on polymer supports.l3 First, it is easy to

49

separate the polymer-supported species from the small molecules present in the reaction
medium by simple filtration. The easy isolation of the polymer bound species in turn
allows for the use of excess soluble starting material to achieve higher yields. Second,
the size of the polymer renders the resin-bound species inaccessible to each other, and
therefore, cross-reactions are avoided. In the porphyrin synthesis, since the substituted
aldehyde is anchored onto the polymer, this advantage excludes the formation of di-, tri-
and tetra-substituted porphyrins. Finally, the use of polymers in the synthesis of mono-
substituted porphyrins eases the purification of the desired product.

In the strategy presented here, the substituted aromatic aldehyde that will
eventually become the mono-substituted portion of the final porphyrin is anchored onto a
polymeric backbone. This allows for the construction of the porphyrin on the polymeric
support with pyrrole and an aldehyde in solution. In this manner free mono-substituted
aldehyde is not utilized and thus di, tri, and tetra-substituted porphyrins are not
synthesized. Therefore, on the insoluble polymer, there is only mono-substituted
porphyrin, and there is only non-substituted porphyrin (tetraphenylporphyrin) in the
reaction solution.

2-2.1. Synthesis of mono-substituted porphyrin on insoluble polymer

The synthesis of mono-substituted porphyrin carboxylic acid 3 on insoluble
polymer is illustrated in Scheme 2-6. 4-Carboxybenzaldehyde 5 was attached to a 2%
crosslinked divinylbezene/polystyrene beads 6, containing 1 mmole/eq. of benzyl
chloride by reported procedures,M which results in 85% conversions from the chloride to
the attached aldehyde. The conversion was determined by releasing the aldehyde from

the polymer support by refluxing in sodium hydroxide. The functionalized polymer

50

support with the bound aldehyde 7 was used in the synthesis of the mono-substituted
porphyrins. Addition of excess of pyrrole and benzaldehyde to the polymer 7 in
refluxing propionic acid followed by DDQ oxidation yielded the mono-substituted
product 8 on the resin and only tetraphenylporphyrin (TPP) in the solution phase. Since
benzaldehyde and pyrrole are inexpensive reagents, the sacrifice in making
tetraphenylporphyrin in solution was well compensated by the fact that only the mono-

substituted porphyrin was synthesized on the polymer.

OCHO

op“ H0200 00:10,,
EtaN, EtOAc

6 reflux, 2 h, 85%

 

CHO

€17 (excess)

propionic acid

reflux, 4 h 0
2- $823322: EDC, 66 + ; tetraphenyl
’ 3 porphyrin in 5

_ solution .
[O = Insoluble polymer] ------------------

   

 

 

p0---- --

 

1. wash resin with CHZCIZ

2. NaOH, THF, reflux,
12 h, 6% for four steps

 

 

Scheme 2-6. Synthesis of mono-substituted porphyrins on insoluble polymer.
The tetraphenylporphyrin generated in solution could be easily removed from the
polymer-supported porphyrins by washing the polymer resin with CHzClz (Soxhlet
extraction). Cleavage of the covalently attached porphyrin from the beads in refluxing

aqueous NaOH solution led to the isolation of the desired porphyrin carboxylic acid 3 in

51

good purity and 6% overall yield. The yield is comparable to solution phase synthesis of
3 in refluxing propionic acid,4 however, the crude isolation is not complicated with
mixtures of multi-substituted porphyrins, and thus the purification proves to be
straightforward from the polymer beads.

The porphyrin synthesis on insoluble polymer had also been tried by using
BF3-Et20 as the acidic catalyst instead. Unfortunately, it was found that BF3-EtzO was
not compatible with the insoluble polymer, since not only there was no porphyrin
carboxylic acid isolated, but also there was no tetraphenylporphyrin (TPP) forming in the
solution. The lack of the tetraphenylporphyrin generation in the reaction indicated that
the BF3-Et20 failed to serves as the acidic catalyst to generate porphyrins. This reaction
was repeated several times, and every time the result was the same. So the porphyrin
synthesis on insoluble polymer can only be carried out in refluxing propionic acid
condition.

2-2.2. Synthesis of mono-substituted porphyrin with soluble polymer

Limitations in the use of solid polymers in synthetic chemistry are pronounced by
the difficulty in using NMR to characterize intermediates, and the heterogeneous nature
of the chemistry that could result in low yields. However, soluble polymers can be used

as an alternative matrix for organic synthesis.'5

These polymers are non-cross-linked
long chains, and exhibit both soluble and insoluble characteristics depending on the
solvent used in the reaction. Synthetic approaches that utilize soluble polymers couple
the advantages of homogeneous solution chemistry (high reactivity, lack of diffusion

phenomena and ease of analysis) with those of solid phase methods (use of excess

reagents and easy isolation and purification of products).16

52

 

Non—cross-linked chloromethylate polystyrene (NCPS) polymer is a soluble
polymer used in the porphyrin synthesis.l7 This copolymer is readily prepared, and the
functional group content easily controlled and quantified via NMR by varying ratios of
starting monomers. NCPS polymer has remarkable solubility properties that are
amenable to organic chemistry. It is soluble in THF, CHzClz, CHCl3, and ethyl acetate
even at low temperatures, and is insoluble in water and alcohols. Consequently, after the
homogeneous reaction of supported intermediates, the polymer and its uniquely bound
product can be easily separated from excess reactants and byproducts by treating with
cold methanol, in which the soluble polymer precipitates as a solid polymer, and the
soluble reagents and byproducts are filtered away. Because the polymer is soluble in
CHC13, NMR analysis of all intermediates may be accomplished in a non-destructive
manner without the need for any specialized NMR techniques.18 All the reaction with the
soluble polymer can be monitored with NMR, and the porphyrin peaks are observable in
NMR spectrum.

As illustrated in Scheme 2-7, the non-cross-linked chloromethylate polystyrene
(NCPS, 9) was prepared as reported by Janda and coworkers by the copolymerization of
styrene with 3 mol% of 4-(chloromethyl) styrene in the presence of 2, 2'-
azobis(isobutyronitrile) (AIBN) in benzene at 70 °C for 40 h.17 After the reaction was
over, most of the solvent was removed under reduced pressure, and the residue was added
to chilled methanol (-30 °C). The soluble polymer precipitated in the methanol as a white
powder, which could be separated from the remaining starting material by suction

filtration.

53

 

AIBN (0. 06 mol%)
6+: Benzene,
reflux, 40 h,

25%

 

Scheme 2-7. Copolymerization to synthesize NCPS polymer.

Having the correct particle size for the precipitated polymer is important for
successful and efficient suction filtration. If the polymer size is too small, the polymer is
likely to go through the filter paper, which can clog the funnel and make the filtration
take several hours to finish. The size of the polymer powder depends on both the amount
of solvent it is dissolved in before it is precipitated with cold methanol (the larger amount
of solvent used, the finer the powder is), and more importantly, on the average molecular
weight of the polymer itself. The molecular weight of the soluble polymer is controlled
by the amount of radical initiator (AIBN) used in the copolymerization reaction. The
smaller amount of initiator, the higher average molecular weight, but at the same time,
the yield of the copolymer decreases. When 0.5 mol% of AIBN was used, the MW avg of
the copolymer isolated was 15,000, and the yield was 50%. The filtration of this polymer

was tedious. By decreasing the initiator to 0.06 mol%, the MWavg of the polymer

increased to 60,000 and a PDI of 2.14 was measured by GPC.19 The yield of

polymerization was 25%.

54

Table 2-1. Synthesis of soluble polymer supported aldehyde l0

OCHO O
m n H020 5 OJKO
o o (YET

 

 

 

9 CI 10
Base Solvent Temp (°C) Period Yield (%)
(day)
NaH DMF 90 1 ---
KH DMF 90 1 80
NaHMDS THF 66 2 ---
i-PrgNEt EtOAc 77 5 25
E13N EtOAc 77 2 60
E13N EtOAc 90 (sealed tube) 3 >95

 

A variety of conditions had been probed to attach 4-carboxybenzaldehyde 5 to the
soluble polymer 9, and the results are outlined in Table 2-1. The best result was obtained
by using triethylamine as the base and heating the reaction mixture in ethyl acetate to 90
°C in a sealed tube. The loading of the aryl aldehyde in 10 was calculated from NMR
integrations of the latter moiety vs. the polymer backbone signals. The conversion was
higher than 95%.

The synthesis of mono-substituted porphyrin on soluble polymer 10 is shown in
Scheme 2-8. Boron trifluoride diethyl etherate (BF3-Et20) was used as the acidic catalyst
in the porphyrin synthesis on soluble polymer, since this condition was much milder and
the yield was higher. The results were quite satisfactory; thus the synthesis was not

attempted with propionic acid condition.

55

O 11aR=Ph

GAO 1. pyrrole / RCHO 11b F1 -.- p-OMePh

Ofij BF3' EtZO, RT, 1 h 110 n = 2-Naphthyl

CHO -
2. CHZClz, p-chloranil

RT, 2 h

 

-----------------

O R E tetraphenyl E 1. wash with MeOH
+ i - - : +-
5 SSHWSMW 2. NaOH,THF

' ................. ' reflux, 12 h

 

 

13a R = Ph 30% overal yield
13b R = p-OMePh 16% overal yield
130 Ft = 2-Naphthyl 15% overal yield

 

[O = soluble polymer]

 

 

Scheme 2-8. Synthesis of mono-substituted porphyrin on soluble polymer.
Condensation of polymer supported aldehyde 10 and pyrrole with excess of either
benzaldehyde 11a, p-methoxybenzaldehyde 11b, or 2-naphthylaldehyde 11c with
BF3-Et20 in CH2C12 (similar to reported procedures for normal porphyrin synthesis in
solution6) and subsequent oxidation with p-chloranil led to the formation of the
corresponding soluble polymer bound porphyrins 12a-c. The product in each case was
isolated and by basic hydrolysis of the ester functionality (2 N NaOH) to yield the
porphyrin acids 13a-c in yields comparable to solution phase synthesis. However, in
comparison to non-polymer supported solution phase synthesis, the purity of the crude

product is substantially improved, and each product is easily purified by a simple silica

56

chromatography. The yield of each porphyrin was calculated based on the amount of
bound aldehyde 10 used in the reaction. The mono-substituted porphyrin acids were
purified by silica column chromatography and eluted with 3% methanol in CHzClz. Non-
aromatic aldehydes such as i-butyl aldehyde, propanal, and acetaldehyde did not yield
any product, presumably due to the polymerization of these reagents under the reaction
conditions.

In conclusion, synthesis of porphyrins on polymeric backbone can be used as a
strategy to selectively synthesize mono-substituted porphyrins without the troublesome

contamination from multiple-substituted products.

57

 

Experimental materials and general procedures:

Anhydrous CH2C12 was dried over C3112 and distilled. Unless otherwise noted,
materials were obtained from commercial suppliers and were used without further
purification. All reactions were performed in dried glassware under nitrogen. Column
chromatography was performed using SiliCycle silica gel (230-400 mesh). 1H-NMR
spectra were obtained on Varian Inova 300 MHz instrument and are reported in parts per
million (ppm) relative to the solvent resonances (8), with coupling constants (J) in Hertz

(Hz).

CH0 conditions

——.——————.

CHO
+

6,7 .

002Me

 

5-(4-Methylcarboxy-phenyl)-10,15,20-triphenylporphyrin 2:

Method 1. To a mixture of 4-carboxymethylbenzaldehyde (0.83 g, 5 mmol),
benzaldehyde (1.62 g, 15 mmol, and Zn(OAc)2 (1.09 g, 5 mmol) in propionic acid (100
mL) was added pyrrole (1.32 g, 20 mmol) when the temperature reached 120 °C. The
solution was refluxed for 4 h with vigorous stirring. The solvent was removed under
reduced pressure, and the dark solid residue was dissolved in CH2C12, and excess amount
of DDQ (0.42 g) was added to the solution. The mixture was refluxed for another 2 h.
The mixture was then cooled to room temperature, and the solution was washed with 6 N

HCl (3 x 40 mL), neutralized with saturated NaHCO3 aqueous solution (3 x 40 mL),

followed by brine (1 x 50 mL), and dried with anhydrous N32804. The solvent was

58

removed under reduced pressure. Product was purified by multiple column
chromatographies (CHzClz) to yield pure 2 (0.26 g, 7%) as deep purple crystals.

Method 2. To a mixture of 4-carboxymethylbenzaldehyde (0.83 g, 5 mrnol) and
benzaldehyde (1.62 g, 15 mmol) in propionic acid (100 mL) was added pyrrole (1.32 g,
20 mmol) when the temperature reached 120 °C. The solution was refluxed for 4 h with
vigorous stirring. The solvent was removed under reduced pressure, and the dark solid
residue was dissolved in CHzClz, and excess amount of DDQ (0.42 g) was added to the
solution. The mixture was refluxed for another 2 h. The solution was then cooled to
room temperature, and the solvent was removed under reduced pressure. Purification by
multiple column chromatographies (CH2C13) yielded pure 2 (0.24 g, 7%) as deep purple
crystals.

Method 3. To a mixture of 4-carboxymethylbenzaldehyde (0.83 g, 5 mmol) and
benzaldehyde (1.62 g, 15 mmol) in freshly distilled CHzClz (100 mL) was added pyrrole
(1.32 g, 20 mmol). Nitrogen gas was purged into the solution with vigorous stirring, and
after 15 min catalytic amount of BF3-E120 (0.1 mL) was added. After 10 min, the
solution turned from colorless to amber. The reaction flask was shielded from light by
aluminum foil, and the reaction was stirred at room temperature for another 2 h. Excess
amount of p-chloranil (3 g) was added to the solution, and it was stirred for an additional
2 h. The solvent was then removed under reduced pressure, and the residue was purified
as described above to yield pure 2 (0.81 g, 25%). 1H NMR (CDCl;, 300 MHz) 5 8.76-
8.85 (m, 8H), 8.42 (d, J=8.3 Hz, 2H), 8.29 (d, J=8.3 Hz, 2H), 8.19-8.21 (m, 6H), 7.71-

7.79 (m, 9H), 4.09 (s, 3H), -2.81 (s, 2H).

59

 

NaOH (2 N) /
THF, reflux,
overnight, 98%

 

 

5-(4-Carboxyphenyl)-10,15,20-triphenylporphyrin 3:

To the solution of 2 (75 mg, 0.1 mmol) in THF (5 mL) was added 2 N NaOH
aqueous solution (1.5 mL), and the solution was refluxed overnight. After the reaction
was cooled down to room temperature, the pH of the reaction solution was adjusted to 4
with 6 N HCl. THF was removed under reduced pressure, and the aqueous layer was
extracted with MeOI-I/CHzClz (1:100, 3 X 20 mL) until there was no purple color left in
the aqueous layer. The organic layers were combined and washed with DI water (2 x 20
mL). The organic layer was then dried with anhydrous NaZSOa, and removed under
reduced pressure to afford the acid 3 in 98% yield. 1H NMR (CDC13, 300 MHz) 8 8.76-
8.85 (m, 8H), 8.42 (d, J=8.3 Hz, 2H), 8.29 (d, J=8.3 Hz, 2H), 8.19-8.21 (m, 6H), 7.71-

7.79 (m, 9H), -2.81 (s, 2H).

60

 

 

EDC, DMAP
CH2012, 48%

   

Porphyrin tweezer 4:

To a solution of acid 3 (90 mg, 0.14 mmol) and 1,5-pentanediol (5 mg, 0.05
mmol) in anhydrous CH2C12 (4 mL) was added EDC (27 mg, 0.14 mmol) and DMAP (18
mg, 0.14 mmol). The reaction mixture was stirred at room temperature overnight, and it
was applied directly to silica gel column and purified (CH2C12) to afford the title
compound 4 (34 mg, 48%). 1H NMR (CDCI3, 300 MHz) 8 8.94—9.04 (m, 16H), 8.48 (d,
J=8.0 Hz, 4H), 8.17-8.23 (m, 16H), 7.46-7.54 (m, 18H), 4.28 (t, J=6. 6 Hz, 4H), 1.65 (m,

4H), 1.48 (m, 2H), -2.81 (s, 2H).

 

Zinc porphyrin tweezer 1:
To a solution of porphyrin tweezer free base 4 (10 mg) in CHzClz (3 mL) was

added Zn(OAc)2 (30 mg, 0.16 mmol). The reaction mixture was stirred at room

61

temperature overnight. The solution was applied directly to silica gel column (CH2C12) to
afford the pure zinc porphyrin tweezer 1 (96 mg, 95%) as a deep purple powder. 1H
NMR (CDCl3, 300 MHz) 8 8.94-9.04 (m, 16H), 8.48 (d, J=8.0 Hz, 4H), 8.17-8.23 (m,

16H), 7.46-7.54 (m, 18H), 4.28 (t, J=6. 6 Hz, 4H), 1.65 (m, 4H), 1.48 (m, 2H).

OCH0 0
Cl H020 MO
‘ 6 EtaN, EtOAc ' 7 CHO

reflux, 2 h

 

Insoluble polymer anchored substituted aldehyde 7:

2% cross-linked divinylbenzene/polystyrene beads 6 (3 g), containing 1
mmole/eq. of benzyl chloride, 4-carboxybenzaldehyde (0.9 g, 6 mmol) and Et3N (1 mL)
were added to EtOAc (20 mL), and the reaction mixture was refluxed for 2 h. The
reaction was then cooled down, and the polymer was washed with ethyl acetate, benzene,
water and acetone, and then dried under vacuum. Since the polymer does not dissolve in

any solvent, the conversion could not be monitored with NMR.

CHO
O 1' (D U (excess)

OAKCL propionic acid
CHO reflux, 4 h t

2. CH2Cl2, DDQ,
reflux, 2 h .

 

 

Insoluble polymer supported porphyrin 8:

Polymer anchored aldehyde 7 (0.5 g, 0.29 mmol), benzaldehyde (190 mg, 1.8
mmol) and pyrrole (155 mg, 2.3 mmol) were added into propionic acid (20 mL), and the
mixture was refluxed for 4 h. The solvent was removed under reduced pressure, and the

dark solid residue was dissolved in CH2C12, and excess amount of DDQ (0.42 g) was

62

added to the solution. The mixture was refluxed for another 2 h. The mixture was then
cooled to room temperature, and the polymer supported porphyrin was separated from
non-substituted porphyrin (tetraphenylporphyrin) by Soxhlet extraction with CHzClz as

solvent.

NaOH, THF,
reflux, 12 h

   

Hydrolysis of insoluble polymer supported porphyrin 8 to acid 3:

To the solution of polymer supported porphyrin 8 (0.5 g) in THF (5 mL) was
added 2 N NaOH (1.5 mL), and the solution was refluxed overnight. After cooling the
solution to room temperature, the polymer was filtered, and washed with THF until all
the purple color was removed. The pH of the filtrate was adjusted to 4 with 6 N HCl.
The THF was removed under reduced pressure, and the aqueous layer was extracted with
MeOH/CH2C12 (1:100, 3 x 20 mL) until there was no purple color left in the aqueous
layer. The organic layers were combined and washed with DI water (2 x 20 mL). The
organic layer was then dried with anhydrous NaZSOa, and removed under reduced

pressure to afford the acid 3. The yield for the two steps, from 7 to 3, was 6%.

63

 

 

 

/ / AIBN (0.06 mol%)
(5 + Benzene,
reflux, 40 h,
Cl

25%

 

Synthesis of soluble polymer NCPS 9:

Styrene with 3 mol% of 4-(chloromethyl)styrene were dissolved in benzene. 2,
2'-azobis(isobutyronitrile) (AIBN) was added into the reaction mixture, and it was stirred
at 70 °C for 40 h. The solvent was removed under reduced pressure to yield an oil like
residue, and it was then added to cold methanol (-30 °C) dropwise with vigorous stirring
to remove the unreacted styrene and 4-(chloromethyl)styrene. The oil like residue
became white powder as soon as cold methanol was added. The precipitate was filtered,
and redissolved with benzene and re-precipitated in cold methanol again. This cycle was
repeated three times until no styrene and 4-(chloromethyl)styrene could be observed by

NMR.

CHO
o o
m n H020 5 OJKO
e o 0U

Cl Sealed tube, 90 °C
9 3 days, 95%

 

10

Soluble polymer supported benzaldehyde aldehyde 10

Soluble polymer NCPS 9 (2 g), 4-carboxybenzaldehyde (0.9 g, 6 mmol) and Et3N
(1 mL) were added to EtOAc (20 mL). The reaction mixture was refluxed for 3 days.
The reaction was then cooled down, and the solvent was removed by reducing pressure.
The oil like residue became white powder as soon as it was added to cold methanol. The

precipitate was filtered, and redissolved with acetyl acetate to precipitated in cold

 

methanol again. This cycle was repeated until all the 4-carboxybenzaldehyde was

 

 

 

removed.
0
O 1. pyrrole / RCHO
BF3‘E120, RT, 1 1'!
CH0 >
10 2. CHZClz, p-chloranil
RT, 2 h
R
NaOH, THF
reflux, 12 h
HOZC O Q n
/
R 133-6

The general procedure for syntheses of mono-substituted porphyrins l3a-13c on

soluble polymer:
To a mixture of the soluble polymer supported benzaldehyde aldehyde 10 (l g)

and aldehyde (15 mmol) in freshly distilled CHzClz (30 mL) was added pyrrole (1.32 g,
20 mmol). Nitrogen gas was purged into the solution with vigorous stirring, and after 15
min, catalytic amount of BF3-Et20 (0.1 mL) was added. After 10 min, the solution turned
from colorless to amber. The reaction flask was shielded from light by aluminum foil,
and the reaction was stirred at room temperature for another 1 h. Excess amount of p-
Chloranil (3 g) was added to the solution, and it was stirred for additional 2 h. Then the

reaction mixture was condensed and the residue was precipitated in cold methanol to

remove the tetraphenylporphyrin-

65

 

To the solution of polymer supported porphyrin 12a-c (0.6 g) in THF (10 mL)
was added 2 N NaOH aqueous solution (5 mL), and the solution was refluxed overnight.
After cooled down to room temperature, the pH of the mixture was adjusted to 4 with 6 N
HCl solution. The THF was removed by reduced pressure, and the aqueous layer was
extracted with MeOH/CHzClz (1:100, 3 X 20 mL) until there was no purple color left in
the aqueous layer. The organic layer was combined and condensed by reducing pressure.
The residue was dissolved in CHzClz (5 mL), and precipitated in cold methanol/THF
(10:1). This cycle was repeated three times, and all the organic solvents were combined.
The solvents were removed under reduced pressure, and the residue was purified by

column chromatography (3% MeOH in CHzClz) to afford the pure porphyrin acids 13a-c.

66

Reference:

1.

2.

10.

ll.

12.

13.

14.

15.

l6.

l7.

Ogoshi, H.; Mizutani, T. Acc. Chem. Res. 1998, 31, 81.

Berova, N.; Nakanishi, K.; Woody, R. W. Circular Dichroism: Principles and
Applications; Wiley-VCH: New York, 2000.

Matile, S.; Berova, N.; Nakanishi, K.; Fleischhuer, J.; Woody, R. W. J. Am.
Chem. Soc. 1996, 118, 5198.

Huang, X. F.; Rickman, B. H.; Borhan, B.; Berova, N.; Nakanishi, K. J. Am.
Chem. Soc. 1998, 120, 6185.

Littler, B. J .; Miller, M. A.; Hung, C. -H.; Lindsey, J. S. J. Org. Chem. 1999, 64,
1391.

Lindsey, J. S.; Schreiman, I. C.; Hsu, H. C.; Kearney, P. C.; Marguerettaz, A. M.
J. Org. Chem. 1987, 52, 827.

Smith, K. M. Porphyrins and Metalloporphyrins; Elsevier Scientific Pub. Co.:
New York, 1975.

Lavallee, D. K. The Chemistry and Biochemistry of N-Substituted Porphyrins;
VCH Publishers: New York, 1987.

Kadish, K. M.; Smith, K. M.; Guilard, R. Synthesis and Organic Chemistry;
Academic Press: San Diego, CA, 2000.

Dolphin, D. The Porphyrins: Structure and Synthesis; Academic Press: New
York, 1979.

Lindsey, J. S., Synthesis of meso-Substituted Porphyrins. In The Porphyrin
Handbook; Kadish, K. M., Smith, K. M., Guilard, R., Eds; Academic Press: San
Diego, CA, 2000.

Leznoff, C. C.; Svirskaya, P. I. Angew. Chem. Int. Ed. Engl. 1978, 17, 947.
Lorsbach, B. A.; Kurth, M. J. Chem. Rev. 1999, 99, 1549.

Merrifield, R. B. J. Am. Chem. Soc. 1963, 85, 2149.

Gravert, D. J .; Janda, K. D. Chem. Rev. 1997, 97, 489.

Wentworth, P.; Janda, K. D. Chem. Commun. 1999, 1917.

Chen, S. Q.; Janda, K. D. J. Am. Chem. Soc. 1997, 119, 11355.

67

18. Fitch, W. L.; Detre, 0.; Holmes, C. P.; Keifer, P. A. J. Org. Chem. 1994, 59,
7955.

19. Yin, M.; Baker, G. L. Macromol. 1999, 32, 7711.

68

Chapter 3

Absolute stereochemical determination of a-chiral carboxylic acids

Developing methodologies that can be used to determine the absolute

stereochemistry of chiral compounds is the goal of our research. Our approach entails

designing and synthesizing chromophoric receptors, which can bind to the chiral

compounds and adopt a helicity through an induced Chirality. The host-guest interactions

between the chiral compounds and the receptors can be observed by CD. In particular,

the Exciton Coupling Circular Dichroic (ECCD) method1 will be used to assign the

stereochemistry of the asymmetric center non-empirically.

According to the number of sites of attachment the chiral molecules possess, the

chiral compounds can be divided into two categories:

Chiral compounds with two sites of attachment: Porphyrin tweezer has been
designed to serve as the receptor, which can interact with the binding molecule
and adopt a chiral conformation as a result of induced Chirality.2

Chiral compounds with only one site of attachment: The stereochemistry of these
compounds is more difficult to determine as compared to the ones with two sites
of attachment. Porphyrin tweezer method cannot be applied directly to these
compounds since the two porphyrin groups cannot be oriented relative to each
other when there is only one attachment instead of two. To overcome this
problem, “carrier” molecules have been designed, which upon derivatization with
chiral molecules, provide two sites of attachment.3

As has been discussed in detail (see Section 1—3.2), the carrier method has been

successfully applied to chiral monoamines to determine their absolute stereochemistry,

69

by using carrier molecule 1 and zinc porphyrin tweezer 2 (Figure 3-1).3 Now, we are
interested in exploring the possibility of extending the same methodology to chiral
compounds with other functional groups, especially a—chiral carboxylic acids.
H O H

N\ CO2H ’kmi N\ ’kni

l H2N OH I N

/ 3 / H CH3

NHBoc — NH2

1 M
O, H F

L
I O /
“5w?" _

Zn Zn N
H
n‘—-

CLOWO O

o o’\/2\/\o 0
Figure 3—1. Stereochemical determination of chiral monoamine with carrier 1.

 

6

L

 

 

 

The most widely employed approach for absolute stereochemical assignment of
a—chiral carboxylic acids is NMR, by using Mosher’s method and its modified versions.4'
7 According to this protocol the chiral carboxylic acids are derivatized to either amides
or esters by using both the (R)- and (S)—enantiomers of a chiral auxiliary reagent. This
auxiliary, such as 9-AHA (9-anthrylhydroxyacetic acid ethyl ester), usually contains an
aryl ring that directs its anisotropic cone selectively towards one of the substituents of the
asymmetric center of the chiral acid. The chemical shifts observed for each of the groups
of the chiral acid (L1, L; in Figure 3-1) reflect their spatial relationship with respect to the
aryl ring. Therefore, the absolute configuration of the chiral center can be obtained from
the difference in the chemical shifts (A8“) measured (Figure 3-2).8 Some of the

commonly used chiral auxiliary reagents are shown in Figure 38.4"7

70

S R

H 0 L1 H H 2
AWJ‘OH + H _|'-|1_2 :29 Akioo :9 ”$9 Chg
R H H H

shieldedC groups
Figure 3-2. The absolute stereochemical assignment of a—chiral carboxylic acids by

NMR.

00C (3 002M.

Ho... ? 'H
H OEt H2NWOM8 HO Ph
0
(R)-9-AHA (R)-PGME (m-methyl mandelate

Figure 3-3. Chiral auxiliary reagents.

When using auxiliary reagents for the stereochemical determination of OL-chiral
carboxylic acids, the conformation of the chiral conjugate has to be well understood.
Figure 3-4 is an example of using PGME (phenylglycine methyl ester) as the auxiliary
reagent.5’ 6 In order for the phenyl group to exert its diamagnetic field effect upon L1 and
L2, coplanarity of C1 to C5 is necessary. Coplanarity of the atoms from position 1 to 4 is
guaranteed due to the s-trans amide linkage, which is well established in peptide
chemistry.9 The planarity can be extended to the methoxycarbonyl group at 5-position,
because of the electrostatic repulsion between carbonyl groups at 2- and S-position,
which has been verified by X-ray crystallography and NOE studies.5

H ' L1 10 3 H5
H H O
H—_J

L m- + E
11.2 COOH HzN/EOMe
coplanar

(FD-PGME

 

Figure 3-4. The derivatization of PGME with chiral acids.

71

However, very few of the chiral auxiliary reagents have been subject to detailed
theoretical or experimental studies.lo From a practical point of view, selecting the most
suitable reagent for a certain substrate is still problematic, and sometimes, the signals in
the NMR spectra of the diastereoisomeric derivatives can be too close for drawing any
safe conclusion.

Another approach for the absolute stereochemical determination of a-chiral
carboxylic acids is by using circular dichroism (CD), in particular, the exciton coupled
circular dichroism (ECCD).ll Although the ECCD method has been applied to several
classes of molecules to determine their absolute stereochemistry, in the case of chiral
carboxylic acids, it can only be applied to a-hydroxy carboxylic acids with free hydroxyl
group”’ '3 or chiral acids carrying an additional chromophore in the molecule. Figure 3-5

is an example of applying the ECCD method for the absolute stereochemical

determination of a—hydroxy carboxylic acids.

0 O
RBJSOH 1 .2equiv. = ROH01 ZequiV-NNRSKO;

(H)H or (S) (F?) or (S)

o
O: 00 0%

 

 

 

Figure 3-5. Two-step derivatization of a-hydroxy carboxylic acids.
After a two-step derivatization with 9-anthryldiazomethane and 2-
naphthoyltriazole, the a-hydroxy carboxylic acid can be converted to a bichromophoric

system.'4 The absolute stereochemistry of the chiral center determines the orientation of

72

the two chromophores in space, which in turn dictates the observed ECCD spectrum. All
the (R) acids exhibit positive ECCD, while (S) acids exhibit negative ECCD.l3

However, in cases where the substrate has no site other than the carboxyl group
for derivatization, application of the ECCD protocol is not straightforward. The 210 nm
n-1t* transition of the carboxyl chromophore has been used in the past for configurational
assignment of the carboxylates.15 As structures shown in Figure 3—6, the CD of a series
of l-carboxyethyl substituted monosaccharides have been measured, and the results
indicated that the absolute stereochemistry of the l-carboxyethyl substituents can be
correlated to the sign of the CD for the carboxyl chromophore with absorption at 210 nm,
with (R) carboxylates having negative CD, while (S) carboxylates exhibit positive CD.15 ‘
16 The absorption bands are weak, and this empirical method can only be applied to l-

carboxylethyl substituted hexoses.

OH
HOOC H00 0
"’1 HO

H3311 OCH3

Figure 3-6. The structure of l-carboxyethyl substituted monosaccharides.

We are interested in developing a general protocol for the non-empirical
determination of the absolute stereochemistry of a—chiral carboxylic acids. In order to
achieve this goal, we are relying heavily on the Exciton Coupling Circular Dichroic
(ECCD)l method by using different carriers in conjunction with porphyrin tweezer.
Nearly at the same time, Nakanishi’s lab has also developed with a very similar
methodology in determining the absolute stereochemistry of oc—chiral carboxylic acids.”

'8 Their system will be discussed later in this chapter.

73

3-1. Stereochemical determination of a-chiral carboxylic acids with carrier A
Carrier 1 has been used by Nakanishi and his coworkers to determine the absolute

stereochemistry of chiral monoamines.3

It contains a carboxylic acid functionality to
derivatize with chiral amines, and the resulting chiral conjugate provides two nitrogen
binding sites to coordinate with zinc porphyrin tweezer 2. We are interested in
stereochemical determination of Ot-chiral carboxylic acids, thus it is reasonable to design
a carrier A with similar structure to that of carrier 1, but having changed the carboxylic

acid functionality in carrier 1 to an amino group, so it can form derivatives with

carboxylic acids (Figure 3-7).

N\ COZH N\ NH2
[g g
NH2 NH2
carrier 1 for chiral carrier A for oz-chiral
monoamines carboxylic acids

Figure 3-7. The structures of carrier 1 and carrier A.

The synthesis of Boc-protected carrier Boc-A is shown in Scheme 3-1. The
synthesis is straightforward, starting with the commercially available 4-
aminomethylpyridine 3. The free amino group in 3 was protected with the Boc
protecting group, and the pyridine nitrogen of compound 4 was oxidized by mCPBA to
form pyridine oxide 5. The cyano functionality was introduced at the ortho position by
treating the pyridine oxide 5 with TMSCN and dimethylcarbamyl chloride in CH2C12 at
room temperature for two days. The cyano group was then hydrolyzed to amide 6 by
refluxing in KOH/t-BuOH mixture for half an hour.19 Finally, the amide functionality in
compound 7 was converted to the amino group via Hofmann rearrangement to form the

desired Boc-protected carrier Boc-A. Two reaction conditions were tested for the

74

Hofmann rearrangement. First was heating the amide 7 with bromine in a solution of
sodium methoxide and methanol, however, no conversion was observed. Second method
entailed heating the amide 7 with bromine and aqueous sodium hydroxide solution. The

second reaction condition worked smoothly and provided the desired amine Boc-A in

 

 

 

 

75% yield.
(BOC)20, TEA NHB NHBoc
””2 THF, rt, OCmCPBA, CH2012 \
I \ overnight, 95% I \ n, 12 h, 80% I ,
N’ i N’ ; g
3 4 5
TMSCN, CHZCIZ.
rt, 48h, 53%
it
‘1}: Cl
NHBOC NaOH’ 3'2 NHBOC KOH/t-BuOH NHBOC
\ 80 °C. 211. 75% | \ reflux, 30 min \
l / i N’ NH2 4% l I
N NH2 0 96% N CN
Boc-A 7 6

Scheme 3-1. The synthesis of Boc-protected carrier A.

The protected carrier Boc-A was then coupled with chiral carboxylic acids by
BOP coupling to generate the chiral conjugates.20 This amide formation reaction was
also examined by using EDC as the coupling reagent; however, EDC did not lead to high
yield of the desired amide. After the formation of the chiral conjugate, the Boc group
was removed by stirring in a mixture of trifluoroacetic acid and CH2C12 to provide the
final chiral complex, which was ready for CD analysis.

NHBoc 1- RchiraICOZH NH2
[a BOP, Et3N, CHZCI: {a TL

N NH2 2.TFA,CHZCIZ N N Cchiral
52-78% H
Boc-A

 

Scheme 3-2. The synthesis of chiral complexes from Boc-A.

75

Compared to carrier 1 derivatized chiral monoamines, the carrier A derivatized 0t-
chiral carboxylic acid complexes are not as rigid in structure. As shown in Figure 3-8,
when carrier 1 is derivatized with a chiral monoamine, the resulting chiral complex is

very rigid. There is no free rotation around the CN-Cczo, Cc=o-NH, and NH—Cchira.

bonds, since all the free rotations around the single bonds are restricted: the anti pyridine
nitrogen/carbonyl oxygen disposition is preferred by 30 lemol, due to electrostatic
repulsion, and the small group (hydrogen) at the chiral center is syn to the carbonyl

oxygen because of pseudo A13 steric interaction.

complexes of carrier 1 complexes of carrier A
with monoamine with monoacid
coordination 1:152 coordination

5'19 \ HN H s'te \ ll R152
I N\ \ O /Syn I N\ %H

/ rotation energy / 0

30 kJ/mol
NH2 NH2
coordination / coordination /
sue sue

Figure 3-8. Comparison of chiral complexes of carrier 1 and carrier A: carrier 1
complexes are locked in a conformation, while carrier A complexes are not so rigid.
However, the complexes formed by carrier A with mono-carboxylic acid are not
as rigid. The orientation of the pyridine nitrogen with respect to the carbonyl oxygen is
not as certain, since the carbonyl is not directly attached to the pyridine ring and the
electrostatic repulsion is not as strong. More importantly, the free rotation around the
Cchiral-Cczo bond is hard to predict. The pseudo A1,; interaction present in carrier A
complexes is not as determinant as the pseudo A13 steric interaction in the carrier 1

complex to freeze the rotation around the Cchiral'CC=O bond. The flexibility of this

76

conjugate may cause problems in correlating the absolute stereochemistry of the chiral
complexes to the observed ECCD signs.

As shown in Scheme 3-2, the carrier A derivatized a-chiral carboxylic acids were
synthesized by coupling Boc-A with different or-chiral carboxylic acids by standard BOP
coupling, followed by Boc deprotection. A pinch of Na2C03 was added into the solution
of the carrier A derivatives to keep the free amino group of the chiral complex from being
protonated. The solution was then added to the zinc porphyrin tweezer in
methylcyclohexane for spectroscopic evaluation. As an example, the ECCD spectra of a

pair of carrier A derivatized enantiomers are shown in Figure 3-9.

-——S-phenoxy propionic
acid

‘ —— R-phenoxy pr0pionic

acid

 

375 425 475

wavelength (nm)
Figure 3-9. The ECCD spectra of carrier A derivatized (R)— and (S)—phenoxypropionic
acid with zinc porphyrin tweezer 2.
The observed ECCD signs and amplitudes of the carrier A derivatized chiral
carboxylic acids upon addition to zinc porphyrin tweezer 2 in methylcyclohexane are

recorded in Table 3-1. The signs of the predicted CD spectrum based on the

77

computational modeling studies are also compared to the observed CD signs. All the CD
measurements listed in Table 3-1 were carried out at room temperature, except for the
conjugate 14A, which was measured at O 0C. This conjugate exhibited no ECCD peak at
room temperature, and the ECCD peak could only be observed at 0 °C.

Table 3-1. CD predictions and amplitudes of carrier A derivatized chiral carboxylic

acids bound to porphyrin tweezer 2.

NHBoc 1- RchiralCOZH NH2
\ BOP, EtaN,CHZC|2A Bi 0
| , ’ JL

 

 

 

N NH2 2. TFA, CHZCIZ N’ ll Cchiral
A 8A-15A
Chiral acids Predicted ECCD As A

5768:): 8 (8) 8A "99 obsllcii/ed

Seizure) 9..., 2336.235.)
“3&3? 1o (8) 1°“ ”99 23:; ((3289)) '57
H3332?” 11 (3) "A neg 1291 (5972)) +124
$5022” 12 (H) ‘2‘ ”S 2223 (:56?) '1 ‘8
£55.02” 13 (H) 13" P°S 21277 ((42113)) '29
Cal'gEZHM (S) ”A ”99 2233 ((4133)) '72

”YCOZH 15A neg 430 (6) _21

H30 ’CHzPh 15 (R) 422 (+15)

 

Initially, a simplistic model based on molecular modeling conformational search
(Monte Carlo) of the (Jr-chiral carboxylic acid conjugated carrier A was used to predict

the anticipated sign of the ECCD spectra. Modeling of the carrier A derivatized

78

complexes was initially performed without bound porphyrin. As shown in Figure 3-10,
the lowest energy conformations of carrier A conjugates predicted a syn arrangement of
the chiral center’s medium group with respect to the amide hydrogen for most of the
derivatives (sizes are based on A values).21 We believe this was predicted in order to
facilitate hydrogen bonding between the heteroatom-containing substituent, which is
usually the medium group, and the amide proton. Also, this arrangement would lead to

staggering of the other two groups at the chiral center (large and small) with respect to

\ 0
IN, ”J11?-

Figure 3-10. Lowest energy conformations of carrier A derivatized complexes based on

the carbonyl.

modeling without binding to zinc porphyrin.

Therefore, our expected ECCD signs were dictated by placing the medium group
syn to the amide hydrogen, which in turn would give rise to a preferred helicity of the
bound porphyrin tweezer as a result of avoiding steric interactions with the large group.
However, this model did not allow for consistent prediction of ECCD signs. We
abandoned this method of analysis since the predictions did not match the observed
ECCD spectra, (see Table 3—1, conjugates 8A, 11A, 12A, and 13A), most probably
because we had not incorporated the binding of the zinc porphyrin to the pyridine
nitrogen atom into our modeling.

Incorporation of the porphyrin into the modeling of carrier A derivatives
produced conformers with inconsistent rotation around the Caryl'Namide bond, as shown in

Figure 3—11. All of the derivatized carriers produced multiple conformations within 1

79

kcal/mol that would lead to opposing ECCD spectra. In many cases, the number of
conformers expected to show positive ECCD was nearly equal to the number of

conformers expected to show negative ECCD (within the lowest 1 kcal/mol).

 

 

A \
V.; .
"N‘ 0 ZnTPP ,
-3.6 I;
W.;.
~ ~r
14A l4A+ZnTPP

Figure 3-11. Based on modeling, binding of the Zn porphyrin to carrier A conjugates
forces the rotation of the Caryl-Namide bond.

We believe the complication is because upon porphyrin binding with the pyridine
nitrogen, unfavorable steric interactions with the amide hydrogen forces the Caryl'Namide
bond rotation, as shown in Figure 3-1 1. The relatively large distance between the site of
porphyrin binding and the chiral center might also lead to unpredictable assignments
since the rotation of multiple single bonds must be assumed. Therefore, it became
apparent that a more rigid carrier should be designed for the stereochemical
determination of OL-chiral carboxylic acids, which could limit the number of possible

conformations upon binding to the zinc porphyrin.

8O

3-2. Stereochemical determination of a—chiral carboxylic acids with carrier B

Carrier B, 1,4-diaminobenzene, was proposed as an alternate carrier in order to
alleviate the problems associated with carrier A. Mono-amidation of 1,4—diaminobenzene
(carrier B) with (it-chiral carboxylic acids provided chiral complexes, which could
coordinate with the zinc porphyrin tweezer via the free amino nitrogen and the carbonyl
oxygen of the resulting amide group.

It was anticipated that the carrier B conjugates would not bind to zinc porphyrin
tweezer as strong as carrier A derivatives, since the amide carbonyl oxygen in carrier B
derivatives was not such a good coordination handle to bind to zinc porphyrins as
pyridine nitrogen in carrier A complexes. However, the disadvantage in binding could be
compensated by other factors. Not only binding through the carbonyl oxygen would
bring the porphyrin moiety closer to the chiral center, but also the Caryl‘Namjde bond
rotation would be inconsequential in this system. The induced helicity within the

porphyrin tweezer would thus be governed only by the rotation of the Ccarbonyl-Cchiral

bond.

fi—r

NH2 RchiraICOZH n Rchiral
O EDC, DMAP,CHZCl2 0’ 76
H2N >< * HzN

 

 

3
NH RchiraICOZH Kl n .
O“ 2 BOP. EtaN.CH20I2 O 76 Chm“
H2N A' 1'1le
3 713-9270
NH2 RchiraICOZH H R .
O PPh3, Et3N,CBr4 O 2; chiral
H2N A’ H2N
B

Scheme 3-3. The synthesis of carrier B derivatized chiral complexes.

81

The different methods that were tried for forming the carrier B derivatized chiral
complexes are summarized in Scheme 3-3. The amidation between the carrier B and
chiral carboxylic acids was first tried with a standard EDC coupling reaction, however,
after the reaction mixture was stirred in the room temperature for 2 days, most of the
starting material remains unreacted. Therefore, EDC was not active enough to form
amide bond between the carrier B and chiral acids. When a more reactive coupling
reagent BOP was used instead, this reaction underwent smoothly, and the starting chiral
acids were fully consumed. Direct transformation of carboxylic acid and amine to
carboxamides could also be achieved by using triphenylphosphine and carbon
tetrabromide. This was a quick reaction, and the yield was also quite good. However, to
separate the resulting phosphonium salt from the amide product was troublesome.

Therefore, for all the chiral acids, BOP coupling reaction was used to form
amides with carrier B. In all the coupling reactions, large excess amount of carrier B,
1,4-diaminobenzene, was used to improve the conversion of the more expensive chiral
acids and to eliminate the generation of diamides. The yield of the coupling reaction
varied from 70-92%.

Prior to investigate the complexes generated by coupling carrier B to chiral acids
by modeling, the binding between the zinc porphyrin tweezer 2 and carrier B complexes
were investigated by UV-vis, N MR and IR.

The carrier B derivatives provide two binding sites to coordinate with zinc
porphyrin tweezer 2, one is the free amino group, without any doubt, and the other one
could either be the amide nitrogen or amide oxygen (Figure 3-12). Although it is

expected that zinc porphyrins would bind to the amide oxygen22 because of its higher

82

electron density, the possibility of binding through amide nitrogen still exists, since
nitrogen has high affinity to zinc. The possibility of binding through the amide nitrogen
would drastically change the course of our studies, so it was important to fully understand

the binding of the carrier B derivatives to zinc porphyrins.

I
I
I
I
O ’
I \ I
’ I
I ‘ 1

*R1

:" : HN
i. Rz vs, I". (:l Rz
‘H2 NH2

binding through binding through amide
amide nitrogen carbonyl oxygen

Figure 3-12. The possible binding of carrier B derivatives to zinc porphyrin tweezer.
Since we were only concerned about the binding of the amide part of carrier B
derivatives to zinc porphyrins, zinc tetraphenylporphyrin (ZnTPP) was used in place of
zinc porphyrin tweezer in the UV-vis study. Two amides with different bulkiness at the
amide nitrogens (t-butylformamide and dimethylforrnarnide) were added to zinc
tetraphenylporphyrin (ZnTPP). The UV-vis spectra of these two amides binding to

ZnTPP are shown in Figure 3-13.

83

 

(a)

     

 

0.6
416 (nm)
0.4 —0 eq.
Ac , —— 50 eq
‘ 100 eq.
0-2 - —_ 200 eq.
425 (nm)
0 7‘ — " J/
360 410 460
(b) wavelength (nm)
0.8

416( nm)

0.6 i
As ‘ —ger.
0.4 ‘ eq'
100 eq.
1 ——2 0 .
0.2 . ///\I\k 0 eq

ow/

 

360 460
wavelength (nm)

Figure 3-13. UV-vis spectra of adding two amides to ZnTPP. (a) titration with (-
butylformamide; (b) titration with dirnethylformamide

In spectrum (a), with the addition of t-butylformamide to ZnTPP, the absorption
intensity of the Soret band at 416 nm decreases, and a new peak at 425 nm can be
observed. This is confirmed by taking the second derivative of the spectrum (Figure 3-
14). The growing shoulder at 425 nm is reasoned to be the bound t-
butylformamide/ZnTPP complex, which significantly red-shifted the Soret band. On the
other hand, as shown in spectrum (b), when dimethylformamide is added to the ZnTPP,
the Soret band at 416 am also decreases; however, the new shoulder at 425 nm is nearly

negligible.

84

 

 

425 (nm)

416 (nm)

360 410 460

wavelength (nm)
Figure 3-14. The second derivative of the UV-vis spectrum of ZnTPP with t-
butylformamide (200 eq.).

The much weaker peak at 425 nm in spectrum (b) as compared to spectrum (a)
indicates that the ZnTPP binds to t-butylformamide stronger than to dimethylformamide
(DMF). The different binding ability of t-butylformamide and dimethylformamide to
ZnTPP suggests that the amides bind to ZnTPP through amide oxygen instead of
nitrogen. If the binding occurred through the nitrogen, DMF should exhibit higher
binding affinity, since, in this case, the nitrogen atom is more accessible.

The binding between amides and zinc porphyrin was also studied by NMR. NMR
of carrier B derivatized (R)-phenylpropionic acid 13B and (S)-methylbutyric acid 14B
were taken before and after the addition of zinc porphyrin tweezer 2, respectively. Since
the NMR spectra of both derivatives showed the same trend, the NMR of one compound
is shown here. Figure 3-15 depicts the chemical shift of derivative 148 before and after
the binding to zinc porphyrin tweezer 2. Spectrum (a) is the proton NMR of carrier B
derivatized (S)-2-methylbutyric acid 148 by itself, and spectrum (b) is the proton NMR

of compound 14B with 1 equivalent of zinc porphyrin tweezer 2.

85

 

 

b) X

 

‘
~
~
~
~
~
~
‘
~
~
~
I~
‘
~
“
~
~

 

 

Figure 3-15. The binding study of the chiral conjugates 14B with zinc porphyrin tweezer
2 by NMR. ((1) 1H-NMR of carrier B derivatized (S)-2-methylbutyric acid 14B; (b) lH-
NMR of compound 14B with zinc porphyrin tweezer (1 equiv) is added. '

As expected, the carrier phenyl protons Ha and Hb are upfield shifted after the

addition of zinc porphyrin tweezer 2; this is due to the anisotropic shielding effect of the

porphyrin rings when the chiral conjugate binds to the porphyrin tweezer. However, the
two sets of carrier phenyl protons shift upfield to a different extent. The aryl protons Ha
are upfield shifted by 0.62 ppm while Hb shifts by 2.48 ppm. The different shift indicates
that these two sets of protons are at different distance from the porphyrin ring. The larger

upfield shift of Hb indicates their closer proximity to the bound porphyrin as compared to

H3. We would have expected a larger upfield shift for H8 if the porphyrin is bound to the

amide nitrogen. Therefore, the NMR study of conjugate 14B indicates that the zinc
porphyrin binds to the amide oxygen instead of amide nitrogen.

There could be another explanation for the observation that the two sets of carrier

phenyl protons, Ha and Hb, upfield shift to different extent. Considering the possibility

86

that the zinc porphyrin tweezer binds to the chiral conjugates through the amide nitrogen,
because of the steric and electronic difference between the free amino nitrogen and the

amide nitrogen, it is anticipated that the two nitrogens will bind the zinc porphyrin with

different affinity, and thus the time averaged chemical shift for H2‘ would experience less

of a change as compared to Hb.

Figure 3-16. The structure of N -methyl-1,4-phenylenediamine.

In order to eliminate this possibility, we have synthesized N-methyl-l,4-
phenylenediamine (Figure 3-16), and recorded its proton NMR before and after the
addition of zinc porphyrin tweezer 2. This compound binds zinc porphyrin tweezer
through the two amino groups; however, because of the steric difference between the two
nitrogens, the primary amino group should bind the zinc porphyrin stronger than the more
sterically hindered secondary amino group. If the binding affinity does affect the

chemical shift of the protons, then we would expect the two sets of protons to upfield

shift to a different extent after the addition of zinc porphyrin tweezer, with Hb upfield

shifted more than Ha. Based on NMR analysis, both aryl protons, H, and Hb, are shifted

upfield from 6.6 ppm to 3.5 ppm. Therefore, based on NMR study, we are reasonably
confident that the carrier B derivatized conjugates bind to the zinc porphyrins through the
amide oxygen.

The conclusion, that chiral conjugates bind to zinc porphyrin tweezer by amide

oxygen, can be drawn based on both UV-vis and NMR studies, and it was also verified

87

by IR study of conjugate 14B. The major change in the IR upon complexation of 14B
with tweezer 2 is a large increase in the intensity of the amide II band. This is also
attributed to the binding of the zinc porphyrin to the carbonyl oxygen, which results in a
larger C-N amide double bond character and thus an increased intensity in the NH

deformation.

HNJKF’R1
F12

I
|
\

NH2

Figure 3-17. The binding of carrier B derivatives and zinc porphyrin tweezer based on
UV-vis, IR and NMR study.

Based on all the above results, we believe that the carrier B derivatized complexes
bind to zinc porphyrin tweezer through the free amino nitrogen and the carbonyl nitrogen,
as shown in Figure 3-17.

Modeling of the carrier B conjugates bound to Zn-TPP (tetraphenylporphyrin),
results in a more uniform set of structures as compared to conjugates of carrier A.
Modeling of the carrier B derivatized complexes by itself produces a greater number of
consistent conformers. With incorporation of ZnTPP, all conformers of 8B-16B
generated within 1 kcal/mol of the lowest energy conformer lead to predictions of the
same ECCD. Modeling of carrier B conjugates also generated fewer conformers,
indicating a smaller range of viable conformations for these conjugates as compared to

the conjugates of carrier A.

88

 

 

Ph—l )
OCH3

ArHN

 

 

 

 

Figure 3-18. Minimized structure of Zn—TPP bound to 9B (most of the hydrogen atoms
deleted for clarity) in which the chiral center has adopted a conformation to minimize
steric strain.

As Figure 3-18 illustrates, in all the structures studied, the rotation of the Carbonyl-
Cchiral bond is such that the large group is almost perpendicular to the carbonyl with the
small group pointing towards the porphyrin plane and the medium group staggered with
respect to the amide proton (insert). We would therefore expect the porphyrin to slide in
the direction of the small group. The relative size of the substituents is based on the A
values for each.

For those compounds with oxygen containing substituents at the chiral center, the
placement of oxygen containing substituents (invariably the medium group in our group
of compounds) is more syn with respect to the amide hydrogen, which can facilitate

hydrogen bonding between the amide hydrogen and the oxygen at the chiral center. This

89

arrangement leads to the large and small groups bisecting the carbonyl oxygen, and once

again dictates the helicity of the bound porphyrin.

S

Q 3 e ECCD
L M

8
<1 I) 69 ECCD
M L
Figure 3-19. Proposed mnemonic for carrier B derivatized chiral acids.

A mnemonic has been proposed for carrier B derivatized chiral acids based on
modeling studies (Figure 3-19). Looking through the CC=O‘Cchiral bond, viewing with the

carboxylate in front, clockwise orientation of the large, medium, and small groups based
on A values leads to a positive ECCD spectrum with carrier B derivatized chiral
carboxylic acids, and vice versa. The predicted ECCD sign based on the proposed
mnemonic for each acid derivative has been compared to the observed sign from CD
measurement with zinc porphyrin tweezer 2 in Table 3-2.

The zinc in the zinc porphyrin tweezer 2 could be introduced to the porphyrin
tweezer free base by adding zinc salts, such as Zn(OAc)2 or ZnBrz. The zinc porphyrin
tweezer synthesized from both zinc salts coordinated with chiral derivatives and
exhibited ECCD spectra. The sign of the ECCD spectra were the same for both zinc
porphyrin tweezers; however, Zn(OAc)2 porphyrin tweezer showed larger ECCD
amplitude upon addition of the chiral substrates as compared to the ZnBrz derived
porphyrin tweezer. To be consistent, all the UV—vis and ECCD studies were carrier out

by using Zn(OAc)2 as the porphyrin tweezer zinc source.

90

Table 3-2. CD predictions and amplitudes of carrier B derivatived chiral carboxylic

acids bound to porphyrin tweezer 2.

NH2 RchiraICOZH kl Rchiral
O BOP, Et3N, CH2012 O 2‘;
HZN H2N

A
7

 

 

B 88—168

Chiral acids Predicted ECCD Mnemonic As A
#53: 8 (S) “B "99 phISZCH, 222; ((3%)) '82
563% eeee are?)
”56531410 (8) ”B "99 H30 H 2A.. 2233 ((3%)) '88
can... saint:
P:0'9C:1C:ZH12(F?) 123 pos Pr: H cH3 4:290:33) +283
”Sni°2”,, (F?) 133 pos H: H p. gag-1355; -190
03136:?" (S) “B "99 0sz H in. :33 (:22?) ‘48
0:1:%%32H16(S ‘63 "99 at, H 2..., 235.122?) 60
garter“, .06; 23.35.22;

 

Table 3-2 lists the predicted and observed ECCD for compounds 8B-16B. As

mentioned before, the binding of the carrier B derivatized conjugates to the zinc

porphyrin tweezer is weaker than the binding of carrier A derivatized conjugates, so all

91

the CD measurements for carrier B derivatized conjugates were performed at 0 °C, in
order to increase the binding of the conjugates with the zinc porphyrin tweezer 2.
Gratifyingly, the sign of the observed ECCD signals matches the predicted values based
on the mnemonic presented above (Figure 3-19), except for the compound 13B, whose
explanation will be provided in detail later on. Compounds 11B and 12B are a pair of
enantiomers, and as expected, they exhibit opposite ECCD spectra.

The ECCD amplitudes of the chiral derivatives at room temperature are much
smaller than those at low temperature, but the signs remain the same, except for the (S)-2-
methylbutyric acid 14B. It shows small positive ECCD couplet at room temperature, and
the sign switches to negative at 0 °C, which matches the predicted sign.

The other acid derivative we need to pay special attention to is the derivative of
(S)-methoxypheny1acetic acid 8B. For this derivative, the ECCD sign switched with the
addition of great excess of chiral substrate. It showed a positive ECCD sign at room
temperature with he,“ 431 nm (A8 +13) and 424 nm (A8 -27), A +40, when the chiral
substrate was less than 15 equivalents. With increasing amounts of substrate added to the
porphyrin tweezer, a new peak growing at 416 nm was observed, and the CD spectrum
looked like an overlap of two ECCD couplets with opposite sign. Addition of excess
chiral substrate did not completely convert the positive ECCD signal to negative ECCD.
When the CD was taken at 0 °C, the same trend was observed (Figure 3—20). When less
than 5 equivalents of chiral substrate was added, it also showed a positive ECCD sign
With he,“ 43] nm (A8 +26) and 424 nm (A8 -40), A +66, and when the chiral substrate
was increased to 60 equivalents, the sign of the ECCD switched, exhibiting a negative

ECCD couplet with hex, 427 nm (A8 -39) and 420 nm (A8 +43), A —82.

92

— 5 eq.
—— 30 eq.
—60 eq.

 

Figure 3-20. The CD spectra of (S)-methoxyphenylacetic acid 8B (different amounts) at
0 °C.

Compound 13B is a special case in carrier B derivatives. If one considers that
phenyl is larger than methyl group, then the results obtained for 138 does not fit the
prediction, and in fact a positive ECCD would be expected instead of the negative ECCD
spectrum that is observed. However, we believe that possible hydrogen-1c interaction
between the phenyl and the amide proton, as suggested by others for similar systems,”26
effectively places the phenyl group syn to the amide hydrogen, thus yielding the observed
ECCD. This placement of phenyl group syn to the amide hydrogen causes the phenyl
group to behave as the medium group when 13B is bound to tweezer 2.

The chemical shift of the amide proton for carrier B conjugates provides further
eVidence for the syn—like arrangement of the phenyl and the amide hydrogen. The
Chemical shift of the amide proton for 14B with ethyl and methyl substituents is at 7.3
ppm and can be considered as the benchmark for the amide proton that does not interact

with the groups at the chiral center. The chemical shift for SB, 9B, 10B, 11B, and 12B,

93

which contain an oxygen substituent ranges from 7.784 ppm. The downfield shift of
these oxygen containing compounds could be indicative of hydrogen bonding between
the oxygen and the amide hydrogen. The chemical shift for the amide proton of 13B
however resonates at 6.9 ppm, and the upfield shift arises from the shielding of the amide
proton by the phenyl group.

A more direct way to test the existence of hydrogen-1t interaction between amide
hydrogen and phenyl ring in compound 13B was to remove the amide hydrogen by
replacing it with methyl group. In this way, the removal of the amide hydrogen would
eliminate the possible hydrogen-1t interaction, and it was expected that this methylated
carrier B derivative Me-13B would show opposite ECCD sign as compared to the

original carrier B derivative 13B.

 

 

 

NH2 NH2 ‘NH
(300120. EtaN HCHO
THF Pd/C, H2
NH2 750/0 NHBOC 41% NHBOC
Boc-B 17
O O O
BOP, EtaN 98%
CHZS'Z NHBoc NH2
32 /° 18 line-138

Scheme 3-4. The synthesis of methylated carrier B derivative Me-13B.
The synthesis of the methylated compound Me-13B is summarized in Scheme 3-
4. The synthesis of compound Me-13B started with carrier B, 1,4-phenylenediamine.
One of the amino groups was protected by Boc, and the free amino group was mono-
methylated to yield compound 17 by reductive amination with formaldehyde, palladium

on carbon and hydrogen gas.” 28 After the generation of the monomethylated amine 17,

94

it was reacted with the chiral (R)-2-phenylpropionic acid to form amide 18 under BOP
coupling reaction condition. Due to the steric hindrance of the methylated amino group,
the reaction did not go to completion even after overnight stirring. The polarities of the
resulting amide product 18 and the methylated starting material 17 were similar, and the
separation by column chromatography was quite hard, but some of the pure product 18
(32%) was isolated. The Boc protecting group in compound 18 was removed by the
standard TFA condition to generate the desired compound Me-13B.

The UV-vis and CD spectra were recorded for the complex Me-13B by adding it
to zinc porphyrin tweezer 2 in methylcyclohexane. The binding between this methylated

complex Me-13B and zinc porphyrin tweezer 2 was very weak. Upon addition of the
chiral complex, the hmax of the Soret band slightly red shifted from 416.0 nm to 416.7

nm, as compared to about 4 nm red shift of the original derivative 13B. In agreement, the

ECCD signal was also very weak, exhibiting a negative ECCD couplet.

C) C)
120 .. HNJkg/ ‘N’ufi/
. [:1 Ph [:1 Ph

j NH2 NH2
60 133 Me—13B

—13B
—Me-13B

 

-120 l
375 425 475

wavelength (nm)
Figure 3-21. The ECCD spectra of compound 13B and Me-13B.

95

As shown in Figure 3-21, contrary to what we were expecting, the ECCD sign of
the methylated conjugate Me-13B was the same as that of the original conjugate 13B,
although the amplitude was much smaller, and thus weakening the possibility hydrogen-1t
interaction postulated between the amide hydrogen and phenyl group. However, we
could not draw this conclusion based on the study with compound Me-13B, since the
ECCD couplet of this compound was so small. This hydrogen-it interaction hypothesis
will be discussed in more detail later.

Another possibility for the abnormal behavior of compound 13B is that the
consideration of A values21 for analysis of 13B is not valid due to other intramolecular
forces, namely stacking interactions. Our modeling of 13B bound to Zn-TPP suggests
that the phenyl group adopts an arrangement parallel to the porphyrin. The probable
stacking interaction of the phenyl group could be responsible for the tilting of the
porphyrin towards it, thus causing it to behave as a small substituent that results in the

observed ECCD.

The coupling between the carrier B and the chiral acids could result in both the
monoarnidation and diamidation products depending on the amount of carrier B used in
the reaction. The diamidated product could be separated from the monoarnidated
Compound by column chromatography.

Compared to the corresponding monoamides, most of the diamides are less
301 uble in organic solvents, but at the concentration needed for CD spectroscopy

(miCromolar), they are sufficiently soluble in methylcyclohexane in order to carry out all

the UV-vis and CD studies. The diamides could bind to zinc porphyrin tweezer 2

through the amide oxygens, however, the binding of diamides to the zinc porphyrin

96

tweezer 2 was not expected to be as strong as the binding of monoarnides. From UV-vis,
it could be observed that the lmax of the Soret band red shifted with the addition of both
monoamides and diamides, however, the wavelength change with the addition of
monoamides is a bit larger than that with diamides. The wavelength change for the
monoamides binding was usually 4 nm, while it was no more than 2 nm upon the
diamides binding.

The chiral diamide of (S)-acetoxypropionic acid 10 exhibited the same ECCD
sign as the corresponding monoamide when added to the zinc porphyrin tweezer (Am 430

nm (At: -15) and 419 run (As +26), A -41), but the amplitude was only half of that of the
monoamide. On the other hand, the diamide derivative of (S)-methylbutyric acid 14 did
not show any ECCD at all. Since taking the ECCD of diamides does not have advantages
over monoamides, and because of the poor solubility of the diamides in organic solvents,
we have not done an extensive study with these compounds.

3-3. Carrier C for the stereochemical study of chiral carboxylic acids

In the carrier B derivatized chiral acids conjugates, the signs of the ECCD spectra

are determined by the free rotation around the CC=O‘Cchiral bonds. Based on modeling

study (Figure 3-18), the rotation of the Ccarbonyl-Cchiml bond was such that the large group

Was almost perpendicular to the carbonyl with the small group pointed towards the
Porphyrin plane and the medium group staggered with respect to the amide proton. For
those compounds with oxygen containing substituents at the chiral center, the placement
of Oxygen containing substituents (invariably the medium group in our group of
conllaounds) staggered with respect to the amide hydrogen also facilitated hydrogen

bondi ng between the amide hydrogen and the oxygen at the chiral center.

97

NH2 OH
NH? NH2
carrier B carrier C

Figure 3-22. The structure of carrier B and C.

In order to further study the conformation of the CC=Q'CChil-a] bond, we chose to
use carrier C, 4-aminophenol. As shown in Figure 3—22, the only difference between
carrier B and carrier C is that one of the amino functionalities in carrier B is replaced by a
hydroxyl group in carrier C. This hydroxyl group in carrier C would be used to
derivatize with chiral acids to form an ester linkage. This new carrier C was proposed
because there is a better understanding of the rotation between carbonyl carbon and the or

. . . . 9
chiral carbon center in esters than in amides.2

O

 

 

 

OH (300)20. OH RchiraICOZH JOL OJLRchiral
TEA, THF BOP, EtaN, CH2C|2 0 Rchiral TFA I CH20|2
86% 73.91% V E: 98% V
NH
2 NHBoc _ NHBOC NH2
Boc-C 8c,12-14c

Scheme 3-5. The synthesis Of carrier C derivatized chiral acids.

The synthesis of carrier C derivatized chiral acids is shown in Scheme 3-5. Since
in carrier C, the amino group was more reactive than the hydroxyl group, it had to be first
protected with the Boc protecting group. The free hydroxyl group was esterified with the
Chiral carboxylic acids activated by BOP, and subsequently the Boc protecting group was

re moved by treatment with TFA to generate the free amine for CD studies.

Carrier C was derivatized with all the chiral carboxylic acids used with the carrier

A eXCCpt for 9, (S)-acetylmandelic acid, and 10, (S)-acetoxypropionic acid. These two

acids bear an ester functionality that could interfere with the binding between the chiral

98

conjugates and the zinc porphyrin tweezer, which in turn will complicate the correlation
of the stereochemistry of the chiral conjugates to the observed ECCD signs.

Carrier C derivatized chiral carboxylic acids bound to zinc porphyrin tweezer
through coordination with the nitrogen of the free amino group, and the carbonyl oxygen.
However, carrier C derivatized conjugates did not bind to zinc porphyrin porphyrin as
strong as carrier B derivatized conjugates. Although, both carrier B and C derivatives
bind to zinc porphyrins through the carbonyl oxygen, the oxygen in carrier C derivatives
is less nucleophilic than the oxygen in carrier B, since the ester oxygens are less electron
donating than the amide nitrogens. Consequently, the binding of the carbonyl oxygen in
carrier C to zinc porphyrins was weaker. This statement could be supported by UV-vis
studies with zinc porphyrin tweezer. The wavelength change of the Soret band upon
addition of the carrier B derivative was usually 4 nm, while it was less than 2 nm upon
the binding to the carrier C derivatives.

Table 3-3 lists the observed ECCD for carrier C derivatized chiral carboxylic
acids. All the measurements were performed at 0 °C in order to increase the binding of
the chiral conjugates to the zinc porphyrin tweezer. There was no observable ECCD
peak when the measurements were performed at room temperature. All the CD
measurements were taken by using methylcyclohexane as the solvent. Other solvents
Were tested, including benzene, dichloromethane, and acetonitrile, but no ECCD peak

Was Observed when using these solvents. ECCD peaks were observable only in
methylcyclohexane and hexane, and since the bisignate ECCD peak was stronger in

methylcyclohexane than in hexane, methylcyclohexane was used as the solvent of choice

for the ECCD studies.

99

Table 3-3. CD amplitudes Of carrier C derivatized chiral carboxylic acids bound to zinc

 

 

porphyrin tweezer 2.
OH
00 Rchiral
1' F‘lchiralcozlmr—. ——*©Z
NH2 NH
Chiral acid 1st MAe) 2nd MAB) A (0 °C)
HrcozH
Ph 'OCH3 8 (S) 432 (-77) 420 (+55) -132
HYCOQH
PhO CH3 12 (F?) 430 (+9) 420 (-4) +13
HYCOZH
H30 ’Ph 13 (H) 432 (+21) 419 (-14) +35
H .COZH

 

The ECCD signs of the carrier C derivatized (R)-methoxyphenylacetic acid 8C,
(R)-phenoxypropionic acid 12C and (S)-methylbutyric acid 14C were the same with the

corresponding carrier B derivatives, and it suggested that the rotation around CC=O'Cchiral

bond remained the same for both the carrier B and C derivatives. However, the ECCD

sign of the carrier C derivatized (R)-2-phenylpropionic acid 13C was Opposite to the
ECCD sign of the carrier B derivative 13B.

In the carrier B derivative 138, if the phenyl is considered larger than methyl

gr 0Up by A value consideration, then the resultant ECCD sign is Opposite to the one

Proposed by the mnemonic (Section 3-2). This contradiction was explained by the

pOSSible hydrogen-1t interaction between the amide proton and the phenyl group. This

hYdl‘ogen-n interaction effectively places the phenyl group syn to the amide hydrogen,

100

instead of perpendicular to the carbonyl as the large group of the other conjugates did,
and the sign predicted from this conformation was the same as the observed ECCD sign.
In Section 3-2, the syn-like arrangement of the phenyl and the amide hydrogen to
facilitate the possible hydrogen-1r interaction between the phenyl and the amide proton
was partially proved by investigating the chemical shift of the amide proton for carrier B
conjugates. The chemical shift for the amide proton of 13B was more upfield shifted

when compared to the other conjugates, which could be explained by the shielding of the

amide proton by the phenyl group.

0 o
HNJIE/ o’u\../
: Ph : Ph

NH2 NH2

133 13c

Figure 3-23. The structure of 13B and 13C.

Here, the switching of the sign between the carrier B derivatized conjugate 13B
and carrier C derivatized compound 13C gave a more direct evidence of the existence of
hydrogen-1t interaction between amide hydrogen and phenyl ring. As the structures
shown in Figure 3-23, the only difference between these two derivatives was the
Changing from an amide bond (in compound 13B) to ester bond (in 13C). That the
elinlination of hydrogen-1: interaction in compound 13C resulted in the opposite of

ECCD sign further suggested the probable hydrogen-1t interaction between amide

hydrogen and phenyl ring in carrier B derivative.

(R)-2-Pheny1propionic acid 13 was also derivatized with carrier C by forming an

an'lide bond with the amino group to generate compound 13C’, structure shown in Figure

3‘24- If the ECCD sign of 13C’ was opposite to that of the original compound 13C, it

101

would also strengthen the argument for the existence of the hydrogen-1t interaction

between amide hydrogen and phenyl ring.

0 O
ofli/ HNJK/
: Ph : [Sh
NH2 OH
130 130'

Figure 3-24. The structure of 13C and 13C’.

A routine BOP coupling of carrier C, 4—aminophenol, with (R)-2-phenylpropionic
acid was not successful, and led to mixtures of ester and amide derivative product.
Precedented in the literature, we have found a way to selectively form the amide bond in
presence of the hydroxyl functionality (Scheme 3-7).30 Acyloxytriphenylphosphonium
salt 19, generated in situ, is a good acylation reagent, which selectively acylated the
amino group of carrier C, when the reaction was carried out at low temperature. If the
reaction was warmed up to room temperature, the hydroxyl group could also react to

form an ester.

 

 

 

0
e HO/Ki/ O
NBS (+3 N Ph 9 e
PPh3 Ph3PHBr DUO PhaHP~oJ\,/ Br
CH2Cl2 ‘NHS 19 [Sh

 

. . O
O P ridine
6) 9 Y
PhaHP‘oJE/ 3' +HO—©-NH2 H00”)?
l5h

-25°c
63% 133'

Scheme 3-6. The coupling of carrier C with acids to form amide bond.
The coupling of carrier C with (R)-2-phenylpropionic acid is shown in Scheme 3-
6’ The acyloxytriphenylphosphonium salt 19 was formed by the treatment of an

equimolar mixture of a (R)-2-phenylpropionic acid and triphenylphosphine with NBS.

102

Adding the salt 19 to the carrier C, 4-aminophenol, at —25 0C generated the
corresponding amides 13C’ in good yield.

No ECCD was observed upon addition of 13C’ to zinc porphyrin tweezer in
methylcyclohexane at 0 0C. This might be due to the weak binding between the zinc
porphyrin and the phenol. Zinc porphyrins are not known to bind hydroxyl groups well,
however, magnesium porphyrins are capable of binding. Similar to metallation of free
base porphyrin tweezer with zinc, the corresponding magnesium porphyrin tweezer was

synthesized. Addition of 13C’ to magnesium porphyrin tweezer yielded a positive CD
couplet, with Xe,“ 429 nm (A8 —17) and 420 nm (A8 +20.8), A -38 at 0 °C. The sign of the

ECCD couplet changed from positive in compound 13C to negative in compound l3C’.

0H H0 0H
H3C :29 e ECCD Ph =9 (9 ECCD H30 =3 e ECCD
ArHN-Ph H3C OAr Ar‘HN-Ph
133 13c 130
120 , § 2 §
60 -
— 133
A8 0 ——-- 130'
—— 130

 

-120
375 425 475
wavelength (nm)

Figure 3-25. The conformations and ECCD of 13B, 13C and 13C’.

103

From the comparison of ECCD signs of the compound 13B, 13C and 13C’
(Figure 3-25), one could deduce that the amide hydrogen does affect the conformation of
the compounds, and the hydrogen-1t interaction between the amide hydrogen and phenyl
group is highly possible.

3-4. Carrier D for the stereochemical study of chiral carboxylic acids

In order to prove that in the comparison of 13C and 13C’, the binding between

the hydroxyl oxygen to magnesium porphyrin is not the cause of the reversal in the

observed ECCD, we chose to utilize carn'er D, which is 1,4-benzenediol (Figure 3-26).

HO-QOH

Figure 3-26. Structure of carrier D.

One of the hydroxyl groups in carrier D would be derivatized with the chiral
carboxylic acids to form an ester bond, and the other hydroxyl group together with the
resulting ester carbonyl oxygen would serve as the binding sites to porphyrin tweezer.
Not surprisingly, no ECCD was observed by addition of carrier D derivatized compounds
to zinc porphyrin tweezers, since oxygens do not bind to zinc as strongly as nitrogens. In
order to improve the binding of the carrier D derivatives with the porphyrin tweezer, we
decided to use magnesium porphyrin tweezer instead of zinc porphyrin tweezer, because
of the high affinity between oxygen and magnesium.

The synthesis of Mg porphryin tweezer was not as easy as the synthesis of Zn
porphyrin tweezer 2. The Zn porphyrin tweezer 2 could be easily synthesized by adding
excess amount of Zn(OAc)2 to the porphyrin tweezer free base in CHzClz and stirring at
room temperature overnight. The zinc porphyrin is very stable and can to be easily

purified by column chromatography.

104

Magnesium porphyrin tweezer was synthesized by adding excess amount of Mglz
to the porphyrin tweezer free base in CHzClz in the presence of bases, such as

triethylamine.3 '

During the synthesis of magnesium porphyrin tweezer, triethylamine
was necessary to keep the reaction mixture basic, because the magnesium ion embedded
in porphyrins is very sensitive to acid, and the weak acidic conditions generated in the
process of inserting magnesium metal into porphyrins is strong enough to pop the
magnesium out. The reaction was completed in 15 minutes and the excess of MgIz could
be removed by filtration. The removal of the triethylamine from reaction mixture was
more of a challenge, since it bound to the porphyrin by coordinating to magnesium. This
rendered the magnesium porphyrin tweezer incapable of coordinating to the chiral
complexes, and therefore, it was essential to remove all the triethylamine from the Mg
porphyrin tweezer system.

Two ways were explored to remove triethylamine from the magnesium porphyrin
system. Methanol was added to the triethylamine coordinated magnesium tweezer and
the mixture was then extracted by CHzClz. Methanol was added in the hope that it would
replace triethylamine by coordinating to the Mg porphyrin tweezer. However, this
method did not work very well. As can be seen from NMR, the replacement was not
quantitative. More over, the methanol bound to magnesium porphyrin could not be
removed under vacuum, and it would also hinder the coordination of magnesium
porphyrins to the chiral complexes.

The other way to remove triethylamine was by washing the triethylamine

coordinated tweezer with 5% NaHC03 aqueous solution and extracting it with CH2C12.31

This method worked out very well, and from NMR analysis, it was clear that all the

105

triethylamine had been removed. A few words have to be said about taking NMRs of
magnesium porphyrins. When taking NMR of the magnesium porphyrin tweezer, small
amount of deuterated methanol is necessary in order to keep the porphyrins from
aggregating.32 Because of the sensitivity of the magnesium porphyrins to acidic
conditions, the magnesium porphyrin tweezer was used without being purified by column
chromatography.

Before using the magnesium porphyrin tweezer for CD study, its extinction
coefficient (8) was determined. The absorption of certain concentration of magnesium
porphyrin tweezer was measured by UV—vis with spectrophotometric solvents, and the
extinction coefficient scould be calculated based on the Lambert-Beer’s Law:33

A 2 8C] (3-1)

where A is the absorbance and it can be read from UV-vis spectrometry, E is the

extinction coefficient, its unit is M'lcm'l, c is the concentration of the sample and l is the
cell length, which is 1 cm in most cases.

Table 3-4 lists the extinction coefficient 8 of Mg porphyrin tweezer in different
solvents.

Table 3-4. The extinction coefficient eof Mg porphyrin tweezer in different solvents.

 

 

Solvent Extinction coefficient (2) Amax (nm)
Methylcyclohexane 460,000 420
Dichloromethane 570,000 422
Chlorofonn 630,000 426
Acetonitrile 580,000 41 6
Benzene 470,000 426

 

The esterification between the chiral carboxylic acids and the carrier D was easy

and straightforward. The conjugates were formed by BOP coupling of the carrier D with

106

chiral carboxylic acids. Again, (S)-acetylmandelic acid 9 and (S)-acetoxypropionic acid
10 were not used, because these two acids had an ester functionality, which could
compete for binding with the magnesium porphyrin tweezer and complicate the observed

ECCD spectra.

Table 3-5. The CD data of carrier D derivatized chiral acids with magnesium porphyrin

 

 

 

tweezer.
0
OH
RchiraICOZH OJLRchiral
BOP, Et3Nv
OH CHZCIZ
78-89°/o 0”
8D,120-14D
Chiral acid 1st MAe) 2nd MAe) A (0 °C)
H .COZH
phECHa 8 (S) 432 (-12) 421 (+27) -39
H cozH
H .COZH
H co H
- 2 No ECCD peaks

C2HstH3 14 (S)

 

The ECCD data of the carrier D derivatized chiral acids with magnesium
porphyrin tweezer are listed in Table 3-5. The measurements were taken in
methylcyclohexane at O 0C. Except for the derivative of (S)-methylbutyric acid 14, which
did not show any ECCD peak under these conditions, all the other tested chiral acids
derivatives had the same ECCD signs as the corresponding carrier C derivatives. The

conservation in the sign of ECCD for these two different carrier derivatives indicated that

107

the binding between hydroxyl oxygen to magnesium porphyrins worked in the same way
as amino nitrogen bound to zinc porphyrins.

Although carn'er C and D could both be used to derivatize chiral carboxylic acids
and exhibit quite consistent ECCD sign, these two carriers had no remarkable advantage
over carrier B. Since carrier B derivatives have quite strong ECCD signals and zinc
porphyrin tweezer is easier to handle than magnesium porphyrin tweezer, carrier B is the

best carrier candidate for the stereochemical determination of (at-chiral carboxylic acids.

3-5. Carrier E and F for the stereochemical study of a—chiral carboxylic acids

As it has been stated above that carrier B, 1,4-diaminobenzene, is the best carrier,
so far, for the stereochemical determination of a—chiral carboxylic acids. In order to have
a better understanding of the interaction of the carrier B derivatives with zinc porphyrin
tweezer, we proposed a new generation of carriers. Carrier E, 4-aminopyridine, is very

similar in structure as carrier B, structures shown in Figure 3-27.

NH2 NH2
\

' (a
/ | /
NH2 N

carrier B carrier E
Figure 3-27. Structures of carrier B and E.

Carrier E could be derivatized with chiral carboxylic acids to form an amide bond
as carrier B. However these two carriers differ from each other with regards to the
geometry of binding of the un-derivatized nitrogen through the zinc porphyrin. Carrier B
derivatives bind to zinc porphyrin tweezer with the free amino nitrogen, while carrier E
derivatives bind to zinc porphyrin tweezer through the pyridine nitrogen. Since the

location of the electron lone pair of the pyridine nitrogen is more defined in comparison

108

to that of the free amino nitrogen, the location of the zinc porphyrin bound to the pyridine
nitrogen of carrier E is more predictable compared to that of carrier B. The comparison
of the two different carrier-derivatized chiral carboxylic acids would enable us to have
better understanding of the coordination between the chiral conjugates and zinc porphyrin
tweezer.

Carrier E was esterified with chiral carboxylic acids by standard BOP reaction
conditions without any problem. For preliminary studies, three chiral carboxylic acids
were derivatized with the new carrier: (S)-2-methoxyphenylacetic acid 8, (R)-

phenylpropionic acid 13, and (S)-2—methylbutyric acid 14.

O
RchiraICOZH

Jk
BOP, 513M CH2C|2 HN Cchiral
\ 1 \
N’ 73849.. 0

NH2

 

7

8E, 13E and 14E

 

375 425 475

wavelength (nm)

Figure 3-28. The ECCD of carrier E derivatized methylbutyric acid 14E.

109

However, carrier E derivatives did not work as well as expected, according to CD
spectra taken by addition of the chiral derivatives to zinc porphyrin tweezer in
methylcyclohexane. The (S)-2-methoxyphenylacetic acid derivative 8E showed a
positive ECCD couplet, which was opposite to the ECCD observed with carrier B
conjugate 8B. On the other hand, the derivatives of (R)-pheny]propionic acid 13E and
(S)-2-methylbutyric acid 14E did not exhibit simple ECCD spectra (as shown in Figure
3-28). Both spectra looked like an overlap of two ECCD couplets with opposite signs,

and they were not improved when the solution was cooled down to 0 °C.

HNJKFQR 5'", HN’HLR

\
\
\
\
‘ IE

d
d

I \ R2
1‘ I ,
NH2 ‘\\ N
Carrier B derivative Carrier E derivative

Figure 3-29. Carrier B and E derivatized amides bind to zinc porphyrin tweezer.

When carrier B and carrier E derivatives were compared more closely, it was
found that actually carrier E did not mimic carrier B very well in binding to zinc
porphyrin tweezer. Zinc porphyrin binds to the nitrogen electron pair, and the location of
the electron pair in turn determines the location of the zinc porphyrin. Therefore, as
shown in Figure 3-29, when carrier E derivatives bound to the zinc porphyrin tweezer,
the two zinc porphyrin planes would approach to the amide substrate in a psuedo-para
fashion, while when it bound to carrier B derivatives, the two zinc porphyrin planes

would approach the substrate in a psuedo-meta fashion.

110

carrier F Carrier F derivative
Figure 3-30. Structure of carrier F and its derivative binds to zinc porphyrin tweezer.

Based on the latter discussion, naturally, 3-aminopyridine was chosen as a new
carrier candidate (carrier F), since the pyridine nitrogen would be meta- to the amide
functionality. As shown in Figure 3-30, the carrier F derivatives would coordinate with
zinc porphyrin tweezer with the resulting amide carbonyl oxygen and pyridine nitrogen.
Since the amino group was at meta position relative to the pyridine nitrogen, the
porphyrin tweezer would approach to the substrate in 3 meta fashion just similar to what
was presumed with carrier B derivatives.

Carrier F was derivatized with chiral carboxylic acids with standard BOP
conditions. The chiral derivatives were then added to zinc porphyrin tweezer solution in
methylcyclohexane, and CD spectra were measured. The ECCD data for carrier F
derivatives are shown in Table 3-6, and are compared to those of the carrier B

derivatives.

lll

Table 3-6. Comparison of the CD data of carrier B and F derivatives

ONHZ RchiraICOZH OHYRChim' mNHZ RchiraICOZH (j TRchiral
HQN BOP, Et3N, HZN O N] BOP, EtaN, I N, O

 

 

B CH2012 CH20I2
83—148, 168 73-86% 8F-14F, 16F
Carrier B derivatives Carrier F derivatives
Chiral acuds As A As A
430 (-68)
-137 r.t
Ph OCH3 3 (s) 420 (+43) ' 426 (+50) +87 (0 °C)
417 (-37)
H ,CozH 432 (-19) 429 (+368)
PthAc 9(8) 419 (+22) '41 420 (—236) +604
H ,CozH 431 (-50) 430 (+160)
H30 ’OAc 1o (S) 420 (+38) ’88 420 (409) +269
H ,002H 429 (+200) 426 (-73)
PhD CH3 12 (H) 420 (-83) +283 418 (+86) '159
H ,’002H 430 (-105) 490 429 (~10) -31
H36 Ph 13 (R) 420 (+85) 420 (+21)
H aC02H 430 (-22) _48 428 (+14) +28
C2H5CH3 14 (S) 420 (+26) 420 (-14)
H CO H
Y 2 428 (-23) _ 429 (+20)
03H70H3 16(8) 418 (+27) 50 421(-20) +40

 

The ECCD amplitudes of the carrier B derivatives shown in Table 3-6 were
measured at 0 °C, since the amplitudes at room temperature were quite small. As for the
Carrier F derivatives, the amplitudes shown in Table 3-6 were measured at room
temperature, except for the derivatives of (R)-phenoxypropionic acid 12 and (S)-2-
methylbutyric acid 14, which were measured at 0 °C. One thing notable about the carrier

F derivatives was that the derivatives of the acetyl containing acids, such as 9F and 10F,

112

showed extraordinary large ECCD peaks, especially carboxylic acid 9, (S)-0—
acetylmandelic acid, which exhibited an amplitude of 604 at room temperature.

The (S)-2-methylbutyric acid derivative 14F showed a small positive ECCD
couplet at room temperature, however when it was cooled to 0 °C, the ECCD couplet
became much stronger, while the sign remained the same. This observation indicated
that the derivative adopted the same conformation at room temperature and at 0 °C, but
the population of the CD active conformation increased with the lowering of temperature.

The (R)-phenoxypropionic acid derivative 12F did not show a clear ECCD
spectrum at room temperature. It looked like an overlap of two ECCD couplets with
opposite signs, which indicated the presence of more than one ECCD active

conformation. When the solution was cooled to 0 °C, a strong negative ECCD couplet

with Xe,“ 426 nm (A8 —73) and 418 nm (As +86), A —159 was observed. The problem of

multiple conformations at room temperature was alleviated, and only one of them was
dominant at low temperature.

Special attention should be paid to the derivatives of acid 8, (R)-
methoxyphenylacetic acid. For its carrier B derivative 8B, as it was mentioned in Section
3-2, the sign of the ECCD couplet changed with the addition of excess chiral substrate.
When less than 5 equivalents of chiral substrate was added, it showed a positive ECCD
Sign and when the chiral substrate increased to 60 equivalents, it showed a negative
ECCD couplet. While for its carrier F derivative 8F, the sign of ECCD switched when it

Was cooled down from room temperature to 0 °C. At room temperature, it showed a
negative ECCD sign with Am 430 nm (A8 -77) and 421 nm (A8 +74), A -151, with a

Small peak at 414 nm. When the solution was cooled to 0 °C, it switched to a positive

113

ECCD couplet with km 426 nm (A8 +42) and 417 nm (A8 -50), A +92, and this time a

small peak could also be observed at 433 nm (Figure 3-31).

80

H OCH3
| \ N‘n/kPh
40 ~ N’ 0
BF
—cold temp.
——room temp.

 

375 425 475
wavelength (nm)

Figure 3-31. The ECCD of compound 8F at different temperatures.

The switching of the ECCD sign from negative at room temperature to positive at
0 0C indicated that the chiral substrate 8F, upon binding to the zinc porphyrin tweezer,
adopted different conformations at room temperature and at O 0C. At room temperature,
the conformation, which showed positive ECCD signal, was the dominant conformation,
and the small peak at 414 nm indicated the existence of small amount of another
conformation, which led to a negative ECCD couplet. At 0 °C, the minor conformation at
room temperature now became the major one, and the sign of the ECCD spectrum
correspondingly switched from negative to positive. The conformation, which used to be
the major conformation at room temperature, could only now been observed as a small

peak at 433 nm.

114

OH HO

MeO vs. Ph
ArHN—Pb MeO~NHAr
hydrogen-1t interaction hydrogen bonding

dominant at room temperature dominant at 0 00
Figure 3-32. Hydrogen bonding vs. hydrogen—n: interaction.

The ECCD sign of compound 8F switched upon the changing of temperature, and
this temperature dependence might be explained by the existence of two competing
interactions (Figure 3-32), the hydrogen bonding between the amide hydrogen and the
methoxy oxygen, and the hydrogen—1c interaction between the amide hydrogen and the
phenyl group. At room temperature, the hydrogen-1t interaction was the major
interaction, and it placed the phenyl group gauche to the amide hydrogen, thus leading to
the observed negative ECCD spectrum. At 0 °C, the hydrogen bonding between the
amide hydrogen and methoxy oxygen dominated, replacing the hydrogen—1t interaction to
become the major contributor to the conformation. This led to the conformational change
that placed the methoxy group gauche to the amide hydrogen, producing a positive
ECCD spectrum. However, at this point, we do not have any proof for this hypothesis.

When comparing the CD data of all the carrier B and carrier F derivatives, what
became obvious was that all the carrier B derivatives exhibited the opposite ECCD as
compared to the carrier F derivatives, except for the (R)-2-phenylpropionic acid 13. We
do not understand the cause of the sign switching between these two systems, however, it

was not likely due to the conformational change around the CC=O'Cchiral bond, since the

amide functionality remained similar for both carrier B and carrier F derivatives. The

different approaches of the zinc porphyrin tweezer adopted upon binding to the substrate

115

might be one of the possible explanations. As shown in Figure 3-33, the zinc porphyrin

tweezer could approach the chiral substrate from opposite side, and this would lead to the
observed opposite ECCD sign, even though the conformation around the CC=o-Cchi,al

bond remained the same for these two derivatives. So far, due to lack of experimental

evidence, this remains a working suggestion.

HNJK'LR Nfifizfii

E H

'. R2 R

1‘ \ 2 ,
~. 11 /
‘Hz

Carrier B derivative Carrier F derivative

—L

Figure 3-33. The different interactions between the Zinc porphyrin tweezer and carrier B
and carrier F derivatives.
3-6. Use of 2-aminopyrazine as the new carrier G

34' 35 (carrier G) for the

We have also tested the ability of 2-aminopyrazine
stereochemical determination of a—chiral carboxylic acids. 2-Aminopyrazine had an
amino group, which could be easily derivatized with chiral carboxylic acids by forming
amide bond, while the pyrazine functionality provided two binding sites to coordinate
with zinc porphyrin tweezer 2.

An advantage of this new carrier over all the other carriers tested so far was that
the binding position of the zinc porphyrin tweezer relative to the chiral complex could be
precisely predicted. Since the direction of the nitrogen lone pairs were fixed, the sign of

the ECCD spectrum should only depend on the rotation around the Cchiral‘CC=O bond

(shown in Figure 3—34).

116

Carrier G derivative 126
Figure 3-34. The binding of carrier G derivatives 12G to zinc porphyrin tweezer.
The coupling between carrier G and (R)-phenoxypropionic acid 12 was easily
achieved by using BOP as the coupling reagent. However, the resulting chiral conjugate
12G did not shown any ECCD peaks when added to zinc porphyrin tweezer 2 in
methylcyclohexane. The CD was taken both at room temperature and O 0C.

From the results of the UV-vis experiments, we believe that the chiral complex
12G did not bind to zinc porphyrin tweezer, since the absorption wavelength Max of the
Soret band did not change upon the addition of the carrier G derivative. One possible
reason contributing to the lack of binding of the complex 12G to zinc porphyrin tweezer
2 is that the two nitrogen binding sites in 12G are very close to each other. In order to

improve the binding, we shortened the distance between the two porphyrin planes by

synthesizing the zinc porphyrin tweezer 2’ with 1,3-propanediol as the linkage, as shown

in Scheme 3-7.

117

 

EDC, DMAP
4 °o

2. Zn(OAc)2

   

Scheme 3-7. The structure of zinc porphyrin tweezer 2’, in which 1,3-propanediol serves
as the linkage.

The newly synthesized zinc porphyrin tweezer 2’ was tested with carrier F
derivatized (S)-O-acetylmandelic acid 9F. This complex exhibited a strong positive
ECCD spectrum, and the sign of the ECCD spectrum was consistent with previous
consideration. However, when the carrier G derivative 12G was added to tweezer 2’, no
ECCD spectrum was observed. UV-vis spectrum also indicated the lack of binding to the
porphyrin tweezer. Therefore, 2-aminopyrazine was abandoned due to lack of binding.
3-7. Design of a new carrier H for chiral carboxylic acids

We have also designed a new carrier H, o-phenylene diamine, for the
stereochemical determination of Ot-chiral carboxylic acids. As it is shown in Scheme 3-8,
carrier H could form benzimidazoles with chiral carboxylic acid under acidic
conditions.36 The chiral benzimidazole could bind to zinc porphyrin tweezer with the two
nitrogen atoms, and the helicity of the zinc porphyrin tweezer would depend on the

rotation around the C—Cchiral single bond.

118

 

H
NH2 0 4 N HCl ©:N>_/—
/ u
(I + H0 reflux N H"’

NH2 H3
Carrier H 14H

Scheme 3-8. The synthesis of chiral benzimidazole.

(S)-2-methylbutyric acid 14 was derivatized with the carrier H, to form chiral
benzimidazoles 14H by refluxing in 4 N HCl. The product could be easily precipitated
from the reaction mixture by adding ammonium hydroxide. Without further purification,
the benzimidazole 14H was added to the zinc porphyrin tweezer, and UV-vis and CD
spectra were obtained. From UV-vis, we observed the red shift of the maximum
absorption wavelength, which indicated the binding of the chiral substrate to zinc
porphyrin tweezer. CD analysis indicated a small positive ECCD curve at room
temperature. Decreasing the temperature did not improve the amplitude of the couplet.
The solubility of the chiral benzimidazole 14H in methylcyclohexane was quite poor, and
it began to precipitate out of the solution when up to 1 equivalent of the compound 14H
was added to the zinc porphyrin tweezer.

Although only preliminary studies were performed with the carrier H, this method
is promising. We plan to optimize the ECCD spectrum by deprotonating the complex to
improve the binding of the substrate to zinc porphyrin tweezer, and by varying the
solvents for CD measurement.

3-8. Conclusion:
We have utilized a carrier strategy to investigate the Chirality of asymmetric

carboxylic acids, and for this purpose, we have design carriers A-H (Figure 3-35).

119

NH2 “”2 OH OH

[Ll
N “”2 NH2 NH2 0H

carrier A carrier B carrier C carrier D

NH2 NH2 N\ NH2 NH2

1 \ \ E I O

N/ I x N N NH2

carrier E carrier F carrier G carrier H

Figure 3-35. Carrier A-H.

Carrier A, which is analogous to the carrier utilized for chiral amines previously
studied,3 does not yield predictable ECCD spectra, most probably due to the
unpredictable single bond rotamers between the porphyrin binding site and the chiral
center. Carrier B, which can generate a more rigid chiral complex, produced consistent
ECCD spectra, which can be rationalized and predicted by molecular modeling. Carriers
C and D have similar structures as carrier B, and their chiral complexes exhibit consistent
ECCD. But their ECCD signals are not as strong as those of carrier B; even carrier D can
only exhibit ECCD with the Mg porphyrin tweezer, which is less stable than the Zn
porphyrin tweezer. The derivatives of carrier F, 3-aminopyridine, also exhibit consistent
ECCD, however, the signs are all opposite to that of the carrier B derivatives. So far, we
cannot explain the reason for the change in sign. Whether the different approach of the
zinc porphyrin tweezer with respect to the chiral substrate could be an explanation is not
clear as of yet. As for carriers E and G, they are not good carriers, since their complexes
exhibit complicated ECCD spectra. Finally, the carrier H, which is based on a totally

different approach, although only briefly studied, provides promising results.

120

In conclusion, of all the tested carriers, carrier B, 1,4-diaminobenzene, is the best
for the absolute stereochemical determination of a—chiral carboxylic acids. The use of
carrier B to determine the stereochemistry of (at—chiral carboxylic acids by zinc porphyrin

tweezer 2 is displayed in Figure 3-36.

 

 

 

 

 

 

Figure 3-36. The stereochemical study of carrier B derivatized a—chiral carboxylic acids
with zinc porphyrin tweezer 2.In summary, the carrier B derivatized chiral carboxylic
acids bind to zinc porphyrin tweezer 2 with the free amino group and the amide oxygen.
When the conjugate binds to porphyrin tweezer, it adopts a conformation to place the
medium group gauche to amide hydrogen, and the large and small group (size wise) point
towards the porphyrin plane. Accordingly, the zinc porphyrin tweezer adopts the
appropriate helicity to minimize the steric interaction with the groups on the chiral center,
and the handedness of porphyrin tweezer can be observed by CD measurements. In turn,
the absolute stereochemistry of the chiral center can be deduced from the sign of the
resulting ECCD spectrum.
3-9. Another approach towards the same goal

As it has been stated at the beginning of this chapter, Nakanishi’s lab has
developed a very similar idea for absolute stereochemical determination of or-chiral

carboxylic acids, by using the Exciton Coupled Circular Dichroic (ECCD) method

121

together with carrier molecules.” '8 The carrier molecule that they used in their studies
is 1,3-diaminopropane. The chiral substrate was derivatized with the achiral carrier X, to
generate a bifunctional amide conjugate that could form a 1:1 host-guest complex with

the zinc porphyrin tweezer. The chiral conjugate coordinated to the tweezer host 2

through the free amino nitrogen and the carbonyl oxygens” 38

NH
NHBOC HOJS-CH O Zinc porphyrin \Zn/O \NH _. l =
M L H tweezer 2 Zn “—
fiuMNHz 1‘ X

NH2x O 0W0 O

 

 

 

e
LAM
2

Figure 3-37. The binding of chiral conjugate to zinc porphyrin tweezer.

As depicted in Figure 3-37, in the most favored conformation adopted by the
chiral conjugate X, the L (large) and M (medium) groups were pointing toward the
porphyrin plane. The absolute configuration of the chiral substrate could be correlated to
the sign of the ECCD couplet in such manner: viewing with the carboxylic group in the
rear, the clockwise orientation of the L, M and H will resulted in a positive ECCD
spectrum, and vice versa.

The system described above is very similar to the one we deve10ped. Both groups
used zinc porphyrin tweezer as the chromophoric host, and the carriers used were
diamino compounds, we used 1,4-diaminobenzene while they used 1,3-diaminopropane.
Both the carriers’ derivatives showed consistent ECCD sign, and mnemonics were
proposed to link the observed ECCD spectra to the absolute stereochemical determination
of (it-chiral carboxylic acids.

However, there was a major difference between these two carrier systems: the

mnemonics proposed, which were consistent with the observed ECCD signs, were

122

opposite to each other. This was due to the different conformations the substrate adopted
upon binding to zinc porphyrin tweezer. Figure 3-38 is an example with (S)-
methylbutyric acid 14. In our system, the chiral substrate adopted the conformation with
the large group (Et) perpendicular to the carbonyl group, while the medium group (Me)
was gauche to the amide hydrogen. The porphyrin tweezer differentiated between the
large (Et) and small group (H) on the chiral center, which led to a negative ECCD. In
their system, the small group (H) was syn to the amide hydrogen, and it was the large (Et)
and the medium group (Me) that were pointing towards the porphyrin plane, which

resulted in a positive ECCD.

O
HN 2H 0 H
—E Et III) negative ECCD
NH2 RHN Me
carrierB
O
/\/\ 5317/14 EMewEt Q positive ECCD
H2N u ._ RHNH
carrierx

Figure 3-38. Opposite ECCD spectra with the two carriers B and X.

We do not yet understand what causes the conformational change in these two
systems, but they work equally well in determining the absolute stereochemistry of or—

chiral carboxylic acids.

123

Experimental materials and general procedures:

Anhydrous CHzClz was dried over CaHz and distilled. The solvents used for CD
measurements were purchased from Aldrich and were spectra grade. All reactions were
performed in dried glassware under nitrogen. Column chromatography was performed
using SiliCycle silica gel (230—400 mesh). 1H-NMR spectra were obtained on Varian
Inova 300 MHz or 500 MHz instrument and are reported in parts per million (ppm)
relative to the solvent resonances (6), with coupling constants (J) in Hertz (Hz). IR
studies were performed on 8 Galaxy series FI'IR 3000 instrument (Matteson). UV—Vis
spectra were recorded on a Perkin-Elmer Lambda 40 spectrophotometer, and is reported
as 71m” [nm]. CD spectra were recorded on a J ASCO J-810 spectropolarimeter, equipped

with a temperature controller (Neslab 111) for low temperature studies, and is reported as

71. [nm] (Ar-2mx [1 mol’1 cm"]). Porphyrin tweezer 2 was synthesized as previously

described.39 (it-Chiral carboxylic acids 8, 9, 10, 13, 14 and 15 were purchased from
Aldrich. Compound 11 and 12 was synthesized as previously reported.

General procedure for CD measurement:

All the CD spectra were taken with 1 11M porphyrin tweezer in
methylcyclohexane (1 mL) with 40 equiv of the carrier/chiral carboxylic acid conjugate.
The CD spectra were measured in millidegrees and normalized into ASIA [nm] units.
Experiments with carrier A, F were carried out at room temperature. However,
measurements with carrier B, C, D and H were performed at 0 °C in order to increase the

binding of the carrier with the porphyrin tweezer.

124

NH2 (300)20, TEA NHBOC
THF, rt,
I \ overnight A | \

N 95% N
3 4

 

4-[(t-Butoxycarbonyl-amino)methyl] pyridine 4:

To a solution of 4-aminomethylpyridine 3 (0.54 g, 5 mmol) in dry THF (10 mL)
was added di-t-butyl dicarbonate (1.10 g, 5 mmol) and triethylamine (0.16 mL). A white
precipitate was formed and redissolved immediately. The reaction mixture was stirred at
room temperature overnight. The solution was filtered and the filtrate was concentrated
to yield an oil like liquid. The residue was purified by flash chromatography (9:1
CH3Cl/MeOH) to yield the title compound 4 (0.9 g, 95%). 1H NMR (300 MHz, CDC13)
8 8.54 (d, J=5.0 Hz, 2H), 7.20 (d, J=5.0 Hz, 2H), 5.10-5.24 (br, 1H), 4.33 (d, J=4.7 Hz,
2H), 1.47 (s, 9H).

NHBoc NHBoc
mCPBA, CHZCIZ \
l \ fl, 12 h, I ,

__‘
v

N 80% c5
4 5

4-[(t-Butoxycarbonyl-amino)methyl] pyridine oxide 5:

To a solution of 4 (0.5 g, 2.4 mmol) in CHzClz (10 mL) was added mCPBA (1.6
g). The reaction was stirred at room temperature for 18 h and then washed with 0.5 N
aqueous NaOH solution (5 mL). The aqueous layer was extracted with CHzClz (10 x 10
mL) until no product was left in the aqueous phase. The organic layers were combined
and dried by NaZSO4. The solvent was removed under reduced pressure, and the residue
was purified by column chromatography (9:1 CH3Cl/MeOH) to yield the pure compound
5 (0.42 g, 80%). 1H NMR (300 MHz, CD3OD) 5 8.27 (d, J=6.6 Hz, 2H), 7.44 (d, J=6.6

Hz, 2H), 4.27 (d, J=5.7 Hz, 2H), 1.44 (s, 9H).

125

NHBoc
TMSCN CH2C12, NHB°°
|\ rt, 48 h, l \

'35 T N’ CN
\ NRC] 53% 6

 

2-Cyano-4-[(t-Butoxycarbonyl-amino)methyl] pyridine 6:
To a solution of 5 (0.65 g, 3.15 mmol) in CHzClz under N2 was added TMSCN

(0.52 mL, 4.09 mmol). After stirring for 5 minutes, dimethylcarbamyl chloride (0.34 mL,
3.44 mmol) was added. The solution was further stirred at room temperature for 48 h,
and then washed with saturated aqueous NaHCO3 solution (5 mL). The aqueous layer
was washed with CHzClz (3 x 10 mL) and the organic layers were combined and dried
over NaZSO4. The solvent was removed under reduced pressure, and purified by column
chromatography (20:1 CHzClleeOH) to yield the title product 6 (0.42 g, 53%). 1H
NMR (300 MHz, CDC13) 6 8.64 (d, J=5.1 Hz, 2H), 7.63 (d, J=6.7 Hz, 1H), 7.44 (d, J=5.1
Hz, 2H), 5.32-5.42 (br, 1H), 4.38 (d, J=6.2 Hz, 2H), 1.47 (s, 9H).

NHB°° KOH / t-BuOH NHBOC
[KL reflux, 30 min; @NH

’ 96% N 2

NCN o
5 7

 

4-[(t-Butoxycarbonyl-amino)methyl] pyridine 2-carboxylic acid amide 7:

Finely powdered KOH (50 mg) was added to a solution of 2-cyano-4-[(tert-
butoxycarbonylamino)methyl] pyridine 6 (50 mg, 0.23 mole) in t-BuOH (2 mL). The
reaction mixture was refluxed for 30 min. The mixture was cooled and poured into an
aqueous sodium chloride solution (10 mL) followed by extraction with CHC13 (4 x 20
mL). The organic layers were combined and then dried over anhydrous NazSO4, and the

solvent was removed under reduced pressure to afford the amide 7 in 96% yield. IH

126

NMR (300MHz, CDC13) 5 8.49 (d, J=4.9 Hz, 1H), 8.09 (s, 1H), 7.80 (br, 1H), 7.36 (d,
J=4.9 Hz, 1H), 5.56 (br, 1H), 5.01 (br, 1H), 4.39 (d, J=6.6 Hz, 2H), 1.42 (s, 9H).
Compound 7 was used in the next reaction without further purification.
\NHBOC NaOH/Br2 ””300
KERN“? 80 °C, 2 h, 75%; (%

O N NH2
7 Boc-A

 

NHBoc-Carrier A:

A solution of NaOH (60 mg) in water (1 mL) was cooled in an ice-bath, and Br;
(25 111) was slowly added to it. The solution was stirred at 0 °C for 15 min, and the amide
4 (100 mg, 0.4 mmole) was added. This mixture was stirred vigorously for 15 min in an
ice-bath and subsequently moved into an oil bath, which was maintained at 75 °C with
constant stirring for an hour. The solution was saturated with sodium chloride, and
extracted with CHC13 (5 x 25 mL). The combined organic layers were dried over
anhydrous N 32804, and the solvent was removed under reduced pressure. The remaining
residue was purified by column chromatography to yield 75% of the pure product. 1H
NMR (300 MHz, CDC13) 5 7.92 (d, J=5.5 Hz, 1H), 6.51 (d, J=5.5 Hz, 1H), 6.36 (s, 1H),

5.08 (br, 1H), 4.51 (br, 2H), 4.16 (d, J=6.0 Hz, 2H), 1.41 (s, 3H).

0 NaH, PhOH o
HOJKE/ THF, rt, 2.5: HO
Br OPh

 

(R)-Phenoxypropionic acid and (S)-phenoxypropionic acid (11 and 12):

NaH (60% suspension, 20 mg, 0.5 mmole, washed 3 times with hexane) was
added to a solution of (S)-2-bromopropionic acid (77 mg, 0.5 mmole for synthesis of 5 or
(R)-2-bromopropionic acid for synthesis of 6) in THF (5 mL). In a separate vessel NaH

(60% suspension, 40 mg, l mmole, washed 3 times with hexane) was added to a solution

127

of phenol (94 mg, 1 mole) in THF (5 mL). The phenolate solution was cannulated into
the 2-bromopropionate solution, and stirred at room temperature for 2.5 h. The reaction
was quenched with NaOH solution (10 mL, 2 N NaOH), and kept stirring at room
temperature for another hour. The reaction was acidified with HCl (2 N) up to pH=1, and
extracted with 320 (3 x 25 mL). The EtzO extract was dried over anhydrous Na2SO4
and concentrated. The residue was purified by column chromatography to yield the
product. The spectroscopic analysis of the product matched with the reported data. The
enantiomeric excess was determined by derivatizing the acids with (S)-

cyclohexylethylamine, and was determined to be in excess of 90% ee for both acids by

NHBoc 1-Rch1ra1002H NH2
[1:1 BOP, Et3N,CHZCl: {% 9L
N’ NH2 2. TFA, CH2012 N’ H Cchiral

52-78%
Boc-A 8A-15A

GC analysis.

 

General procedure for preparation of carrier Alcarboxylic acid conjugates (8A-
15A):

To a solution of carrier A (1 equiv) in CH2C12 (2 mL), BOP (1.1 equiv),
triethylamine (5 equiv), and the chiral carboxylic acid (1 equiv) were added. The
solution was stirred at room temperature overnight. After removal of most of the solvent
under reduced pressure, the product was purified by silica gel column chromatography to
yield the Boc-protected conjugate. The Boc group was removed by stirring overnight in
TFA/CHzClz (1:4) at room temperature. The solvent was then removed by purging
nitrogen. The residue was redissolved in anhydrous CHzClz, and anhydrous NazCO3 (20

mg) was added. This solution was stirred at room temperature for an hour, and the

128

NazCO3 was filtered. The solvent was evaporated under reduced pressure to provide

NHBoc
{L1 0
N’ NJk/Ph

H OCH3
Boc-8A

analytically pure product.

Boc protected carrier Alcarboxylic acid conjugates Boc-8A:
1H NMR (300 MHz, CDC13) 5 9.17 (br, 1H), 8.24 (d, J=4.9 Hz, 1H), 8.14 (s, 1H), 7.45

(m, 5H), 7.00 (d, J=4.9 Hz, 1H), 5.03 (br, 1H), 4.73 (s, 1H), 4.30 (d, J=6.5 Hz, 2H), 3.43

NH2
(£1 0
N’ NJVPh

H OCH3
8A

(5, 3H), 1.46 (s, 9H).

Carrier Alcarboxylic acid conjugates 8A:
lH NMR (300 MHz, CDCl;,) 5 9.17 (br, 1H), 8.24 (d, J=4.9 Hz, 1H), 8.14 (s, 1H), 7.45
(m, 5H), 7.00 (d, J=4.9 Hz, 1H), 4.73 (s, 1H), 3.84 (s, 2H), 3.43 (s, 3H), 1.58 (br, 2H).

NHBoc
{LI 0
N’ N’UYP"

H OCOCH3
Boc-9A

Boc protected carrier Alcarboxylic acid conjugates Boc-9A:
'H NMR (300 MHz, CDCl3) 5 8.75 (br, 1H), 8.23 (d, J=4.9 Hz, 1H), 8.14 (s, 1H), 7.50
(m, 2H), 7.41 (m, 3H), 7.03 (d, J=4.9 Hz, 1H), 6.21 (s, 1H), 5.02 (br, 1H), 4.33 (d, J=6.6,

Hz, 2H), 2.27 (s, 3H), 1.47 (s, 9H).

129

NH2
(L1 0
N2 J|\_,Ph

H OCOCH3
9A

Carrier Alcarboxylic acid conjugates 9A:
lH NMR (300 MHz, CDC13) 5 8.75 (br, 1H), 8.23 (d, J=4.9 Hz, 1H), 8.13 (s, 1H), 7.51
(m, 2H), 7.41 (m, 3H), 7.07 (d, J=4.9 Hz, 1H), 6.19 (s, 1H), 3.90 (s, 2H), 2.26 (s, 3H).

NHBoc
(£1 0
/ ,lk/

N N ,
H OCOCH3
Boc-10A

Boc protected carrier Alcarboxylic acid conjugates Boc-10A:
lH NMR (300 MHz, CDCl3) 5 8.66 (br, 1H), 8.23 (d, J=4.9 Hz, 1H), 8.14 (s, 1H), 7.03
(d, J=4.9 Hz, 1H), 5.35 (q, J=6.8 Hz, 1H), 5.02 (br, 1H), 4.33 (d, J=6.6 Hz, 2H), 2.22 (s,

3H), 1.56 (d, J=6.8 Hz, 3H), 1.48 (s, 9H).

NH2
(3°
, Jk/

NH,
OCOCH3
10A

Carrier Alcarboxylic acid conjugates 10A:
'H NMR (300 MHz, CDC13) 5 8.43 (br, 1H), 8.22 (d, J=4.9 Hz, 1H), 8.17 (s, 1H), 7.07
(d, J=4.9 Hz, 1H), 5.38 (q, =6.6 Hz, 1H), 3.90 (s, 2H), 2.20 (s, 3H), 1.55 (d, J=6.6 Hz,

3H).

130

NHBoc
6:1 0
N’ N’u\./
H OPh
Boc-1 1 A
Boc protected carrier Alcarboxylic acid conjugates Boc-1 lA/12A:

1H NMR (300 MHz, CDC13) 5 8.66 (br, 1H), 8.16 (m, 2H), 7.24 (m, 3H), 6.99 (m, 3H),

4.95 (br, 1H), 4.79 (q, J=6.9 Hz, 1H), 4.33 (d, J=6.6 Hz, 2H), 1.61 (d, J=6.9 Hz, 3H),

NH2

[£1 0
N’ N’K/
OPh

11A

1.45 (s, 9H).

Carrier Alcarboxylic acid conjugates 11A/12A:
1H NMR (300 MHz, CDC13) 5 8.82 (br, 1H), 8.23 (m, 2H), 7.28 (m, 3H), 6.99 (m, 3H),
4.78 (q, J=6.9 Hz, 1H), 3.91 (s, 2H), 1.63 (d, J=6.9 Hz, 3H).

NHBoc
(L1 0
N’ NJE/

H 1511
Boc-13A

Boc protected carrier Alcarboxylic acid conjugates Boc-13A:
1H NMR (300 MHz, CDC13) 5 8.54 (br, 1H), 8.14 (d, J=5.1 Hz, 1H), 7.96 (s, 1H), 7.38
(m, 5H), 6.93 (d, J=5.1 Hz, 1H), 4.93 (br, 1H), 4.30 (d, J=6.5 Hz, 2H), 3.72 (q, J=7.2 Hz,

1H), 1.57 (d, J=7.2 Hz, 3H), 1.44 (s, 9H).

131

13A
Carrier Alcarboxylic acid conjugates 13A:
lH NMR (300 MHz, CDC13) 5 8.22 (br, 1H), 8.15 (d, J=5.1 Hz, 1H), 7.80 (s, 1H), 7.38
(m, 5H), 7.01 (d, J=5.1 Hz, 1H), 3.90 (s, 2H), 3.74 (q, J=7.2 Hz, 1H), 1.60 (d, J=7.2 Hz,
3H).
NHBoc
r5 0
N’ HJH/
Et
Boc-14A
Boc protected carrier Alcarboxylic acid conjugates Boc-14A:
1H NMR (300 MHz, CDCl3) 5 8.21 (m, 2H), 7.99 (br, 1H), 6.95 (d, J=4.9 Hz, 1H), 4.95
(br, 1H), 4.36 (d, J=6.5 Hz, 2H), 2.27 (m, 1H), 1.78 (m, 1H), 1.51 (m, 1H), 1.43 (s, 9H),

1.22 (d, J=6.9 Hz, 1H), 0.94 (t, J=6.9 Hz, 3H).

NH2
N’ ”JV
Et

14A

Carrier Alcarboxylic acid conjugates 14A:
lH NMR (300 MHz, CDC13) 5 8.24 (m, 2H), 7.95 (br, 1H), 7.01 (d, J=4.9 Hz, 1H), 3.88
(s, 2H), 2.27 (m, 1H), 1.72 (m, 1H), 1.52 (m, 1H), 1.22 (d, J=6.9 Hz, 1H), 0.94 (t, J=6.9

Hz, 3H).

132

NHBoc
{L1 0
N’ ukan

Boc-15A
Boc protected carrier Alcarboxylic acid conjugates Boc-15A:
IH NMR (300 MHz, CDCl3) 5 8.12 (d, J=5.1 Hz, 1H), 7.81 (s, 1H), 7.25 (m, 6H), 6.92
(d, J=5.1 Hz, 1H), 5.03 (br, 1H), 4.30 (d, J=6.5 Hz, 2H), 3.08 (m, 1H), 2.66 (m, 2H), 1.60

(d, J=7.2 Hz, 3H), 1.46 (s, 9H).

NH2
{L1 0
N’ 11*?”

15A

Carrier Alcarboxylic acid conjugates 15A:
1H NMR (300 MHz, CDCl3) 5 8.12 (d, J=5.1 Hz, 1H), 7.70 (s, 1H), 7.25 (m, 5H), 6.92

(d, J=5.1 Hz, 1H), 3.87 (s, 2H), 3.08 (m, 1H), 2.66 (m, 2H), 1.60 (d, J=7.2 Hz, 3H).

NH2 RauncozH n R .
O BOP, Eth, CH2012 O \n’ Chiral

3 70-92% 2 33.153

 

General procedure for preparation of carrier B/carboxylic acid conjugates:

To a solution of carrier B (5 equiv) in CHgClz (4 mL), BOP (1.1 equiv),
triethylamine (5 equiv), and the chiral carboxylic acid (1 equiv) were added. The
solution was stirred at room temperature overnight. After removal of most of the solvent
under reduced pressure, the product was purified by silica gel flash column

chromatography to yield the carrier Blacid conjugate ready for use for CD analysis.

133

o
HNJk/Ph

OCH3
NH2 88
Carrier B/carboxylic acid conjugates 8B:
1H NMR (300 MHz, CDC13) 5 8.34 (br, 1H), 7.39 (m, 7H), 6.62 (d, J=8.8 Hz, 2H), 4.68

(s, 1H), 3.41 (s, 3H), 2.61 (br, 2H).

0
HNkPh

OCOCH3

NH2 93
Carrier B/carboxylic acid conjugates 9B:
lH NMR (300 MHz, CDCl3) 5 7.78 (br, 1H), 7.51 (m, 2H), 7.40 (m, 3H), 7.28 (d, J=8.2

Hz, 2H), 6.63 (d, J=8.2 Hz, 2H), 6.18 (s, 1H), 3.31 (br, 2H), 2.23 (s, 3H).

N’K/

o
E: OCOCH3

NH2 108

I

Carrier B/carboxylic acid conjugates 10B:
lH NMR (300 MHz, CDC13) 5 7.68 (br, 1H), 7.28 (d, J=8.8 Hz, 2H), 6.63 (d, J=8.8 Hz,

2H), 5.30 (q, J=7.14 Hz, 1H), 3.63 (br, 2H), 2.15 (s, 3H), 1.51 (d, J=7.14 Hz, 3H).

134

O
HNJkg/
E: OPh
NH, 118/128
Carrier B/carboxylic acid conjugates 11B/12B:

1H NMR (300 MHz, CDCl3) 5 8.06 (br, 1H), 7.33 (m, 4H), 7.02 (m, 3H), 6.65 (d, J=8.8

Hz, 2H), 4.77 (q, J=6.6 Hz, 1H), 3.63 (br, 2H), 1.66 (d, J=6.6 Hz, 3H).

1‘

HN

©

NH2 138

O
p

h

Carrier B/carboxylic acid conjugates 13B:
1H NMR (300 MHz, CDC13) 5 7.34 (m, 5H), 7.14 (d, J=8.8 Hz, 2H), 6.81 (br, 1H), 6.58

(d, J=8.8 Hz, 2H), 3.66 (q, J=7.1 Hz, 1H), 3.55 (br, 2H), 1.57 (d, J=7.1 Hz, 3H).

0
”Y
Et

I
Z

2

H2 148

Carrier B/carboxylic acid conjugates 14B:

lH NMR (300 MHz, CDCl3) 5 7.31 (br, 1H), 7.09 (d, J=8.8 Hz, 2H), 6.41 (d, J=8.8 Hz,
2H), 3.39 (br, 2H), 2.03 (m, 1H), 1.51 (m, 1H), 1.26 (m, 1H), 0.99 (d, J=6.6 Hz, 1H),

0.73 (t, J=6.6 Hz, 3H).

135

o
HNJK/

(:1 CHgPh

NH2 158
Carrier B/carboxylic acid conjugates 15B:
1H NMR (300 MHz, CDC13) 5 7.22 (m, 5H), 7.07 (d, J=8.8 Hz, 2H), 6.69 (br, 1H), 6.57

(d, J=8.8 Hz, 2H), 2.99 (m, 1H), 2.71 (m, 1H), 2.04 (m, 1H), 1.23 (d, J=6.6 Hz, 1H).

0

J‘xrv

H2 168

I
Z

2

Carrier B/carboxylic acid conjugates 16B:
1H NMR (CDC13, 300 MHz) 5 7.30 (d, J=8.8 Hz, 2H), 7.12 (br, 1H), 6.67 (d, J=8.8Hz,
2H), 3.50 (br, 2H), 2.31 (m, 1H), 1.73 (m, 1H), 1.35 (m, 2H), 1.22 (d, J=6.6 Hz, 3H),

0.94 (d, J=7.7 Hz, 3H).

 

(BOC)QO, ElgN
THF
NH2 75% NHBoc

Boc-B
Boc-protected carrier B:
To a solution of carrier B, 1,4-diaminobenzene, (0.54 g, 5 mmol) in dry THF (10
mL) was added di-t—butyl dicarbonate (1.10 g, 5 mmol) and triethylamine (0.16 mL). A
white precipitate was formed and redissolved immediately. This reaction mixture was
stirred at room temperature overnight. The reaction was concentrated under reduced

pressure, and the residue was purified by flash chromatography (75% yield). lH NMR

136

(CDC13, 300 MHz) 5 6.98 (d, J=7.7 Hz, 2H), 6.50 (d, J=8.2 Hz, 2H), 3.35 (br, 2H), 1.34

 

(s, 9H).
NH2 ‘NH
HCHO
Pd/C, H2
NHBoc 41% NHBoc
Boc-B 17

Methylated Boc-protected carrier B 17:

The protected carrier Boc-B (41.6 mg, 0.2 mmol) was dissolved in EtOAc (1 mL).
Formaldehyde (0.25 mmol) and Pd/C (10 mg) were added. This solution was then
hydrogenated with atmospheric pressure H2 for 20 h. The black powder was filtered, and
the filtrate was concentrated under reduced pressure, and purified by column
chromatography to yield the title product 17 (41%). 1H NMR (CDC13, 300 MHz) 5 7.14

(d, J=7.7 Hz, 2H), 6.51 (d, J=8.8 Hz, 2H), 3.51 (br, 1H), 2.74 (s, 3H), 1.48 (s, 9H).

0 0
‘NH Hok/ \NJK/
Ph Ph

 

BOP, Et3N
NHBoc CHZC'Z NHBoc

Methylated Boc-B/carboxylic acid conjugate 18:

To a solution of Boc protected carrier compound 17 (111 mg, 0.5 mmol) in
CH2C12 (6 mL), BOP (250 mg, 0.55 mmol), triethylamine (1 mL), and the chiral acid (75
mg, 0.5 mmol) were added. The solution was stirred at room temperature overnight.
After removal of most of the solvent under reduced pressure, the product was purified by
silica gel flash chromatography (32% yield). 1H NMR (CDCl3, 300 MHz) 5 7.28 (d,
J=8.8 Hz, 2H), 7.04 (d, J=7.7 Hz, 2H), 6.94 (d, J=8.2 Hz, 2H), 6.75 (m, 2H), 6.85 (d,

J=8.2 Hz, 2H), 4.56 (q, J=6.6 Hz, 1H), 3.07 (s, 3H), 1.36 (s, 9H), 1.28 (d, J=6.0 Hz, 3H).

137

O

 

 

0
OH RchiralCOZH JL 0 R h. .
BOP, Et3N, CHZCIZ 0 Rchiral TFA/CH2C12 ° "8
73-91% 7 98% 7
NHBoc NHBoc NH2
Boc-C 8c, 12-14c

General procedure for preparation of carrier C/carboxylic acid conjugates:

To a solution of carrier Boc-C (5 equiv) in CHzClz (4 mL), BOP (1.1 equiv),
triethylamine (5 equiv), and the chiral carboxylic acid (1 equiv) were added. The
solution was stirred at room temperature overnight. After removal of most of the solvent
under reduced pressure, the product was purified by silica gel column chromatography to
yield the Boc-protected acid conjugate. The Boc group was removed by stirring
overnight in TFA/CHzClg (1:4) at room temperature. The solvent was then removed by
purging nitrogen. The residue was redissolved in anhydrous CHzClz, and anhydrous
N32CO3 (20 mg) was added. This solution was stirred at room temperature for an hour,
and the NazCO3 was filtered. The solvent was evaporated under reduced pressure to

provide analytically pure product.

[:1 OCH3

HBOC Boc-8C

o
OJKrPh
N

Boc protected carrier C/carboxylic acid conjugates Boc-8C:
1H NMR (CDC13, 300 MHz) 5 7.56 (m, 2H), 7.40 (m, 3H), 7.32 (d, J=8.8 Hz, 2H), 6.92

(d, J=8.8 Hz, 2H), 6.62 (br, 1H), 5.00 (s, 1H), 3.51 (s, 3H), 1.49 (s, 9H).

138

O
OJKrPh
E: OOH3
Carrier C/carboxylic acid conjugates 8C:

1H NMR (CDC13, 300 MHz) 5 7.56 (m, 2H), 7.40 (m, 3H), 6.78 (d, J=8.8 Hz, 2H), 6.61

(d, J=8.8 Hz, 2H), 4.99 (s, 1H), 3.52 (s, 3H).

O
OJYOPH

1 Boc-12C

NHBoc

Boc protected carrier C/carboxylic acid conjugates Boc-12C:
1H NMR (CDC13, 300 MHz) 5 7.35 (m, 4H), 6.96 (m, 5H), 6.54 (br, 1H), 4.98 (d, J=7.1

Hz, 1H), 1.78 (q, J=7.1 Hz, 3H), 1.50 (s, 9H).

O
OJKFOPh

12C

NH2
Carrier C/carboxylic acid conjugates 12C:
1H NMR (CDC13, 300 MHz) 5 7.33 (m, 2H), 6.98 (m, 3H), 6.82 (m, 4H), 5.76 (br, 1H),

4.95 (q, J=7.1 Hz, 1H), 1.78 (d, J=7.1 Hz, 3H).

139

O
OJKrPh

1 Boc-13C

NHBOC

Boc protected carrier C/carboxylic acid conjugates Boc-13C:

1H NMR (CDC13, 300 MHz) 5 7.40-7.31 (m, 7H), 6.94 (d, J=8.8 Hz, 2H), 6.58 (br, 1H),

3.98 (q, =7.1 Hz, 1H), 1.64 (d, J=7.1 Hz, 3H), 1.49 (s, 9H).

O
OJth
1 130

NH2

Carrier C/carboxylic acid conjugates 13C:

lH NMR (CDC13, 300 MHz) 5 7.36 (m, 5H), 6.79 (d, J=8.8 Hz, 2H), 6.64 (d, J=8.8 Hz,

2H), 3.92 (q, J=7.1 Hz, 3H), 3.32 (br, 2H), 1.62 (t, J=7.1 Hz, 3H).

05A

1
1 800-140
0

NHBO

Boc protected carrier C/carboxylic acid conjugates Boc-14C:
‘H NMR (CDCl3, 300 MHz) 5 7.36 (d, J=8.8 Hz, 2H), 6.98 (d, J=8.8 Hz, 2H), 6.64 (br,

1H), 2.61 (m, 1H), 1.83 (m, 1H), 1.63 (m, 1H), 1.50 (s, 9H), 1.27 (q, J=7.1 Hz, 3H), 1.02
(t, J=7.1 Hz, 3H).

140

 

O
0%
E22 14c
Carrier C/carboxylic acid conjugates 14C:
1H NMR (CDC13, 300 MHz) 5 6.88 (d, J=8.8 Hz, 2H), 6.69 (d, J=8.8 Hz, 2H), 3.39 (br,
2H), 2.60 (m, 1H), 1.81 (m, 1H), 1.64 (m, 1H), 1.26 (q, J=7.1 Hz, 3H), 1.02 (t, J=7.1 Hz,

3H).

H Ph

0
N
: 13C'

OH

The synthesis of compound 13C’:

To a stirred solution of PPh3 (26.8 mg, 0.10 mmol), (R)-pheny1propionic acid (10
mg, 0.1 mmol) in anhydrous CHzClz (0.5 mL) at 0 °C was added NBS (20 mg, 1.12
mmol) in one portion. The mixture was set aside at room temperature, while making the
other solution. 4-Aminophenol (10.9 mg, 0.10 mmol) and pyridine (10.1 mg, 1.28 mmol)
was dissolved in THF (0.5 mL), and the mixture was cooled to —25 0C. Then the first
solution was added dropwise. The cooling bath was removed after 15 min and warm to
room temperature. The solvent was removed under reduced pressure, and the residue
was separated by column chromatography to yield the title compound (63%). lH NMR
(CDC13, 300 MHz) 5 7.39 (m, 5H), 7.22 (d, J=8.8 Hz, 2H), 5.97 (br, 2H), 6.75 (d, J=8.8

Hz, 2H), 5.76 (br, 2H), 3.72 (q, J=7.1 Hz, 3H), 1.57 (t, J=7.1 Hz, 3H).

141

O

 

OH
RchiralCO2H OJLRchiral
BOP, E13N
OH CH2012
78-89% 0”
80, 120-140

General procedure for preparation of carrier D/carboxylic acid conjugates:

To a solution of carrier D (5 equiv) in CHzClz (4 mL), BOP (1.1 equiv),
triethylamine (5 equiv), and the chiral acid (1 equiv) were added. The solution was
stirred at room temperature overnight. After removal of most of the solvent under

reduced pressure, the product was purified by silica gel column chromatography.

0
OJKrPh

OCH3

OH 80

Carrier D/carboxylic acid conjugates 8D:
1H NMR (CDC13, 300 MHz) 5 7.56 (m, 2H), 7.42 (m, 3H), 6.84 (d, J=8.8 Hz, 2H), 6.75

(d, J=8.8 Hz, 2H), 5.73 (br, 2H), 4.99 (s, 1H), 3.50 (s, 3H).

O
OJH/OPh

: 120

CH

Carrier D/carboxylic acid conjugates 12D:
lH NMR (CDC13, 300 MHz) 5 7.32 (m, 3H), 7.02 (m, 2H), 6.87 (d, J=8.8 Hz, 2H), 6.79

(d, J=8.8 Hz, 2H), 4.98 (q, J=7.1 Hz, 1H), 1.78 (d, J=7.1 Hz, 3H).

142

Carp.
©

13D
OH

Carrier D/carboxylic acid conjugates 13D:
IH NMR (CDC13, 300 MHz) 5 7.36 (m, 5H), 6.82 (d, J=8.8 Hz, 2H), 6.76 (d, J=8.8 Hz,
2H), 5.23 (br, 2H), 3.96 (q, J=7.1 Hz, 3H), 1.62 (t, J=7.1 Hz, 3H).
0
0%
E21 140

Carrier D/carboxylic acid conjugates 14D:
1H NMR (CDC13, 300 MHz) 5 6.90 (d, J=8.8 Hz, 2H), 6.77 (d, J=8.8 Hz, 2H), 6.38 (br,
2H), 2.62 (m, 1H), 1.83 (m, 1H), 1.67 (m, 1H), 1.28 (q, J=7.1 Hz, 3H), 0.99 (t, J=7.1 Hz,

3H).

0
RchiraICOZH

JL
BOP, Et3N, OHch2 HN Cchiral
\ - \
N, 73-84% (I:

NH2

 

7

I
BE, 13E and 14E

General procedure for preparation of carrier E/carboxylic acid conjugates:

To a solution of carrier E (5 equiv) in CH2C12 (4 mL), BOP (1.1 equiv),
triethylamine (5 equiv), and the chiral carboxylic acid (1 equiv) were added. The
solution was stirred at room temperature overnight. After removal of most of the solvent

under reduced pressure, the product was purified by silica gel column chromatography.

143

O
HNJKrPh

I OCH3
as

z\/

Carrier E/carboxylic acid conjugates 8E:
1H NMR (CDCl3, 300 MHz) 5 8.79 (br, 1H), 8.45 (d, J=5.5 Hz, 2H), 7.56 (d, J=5.5 Hz,

2H), 7.37 (m, 5H), 4.74 (s, 1H), 3.41 (s, 1H).

O
N’HA
\

N 13E

I

Carrier E/carboxylic acid conjugates 13E:
1H NMR (CDC13, 300 MHz) 5 8.45 (d, J=5.5 Hz, 2H), 7.64 (br, 1H), 7.58 (d, J=5.5 Hz,

2H), 2.39 (m, 1H), 1.79 (m, 1H), 1.57 (m, 1H), 1.24 (d, J=6.6 Hz, 3H), 0.99 (d, J=7.7 Hz,

3H).
0
HN Ph
| \
N/ 14E

Carrier E/carboxylic acid conjugates 14E:
lH NMR (CDC13, 300 MHz) 5 8.30 (d, J=5.5 Hz, 2H), 7.44 (br, 1H), 7.57 (d, J=5.5 Hz,

2H), 7.27 (m, 5H), 3.77 (q, J=6.6 Hz, 1H), 1.62 (d, J=6.0 Hz, 3H).

WNW RchiraICO2H m Tachiral
/ /) O

 

N BOP Et3N,
F CH2C|2
73-86% 8F-14F, 16F

General procedure for preparation of carrier F/carboxylic acid conjugates:

To a solution of carrier F (5 equiv) in CHzClz (4 mL), BOP (1.1 equiv),

triethylamine (5 equiv), and the chiral carboxylic acid (1 equiv) were added. The

144

solution was stirred at room temperature overnight. After removal of most of the solvent

under reduced pressure, the product was purified by silica gel column chromatography.

0
HNJKrOCHa
6 Ph
/N 8F
Carrier F /carboxylic acid conjugates 8F:
'H NMR (CDC13, 300 MHz) 5 8.64 (m, 2H), 8.39 (d, J=4.9 Hz, 1H), 8.13 (d, J=9.4 Hz,

1H), 7.58 (m, 5H), 7.23 (m, 1H), 4.80 (s, 1H), 3.45 (s, 3H).

O
HNJH’OAC
6 Ph
/N 9F
Carrier Flcarboxylic acid conjugates 9F:
1H NMR (CDC13, 300 MHz) 5 8.57 (m, 2H), 8.34 (d, J=4.9 Hz, 1H), 8.18 (d, J=9.4 Hz,
1H), 7.51 (m, 2H), 7.40 (m, 3H), 7.23 (m, 1H), 6.19 (s, 1H), 2.27 (s, 3H).

0
HNJKFOAC
O
I ,N 10F
Carrier Flcarboxylic acid conjugates 10F:
1H NMR (CDC13, 300 MHz) 5 8.61 (s, 1H), 8.39 (d, J=4.9 Hz, 1H), 8.22 (d, J=8.2 Hz,

1H), 8.10 (br, 1H), 7.30 (m, 1H), 5.37 (q, J=7.1 Hz, 1H), 2.12 (s, 1H), 1.58 (d, J=7.1 Hz,

3H).

145

C)
HN

b OPh

/N 12F

Carrier Flcarboxylic acid conjugates 12F:

lH NMR (CDC13, 300 MHz) 5 8.62 (s, 1H), 8.37 (m, 2H), 8.20 (d, J=9.4 Hz, 1H), 7.40

(m, 3H), 7.23 (m, 3H), 4.85 (q, J=7.1 Hz, 1H), 1.69 (d, J=6.6 Hz, 3H).

0

b Ph
,N 13F

Carrier Flcarboxylic acid conjugates 13F:
1H NMR (CDC13, 300 MHz) 5 8.45 (s, 1H), 8.32 (d, J=4.4 Hz, 1H), 8.20 (d, J=8.2 Hz,

1H), 7.48 (m, 6H), 3.78 (q, J=7.1 Hz, 1H), 1.63 (d, J=7.1 Hz, 3H).

O
HNJKl/\
| \

,N 141:

Carrier Flcarboxylic acid conjugates 14F:
lH NMR(CDC13, 300 MHz) 8.61 (s, 1H), 8.37-8.22 (m, 2H), 7.43 (br, 1H), 7.33 (m, 1H),

2.37 (m, 1H), 1.79 (m, 1H), 1.57 (m, 1H), 1.22 (d, J=6.6 Hz, 3H), 1.01 (d, J=7.2 Hz, 3H).

O
HN)I\'/\/
\
| ,N 16F

Carrier Flcarboxylic acid conjugates 16F:
1H NMR (CDC13, 300 MHz) 5 8.53 (s, 1H), 8.37 (d, J=4.4 Hz, 1H), 8.19 (d, J=7.7 Hz,
1H), 7.37 (br, 1H), 7.26 (m, 1H), 2.36 (m, 1H), 1.70 (m, 1H), 1.42 (m, 2H), 1.22 (d,

J=7.1 Hz, 3H), 0.89 (d, J=7.1 Hz, 3H).

146

O
Ho’ufi/ OPh

N\ NH2 OPh _ N\ (We
[NJ BOP, Eth, 7 EN; 0
CHzCIQ 76% 126

 

Carrier G derivative 12G:

To a solution of carrier G (5 equiv) in CHzClz (4 mL), BOP (1.1 equiv),
triethylamine (5 equiv), and the chiral acid (1 equiv) were added. The solution was
stirred at room temperature overnight. After removal of most of the solvent under
reduced pressure, the product was purified by silica gel flash chromatography. 1H NMR
(CDC13, 300 MHz) 5 9.71 (s, 1H), 8.78 (br, 1H), 8.37 (d, J=2.2 Hz, 1H), 8.21 (d, J=2.2

Hz, 1H), 7.28 (m, 2H), 6.98 (m, 3H), 4.82 (q, J=7.1 Hz, 1H), 1.64 (d, J=6.6 Hz, 3H).

CHM” 0 “”0 (Inf
. / a,

Carrier H 14H

 

Carrier H derivative 14H:

o-Phenyldiamine (54 mg, 0.5 mmole) and (S)-2-methylbutyric acid (51 mg, 0.5
mmol) were dissolved in 4 N HCl (2 mL), and the solution was refluxed for 8 h, until it
turned to dark red. It was then cooled to room temperature, and enough ammonium
hydroxide was added in order to get precipitate. The precipitate was filtered and washed
with water. It was then dried with a vacuum, however, the yield of this reaction was not
determined because of the very small amount of the product. 1H NMR (CDC13, 300
MHz) 5 7.60 (br, 1H), 7.26 (m, 4H), 3.04 (q, J=6.9 Hz, 1H), 1.86 (m, 2H), 1.46 (d, J=6.9

Hz, 3H), 0.96 (t, J=7.2 Hz, 3H).

147

 

 

 

P0

01

0)

O

2. Zn(OAc)2

   

Porphyrin tweezer 2’:

To a solution of porphyrin acid (90 mg, 0.14 mmol) and 1,3-propanediol (5 mg,
0.05 mmol) in anhydrous CHzCll (4 mL) was added EDC (27 mg, 0.14 mmol) and
DMAP (18 mg, 0.14 mmol). The reaction mixture was stirred at room temperature
overnight and Zn(OAc)3 (50 mg) was added. The reaction was stirred for another 2 h,
and it was then applied directly to silica gel column and purified (CH2C12) to afford the
title compound 2’ (48%). ]H NMR (CDC13, 300 MHz) 5 8.94-9.04 (m, 16H), 8.48 (d,
J=8.0 Hz, 4H), 8.17-8.23 (m, 16H), 7.46-7.54 (m, 18H), 4.71 (t, J=6.6 Hz, 4H), 2.43 (t,

J=6.6 Hz, 2H).

148

Reference:

1.

10.

11.

12.

13.

14.

15.

l6.

l7.

Harada, N.; Nakanishi, K. Circular Dichroic Spectroscopy: Exciton Coupling in
Orgainc Stereochemistry, ed.; University Science Books: Mill Valley, 1983.

Huang, X. F.; Rickman, B. H.; Borhan, B.; Berova, N.; Nakanishi, K. J. Am.
Chem. Soc. 1998, 120, 6185.

Huang, X.; Borhan, B.; Rickman, B. H.; Nakanishi, K.; Berova, N. Chem. Eur. J.
2000, 6, 216.

Tyrrell, E.; Tsang, M. W. H.; Skinner, G. A.; Fawcett, J. Tetrahedron 1996, 52,
9841.

Nagai, Y.; Kusumi, K. Tetrahedron Lett. 1995, 36, 1853.
Kusumi, K.; Yabuuchi, T.; Ooi, T. Chirality 1997, 9, 550.
Mizutani, J .; Tahara, S. Tetrahedron Lett. 1996, 37, 4737.

Ferreiro, M. J .; Latypov, S. K.; Quinoa, E.; Riguera, R. J. Org. Chem. 2000, 65,
2658.

Barker, R. Organic Chemistry of Biological Compounds; Prentice-Hall: New
Jersey, 1971.

Seco, J. M.; Quinoa, E.; Riguera, R. Tetrahedron: Asymmetry 2001, 12, 2915.

Berova, N.; Nakanishi, K. Circular Dichroism, Principles and Applications, 2nd
ed.; Wiley-VCH: New York, 2000.

Rickman, B. H.; Matile, S.; Nakanishi, K.; Berova, N. Tetrahedron 1998, 54,
5041.

Hor, K.; Gimple, O.; Schreier, P.; Humpf, H. -U. J. Org. Chem. 1998, 63, 322.
Nakaya, T.; Tomomoto, T.; Irnoto, M. Bull. Chem. Soc. Jpn. 1967, 40, 691.
Andersson, 1...; Kenne, L. Carbohydr. Res. 2003, 338, 85.

Barth, G.; Voelter, W.; Mosher, H. S.; Bunnenberg, 13.; Djerassi, C. J. Am. Chem.
Soc. 1970, 92, 875.

Proni, G.; Pescitelli, (3.; Huang, X.; Quraishi, N. Q.; Nakanishi, K.; Berova, N.
Chem. Commun. 2002, 1590.

149

18.

19.

20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

30.

31.

32.

33.

34.

35.

Proni, G.; Pescitelli, G.; Huang, X.; Nakanishi, K.; Berova, N. J. Am. Chem. Soc.
2003, 125, 12914.

Hall, J. H.; Gisler, M. J. Am. Chem. Soc. 1976, 41, 3769.
Fehrentz, J. A.; Castro, B. Synthesis 1983, 676.

Eliel, E. L.; Wilen, S. H. Stereochemistry of Organic Compounds; John Wiley &
Sons, Inc.: New York, 1994.

Sigel, H.; Martin, R. B. Chem. Rev. 1982, 82, 385.

Kurtan, T.; Nesnas, N.; Li, Y. Q.; Huang, X. F.; Nakanishi, K.; Berova, N. J. Am.
Chem. Soc. 2001, 123, 5962.

Jimenez, A. 1.; Cativiela, C.; Gomez-Catalan, J .; Perez, J. J .; Aubry, A.; Paris, M.;
Marraud, M. J. Am. Chem. Soc. 2000, 122, 5811.

Adams, H.; Harris, K. D. M.; Hembury, G. A.; Hunter, C. A.; Livingstone, D.;
McCabe, J. F. Chem. Commun. 1996, 2531.

Crisma, M.; Formaggio, F.; Valle, G.; Toniolo, C.; Saviano, M.; Iacovino, R.;
Zaccaro, L.; Benedetti, E. Biopolymers 1997, 42, 1.

Klotz, P.; Chatrenet, B.; Coppo, M.; Rousseau, B.; Goeldner, M.; Hirth, C. J.
Labeled Compd. Radiophann. 1991, XXIX, 149.

Kessler, P.; Chatrenet, B.; Goeldner, M.; Hirth, C. Synthesis 1990, 1065.

Latypov, S. K.; Galiullina, N. F.; Aganov, A. V.; Kataev, V. E.; Riguera, R.
Tetrahedron 2001, 57, 2231.

Froyen, P. Tetrahedron Lett. 1997, 38, 5359.
Lindsey, J. S.; Woodford, J. N. Inorg. Chem. 1995, 34, 1063.

O'Shea, D. F.; Miller, M. A.; Matsueda, H.; Lindsey, J. S. Inorg. Chem. 1996, 35,
7325.

Pecsok, R. L.; Shields, L. D.; Cairns, T.; Mcwilliam, I. B. Modern Methods of
Chemical Analysis, 2nd. ed.; John Wiley & Sons, Inc.: New York, 1976.

Hara, H.; van der Plas, H. C. J. Heterocyclic Chem. 1982, 19, 1285.

van der Haak, H. J. W.; van der Plas, H. C.; van Veldhuizen, B. J. Org. Chem.
1981, 46, 2134.

150

 

 

 

 

 

39.

36.

37.

38.

39.

Wang, H.; Partch, R.; Li, Y. J. Org. Chem. 1997, 62, 5222.
Weiss, J. J. Inclusion Phenom. Macrocyclic Chem. 2001, 40, l.

Mizutani, J .; Kurahasci, T.; Murakami, T.; Matsumi, N.; Ogoshi, H. J. Am. Chem.

Soc. 1997, 119, 8991.

Huang, X.; Rickman, B.; Borhan, B.; Berova, N.; Nakanishi, K. J. Am. Chem.

Soc. 1998, 120, 6185.

151

Chapter 4
The stereochemical determination of di,-chiral halocarboxylic acids
In chapter 3, we have shown in detail the absolute stereochemical determination
of a—chiral carboxylic acids by using different carriers in conjunction with porphyrin
tweezer. The extension of this methodology was applied to the stereochemical
determination of or—chiral halocarboxylic acids. a-Chiral halocarboxylic acids were first
derivatized with carrier A and carrier B. The syntheses of (R)-2-bromopropionic acid

derivatives, A-l-Br and B-l-Br, are shown in Scheme 4-1.

 

 

NHBOC O 1. BOP, EtaN, NH2
CH O)
' N2 NH H3C Br 2. TFA, SHZCIZ N’ NJ‘YH
A 1-Br A-1-Br
BOP, E13N, HBr CH
NH2 BY” CHZCIZ N ~H3
+ H0 o, 4+ O
HZN H30 Br 82% H2N O
B 1-Br B-1-Br

Scheme 4-1. The derivitization of (R)-2-bromopropionic acid l-Br to carrier A and B.
The carrier A derivative A-l-Br showed a small positive ECCD sign with he,“
427 nm (A8 +9) and 420 nm (A8 ~13), A +21 at room temperature. But since carrier A
derivatives were found unreliable for stereochemical assignment in our previous studies,

further use of this carrier was abandoned.

The carrier B derivative B-l-Br exhibited a very strong negative ECCD sign with
74111430 nm (A8 -236) and 421 nm (A8 +136), A -372 at room temperature, and the signal

became even stronger (A = ~643) when it was cooled to 0 OC. However, this sign was

opposite to the predicted sign from the proposed mnemonic in Section 3-2, shown in

152

Figure 4-1. According to A values,l bromo was considered as the medium group, and
methyl group as large, thus a positive ECCD was predicted.

H O H 6
Me 5 Me CXZD positive ECCD

Br
NHAr 3'

Figure 4-1. The contradiction of the observed ECCD sign vs. the predicted sign based
on the mnemonic.

4-1. Synthesis of a-chiral halocarboxylic acids and the ECCD data of carrier B

derivatives.

The deviation observed with (R)-2-bromopropionic acid derivative B-l-Br was
investigated with a series of (it-chiral halocarboxylic acids. In order to study the
behaviors of the or-chiral halocarboxylic acids derivatives in more detail, a series of or-
chiral halocarboxylic acids were synthesized from chiral amino acids and derivatized
with carrier B. The general procedures for the synthesis of or-chiral halocarboxylic acids

from amino acids are shown in Scheme 4-2.

 

 

 

HXNHz KBr, NaNO: HxBr
Fl COZH 2.5Nsto4 H cozH
Us) -10°C,3h (S)
H NH NaN02 H Cl
.~ 2 > .‘
RXCOZH 6 N HCI RXC02H
0
L18) 0 0’4“ <5)
/
(HF) N3
HXNHZ X > HXF
R COZH NaNO2 H 002H
L13) ”:3“ (8)

Scheme 4-2. The synthesis of a-chiral halocarboxylic acids.

153

The Ot-chiral halocarboxylic acids could be easily synthesized from optically pure
amino acids.2 Bromination of the L-(S)—amino acids with potassium bromide and sodium
nitrite in 2.5 N sulfuric acid afforded or-chiral bromocarboxylic acids. The chiral 0t-
chlorocarboxylic acids and or—fluorocarboxylic acids could also be synthesized in a
similar manner. or-Chloro chiral carboxylic acids were produced by adding sodium
nitrite to a vigorously stirred solution of L-(S)-amino acids in 2.5 N HCl, and or-fluoro
chiral carboxylic acids were synthesized by treating L-(S)—amino acids with sodium
nitrite and hydrogen fluoride-pyridine as the reaction solvent. These reactions proceeded
with retention of stereochemistry,2 and it was the result of double inversion at the chiral

center, as shown in Scheme 4-3.

 

 

 

+ G) G
HXNHz M HxNz __z 8 .H X HXX
R COZH R COZH \C‘,>=O R COZH
L13) (5)

Scheme 4-3. The mechanism of the (it-chiral halocarboxylic acids synthesis.

Under the acidic reaction conditions, sodium nitrite reacted with the amino group
to generate a diazo intermediate. The diazo moiety was a good leaving group, which was
subsequently replaced by the free carboxylic acid functionality to form a very unstable
lactone intermediate, which upon nucleophilic attack of the halide formed the chiral
halocarboxylic acids.

The reactions were completed in four hours, and the products were easily
separated from the reaction mixture by extraction with dichloromethane. The crude yield
of these chiral carboxylic acids varied from 42% to 67%. Without further purification,

these chiral acids were coupled with carrier B, 1,4-diaminobenzene. The amino acids

154

utilized for our studies were L-alanine, L-valine, L-leucine, L-isoleucine and L-
phenylalanine.

The enantiomeric ratio for each of the Ot-chiral halocarboxylic acids was
determined by derivatizing the chiral acids with chiral (S)-methylbenzylamine by BOP
coupling reaction. The reaction was stirred at room temperature overnight, and was
analyzed by GC-MS to determine the retention time of the two diastereomers. The
diastereomeric ratio was then determined by calculating GC peak areas. The

enantiomeric ratio of the a-chiral halocarboxylic acids is listed in table 4-1.

Table 4-1. The enantiomeric ratio for chiral a—halogen carboxylic acids.

 

Chiral acids X=Br X=Cl X=F
0
Ho’"\,/ 1 99% 99% 50%
X

 

O
HO’IK/kz 90% 90% 64%

HOJ‘YY 90% 90% >80%
x 3
Lb
HO ; 800/0 800/0 800/0

0
HOJ'YWH.5 90% 75% 56%
X

 

The synthesized Ot-chiral halocarboxylic acids were coupled to carrier B with
standard BOP coupling reaction, and all the reactions underwent smoothly. The carrier B
derivatized complexes were then added to zinc porphyrin tweezer for spectroscopic

evaluation.

155

Table 4-2. The CD data of carrier B derivatives of or-chiral halocarboxylic acids.

 

 

 

NH 0 BOP, Et3N, H x
O : Ho’u\ri"l CHZC'Z > 0'“)ka
H2N x 54-82% H2N O
Chiral acids X = Br X = Cl X = F
derivatives Ag A As A A8 A
L (H) (S) (8)
HO , 430 (-236) -372 429 (+25) +63 428 (+34) +65
X 1 421 (+136) 421 (-38) 419 (-31)
0
AA 429 (+103) 429 (+132) 428 (+114)
”0 ,5, 2(3) 421 (-91) +194 421 (-82) +214 420 (-98) +207

0 429 (+107) 429 (+24) 429 (+31)
HOW +182 +54 +61
x 3(8)

421 (-75) 421 (-30) 420 (-30)
O
W 429 (+94) 429 (+108) 429 (+96)
. +175 +232 +162
H00)? 4(3) 421 (-81) 421 (-124) 420 (-66)
JK» 429 (+47) 429 (+22) 428 (+18)
”0 i P“5(S) 421 (-43) +90 421 (-19) +41 419 (-22) +40

 

The CD data for all the chiral (it-halogen carboxylic acids derivatized with carrier
B, 1,4-diaminobenzene, are summarized in Table 4-2. The CD3 were taken by adding the
chiral conjugates to zinc porphyrin tweezer in methylcyclohexane at room temperature.
All the derivatives exhibited ECCD spectra that were consistent for all or-chiral
halocarboxylic acids, i.e., ECCD for (R)-acids were negative and ECCD for (S)-acids
were positive.

A chiral iodocarboxylic acid was also synthesized to see that if the ECCD sign for
the chiral iodo compounds would be consistent with the other halogen containing
carboxylic acids derivatives. The (Jr-chiral bromo-, chloro-, and fluoro- carboxylic acids
were synthesized from natural amino acids in a single step,2 in which the free amino

group was transformed to diazo intermediate, and then replaced by halogens. However,

156

since the iodo functionality was too reactive, the chiral Cit-iodocarboxylic acid could not
be synthesized by this simple method.

Alternatively, the chiral iodocarboxylic acids could be synthesized by the
Finkelstein3 reaction starting with the corresponding chiral bromo carboxylic acids. As
shown in Scheme 4-4, the (S)—2-iodopropionic acid 1-1 was synthesized by refluxing (R)-
2-bromopr0pionic acid l-Br in acetonitrile overnight with excess amount of potassium

iodide. In this process, the stereochemistry at the chiral center was inverted.

 

O
NH
0 KI,CH3CN L 2 BOP, Et3N HNJKi/
+ +*
'40)}: reflux HO 1 CHZCIZ (:I
840/0 NH
1-Br 1-1 2 NH2 B-1-I

Scheme 4-4. The attempted synthesis of carrier B derivatized iodocarboxylic acid B-l-I.

Without any further purification, the (S)-2-iodopropionic acid l-I was coupled
with carrier B, 1,4-diaminobenzene, by a standard BOP coupling. However, this routine
amidation reaction did not yield the desired product and led to many unidentified
compounds. The problem could be that the iodide was too reactive, and might be
displaced by nucleophilic attack with another molecule of 1,4-diaminobenzene.

In order to synthesize the carrier B derivatized chiral iodocarboxylic acid B-l-I,
the order of the reactions was changed. As illustrated in Scheme 4-5, the free amino
group in the carrier B was first protected with Boc, and then esterified under BOP
coupling condition with (R)-bromopropionic acid to form BocB-l-Br. The synthesis of
desired compound B-l-I was finished by replacing the bromo with iodo group by the

Finkelstein reaction,3 and followed by deprotection.

157

O O

 

 

HOJKK + BOP, Et3N B, KI,CH3CN i CH2012 i
Br NHB CH2012 fBflUX 9870

0° 83% NHBoc 90% NHBoc NH2

1-Br BocB-1-Br B-1-l

Scheme 4-5. The synthesis of carrier B derivatized (S)-2-iodopropionic acid B-l-I.

The complex B-l-I was added to zinc porphyrin tweezer in methylcyclohexane

and led to a positive ECCD sign at 0 °C (km 431 nm (A8 +21) and 417 nm (A8 -28), A

+48), which is consistent with all the other carrier B derivatized Ot-chiral halocarboxylic
acids. However, the iodo complex B-l-I was unstable and reproducible ECCD could not
be obtained if the iodo compound was stored for days.

As described above, carrier B derivatives of chiral a-halogen carboxylic acids
exhibit consistent ECCD spectra for all types of halogens tested thus far, with negative
ECCD for (R) and positive ECCD for (S) Chirality. However, the observed ECCD of
carrier B derivatives of chiral or-halocarboxylic acids are opposite to the sign predicted
based on the mnemonic proposed for carrier B (see Section 3—2).

Figure 4-2 illustrated the observed and predicted ECCD of carrier B derivatized
(S)-2-bromo-3-methylbutyric acid B-2-Br. According to the proposed mnemonic for
carrier B, the large i-Pr group was perpendicular to the carbonyl group, and the medium
bromo group was gauche to amide hydrogen. The counterclockwise orientation from

large to medium to small group would result in a negative ECCD couplet. However, the

chiral complex exhibited a strong positive ECCD couplet, with 71.6,“ 429 nm (A8 +103)

and 421nm(Ae -91), A +194.

158

C) 11 ii

that} :2. +Pr—Q 1:1) 6 ECCD

 

 

 

ArHN Br Br
'20 fig (8) B-2-Br
w .
A8 0
'60 ‘
-120
375 425 475

wavelength (nm)

Figure 4-2. The observed and predicted ECCD of B-2-Br.
4-2. Explanations for the unexpected ECCD of carrier B derivatized a-chiral
halocarboxylic acids

In order to explain the inconsistency between the observed ECCD of the carrier B
derivitized chiral a—halocarboxylic acids and the predicted signs based on the proposed
mnemonic (see Section 3-2), three hypotheses were proposed (Figure 4-3).

(a) Electronic factors overcome the steric control of the substituents in
determining the sign of the ECCD couplet, by causing the halogens to behave as the large
group and positioning themselves perpendicular to carbonyl;

(b) The halogen on the chiral center and not the carbonyl oxygen coordinates with
the zinc porphyrin tweezer, resulting in a new orientation between the two interacting
porphyrins;

(c) The carbonyl oxygen and the halogen on the chiral center cooperatively bind
to the zinc porphyrin tweezer in a bi-dentate fashion leading to a new paradigm for

binding porphyrins.

159

 

Figure 4-3. Possible conformations leading to the observed ECCD sign: (a) halogen
behaves as the large group and is perpendicular to carbonyl; (b) the halogen binds to zinc
porphyrin tweezer; (c) cooperative binding of halogen and the carbonyl oxygen to zinc
porphyrin tweezer.

4-2.1. Determining the rotation of C¢,.-Cc=o bond by NMR study

From the latter discussion, it was clear that the configuration of the Cchirai-Cc=o

bond could adopt various rotomers. The rotation around the Cchiral-Cc=o bond directly
dictated the observed ECCD sign, so it was critical to understand the rotation around the
Cchiral‘CC=O bond. We relied on NMR of carrier B derivatives and other amides to

investigate the single bond conformation.

The (R)-2—bromopropionic acid l-Br was derivatized with (R) and (S)-2-
phenylmethylamine 6. The preferred conformation of the amide bond has been
extensively studied and understood, placing the hydrogen on the chiral amine center syn

4-7

to the carbonyl oxygen due to Am strain (Figure 4-4). With the amide conformation

locked, we were interested in determining the preferred conformation around the Cchiral-

Cc=o bond using NMR. Depending on the favored conformation of the Cchiral-Cc=o bond,

the substituents on the chiral center would be placed either inside or outside the

160

anisotropic cone of the neighboring phenyl group, experiencing different shielding

effects.
H O H
0,0Me yr ,0Ph
Cchiral/1L1?l Cchiral 1)] Me
H H
SR GS

Figure 4-4. The conformation of the amide part in the derivatives.
Standard BOP coupling between the (R)-2-bromopropionic acid l-Br and chiral
amines (R) and (S)-2-phenylmethylamine 6 provided diastereomers 6R and 68, the NMR
spectra of which (in CDCl3) are shown in Figure 4-5.
M:_ra"~Nr'kb~Meb
H Ph
Bren (H, H)

Al -..J

 

H:~r3u‘N’l\'b’ 'Ph

 

 

 

H Meb
Bresms
4L 1 _ L
lllllIlllljlrrlllllllllllllllljllll1rlTFITllljlljlllllrllrlTrlllllllll

8.0 7.5 7.0 6.5 6.0 5.5 5.0 4.5 4.0 3.5 3.0 2.5 2.0 1.5
Figure 4-5. NMR of compounds 6R and 68.

In the spectra shown in Figure 4-5, Ha appears at ~4.4 ppm, and the methyl group
(Mea) resonated at ~l.8 ppm. Ha shifts downfield in the 68 (R, S) diastereomer, while
Mea has slightly upfield shifted. Same results were obtained both at room temperature

and 0 0C. These results suggested that, as shown in Figure 4-5, in the 6R (R, R)

diastereomer case, the methyl group on the carboxylic acid chiral center (Mea) is on the

161

same side of the phenyl group, while the corresponding hydrogen (Ha) is on the same side

as the methyl group of the amide chiral center (Meb).

Although the above study provided important indication for the preferred
conformation of the free 6R and 68 in solution, we were reluctant in using these results to
extrapolate conformational information for the complexed porphyrin tweezer system.

Upon binding to the zinc porphyrin tweezer, the Cchiral'CCzo bond conformation could

change, making the above analysis ambiguous. Therefore, these results were not
conclusive to be of help for the actual system.
4-2.2. The study with carrier B derivative of 2-trifluoromethylpropionic acid

As discussed before, we were interested in determining the cause of the unique
behavior of Ot-chiral halocarboxylic acids as compared to the other Ot-chiral carboxylic
acids we had already studied (see Section 3-2). One of the suggestions was that the
strong electronegativity of the halogens overcame the steric factor by forcing the halogen
substituents to behave as the large group.8’ 9

In order to address this issue, we were interested in synthesizing or-
tn'fluoromethyl containing chiral acids, and derivatize them with carrier B. These
carboxylic acids have a strong electron withdrawing group, -CF3, and their complexation
with porphyrin tweezers were expected to give us some insight into our hypotheses for
the a—chiral halocarboxylic acids derivatives. In particular, if the ECCD signs of the
carrier B derivatized Ot-trifluoromethyl chiral carboxylic acids were consistent with those
of chiral halocarboxylic acid derivatives, the electronegativity argument expressed before

could be valid. If not, we need to address this issue otherwise.

162

4 1. SOCI2
I—i + How/4R 2. n-BuLi r—\'
O

01”“ Mica

 

 

 

 

 

 

o O o 3 /
“ Si
1. LDA ’7‘
N
2. CF31 ,Eth or VB
7 8 0~ 9
H2804 \ L1
CF:3 reflux 3 CF
t—\ 3
Howl/BR O N R
O O 0

Scheme 4-6. The proposed synthesis of chiral 2-trifluoromethyl carboxylic acids.
The plan for synthesizing the trifluoromethyl containing chiral acids is shown in
Scheme 4-6. In order to introduce the trifluoromethyl group at the or position in a
stereoselective manner, a chiral auxiliary is necessary. After the carboxylic acid was
coupled with the Evans’ chiral auxiliary, a chiral imide enolate could be generated by
treatment with LDA. Trifluoromethylation was then accomplished by addition of excess
amount of iodotrifluoromethane 7 in the presence of triethylborane 8, which was to

0 The trifluoromethyl radical thus adds with a C(a)-si-face

accelerate this reaction.1
preference to the Li-chelated transition state. At the final step, the chiral auxiliary could
be removed by hydrolysis to provide the desired chiral 2-trifluoromethyl carboxylic
acids.4 Although the aforementioned synthetic scheme looks promising, several
problems were encountered when the chemistry was actually applied.

Both of the reagents used in the above scheme, trifluoromethyl iodide 7 and
triethylborane 8, are commercially available. However, due to their high expense, we
decided to synthesize them. The syntheses of those two compounds have been previously

reported and are not complicated.1 " '2' '3

163

Et
3EtBr + 3Mg + BF3-0Et2 ——2—9—> EtaB + 3FMgBr
a

Scheme 4-7. The synthesis of triethylborane 8.

As shown in Scheme 4-7, triethylborane 8 could be easily synthesized by reacting
trifluoroborane etherate with ethyl magnesium bromide, generated in situ.“’ '2 The
challenge lies in the separation and purification of the product from the reaction mixture,
mostly because triethylborane is pyrophoric. Upon exposure to air, even during
transferring between flasks, triethylborane caught fire, which made its purification by
distillation impossible. So finally, triethylborane was purchased in order to move on with

the rest of the synthesis.

DMF
reflux

 

CFgCOZNa + '2 CF3I

7
Scheme 4-8. The synthesis of trifluoromethyl iodide 7.

At the same time, trifluoromethyl iodide 7 could be synthesized by refluxing
iodine with sodium salt of trifluoroacetic acid in dry dimethylformamide, as illustrated in
Scheme 4-8.l3 Since the boiling point of the resulting trifluoromethyl iodide is about —40
°C, it could be distilled from the reaction mixture and collected by cooling the collecting
flask at —78 oC. Although the synthesis of the trifluoromethyl iodide 7 proceeded
smoothly, its storing and handling for further use was problematic. The rubber septum
used to cover the trifluoromethyl idodide container became very brittle and stiff when the
flask was kept in the —80 °C freezer. Moreover, since the trifluoromethyl iodide was a
gas at room temperature, it was very hard to determine the extract amount used in the

trifluoromethylation reaction.

164

CF3 01:3
+10% HOW
O O
9 10
Figure 4-6. The structures of 2-trifluoromethylpropionic acid 9 and 2-
trifluoromethylbutyric acid 10.
With the reagents 7 and 8 in hand, we were ready to synthesize 2-trifluoromethyl

chiral carboxylic acids. The first two targets were 2-trifluoromethylpropionic acid 9, and

2-trifluoromethylbutyric acid 10, the structures of which are shown in Figure 4-6.

 

 

O O O
1. n-BuLi, -78°C
W 2. L \—$_
Cl
11 85% 12 14
,?L O M CCOCI Et N i k
93 : 3
+
O NH HUM . z 0 N
13 n-BULI,THF
11 79% 15

Scheme 4-9. The amidation with Evans’ chiral auxiliary.

The first step for the synthesis of trifluoromethyl containing chiral acids involved
the coupling of the acids or acid derivatives, with the Evans’ chiral auxiliary (S)-4-
isopropyl-2-oxazolidinone 11. In our case propionyl chloride 12 and butyric acid 13
were used, as shown in Scheme 4-9, in order to form the chiral amides with
oxazolidinone 11. Propionyl chloride 12 could directly react with the Evans’ chiral
auxiliary 11 under strong basic conditions to generate the chiral amide 14 at low
temperature.4 On the other hand, butyric acid 13 had to be activated with pivaloyl
chloride to form a more reactive mixed anhydride first, which was then coupled with the

deprotonated chiral auxiliary 11 to form the chiral amide 15.'4

165

Although both of the reactions worked, column chromatography purification was
needed to acquire pure products 14 and 15. However, the identification of the product
eluted was problematic, since all the starting materials and products of the above
reactions were neither UV active or showed strong color in the presence of different dyes.
Therefore, GC-MS was used to determine the component in each of the fractions of the
column, instead of normal TLC analysis, making the process very tedious. Ultimately,

pure products were separated from the reaction mixture at decent yields (Scheme 4-9).

 

 

I—\ o 3
O N 2.CF3|,Et3B,-20 C O N
1r 1“ = 1r 1*
o o 20% o o
14 16
\— 1. LDA, -78 00 }—CF
m 2. CF lEt B -2o°c F“ 3
N 3' 3' o N
O O o O
15 17

Scheme 4-10. The introduction of trifluoromethyl group to the CL position.

The critical step for the synthesis of chiral trifluoromethyl carboxylic acids was
the introduction of the trifluoromethyl group to the Qt position of the amides 14 and 15 in
a chiral fashion, as shown in Scheme 4-10. The enolate formed by LDA, which was
generated in situ by treating diisopropylamine with n-BuLi at —78 °C, reacted with
trifluoromethyl iodide 7 and triethylborane 8 at —20 °C. As it had been mentioned before,
trifluoromethyl iodide 7 is a gas at room temperature, and it could not be added to the
reaction vessel by syringes, so the cold trifluoromethyl iodide 7 was poured into the
reaction mixture, which inevitably led to a brief contact with air. Since these reactions

were carried out at very small scales (1 mmol), they were extremely sensitive to moisture.

166

When two reactions in Scheme 4-10 were set up side by side, compound 16 was obtained
at roughly 20% yield, but only traces of amide 17 could be identified.

The separation and purification of compound 16 proved to be troublesome, due to
the very similar polarities of the starting amide 14 and the product 16. In addition, none
of the two compounds were neither UV active or show visible spot with any dyes. Once
again, GC-MS was used for the determination of the components of each of the column
fractions. Unfortunately, after several attempts the product, amide 16, could not be
separated from the starting amide 14 by column chromatography, since each of the
fractions was a combination of the two at similar ratios. Although pure product 16 could
not be isolated at this point, it was decided to proceed to the next step using the amide
mixture. The logic behind it was that after coupling with carrier B, 1,4-diaminobenzene,

the isolation of the desired product would be easier, since now it is UV active.

 

 

. NH
..4~ CF H202.LiOH 0,:3 2 833.3311. HN’KE/
CNN 3 THF/H20 1.10% 2 2 _ cr=3
1r 1* ’
o 00°C
NH2
16 NH2 18

Scheme 4-11. The synthesis of carrier B derivatized 2-trifluoromethylpropionic acid 18.

The coupling of carrier B with 2-trifluoromethylpropionic acid 9 is shown in
Scheme 4-11. The removal of the chiral auxiliary was achieved by hydrolyzing the
mixed amides l4 and 16 with lithium hydroxide and hydrogen peroxide at 0 °C, which
provided a mixture of the desired chiral carboxylic acid 18 and propionic acid.4 Finally
the mixed carboxylic acids were coupled with carrier B under standard BOP coupling
condition. The reaction provided a rather complicated mixture of UV active compounds,

which underwent separation by column chromatography to yield the pure conjugate 18,

167

as identified by MS. Since the starting materials of the above reactions were mixtures,
the yields of these reactions were not determined.

With the pure compound 18 in hand, we were ready to evaluate its binding to zinc
porphyrin tweezer under stande condition, methylcyclohexane at 0 °C, and compared

the result with those obtained from Ot—chiral carboxylic acid derivatives. The ECCD
spectrum of 18 binding to the tweezer revealed a weak negative couplet with A“, 423 nm

(A8 -13.3) and 414 nm (A8 +10), A —23.3.

 

O 0 Me
CF3 :1) positive ECCD F3C CD negative ECCD
Me NHAr 13 ArHN H 18
conformation of the conformation when
proposed mnemonic considering electronic effect
15 4|
As 0 ~
-15
375 425 475

wavelength (nm)

Figure 4-7. The two possible conformations for compound 18 and its ECCD spectrum.
As shown in Figure 4-7, according to A values,l trifluoromethyl group (CF3) is
larger than methyl group (CH3) in size, and thus the observed ECCD of this complex is
opposite to the predicted sign based on the proposed mnemonic for carrier B derivatives.

If the electronic effect of the trifluoromethyl group is considered, the conformation of the

CC=0'CChira] bond changes. Although CF3 is still the large group and perpendicular to the

168

amide carbonyl, the small group (H) is now staggered with respect to the amide proton,
and the medium group (CH3) points to the porphyrin plane. This new conformation
would lead to the observed negative ECCD spectrum. This experiment might suggest
that the electronic effect of the strong electron withdrawing group at the chiral center play
a role in the stereochemical determination of tat-chiral halocarboxylic acid.

However, the ECCD of carrier B derivatized 2-trifluoromethylpropionic acid 18
was not reproducible after the compound was stored on sodium carbonate (to keep the
amino from being protonated) for a day. This might be due to the rapid epimerization at
the chiral center, since the or hydrogen, which is also adjacent to the strong electron
withdrawing trifluoromethyl group, is very acidic and could epimerize in either acidic or
basic condition.

4-2.3. The possibility of o*-—1t interaction

As mentioned before, the strong electron withdrawing effect of halogen atoms
could also affect the ECCD signs of carrier B derivatized chiral halocarboxylic acids by
O’*-1[ interaction,15 which would cause the halogens to behave as the large group instead
of medium group by electronic effects.

When Heathcock and coworkers studied the addition of nucleophiles to chiral
aldehydes according to Felkin-Ahn model,16 they observed that “the simple steric effects
are at least as important as 0*-orbital energies in determining which was the ‘large’
group for the purpose of applying the Felkin model.” Since carbon-heteroatom bonds
have significant lower o*-—orbital energies than carbon-carbon bond, the major conformer

should have the heteroatom perpendicular to the carbonyl, acting as the large group.

169

According to their results, they determined the order of ligand preferences for the
perpendicular position: MeO> t-Bu> Ph> i-Pr> Et> Me> H.

When we first predicted the ECCD signs of carrier B derivatized 0t-
halocarboxylic acids based on the proposed mnemonic for carrier B derivatives, the large,
medium and small group on the chiral center were defined based on their A values.l
Halogens are smaller than alkyl groups but larger than hydrogen, so they are considered
as the medium group. But when the 6*—-1t interactions are considered, the halogens could
behave as the large group by electronic effect and could be perpendicular to the carbonyl
group. If this approach is valid, our previously proposed mnemonic could also be
successfully applied to the Ot-chiral halocarboxylic acid derivatives.

Although this o*—1t interaction theory could give an explanation for our results of
the or-chiral halocarboxylic acid derivatives, it could at the same time complicate some of
the previously discussed cases where heteroatoms, besides halogens, were involved. For
example, as discussed in the previous chapter, the derivatives of oxygen containing acids,
such as (S)-acetylmandelic acid derivative 9B and (S)-acetoxypropionic acid derivative
10B, did not place the heteroatom, oxygen, perpendicular to the carbonyl group.

A possible explanation for this different behavior could be the hydrogen bonding
between the amide hydrogen and the heteroatom oxygen in preference to the possible
0*—1t interaction. Because of the hydrogen bonding, the oxygen containing substituent is
locked in the gauche position relative to the CC=o-CC-N bond, instead of being

perpendicular to carbonyl group, as the 0*—1t interaction would predict (Figure 4-8).

170

so 08

X VS. L
L NHAr ArHN on
GM: interaction hydrogen bonding

Figure 4-8. The conformations dictated by oak—1t interaction and hydrogen bonding.

If the above hypothesis is true, when the specified hydrogen bonding is eliminated
from the system, the complex should adopt a conformation in order to put the oxygen
perpendicular to the carbonyl group in favor of the O*—1t interaction. In order to test this
statement, we decided to methylate the amide nitrogen and therefore remove the
hydrogen available for hydrogen bonding. (S)-2-phenoxypropionic acid was the substrate
of choice, since its carrier B derivative 11B exhibited the strongest ECCD observed (see

Section 3-2).

0 O

N ; ;
OPh $93. OPh
BOP, Et3N 98%

NHB°° C2532 NHBoc NH2
19 Boc-Me-11B Me—11B

Scheme 4-12. The synthesis of methylated carrier B derivative Me-l 1B.

The synthesis of the monomethylated carrier B derivatized (S)-2-
phenoxypropionic acid Me-11B is shown in Scheme 4—12. The N—methylated carrier 19
was coupled with the (S)-2-phenoxypropionic acid 11 under standard BOP coupling
condition, and subsequently Boc deprotection with trifluoroacetic acid yielded the N-
methylated compound Me-llB. The chiral complex Me-llB was then added to zinc

porphyrin tweezer in methylcyclohexane to measure its ECCD. To our satisfaction, the

171

methylated carrier B derivative Me-llB showed a positive ECCD spectrum, opposite to

that of the original carrier B derivative 11B (Figure 4-9).

C)p{ C) [4
H30 E H3C‘Q/ l::> negative ECCD
ArHNOPh OPh

carrier 8 derivative 118

H 0 H e
oph E QOPh :1) positive ECCD
H3C NMeAr H3C

modified carrier 8
derivative Me-1 1 B

 

200 ;
100
A8 0 . ._ —— 11B
—— Me—1 1 B
-100
-200
375 425 475

wavelength (nm)
Figure 4-9. The conformations and ECCD of 11B and Me-llB.

According to the above observations, a conformational change of the chiral
complexes is implied (Figure 4-9). In particular, for the original carrier B derivative 11B,
which contains an amide hydrogen, the oxygen on the chiral center is gauche to the Cc=o-
CON bond to form a hydrogen bond. This conformation predicts a negative ECCD
spectrum, which is consistent with the observation. In the case of the modified derivative
Me-llB, however, the amide hydrogen is eliminated, and according to G*-n interaction,
the oxygen now becomes the large group because of the electronic effects. This new
conformation dictates a positive ECCD spectrum, which is also consistent with the

observed ECCD sign.

172

In order to obtain additional evidence for the validity of the above theory, the
carrier B derivatized (R)-2-bromopropionic acid 1Br was chosen to undergo methylation.
In this case, since there was no chance for hydrogen bonding in either the original

derivative B-l-Br or the methylated derivative MeB-l-Br, we did not expect to see an

 

 

ECCD sign switch.
\ TFA O
NH \
Br 18' NJ‘H/ 0112012 ”5;
BOP, Et3N ©Br 98°/°
NHBoc 0H2012 46% NHBOC NH2 MeB-1-Br

Scheme 4-13. The synthesis of modified carrier B derivative MeB-l-Br.

(R)-2-brom0propionic acid lBr was coupled with the methylated carrier B as
shown in Scheme 4-13. However, the chiral complex MeB-l-Br did not show any
ECCD spectrum when it was added to zinc porphyrin tweezer in methylcyclohexane,
either at room temperature or at 0 °C.

According to the experiments performed so far, the o*—rt interaction could be a
possible explanation for the behavior of the carrier B derivatized or-chiral halocarboxylic
acids. However, although we had some evidence, we were still not able to provide a
strong proof for this theory.

4-2.4. The N MR study to determine the interaction between zinc porphyrin tweezer
and carrier B derivatized a-chiral halocarboxylic acids

During all the previous studies, we were trying to consider the electronic effect of
the halogens in the carrier B derivatized chiral halocarboxylic acid cases. As mentioned
before, the halogens could behave as the large group due to their strong electronegativity

or due to 6*—rt interactions as shown in Figure 4-3a.

173

Besides the electron effect of the halogens, there are two other possible
explanations for the observed ECCD of the carrier B derivatized Ot-chiral halocarboxylic
acids. Both cases involved binding between the chiral complex and the zinc porphyrin
tweezer. When the carrier B derivatized Ot-chiral halocarboxylic acids bind to zinc
porphyrin tweezer, it is certain that the free amino group of the conjugate would bind to
one of the zinc porphyrins of the tweezer. What we were not so sure about is, which
atom of the chiral complex would preferentially bind to the other porphyrin of the
tweezer. As mentioned before, it is possible that the halogen atom binds to the zinc
porphyrin, instead of the carbonyl oxygen (Figure 4-3b), or that there is a cooperative
binding of both halogen atom and carbonyl oxygen to zinc porphyrin (Figure 4-3c). In
order to investigate these two hypotheses (Figure 4-3b and c), we undertook a NMR
study of a series of different amides binding to zinc tetraphenylporphyrin (ZnTPP).

When a substrate binds to ZnTPP, because of the strong diamagnetic anisotropy
of the porphyrin ring, the protons on the substrate will shift upfield. The tighter the
substrate binds to ZnTPP, the closer it is to the porphyrin plane, and the larger upfield
shifts the protons will “sense”. Therefore, by comparing the change of the proton

chemical shift of various substrates, we could estimate how well they bind to ZnTPP.

Table 43. Change in proton chemical shift upon addition to ZnTPP.

 

8 C

echba echrga ©2:Br (jgbfa g bar 00:
1

 

d
20 2 24 25
a 0.019 0.093 0
b 0.084 0.177 0.023 0.004 0.014 0.029
c 0.068 0.104 0.008 0.017 0.028
d 0 0.036 0.013 0.015
e 0 0.024 0.010 0.013

 

174

The changes in proton chemical shift of these substrates upon binding to ZnTPP
are shown in Table 4-3. All the NMR studies were performed in CDC13 at room
temperature, and the ratio between the substrates and ZnTPP was roughly 1: 1.

All the substrates underwent an upfield shift upon addition to ZnTPP, which
indicated that all the substrates bound to ZnTPP to some extent. Among them, substrate
21, the or-bromo amide, showed the most significant upfield shift, which suggested the
tightest binding. Substrate 20, which lacked the Ot-bromo moiety, also showed binding to
ZnTPP, but not as strong as its 0t-bromo analogue, which suggests that the halogen atom
participates in the binding. On the other hand, when substrates 22 and 23, which bore a
bromine atom but lacked a carbonyl group were tested, they experienced very small shift
upon introduction to ZnTPP. This indicated that the bromine by itself did not bind to
ZnTPP. The next substrates tried were esters 24 and 25, one with a (it—bromo and the
other without. Based on the small change of the chemical shift of these two substrates,
we could conclude that they both did not bind to ZnTPP strongly, and the bromine atom
does not seem to participate in the binding of esters to ZnTPP.

Overall, the results presented in Table 4—3 suggested the cooperative binding of
the or-bromo and the carbonyl oxygen of amides to ZnTPP. Interestingly, the above
statement was not true in the case of esters.

In order to investigate whether the same cooperative binding occurs for all the Ot-
halo amides and that it is not specific for a—bromo amide, the a—chloro and a—fluoro
amides were also tested. Just as the a—bromo case, the NMRs were taken in CDC13 at

room temperature, with the ratio between the amide substrates and the ZnTPP being

175

about 1: l. The binding results, expressed as the change of the haloamides chemical shifts

upon addition to ZnTPP, are shown in Table 4-4.

Table 4-4. The change in chemical shifts of haloamides upon addition of ZnTPP.

 

:B
edc

Kl

1°

Kl

GI 4‘?

ll 0
cFI

 

'b a b a b a
21 26 27
a 0.093 0.080 0.092
b 0.177 0.162 . 0.156
c 0.104 0.095 0.104
d 0.036 0.022 0.021
9 0.024 0.017 0.016

 

According to the numbers obtained in Table 4-4, all the haloamides bound to
ZnTPP with the same relative affinity. The proposed cooperative binding in or-
bromoamides seems to exist for both a—fluoroamides and a—chloroamides.

The NMR studies listed in Table 4-3 and 4-4 suggests that the carrier B
derivatized Ot-chiral halocarboxylic acids bind to zinc porphyrin tweezer with both the
carbonyl oxygen and the halogen atom at the chiral center. In order to have a better
understanding of the rotation around the Cczo-Cchiral bond, a tertiary or-bromoamide 28

was synthesized, and the change of its chemical shift upon addition to ZnTPP was

followed by NMR (Figure 4-10).

A8 (0.049)
3'.
O 2 W/ A8 (0. 017)
A8 (0. 006)\
A8 (0.031)
A8 (0.006) 28 A8 (0.033)

Figure 4-10. The change of chemical shift of bromoamide 28 upon addition of ZnTPP.

176

Once more, NMR experiments were carried out in CDC13 at room temperature,
with the ratio between the amide substrate and ZnTPP being nearly 1:1. The fact that all

the protons of amide 28 shifted upfield suggested that the amide substrate bound to
ZnTPP. In order to understand the rotation around the CCzO'Cchiral bond, we needed to

determine the positions of the methyl group (S) and the ethyl group (L) relative to the
porphyrin plane. Therefore, we were interested in protons a, the protons of the methyl
group, and protons b, the protons on the methylene of the ethyl group. A comparison of
the change of the chemical shift between these two protons was a direct measurement of
the proximity of these two sets of protons to the porphyrin ring. The larger the change of
the chemical shift, the closer the protons were to the porphyrin plane. As shown in
Figure 4-10, protons b experienced a larger change of chemical shift as compared to
protons a, which suggested that the ethyl group was in closer proximity to the porphyrin

ring as compared to the methyl group.

 

Figure 4-11. The interaction of carrier B derivatized (It-chiral halocarboxylic acids with
zinc porphyrin tweezer.

According to the above experiments, the suggested conformation of the carrier B
derivatized (it-chiral halocarboxylic acids upon binding to zinc porphyrin tweezer is

shown in Figure 4-11: (1) There is a cooperative binding of the (it-halogens and carbonyl

177

oxygen to zinc porphyrin; (2) The CC=O"Cchiral bond adopts such a conformation that the
large group is gauche to the carbonyl and the small group is syn to Cc=o-CC-N bond.
4-3. The study of the interaction between the zinc porphyrin tweezer and carrier B
derivatized ct-chiral halocarboxylic acids by forming five-membered a-halolactams
Although, according to the results presented above we have acquired some NMR
evidence to show the interactions between the zinc porphyrin tweezer and carrier B
derivatized Ot-chiral halocarboxylic acids, we thought we should support our suggestion
of the novel cooperative binding through another approach. In particular, we were
interested in constraining the free rotation around the CCfl'CChjra] bond, so that, by

removing this variable, the experimental results would reflect directly the nature of the

cooperative binding.

Fl
.192
n . as
0 £92 VS' N
HZN H'R1 H N
2

Figure 4-12. The carrier B derivatives versus the five-membered chiral 0t-lactams.

In order to achieve that, we decided to utilize five-membered chiral a—lactams
(Figure 4-12). For these systems, there is no rotation around the CC=O'Cchiral bond, since
that bond is now frozen in a five-membered ring. Therefore, the ECCD sign will be
dictated only by the interaction of the chiral center with the zinc porphyrins.

As shown in Figure 4-13, for most chiral five-membered lactams, when they bind
to zinc porphyrin tweezer, the porphyrin plane should “slide away” from the large group
on the chiral center and cover the small group to minimize the steric interaction between

the chiral substituent and the porphyrin plane. However, when one of the substituents on

178

the chiral center is a halogen and the other one is a hydrogen, and if the stated
cooperative binding of the halogen and carbonyl oxygen to zinc porphyrin really exists,
the porphyrin should slide toward the halogen, even though it is the large group.
Therefore, if cooperative halogen binding exists, this specific a—halolactam and all the .
other five-membered (Jr-chiral lactams should induce the opposite Chirality upon binding

to zinc porphyrin tweezer, which will be exhibited as opposite ECCD sign

 

Figure 4-13. The comparison of interaction of a five-memered lactam and the or-
halolactam with zinc porphyrin tweezer.

4-3.1. The synthesis of chiral (Jr-substituted-y-butyrolactams

Mitsunobo

O .
Ba "33°19" Ar, 0H:> 2 .4 ArNHAIMe2
ArN .91 1:{> N’u),/\’ 03:;
H R2131
29
Figure 4-14. The synthesis of five-membered (it-chiral lactams from

a—chiral-y—butyrolactones.

The synthesis of five-membered a—chiral lactams 29 has been previously
reported,l7 starting with Ot-chiral—y—butyrolactones 31 (Figure 4-14). The five-
membered or-chiral lactams 29 could be synthesized by Mitsunobo reaction'8 from 4-

hydroxy-butyramide 30, which was synthesized by opening an or—chiral-y—butyrolactone

19

31 with aromatic amine and trimethylaluminum. During this process, the

179

stereochemistry of the chiral center is retained. Since we need to synthesize a library of
five-membered (It-chiral lactams, the real task lies in the synthesis of different
or—chiral—y—butyrolactones.

Our first synthetic target was chiral or—chloro—y—butyrolactone 32, which would
eventually afford the corresponding lactam. As shown in Scheme 4-14, the proposed
synthesis of 32 started with the reaction of chiral homoserine 33 with sodium nitrite, in
aqueous hydrochloric acid solution, which was a general method for converting amino
groups to halogens,2 with overall retention of stereochemistry. Treatment of the chiral

alcohol 34 with acid would then lead to the formation of lactone 32.

 

 

NH2
33 34

HOJOKKVOH NaNo2 / Hg HOjLiA’OH 1N HCI 0&0
Cl
32
Scheme 4-14. The synthesis of chiral a—chloro—v—butyrolactone 32.

However, the aforementioned synthesis did not proceed as expected. When L-
homoserine 33 was treated with sodium nitrite in the presence of hydrochloric acid,
besides the anticipated conversion of the amino group to chloro, the condition was acidic
enough for the ring closure to occur. This reaction could cause a serious problem, since,
although we obtained the desired butyrolactone 32, we could not be certain of whether or
not the stereochemistry of the chiral center was retained.

As stated before (see Section 4-1), in the process of converting amino groups to
halogens, the stereochemistry of the chiral center is retained because of the participation
of the adjacent carboxylic acid. However, in the above reaction, when direct conversion

of the homoserine 33 to butyrolactone 32 occurred, the acid functionality also reacted

with the free hydroxyl group to form the five-membered lactone ring, and this

180

complicated the stereochemical determination of the chiral center. If the lactone
formation was faster than the substitution reaction, the stereochemistry at the chiral
center would be reversed, while if the replacement of the diazo group by acid was faster
than the lactone formation, then the stereochemistry should be retained.

In order to avoid this problem, another approach was adopted, which involved the
closing of the five-membered lactone ring before converting the amino groups to
halogens under the standard sodium nitrite method. In this way, the stereochemistry of
the chiral center should be reversed. The commercially available (S)-homoserine lactone
hydrochloride 35 was used as the starting material. As shown in Scheme 4-15, the (S)-
homoserine lactone 35 was converted to a-chloro-y—butyrolactone 32 by treatment with
potassium nitrite and aqueous hydrochloric acid, and the stereochemistry of the chiral

center switched from (S) to (R).

KNOZ,

o
C{2mm HCl GNHC' Ogre:
68°/

32
Scheme 4-15. The synthesis of a-chloro-y-butyrolactone 32.

The next target was chiral a-phenoxy-y-butyrolactone 36, which could be
synthesized from a-chloro-y-butyrolactone 32 by replacing the chloro group with a
phenoxy group (Scheme 4—16). In order to eliminate the possible racemization at the
chiral center under strong basic condition, excess amount of phenol was pretreated with

sodium hydride, and then the a—chloro-y—butyrolactone 32 was added into the reaction.

 

o PhOH, 0
05,01 NaH THI; OfiOPh
32 55 /o 36

Scheme 4-16. The synthesis of a—phenoxy-y—butyrolactone 36.

181

With the two of chiral butyrolactones 32 and 36 in hand, we were ready to convert
them to the corresponding chiral lactams 37 and 38. Scheme 4-17 presents the

conversion of (R)-ct-phenoxy-y-butyrolactone 36 to its corresponding lactam 38.

PhD PhO

 

 

 

o
NH2 AIM93 HNJK/VOH i 1 TFA 2 9
01-12012 5P“ n-Bu3P/DIAD o N CH20l2 O N
OPh... =
(15’ 67% EN: THF, 78% E; 98%
NHB‘” NHBoc NHBOC NH2
39 40 38

Scheme 4-17. Converting butyrolactone to butyrolactam.

The Boc-protected carrier B was first reacted with trimethylaluminum in order to
obtain the corresponding dimethylaluminum amide, which was then treated with (R)-01-
phenoxy-y—butyrolactone 36 to form the amide 39. Cyclodehydration of amide 39 was
achieved by Mitsunobu reaction to yield the Boc-protected lactam 40. The Boc
protecting group on lactam 40 was then removed by stirring in TFA and CHzClz to yield
the desired chiral butyrolactam 38. The chiral a—chloro-y-butyrolactone 32 could be
converted to its corresponding lactam 37 in the same manner.

The pure butyrolactams 37 and 38 were then added to zinc porphyrin tweezer in
methylcyclohexane, and their ECCD were measured. The (R)-(x-chloro-y-butyrolactam
37 exhibited a positive ECCD curve at both room temperature and 0 oC, upon addition to
zinc porphyrin tweezer. Although the ECCD amplitude at 0 °C was larger than that at
room temperature, the shape of the peak was not as symmetric. On the other hand, the
(S)—a—phenoxy-y—butyrolactam 38 showed a negative ECCD spectrum at room

temperature, which disappeared when cooled to 0 °C. The enantiomeric ratios for these

182

two compounds before Boc deprotection were determined by chiral HPLC, to be ~80%
for (R)-a—chloro-'y-butyrolactam 37, but only 33% for (S)-(x~phenoxy-'y-butyrolactam 38.
Although 80% is an acceptable value, 33% ee for compound 38 is not. Since it
was derived from (R)-01-chloro-'y-butyrolactone 32 by replacing the chloro group with
phenoxy, racemization must have taken place during that process. The possibility of
racemization in the substitution step was foreseen, and in order to minimize it, less than
one equivalent of sodium phenoxide was used. However, the precaution was obviously
not enough. In order to improve the enantioselectivity of the substitution reaction, the
chiral cr-bromo-y-butyrolactone 39 was synthesized. Since bromide is a better leaving
group than chloride, the substitution reaction described above should work better. The
synthesis of (S)-0t-phenoxy-'y—butyrolactam 38 through the a—bromo-y—butyrolactone 41 is

shown in Scheme 4-18.

 

 

PhO
KN02 / KBr O PhOH, 02;)
O§NH§ 2H8r 2. 5 N H2804 OfiBr rNaH Ob’ophfi _. E:
74517057?0
41 NH2 38

Scheme 4-18. Second route for the synthesis of (S)-a-phenoxy-y-butyrolactam 38.
The (S)~01-bromo-‘y-butyrolactone 41 was synthesized in the standard way by
stirring chiral homoserine lactone hydrobromide 35 with sodium nitrite and potassium
bromide in diluted sulfuric acid solution. The product was extracted and used without
further purification. The bromide compound 41 was then combined with 1 eq. of sodium
phenoxide to generate (S)-0t—pheonxy-’y-butyrolactone 36, which was subsequently

converted to the (S)-0t-phenoxy-'y—butyrolactam 38 as described before. To our surprise,

183

the enantiomeric excess determined by chiral HPLC was only 7%, even worse than what
was previously obtained.

So far, the attempts to synthesize chiral a—phenoxy-y—butyrolactone 36 by
replacing halogens with sodium phenoxide to secure high enantiomeric ratio had failed.
Another approach could be the use of Mitsunobu reaction, which could then lead to an

enantiomerically pure lactam 38 (Scheme 4-19).

PhO
H2 1. AlMea, CH2Cl2 or)

N
OH
0 cm DIAD,PPh3 O OPh 2 N
0:5“ 0 o: 7’ + 2. 01/10, PPh3, THF
THF 55 /° NHBOC 3. TFA, CHZCIZ
36 N

42 H2

 

 

38

Scheme 4-19. Third route for the synthesis of (S)-a-phenoxy-y-butyrolactam 38.

The synthesis of chiral a—phenoxy-y—butyrolactone 36 by Mitsunobu reaction'8
started with (R)-01-hydroxy-‘y-butyrolactone 42 reacting with phenol in the presence of
DIAD and PPh3. The desired chiral (S)-0t-phenoxy-y—butyrolactone 36 was separated
(56% yield).20 The (S)-0t-phenoxy-y-butyrolactone 36 was then reacted with the Boc
protected carrier B in the presence of trimethylaluminum, followed by an another
Mitsunobu reaction to close the lactam ring, and deprotection with TFA and CH2C12
afforded the final (S)-a-phenoxy-y—butyrolactam 38.17 This time, a greater than 99% ee
was determined for lactam 38, before Boc deprotection. When lactam 38 was added to
the zinc porphyrin tweezer in methylcyclohexane at room temperature, a negative ECCD
spectrum was obtained.

By using or-hydroxy-y—butyrolactone 42 as the starting material, more oxygen
containing chiral butyrolactones could be synthesized, since the hydroxyl group could be

easily converted to other functionalities.

184

M60,

OI)

N
N

 

 

0 A920, CH3CN 0 __.
obpw Mel 4 050m
reflux 70%
44

42 H243
Scheme 4-20. The synthesis (R)-0t-methoxy-y—butyrolactam 43.
In particular, the next target was or-methoxy—y—butyrolactam 43, and its synthesitic
plan is depicted in Scheme 4-20.21 The corresponding (R)-0t-methoxy-y—butyrolactone 44
could be easily synthesized from (R)-0t-hydroxy-y-butyrolactone 42 by reacting with
methyl iodide under refluxing condition in the presence of silver oxide. In this process,
the stereochemistry of the chiral center should be retained.” The (R)-
methoxybutyrolactone 44 was then converted to its corresponding (R)-ct-methoxy-y-
butyrolactam 43 according to the protocol previously describe by using AlMe3 and then
Mitsunobu reaction conditions.'7
Since the chiral a—methoxy-y-butyrolactone 44 was easily synthesized from the
chiral a—hydroxy-y—butyrolactone 42 by using methyl iodide and silver oxide,21 it was
expected that more alkoxybutyrolactones could be synthesized in the same manner, and

the attempted syntheses are shown in Scheme 4-21.

0 A920, CH3CN .
035‘}. + ; no reaction
reflux
42

A920, CH3CN

O E l 0
ob reflux 0; I?
45

42
0 OH + EtOH DlAD,PPh3 0
.~ > OEt
01:“) THF 015
42 45

 

 

 

Scheme 4-21. The attempted synthesis of a-alkoxy-y-butyrolactones.

185

Cyclohexyl iodide reacted with or—hydroxy-y—butyrolactane 42 in the presence of
silver oxide. The reaction mixture was refluxed in acetonitrile for over 2 days; however,
there was no conversion to the desired product. This might be due to the fact that the
secondary iodide was too sterically hindered, and therefore a less sterically hindered
primary iodide, ethyl iodide, was tested.

The reaction condition with ethyl iodide was similar to that of methyl iodide,
however, the product separated from the reaction mixture was not the desired alkylated
lactone, but the (LB—unsaturated lactone 45 with the hydroxyl group being converted to
ethyl ether. The mechanism leading to this unexpected product is still unclear, however,
it is obvious that even ethyl iodide was not active enough to convert the or-hydroxy-y—
butyrolactone 42 to the corresponding ethoxy lactone by using the silver oxide method.
We also tried to synthesize the ethoxy lactone by Mitsunobu reaction of ethanol and 0t-
hydroxy-y-butyrolactone 42. Once more, the only product obtained was the same
(LB-unsaturated lactone 45. Mitsunobu reaction works well when an alcohol reacts with
a carboxylic acid, but not when two alcohols react with each other.22 In the reaction with
phenol (Scheme 4-19), since phenol is quite acidic, the reaction provided the desired
phenoxy lactone 36. However, since ethanol is not as acidic, the reaction failed to
provide the desired Mitsunobu product.

Up to this point, all attempts to obtain chiral a-ethoxy-y-butyrolactone had failed.
Since even ethyl iodide was not active enough to convert the a-hydroxy-y-butyrolactone
42 to ethoxy lactone by using silver oxide reaction conditions, more reactive iodides such

as allylbromide and benzylbromide were tried (Scheme 4-22).

186

 

0 A920, CH30N 0
035.4% + ”Br 4, qu
reflux 67%
46

 

42
o A o
920, CH3CN
ObAOH + BnBr = (25108”
reflux 51%
42 47

Scheme 4-22. The synthesis of chiral benzoxy and allyoxy lactones.

When allylbromide or benzylbromide reacted with (R)-0t-hydroxy-y-butyrolactone
42 under silver oxide conditions, both reactions yielded a mixture of products with the
desired alkoxy compound 46 and 47 being the major product with satisfactory yield.
However, for both these reactions, the 0t,B-unsaturated lactones were generated also.

In addition, two more chiral alkoxy, (S)-01—naphthyloxy-y—butyrolactone 48 and
(S)-0t-p-tert—butylphenoxy-y-butyrolactone 49 were synthesized by using Mitsunobu
reaction, as shown in Scheme 4-23. One thing to be noticed with these Mitsunobu

reactions is that the reaction did not work when n-Bu3P was used instead of PPh3. '

OH DIAD, n-BU3P

O
,OH + a- no reaction
0 THF

OH DIAD, PPha

o O
~~°H ——’ 5°
05 + THF 51% O
42 48
OH
0 DIAD, PPh3 o
..OH + ———- is
05 Q THF 58% 0 O
42

t'BU 49 t‘BU

 

Scheme 4-23. The synthesis of (S)-0t-naphthyloxy-y-butyrolactone 48 and (S)-0t-p-tert-
butylphenoxy-‘y—butyrolactone 49.

With these new chiral lactones in hand, we moved on to the syntheses of the

corresponding carrier B derivatized chiral butyrolactams, with the protocols described

187

before. The conversions were uneventful, and hydrogenation of (R)-0t-allyloxy-'y-
butyrolactam 50 with Pd/C yielded (R)-a-propoxy-y—butyrolactam 51 (Scheme 4-24).
Upon addition of these derivatized chiral lactams to zinc porphyrin tweezer in

methylcyclohexane at room temperature, the ECCD data were obtained and are shown in

Table 4—5.

f §

 

0,, 0,,
( 3 H2, Pd/C ( 5
O N O N
EtOAc
NH2 NH2
50 51

Scheme 4-24. The hydrogenation of lactam 50 to 51.

188

 

Table 4-5. The CD data of chiral butyrolactams.

 

 

 

R
o MHZ 1.A|Me3,CH2Cl2 01;)
off“ + 2. DIAD, PPhg, THF
NHBOC 3. TFA, CH2012
NH2
Exgiegied 1sti.(Ae) 2nd MAE) Amplitude
O
114mb“) 52(3) negative 429 (-32) 417 (+30) -62
O
ArHano 50(F?) negative 429 (-34) 417 (+19) -53
O
,1,me 51(6) negative 429 (-23) 417 (+27) -50
0
1141115303" 53(6) negative 431(-12) 416 (+12) -24
0 .
ArHNb“‘C| 370‘?) positive 426 (+34) 418 (-23) +57
0
ArHN 0"“ 38(5) positive 427 (-17) 418 (+25) -42
O
O . .
ArHNb’ posrtive 427 (-17) 418 (+34) -51
00 54(5)
0
O
ArHN ..
osrtlve 427 -14 417 14 -28
01.3.. p ( ) 1+ )
55(8)

 

As it is shown in Figure 4-15, when the four chiral (R)-0t-alkoxy-y—butyrolactams
50-53 bind to zinc porphyrin tweezer, the two substitutions, the alkoxy group and
hydrogen, on the chiral center are pointing toward the porphyrin plane. Since the alkoxy

groups are larger in size than hydrogen, the porphyrin slides away from the alkoxy group

189

and covers the small hydrogen group. The counterclockwise orientation of the two

porphyrin rings leads to a negative ECCD sign for all four of them.

E negative ECCD

 

Figure 4-15. The conformation of (R)-0t-alkoxy-y-butyrolactams 50-53 binding to zinc
porphyrin tweezer.

As for the (R)-0t-chloro—7-butyrolactam 37, although its stereochemistry is the
same as the four chiral alkoxybutyrolactams 50-53, the sign of its ECCD spectrum is
opposite to those of the chiral alkoxybutyrolactams, and this is due to the cooperative
binding of the chlorine and the carbonyl oxygen with the zinc porphyrin tweezer. As
shown in Figure 4-16, although chlorine is larger than hydrogen in size, because of the
cooperative binding, the bound porphyrin slides away from the small hydrogen to cover
the large chlorine. The clockwise-orientated porphyrin tweezer results in a positive

ECCD sign.

SdD/ :— positive ECCD

 

Figure 4-16. The conformation of (R)-a-chloro-'y-butyrolactam 37 binding to zinc

porphyrin tweezer.

190

For the last three chiral butyrolactams 38, 54 and 55, since there is no halogen
present in the molecules, there should be no cooperative binding and thus it should follow
the same trend as butyrolactam 50-53 by adopting the same conformation as shown in
Figure 4-15. However, they exhibited opposite ECCD sign as we expected. In order to
explain this contradiction, we turn to the possible tit—stacking interaction between the
aromatic group on the chiral center and the porphyrin ring. As shown in Figure 4-17, this
n—stacking interaction brings the porphyrin plane close to the aromatic substitution at the
chiral center, in spite of the fact that it is larger than hydrogen. The aryl ring in essence
creates a modified bi-dentate mode of binding, leading to a counterclockwise orientation

of the porphyrin tweezer that results in a negative ECCD spectrum.

E negative ECCD

 

Figure 4-17. The conformation of chiral Ot-alkoxy-y-butyrolactams 38, 54 and 55
binding to zinc porphyrin tweezer.
4-3.2. The synthesis of chiral a-alkyl-y-butyrolactams

The chiral butyrolactams synthesized contain either halogen, or alkoxy groups at
the chiral center. However, in order to complete our library, we needed to obtain a few

chiral Ot-alkyl-y-butyrolactams. For the synthesis of the chiral a—alkyl-y—butyrolactams, a

similar scheme as before was followed, by starting with chiral a—alkyl-y-butyrolactones.

191

0321

N

 

O O chULi O
ObnOH —> qbnOTs 03542 E

NH2
Scheme 4-25. The synthesis of chiral Ot-alkyl-y-butyrolactams using cuprates.

The first attempt to synthesize the chiral a-alkyl-y—butyrolactones, shown in
Scheme 4-25, was by placing a good leaving group, such as a tosylate or triflate, at the
chiral center, which was then replaced by a cuprate. Since, replacement of an Ot-tosylate
with different kinds of cuprates was reported,23 we decided to follow that protocol and

make chiral a-tosyl-y-butyrolactone 54 from chiral a—hydroxy-y—butyrolactone 42.

O T Cl P 'd' O 0%
Ob‘OH s . yri ine= Ob'on‘w 0 Cl
E120 36%
42 56 32

Scheme 4-26. The synthesis of (R)-0t-tosyl-y-butyrolactone 56.

 

As shown in Scheme 4—26, the first conditions tried in order to synthesize (R)-Ot-

tosyl-y-butyrolactone 56 was to react (R)-0t-hydroxy-y-butyrolactone 42 with TsCl, using
pyridine as the base in ether. The reaction was completed overnight, followed by diluted
HCl solution to remove the pyridine. However, after column chromatography, two
products were separated with close polarity and close yield (nearly 1:1 ratio). One of the
compounds was the desired tosylate 56, while the other one was the unexpected
chlorobutyrolactone 32. The generation of the unexpected chlorobutyrolactone 32 was
rationalized by the addition of diluted HCl in the quenching step. The hydrochloric acid

could replace the tosylate group in product 56 and resulted in the formation of the chloro

192

byproduct 32. Also, the chloride generated during the reaction of the alcohol with tosyl
chloride could have resulted in the formation of 32.

In an attempt to investigate if HCl was indeed the source of C1, the same reaction
was repeated, using diluted H2SO4 instead of HCl to neutralize the excess amount of
pyridine. However, as before, the same byproduct, chlorobutyrolactone 32, was formed
in the reaction with similar yields. Therefore, we knew that the neutralization step did
not lead to the formation of the chloro byproduct 32. Since HCl was not the source of C1,
the only other explanation could be that the chlorine came from the starting material
TsCl. Since TsCl was used in the reaction as the source of the tosylate group, the

generation of the chloro byproduct 32 was inevitable under this reaction condition.
0 T Cl E N O
05.1% S ' '3 2 oéuors
THF 83%
56

Scheme 4-27. Second route for the synthesis of (R)-0t-tosyl-y-butyrolactone 56.

 

Another reaction condition was tried to eliminate the generation of the unwanted
chlorobutyrolactone byproduct (Scheme 4-27). The new reaction utilized THF as the
solvent, and triethylamine as the base instead of pyridine. Under this new reaction
condition the only product generated was the desired tosylate lactone 56, without any

formation of the chloro byproduct 32.

MeZCUL' decompostion of
O 5670 ‘OTS——_’ staring material
E120, -10 00

Scheme 4-28. The first reaction of (R)-01-tosy1-'y-butyrolactone 56 with dimethylcuprate.

The (R)-0t-tosyl-y-butyrolactone 56 was ready to be converted to the chiral

alkylbutyrolactones by reacting with organocuprates. As shown in Scheme 4-28, the first

193

organocuprate used was lithium dimethylcuprate. Generally, the reactions with
organocuprates are kept at —78 °C to keep the organocuprates from decomposing.
However, the lithium dimethylcuprate is quite stable, and the reaction could be carried
out at relatively high temperature (-10 °C) to facilitate the replacement of the leaving
group.23 The lithium dimethylcuprate was made by adding methyllithium into cuprous
iodide at 0 0C in ether. The reaction initially showed a bright yellow color, and with the
addition of methyllithium, the yellow color gradually disappeared. After the generation
of lithium dimethylcuprate, the (R)—a-tosyl-y—butyrolactone 56 was dissolved in ether and
added to the cuprate solution at -20 °C in salt-ice bath. Unfortunately, the (R)-0t-tosyl-'y-
butyrolactone 56 does not dissolve in ether well, and when it was added into
dimethylcuprate, it became very sticky and precipitated out of the reaction mixture. The
reaction was kept at —10 °C, and at this temperature, the starting material (R)-0t-tosyl-'y-

butyrolactone 56 decomposed.

 

0 M92CuLi 0
.11OTS A’ Me
0:5 00
56 -40 °c

Scheme 4-29. The second reaction of compound 56 with dimethylcuprate.

The reaction was repeated, only this time, as shown in Scheme 4—29, the reaction
was kept at lower temperature to prevent the decomposition of 56. The solvent mixture
of EtZO and THF (1:1 ratio) was used to increase the solubility of the starting material.
All the starting tosylate 56 disappeared in 4 hours, and from GC-MS, we could observe
four major peaks, with one of them being possibly the desired a—methyl-y—butyrolactone.
However, the isolation of the desired product from the reaction mixture by column

chromatography was not successful. After the reaction mixture was loaded to silica

194

m— -— -_-_-_~_~-__.. .4- .4- _ .. _

 

 

column, it could not be recovered in any of the fraction. This might due to the low
boiling point of the product, which could result in the vaporization of the compound in
the process of removing solvents under reduced pressure.

In order to increase the boiling point of the product, we decided to use lithium
dibutylcuprate instead to react with the (R)-0t-tosyl-y—butyrolactone 56. The procedure
for this reaction is the same as dimethylcuprate, with the in situ generation of lithium
dibutylcuprate first, and then react it with (R)-0t-tosyl-y-butyrolactone. Since the
dibutylcuprate is not as stable as dimethylcuprate, the reaction temperature was kept at —
78 oC. Since, as mentioned before, the (R)-0t-tosyl-y—butyrolactone 56 was not very
soluble in ether, the reaction took place in a solvent mixture of ether and THF (Scheme 4-

30). The reaction mixture was kept for 4 h, and only starting material 56 was recovered.

O BuchI-i starting material
0‘5 \OTS E120 / THF recoverd
56 -78 °c

 

Scheme 4-30. The reaction of (R)-0t—tosyl-‘y—butyrolactone 56 with dibutylcuprate.

Attempts to synthesize chiral (x-alkyl-y-butyrolactones by reacting organocuprate
with tosylate 56 had failed. An alternative route for the synthesis of these compounds
could be the Reformatsky reaction. The general Reformatsky reaction calls for (it-bromo
compounds reacting with aldehydes in the presence of activated zinc. However, there
have been no examples of using alkyl halides instead of aldehydes in the Reformatsky
reaction. From literature search, a palladium catalyzed coupling reaction of a
Reformatsky reagent with allylic acetates to add the allylic group to the a—position of the

Reformatsky reagent was found,24 as shown in Scheme 4-31.

195

 

1 4
H‘YCOZHZ+ R3 0A0 1300313113)4 (:13 R + Fla/w”
”\r = choznz
ZnBr R4 R4 H‘ 002152
Scheme 4-31. A palladium catalyzed coupling reaction of Reformatsky reagent with
allylic acetates.

The same palladium catalyzed coupling reaction could be used to in our synthesis

to generate the chiral a—allyl-y—butyrolactones by reacting the chiral or-bromo-y-

butyrolactone 39 with allyl acetate, as shown in Scheme 4-32.

0 O 0
Oanr 29'6" oblan WON: M OW
THF

41
Scheme 4-32. The proposed synthesis of chiral a-allyl-y-butyrolactones by palladium
catalyzed Reformatsky reagent utilizing zinc on graphite.

One critical factor for the Reformatsky reaction is the preparation of highly
reactive zinc. There are several ways to activate zinc metal. One way is by the
potassium-graphite (CgK) reduction of the corresponding zinc halide (ZnClz)25 in ethereal
solvents according to equation 3-1.26’ 27

2C8K +ZnClz -> CmZn + 2KCI (3-1)

Potassium-graphite (CgK) was prepared by heating graphite powder under argon
to about 150 0C, and adding freshly cleaned potassium in pieces. The heating did not
stop until the potassium melts, which generated the pyrophoric potassium-graphite. It
was then covered with THF, and solid anhydrous ZnClz was added. The reaction was
stirred at 70 0C for half an hour, at which time the zinc-graphite was made. This was a
dangerous reaction to perform, since potassium is pyrophoric especially at high

temperatures. After the zinc-graphite was made, it was cooled to 0 °C, and the chiral Ct-

196

bromo-y-butyrolactone 41 was added dropwise to generate the Reformatsky reagent. The
solution was then warmed up to room temperature, and the palladium catalyst and allyl
acetate were added. The mixture was stirred for an hour before it was quenched by
aqueous acid. However, this reaction failed, and the starting a-bromo-y-butyrolactone 41
was recovered. One of the possible reasons for the failure could be the unsuccessful
activation of zinc by potassium-graphite.

In the mean time, a much milder way to prepare highly reactive zinc was found,
formed by the reduction of anhydrous zinc chloride with 2.1 equivalent of lithium and a
small amount of naphthalene in freshly distilled glyme under an argon atmosphere.28 In
this activation, lithium serves as the reducing agent,29 and naphthalene acts as an electron
carrier. The best solvent for the reduction is glyme, since the use of THF leads to a zinc
powder of much reduced reactivity while reduction does not proceed at all in ether. After
the mixture was stirred overnight, very fine black particles could be observed at the
bottom of the reaction flask, which indicated the generation of activated zinc powder.
The glyme was then removed by syringe, and THF was injected into the flask serving as

the solvent for the following palladium catalyzed coupling reaction.

0 O 0
.Br activated Zn .4ZnBr + WOAC Pd(PPh3)4
O O ——_—’ O
THF
41 57

 

Scheme 4-33. Palladium coupling reaction of Reformatsky reagent with allyl acetate.
The coupling reaction was carried out in the same way as described above
(Scheme 4-33), but this time, the separated compound from the reaction mixture was the

plain butyrolactone 57. The generation of butyrolactone 57 indicated the successful

197

 

iii. '*---—._--a— _—-~-

formation of the Reformatsky reagent, however, the palladium coupling reaction
provided the reduced product instead of the coupling product.

Since the synthesis of chiral or-alkyl-y-butyrolactones with palladium catalyzed
reaction failed, an alternative way was proposed, which involved the use of chiral

auxiliaries. The pr0posed synthesis is shown in Scheme 4-34.

 

 

O - O O
o SOC'Z A H ""8”“ Me JL JJ\/\/OP
+M 8‘N N' ‘N N
HOJK/Vop CIMOP H
58 59 Me 60 Ph Me Ph 61
n»BuLi,FfX

O
37);} __ Me NJLNWP
0 Me) 5 pi?

Scheme 4-34. The proposed synthesis of chiral a—alkyl-y—butyrolactones with chiral
auxiliary.

According to the proposed scheme, the protected 'y-hydroxy-carboxylic acid 58
was converted to the more reactive acyl chloride 59, which then reacted with chiral
auxiliary 6030 in the presence of n-BuLi. The R group (R=Me, allyl, and benzyl) could
be introduced in a chiral fashion to the or position by n—BuLi treatment and addition of the
corresponding alkyl halide. The standard deprotection of the hydroxyl group, and the

removal of the chiral auxiliary would result in the formation of the desired chiral a—alkyl-
y—butyrolactones.
Cl KOH O
reflux H0
65% 59

Scheme 4-35. The synthesis of benzyl protected 'y-hydroxy-carboxylic acid 58.

198

 

The benzyl protected y—hydroxy-carboxylic acid 58 was synthesized following
published procedures,3 l as shown in Scheme 4-35. Butyrolactone was refluxed in toluene
in the presence of benzyl chloride and potassium hydroxide, and in this way, the hydroxyl

group could be selectively protected in the presence of the carboxylic functionality.

 

+
OHJK/VOBn__—. reflux (25
58
O (COC12)2 / DMFS7 @Cl
HC3’u\/\’OBn hexane ((5)574-
58

Scheme 4-36. The attempted conversion of the acid 58 to acyl chloride.

As shown in Scheme 4-36, the benzyl protected y-hydroxy-carboxylic acid 58
could then be converted to the corresponding acyl chloride 59 by refluxing in $002.
However, the starting material 58 decomposed under the reaction condition, and benzyl
chloride and butyrolactone 57 were isolated from the reaction instead. Another way to
convert carboxylic acid 58 to the acyl chloride 59 was by using oxalyl chloride and

32 This reaction was carried out at both room

catalytic amount of DMF in hexane.
temperature and refluxing condition. However, in both cases, only benzyl chloride and

butyrolactone 57 were generated, instead of the desired acid chloride 59.

 

FL 0 AIM
. e
Me‘N NH 4. 035 3 no reaction

H

Me 60 Ph 57

Scheme 4-37. The attempted reaction between chiral auxiliary 60 and butyrolactone 57.
Since the conversion of benzyl protected y—hydroxybutyic acid 58 to acyl chloride
59 failed, we thought we might be able to couple it to the chiral auxiliary 60 directly.

Since butyrolactones can be opened with amines in the presence of AlMe3 (Scheme 4—

199

16),19 we tried to react the chiral auxiliary 60 and butyrolactone directly, as shown in
Scheme 4-37. However, this reaction did not work, which might be due to the fact that

the anion generated on the chiral auxiliary 60 was too stable to react with the

 

butyrolactone.
o O O
Me.NJLN.H 0 Pivaloyl chloride M8~NJLN/U\/\/0H
: I ,4 HOJ|\/\,orain 4
Me Ph EI3N, Toluene Me Ph
60 58 61

Scheme 4-38. The coupling of chiral auxiliary 60 with acid 58 by generating mixed
anhydride.

As shown in Scheme 4-38, another way to couple the chiral auxiliary 60 with the
benzyl protected y—hydroxybutyic acid 58 was by converting the carboxylic acid to a
more reactive mixed anhydride by treating it with pivaloyl chloride.33 This reaction was
refluxed in toluene, and the desired conjugate 61 was formed along with some
byproducts. This reaction, although promising, had to be optimized, and at the same
time, we came across a published procedure on the synthesis of chiral or-alkyl-y-
butyrolactones, which we decided to follow.

After many unsuccessful attempts, the chiral a—alkyl-y—butyrolactones were
finally synthesized by following A. I. Meyers’ method, which utilizes chiral 2-
methyloxazoline as the chiral auxiliary.34 The chiral 2-methyloxazoline 62 is
commercially available, and could be used to prepare either enantiomer of the or-alkyl-y-
butyrolactones as shown in Scheme 4-39. For our purpose, we follow the procedures to

synthesize chiral lactones with (S) configuration.

200

Ph
”Bu”, (3) Me-(\ ”T1 LDA

 

 

 

 

 

 

MegsiCl 52 OMe RX
0 Ph
TMSO’W NT? RANT,”
II 63
OMe OMe
LDA, RX LDA
|/\,OTMS
64
TMSO\_¢ 0 Ph R, (,an
HTN]; H KN ""1
OMe TMSO OMe
(a 65
H H@
o o
H. a: H
H\‘ O R‘ O
(H) (8)

Scheme 4-39. The synthesis of chiral butyrolactones with chiral 2-methyloxazoline 62.

According to the published scheme,34 alkylation of chiral 2-methyloxazoline 62
forms the 2-alkyloxazoline 63. Ethylene oxide could not be introduced to 2-
alkyloxazoline 63 after LDA deprotonation due to its poor reactivity at —78 0C to —98 °C,
and therefore, a synthetic equivalent, 2-(trimethylsilyloxy)ethyl iodide 64 was used
instead, affording compound 65. The subsequent acidic hydrolysis produces the (S)
lactones.

Scheme 4-40 rationalizes the absolute configuration of chiral butyrolactones.34
The deprotonation of the 2-alkyloxazoline 63 with LDA results in the generation of two

enolates with a ratio of 95:5, favoring the Z enolate. These two enolates do not

equilibrate under the reaction condition, and are subjected to bottom-side alkylation

201

A,

leading to compound 65 as the major product. Upon acidic hydrolysis, the chiral lactones
with (S) configuration are produced. However, a small amount of chiral lactones with the
opposite configuration could be generated in the same process. Bottom-side alkylation of
the minor enolate and some (~10%) top-side alkylation of the major enolate provide 15-
20% of the opposite enantiomer.
O Ph 0 Ph 0 Ph
R “N11 LDA 39:09} + :HNI'oI

- 0 l
OMe 78 C Li‘OMe Li *OMe

(95%) (6%)
I RX

Ph
R O
,__ j
061
R Li‘-OMe

 

63

l -98 °c

O

H . 0 H6140];

R“ R. N
(S) 65 OMe

 

Scheme 4-40. Rationale for the absolute configuration of chiral butyrolactones.

The syntheses of (S)-2-methyl-'y-butyrolactone 66 and (S)-allyl-y-butyrolactone 67
were achieved following the published procedure, with one modification. In stead of
using 2-(trimethylsilyloxy)ethyl iodide 63 as the alkylation reagent, 2-(OTBS)ethyl
iodide 68 was used, because of the unsuccessful synthesis of 2-(trimethylsilyloxy)ethyl

iodide 64.

Ph _ Toluene
Phx '1’ EtzNSlMe3 + 2M61

80-90°C

69 64

 

l/\,OTMS

Scheme 4-41. The published synthesis of 2-(trimethylsilyloxy)ethyl iodide 64.

202

 

We have tried to synthesize 2-(trimethylsilyloxy)ethyl iodide 64 according to the
published synthesis shown in Scheme 4-41.3S' 36 The 1:2 mixture of EtzNSiMe3 and Mel
could act as synthetic equivalent of Me3SiI,37 and they could be handled without special
care. According to the reference, the reaction of benzophenone ethylene acetal 69 with 1
equivalent of EtzNSiMe3 and 2 eq. of Mel could afford the desired 2-

(trimethylsilyloxy)ethyl iodide 64 as the major product.

0 p—TsOH 10 mol% Ph 0

 

 

/—\ : unreacted
HO OH + phiph CH20'2» Phxo:I + benzolphenone 70
70 sealed tube 69
T H1 l°o
/__\ + O p- 80 0mo/; PhXO] + unreacted
70 Dean-Stark 59

Scheme 4-42. The reactions for the synthesis of benzophenone ethylene acetal 69.

The benzophenone ethylene acetal 69 could be synthesized by reacting
benzophenone 70 with ethylene glycol. However, since the benzophenone 70 is sterically
crowded, the synthesis of benzophenone ethylene acetal 69 was not very efficient.
Benzophenone was condensed with ethylene glycol, using TsOH as the catalyst. Two
different reaction conditions were tried (Scheme 4—42), either using CH2C12 as the solvent
and heating in a sealed tube, or using Dean-Starks trap with benzene as the solvent. Both
the reactions worked, however, did not go to completion, and the starting material 70
could not be separated from the product 69.

Since the benzophenone ethylene acetal 69 could not be synthesized efficiently,
and 2-(trimethylsilyloxy)ethyl iodide 64 is too reactive to survive the column if we

synthesize it from ethylene glycol by iodonation followed by silylation, we turned to a

203

 

 

more stable synthetic equivalent, the 2—(OTBS)ethyl iodide 68. The conventional

synthesis of the 2-(OTBS)ethyl iodide 68 from ethylene glycol is shown in Scheme 4-43.

 

 

 

N H, TBSCI TsCl, Et N Nal/CchN
HO’\’OH a HONOTBS 3 TsO’VOTBS fiA/OTBS
THF CH2C12 reflux

87% 71 84% 72 96% 68

Scheme 4-43. The synthesis of 2-(OTBS)ethyl iodide 68.

The synthesis of 2-(OTBS)ethyl iodide 68 started with ethylene glycol.
Monosilylation of the ethylene glycol to compound 71 could be achieved by addition of
NaH and TBSCl in THF. The free hydroxyl group was then converted to tosylate 72,
which was a good leaving group, and was replaced by iodide, according to the Finkelstein
reaction,3 to yield the desired 2-(OTBS)ethyl iodide 68.

He
LDA Ph

0

4 4 13‘\H1 (\N E H. &

Me—( (1:13,? Allybromide WM l/\,OTBs; O
f—

67

52 OMe -78 0C 890/0 73 OMe OT7BSOMe

-98 °C 82%

 

 

Scheme 4-44. The synthesis of (S)-2-allyl-y-butyrolactone 67.

With the 2-(OTBS)ethyl iodide 68 in hand, (S)-2-methyl-'y—butyrolactone 66 and
(S)-allyl-y—butyrolactone 67 were synthesized in the same manner; the synthesis of (S)-2-
allyl-y—butyrolactone 67 is shown in Scheme 4-44. The alkylation of 2-alkyloxazoline 62
with allylbromide provided compound 73 with 89% yield. The following alkylation of
compound 73 with 2-(OTBS)ethyl iodide 68 was the critical step. It is important to keep
the alkylation below —98 °C, since when the reaction was tried at —78 °C, product 74 was
not generated but instead starting material 73 was recovered. This could be due to E2
elimination of iodide 68 at warmer temperature. The alkylation reaction between

compound 73 and 68 finished in 2 h at —98 °C. This was a clean reaction, and the pure

204

 

 

alkylation product 74 could be easily separated by column chromatography. The
alkylation product 74 was then hydrolyzed under acidic conditions to form the (S)-2-
allyl-'y-butyrolactone 67. The hydrolysis was carried out at refluxing 4.5 N HCl for only
15 minutes to avoid possible racemization.

However, the synthesized (S)-2-allyl-‘y-butyrolactone 67 was not very pure by
NMR. This was unexpected, since the starting material 74 was very clean, and the
hydrolysis should not provide any other product than the desired butyrolactone 67.
However, despite the fact that the (S)-2-allyl-y-butyrolactone 67 was not very pure, it was
still used as starting material for the synthesis of (S)-2-allyl-y-butyrolactam 75. It was

thought that perhaps the impurity would be easier to remove in a later step.

\
O CH2012 HN 5 0H1.n-Bu3P/DIAD o N

H ' E
//_.>(~- 0, + 59% E: _\\ 2. TFA/CH20|2 E:
67 N

 

 

74%
NHBOC NHBoc H
2

76 75
Scheme 4-45. The synthesis of (S)-2-allyl-y—butyrolactarn 75.

Scheme 4-45 shows the synthesis of (S)-2-allyl-y—butyrolactam 75 from the chiral
lactone 67 . The (S)-2-allyl-y-butyrolactone 67 reacted with Boc-protected carrier B in the
presence of trimethylaluminum, and the cyclodehydration of the resulting amide 76 by
Mitsunobu reaction formed the chiral lactam 75. Although the desired lactam 75 was the
major product, the pure compound could not be separated from the reaction mixture.
There were several other compounds with very close polarity as the desired lactam, and

the desired product 75 could not be separated from the byproducts by column

205

chromatography. Although it was not surprising that other products were present, since
the starting material was not pure, we did not anticipate the separation problem.

In order to synthesize (S)-2-allyl-y-butyrolactone 67, we closely followed the
published procedure, however, with one small difference. In the published protocol, the
alkylation product 74 of 2-(OTBS)ethyl iodide 68 reacting with 2-allyl-oxazoline 73 was
hydrolyzed without any purification, while in our synthesis, it was purified by column
chromatography before undergoing acidic hydrolysis. Perhaps, the alkylation product
decomposed easily after subjecting the silicon column, which could yield all the
byproducts we could not separate.

The reactions for the synthesis of (S)-2-allyl-y—butyrolactone 67 were repeated as
shown in Scheme 4-44. This time, the alkylation product 74 was hydrolyzed under acidic
condition as soon as it was extracted from the reaction mixture without any further

purification. After the hydrolysis reaction, the desired chiral butyrolactone 67 could be
easily purified by column chromatography. After the pure (S)-2—allyl-y-butyrolactone 67
was secured, it was subsequently converted into the corresponding (S)-2-allyl-y-
butyrolactam 75 as shown in Scheme 4-45.

The absolute stereochemistry of the (S)-2-allyl-y—butyrolactone 67 was proved by
taking the optical rotation in ethanol.38 Since the conversion of the chiral lactone to
chiral lactam did not affect the stereochemistry of the chiral center,17 the final 2-allyl-y—

butyrolactam 75 should also have the (S) configuration.

206

’ _

 

 

Scheme 4-46. The synthesis of (S)-2-propyl-'y-butyrolactam 77.

The (S)-2-allyl-y—butyrolactam 75 could be easily transformed to (S)-2-propyl-'y-

butyrolactam 77 by hydrogenation with Pd/C and hydrogen gas. This reaction finished

 

overnight, yielding compound 77 as the sole product, not requiring further purification.
Another chiral 2-alkyllactam, (S)-2-methyl-7-butyrolactam 78, was also synthesized in
the same manner, through (S)-2-methyl-y-butyrolactone 65, by using Mel as the
alkylation reagent, instead of allylbromide. These chiral 2-alkyllactams were added to
zinc porphyrin tweezer solution at room temperature, and the ECCD data were recorded
and shown in Table 4-6.

Table 4-6. The ECCD data of (S)-2-alkyl-y-butyrolactams

H
H
2 1. AlMeg, 0H2c12 01—)

N
O A N
0: 7R + 2. DIAD, PPh3, TH
N 3. TFA, CH2C|2
N

HBoc
H2

 

 

1stA(A8) 2nd MAe) Amplitude

 

 

ArHNO “4978(5) 428 (-47) 419 (+36) -83
O
ArHN /75(S) 427(-41) 418(+38) -79
O
ArHN 77(8) 428 (-24) 420 (+37) -61
207

As listed in Table 4-6, the ECCD signs for the chiral 2-alkyl-y-butyrolactams 75,
77 and 78 were very consistent, with (S) configuration showing negative ECCD.
However, this sign was opposite to what we expect for the chiral butyrolactams, as shown
in Figure 4-18. At this point, we cannot explain this contradiction. We believe that the
assigned stereochemistry of the synthesized compounds is correct, since their synthesis
was based on well established protocols. However, in an attempt to eliminate the
possibility that the above result is due to the wrong stereochemistry assignment, we are
currently involved in trying to verify the absolute stereochemistry of the chiral center by
synthesizing these chiral substrates following an independent synthetic route. If the
results of the second independently synthesized compounds agree with what previously

observed, we will have to consider other possibilities for the unexpected observations.

__ observed
E POSlt'Ve ECCD iii) negative ECCD

 

Figure 4-18. The observed ECCD sign is opposite to the predicted sign based on model.
Conclusion:

The absolute stereochemistry of (Jr-chiral halocarboxylic acids can be determined
by derivatizing them with carrier B, 1,4-diaminobenzene. Upon addition to porphyrin
tweezer, (R) configuration acids show negative ECCD and (S) configuration exhibits
positive ECCD. The observed ECCD is opposite to the predicted sign based on the

proposed mnemonic in Section 3-2, which we propose is due to the cooperative binding

208

h

 

of the halogen and carbonyl oxygen to zinc porphyrin. This statement has been
supported by UV-vis, NMR and IR studies.

We have also tried to prove the above novel binding by synthesizing (it-chiral
butyrolactams, which can be easily obtained from (Jr—chiral butyrolactones. We compared
the sign of the or-chloro butyrolactam with that of the other lactams, and the results match
our hypothesis. However, when the substituent on the chiral center is an alkyl group, the
observed ECCD sign is opposite to our expectation. We cannot find a plausible

explanation for this observation.

209

 

 

Experimental materials and general procedures:

Anhydrous CHzClz was dried over CaHz and distilled. The solvents used for CD
measurements were purchased from Aldrich and were spectra grade. All reactions were
performed in dried glassware under nitrogen. Column chromatography was performed
using SiliCycle silica gel (230-400 mesh). 1H-NMR spectra were obtained on Varian
Inova 300 MHz or 500 MHz instrument and are reported in parts per million (ppm)
relative to the solvent resonances (8), with coupling constants (J) in Hertz (Hz). IR
studies were performed on a Galaxy series FTIR 3000 instrument (Matteson). UV/Vis
spectra were recorded on a Perkin-Elmer Lambda 40 spectrophotometer, and is reported
as Amax [nm]. CD spectra were recorded on a JASCO J -8 10 spectropolarimeter, equipped
with a temperature controller (Neslab 1 l 1) for low temperature studies, and is reported as
3. [nm] (At:max [1 mol'1 cm"]). Porphyrin tweezer 2 was synthesized as previously
described.39

General procedure for CD measurement:

All the CD spectra were taken with 1 11M porphyrin tweezer in
methylcyclohexane (1 mL) with 40 equiv of the carrier/chiral acid conjugate. The CD

spectra were measured in millidegrees and normalized to ASIA [nm] units.

NHBOC O 1. BCOHP’CEtaN, NH2
\ )SrH 2 2 E \ O
I + H0 , |
, H3C Br 2. TFA, CHZCIQ N’ ”711

 

N

A 1-Br A-1-Br
Carrier A conjugated acid A-l-Br:
To a solution of carrier A (1 equiv) in CH2C12 (2 mL), BOP (1.1 equiv),

triethylamine (5 equiv), and the chiral acid (1 equiv) were added. The solution was

210

stirred at room temperature overnight. After removal of most of the solvent under
reduced pressure, the product was purified by silica gel flash chromatography to yield the
Boc-protected conjugate. The Boc group was removed by stirring overnight in
TFA/CH2C12 (1:4) at room temperature. The solvent was then removed by purging
nitrogen. The residue was redissolved in anhydrous CHzClz, and anhydrous Na2CO3 (20
mg) was added. This solution was stirred at room temperature for an hour, and the
N32CO3 was filtered. The solvent was evaporated under reduced pressure to provide
analytically pure product.

Boc-A-l-Br:

1H NMR (300 MHz, CDC13): 5 8.66 (br, 1H), 8.16 (m, 2H), 7.24 (m, 3H), 6.99 (m, 3H),
4.95 (br, 1H), 4.79 (q, J=6.9 Hz, 1H), 4.49 (d, J=6.6 Hz, 2H), 1.88 (d, J=6.9 Hz, 3H),
1.45 (s, 9H).

A-l-Br:

1H NMR (300 MHz, CDCl3): 5 8. 82 (br, 1H), 8.2 (m, 2H), 7.28 (m, 3H), 6.99 (m, 3H),
4.68 (q, J=6.9 Hz, 1H), 3.91 (s, 2H), 1.92 (d, J=6.9 Hz, 3H).

H _,NH2 KBr, NaNOZA HxBr

RXCOZH 2.5 N stof H cozH

-10 °C. 3 h 28r-58r
45-57%

 

The general procedure to synthesize chiral a-bromocarboxylic acids:

Sodium nitrite (0.22 g, 3.24 mmol) was added to a solution of amino acid (1
mmol) and potassium bromide (0.78 g, 6.56 mmol) in 2.5 N sulfuric acid (9 mL) at —10
°C, and the reaction mixture was stirred for 3 h, and then extracted with CHzClz (3 x 25
mL). The combined organic layers were dried over sodium sulfate and evaporated. The

Crude acid (45-57%) was ready to couple with the carrier B without further purification.

211

H.

 

NaNo2
H ,NH2 4 H ..CI

7

R COzH 6NHCl,0°C, RXCOzH
4 h 52-67% 201-501

The general procedure to synthesize chiral a—chlorocarboxylic acids:

To a vigorously stirred solution of amino acid (1 mmol) in 6 N HCl (3 mL) at 0
°C was added freshly pulverized sodium nitrite in small portions. This mixture was
stirred for 4 h at 0 °C and then extracted with CH2C12 (3 x 25 mL). The combined
extracts were dried over sodium sulfate and solvent was removed under reduced pressure.

The crude acid (52%-67%) was coupled with carrier B without further purification.

(HF)xN:\>

 

H ,NH2 : HXF
R COQH NaN02, rt, 3 h R C02H
(BB-55% 1F-5F

The general procedure to synthesize chiral a—fluorocarboxylic acids:

Sodium nitrite (0.26 g, 3.77 mmol) was added to a solution of amino acid (1.1
mmol) in hydrogen fluoride-pyridine (2 mL) at —10°C, and the reaction mixture was
stirred for 3 h at room temperature, at which time the reaction was quenched with water
(10 mL). The resulting solution was adjusted to pH 4.0 with solid sodium bicarbonate.
The mixture was extracted with CHzClz (3 x 25 mL). The combined organic layers were
dried over sodium sulfate and solvent was removed under reduced pressure. The crude

acid was ready to couple with the carrier B without further purification.

212

 

BOP, Et N,
O 3 N

”“2 CH Ci
2 2
O .HeJ’yR 4 O
H2N

x 63-88% 7H2N

 

x
ii/H‘1
0

General procedure for preparation of carrier B/halocarboxylic acid conjugath

To a solution of carrier B (5 equiv) in CH2C12 (4 mL), BOP (1.1 equiv),
triethylamine (5 equiv), and the chiral acid (1 equiv) were added. The solution was
stirred at room temperature overnight. After removal of most of the solvent under
reduced pressure, the product was purified by silica gel flash chromatography to yield the

carrier B / acid conjugate ready for use for CD analysis.

0

HNJK/Br

NH2 B'1'BI'
Carrier B/bromocarboxylic acid conjugates B-l-Br:
lH NMR (CDC13, 300 MHz) 8 7.91 (br, 1H), 7.30 (d, J=8.8 Hz, 2H), 6.67 (d, J=8.8 Hz,

2H), 4.56 (q, J=7.1 Hz, 1H), 3.63 (br, 2H), 1.98 (d, J=7.1 Hz, 3H).

H2N O

B-2-Br
Carrier B/bromocarboxylic acid conjugates B-2-Br:
1H NMR (300Hz, CDC13) 5 8.12 (br, 1H), 7.30 (d, J=8.8 Hz, 2H), 6.68 (d, J=8.8 Hz, 2H),

4.43 (d, J=4.4 Hz, 1H), 2.48 (m, 1H), 1.12 (d, J=6.6 Hz, 3H), 1.05 (d, J=6.6 Hz, 3H).

213

n Br
We N

O B-3-Br
Carrier B/bromocarboxylic acid conjugates B-3-Br:
lH NMR (300Hz, CDC13) 8 8.02 (br, 1H), 7.30 (d, J=8.8 Hz, 2H), 6.68 (d, J=8.8 Hz, 2H),

4.44 (m, 1H), 1.98 (m, 3H), 1.00 (d, J=6.6 Hz, 3H), 0.95 (d, J=6.6 Hz, 3H).

N Br
go if)“

B-4-Br

Carrier B/bromocarboxylic acid conjugates B-4-Br:
1H NMR (300Hz, CDC13) 8 8.04 (br, 1H), 7.30 (d, J=8.8 Hz, 2H), 6.60 (d, J=8.8 Hz, 2H),
4.43 (d, J=4.9 Hz, 1H), 2.25 (m, 1H), 1.61 (m, 1H), 1.35 (m, 1H), 1.09 (d, J=6.6 Hz, 3H),

0.95 (y, J=7.5 Hz, 3H).

H2N

Carrier B/bromocarboxylic acid conjugates B-5-Br:
1H NMR (300Hz, CDC13) 8 7.76 (br, 1H), 7.30 (m, 5H), 7.21 (d, J=8.8 Hz, 2H), 6.65 (d,

J=8.8 Hz, 2H), 4.60 (m, 1H), 3.62 (m, 1H), 3.35 (m, 1H).

O

HNJH/Cl

©

NH2 B-1-Cl
Carrier B/bromocarboxylic acid conjugates B-l-Cl:
1H NMR (CDC13, 300 MHz) 8 7.89 (br, 1H), 7.31 (d, J=8.8 Hz, 2H), 6.67 (d, J=8.8 Hz,

2H), 4.60 (q, J=7.1 Hz, 1H), 3.64 (br, 2H), 1.95 (d, J=7.1 Hz, 3H).

214

, .

,3 Cl
MO W

B-2-CI
Carrier B/chlorocarboxylic acid conjugates B-2-Cl:
1H NMR (300Hz, CDC13) 8 8.20 (br, 1H), 7.31 (d, J=8.8 Hz, 2H), 6.67 (d, J=8.8 Hz, 2H),

4.42 (d, J=3.3 Hz, 1H), 2.66 (m, 1H), 1.12 (d, J=6.6 Hz, 3H), 1.01 (d, J=6.6 Hz, 3H).

...0W

0 B-3-CI

 

Carrier B/chlorocarboxylic acid conjugates B-3-Cl:
1H NMR (300Hz, CDC13) 8 8.12 (br, 1H), 7.32 (d, J=8.8 Hz, 2H), 6.68 (d, J=8.8 Hz, 2H),

4.46 (m, 1H), 1.98 (m, 3H), 1.00 (m, 6H).

Cl
”MW

B-4-Cl

Carrier B / chlorocarboxylic acid conjugates B-4-Cl:
lH NMR (300Hz, CDC13) 8 8.20 (br, 1H), 7.30 (d, J=8.8 Hz, 2H), 6.68 (d, J=8.8 Hz, 2H),
4.45 (d, J=3.7 Hz, 1H), 2.39 (m, 1H), 1.53 (m, 1H), 1.35 (m, 1H), 1.11 (d, J=7.1 Hz, 3H),

0.95 (t, J=6.7 Hz, 3H).

H2N

Carrier B/chlorocarboxylic acid conjugates B-S-Cl:
1H NMR (300Hz, CDC13) 8 7.98 (br, 1H), 7.32 (m, 5H), 7.23 (d, J=8.8 Hz, 2H), 6.67 (d,

J=8.8 Hz, 2H), 4.68 (m, 1H), 3.54 (m, 1H), 3.31 (m, 1H).

215 I

 

o
HN F
N: H2 B-1-F

Carrier B/fluorocarboxylic acid conjugates B-l-F:

1H NMR (CDC13, 300 MHz) 8 7.84 (br, 1H), 7.34 (d, J=8.8 Hz, 2H), 6.69 (d, J=8.8 Hz,

2H), 5.19—5.03 (dd, 1H), 3.66 (br, 2H), 1.71 (m, 3H).

,3 F
MO 2W

B-2-F
Carrier B/fluorocarboxylic acid conjugates B-2-F:
1H NMR (300Hz, CDC13) 8 7.81 (br, 1H), 7.35 (d, J=8.8 Hz, 2H), 6.69 (d, J=8.8 Hz, 2H),
4.86 (dd, 1H), 2.41 (m, 1H), 1.15 (d, J=7.1 Hz, 3H), 1.01 (d, J=7.1 Hz, 3H).
H.911
H2NO 0 B-3-F
Carrier B/fluorocarboxylic acid conjugates B-3-F:
lH NMR (300Hz, CDC13) 8 7.85 (br, 1H), 7.35 (d, J=8.8 Hz, 2H), 6.68 (d, J=8.8 Hz, 2H),
5.04 (m, 1H), 1.98 (m, 3H), 1.02 (m, 6H).
n F
M
“NO 0 B-4-F
Carrier Blfluorocarboxylic acid conjugates B-4-F:
1H NMR (300Hz, CDC13) 8 7.83 (br, 1H), 7.35 (d, J=8.8 Hz, 2H), 6.69 (d, J=8.8 Hz, 2H),
4.89 (m, 1H), 2.16 (m, 1H), 1.57 (m, 1H), 1.30 (m, 1H), 1.11 (d, J=7.1 Hz, 3H), 0.95 (t,

J=7.1 Hz, 3H).

216

I

Carrier B/fluorocarboxylic acid conjugates B-S-F:
1H NMR (300Hz, CDCl3) 8 7.62 (br, 1H), 7.32 (m, 5H), 7.23 (d, J=8.8 Hz, 2H), 6.63 (d,

J=8.8 Hz, 2H), 5.22-5.16 (dd, 1H), 3.44-3.21 (m, 2H).

0

HNJW’B"

K5

NHBOC BOCB'1'BI'
Carrier B/bromocarboxylic acid conjugates BocB-l-Br:
1H NMR (CDC13, 300 MHz) 8 7.99 (br, 1H), 7.44 (d, J=8.8 Hz, 2H), 7.38 (d, J=7.7 Hz,

2H), 6.44 (br, 1H), 4.50 (q, J=7.1 Hz, 1H), 1.98 (q, J=7.1 Hz, 1H), 1.48 (s, 9H).

0 O
HN Br HNJKI’l
Kl, CH30N
reflux
NHBoc 90% NHBoc
BocB-1-Br BocB-1-I

Carrier Blbromocarboxylic acid conjugates BocB-l-I:

The bromo compound BocB-l-Br (34 mg, 0.1 mmol) and K1 (83 mg, 0.5 mmol)
was dissolve in acetonitrile. The mixture was refluxed overnight, and the precipitate was
filtered away, and the filtrate was concentrated under reduced pressure. The residue was
purified by column chromatography to provide the title compound (90%). 1H NMR
(CDC13, 300 MHz) 8 7.79 (br, 1H), 7.24 (d, J=8.8 Hz, 2H), 7.15 (d, J=7.7 Hz, 2H), 6.30

(br, 1H), 4.41 (q, J=7.1 Hz, 1H), 1.90 (q, J=7.1 Hz, 1H), 1.41 (s, 9H).

217

1

 

 

o H o H
CHZCIZ H
\ OH + H N "Me a \ N 'Me
M TIL 2 76% M H Ph
SRIGS

Coupling of (R)-2-bromopropionic acid with (R)/(S)-methylbenzylamine 6R/6S:
(R)-2-bromopropionic acid (15.3 mg, 0.1 mmol) and (R)—methylbenzylamine
(12.1 mg, 0.1 mmol) were dissolved in anhydrous dichloromethane (2 mL). To this
solution was added EDC (20 mg, 0.11mmol) and DMAP (15 mg, 0.12 mmol). This
mixture was stirred at room temperature overnight. The solvent was removed, and the
product was purified by column chromatography (76%). lH NMR (300Hz, CDC13) 8
7.31 (m, 5H), 6.58 (br, 1H), 5.15 (m, 1H), 4.37 (q, J=6.8 Hz, 1H), 1.87 (d, J=7.8 Hz, 3H),

1.50 (d, J=6.8 Hz, 3H).

DMF reflux

 

CF3C02Na + :2 CF3I

7
Synthesis of trifluoromethyl iodide 7:

A mixture of iodine (0.15 mol) and dry sodium salt of trifluoroacetic acid (0.1
mol) was mixed in of dry dimthylformamide (40 mL) in a round-bottomed flask and
heated to reflux under a condenser cooled with water maintained above the boiling point
of the fluoroalkyl iodide product. A syringe was inserted to top of the condenser, and the
other end was in a small round-bottom flask, which was cooled with dry ice—acetone bath.
The gaseous product was collected over an hour period. The product was kept at —78 °C

freezer, and used without further purification.

218

Y.

 

 

 

O O O
1. n-BuLi, -78°C
L3” CI/K/ LST

11 70% 14

Synthesis of chiral amide 14:

To (534-1sopropyl-2-oxazolidinone (260 mg, 2 mmol) dissolved in dry THF (4
mL), was added n-BuLi (2.5 N in hexane, 1 mL) at -78 °C. After it was stirred for half
an hour, propionyl chloride (200 mg, 2.1 mmol) was added into the solution. The
reaction was stirred at -—78 °C for 2 h. The reaction was quenched with 1 mL of saturated
aqueous sodium bisulfate, and then extracted with ether (3 x 10 mL). The ether layer was
dried with NaZSO4, and concentrated. The residue was purified by column
chromatography, to yield the compound (70% yield). 1H NMR (300Hz, CDC13) 8 4.44-
4.06 (m, 3H), 2.91 (q, J=6.6 Hz, 2H), 2.38 (m, 1H), 1.18 (t, J=6.6 Hz, 3H).

0 o o
)L + 0 Me3CCOCI, Et3N JL ,u\/\
0 NH Ho’“\/\ _ o N

W n-BuLi,THF V W
11 65% 15

Synthesis of chiral amide 15:

 

To a stirred solution of butyric acid (176 mg, 2 mmol) in THF (1 mL) at -20 °C
was added Me3CCOC1 (0.24 mL, 2.2 mmol) dropwise over 2 min. After 10 min, Et3N
(0.33 mL, 2.4 mmol) was added and the resulting mixture was allowed to warm to -10 °C
for an hour. In a separate flask, commercially available (S)-4-isopropyl-2-oxazolidinone
11 (260 mg, 2 mmol) was dissolved in dry THF (4 mL), the resulting solution was cooled
to —-78 °C, and to this solution was added n-BuLi in hexane (2.5 N solution, 1 mL). This
cold gel-like solution was poured into the mixed anhydride prepared above. After this

mixture was stirred at —78 °C for 30 min, it was quenched with aqueous sodium bisulfate

219

(l N, 2 mL). THF was removed under reduced pressure, and the residue was extracted
with ethyl acetate (3 x lOmL). The organic layer was dried over Na2804, and
concentrated. The residue was purified by column, and afforded 15 in 65% yield. lH
NMR (300Hz, CDC13) 8 4.44-4.06 (m, 3H), 2.93 (m, 1H), 2.38 (m, 1H), 1.79 (m, 1H),

1.57 (m, 1H), 1.24 (d, J=6.59 Hz, 3H), 0.99 (d, J=7.69 Hz, 3H).

 

 

 

.3— 1. LDA ,3“— ‘
’ ‘ 2.01:31, 5133 ’_‘N CF3
o N

O 020%

14

Synthesis of trifluoromethyl amide l6:

Diisopropylamine (0.2 mL, 1.4 mmol) was dissolved in THF (4 mL), and cooled
it to —78 °C. n-BuLi in hexane (2.5 N solution, 0.5 mL) was added into the solution. The
solution was briefly warmed to 0 °C, and then cooled back to —78 °C. Amide 14 (185
mg, 1 mmol) dissolved in 1 mL of THF was added to this solution via cannula, and the
solution was then stirred at —78 °C for an hour. CF31 was added into the solution with
triethylborane, and was stirred at —78 °C for another 10 minutes, before it was warmed to
—20 °C for 2 hours. The reaction was quenched with saturated aqueous sodium bisulfate
(2 mL). THF was removed under reduced pressure, and the residue was extracted with
ethyl acetate (3 x lOmL). The organic layer was dried over NaZSO4, and concentrated.
The residue was purified by column, but still was a mixture of starting amide 14 and the

desired product 16.

220

 

L

.3 H O , LiOH CF

CF 2 2 3

O?N11’K3 THF/ H20 HOW/K
o o 0
16 9

Synthesis of (R)-trifluoromethylpropionic acid 9:

To the mixture of amide 16 and 14, was added THF and water (4 mL, 3:1), and
then H202 (0.3 mL) and aqueous LiOH (1 mmol, in lmL of water) were successively
added into the reaction mixture. The reaction was stirred at 0 °C for an hour, and then it
was quenched with sodium sulfate solution (1 mL). THF was removed under reduced
pressure, and the residue was extracted with ethyl acetate (3 x lOmL). The organic layer

was dried over NaZSO4, and concentrated to afford the mixed carboxylic acids.

 

\ o
N” o BOP, 513N, ‘NJTOP"
CH Cl
+ HOJKrOPh 2 2 :
56%
NHB°° NHBoc
Me—11B

Methylated Boc-B/carboxylic acid conjugate 11 Me-l 1B:

To a solution of Boc protected carrier B (111 mg, 0.5 mmol) in CHZClz (6 mL),
BOP (250 mg, 0.55 mmol), triethylamine (1 mL), and the chiral carboxylic acid (83 mg,
0.5 mmol) were added. The solution was stirred at room temperature overnight. After
removal of most of the solvent under reduced pressure, the product was purified by silica
gel flash chromatography (56% yield). 1H NMR (CDC13, 300 MHz) 8 7.28 (d, J=8.8 Hz,
2H), 7.04 (d, J=7.7 Hz, 2H), 6.94 (d, J=8.2 Hz, 2H), 6.75 (m, 2H), 6.85 (d, J=8.2 Hz,

2H), 4.56 (q, J=6.6 Hz, 1H), 3.07 (s, 3H), 1.36 (s, 9H), 1.28 (d, J=6.0 Hz, 3H).

221

 

 

\
NH BOP, Et3N, mks/Br

O
+ HOJKE/Br CH20'2
46%

 

NHB°° NHBoc

MeB-1-Br

Methylated Boc-Blcarboxylic acid conjugate MeB-l-Br:

To a solution of Boc protected carrier B (111 mg, 0.5 mmol) in CHzClz (6 mL),
BOP (250 mg, 0.55 mmol), triethylamine (1 mL), and the chiral carboxylic acid (76 mg,
0.5 mmol) were added. The solution was stirred at room temperature overnight. After
removal of most of the solvent under reduced pressure, the product was purified by silica
gel flash chromatography (46% yield). 1H NMR (CDC13, 300 MHz) 8 7.38 (d, J=8.8 Hz,
2H), 7.18 (d, J=7.7 Hz, 2H), 6.60 (br, 1H), 4.56 (q, J=6.6 Hz, 1H), 3.12 (s, 3H), 1.68 (d,

J=6.0 Hz, 3H),1.48 (s, 9H).

KNOz,

0
05mm, HCI GNHC' (able:
68°/

32
The synthesis of a-chloro-y-butyrolactone 32:

Sodium nitrite (0.22 g, 3.24 mmol) was added to a solution of homoserine 35 (137
mg, 1 mmol) and potassium bromide (0.78 g, 6.56 mmol) in 2.5 N sulfuric acid (9 mL) at
—10 °C, and the reaction mixture was stirred for 3 h, and then extracted with CH2C12 (3 x
25 mL). The combined organic layers were dried over sodium sulfate and evaporated.
The residue was purified by column chromatography and provided the pure title
compound (68% yield). 1H NMR (CDC13, 300 MHz) 8 4.51 (m, 1H), 4.38 (m, 2H), 2.75

(m, 1H), 2.46 (m, 1H).

222

 

Ob OH+ ArOH D'AD Pph3 059°”
THF 51-58%
36, 48 and 49
General procedure for the synthesis of (S)-a-aryloxy-'y-butyrolactones:

Aromatic alcohol (0.5 mmol), (R)-a—hydroxy-y—butyrolactone 42 (51 mg, 0.5
mmol) and PPh3 (131 mg, 0.5 mmol) were dissolved in THF (3 mL) and cooled to 0 °C.
DIAD (106 mg, 0.5 mmol) were added to the reaction mixture dropwise. The reaction
was slowly warmed up to room temperature, and stirred overnight. The reaction was
quenched with saturated NaHCO3 (5 mL) and extract with CH2C12 (3 x 10 mL). The
organic layers were combined and dried over NaZSO4. The solvent was removed under
reduced pressure, and the residue was purified by column chromatography to provide the

desired compound.
0
Diff)
36
(S)-a-phenoxy-y-butyrolactone 36:
1H NMR (CDC13, 300 MHz) 8 7.29 (m, 2H), 7.03 (m, 3H), 4.93 (t, J=7.7 Hz, 1H), 4.46

(m, 1H), 4.30 (m, 1H), 2.68 (m, 1H), 2.42 (m, 1H), 1.28 (s, 9H); 13C NMR (CDC13, 300

MHz) 8 173.3, 157.1, 129.3, 122.0, 115.9, 72.3, 65.2, 29.9.

08;“

(S)-a-naphthyloxy-'y-butyrolactone 48:
'H NMR (CDC13, 300 MHz) 8 7.75 (m, 3H), 7.44-7.18 (m, 4H), 5.05 (t, J=7.7 Hz, 1H),

4.46 (m, 1H), 4.33 (m, 1H), 2.74 (m, 1H), 2.42 (m, 1H); 13C NMR (CDC13, 300 MHz) 8

223

1‘

173.1, 154.8, 134.1, 129.5, 129.4, 127.4, 126.7, 126.3, 124.0, 118.5, 108.7, 72.0, 65.2,

29.4.
0
037%)
49 t-Bu
(S)-a-p-tert-butylphenoxy-y-butyrolactone 49:
lH NMR (CDC13, 300 MHz) 8 7.30 (d, J=8.8 Hz, 2H), 6.93 (d, J=8.8 Hz, 2H), 4.90 (t,
J=8.2 Hz, 1H), 4.47 (m, 1H), 4.33 (m, 1H), 2.65 (m, 1H), 2.40 (m, 1H), 1.28 (s, 9H); 13C
NMR (CDC13, 300 MHz) 8 173.3, 154.5, 145.1, 126.1, 114.6, 72.2, 64.4, 33.9, 31.0, 29.3.

0 A920, CH3CN 0

015.0?” RX : (15.01%
reflux 51-70°/o

42

44, 46 and 47

 

General procedure for the synthesis of (R)-a-alkoxy-y-butyrolactones 44, 46 and 47:
(R)-a-hydroxy-y—butyrolactone 42 (51 mg, 0.5 mmol), alkyl halides (2 mmol) and

silver oxide (2 mmol) was refluxed in acetonitrile (5 mL) overnight. The precipitate was

filtered away, and the filtrate was concentrated under reduced pressure. The residue was

purified by column chromatography to yield the pure title product.

0
0‘5. \OMe
44

(R)-a-methoxy-'y-butyrolactone 44:
'H NMR (CDC13, 300 MHz) 8 4.39 (m, 1H), 4.28 (m, 1H), 4.07 (m, 1H), 3.58 (s, 3H),

2.49 (m, 1H), 2.11 (m, 1H).

224

‘1‘

(R)-a-allyloxy-y-butyrolactone 46:
lH NMR (CDC13, 300 MHz) 8 5.79 (m, 1H), 5.15 (m, 2H), 4.38—4.06 (m, 5H), 2.39 (m,
1H), 2.12 (m, 1H); 13C NMR (CDC13, 300 MHz) 8 174.7, 133.3, 117.7, 72.3, 70.8, 65.0,
29.3.

° Q

035.0
47

(R)-a-benzoxy-y-butyrolactone 47 :
1H NMR (CDC13, 300 MHz) 8 7.34 (m, 5H), 4.90 (d, J=11.5 Hz, 1H) 4.70 (d, J=11.5 Hz,
1H), 4.36 (m, 1H), 4.16 (m, 2H), 2.41 (m, 1H), 2.23 (m, 1H); 13C NMR (CDC13, 300

MHz) 8 174.7, 133.6, 128.2, 127.7, 72.7, 71.8, 65.1, 29.5.

0
H2 AlM83 HNJKA/OH

H + R
0 55-76%

N
NHBoc NHBoc

 

General procedure for the synthesis of amides:

To a solution of Boc-B (1.5 eq.) in anhydrous CH2C12 (5 mL) at room temperature
was added trimethylaluminum (2.0 N in hexane, 1.6 eq.) dropwise, under nitrogen. The
resulting mixture was stirred for 15 min, then the a-chiral butyrolactone (1eq.) in CH2C12
(2 mL) was added slowly and stirring was continued at room temperature overnight. The
reaction was quenched with aqueous citric acid (2 N, 2 mL), glycerin (5 drops) and
saturated aqueous potassium sodium tartrate (10 mL), and kept stirring for another 6 h.

The reaction mixture was extracted with CH2C12 (3 x 10 mL). The combined organic

225

’ -

extract were dried over NaZSO4, and concentrated. The residue was purified by column

chromatography to yield the pure amides.

O H
OH TFA U
HNJKR/V n-Bu3P / DIAD b CH2Cl2 o

‘ O N N

 

 

General procedure for the synthesis of chiral butyrolactams:

Tri-n-butylphosphine (1.1 eq.) was added to a solution of DIAD (1.1 eq.) in dry
THF (2 mL) at room temperature. The resulting mixture was stirred for 5 min, then
added dropwise to a solution of chiral amide (1 eq.) in THF (1 mL) at 0 °C under
nitrogen. The reaction mixture was allowed to warm slowly to room temperature and
stirred overnight. The reaction was quenched with saturated NaHCO3 (5 mL) and
extracted with CH2C12 (2 x 10 mL). The combined organic extract were dried over
NaZSO4, and concentrated. The residue was purified by column chromatography to yield
the Boc protected lactam as a white solid. Some times, the lactam could not be separated
from the byproduct from DIAD. The Boc group was removed by stirring overnight in
TFA/CHzClz (1:4) at room temperature. The solvent was then removed by purging
nitrogen. The residue was redissolved in anhydrous CH2C12, and anhydrous Na2CO3 (20
mg) was added. This solution was stirred at room temperature for an hour, and the
Na2C03 was filtered. The solvent was evaporated under reduced pressure to provide

analytically pure product.

226

O

HNJH/VOH

(>01

NHBoc amide 37
Amide 37:
lH NMR (CDC13, 300 MHz) 8 9.04 (br, 1H), 7.31 (d, J=8.8 Hz, 2H), 7.22 (br, 1H), 7.16
(d, J=8.8 Hz, 2H), 4.48 (m, 1H), 3.62 (m, 2H), 2.13 (m, 1H), 1.97 (m, 1H), 1.36 (s, 9H);
l3C NMR (CDC13, 300 MHz) 8 167.5, 153.0, 135.2, 132.0, 120.8, 118.9, 57.9, 56.7, 48.6,

37.4, 28.0.

on,

O (N:

NHBOC Boc-37

Boc-protected lactam Boc-37:

1H NMR (CDC13, 300 MHz) 8 7.54 (d, J=8.8 Hz, 2H), 7.38 (d, J=8.8 Hz, 2H), 6.57 (br,
1H), 4.51 (m, 1H), 3.98 (m, 1H), 3.79 (m, 1H), 2.67 (m, 1H), 2.35 (m, 1H), 1.49 (s, 9H);

13C NMR (CDCI3, 300 MHz) 8 168.6, 152.5, 135.5, 133.6, 120.6, 118.6, 80.4, 55.8, 46.0,

29.3, 28.1.
_0
HNJK/VOH
00
NHBoc amide 39
Amide 39:

1H NMR (CDCI3, 300 MHz) 8 8.17 (br, 1H), 7.41 (d, J=8.8 Hz, 2H), 7.30 (m, 3H), 7.03-
6.95 (m, 4H), 6.47 (br, 1H), 4.85 (t, J=6.6 Hz, 1H), 3.87 (m, 2H), 2.32 (br, 1H), 2.24 (m,

2H), 1.49 (s, 9H).

227

 

8°16

Z

40

Z
I
8’
o

Boc-protected lactam 40:
1H NMR (CDC13, 300 MHz) 8 7.43 (d, J=9.3 Hz, 2H), 7.22 (d, J=8.8 Hz, 2H), 7.18 (m,

2H), 6.95-6.80 (m, 3H), 6.49 (br, 1H), 4.83 (t, J=7.6 Hz, 1H), 3.82 (m, 2H), 2.64 (m, 1H),

 

2.33 (m, 1H), 1.48 (s, 9H).

0
A

©

NHBOC amide 50

Amide 50:

'H NMR (CDC13, 300 MHz) 5 8.43 (br, 1H), 7.40 (d, J=8.8 Hz, 2H), 7.27 (d, J=8.8 Hz,
2H), 6.99 (br, 1H), 5.89 (m, 1H), 5.25 (m, 2H), 4.06 (m, 3H), 3.71 (m, 2H), 3.19 (br, 1H),
1.99 (m, 2H), 1.49 (s, 9H); 13C NMR (CDC13, 300 MHz) 5 170.7, 152.6, 134.9, 133.1.

131.8, 120.4, 118.9, 118.2, 80.0, 77.8, 71.6, 58.7, 35.2, 28.0.

228

 

35"“

N

NHBOC BOG-50

Boc protected lactam Boc-50:
lH NMR (CDC13, 300 MHz) 8 7.59 (d, J=8.8 Hz, 2H), 7.35 (d, J=8.8 Hz, 2H), 6.57 (br,

1H), 5.97 (m, 1H), 5.26 (m, 2H), 4.43 (m, 1H), 4.25-4.17 (m, 2H), 3.87-3.60 (m, 2H),

2.42 (m, 1H), 2.06 (m, 1H), 1.48 (m, 9H).

.1
01—1

N

NH2 50

Lactam 50:

lH NMR (CDC13, 300 MHz) 8 7.35 (d, J=8.8 Hz, 2H), 6.66 (d, J=8.8 Hz, 2H), 5.96 (m,

1H), 5.26 (m, 2H), 4.43 (m, 1H), 4.20 (m, 2H), 3.87-3.60 (m, 2H), 2.44 (m, 1H), 2.06 (m,

1H).

229

 

?

5

O
029
a: .

2

Lactam 51:
1H NMR (CDC13, 300 MHz) 8 7.37 (d, J=8.8 Hz, 2H), 6.65 (d, J=8.8 Hz, 2H), 4.11 (m,
1H), 3.87 (m, 1H), 3.74 (m, 1H), 3.65 (m, 1H), 3.53 (m, 1H), 2.41 (m, 1H), 2.06 (m, 1H),

1.60 (m, 2H), 0.92 (t, J=7.1 Hz, 3H).

0

HNJH/VOH

E: OMe

NHBoc amide 52
Amide 52:
1H NMR (CDCl3, 300 MHz) 8 8.33 (br, 1H), 7.46 (d, J=8.2 Hz, 2H), 7.31 (d, J=8.2 Hz,
2H), 6.62 (br, 1H), 3.88 (d, J=6.6 Hz, 1H), 3.77 (d, J=6.0 Hz, 2H), 3.48 (s, 3H), 2.04 (m,

2H), 1.49 (s, 9H).

MeO

0479
N

HBOC BOC'52

1H NMR (CDC13, 300 MHz) 8 7.52 (d, J=8.8 Hz, 2H), 7.33 (d, J=8.8 Hz, 2H), 6.57 (br,

Boc protected lactam Boc-52:

1H), 4.05 (t, J=7.7 Hz, 1H), 3.71 (m, 2H), 3.57 (s, 3H), 2.45 (m, 1H), 2.04 (m, 1H), 1.49

(s, 9H).

230

 

HNJKl/V

(:1 OVPhH

HBoc Amide 53
Amide 53:
1H NMR (CDC13, 300 MHz) 8 8.37 (br, 1H), 7.34 (m, 9H), 6.62 (br, 1H), 4.62 (s, 2H),
4.13 (t, 1260 Hz, 1H), 3.76 (m, 2H), 2.62 (br, 1H), 2.05 (m, 2H), 1.49 (s, 9H), 13C NMR
(CDC13, 300 MHz) 8 170.5, 152.5, 136.1, 134.7, 131.9, 128.5, 128.2, 127.8, 120.4, 118.7,

80.3, 78.5, 73.1, 59.1, 35.3, 28.0.

Lactam 53:
lH NMR (CDC13, 300 MHz) 8 7.36 (m, 7H), 6.65 (d, J=8.8 Hz, 2H), 4.99 (d, J=12.1 Hz,

1H), 4.79 (d, J=12.l Hz, 1H), 4.20 (t, J=7.7 Hz, 1H), 3.87-3.62 (m, 2H), 2.35 (m, 1H),

2.12 (m, 1H).
0
HN’U\/\/0H
NHBOC amide 54
Amide 54:

1H NMR (CDC13, 300 MHz) 8 8.27 (br, 1H), 7.79-7.63 (m, 3H), 7.43 (m, 4H), 7.22 (m,
4H), 6.60 (br, 1H), 4.50 (t, J=6.0 Hz, 1H), 3.87 (m, 2H), 2.72 (br, 1H), 2.28 (m, 2H), 1.49
(s, 9H); l3C NMR (CDC13, 300 MHZ) 8 169.4, 154.3, 152.6, 135.1, 133.8, 131.5, 129.8,

129.4, 127.3, 126.8, 126.4, 124.2, 120.7, 118.7, 118.0, 108.9, 80.2, 60.2, 58.3, 35.1, 28.1.

231

 

z 0”

Z
I
N

Lactam 54:
IH NMR (CDC13, 300 MHz) 8 7.73 (m, 3H), 7.41-7.20 (m, 6H), 6.73 (d, J=8.8 Hz, 2H),

5.15 (t, J=6.0 Hz, 1H), 3.85 (m, 1H), 2.77 (br, 1H), 2.23 (m, 1H), 1.49 (s, 3H).

0
HNJKA/OH

r: '0...

6
NHBOC amide 55
Amide 55:
lH NMR (CDC13, 300 MHz) 8 8.28 (br, 1H), 7.40 (d, J=8.8 Hz, 2H), 7.28 (m, 4H), 6.89
(d, J=8.8 Hz, 2H), 6.72 (br, 1H), 4.80 (t, J=6.0 Hz, 1H), 3.83 (m, 2H), 2.86 (br, 1H), 2.21
(m, 2H), 1.49 (s, 9H), 1.25 (s, 9H); 13C NMR (CDC13, 300 MHz) 8 169.9, 154.5, 152.3,

144.8, 135.1, 131.6, 126.3, 120.9, 118.7, 114.3, 101.2, 80.2, 58.5.

232

 

 

t—Bu

Q

0

b

O

NHBoc BOG-55

Boc protected lactam Boc-55:
lH NMR (CDC13, 300 MHz) 8 7.66 (d, J=9.09 Hz, 2H), 7.42 (d, J=8.79 Hz, 2H), 7.35 (d,
J=9.09 Hz, 2H),7.05 (d, J=8.79 Hz, 2H), 6.51 (br, 1H), 5.03 (t, 1:7 .62 Hz, 1H), 3.89 (m,

2H), 2.70 (m, 1H), 2.33 (m, 1H), 1.55 (s, 9H), 1.34 (s, 9H).

 

N H, TB I
THF 71
87%

Monosilylated ethylene glycol 71:

NaH (60% suspension, 80 mg, 2 mole, washed 3 times with hexane) was added
to a solution of ethylene glycol (124 mg, 2 mole) in THF (5'mL) at 0 °C. After this
mixture was stirred at room temperature for an hour, TBSCl (300 mg, 2 mmol) was
added into the solution, and the reaction was stirred at room temperature overnight. The
reaction was quenched with brine, and extracted with CHZCIZ (3 x 20 mL). The organic
layer was combined and dried over Na2804. The residue was purified by column
chromatography to yield the title compound (87%). lH NMR (CDC13, 300 MHz) 8 3.69

(m, 2H), 3.60 (m, 2H), 0.81 (s, 9H), 0 (s, 6H).

233

 

N

71 84% 72

 

Tosylate 72:

Monosilylated ethylene glycol 71 (176 mg, 0.1 mmol), and Et3N (1 mL) were
dissolved in CH2C12 (10 mL). TsCl (380 mg, 0.2 mmol) was slowly added in small
portion at 0 °C. The reaction was allowed to warm to room temperature, and stirred
overnight. The solvent was removed under reduced pressure, and the residue was
purified by column chromatography to yield the title compound (84%). lH NMR
(CDC13, 300 MHz) 8 7.79 (d, J=8.2 Hz, 2H), 7.32 (d, J=8.2 Hz, 2H), 3.81 (t, J=6.6 Hz,

2H), 3.81 (t, J=6.6 Hz, 2H), 2.43 (s, 3H), 0.81 (s, 9H), 0 (s, 6H).

 

Nal / CH3CN
OTBS OTBS
TSO/V reflux IN
72 96% 68

TBS-protected iodoethanol 68:

Tosylate 72 (165 mg, 0.5 mmol) was dissolved in CH3CN (10 mL) together with
Nal (10 mg, 1 mmol). This mixture was refluxed overnight, and the precipitate was
filtered away, and the filtrate was concentrated under reduced pressure. The residue was
purified by column chromatography to yield the title compound (96%). 1H NMR
(CDC13, 300 MHz) 8 3.81 (t, J=6.59 Hz, 2H), 3.18 (t, J=6.59 Hz, 2H), 0.89 (s, 9H), 0 (s,

6H).

234

 

 

 

Ph 0 Ph
M978 N]: Allybromide WM]:

62 (Ne -78°C 89% 73 5M6

The alkylated product 73:

To a solution of lithium diisopropylamide (LDA) (1.55 mmol) in THF was added
a solution of chiral auxiliary (318 mg, 1.55 mmol) 62 in THF (2 mL) dropwise at —78 °C.
The mixture was stirred for 45 min at —78 °C, and a solution of allylbromide (0.48 g, 4
mmol) in THF was added slowly. After complete addition, the mixture was stirred an
additional 4 h, allowed to slowly warm to —50 °C, and quenched by pouring into ice-
water. The aqueous mixture was extracted (3 x 30 mL) with ether, and the combined
ether layers were dried with NaZSO4, and concentrated. Following column
chromatography affords the title compound as clear oil (89% yield). 1H NMR (CDC13,
300 MHz) 8 7.32 (m, 5H), 5.83 (m, 1H), 5.23 (d, J=6.6 Hz, 2H), 5.03 (m, 2H), 4.09 (m,
1H), 3.58 (m, 1H), 3.48 (m, 1H), 3.38 (s, 3H), 2.50-2.40 (m, 4H).

0 Ph 1.LDA H :70
WNTSQI INOTBS 68 >/

73 OMe -98°C
2. HEB

 

56%

(S)-2-Allyl-y-butyrolactone 67:

To a solution of LDA (1.1 eq.) was added the compound 73 (246 mg, 1 mmol) in
THF dropwise at —78 °C, and the mixture was stirred at —78 °C for 45 min. The reaction
was then cooled to —98 °C, and 68 (1.03 g, 4 mmol) was slowly added. The reaction was
kept at —98 °C for 2 h, and then at -78 °C for 4 h. After being warmed slowly to 0 °C, the
solution was quenched with ammonium chloride solution, and extracted with ether (4 x

20 mL). The ether extract was dried over Na2S04, and concentrated. The crude alkylated

235

.s - .. .
nun—ah. -~.‘_—_~.._—v-_4 a.

product was hydrolyzed immediately in 20 mL of refluxing 4.5 N HCl for 15 min. The
solution was cooled down to room temperature, and extracted with ether (4 x 20 mL).
The ether extract was dried over Na2804, and concentrated. Column chromatography
provided the pure title product in 56% yield. 1H NMR (CDCl3, 300 MHz) 8 5.77 (m,
1H), 5.11 (m, 2H), 4.34-4.16 (m, 2H), 2.63 (m, 2H), 2.42 (m, 2H), 1.99 (m, 1H); 13C

NMR (CDC13, 300 MHz) 8 182.6, 134.1, 117.4, 66.2, 38.5, 34.0, 27.5.

0
V

©

NHBOC 75

Allyl amide 76:

1H NMR (CDC13, 300 MHz) 8 7.67 (br, 1H), 7.37 (d, J=8.8 Hz, 2H), 7.26 (d, J=8.8 Hz,
2H), 6.50 (br, 1H), 5.78 (m, 1H), 5.09 (m, 2H), 3.70 (m, 2H), 2.50 (m, 2H), 2.24 (m, 1H),
1.95 (br, 1H), 1.81 (m, 2H), 1.49 (s, 9H); 13C NMR (CDC13, 300 MHz) 8 168.1, 153.2,

135.2, 134.3, 132.7, 120.5, 128.9, 117.0, 80.2, 60.0, 44.2, 36.5, 34.4, 28.0.
i?
O N
(3430C BOG-75
Boc protected lactam Boc-75:
1H NMR (CDC13, 300 MHz) 8 7.52 (d, J=8.8 Hz, 2H), 7.34 (d, J=8.8 Hz, 2H), 6.49 (br,
1H), 5.78 (m, 1H), 5.07 (m, 2H), 3.71 (m, 2H), 2.65 (m, 2H), 2.25 (m, 2H), 1.79 (m, 1H),

1.48 (s, 9H); 13(3 NMR (CDC13, 300 MHz) 5 174.7, 152.5, 135.1, 134.3, 120.1, 118.5,

116.7, 80.1, 46.6, 42.5, 35.1, 28.0, 23.6, 21.6.

236

’1'

33
O N

NH2 75
Allyl lactam 75:

l H NMR (CDC13, 300 MHz) 8 7.42 (d, J=8.8 Hz, 2H), 6.70 (d, J=8.8 Hz, 2H), 5.83 (m,

1H), 5.17 (m, 2H), 3.75 (m, 2H), 2.70 (m, 2H), 2.31 (m, 2H), 1.82 (m, 1H).

Propyl lactam 77:

To the allyllactam 75 (21.6 mg, 0.1 mmol) in EtOAc (5 mL) was added Pd/C (5
mg). This reaction was then hydrogenated at atmospheric pressure for 20 h. The black
powder was filtered, and the filtrate was concentrated under reduced pressure to yield the
pure product (97%). IH NMR (CDCl3, 300 MHz) 8 7.51 (d, J=8.8 Hz, 2H), 7.32 (d,
J=8.8 Hz, 2H), 6.57 (br, 1H), 3.72 (m, 2H), 2.55 (m, 1H), 2.27 (m, 1H), 1.77 (m, 1H),
1.48 (s, 9H), 1.22 (d, J=7.14 Hz, 3H), 0.99 (m, 4H); l3C NMR (CDCl3, 300 MHz) 8

175.2, 152.3, 134.1, 120.1, 118.6, 80.1, 46.6, 42.8, 33.0, 28.0, 24.4, 21.6, 13.7.

237

 

 

 

 

4.:‘4 ”LAJ— I.“ +J_...-'_,__ ‘_

Ph

 

Ph
Me—<\ (LT M91 Viz?"
I

62 OMe -78 °c 82% 63 (Me

The alkylated product 63:

To a solution of lithium diisopropylamide (LDA) (1.55 mmol) in THF was added
a solution of chiral auxiliary (318 mg, 1.55 mmol) 62 in THF (2 mL) dropwise at —78 °C.
The mixture was stirred for 45 min at —78 °C, and a solution of methyl iodide (0.64 g, 4
mmol) in THF was added slowly. After complete addition, the mixture was stirred an
additional 4 h, allowed to slowly warm to —50 °C, and quenched by pouring into ice-
water. The aqueous mixture was extracted (3 X 30 mL) with ether, and the combined
ether layers were dried with Na2SO4, and concentrated. Column chromatography affords
the title compound as clear oil (82% yield). lH NMR (CDC13, 300 MHz) 8 7.30 (m, 5H),
5.24 (d, J=6.6 Hz, 2H), 4.07 (m, 1H), 3.58 (m, 1H), 3.48 (m, 1H), 3.38 (s, 3H), 2.38 (t,

J=7.7 Hz, 2H), 1.21 (t, J=7.7 Hz, 3H).

 

OTPh 1. LDA H £0
\/<‘ . > 0,--
N "1' IA/OTBS 68
53 0M9 -98°C 66
47°
2. H69 A

(S)-2-Methyl-7-butyrolactone 66:

To a solution of LDA (1.1 eq.) was added the compound 63 (220 mg, 1 mmol) in
THF dropwise at —78 °C, and the mixture was stirred at —78 °C for 45 min. The reaction
was then cooled to -98 °C, and 68 (1.03 g, 4 mmol) was slowly added. The reaction was
kept at —98 °C for 2 h, and then at —78 °C for 4 h. After being warmed slowly to 0 °C, the
solution was quenched with ammonium chloride solution, and extracted with ether (4 x

20 mL). The ether extract was dried over Na2804, and concentrated. The crude alkylated

238

product was hydrolyzed immediately in 20 ml. of refluxing 4.5 N HCl for 15 min. The
solution was cooled down to room temperature, and extracted with ether (4 x 20 mL).
The ether extract was dried over NaZSO4, and concentrated. Column chromatography
provided the pure title product in 47% yield. 1H NMR (CDCl3, 300 MHz) 8 4.40-4.22

(m, 2H), 2.62-2.53 (m, 2H), 1.96 (m, 1H), 1.32 (d, J=7.1 Hz, 3H).
NHBOC

1H NMR (CDC13, 300 MHz) 8 7.53 (d, J=8.8 Hz, 2H), 7.34 (d, J=8.8 Hz, 2H), 6.46 (br,

Boc-78

Boc protected lactam Boc-78:

1H), 3.74 (m, 2H), 2.63 (m, 1H), 2.34 (m, 1H), 1.73 (m, 1H), 1.48 (s, 9H), 1.27 (d, J=7.1

Hz, 3H); 13C NMR (CDC13, 300 MHz) 8 174.7, 158.8, 134.2, 133.0.1, 119.0.4, 75.3, 60.0,

38.4, 36.2, 28.0, 17.4.

239

 

 

Reference:

1.

10.

11.

12.

13.

14.

15.

16.

17.

18.

19.

Elie], E. L.; Wilen, S. H. Stereochemistry of Organic Compounds; John Wiley &
Sons, Inc.: New York, 1994.

Iida, K.; Kajiwara, M., J. Labeled Compd. Radiopharm. 1991, XXIX, 201.

Matins, M. A. P.; Sinhorin, A. P.; Zimmerrnann, N. E. K.; Znatta, N.; Bonacorso,
H. G.; Bastos, G. P. Synthesis 2001, 13, 1959.

Tyrrell, E.; Tsang, M. W. H.; Skinner, G. A.; Fawcett, J. Tetrahedron 1996, 52,
9841.

Nagai, Y.; Kusumi, K. Tetrahedron Lett. 1995, 36, 1853.
Kusumi, K.; Yabuuchi, T.; Ooi, T. Chirality 1997, 9, 550.
Mizutani, J .; Tahara, S. Tetrahedron Lett. 1996, 3 7, 4737.

Proni, G.; Pescitelli, G.; Huang, X.; Quraishi, N. Q.; Nakanishi, K.; Berova, N.
Chem. Commun. 2002, 1590.

Proni, G.; Pescitelli, G.; Huang, X.; Nakanishi, K.; Berova, N. J. Am. Chem. Soc.
2003, 125, 12914.

Iseki, K.; Nagai, T.; Kobayashi, Y. Tetrahedron Lett. 1993, 34, 2169.
Liou, K.; Yang, P.; Lin, Y. J. Organomet. Chem. 1985, 294, 145.

Jadhav, P. K.; Bhat, K. S.; Perumal, P. T.; Brown, H. C. J. Am. Chem. Soc. 1992,
114, 385.

Paskovich, R.; Gaspar, P.; Hammond, G. S. J. Org. Chem. 1967, 32, 833.
Ghosh, A. K.; Hussain, K. A.; Fidanze, S. J. Org. Chem. 1997, 62, 6080.
Anh, N. T.; Eisenstein, O. Nouv. J. Chem. 1977, 1, 61.

Lodge, E. P.; Heathcock, C. H. J. Am. Chem. Soc. 1987, 109, 3353.

Bell, 1. M.; Beshore, D. C.; Gallicchio, S. N.; Williams, T. M. Tetrahedron Lett.
2000, 41, 1141.

Mitsunobu, 0. Synthesis 1981, 1.

Basha, A.; Lipton, M.; Weinreb, S. M. Tetrahedron Lett. 1977, 18, 4171.

240

20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

30.

31.

32.

33.

34.

35.

36.

37.

38.

Mitsunobu, 0.; Eguchi, M. Bull. Chem. Soc. Jpn. 1971, 44, 3427.
Finch, N.; Fitt, J. J .; Hsu, I. H. S. J. Org. Chem. 1975, 40, 206.
Mitsunobu, 0.; Yamada, M. Bull. Chem. Soc. Jpn. 1967, 40, 935.
Johnson, C. R.; Dutra, G. A. J. Am. Chem. Soc. 1973, 95, 7777.

Boldrini, G. P.; Mengoli, M.; Tagliavini, E.; Trombini, C.; Umani-ronchi, A.
Tetrahedron Lett. 1986, 27, 4223.

Boldrini, G. P.; Savoia, D.; Tagliavini, E.; Trombini, C.; Umani-ronchi, A. J. Org.
Chem. 1983, 48, 4108.

Braga, D.; Ripamonti, A.; Savoia, D.; Trombini, C.; Umani-ronchi, A. J. Chem.
Soc., Chem. Commun. 1978, 927.

Savoia, D.; Tagliavini, E.; Trombini, C.; Umani-ronchi, A. J. Org. Chem. 1981,
46, 5340.

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

Rieke, R. D; Kavaliunas, A. V.; Rhyne, L. D.; Fraser, D. J. J. Am. Chem. Soc.
1979, 101, 246.

Cardillo, G.; D'Amico, A.; Orena, M.; Sandri, S. J. Org. Chem. 1988, 53, 2354.

Lermer, L.; Neeland, E. G.; Ounsworth, J. P.; Sims, R. J .; Tischler, S. A.; Weiler,
L. Can. J. Chem. 1992, 70, 1427.

Eberlein, T. H. J. Org. Chem. 1992, 57, 3479.

Prashad, M.; Kim, H.; Har, D.; Repic, 0.; Blacklock, T. J. Tetrahedron Lett.
1998, 39, 9369.

Meyers, A. 1.; Yamamoto, Y.; Mihelich, E. D.; Bell, R. A. J. Org. Chem. 1980,
45, 2792.

Iwata, A.; Tang, H.; Kunai, A. J. Org. Chem. 2002, 67, 5170.
Detty, M. R.; Seidler, M. D. J. Org. Chem. 1981, 46, 1283.

Ohshita, J .; Iwata, A.; Kanetani, F.; Kunai, A.; Yamamoto, Y.; Matui, C. J. Org.
Chem. 1999, 64, 8024.

Sato, T.; Hanayama, K.; Tujisawa, T. Tetrahedron Lett. 1988, 29, 2197.

241

 

39. Huang, X.; Rickman, B.; Borhan, B.; Berova, N.; Nakanishi, K. J. Am. Chem.
Soc.1998,120,6185.

242

Chapter 5
The chiral aggregation of porphyrin acids

It is well-known that porphyrins tend to self-align into one-dimensional aggregate
under appropriate conditions, creating novel supramolecular architectures."3 Three types
of interactions have been proposed for porphyrin aggregationz4’ 5 (1) free-base porphyrins
aggregate via a weak 1t-1t stacking and/or van der Waals interaction;6 (2) some
metalloporphyrins exhibit a strong metal-1t interactions that can be abolished by ligand
binding to the metal and a strong metal-side chain binding; (3) the aggregation of
porphyrins can also be reinforced by hydrogen-bond interactions between the substituents
on the porphyrin periphery.7'9

Upon self-assembly, the porphyrins can form two different kinds of aggregates:
the “face-to-face” H-type and the “edge-to-edge” J—typelo (Figure 5-1). It is known that
the Soret bands of H-type and J-type aggregates appear at shorter and longer wavelength,

respectively, compared to the Soret bands of the corresponding porphyrin monomers.l " ‘2

This could be interpreted on the basis of the exciton coupling mechanism.” '4

 

H-aggregation J- aggregation
Figure 5-1. The self-assembly of porphyrins.

Some porphyrin aggregates have attracted a lot of attention for mimicking the

self-assembly of bacteriochlorophyll c,'5' '6

the main pigment of chlorosomes (green
sacs), the light-harvesting organelles of green photosynthetic bacteria. The study of

porphyrin aggregates could potentially give access to artificial and robust devices for

243

photochemical energy conversion.” ‘8 Porphyrin aggregatesl9 have also been studied for
serving as a new generation of optoelectronic devices for telecommunications, optical
switching and information storage.”22

Since porphyrins can easily aggregate, we were interested in investigating if the
porphyrins could aggregate in a chiral fashion upon addition of chiral substances. If so,
we would like to investigate the possibility of determining the stereochemistry of chiral
compounds by studying the helicity induced within porphyrin aggregates, analyzed by
ECCD.

In order to investigate the above hypothesis, zinc porphyrin acid was synthesized.
Porphyrin acids should readily aggregate in non-polar solvents, such as hexane, and
methylcyclohexane, while the acid functionality should align the porphyrin planes

through hydrogen bonds and polar interactions. In addition, as we saw before for the zinc

porphyrin tweezer case, the zinc atom can serve as a binding site to coordinate with chiral

substrates.

5‘ O
K“ d

R

\R‘ ‘.

Z‘8R2 _. ® or
/
O O-H ------- O OH positive ECCD negative ECCD

Figure 5-2. Porphyrin acid aggregate as a chromophoric host to determine the
stereochemistry of chiral substrates.

A schematic representation of the postulated aggregation idea is shown in Figure
5-2. Upon binding of the chiral substrate, the porphyrin aggregate will adopt the Chirality

of the substrate, resulting in a preferred helicity, which can be observed as an ECCD

244

spectrum. The sign of the observed ECCD spectrum, in turn, reflects the absolute
stereochemistry of the chiral substrate.
5-1. Determination of extinction coefficient (a) of zinc porphyrin acid

In order to determine the exact concentration of the zinc porphyrin acid aggregate,
it is important to know the extinction coefficient (8) of the zinc porphyrin acid. We have
found two published extinction coefficients (8) for zinc porphyrin acid. $353,000 M'
'cm" in methanol,23 and $602,559 M'cm‘l in CH3C1.24 Although the extinction
coefficient of the same compound varies when measured in different solvents, it should
not be so different as what we found in these two reports. In fact, when these numbers
were compared with our experimental results, with the same amount of porphyrin being
added to different solvents (MeOH and CHC13), and the UV-vis was recorded, we did not
observed the degree of difference as reported. Since the published numbers contradicted
each other, we decided to determine the e for the porphyrin carboxylic acid ourselves.
The process, which was more troublesome than anticipated, is described in the next part.
5-l.l. Determination of extinction coefficient (8) by Lambert-Beer’s Law

The first way to determine an extinction coefficient was by using Lambert-Beer’s
law (6-1):

A 2 ed (6-1)

where A is the absorption observed by UV-vis, 8 is the extinction coefficient, c is

the concentration of the compound, and l is the length of the UV cell, which is 1 cm.

According to this equation, the extinction coefficient (8) of a compound could be

. . . . . . A
easrly deterrmned 1f the concentration of the solution 18 known: 8 = 7.
c

245

Zinc porphyrin acid can be prepared by addition of zinc acetate to porphyrin acid
free-base in dichloromethane. Since the zinc porphyrin acid decomposes very easily, it
was purified by column chromatography before use. The dried zinc porphyrin acid was
then carefully weighted, and known volume of freshly distilled dichloromethane was
added to make a porphyrin acid stock solution of known concentration. A known amount
of that was then diluted with methylcyclohexane and the spectrum was recorded. Since
the concentration of the zinc porphyrin acid was known, the 80f zinc porphyrin acid in
methylcyclohexane could be calculated according to the above equation.

The measurement was repeated twice and the extinction coefficient (8) was
calculated. However, the results of those two measurements were not in good agreement.
In one case, a e of 150,000 M'lcm'l was calculated, while in the other case, the number
dropped to only 83,000 M'lcm’l. Compared to the reported extinction coefficient (8) of
zinc porphyrin methyl ester ($550,000 M'lcm’l in CHzClz),25 our measured 8 were
considered inaccurate. The inconsistency of the two measurements urged us to try to
rationalize it, and we believe that it is due to the aggregation of zinc porphyrin acid in
methylcyclohexane. The self-aggregation of zinc porphyrin acid in solution would result
in the deviation from the Lambert-Beer’s law, and consequently, the extinction
coefficient of porphyrin acid cannot be determined this way.

5-1.2. Determination of extinction coefficient (5) of zinc porphyrin acid based on
that of zinc porphyrin methyl ester

Since the determination of extinction coefficient (8) of zinc porphyrin acid by
Lambert-Beer’s law failed because of the inevitable self-aggregation of porphyrin acid in

methylcyclohexane, we had to find another way to determine it.

246

Zinc porphyrin acid could be easily converted to zinc porphyrin methyl ester by
addition of a methylation reagent, such as, TMSCHNz. Since the extinction coefficient
(8) of zinc porphyrin methyl ester is known, we could calculate its exact concentration in
the solution and therefore, upon conversion to the methyl ester, we could determine the
extinction coefficient of zinc porphyrin acid.

acid + TMSCHN 2 —) methylester (6-2)
For a diluted zinc porphyrin acid, its absorption Ada-d could be determined by UV-

vis. By addition of TMSCHNZ, it was converted to the corresponding methyl ester, while
the excess of TMSCHNz could be easily removed by purging the cell with nitrogen gas.

When the residue was dissolved in fresh solvent, the only thing left was the zinc
porphyrin methyl ester, whose absorption Ame, could also be recorded by UV-vis.
During this conversion, the concentration of the starting acid should be equal to that of

the product methyl ester, and based on the Lambert-Beer’s equation: A = eel , we could

deduce this equation (6-3):

. Aacid
Eaad = m x gester (6-3)

Ada-d and Ame, are the absorption of the porphyrin acid and methyl ester,
respectively, which can be obtained from the UV-vis spectra, and cam; and 8“,” are the

extinction coefficient of porphyrin acid and methyl ester, with Ees,e;550,000 M'lcm’l in

CH2C12.25
Based on this method, the extinction coefficient of zinc porphyrin acid in CHzClz
was found to be 580,000 M'lcm’l. Since it is quite close to the published figure

$602,559 M’l cm'1 in CH3Cl, we are quite satisfied with it. Since the zinc porphyrin

247

 

 

 

acid aggregates in methylcyclohexane, there is no need to deduce its extinction
coefficient.
5-2. The zinc porphyrin acid aggregate with carrier F derivatized (S)-O-
acetylmandelic acid 9F

The idea of using zinc porphyrin aggregate as chromophoric host was tested
briefly, but we have observed some interesting phenomena. The first chiral complex
tested for the zinc porphyrin acid aggregate system was the carrier F derivatized (S)-O-
acetylmandelic acid 9F, since this substrate exhibited the largest ECCD amplitude when
added to the zinc porphyrin tweezer in methycyclohexane, (A = +604). The structures of
carrier F derivatized (S)-O-acetylmandelic acid 9F and zinc porphyrin acid are illustrated

in Figure 5—3.

 

Figure 5-3. The structures of the carrier F derivatized (S)-O-acetylmandelic acid 9F and
zinc porphyrin acid.

When the carrier F derivatized (S)-O-acety1mandelic acid 9 was added to the
freshly made zinc porphyrin acid in methylcyclohexane at 0 °C, a positive ECCD

spectrum was obtained (Figure 5-4).

248

 

 

 

 

375 425 475
wavelength (nm)

 

Figure 5-4. The addition of chiral substrate 9F to zinc porphyrin acid in

methylcyclohexane leads to a positive ECCD.

"1'

However, this porphyrin acid aggregate was not as stable as the zinc porphyrin
tweezer system. As shown in Figure 5-5, the amplitudes of the ECCD spectra decreased
significantly with time. After an hour, the amplitude dropped to nearly one third of the

original.

80

— 0 minute

—— 15 minutes
— 30 minutes
——-45 minutes
—— 60 minutes
— 75 minutes

 

.30
375 425 475

wavelength (nm)

Figure 5-5. The time study of zinc porphyrin acid aggregate with 9F .
We had not observed the decrease in amplitude when zinc porphyrin tweezer was

used as the chromophoric host. The decreasing of the ECCD amplitude of zinc porphyrin

249

 

 

 

acid aggregate might suggest the instability of the zinc porphyrin acid aggregate.
However, the change in the amplitude could also be caused by photo bleaching: the
degradation of the zinc porphyrin acid under the prolonged light radiation. In order to
test this possibility, we added the chiral substrate 9F to zinc porphyrin acid, and divided
this solution into two. One portion was kept in the CD machine, under the radiation of
light all the time, while the other one was shielded away from light. After an hour, CD

spectra were measured for both solutions, and the amplitudes for both solutions were

 

much smaller than that taken an hour ago (Figure 5-6). This experiment excluded the

possibility of photo bleaching.

30.

 

15 j
: —— stock solution
A8 0 — shielded from light
in an hour
—- in CD cell in an
I hour
-15
-30 i
375 425 475

wavelength (nm)
Figure 5-6. The photo bleaching study.
Some preliminary studies were also performed for the binding between zinc
porphyrin acid and the carrier F derivatized (S)-O-acetylmandelic acid. As shown in
Figure 5-7, the addition of the carrier F derivatized (S)-O—acetylmandelic acid 9F to zinc

porphyrin acid exhibited positive ECCD spectra at 0 0C, while the amplitudes of ECCD

250

spectra varied with the amount of chiral complex present. The maximum amplitude was
reached with the addition of 1 equivalent of chiral complex. From the ECCD spectra, it
was clear that the ECCD signals were sensitive to the amount of chiral complex in the
system, since there was nearly no ECCD signal upon addition of more than 10
equivalents of chiral complex. On the other hand, for the system with zinc porphyrin
tweezer as the chromophoric host, more than 100 equivalent of chiral complex has to be

added before the disappearance of ECCD peaks.26

 

-80
375 425 475

wavelength (nm)
Figure 5-7. The titration of zinc porphyrin acid with 9F.

When zinc porphyrin acid aggregates coordinate with the chiral substrate, it is the
zinc atom, which serves as the binding site. This can be proved by adding the chiral
substrate to free-base porphyrin acid, in which no CD was observed. From UV-vis, it
could be seen that there was no binding between the substrate and porphyrin acid free

base, since the absorption wavelength kmax of the porphyrin acid Soret band remained the

251

same before and after the addition of the chiral substrate. However, when the substrate
was added into zinc porphyrin acid, a new peak was observed at 427 nm in the UV-vis
spectrum.

The above experiment eliminates the possibility of the carboxylic acid
functionality on the porphyrin periphery binding to the chiral substrate. Instead, the
carboxylic acid functionality on the porphyrin plays an important role in the porphyrin
aggregation. When the acid functionality was esterified as a methyl ester, as shown in
Figure 5-8, the amplitude of the ECCD spectrum decreased significantly upon addition of

9F, even though the sign stayed the same.

10

 

375 425 475

wavelength (nm)
Figure 5-8. The ECCD of 9F in zinc porphyrin methyl ester.

The above studies indicated that the carboxylic acid functionality could help the
aggregation and alignment of the zinc porphyrin aggregates by hydrogen bonding. The
importance of hydrogen bonding could also be seen by addition of base, which resulted in
reducing the hydrogen bonds and increase of charge repulsion between aggregating

carboxylates. When an inorganic base, such as sodium carbonate, was added into the

252

 

?_.

 

’9', m“ u Git-51L”

 

 

mixture of zinc porphyrin acid and chiral substrate 9F, the amplitude of the ECCD
spectrum decreased (Figure 5-9). However, the addition of sodium carbonate was not as
effective as expected in disrupting the hydrogen-bonding network. This might be due to
the limited solubility of the base and also the porphyrin carboxylate in
methylcyclohexane. On the other hand, as shown in Figure 5-10, no ECCD was
observed, when the zinc porphyrin tweezer was pre-treated with sodium carbonate before

the addition of the chiral substrate 9F.

80
40 j
' -—— no base
0 i -\ —added base-1
Ac ' a . g _
. 7 v .. ‘ .‘ \‘. w . added base-2
. \ . w ~J~ —— added base-3
‘40 ‘3‘ ,1'
; 1
-80
375 425 475

wavelength (nm)

Figure 5-9. The addition of sodium carbonate to zinc porphyrin tweezer.

 

375 425 475
wavelength (nm)

Figure 5-10. The CD spectrum of 9F adds to pre-treated zinc porphyrin acid.

253

 

T .

Furthermore, when an organic base such as diisopropylmethylamine was added to
the zinc porphyrin acid solution with the chiral substrate present, the ECCD peak
decreased very quickly, and finally disappeared. This could be due to two separate
reasons: (a) diisopropylmethylamine deprotonates the carboxylic acid, disrupting the
hydrogen bond network, which we believe is responsible for the alignment of the
porphyrin aggregates, and therefore is lost; (b) diisopropylmethylamine displaces the
chiral conjugate and bind to zinc, which again leads to the loss of ECCD. Therefore,
more time needs to be spent on this subject to verify the nature of the chiral species

formed.

50

— no amine

-——1 eq. at 0 minute
——1 eq. at 5 minutes
——1 eq. at 10 minutes

 

375 425 475
wavelength (nm)

Figure 5-11. Adding diisopropylmethylamine (1eq.) to zinc porphyrin acid with 9F.
5-3. Other chiral substrates for the zinc porphyrin acid aggregates
The zinc porphyrin acid system exhibited ECCD upon addition of carrier F
derivatized (S)-O-acetylmandelic acid 9F, and all the other carrier F derivatized chiral
acids (see Section 3-5) were also tested, however, none of them exhibited a distinctively

clear ECCD signal, except for the (S)-methoxyphenylpropionic acid’s derivative 8F. A

254

 

positive ECCD was observed, however, not as strong as that of the substrate 9F. At the
same time, chiral monoamines and diamines were also tested, and only L-lysine methyl
ester showed a small positive ECCD spectrum with the addition of zinc porphyrin acid.

The structures of SF and L-lysine methyl ester are shown in Figure 5-12.

H OCH3 NHz
\ Njfkph ‘ OCH3
l HZNMW
N’ O 0

8F L-Iysine methyl ester

Figure 5-12. Structures of (S)-methoxypheny1propionic acid derivative 8F and L-lysine
methyl ester.
5-4. Porphyrin aggregates with other porphyrin compounds

As mentioned before, the zinc porphyrin acid can aggregate in
methylcyclohexane, and upon the binding of a chiral complex, the aggregate can adopt a
preferred helicity, which can be observed as ECCD spectrum. In the process, zinc serves
as the binding site to coordinate with the chiral substrate, and the carboxylic functionality
helps the alignment of the porphyrin aggregates by forming hydrogen bonds. It is natural
to wonder if porphyrin compounds with other functionalities, such as alcohols, can also
exhibit the same behavior. In order to test this, we have synthesized two porphyrin

mono-alcohols by derivatizing porphyrin acid with diols.

255

“9'!-

 

 

 

1. 1.5-pentanediol
EDC, DMAP

2. Zn(0Ac)2
8 0

6 /o

  
 

-1

1. ethylene glycol
DC, DMAP

2. Zn(OAc)2
85%

 

Scheme 5-1. The synthesis of porphyrin mono-alcohols 1 and 2.

As shown in Scheme 5-1, the porphyrin acid was derivatized with 1.5-pentanediol
and ethylene glycol, to generate two porphyrin mono-alcohols 1 and 2, respectively. Zinc
was inserted into these two porphyrin alcohols by addition of zinc acetate to the
porphyrin solution. The carrier F derivatized (S)-O-phenylmandelic acid 9F was then
added to the resulting zinc porphyrin mono-alcohols and CD spectra were recorded.
When the chiral complex was added to porphyrin mono-alcohol 1, an asymmetric

positive ECCD peak was observed he,“ 430 nm (As +11) and 422 nm (At: -l7), A +28,

while when it was added to the mono-alcohol 2, the ECCD signals became stronger. As
shown in Figure 5-13, although the ECCD peak for the zinc porphyrin mono-alcohol 2
was not as strong as that of the zinc porphyrin acid, the system was more stable as the

ECCD signal did not decrease with time.

256

30‘

—-60 minute
——- 0 minute

 

375 425 475

wavelength (nm)
Figure 5-13. The time study of zinc porphyrin mono-alcohol 2 with carrier F derivatized
(S)-O-phenylmandelic acid 9F .
Conclusion:

Although we have only carried out preliminary studies on the porphyrin
aggregates, the obtained results are very encouraging. It appears that it is possible to
transfer the Chirality of a small molecule to a zinc-porphyrin aggregate, which can be
observed by ECCD. As a proof of concept, upon addition of derivatized carboxylic acids
8F and 9F as well as L-Lysine methyl ester to a zinc porphyrin carboxylic acid solution,
an ECCD spectrum was observed. However, the system is very dynamic, changing
significantly with time, while it fails to produce positive results with other chiral
compounds. At the same time, it was observed that the acid functionality on the
porphyrin periphery is very critical for the success of the system, since if replaced by the
corresponding methyl ester the ECCD is reduced. We believe that the hydrogen bonding
between the carboxylic acid groups dictates directionality to the aggregate, essentially for

the formation of an ECCD active complex. In an attempt to explore the above

257

 

hypothesis, alcohols were used to direct the aggregate formation and the preliminary
results were very promising (currently further studies were carried out in the lab (Marina
Tanasova) by using polyols as the aggregate directing group).

The porphyrin aggregate system, although promising, has to be further optimized
and proven valid for a wide range of substrates. If consistent results are obtained it could
be potentially used as an alternative to the porphyrin tweezer system having the

advantage of requiring less steps for its synthesis.

258

Experimental materials and general procedures:

Anhydrous CH2C12 was dried over CaHz and distilled. The solvents used for CD
measurements were purchased from Aldrich and were spectra grade. All reactions were
performed in dried glassware under nitrogen. Column chromatography was performed
using SiliCycle silica gel (230-400 mesh). IH—NMR spectra were obtained on Varian
Inova 300 MHz or 500 MHz instrument and are reported in parts per million (ppm)
relative to the solvent resonances (8), with coupling constants (J) in Hertz (Hz). UV-Vis
spectra were recorded on a Perkin—Elmer Lambda 40 spectrophotometer, and is reported
as lmax [nm]. CD spectra were recorded on a J ASCO J -8 10 spectropolarimeter, equipped
with a temperature controller (Neslab l 11) for low temperature studies, and is reported as
A [nm] (mama, [1 mt1 cm"]). Porphyrin acid was synthesized as described in Chapter 2,

and all the other starting materials were purchased from Aldrich.

1. 1.5-pentanediol
EDC, DMAP

2. Zn(OAc)2
8 0

6 /o

   

Porphyrin alcohol 1:

To a solution of porphyrin acid (90 mg, 0.14 mmol) and 1,5-pentanediol (50 mg,
0.5 mmol) in anhydrous CH2C12 (4 mL) was added EDC (27 mg, 0.14 mmol) and DMAP
(18 mg, 0.14 mmol). The reaction mixture was stirred at room temperature overnight and
Zn(OAc)2 (50 mg) was added. The reaction was stirred for another 1 h, and it was

applied directly to silica gel column and purified (CHZClz) to afford the title compound 1

259

(86%). lH NMR (CDCl_;, 300 MHZ) 8 8.76—8.85 (m, 8H), 8.42 (d, J=8.3 HZ, 2H), 8.29
(d, J=8.3 Hz, 2H), 8.19—8.21 (m, 6H), 7.71—7.79 (m, 9H), 4.71 (d, J=6.0 Hz, 2H), 3.82 (d,

J=6.0 Hz, 2H), 2.13 (m, 4H), 1.91 (m, 2H).

1. ethylene glycol
EDC, DMAP

2. Zn(OAc)2
85%

 

   

Porphyrin alcohol 2:

To a solution of porphyrin acid (90 mg, 0.14 mmol) and ethylene glycol (62 mg, 1
mmol) in anhydrous CH2C12 (4 mL) was added EDC (27 mg, 0.14 mmol) and DMAP (18
mg, 0.14 mmol). The reaction mixture was stirred at room temperature overnight and
Zn(OAc)2 (50 mg) was added. The reaction was stirred for another 1 h, and it was
applied directly to silica gel column and purified (CH2C12) to afford the title compound 2
(86%). 1H NMR (CDC13, 300 MHz) 8 8.76—8.85 (m, 8H), 8.42 (d, J=8.3 Hz, 2H), 8.29
(d, J=8.3 Hz, 2H), 8.19—8.21 (m, 6H), 7.71-7.79 (m, 9H), 4.54 (d, J=6.0 Hz, 2H), 4.41 (d,

J=6.0 Hz, 2H).

260

 

Reference:

1.

2.

10.

11.

12.

13.

14.

15.

16.

17.

Engelkamp, H.; Middelbeek, S.; Nolte, R. J. M. Science 1999, 284, 785.

van Nostrum, C. F.; Picken, S. J .; Schouten, A. -J.; Nolte, R. J. M. J. Am. Chem.
Soc. 1995, 117, 9957.

Drager, S. A.; Zangmeister, R. A. P.; Armstrong, N. R.; O'Brien, D. F. J. Am.
Chem. Soc. 2001, 123, 3595.

Abraham, R. J.; Eivazi, F.; Pearson, H.; Smith, K. M. J. Chem. Soc., Chem.
Commun. 1976, 698.

Abraham, R. J.; Fell, S. C.; Pearson, H.; Smith, K. M. Tetrahedron 1979, 35,
1759.

Tamaru, S.; Uchino, S.; Takeuchi, M.; Ikeda, M.; Hatano, T.; Shinka, S.
Tetrahedron Lett. 2002, 43 , 3751.

Luboradzki, R.; Gronwald, O.; Ikeda, M.; Shinka, S.; Reinhoudt, D. N.
Tetrahedron 2000, 56, 9595.

Melendez, R. E.; Carr, A. J.; Linton, B. R.; Hamilton, A. D. Struct. Bond. 2000,
96, 31.

Seto, C. T.; Whitesides, G. M. J. Am. Chem. Soc. 1993, 115, 905.

Leighton, P.; Cowan, J. A.; Abraham, R. J .; Sanders, J. K. M. J. Org. Chem 1988,
53, 733.

Akins, D. L.; Zhu, H. R.; Guo, C. J. Phys. Chem. 1996, 100, 5420.

Maiti, N. C.; Ravikanth, M.; Mazumdar, S.; Perisasmy, N. J. Phys. Chem. 1995,
99, 17192.

Kasha, M. Physical Processes in Radiation Biology; Academic Press: New York,
1964.

Davydov, A. S. Theory of Molecular Excitions; Plenum Press: New York, 1971.
Balaban, T. S.; Eichhofer, A.; Lehn, J. -M. Eur. J. Org. Chem. 2000, 6, 4047.

Balaban, T. S.; Leitich, J.; Holzwarth, A. R.; Schaffner, K. J. Phys. Chem. B
2000, 104, 1362.

Gratzel, M. Nature 2001, 414, 338.

261

-2

 

18.

19.

20.

21.

22.

23.

24.

25.

26.

Brabec, C. J.; Shaheen, S. E.; Fromherz, T.; Padinger, P.; Hummelen, J. C.;
Dhanabalan, A.; Hanssen, R. A. J .; Sariciftci, N. S. Synth. Met. 2001, 121, 1517.

Plater, M. J .; Aiken, S.; Bourhill, G. Tetrahedron 2002, 58, 2405.

Chemla, D. S.; Zyss, J. Nonlinear Optical Properties of Organic Molecules and
Crystals; Academic: Orlando, FL, 1987.

Prasad, P. N.; Williams, D. J. Introduction to Nonlinear Optical Eflects in
Molecules and Polymers; Wiley: New York, 1991.

Burland, D. M.; Miller, R. D.; Walsh, C. A. Chem. Rev. 1994, 94, 31.
Matile, S.; Berova, N.; Nakanish, K. J. Am. Chem. Soc. 1995, 117, 7021.

Faustino, M. A.; Neves, M. G. P. M. S.; Vicente, M. G. H.; Cavaleiro, J. A. S.;
Neumann, M.; Brauer, H. -D.; Jori, G. Photochem. Photobiol. 1997, 66, 405.

Berova, N.; Nakanish, K.; Woody, R. W. Cicular Dichroism: Principles and
Applications, 2nd ed.; John Wiley & Sons, Inc. Publication: New York, 2000;

Huang, X. F.; Rickman, B. H.; Borhan, B.; Berova, N.; Nakanishi, K. J. Am.
Chem. Soc. 1998, 120, 6185.

262

 

 

 

Chapter 6

The attempted synthesis of boronic porphyrin tweezer

Since it has been very well known that boronic acids can form stable complexes
with sugars and diols even in aqueous solution,"3 numerous arylboronic acids have been
prepared and studied as sugar receptors.4‘ 5 However, the simple arylboronic acids form
stable complexes with a variety of sugars and thus they are not useful as specific
receptors or sensors for a specific sugar.

A specific carbohydrate can be recognized by appropriate manipulation of two
boronic acids in the same molecule, with the precise placement of the recognition
element in the proper position and orientation for optimal complementarity to the guest.
Figure 6-1 is an example of a computer—based designed receptor, specifically for
glucopyranose.3 Its high affinity to glucose has been evaluated by fluorescence
experiments, exhibiting a 400-folder greater affinity for glucose than any of the other

sugars.

 

receptor glucose complex

Figure 6-1. A receptor for glucopyranose and its glucose complex.
it has been demonstrated that porphyrins, bearing boronic acid functionalities, can
bind to carbohydrates and the binding could be observed by ECCD spectra (Figure 6-2).6’
7

However, they have high affinities for all the carbohydrates, and are not specific

receptors for a single sugar.

263

 

 

 

R R = 4-C8H17OC6H5

Figure 6-2. The boronic porphyrin and its binding to sugars.
We are interested in designing artificial carbohydrate receptors, which show
selectivity for a specific carbohydrate. Of particular interest are glucose sensors for
potential application in the maintenance of blood glucose levels in patients with

diabetes.8’ 9

In order to achieve this goal, we wanted to design boronic acid based
porphyrin receptors, which upon binding to specific substrates will exhibit ECCD
spectra. In the previous chapters, we have demonstrated the ability of zinc porphyrin
tweezer for the stereochemical detemiination of chiral substrates. An extension of these

is boronic porphyrin tweezers 1 and 2 designed specifically as receptor molecules for

carbohydrates.

B(OH)2

Q I3(0le

 

Figure 6-3. The structure of boronic porphyrin tweezer l and 2.

264

The boronic porphyrin tweezers 1 and 2 can bind to carbohydrates by forming
covalent bonds between the boron on the porphyrin periphery and the oxygen of the
sugars. Since the sugars are all chiral, the binding to the boronic porphyrin tweezer 1 or
2 will induce helicity of the porphyrin host, which can provide fingerprint spectra for the
carbohydrates. The boronic porphyrin tweezers can be synthesized from the para-
substituted porphyrin carboxylic acids by a standard EDC coupling with a diol. The
challenge lies in the synthesis of the para-substituted porphyrin carboxylic acid 3 and 4,
which are proposed to be synthesized by MacDonald “2+2” cyclization method.'0

The synthesis of porphyrin acid 3 relies on introducing the 10-, 20- meso carbon
of the porphyrin as forrnyl group at the 1-, and 9- position of one dipyrromethane 5, and
then condensing it with another dipyrromethane 5, which has no substitution at 1-, and 9-

position, under acidic condition (Scheme 6-1).”

0.3.0

(Q) p-TsOH / MeOH O
B. -

r

 

   

COZH —”
\ / /
\NHHN/
5 0H0 eno O
6
0 OH
3

Scheme 6-1. Proposed synthesis of porphyrin tweezer l by MacDonald “2+2”
cyclization.
As for tweezer 2, since there are two extra phenyl groups at 5- and 15- position,

its “2+2” cyclization synthesis requires the condensation of dipyrromethane-dicarbinol 7,

265

 

instead of dipyrromethane-dialdehyde 6, with dipyrromethane 5. Its proposed synthesis

is illustrated in Scheme 62”

 

Scheme 6-2. Proposed synthesis of porphyrin tweezer 2 by MacDonald “2+2”
cyclization.
6-1. Synthesis of porphyrin tweezer 1

For the synthesis of porphyrin tweezer 1, dipyrromethane 5 could be easily
synthesized by reacting protected 4-boronic benzaldehyde with pyrrole under acidic
condition, as shown in Scheme 6-3. The boronic acid functionality of the 4-boronic

benzaldehyde was protected with 1,3-propanediol12 before it was condensed with excess

266

 

 

amount of distilled pyrrole to form the 5-phenyldipyrromethane 5, by using TFA as the
acidic catalyst. Excess pyrrole is required to maximize formation of dipyrromethane, and

lowering the pyrrole/benzaldehyde ratio favors formation of higher oligomer and

 

polymer.
CH0 9 03.0
L l"
'B‘ TFA,CH2C|2
U 35% \ /
\ NH HN ,
5

 

Scheme 6-3. Synthesis of dipyrromethane 5.

Scheme 6-4 shows the proposed synthesis of dipyrromethane-dialdehyde 6 from
2-pyrrole-carboxylic acid 8. The synthesis of dipyrromethane-dialdehyde 6 started by
using 2-pyrrole-carboxylic acid 8 as the starting material. After it was condensed with 4-
carboxymethylbenzaldehyde 9 to form the dipyrromethane 10, the methyl ester could be
converted to carboxylic acid functionality by hydrolysis, and consequently, the
carboxylate functionality in compound 11 would be reduced to aldehyde to finish the

synthesis of dipyrromethane-dialdehyde 6.'3

267

 

 

 

 

—fl___‘..—.L.'._—. ,—

CHO

 

 

 

H
N cozH 9
U COOMe _
EtOH/HCI
8 reflux H020
CH(OMe)3 / TFA
COOH
\ /
\ NH NH /

OHC 6 CHO

Scheme 6-4. The proposed synthesis of dipyrromethane-dialdehyde 6 from 2-pyrrole-
carboxylic acid 8.

As shown in Scheme 6-4, the 2-pyrrole-carboxylic acid 8 was condensed with 4-
carboxymethylbenzaldehyde 9 in refluxing HCl and ethanol to form dipyrromethane 10.14
However, under this harsh condition, only black polymeric like material was generated
instead of the desired dipyrromethane product 10. Thus the carboxylic acid functionality
in acid 8 had to be protected by forming esters to survive the strong acidic condition for

the synthesis of dipyrromethane 10.

 

H H

N COZH BC” I H2804: N COzEt
U reflux U

8 10% 12

Scheme 6-5. The esterification of 2-pyrrole-carboxylic acid 8.
The conversion of carboxylic acid 8 to ethyl ester 12 was carried out according to
published procedures by refluxing in an acidic solution of ethanol (Scheme 6-5).'5

However, only very small amount of desired product 12 (the yield was about 10%) was

268

‘1

 

generated, with the majority of the isolated product being some black powder, which

probably was a polymer since it did not dissolve in any solvent.

 

 

CHO COOMe
1;] 9
(\ ,7/0025‘ COOMe _
EtOH / HCI
12 reflux

Scheme 6-6. The condensation of ethyl ester 12 with aldehyde 9.

The ethyl ester 12 and 4-carboxymethylbenzaldehyde 9 were then condensed to
generate dipyrromethane 13, by refluxing in an acidic solution of ethanol for 4 hours
(Scheme 6-6). The desired product 13 was afforded as a precipitate.'4 However, the
synthesis of dipyrromethane-dialdehyde 6 as outlined in Scheme 6-4 was not pursued
because of perceived problems. First of all, since compound 11 with three carboxylic
acid functionalities, was going to be synthesized by hydrolysis of compound 13, it would
be very hard to extract it from the water. Secondly, to convert compound 11 to
dipyrromethane-dialdehyde 6, the reduction of the carboxylic acid functionality on the
pyrrole ring could also compromise the carboxylic acid functionality on the phenyl ring.
Since the dipyrromethane-dialdehyde 6 might also be synthesized by using pyrrole as

starting material, we focused on this strategy.

CHO

COOR

COOR

 
    

H 9a R=H
U coon 9" Rzm‘:

POQb/DMF
14 TFA/CH2C12 "

 

 

\ /

\ NH HN
OHC CH0
15a Fi=H 6a R=H

15b F1=Me 6b B=Me

  

 
     

Scheme 6-7. Unsuccessful synthesis of dipyrromethane-dialdehydes 6a and 6b from

pyrrole.

269

As depicted in Scheme 6-7, pyrrole 14 was condensed with 4-
carboxybenzaldehyde 9a and 4-methoxycarbonylbenzaldehyde 9b by using TFA as
acidic catalyst to furnish the dipyrromethane 15a and 15b.16 Compared to the harsh
condition of refluxing in EtOH and HCl, the TFA condition is a much milder way to
generate dipyrromethane and provides higher yields. Unfortunately, this reaction did not
work on 2-substituted pyrrole, such as the 2-pyrrolcarboxylic acid 8 or its corresponding
ethyl ester 12. The Vilsmeier reaction to install formyl groups onto the dipyrromethane
15a and 15b,” '8 failed to provide the desired product 6a or 6b.

The two different schemes (Scheme 6-4 and 6-7) to synthesize dipyrromethane-
dialdehyde 6 had failed, and therefore, at this point, we directed our attention toward the
synthesis of boronic porphyrin tweezer 2.

6-2. Synthesis of boronic porphyrin tweezer 2
6-2.1. Synthesis of boronic porphyrin tweezer 2 by MacDonald “2+2” cyclization

The synthesis of porphyrin tweezer 2 by MacDonald “2+2” cyclization method
requires the condensation of dipyrromethane 5 and dipyrromethane-dicarbinol 7. The
synthesis of dipyrromethane 5 is demonstrated in Scheme 6-3, and the challenge here was
to synthesize the dipyrromethane-dicarbinol 7. The synthesis of boronic porphyrin acid 4
through the condensation of dipyrromethane 5 and dipyrromethane-dicarbinol 7 is shown

in Scheme 6-8.

270

 

_,1

 

 

COOH
EtMgBr
PhCOCl

 

\\ ’Ctngifid
NH HN o

27 /° PhOC

15a

 

COOH

0B 0 BF3 EtZO/NHacl
oflTA '
h\NHHN/P
\NHSHN/

Scheme 6-8. Proposed synthesis of boronic porphyrin acid 4.

 

 

Scheme 6-8 depicts the proposed synthetic pathway to secure the protected
boronic acid 4. The treatment of dipyrromethane 15a with excess amount of EtMgBr and
benzoyl chloride successfully introduced benzoyl groups to the 1- and 9- position of
compound 15a. This process generated both the mono-acylated and di-acylated
dipyrromethane, which were easily separated by column chromatography.“5 According
to literature precedent,l6 the ratio between the mono- and di-acylated products should be
1:5. It was found that prolonging the reaction time increased the yield of the di-acylated
product; even so a low yield of 16 was isolated. The reduction of diacyl dipyrromethane

16 to dicarbinol 7 was achieved with excess NaBHa in THF/methanol.l9

This reduction
reaction proceeded well, however, we were unable to extract product 7 from the reaction

mixture, due to its high polarity.

271

 

 

 

Since the carboxylic acid 7 was too polar, we planned to decrease its polarity by
synthesizing the corresponding methyl ester 17 instead. As shown in Scheme 6-9, the
dipyrromethane methyl ester 15b was used as the starting material. When
dipyrromethane 15b was treated, under the Friedel-Craft reaction condition, with benzoyl

chloride in the presence of AlCl3,'9’ 20

it yielded both mono- and di-acylated products,
which were easily separated by column chromatography. Unfortunately, the diacylated

product that was isolated from reaction mixture was not the desired compound 17, but

byproduct 18, with one substituent at the or-position and the other at B-position.

 

COOMe COOMe COOMe
AIC13/ PhCOCl
CHQCIZ, rt, 4 h +
\ / \ /
\\NHHN// \ NHHN / COP“ \ NHHN /
PhOC PhOC COPh
15b 18 17

Scheme 6-9. Synthesis of dipyrromethane-dicarbinol 17.

The reaction to synthesize symmetrically substituted compound 17 by Friedel-
Crafts condition was repeated several times. In one of the reactions, it was noticed that
on TLC, there was a spot just right above the mono-acylated product. It did not seem to
be the desired di-acylated product, since it lacked the characteristic bright color most the
dipyrromethane compounds exhibit in bromine vapor, and also because according to the
literature,‘9 generally di-acylated products moved slower than mono-acylated products.
However, this spot was isolated, and it was identified by NMR to be the desired
symmetric compound 17.

The yield of the 17 varied with reaction time and the amount of Lewis acid and
benzoyl chloride used in the reaction. It seems like, the longer the reaction time; the

lower the yield of the desired symmetrically substituted product 17 , while the yield of the

272

 

 

asymmetrically substituted byproduct 18 was increasing. The yield of compound 17

ranged from 15% to 40%. We had also tried to introduce the two benzoyl groups of 17 in
a step-wise fashion, however, the reaction stopped at the mono-acylated product stage.
We tried different conditions, but we could not add the second benzoyl group to the
mono-acylated compound, and at the end of the reactions, the mono-acylated starting
material was recovered.

After the symmetrically diacylated dipyrromethane 17 was made, it was ready to
condense with dipyrromethane 5 to synthesize the para-substituted porphyrin 4. Since
the final target of our synthesis was the porphyrin tweezer 2, there were two pathways to
synthesize it by using the diacylated dipyrromethane 17 as the starting material (Scheme
6-10): pathway I was to synthesize a porphyrin carboxylic acid 4 first, and then form the
linkage between the two porphyrins by simple esterification; pathway II would introduce
the linkage between two of the dipyrromethane 17 first, and then build two porphyrins on

the dipyrromethane dimer 19 in a single step.

273

 

 

    
  
 
 

pathway l

/

pathway II

\\NH HN//
PhOC
Scheme 6-10. Two pathways for the synthesis of porphyrin tweezer 2.

To synthesize porphyrin tweezer 2, pathway II was attempted first, which
required the synthesis of the dipyrromethane dimer 19. With dipyrromethane carboxylic
acid 16 in hand, esterification with 1,5-pentanediol utilizing EDC as the coupling reagent
should afford the dipyrromethane dimer 19. However, the standard esterification reaction
failed, there was no formation of either the diester 19, or the monoester (Scheme 6-11).
This might be due to the impurities in the carboxylic acid 16, since the purification of the
acid 16 by column chromatography was not very effective, and it had to be precipitated

out of the column fractions to get purer product.

274

 
  
 

1 ,5-p\e/ntanediol

EDC/{DMAP

 

 

Scheme 6-11. The unsuccessful EDC coupling of compound 16 and 1,5-pentanediol.

In order to obtain pure carboxylic acid 16, its corresponding methyl ester 17 was
hydrolyzed under basic condition, as shown in Scheme 6-12. The routine hydrolysis, by
refluxing the dipyrromethane methyl ester 17 in a mixture of THF and aqueous sodium
hydroxide solution overnight, provided a very polar compound as the only product. This
product was so polar that it did not move on the TLC plate by even using 10% MeOH in
CHzClz as the developing solvent. However, it was demonstrated not to be the desired
carboxylic acid 16 by NMR. This hydrolysis reaction was repeated several times, and

each time same product was observed.

COOMe
NaOH / THF

   

Scheme 6-12. The unsuccessful hydrolysis of methyl ester 17 to acid 16.

Since the attempt of securing dipyrromethane carboxylic acid 16 failed, we turned
to an alternate way to synthesize the dipyrromethane dimer 19. As shown in Scheme 6-
13, a simpler dipyrromethane dimer 20 was synthesized first, and then the benzoyl groups
would be introduced in the a-position of pyrroles of compound 20 by Friedel-Crafts

reaction. However, this was a risky plan, since installing four benzoyl groups in one

275

 

single reaction could result in many byproducts and the yield of the desired product 19
was sure to be low. Having that in mind, this method was still worth trying because the

starting material 15a was very easy to synthesize.

COOH
EDC, DMAP

CH2C|2 T

\

 
  
  

 

\\NHHN//
15a
Scheme 6-13. Alternate way to synthesize dimer 19.
As shown in Scheme 6-14, the dipyrromethane dimer 20 was easily synthesized
from the dipyrromethane carboxylic acid 15a by esterifying with 1,5-pentanediol under

standard EDC reaction condition. Synthesizing dimer 20 stepwise from carboxylic acid

15a could improve the yield of the dimer 20.

COOH

1 ,5-pentanediol

EDC, DMAP 7
CH2C|2 35%

 

         

\ /
15a 20 \ N” ”N /
Scheme 6-14. The synthesis of dimer 20.

As shown in Scheme 6-15, the isolated dimer 20 was then treated under Friedel-
Crafts reaction condition to install the benzoyl groups. Although the Friedel-Crafts
reaction was successfully applied to introduce benzoyl groups with the dipyrromethane

methyl ester 15b, it did not work for the dimer 20. This reaction was repeated a couple of

times, and in all the reactions starting material 20 was recovered.

276

 

 

Scheme 6-15. The unsuccessful Friedel-Crafts reaction to synthesize dimer 19.
Since we were unable to introduce four benzoyl groups with dimer 20, we turned
back to the original idea of installing benzoyl groups first, and then forming dimer 19.
As is illustrated in Scheme 6-11, the EDC coupling reaction of carboxylic acid 16 with
1,5-pentanediol failed to provide compound 19, and methyl ester 17 could not be
hydrolyzed to provide the carboxylic acid 16 (Scheme 6-12), an alternate method was
needed to form the dimer 19 directly from the methyl ester 17, without forming the

carboxylic acid 16 first.

COOMe

1 ,5-pentanediol

r

 

 

\ / TsOH, 120°C,
\ NH HN / overnight
PhOC COPh 70% PhOC
17

Scheme 6-16. The first attempt to synthesize compound 19 from methyl ester 17.

As shown in Scheme 6-16, we planned to synthesize the monoester 21 from the
methyl ester 17 by transesterificating with 1,5-pentanediol first, and then transform the
monoester 21 to the diester 19 by reacting with equal amount of methyl ester 17 in
presence of a base, such as sodium hydride. The monoester 21 was easily synthesized
from methyl ester 17 via transesterification, by heating in 1,5-pentanediol in the presence

of TsOH as the acidic catalyst. But once again, the attempt to synthesize the diester 19

277

 

from the monoester 21 was unsuccessful. Even after stirring at room temperature
overnight, the reaction did proceed, and heating it to reflux resulted in the decomposition
of starting material 21.

According to the literature, methyl esters can be directly transformed to diesters
using sodium metal, as shown in Scheme 617.” Although the stoichiometry of the
reaction calls for half equivalent of the diol, since the reaction was performed in a large
scale, the diol was used as a solvent. In our case, the working scale did not allow for the
use of 1,5-pentanediol as a solvent. Therefore, THF was chosen to serve as the solvent.
Treatment of methyl ester 17 with sodium metal and half equivalent of 1,5-pentanediol in
THF generated the desired diester 19 in 30% yield. The addition of THF drastically
decreased the reaction rate and the yield of the desired product 19 was much worse than
the reported.

COOMe

1 ,5-pentanediol

/ Na, THF

\
\ NH HN / 68 °C 2 h
PhOC COPh 30% PhOC

17

 

Scheme 6-17. Directly synthesis of diester 19 from methyl ester 17.

With compound 19 in hand, it seemed we were not far away from securing the
final porphyrin tweezer 2. It only required the MacDonald “2+2” cyclization reaction to
build two porphyrin planes by condensing compound 19 with dipyrromethane 5. Before
the synthesis of the porphyrin tweezer 2 with compound 19, the MacDonald “2+2”

cyclization reaction was tested with a model compound 17.

278

As shown in Scheme 6-18, the diketone 17 was reduced with NaBH4 to a diol,
affording the starting material for MacDonald “2+2” porphyrin synthesis.lo Since the
diol intermediate was unstable, it had to be used immediately for the following “2+2”
cyclization. The coupling of the diol and dipyrromethane 5 was achieved with TFA as

acidic catalyst to yield porphyrin products in 40% yield.

 

 

 

 

COOMe l“ COOMe 7 1. TFA, CH3CN
NaBH4 '
TFA/MeoH Q ’0
- \ / B
\ / \ NH HN
\ N“ ”N / Ph Ph
PhOC 17 COPh OH Ho
_ T \ /
\NHHN/ 5
2. DDQ 40%

 

O OMe

Scheme 6-18. The MacDonald “2+2” cyclization of compound 5 and 17.

The purple color of the isolated product suggested that the desired porphyrin
compound was generated in this cyclization reaction. From NMR spectrum, we could
clearly observe the proton peaks characteristic of porphyrins, however, there were no
signals belonging to the boronic acid protecting group. It was shown from NMR and MS

studies that what we separated was a mixture of desired porphyrin 22 and the reduced

279

porphyrin 23, which suggested that the boronic acid functionality was removed and lost
during the reaction.

To better understand the fate of the boron, the reaction shown in Scheme 6-18
was repeated and the crude product was analyzed by MS and llB-NMR. The MS
analysis revealed the presence of the molecular ion peaks for both the desired boronated
porphyrin 22 and the de-boronated porphyrin 23, with a ratio of about 2.5:]. llB-NMR of
the reaction mixture exhibited a small boron peak. A small portion of the reaction
mixture was subjected to a quick column to isolate some porphyrin product. MS analysis
of the isolated porphyrin indicated the presence of both the desired boronated porphyrin
22 and the de-boronated porphyrin 23, however, the ratio of the boronated porphyrin 22
to the de-boronated porphyrin 23 decreased from 2.5:1 before column to 1:2 after
column. The rest of crude mixture was stored in the freezer overnight, and MS was
measured. This time, the ratio of the boronated porphyrin 22 and the de-boronated
porphyrin 23 decreased from 2.5 :1 to l: 1.

The comparison of these MS data makes it obvious that the 1,3-propanediol
protected boronic functionality is unstable, and the Caryl-B bond in porphyrins is cleaved
during the acidic MacDonald “2+2” cyclization reaction conditions and on silica column.
In order to improve its stability, the boronic acid should be protected with a bulkier diol,

such as 2,2-dimethyl-1,3-propanediol.

280

  
  

2. TFA / CH3CN

/\

 

1. NaBH,, O (W
TFA / MeOH 0‘30

/
Ph / Ph

0

0 0
Scheme 6-19. The attempted synthesis of boronic porphyrin tweezer 2.

The same reaction was attempted with dimer 19 as shown in Scheme 6-19. The
MacDonald “2+2” cyclization reaction of the diester 19 was not as clean as that with
methyl ester 17, and provided at least three separable porphyrin products. Since the three
compounds had very similar NMR spectra, it was hard to analyze the products. When
submitted for FAB-MS, however, none of them were found to have the right molecular
weight of the desired porphyrin tweezer 2. The biggest problem with the MacDonald
“2+2” cyclization reaction of diester 19 was that there were many porphyrin byproducts
generated in the process, the separation of which was tedious.

From the unsuccessful attempt of synthesizing porphyrin tweezer 2 from
dipyrromethane diester 19 (Scheme 6-19), it was quite clear that it was not an efficient
method to build two porphyrin rings in one reaction. 80 for the synthesis of porphyrin
tweezer 2, we investigated pathway 1, shown in Scheme 6-10, to synthesize the boronic
porphyrin acid 4 first, and then couple it with 1,5-pentanediol to yield porphyrin tweezer
2 by EDC coupling reaction. As stated above, the 1,3-pentanediol was too labile, so we

chose a bulkier group, 2,2-dimethyl-l,3-propanediol, to protect the boronic acid.

281

 

1 . NaBH4
COOMe TFA / MeOH
2. TFA / CH3CN

   

 

\ /
\ NH HN / 26
3. BBQ 20%

0 OMe
Scheme 6-20. The synthesis of boronic porphyrin methyl ester 24.

Scheme 6-20 shows the synthesis of 2,2-dimethyl-1,3-propanediol protected
porphyrin methyl ester 24. With the newly protected boronic benzaldehyde 25 in hand, it
was condensed with pyrrole to synthesize the dipyrromethane 26, which subsequently
underwent a MacDonald “2+2” cyclization with dipyrromethane-dicarbinol 17 to form
the porphyrin methyl ester 24. This time, the desired boronic porphyrin methyl ester 24
was isolated from the reaction mixture by column chromatography, and was observed by
MS.

In order to synthesize the porphyrin tweezer, the porphyrin methyl ester 24 needs
to be converted to its corresponding carboxylic acid 27 . However, the porphyrin methyl
ester could not be converted to the carboxylic acid by standard reaction conditions, such
as refluxing in the mixture of NaOH aqueous solution and THF or acidic hydrolysis,
since the boron protecting group could not survive. As shown in Scheme 6-21, LiI and

DMF could be used as mild conditions specifically for the removal of methyl esterszz’ 23

282

The porphyrin methyl ester 24 was refluxed in a mixture of LiI and DMF for two hours,
and NMR of the reaction mixture was taken. Based on NMR, the deboronated product

was once again observed.

Lil / DMF

reflux

 

24

Scheme 6-21. The unsuccessful removal of methyl ester of compound 24.

Since the attempt to transform the porphyrin methyl ester 24 to carboxylic acid 27
failed, the porphyrin tweezer 2 had to be synthesized from the methyl ester 24 directly
instead of by transesterification. A transesterification reaction had been used in Scheme
6-17 to transfer a methyl ester 17 to a diester 19.21 The same condition was used to
synthesize porphyrin tweezer 2 from methyl ester 24, as shown in Scheme 6-22.
However, after the reaction mixture was heated for a day, most the starting material 24
still remained in the reaction. Although a product was isolated by column

chromatography, from NMR analysis, it was not the desired porphyrin tweezer 2.

283

 

Scheme 6-22. The attempted synthesis of porphyrin tweezer 2.

6-2.2. Synthesis of boronic porphyrin tweezer 2 by Pd catalyzed coupling reaction
We have tried to synthesize boronic porphyrin tweezer by MacDonald “2+2”
cyclization method; however, the boronic functionality is too labile to be carried through
in the porphyrin tweezer synthesis. Although in most cases, the carbon-boron bond is not
very unstable, in our porphyrin system, the Caryl-B bond becomes very unstable.
Therefore, it is best to introduce the boronic functionality at the end of the synthesis. As
shown in Scheme 6-23, the boronic functionality can be introduced by palladium

coupling of boronic compounds, such as pinacolborane, with aryl bromide, iodide and

R H—B’O 8%
(D O
PdC 2(PPh3)2 ©
R=Br, OTf, |

Scheme 6-23. Pd catalyzed coupling of pinacolborane with aryl halides.

inflates.” 25

 

284

1. H'B. :
0

Ph
Pd catalyst, Et3N
dichloroethane

‘
7

Ph 2. deprotection

 

Fl: Br, I, or OTf

 

O

Scheme 6-24. The synthesis of porphyrin tweezer 2 by Pd coupling reaction.
Scheme 6-24 depicts the strategy of introducing the boronic acid functionality at
the end of the synthesis. The synthesis of bromoporphyrin tweezer 28 requires securing
the para-substituted bromoporphyrins. As shown in Scheme 6—25, Lindsey’s method was
used to synthesize the para-substituted bromoporphyrin alcohol 29,26 by condensing 4-
hydroxybenzaldehyde, 4-bromobenzy1aldehyde, and phenyldipyrromethane 3027 in

CH3C1 at 1:1:2 ratio, with BF3-Et20 as the acidic catalyst.

CHO+ Q TFA,CH2C12
49% \ /
\ NH HN / B,

 

 

Scheme 6-25. The synthesis of p-bromoporphyrin alcohol 29.

285

This reaction provided a mixture of di-hydroxyl, di-bromo, and the desired
bromoporphyrin alcohol 29, and correspondingly, the yield of the desired product 29 was
not high. The separation of these three porphyrin products was easy enough, because the
polarities of bromide and hydroxyl functionality were very different. The alcohol
functionality in porphyrin 29 could be used to generate a tweezer by forming either an
ether bond or an ester bond. Because of the availability of 1,3-dibromopropane, the
porphyrin alcohol 29 was reacted with 1,3-dibromopropane under basic condition to form
dibromo tweezer 31. The attempted reaction between porphyrin alcohol 29 and 1,3-
dibromopropane is shown in Scheme 6-26. The product of the reaction, even though a

porphyrin like compound, was not analyzed further.

 

Scheme 6-26. The attempted synthesis of dibromo tweezer 31.

Since the synthesis of bromoporphyrin tweezer via an ether bond failed, we
planned to synthesize the bromoporphyrin 32, containing a carboxylic acid functionality

to react with 1,5-pentanediol by standard EDC facilitated esterification.

286

 

 

J

1. BF3'E20, CH3CI
x e
2. DDQ

 

   

Scheme 6-27. The unsuccessful synthesis of bromoporphyrin carboxylic acid 32.

The synthesis of bromoporphyrin carboxylic acid 32 is shown in Scheme 6-27. It
was synthesized by the same method as porphyrin alcohol 29, condensing 4-
carboxybenzaldehyde, 4-bromobenzylaldehyde, and phenyldipyrromethane 30 in CH3C1
(1:1:2 ratio) with BF3'E120 as the acidic catalyst. This reaction provided a mixture with
the unwanted di-bromoporphyrin as the major product. The generation of the desired
porphyrin carboxylic acid 32 could be observed by TLC plate, however, it was hard to
purify because of its high polarity. Since the carboxylic acid 32 was hard to synthesize
directly, the corresponding methyl ester 33 was synthesized, which could be subsequently

converted to the acid 32 by standard hydrolysis under basic conditions.

1. BFa-EZO, CH3CI
2. DDQ10°/o 7

 

   

Scheme 6-28. The synthesis of porphyrin methyl ester 33.

287

"1

 

 

As shown in Scheme 6-28, porphyrin methyl ester 33 was synthesized by
condensing 4-methoxycarbonylbenzaldehyde, 4-bromobenza1dehyde and 5-
phenyldipyrromethane 30 in CHCl3 with BF3-EtzO as the catalyst. This reaction provided
a mixture of porphyrin products with the desired compound 33 having the medium
polarity. The ratio between the dibromo porphyrin, compound 33 and diester porphyrin

was about 2:421.

NaOH ITHF
reflux, overnight

99%

 

 

Scheme 6-29. Hydrolysis of methyl ester 33 to acid 32.
The methyl ester 33 was then hydrolyzed under basic conditions (Scheme 6-29).
The resulting carboxylic 32 dissolved quite well in organic solvents, such as CHzClz and
CH3C1. After carboxylic acid 31 was extracted, it was dried and used without further

purification for the esterification reaction to synthesize the dibromo tweezer 34.

288

1 ,5-pentandiol
EDC, DMAP

 

Ph Ph

 

35 major 34 minor 5% 0

Scheme 6-30. EDC coupling of acid 32 with 1,5-pentanediol.

The esterification of the porphyrin acid 32 with 1,5-pentanediol generated the
dibromo porphyrin tweezer 34, but not as well as expected (Scheme 6-30). Surprisingly,
the major product of the reaction was the porphyrin ethyl ester 35. The only step that
could have introduced a trace of ethanol into the reaction was during the synthesis of the
porphyrin methyl ester 33. In that process, chloroform was used as the solvent, which
could have been contaminated with ethanol. However, the porphyrin methyl ester 33 was
purified by column chromatography, and the small amount of ethanol in chloroform
should have been removed by the column. So the source of ethanol still remains a
mystery.

Having at hand the dibromo tweezer 34, the palladium coupling reaction could be
probed. However, before undergoing the palladium coupling reaction, the porphyrin
tweezer free base 34 was converted to its corresponding zinc porphyrin tweezer, since the
palladium could be inserted, if the porphyrin free base was utilized. Following

established procedures, zinc ion was introduced by adding zinc acetate to compound 34

289

in chloroform. The zinc porphyrin tweezer was purified by column chromatography

before the palladium coupling reaction.

1. 2n(0Ac)2 / CH301

2

' o

H_B. :é Ph /
Ph 0

Pd(PPh3)4 Ph
Et3N 0
Cl’ ‘01

5% O O

    

PinacoI-2 O 0
Scheme 6-31. The Pd catalyzed coupling of dibromo tweezer.

As shown in Scheme 6—31, the palladium catalyzed coupling reaction between 34
and pinacolborane28 completed in four hours as monitored by TLC. It provided at least
four different kinds of porphyrin products, which were separated by column
chromatography. These porphyrin products were analyzed by FAB-MS, and one of the
four products seemed to be the desired porphyrin tweezer Pinacol-2. Therefore, the
palladium catalyzed coupling reaction is a reasonable method to synthesize boronic
porphyrin tweezer. In order to optimize the reaction condition to improve the yield of the
desired coupled product, modeling studies were performed with the bromoporphyrin
methyl ester 33.

Bromoporphyrin methyl ester free acid 33 was converted to its zinc porphyrin,
and then coupled with pinacolborane by using PdC12(PPh3)3 as the catalyst. The addition

of all the starting material to the Schlenk reaction flask was conducted in a dry box, since

290

the palladium catalyst is very sensitive to moisture and oxygen. The reaction mixture

was heated to 90 °C overnight before all of the starting material 33 was consumed.

1. Zn(OAc)2 / CHacl

2.
Ph H-Biofi
0 Ph

PdCl2(PPh3)2
13N
CFC!

   

Scheme 6-32. The coupling reaction of porphyrin methyl ester 33.

As shown in Scheme 6-32, this coupling reaction provided two products, with the
major one having very similar polarity as the starting porphyrin methyl ester 33. The two
products were separated by column chromatography, and based on MS, the major product
was suggested to be the reduced product 23, and the minor product as the desired product
36. The ratio between these two products was 10:1, and increasing the amount of
triethylamine did not affect the ratio. According to known mechanism, the new hydrogen
in the reduced product 23, was transferred from pinacolborane. The formation of the
reduced product could potentially be eliminated if the hydrogen was removed from the
boron species.24

Bis(pinacolato)diboron is another commonly used coupling reagent for palladium
coupling reaction. Several reaction conditions were tried to couple this boron compound
to bromoporphyrin 33,29‘ 30 but unfortunately, porphyrin 36 was not isolated (Scheme 6-

33).

1311012819902
DMSO, KOAC
90 °C,overnight

   

O OMe

Scheme 6-33. The unsuccessful coupling with bis(pinacolato)diboron.

It is understood that for the palladium catalyzed coupling reactions, electron rich
aryl halides systems proceed more smoothly than electron poor system.31 In our
porphyrin synthesis, the presence of Zn(H) makes this systems more electron deficient as
compared to the porphyrin free base. Zinc was introduced to hinder any insertion of
palladium into the porphyrin. However, as it turns out palladium insertion was not
observed in the coupling step, and thus the conversion of porphyrin free base to zinc
porphyrin was not necessary. Therefore, for the rest of the coupling reactions, porphyrins
were used as porphyrin free bases instead of metalloporphyrins.

The studies with bromoporphyrins show that the bromoporphyrin system is not a
good coupling partner for the palladium catalyzed reaction. Since aryl triflates and aryl
iodides are better coupling reagents than aryl bromides, they become our next targets.” 33
The reactions of aryl triflates and iodides require lower reaction temperature, the rate of
the reaction is faster, while the ratio between the desired coupling product and the

reduced by—product is expected to improve.

292

 

 

   
    

 

 

 

 

CH0 CH0
H OTf
OH OTf N
Tf20. Py (2 eq.) U
(3112ch 790/0 TFA. CHZCIZ \ /
CHO CHO 51% \ NH NH
37 38
NaBH4 Tl
TFA/ MeOH
= Ph
\ /
\ / \
\ NH HN Ph N” H” Ph \ /
17 ‘
38
2. DDQ 10%

 

Scheme 6-34. Synthesis of triflate porphyrin methyl ester 39.

As shown in Scheme 6-34, the synthesis of the triflate porphyrin methyl ester 39
started with the conversion of the hydroxyl group to the aryl triflate 37. When pyridine
was used as solvent, no product was generated.34 When less amount of pyridine was
applied (2 equivalents) and dichloromethane was used as the solvent,34 this reaction
yielded the aryl triflate 37 as the sole product after 30 h. The aryl triflate 37 was
condensed with excess pyrrole in acidic condition to yield the dipyrrolemethane 38,
which was reacted with dipyrromethane-dicarbinol 17 by MacDonald “2+2” cyclization
to synthesize the triflate porphyrin methyl ester 39. At the same time, the de-triflated
porphyrin was also isolated from the MacDonald “2+2” cyclization. Triflate porphyrin
methyl ester 39 was then coupled with pinacolborane under routine palladium conditions
(Scheme 6-35). After heating at 90 °C in a Schlenk flask overnight, the formation of the

desired product 36 was not observed, and only the starting material 39 was recovered.

293

/ . .
N
ll

Pd(PPh;)\4
CHZCIZ, Et3N,
90°C, overnight

 

36

Scheme 6-35. The attempted Pd coupling of triflate 39.

Since the triflate porphyrin methyl ester 39 failed to couple with pinacolborane
under the palladium coupling condition, iodoporphyrins 41 was synthesized. The
synthesis Of iodoporphyrin methyl ester 41 was very similar to the synthesis of
bromoporphyrin methyl ester 33 (Scheme 6-36). It started with the conversion of 4-
formylphenylboronic acid to 4-iodobenzaldehyde 40.35 The boronic acid functionality
was converted to iodide by refluxing with NIS in dry acetonitrile, and the 4-
iodobenzaldhyde 40 was then condensed with methyl 4-formylbenzoate and 5-
phenyldipyrromethane 30 by using BF3-Et20 as catalyst to synthesize the iodoporphyrin

methyl ester 41.

294

B(OH)2

INS / CH30N ©
reflux 55%

I
HO CHO 4°

 

C

1. BF3.E20,
CH3CI

2. DDQ14°/or

 

   

Scheme 6-36. Synthesis of iodoporphyrin methyl ester 41.
Several reactions were tested to find the optimal palladium catalyzed reaction
conditions for coupling iodoporphyrin methyl ester 41 with pinacolborane. The results
are shown in Table 6-1.

Table 6-1. The palladium catalyzed coupling of iodoporphyrin 41 with pinacolborane.

 

Palladium Catalystv Ph
Dichloroethane
Et3N, overnight

 

 

Palladium Catalyst Temperature (°CL Ratio (36:23)

 

PdCl2(PPh3)2 140 1:1
PdCl2(PPh3)2 90 128
Pd(PPh3)4 140 1 :7

 

295

As shown in Table 6-1, the highest yield and the best ratio between the desired
coupling product 36 and the reduced product 23 was achieved by using PdC12(PPh3)2 as
the catalyst, and the reaction temperature at 140 °C. In this Optimal condition, the yield
of the desired boronic porphyrin 36 was 30%, and the ratio between the desired porphyrin
36 and the reduced porphyrin 23 was nearly 1:1. With PdC12(PPh3)2 as the catalyst,
lowering the temperature from 140 0C to 90 °C, drastically reduced the yield, and the
ratio between the desired porphyrin 36 and the reduced product 23 was 1:8. When
Pd(PPh3)4 was used as the catalyst, the yield of the desired product 36 decreased, and the
ratio between the desired porphyrin 36 and reduce product 23 was roughly 1:7. It is hard
to discriminate between the iodoporphyrin 41, the boronic porphyrin 36 and the reduced
porphyrin 23 based on NMR, since they have very similar NMR spectra. Mass Spectrum
is a better way to determine the formation of the products. The porphyrin compounds
have strong molecular ion peaks, and the formation of the product can be confirmed by
comparing the molecular ion mass with the calculated molecular weight.

After optimization of the palladium catalyzed coupling reaction between
iodoporphyrins and pinacolborane, two things were left before the synthesis of the final
boronic porphyrin tweezer 2 could be achieved. The first thing was the synthesis of the
iodoporphyrin tweezer 42. Since, the Caryl-B bond was very easy to break, it was better
to form the diiodoporphyrin tweezer 42 first, and then couple with pinacolborane to
afford the boron protected tweezer Pinacol-Z. The second thing was to find a mild and
efficient way to deprotect the boronic acid. The challenge lied in finding reaction
conditions to selectively deprotect the boronic acid functionality in the presence of

another ester moiety.

296

The first task of synthesizing the diiodoporphyrin tweezer 42 was straightforward.
As shown in 6-37, after the iodoporphyrin methyl ester 41 was hydrolyzed by refluxing
in the mixture of THF and aqueous sodium hydroxide solution, the resulting pure
porphyrin acid 43 was then esterfied with 1,5-pentanediol by EDC coupling reaction to

provide the diiodoporphyrin tweezer 42.

 

Scheme 6-37. The synthesis of diiodoporphyrin tweezer 42.

As shown in Scheme 6-38, the diiodoporphyrin tweezer was coupled to
pinacolborane by palladium catalyzed coupling reaction. Coupling the diiodoporphyrin
tweezer 42 and pinacolborane, with PdC12(PPh3)2 as the catalyst at 140 °C, provided the
fully reduced porphyrin, half reduced porphyrin and the desired diboron porphyrin
tweezer Pinacol-2. The separation of these three products was not hard, with the desired
porphyrin tweezer Pinacol-2 being the least polar on TLC. From NMR, there was no
difference between these compounds in the aromatic region, and only by FAB-MS, the
formation of the desired compound could be confirmed. Coupling porphyrin tweezer 42
and pinacolborane with PdC12(PPh3)2 as the catalyst successfully provided the desired

boronic porphyrin tweezer pinacol-Z, but the yield was quite poor. However, in our

297

‘Mtnr ~‘ ~-

study, since only very small amount of boronic porphyrin tweezer is needed, the poor

yield was acceptable.

O o
Ph Ph :E .B-H

 

 

0 = Ph Ph
/ Ph Ph FdCI2(PPh3)2
O Dichloroethane Ph
/ Et3N, 140°C /

O O ovemlght /
‘\_\_\ 23% O O O

O _\_\_\

42 O pinacol-2 O
0

Scheme 6-38. The Pd coupling of diiodoporphyrin tweezer 42 to synthesize porphyrin
tweezer pinacol-2.

After the synthesis of protected boronic porphyrin tweezer pinacol-2, the only
task left was to free the boronic acid from its protection. Some studies were performed to

search for the conditions that were strong enough to remove the pinacol protecting group

but mild to preserve the labile Caryl-B bond.

6-2.3. Hydrolysis of Pinacolborane

Most of the boronic esters are quite susceptible to hydrolysis. In aqueous
solution, the ester could easily fall off at a wide range of pH values. Some boronates can
even fall off on silica gel column because of its acidity. However, as the steric demands
of the protecting diol increase, the rate of hydrolysis decreases. Pinacole, which is a
bulky protecting group, is rather difficult to hydrolyze. Generally, there are three ways to

hydrolyze pinacol boronate. Stirring the boronates in acidic or basic aqueous solution is

298

the most common way to hydrolyze any boronic ester, including pinacol boronate. The
second is reducing the boronates with lithium aluminum hydride in THF, and the third is

oxidizing the boronates with sodium periodate in a mixture of acetone and water.

   

NaOH (aq.), THF, rt

O or TFA/0112012, l1;
10% H2804 / THF, n

 

: no reaction

0

  
 

36
C02 Me

Scheme 6-39. The unsuccessful hydrolysis of boronic porphyrin 36.

The first attempt to deprotect the pinacol boronate was by hydrolysis, as shown in
Scheme 6-39, in both basic and acidic condition. After the pinacol boronate porphyrin
methyl ester 36 was stirred in the mixture of 2 N sodium hydroxide aqueous solution and
THF (1:2) for 2 days, there was no sign of the free boronic acid product. The formation
of porphyrin boronic acid could be easily monitored by TLC, since the product should be
more polar than the starting pinacol boronate, and the lack of purple spot at baseline
excluded the formation of porphyrin boronic acid in the reaction. The hydrolysis was
also tried in a variety of acidic conditions, including trifluoroacetic acid in
dichloromethane, and 10% aqueous sulfuric acid in THF.36 The reactions were kept at
room temperature for 2 days; however, no free boronic acid product was observed for
either reaction. Refluxing is very common for hydrolysis of boronates in aqueous media;

however, the porphyrin boronate 36 could not be heated under acidic or basic aqueous

299

conditions, due to the presence of a methyl ester, which would have been readily
hydrolyzed.

Coutts and coworkers have published on cleaving pinanediol boronate esters to
yield the free boronic acids.37 Pinanediol is a very bulky group, and the cleavage of the
pinanediol group from boronate esters is known to be difficult. As shown in Scheme 6-
40, they have developed two mild techniques to successfully remove pinanediol from the
boronate. When comparing the structures of pinanediol and pinacol, it is quite obvious

that the pinanediol is even bulkier than pinacol, and correspondingly, it should be harder

 

 

to cleave.
MethodA O
860% 1 N 10 B HN B<OH)2 + /%
BocHN - a 4 00
a 0 2. H01 F5, 0 OHC
R
E} , . Method B 9‘
8. g . 3(0le
O : + Ph-Br
300““, o 1.Ph-B(OH)2 ”C'HzN 5 O 0
B 2. HCI

Scheme 6-40. The cleavage of pinanediol boronate to boronic acids.

As shown in Scheme 6-40, there are two methods to get free boronic acids from
of pinanediol boronate esters. Method A relies on the oxidative cleavage of the
pinanediol by using sodium periodate. Although boronic acids could be oxidized by
nucleophilic oxidants, such as alkaline hydrogen peroxide, periodate is an eletrophilic
oxidant, and it could not oxidize free boronic acid. This method is compatible even with
Boc protected dipeptides, and it should be compatible with the methyl ester functionality
as well.

Method B is based on the transesterification with phenylboronic acid and the

removal of the pinanediol as the phenylboronate ester. It is particularly applicable to

300

water soluble boronates. In this process, a biphasic system is used, and the starting
pinanediol ester and the free boronic acids, both the starting phenylboronic acid and the
resulting boronic acid product, are dissolved in the aqueous phase with the pinanediol
phenylboronate ester in the organic phase. This procedure allows for the easy recovery
of the valuable pinanediols from the organic layer. However, our target molecule, the
boronic porphyrin methyl ester 36, is not water soluble, and Method B was not applicable
to this system. Method A was used for the deprotection of compound 36 to form the free

porphyrin boronic acid.

NalO4, NH4OAC‘ Too many spots in TLC.

' hard to tell which one is
acetone, water the desired free boronic
rt, 48 h acid

 

 

Scheme 6-41. The attempted oxidation of porphyrin methyl ester 36.
According to the reference,37 the solvent for this oxidation reaction is a mixture of
acetone and water at 1:1 ratio, but since porphyrins do not dissolve in this mixture, the
amount of acetone used in the solvent was increased. As shown in Scheme 6-41, after
this mixture was stirred at room temperature for 48 h, three bands could be observed from
TLC plates. The first band was the unreacted porphyrin pinacol boronate 36. The second
and the third band each contained two bands, and these two bands were very close to

each other. Since pure product could not be isolated from these two bands, it was

301

impossible to identify them. This reaction was repeated several times, and the results
were the same.

Since the porphyrin methyl ester 36 is a quite complex molecule, a simpler
pinacol boronate 44 was chosen as the model compound to find the best reaction
condition for the removal of pinacol from the protected boronates. As shown in Scheme
6-42, different reaction conditions were applied to this model compound 44. When
sodium peroidate served as the oxidant and NH40AC as the base, after two days of
stirring in room temperature, the desired boronic acid product was generated with some
unreacted starting material 44. When NHaOAC was not used in the reaction, the reaction
rate significantly slowed down, and there was more starting material 44 left unreacted.
When other oxidants were used, such as oxone and hydrogen peroxide, the carbon boron

bond was cleaved, and the boronate 44 was oxidized to 3,5-dimethylphenol.

0.840 Nal04, NH.,OAc HOB'OH

Q acetone, water m
I1, 48 h

 

 

44
oxone (or H202),
O'B’O NH4OAC OH
acetone, watef Q
rt, 6 h
44

Scheme 6-42. The oxidation studies on model molecule 44.
The studies with 3,5-dimethylphenyl pinacolboronate 44 (Scheme 6-42) showed
that the oxidation with sodium periodate could afford free boronic acid from
pinacolboronate. However, this method did not work for the porphyrin pinacolboronates.

Then the only method left to deprotect the pinacol boronates and form free boronic acid

302

was by hydrolysis. The hydrolysis of boronates often occurs in refluxing acidic or basic
aqueous solution, however, the methyl ester in porphyrin 35 could not survive these
conditions. Therefore, it was planned to reduce the methyl ester 36 to ether first, and
then hydrolyze the new compound in refluxing acidic or basic aqueous condition to

afford free boronic acid compound.

LAH (1.3 eq.),

BFa'Etzo (3 eq.)_ From NMB, the
EtZO / THF product 18 not an

, ether
reflux, overnight

 

 

Scheme 6-43. The attempted reduction of methyl ester 36.

A mixture of lithium aluminum hydride and BF3 etherate can be used to reduce
esters to ethers directly.38 As shown in Scheme 6-43, pinacol boronate porphyrin 36, 1.3
equivalent of lithium aluminum hydride, and 3 equivalents of BF3-EtzO were added to a
solution of ether and THF. This reduction is typically performed in pure ether, but since
porphyrins do not dissolve in ether very well, minimum amount of THF was used to
improve the solubility of the porphyrin. When this mixture was stirred at room
temperature, no reaction occurred. After this mixture was refluxed overnight, the starting
material 36 decomposed.

So far, all the attempts to deprotect the pinacol borates had failed. Although, we
had successful synthesized the pinacol protected boronic porphyrin tweezer pinacol-2,

we could not find a way to remove the protecting group.

303

6-3. Transmetallation to introduce boronic acid

Transmetallation of aryl halides and reaction of the lithiated species with borates
could be an alternative method for the introduction of boronic acid.39 As shown in
Scheme 6-44, boronic acids could be transformed from the corresponding aryl halides by
transmetallation. Grignards, which are easily accessible from aryl halides, readily react
with B(OMe)3, and the subsequent treatment under acidic aqueous condition generates

the boronic acid.

Br 1. n-BuLi 3(Ole

2. B(OMe)3
3. HCI

Scheme 6-44. The transmetallation with bromobenzene.

As shown in Scheme 6-45, the same transmetallation reaction condition was
applied to the bromoporphyrin methyl ester 33. However, MS analysis did not indicate a
peak having the same molecular weight as the expected porphyrin boronic acid, and in
fact most of the starting bromoporphyrin 33 remained unreacted. Although the same
transmetallation reaction worked well for simple bromobenzene, it did not work for the
bromoporphyrin 33. The reason could be that the porphyrin stabilizes the negative
charge better than a simple benzene ring. After transmetallation with n-BuLi, the carbon

anion is too stable to attack B(OMe)3.

304

1. n-BuLi

2. B(OMe)3
3. HCI

no reaction

 

 

Scheme 6-45. The attempted transmetallation with bromoporphyrin 33.

So far, three different methods were tried to synthesize the boronic porphyrin
tweezer 2: the MacDonald’s “2+2” cyclization, the palladium catalyzed coupling reaction
and the transmetallation of bromoporphyrins. However, none of them provided the
desired boronic porphyrin tweezer 2.

6-4. Synthesis of water soluble boron containing porphyrin 45

There is very limited number of papers on the synthesis of para-substituted
boronic porphyrin, and our experience with this porphyrin suggested that the boronic
porphyrin tweezer 2 was a more difficult target to synthesize than we had anticipated.
We decided to change the structure, and make a more accessible boronic porphyrin

tweezer 45; the structure of which is shown in Figure 6-4.

305

 

Figure 6-4. The structure of a new porphyrin tweezer 45.

Porphyrin tweezer 45, shown in Figure 6-4, is a water-soluble salt, and can be
synthesized from the porphyrin methyl ester 46. Shinkai in Kyushu University has
synthesized a very similar molecule, which was used in binding with
oligocarbohydrates.40 As illustrated in Scheme 6-46, the phenylboronic part of molecule

46 is introduced by the pyridine porphyrin 47 reacting with the primary bromide 48.

 

Scheme 6-46. The synthesis of boronic porphyrin methyl ester 46.

The synthesis of the pyridine porphyrin 47 was routine, and is shown in Scheme

6-47. Porphyrin 47 was synthesized by condensing one equivalent of 4-formylpyridine

306

 

with one equivalent of methyl 4-formybenzoate and two equivalents of dipyrrolemethane

30, using BF3-Et20 or TFA as the acidic catalyst.

1. Bl=3 Ego,
CHZCIZ / EtOH

2. DDQ
5%

 

   

Scheme 6-47. The synthesis of pyridine porphyrin 47 .

The only difference in the synthesis of the pyridine porphyrin 47 and all the other
porphyrins was the much longer reaction time it required.“ The reaction took 24 h, while
it only required one or two hours for the synthesis of the other porphyrins. Because the
nitrogen atom of the starting material, 4-formylpyridine, was able to coordinate with the
catalyst, either BF3-Et20 or TFA, the reaction rate was slowed down. The separation of
this reaction mixture was hard, because of the generation of many byproducts, and the
yield was worse compared to the other porphyrins synthesized before.

13(01-1)2 B(OH)2

; 312. 1“! ©
49 LBr 48

Scheme 6-48. The bromination of compound 48.
The primary bromide 48 could either be synthesized or be purchased, however,
since it is expensive ($130 per gram), the compound was synthesized. As shown in
Scheme 6-48, it could be transformed from 4-methlyphenylboronic acid 49 by simple

bromination with bromine under light.42 The reaction worked smoothly when the scale of

307

 

this reaction was small, and the desired product 48 precipitated out of the mixture at the
end of the reaction and was therefore, easily purified. But the same reaction did not work
as well for larger amounts of starting material. In the reaction mixture, two products
were observed, the desired mono-brominated product 48 together with the unwanted
dibrominated product. The mixture was applied to column chromatography several
times, but we were unable to separate them due to their close polarity. The compound 48

used in the reaction was purchased from herein.

 

B(0H)2 8(0le
.1.
/ O
DMF / \
N
3, <2
48 50

Scheme 6-49. The reaction between pyridine and primary bromide 48.
As previously reported, the free boronic acid of compound 48 is first protected by
1,3-pentanediol, and then reaction with pyridine porphyrin 47 leads to the boronic

40, 43

containing porphyrin 46. We wanted to investigate the possibility of using this
reaction, without protecting the free boronic acid in compound 48. The reaction was first
checked with pyridine, as shown in Scheme 6-49. Pyridine was mixed with boronic acid
48, and dissolved in DMF. This reaction was stirred at 70 °C for one day, and the solvent
was removed with residue being washed with ether and acetone. The white precipitate
left in the flask was the desired coupling product 50. Because it was a salt, it did not

dissolve in organic solvent, and was easy to separate and purify from the reaction

mixture.

308

 

C02 Me 47

 

Scheme 6-50. The synthesis of water soluble porphyrin methyl ester 46.

Since the model reaction proceeded very well, the same method was applied to
the synthesis of pyridine porphyrin 46 (Scheme 6-50). After stirring at 80 °C for one day,
some of the product 46 had formed, but there was still unreacted starting pyridine
porphyrin 47 present. The reaction temperature was increased to 100 °C in order to
convert all the starting material to the product. However, after heating this solution at
100 °C for one day, there was no formation of the desired product observed. Prolonging
the reaction time at 100 °C resulted in the decomposition of the starting pyridine
porphyrin 47 . We needed to optimize the reaction between the pyridine porphyrin 47 and
the primary bromide 48. After the synthesis of the pyridine porphyrin tweezer,.the same
condition would be applied to achieve the synthesis of the boronic porphyrin tweezer 45.

However, we have not finished the synthesis of the boronic porphyrin tweezer 45.
After a year of working on the synthesis of the phenyl boronic acid porphyrin tweezer, I

moved on to another project.

309

 

Experimental materials and general procedures:

Anhydrous CHzClz was dried over CaHz and distilled. Unless Otherwise noted,
materials were obtained from commercial suppliers and were used without further
purification. All reactions were performed in dried glassware under nitrogen. Column
chromatography was performed using SiliCycle silica gel (230-400 mesh). 1H-NMR
spectra were obtained on Varian Inova 300 MHz instrument and are reported in parts per

million (ppm) relative to the solvent resonances (8), with coupling constants (J) in Hertz

 

(Hz).
CHO 0.?0
Q
030 TFA
K) \ /
\NHHN/
5

5-(4-Boronated)-phenyldipyrromethane 5:

A mixture of pyrrole (7 mL, 100 mmol) and 4-boronated-benzaldehyde (0.38 g, 2
mmol) was flushed with N2 for 5 min and treated with TFA (15 1.11, 2 mmol). This
mixture was stirred vigorously for 5 min. 0.1 N NaOH and ethyl acetate were added, and
the layers were separated. The aqueous layer was acidified and extracted with ethyl
acetate twice. Organic layers were combined and dried with anhydrous Na2804. The
solution was concentrated under reduced pressure. Column chromatography afforded a
pale yellow solid 5 (35%). 1H NMR (300 MHz, CDCl3): 8 7.74 (d, J=8.1 Hz, 2H), 7.22
(d, J=8.1 Hz, 2H), 6.70 (m, 2H), 6.17 (m, 2H), 5.93 (m, 2H), 5.50 (s, 1H), 4.19 (q, J=5.4

Hz, 4H), 2.09 (t, J=5.4 Hz, 2H).

310

 

Kl OOH EtOH/HQSOi H
U 2 reflux 10% V U

8 12

002Et

Ethyl ester of 2-pyrrol-carboxylic acid 12:

2-Pyrrol-carboxylic acid (100 mg) 8 was dissolved in 5 mL of ethanol, containing
1 mL concentrated H2S04. This solution was refluxed for 24 h, and the reaction mixture
was extracted with CHCl3 and water. The organic layer was separated and dried over
anhydrous N aZSO4. The solvent was removed under reduced pressure. The product was
purified by column chromatography, and the product was isolated as a solid in 10% yield.
1H NMR (300 MHz, CDC13): 8 9.20 (br, 1H), 6.96 (m, 2H), 6.28 (m, 1H), 4.34 (q, J=6.9

Hz, 2H), 1.38 (t, J=6.9 Hz, 3H).

CHO COOMe

9
$575023 COOMe A

EtOH / HCI
12 reflux

 

 

1,9-Bis(ethyloxycarbonyl)-5-(4-carboxy)-phenyldipyrromethane 13:

The ethyl ester 12 (70 mg, 0.5 mmol) and 4-carboxymethylbenzaldehyde 9 (33
mg, 0.25 mmol) was dissolved in absolute ethanol (4 mL), containing concentrated HCI
(1 mL). The reaction mixture was refluxed under nitrogen for 4 h and cooled in an ice-
bath. The precipitate was filtered and washed with cold methanol to afford the title

compound 13 as a solid.

311

CHO

 

COOH
R=M
€57 COOH 9" :
TFA \ /
14 \ NH HN /
15a R=H
15b B=Me

5-(4-Carboxy)-phenyldipyrromethane 153:

A mixture of pyrrole 14 (14 mL, 200 mmol) and 4-carboxybenzaldehyde 9a (0.6
g, 4 mmol) was flushed with N2 for 5 rrrin and treated with TFA (30 u], 4 mmol). This
mixture was stirred vigorously for 5 min. 0.1 N NaOH and ethyl acetate were added, and
the layers were separated. The aqueous layer was acidified and extracted with ethyl
acetate twice. Organic layers were combined and dried with anhydrous Na2804. The
solution was concentrated under reduced pressure. Column chromatography afforded a
pale brown solid 15a (51%). 1H NMR (300 MHz, CDCl3): 8 9.76 (br, 1H), 7.94 (d, J=8.1
Hz, 2H), 7.32 (d, J=8.1 Hz, 2H), 6.69 (m, 2H), 5.98 (m, 2H), 5.74 (m, 2H), 5.54 (s, 1H).
5-(4-Methoxycarbonyl)-phenyldipyrromethane 15b:

A mixture of pyrrole (14 mL, 200 mmol) and 4-methoxycarbonylbenzaldehyde
12b (1 g, 6 mmol) was flushed with N2 for 5 rrrin and treated with TFA (45 pl, 6 mmol).
This mixture was stirred vigorously for 5 min. 0.1 N NaOH and ethyl acetate were
added, and the layers were separated. The aqueous layer was acidified and extracted with
ethyl acetate twice. Organic layers were combined and dried with anhydrous Na2S04.
The solution was concentrated under reduced pressure. Column chromatography
afforded a pale brown solid 15b (50%). 1H NMR (300 MHz, CDC13): 8 7.99 (d, J=8.1
Hz, 2H), 7.29 (d, J=8.1 Hz, 2H), 6.73 (m, 2H), 6.18 (m, 2H), 5.91 (m, 2H), 5.54 (s, 1H),

3.92 (s, 3H).

312

 

COOH
EtMgBr

PhCOCI

 

Toluene, rt

\ / .
oveml ht
\ NH HN / 270/09

 

15a
1,9-Bisbenzoyl-5- (4-carboxy)-phenyldipyrromethane 16:

A solution of EtMgBr in 320 (5 mL, 5 mmol) was slowly added to a stirred, ice-
water bath cooled flask containing a solution of 5-(4-carboxy)-phenyldipyrromethane 15a
(133 mg, 0.5 mmol) in toluene (10 mL) under nitrogen atmosphere. This solution was
stirred in room temperature for 30 min, and then a solution of benzoyl chloride (300 mg,
2.1 mmol) in toluene (2 mL) was added slowly. The reaction was stirred overnight. It
was then quenched with saturated aq. NHaCl, and extracted with ethyl acetate (3 x 20
mL). The organic layers were combined and washed with brine, and dried with
anhydrous NaZSOa. Solvent was remove under reduced pressure, and the residue was
purified by column chromatography. The fraction contained the product was
concentrated and dissolved with CHzClz. Precipitation with hexane to afforded a white
solid 16 (27%). 'H NMR (300 MHz, CDC13): 8 11.12 (br, 1H), 7.93 (m, 4H), 7.84 (m,

4H), 6.89 (m, 2H), 6.26 (m, 2H), 5.76 (s, 1H).

COOMe

COOMe

AICI3 / PhCOCI
CHZCIZ, rt, 4 h

35%

 

   

15b

1,9-Bisbenzoyl-5- (4-methoxycarbonyl)-phenyldipyrromethane 17:
A solution of 5-(4-methoxycarbonyl)-phenyldipyrromethane 15b (140 mg, 0.5

mmol) in CH2C12 (1.4 mL) was added to ice-cold solution of benzoyl chloride (210 mg,

313

1.5 mmol) and AlCl3 (0.25 g, 1.8 mmol) in CHzClz (14 mL) under nitrogen. The reaction
was stirred in room temperature for 4 h, and quenched with saturated aq. NaHCO3 (5
mL). The mixture was filtered through Celite to remove the insoluble salt, and the
organic layer was separated and washed with NaOH and water. The organic layer was
dried with anhydrous NazSO4, and removed under the reduced pressure. The product was
purified by column chromatography as a solid in 35% yield. 1H NMR (300 MHz,
CDC13): 8 11.95 (br, 2H), 8.04 (d, J=8.1 Hz, 2H), 7.72 (d, J=7.5 Hz, 4H), 7.63 (d, J=8.1

Hz, 2H), 7.46 (d, J=7.2 Hz, 2H), 7.37 (m, 4H), 6.51 (m, 2H), 5.93 (m, 2H), 5.77 (s, 1H),

 

  

3.91 (s, 3H).
COOH
1 ,5-pentanediol
EDC, DMAP:
\\NH HN// CHZCIZ 35% \ /
15a 20 \ NH HN /

1,5-Bis (5’-p-carboxyphenyldipyrromethanyl) pentanoate 20:
5-(4-Carboxy)-phenyldipyrromethane 15a (130 mg, 0.5 mmol), 1,5-pentanediol
(20.8 mg, 0.2 mmol), EDC (100 mg, 0.5 mmol) and DMAP (72 mg, 0.6 mmol) was
dissolved in CH2C12 (3mL), and the mixture was stirred in room temperature overnight.
From TLC, there was no starting material left. The reaction mixture was directly applied
to column chromatography, and the product was isolated as a solid in 35% yield. 1H
NMR (300 MHz, CDC13): 8 7.94 (d, J=8.1 Hz, 4H), 7.24 (d, J=8.1 Hz, 4H), 6.69 (m, 4H),
6.14 (m, 4H), 5.87 (s, 4H), 5.48 (s, 2H), 4.30 (t, J=6.6Hz, 4H), 1.82 (m, 4H), 1.55 (m,

2H).

314

 

COOMe O O\/\/\/OH

1 ,5-pentanediol

 

\ / TsOH, 120 °C,
\ /
PhOC COP" PhOC COPh
17 21

1- (1’,9’-Bisbenzoyl-5’-p-carboxyphenyldipyrromethanyl) pentanoate 21:

1,9-Bisbenzoyl-5-(4-methoxycarbonyl)-phenyldipyrromethane 17 and catalytic
amount of TsOH was dissolved in 1,5-pentanediol (4 mL). Heating this mixture at 120
0C overnight and the mixture turned black. The reaction mixture was extracted with
water and chloroform (3 x 20 mL). The organic layers were combined and dried over
anhydrous Na2804. The solvent was removed under reduced pressure, and the product
was purified by column chromatography as a solid in 70% yield. IH NMR (300 MHz,
CDCl3): 8 11.59 (br, 2H), 8.07 (d, J=8.1 Hz, 2H), 7.76 (d, J=7.5 Hz, 4H), 7.64 (d, J=8.1
Hz, 2H), 7.51 (d, J=7.2 Hz, 2H), 7.44 (m, 4H), 6.59 (m, 2H), 5.99 (m, 2H), 5.77 (s, 1H),
4.36 (t, J=6.6 Hz, 2H), 3.68 (t, J=6.6 Hz, 2H), 1.82 (m, 4H), 1.55 (m, 2H).

COOMe

1 ,5-pentanedioi

 

/ Na, THF

\
\ NH HN / 68 °C 2 h
PhOC COPh 30% PhOC

17

 

1,5-Bis (1’, 9’-bisbenzoyl-5’-p-carboxyphenyldipyrromethanyl) pentanoate 19:
l,9-Bisbenzoyl-5-(4-methoxycarbonyl)-phenyldipyrromethane 17 (120 mg, 0.25

mmol), 1,5-pentanediol (14 mg, 0.13 mmol) and sodium (3 mg, 0.13 mmol) were

dissolved in THF (1.5 mL). The reaction was heated at 68 °C for 2 h, and the solvent was

removed under reduced pressure. HCl (6 N) was added to neutralize the sodium, and it

315

was extracted with dichloromethane (3 x 20 mL). The organic layers were combined,
and dried over anhydrous Na2SO4. The solvent was removed under reduced pressure,
and the product was isolated by column chromatography as solid in 30% yield. 1H NMR
(300 MHz, CDC13): 8 11.88 (br, 2H), 8.01 (d, J=8.1 Hz, 2H), 7.75 (d, J=7.5 Hz, 4H),
7.64 (d, J=8.1 Hz, 2H), 7.49 (d, J=7.2 Hz, 2H), 7.39 (m, 4H), 6.54 (m, 2H), 5.95 (m, 2H),

5.80 (s, 1H), 4.36 (t, J=6.6 Hz, 2H), 1.82 (m, 4H), 1.55 (m, 2H).

 

     

O.B.O U 0'30
TFA
66°o
CHO \\ //
25 NH HN 26

5-(4-Boronated)-phenyldipyrromethane 26:

A mixture of pyrrole (11.5 mL, 165 mmol) and 4—boronated-benzaldehyde 25
(0.72 g, 3.3 mmol) was flushed with N2 for 5 min and treated with TFA (25 1.1.1, 3.3
mmol). This mixture was stirred vigorously for 5 min. 0.1 N NaOH and ethyl acetate
were added, and the layers were separated. The aqueous layer was acidified and
extracted with ethyl acetate (2 x 20 mL). Organic layers were combined and dried with
anhydrous Na2SO4. The solution was concentrated under reduced pressure. Column
chromatography afforded a pale yellow solid 26 (66%). 1H NMR (300 MHz, CDCl3): 8
7.84 (d, J=8.1 Hz, 2H), 7.23 (d, J=8.1 Hz, 2H), 6.65 (m, 2H), 6.20 (m, 2H), 5.93 (m, 2H),

5.42 (s, 1H), 3.92 (s, 4H), 1.10 (s, 6H).

316

1 . NaBH4
COOMe TFA / MeOH
2. TFA / CH3CN

       

 

3. DDQ O OMe

Boronic porphyrin methyl ester 24:

NaBI-I4 (100 mg, 2.6 mmol) was added to the solution of dipyrromethane 17 (24.4
mg, 0.05 mmol) in MeOH/THF (1:3, 4 mL) at small portions. This mixture was stirred at
room temperature for an hour, and quenched with saturated aq. ammonium chloride. The
mixture was extracted with dichloromethane (3 x 20 mL), and the organic layers were
washed with water (1 x 20 mL). The organic layer was dried over anhydrous Na2SO4,
and was removed under reduced pressure. The residue was dissolved with dry
acetonitrile (20 mL), and 5-(4-boronated)-phenyldipyrromethane 26 (16 mg, 0.05 mmol)
was added into this solution. Purged N2 into the mixture for 10 min, and TFA (0.05 mL,
0.65 mmol) was added into the reaction mixture. The deep red solution was stirred at
room temperature for 1 h, and DDQ (0.35 g, 0.15 mmol) was added. The reaction
mixture was stirred at room temperature for another hour, and all the solvent was
removed under reducing pressure. The product was purified by column chromatography
as purple solid in 20% yield. 1H NMR (300 MHz, CDC13): 8 8.98 (m, 8H), 8.46 (d, J=8.1
Hz, 2H), 8.32 (d, J=8.1 Hz, 2H), 8.24 (m, 6H), 7.80 (m, 9H), 4.13 (s, 3H), 3.92 (s, 4H),

1.11 (s, 6H); -2.79 (s, 2H); MS: m/z 784.

317

 

 

5-Phenyldipyrromethane 30:

A mixture of pyrrole (21 mL, 300 mmol) and benzaldehyde (5.3 g, 50 mmol) was
flushed with N2 for 5 rrrin and treated with TFA (0.45 mL, 50 mmol). This mixture was
stirred vigorously for 5 rrrin. 0.1 N NaOH and ethyl acetate were added, and the layers
were separated. The aqueous layer was acidified and extracted with ethyl acetate (2 x
100 mL). Organic layers were combined and dried with anhydrous Na2SO4. The
solution was concentrated under reduced pressure. Column chromatography afforded a
pale yellow solid 26 (49%). lH NMR (300 MHz, CDC13): 8 7.90 (br, 2H), 7.26 (m, 5H),

6.69 (br, 2H), 6.15 (d, J=2.8 Hz, 2H), 5.92 (br, 2H), 5.47 (s, 1H).

CH0 CH0
1. BF3‘E20, CH3CI

2. DDQ

 

OH Br \ NHHN /

 

Monohydroxyl-monobromo porphyrin 29:

4-Hydroxybenzaldehyde (30.5 mg, 0.25 mmol), 4-bromobenza1dehyde (46 mg,
0.25 mmol) and S-phenyldipyrromethane 30 (111 mg, 0.5 mmol) were dissolved in
CHC13 (50 mL). The solution was purged with N2 for 10 minutes, and BF3-Et2O (20 111)

was added. The reaction was stirred for an hour, and DDQ (0.20 g, 0.09 mmol) was

318

 

added. The reaction mixture was stirred for another hour, and subjected directly to
column chromatography. The product was isolated as purple solid in 10% yield. 1H
NMR (300 MHz, CDC13): 8 8.89 (m, 8H), 8.22 (m, 4H), 8.04 (m, 4H), 7.90 (d, 2H), 7.75

(m, 6H), 7.13 (d, 2H), -2.81 (s, 2H).

1. BF3’ Ezo, CH3C1
2. DDQ 10% 7

 

   

Porphyrin methyl ester 33:

4-Methoxycarbonyl-benzaldehyde (41 mg, 0.25 mmol), 4-bromobenzaldehyde
(46 mg, 0.25 mmol) and 5-phenyldipyrromethane 30 (111 mg, 0.5 mmol) were dissolved
in CHCl3 (50 mL). The solution was purged with N2 for 10 minutes, and BF3-Et2O (20
111) was added. The reaction was stirred for an hour, and DDQ (0.20 g, 0.09 mmol) was
added. The reaction mixture was stirred for another hour, and was subjected directly to
column chromatography. The product was isolated as purple solid in 10% yield. The
ratio between di-bromo, 33, and di-ester is 2:4: 1. 1H NMR (300 MHz, CDC13): 8 8.91(m,
8H), 8.49 (d, J=8.1 Hz, 2H), 8.35 (d, J=8.1 Hz, 2H), 8.25 (m, 4H), 8.11 (d, J=8.1 Hz,

2H), 7.91 (d, J=8.1 Hz, 2H), 7.80 (m, 6H), 4.35 (s, 3H), -2.81 (s, 2H).

319

 

NaOH / THF
reflux, overnight

99%

 

 

Porphyrin acid 32:

Porphyrin methyl ester 33 (35 mg) was added into NaOH (2 N, 1.5 mL) and THF
(5 mL), and the reaction mixture was refluxed for 12 h. The solution was acidified with 2
N HCl to pH of 4. The reaction mixture was extracted with CH2C12 until the aqueous
layer became colorless. The organic layers were combined and washed with water (2 x
20 mL). The organic layer was dried over anhydrous Na2804, and the removal of the
solvent under reduced pressure provided the title compound as purple solid in 99% yield.
1H NMR (300 MHz, CDC13): 8 8.91 (m, 8H), 8.56 (d, J=8.1 Hz, 2H), 8.39 (d, J=8.1 Hz,

2H), 8.23 (m, 4H), 8.10 (d, J=8.1 Hz, 2H), 7.92 (d, J=8.1 Hz, 2H), 7.81 (m, 6H).

320

 

 

1 ,5-pentanediol
EDC, DMAP Ph

 

 

Dibromo porphyrin tweezer 34:

Porphyrin acid 32 (30 mg, 0.04 mmol), 1,5-pentanediol (2.12 mg, 0.02 mmol),
EDC (10 mg, 0.05 mmol) and DMAP (7.2 mg, 0.06 mmol) were dissolved in CH2C12 (5
mL). The mixture was stirred in room temperature overnight, and applied directly to
column chromatography. The product was isolated as a purple solid in 5% yield. 1H
NMR (300 MHz, CDCl3): 8 8.85 (m, 16H); 8.52 (d, J=8.1 Hz, 4H), 8.35 (d, J=8.1 Hz,
4H), 8.14 (m, 12H), 7.78 (m, 16H), 4.68 (t, J=6.6 Hz 4H) 1.65 (m, 4H), 1.48 (m, 2H), -

2.81 (s, 2H).

321

3’

 

1. Zn(OAc)2 / CH3C|

2
' .0 Ph
..-e, 2
Ph 0

Pd(PPh3)4
Et3

Cl CI

   

Boronate porphyrin tweezer 34:

Porphyrin tweezer free base 34 (6 mg) was dissolved in 1.5 mL of chloroform,
and zinc acetate (10 mg) was added. This mixture was stirred in room temperature for 2
h, and applied to column chromatography to isolate the pure zinc porphyrin tweezer. A
50 mL Schlenk flask was charged with zinc porphyrin tweezer (6 mg), pinacolborane (20
111, 0.14 mmol), triethylamine (30 111, 0.22 mmol), tetraphenylphosphinepalladium (1
mg), and 2 mL of 1,2-dichloroethane in dry box. This mixture was stirred at 90 °C for 4
h, and it was then quenched with saturated aq. KCl (1 mL). The mixture was separated
and the organic layer was washed with water and dried over anhydrous Mg2804. The
solvent was removed under reduced pressure, and the product was isolated as a purple

solid in 5% yield. MS: m/z 1764.

322

 

H".
Ph

PdCl2(PPh3)2
Et3N
C(“Cl

 

Coupling of bromoporphyrin 33 with pinacolborane:

Bromoporphyrin methyl ester free base 33 (15 mg) was dissolved in 1.5 mL of
chloroform, and zinc acetate (20 mg) was added. This mixture was stirred at room
temperature for 2 h, and applied to column chromatography to isolate the pure zinc
porphyrin tweezer. A 50mL Schlenk flask was charged with zinc porphyrin (15 mg),
pinacolborane (20 1.11, 0.14 mmol), triethylamine (30 1.11, 0.22 mmol), PdCl2(PPh3)2 (1
mg), and 3 mL of 1,2-dichloroethane in dry box. This mixture was stirred at 90 °C
overnight before it was quenched with saturated aq. KCl (1 mL). The mixture was
separated and the organic layer was washed with water and dried over anhydrous
Mg2SO4. The solvent was removed under reduced pressure, and the product was isolated
as a purple solid in 6% yield. 1H NMR (300 MHz, CDC13): 8 8.87 (m, 8H), 8.46 (d,
J=8.2 Hz, 2H), 8.33 (d, J=8.2 Hz, 2H), 8.25 (m, 8H), 7.79 (m, 8H), 4.13 (s, 3H), 1.28 (s,

9H); MS: m/z 862.

323

‘1

 

OH
szo. 131/(299.)

CH20|2
CHO CHO 37

 

4-Formylphenyl trifloromethanesulfonate 37 :

Suspended 4-hydroxylbenzaldehyde (1.22 g, 10 mmol) and pyridine (0.9 mL) in
CH2C12 (20 mL), the mixture was slowly added to ice-bath colded Tf2O (2 mL, 12
mole) in CH2C12 (15 mL). This mixture was stirred at 0 °C for 10 minutes, and then
warmed up to room temperature, and stirred for 30 h. The solvent was removed, and the
residue was purified by column chromatography to yield the title compound 37 (79%).
The spectrum matches the reported literature.34 1H NMR (300 MHz, CDC13): 8 10.1 (s,

1H), 8.01 (d, J=8.7 Hz, 2H), 7.47 (d, J=8.7 Hz, 2H).

OTf
kl

OTf 1.17

TFA, CH20|2 \ /
CHO 51 °/o \ NH NH
37 38

 

5-(4-Triflate)-phenyldipyrromethane 38:

A mixture of pyrrole (2.5 mL, 31 mmol) and aldehyde 37 (0.12 g, 0.55 mmol)
was flushed with N2 for 5 min and treated with TFA (5 u], 0.66 mmol). This mixture was
stirred vigorously for 5 min. 0.1 N NaOH and ethyl acetate were added, and the layers
were separated. The aqueous layer was acidified and extracted with ethyl acetate (2 x 30
mL). Organic layers were combined and dried with anhydrous Na2SO4. The solution
was concentrated under reduced pressure. Column chromatography afforded a pale
yellow solid 38 (51%). lH NMR (300 MHz, CDC13) 8 7.99 (br, 2H), 7.27 (m, 4H), 6.72

(m, 2H), 6.21 (m, 2H), 5.90 (m, 2H), 5.50 (s, 1H).

324

 

COOMe " COOMe — 1. TFA, CHacN

 

 

 

 

   

NaBH4 OTf
TFA/MeOH
- Ph
\ /
\ NH HN / \ /
Ph Ph
COP“ _ OH HO __ \ NH HN /
38
2. DDQ10%

Triflate porphryin methyl ester 39:

NaBHa (200 mg, 5 mmol) was added to the solution of phenyldipyrromethane 17
(48.8 mg, 0.1 mmol) in MeOH/THF (1:3, 4 mL) at small portions. This mixture was
stirred at room temperature for an hour, and quenched with saturated aq. ammonium
chloride. The mixture was extracted with dichloromethane (3 x 20 mL), and the organic
layer was washed with water (1 x 20 mL). The organic layer was dried over anhydrous
Na2804, and was removed under reduced pressure. The residue was dissolved with dry
acetonitrile (20 mL), and 5-(4-triflate)-phenyldipyrromethane 38 (37 mg, 0.1 mmole) was
added into this solution. Purged N2 into the mixture for 10 min, and TFA (0.05 mL, 0.65
mmol) was added into the reaction mixture. The deep red solution was stirred at room
temperature for 1 11, and DDQ (0.35 g, 0.15 mmol) was added. The reaction mixture was
stirred at room temperature for another hour, and all the solvent was removed under
reducing pressure. The product was purified by column chromatography as purple solid

in 10% yield. MS: m/z 820.

325

 

3(0le

NIS / CH3CN
reflux 55%

I
CHO CHO 4°

 

Synthesis of 4-iodobenzaldehyde 40:

4-Formylphenylboronic acid (300 mg, 2 mmol) and NIS (612 mg, 2.4 mmol) was
dissolved in 10 mL of dry MeCN, and this mixture was refluxed for 16 h. The solvent
was then removed reduced pressure, and the product was separated by column
chromatography as a solid in 55% yield. The NMR spectrum matches the reported

literature.3S 1H NMR (300 MHz, CDC13): 5 10.1 (s, 1H), 8.06 (d, J=8.7 Hz, 2H), 7.52 (d,

 

J=8.7 Hz, 2H).
CHO CHO 1 Bl=3 Ego,
CH3CI
+ +
\ / 2 DDQ
002Me I \ NH HN / 14%

 

Synthesis of iodoporphyrin methyl ester 41:

4-Methoxycarbonylbenzaldehyde (73 mg, 0.5 mmol), 4-iodobenzaldehyde 40
(115 mg, 0.5 mmol) and 5-phenyldipyrromethane 30 (222 mg, 1 mmol) were dissolved in
CH2Cl2 (100 mL). The solution was purged with N2 for 10 minutes, and BF3-Et2O (40
pl) was added. The reaction was stirred for an hour, and DDQ (0.20 g, 0.09 mmol) was
added. The reaction mixture was stirred for another hour, and subjected directly to

column chromatography. The product was isolated as purple solid in 14% yield. IH

326

 

 

NMR (300 MHz, CDC13): 8 8.91 (m, 8H), 8.48 (d, J=8.2 Hz, 2H), 8.35(m, 2H), 8.25 (m,

4H), 8.11 (d, J=8.2 Hz, 2H), 7.95 (m, 2H), 7.81 (m, 6H), 4.15 (s, 3H), -2.81 (s, 2H).

0.0

B
O
:OB-H
O
PdCl2(PPh3)2 phmph
Dichloroethane
Et3N, 140 °C /

overnight 0

O OMe
41 36

 

 

Coupling of iodoporphyrin with pinacolborane by PdC12(PPh3)2:

A 50 mL Schlenk flask was charged with iodoporphyrin methyl ester free base 41
(20 mg), pinacolborane (50 111, 0.35 mmol), triethylamine (100 111, 0.7 mmol),
PdCl2(PPh3)2 dichlorodi(triphenyl)phosphinepalladium (1 mg), and 2 mL of 1,2-
dichloroethane in dry box. This mixture was stirred at 140 °C overnight before it was
quenched with aq. KCl (1 mL). The mixture was separated and the organic layer was
washed with water and dried over anhydrous Mg2S04. The solvent was removed under
reduced pressure, and the product was isolated as a purple solid in 30% yield. 1H NMR
(300 MHz, CDC13): 8 8.87 (m, 8H), 8.46 (d, J=8.2 Hz, 2H), 8.33 (d, J=8.2 Hz, 2H), 8.25

(m, 8H), 7.79 (m, 8H), 4.13 (s, 3H), 1.28 (s, 9H); MS: m/z 862.

327

 

 

NaOH,
THF

reflux, ph
overnight
98%

 

COzMe
41

 

Synthesis of iodoporphyrin acid 43:

Iodoporphyrin methyl ester 41 (65 mg) was added into NaOH (2 N, 1.5 mL) and
THF (5 mL), and refluxed for 12 h. The solution was acidified with 2 N HCI to pH of 4.
The reaction mixture was extracted with CH2C12 until the aqueous layer became
colorless. The organic layers were combined and washed with water (2 x 20 mL). The
organic layer was dried over anhydrous Na2SOa, and the removal of the solvent under
reduced pressure provided the title compound as purple solid in 98% yield. 1H NMR
(300 MHz, CDCl3): 8 8.88 (m, 8H), 8.58 (d, J=8.2 Hz, 2H), 8.38 (d, J=7.7 Hz, 2H), 8.24

(m, 4H), 8.13 (d, J=7.7 Hz, 2H), 7.97 (d, J=8.2 Hz, 2H), 7.80 (m, 6H), -2.74 (s, 2H).

328

 

 

 

 

1 ,5-pentanediol Ph
EDC, DMAP / Ph
/
O C>‘\_\j O
42 o 0

Synthesis of diiodoporphyrin tweezer 42:

Porphyrin acid 43 (57 mg, 0.07 mmol), 1,5-pentanediol (3.78 mg, 0.036mmole),
EDC (19.1 mg, 0.1 mole) and DMAP (13.5 mg, 0.11 mole) were dissolved in
CH2C12, and the mixture was stirred in room temperature overnight. From TLC, there
was no starting material left. The product was purified by column chromatography as a
purple solid (54%). 1H NMR (300 MHz, CDCl3): 8 8.88 (m, 16H), 8.58 (d, J=8.2 Hz,
2H), 8.38 (d, J=7.7 Hz, 4H), 8.24-8.13 (m, 16H), 7.97-7.80 (m, 18H), 4.28 (t, J=6.6 Hz,

4H), 1.65 (m, 4H), 1.48 (m, 2H), -2.81 (s, 2H).

329

-1

 

O o
Ph Ph : _B-H
0 P

 

/ Ph PdCl2(PPh3)2
O Dichloroethane
/ Et3N, 140°C
0 O O ovemlght
42 O

 

Synthesis of Boron porphyrin tweezer:

A 50 mL Schlenk flask was charged with diiodoporphyrin tweezer free base 40
(20 mg), pinacolborane (50 1.11, 0.35 mmol), triethylamine (100 111, 0.7 mmol),
PdCl2(PPh3)2 (1 mg), and 2 mL of 1,2-dichloroethane in dry box. This mixture was
stirred at 140 °C overnight before it was quenched with aq. KCl (1 mL). The mixture was
separated and the organic layer was washed with water and dried over anhydrous
Mg2SOa. The solvent was removed under reduced pressure, and the product was isolated
as a purple solid in 23% yield. 1H NMR (300 MHz, CDCl3): 8 8.87 (m, 16H), 8.46 (d,
J=8.2 Hz, 4H), 8.33-8.25 (m, 16H), 7.79-8.24 (m, 18H), 4.28 (t, J=6.6 Hz, 4H), 1.65 (m,

4H), 1.48 (m, 2H), -2.81 (s, 2H); MS: m/z 1636.7.

330

 

 
   
 

1. BF3-E20,
CH2012/ EtOH

2. DDQ T
5%

O

 

Synthesis of pyridine porphyrin 47:

To a degassed solution of dipyrrolmethane 29 (444 mg, 2 mmol) in CH2Cl2/EtOH
(100 mL, 95:5) was successively added methyl 4-formylbenzoate (146 mg, 1 mmol),
BF3-Et2O (231 1.11) and 4-pyridinecarboxaldehyde (107 mg, 1 mmol). This solution was
protected from light and stirred under N2 for 36 hours. Then DDQ (340 mg, 1.5 mmol)
was added and the stirring was continued for another 2 hours. The solvent was removed
under reduced pressure, and the product was separated by column chromatography as a

purple solid in 5% yield. Mass Spectrum: m/z 873.

331

Reference:

1.

2.

10.

ll.

12.

13.

14.

15.

16.

17.

18.

19.

20.

Lorand, J. P.; Edward, J. O. J. Org. Chem. 1959, 24, 769.

Sugashi, A.; Ikeda, M.; Takeuchi, M.; Shinkai, S. Angew. Chem. Int. Ed. 2000,
39, 3839.

Yang, W.; He, H.; Drueckhammer, D. G. Angew. Chem. Int. Ed. 2001, 40, 1714.
Cooper, C. R.; James, T. D. Chem. Lett. 1998, 883.
McNichols, R. J .; Cote, G. L. J. Biomed. Opt. 2000, 5, 15.

Gardiner, S. J.; Smith, B. D.; Duggan, P. J.; Karpa, M. J.; Griffin, G. J.
Tetrahedron 1999, 55, 2857.

Arimori, 8.; Takeuchi, M.; Shinkai, S. J. Am. Chem. Soc. 1996, 118, 245.

Bartlett, P. A.; SHea, G. T.; Telfer, S. J.; Waterman, S. Molecular Recognition:
Chemical and Biological Problems; Royal Society of Chemistry: London, 1989.

Bartlett, P. A.; Lauri, G. J. Comput. Aided Mol. Des. 1994, 8, 51.

Arsenault, G. P.; Bullock, E.; MacDonald, S. F. J. Chem. Soc. 1960, 82, 4384.
Boyle, R. W.; Clarke, O. J. Tetrahedron Lett. 1998, 39, 7167.”

Hughes, M. P.; Shang, M.; Smith, B. D. J. Org. Chem. 1996, 61, 4510.
Valasinas, A.; Hurst, J .; Frydman, B. J. Org. Chem. 1998, 63, 1239.

Chang, C. J.; Deng, Y. Q.; Heyduk, A. F.; Chang, C. K.; Nocera, D. G. Inorg.
Chem. 2000, 39, 959.

Soufiaoui, C. Tetrahedron 1989, 45, 3351.

Rao, P. D.; Dhanalekshmi, 8.; Linsey, J. S. J. Org. Chem. 2000, 65, 7323.
Sessler, J. L.; Hugdahl, J .; Johnson, M. R. J. Org. Chem. 1986, 51, 2838.
Khoury, R. G.; Jaquinod, L.; Smith, K. M. Chem. Commun. 1997, 1057.

Cho, W.; Kim, H.; Littler, B. J.; Miller, M. A.; Lee, C. -H.; Lindsey, J. S. J. Org.
Chem. 1999, 64, 7890.

Chen, Q.; Lightner, D. A. J. Org. Chem. 1998, 63, 2265.

332

--v

 

21.

22.

23.

24.

25.

26.

27.

28.

29.

30.

31.

32.

33.

34.

35.

36.

37.

38.

39.

Djurendic, A.; Suranyi, T. M.; Milikovic, D. A. Chzech. Chem. Commun. 1990,
55, 1763.

Martinez, A. G.; Barcina, J. O.; Veccio, H. D.; Hanack, M.; Subramanian, L. R.
Tetrahedron Lett. 1991, 32, 5931.

Zieglar, T.; Lemanski, G.; Hurttlen, J. Tetrahedron Lett. 2001, 42, 569.

Miyaura, N .; Ishiyama, T.; Sasaki, H.; Ishikawa, M.; Satoh, M.; Suzuki, A. J. Am.
Chem. Soc. 1989, III, 314.

Ohe, T.; Ohe, K.; Uemura, S.; Sugita, N. J. Organomet. Chem. 1988, 344, C5.

Littler, B. J .; Miller, M. A.; Hung, C. -H.; Lindsey, J. S. J. Org. Chem. 1999, 64,
1391.

Boyle, R. W.; Bruckner, C.; Posakony, J .; James, B. R.; Dolphin, D. Org. Synt.
1999, 76, 287.

Hyslop, A. G.; Kellett, M. A.; Iovine, P. M.; Therien, M. J. J. Am. Chem. Soc.
1998, 120, 12676.

Ishiyama, T.; Ishida, K.; Miyaura, N. Tetrahedron 2001, 5 7, 9813.

Ishiyama, T.; Murata, M.; Miyaura, N. J. Org. Chem. 1995, 60, 7508.

Murata, M.; Oyarna, T.; Watanabe, S.; Masuda, Y. J. Org. Chem. 2000, 65, 164.
Ishiyama, T.; Itoh, Y.; Kinato, T.; Miyaura, N. Tetrahedron Lett. 1997, 38, 3447.
Nakamura, H.; Fujiwara, M.; Yamamoto, U. Bull. Chem. Soc. Jpn. 2000, 73, 231.
Echavarren, A. M.; Stille, J. K. J. Am. Chem. Soc. 1987, 109, 5478.

Thiebes, C.; Prakash, G. K. S.; Petasis, N.; Olah, G. A. Syn. Lett. 1998, 141.
Takeuchi, M.; Chin, Y.; Irnada, T.; Shinkai, S. Chem. Commun. 1996, 1867.

Coutts, S. J .; Adams, J .; Drolikowski, D.; Snow, R. J. Tetrahedron Lett. 1994, 35,
5109.

Pettit, G. R.; Kasturi, T. R. J. Org. Chem. 1960, 25, 875.

Boisnard, A.; Carbonnelle, A. -C.; Zhu, J. Org. Lett. 2001, 3, 2061.

333

 

 

40.

41.

42.

43.

Sugasaki, A.; Ikeda, M.; Takeuchi, M.; Koumoto, K.; Shinkai, S. Tetrahedron
2000, 56, 4717.

Rucareanu, S.; Mongin, 0.; Schuwey, A.; Hoyler, N.; Gossauer, A. J. Org. Chem.
2001, 66, 4973.

Thomas, A. P.; Robert, D. A.; Thomason, D. A. Bioorg. Med. Chem. Lett. 1994,
4, 2615.

Arimori, S.; Murakami, H.; Takeuchi, M.; Shinkai, S. J. Chem. Soc., Chem.
Commun. 1995, 961.

334

 

APPENDIX

335

 

 

 

 

 

 

 

Au

 

 

 

 

 

 

_.N

B aoPeN as, 5 6:38:60 3
L42: a 658.58 A6
O 3
achN 9. confine cog: Fm ncsanoo CO 325650 565 E 69ch

 

336

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

o F m m e m e N m m
_h_LL_»_______~_____p___bpbg________________$____$___V_
d 33. .. 4 .4 w J\
Am
4 _ D
s or L a 45: a
L
mu
0 data £5, ca 65858 3
.0 on 958800 E
\ffiz: 3

O achN 9 5:63 com: on 23888 CO 32:55 cocoa E omcmzo

337

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

o F .o. m e m e H e
_W~____hh_h__g__dp__p___L_p_$_»____________h_k_____L_F
_ La - 5 UN
L
3
R
O aaPeN £5, R 6:38.58 3
u R 9:59.80 Am
\ffiz:
O 3

anficN 9 confine com: um ocaanoo Co _mo_Eoco cocoa :_ oacmco

 

338

 

 

 

 

 

 

 

 

 

 

 

mu

.m
Cf}:

 

 

 

.. 5; ill
ochN 5;, 8 2.8980 3
mm Ocsanoo Am

achN 2 confine :8: am 95888 CO 62820 cocoa E 3:20

 

339

AEcV camco_o>m>>
m5» mmv

mum
fl com.

1 omT

 

N
5.7m 12

cm 0 + 53625 53:98 cN
\ffiz:

O

..m- Tm OCBOQEOO .6 Enzooqw 000m

w<

 

r om?

340

A55 £92923

min mmv mum
1 ON T

l owl

F‘-
.—

 

tom

 

t omw
N_._z
hm.~.m

cm 0 + cowoozz 5:598 :N
12...
O

..m-~-m 3:09:00 Co 83.86% 000m

341

 

 

.rilltth

AEcv £82825

when mNV

mum

1

 

..mtnrm

O + .885 5.38.8 :N

..métm 8:868 do 83:88 000m

 

ONT

om

our

w<

342

A85 £m:m.m>m>>

mt» mmw mum
fi oo 7

I
O
“P

lllfll til Ithvl/
w<

om

T

 

. ooF

N...z
..méé

.m 0 + $395 $3598 cN

>_\./=\ZI
O

.méé 23888 B 528QO 000m

343

mhw

ASE £9563;

va mum
1 00...

T om!

 

w<

[om

 

«:2 oo

.méfi
..m 0 + .385 53:98 cN

2n. . 21
/\/=\

O

5-9m 2:09:00 6 E2823 000m

344

2.5 595633

 

 

 

mhw mNV mnm
a 0v-
7 ON!
Y‘all il LII I 1’14 .J.III Ir 1 o
w<
1 ON
NIZ I 0.?

5.7m
_o O + 5323 53:88 :N

\./=\z:

O

.0- Tm nczanoo ho Eszomqm 000m

345

min

p.—

Es £929;

mmv

 

 

mnm
n om T

I m“.-

 

N_._z

6-3
.0 + .3me 23:98 :N

J\./=\z_._
o

_o-~-m 23888 6 83:0QO 000m

 

r om?

346

AEE 2&co_m>m>>

 

 

mnv mmv mum
1 ow-
,. om-
F111: _ f : o 2
,. om
~12 [ ov

_O.n.m

.0 O + 5395 c2398 cN
52...
O

_o-m-m nczanoo 6 83:0QO 000w

347

 

mnv

 

AEE £98632,
mmv

I I. b

N...z

_0 O + .3me 5:398 cN

\J\./=\ZI
O

_0-¢.m 33888 .6 83:0QO 000m.

_O.¢.m

mum

1

I

 

om 7

mm

omw

348

mnv

 

A85 £mc¢.m>m>>

 

 

va mum
- cm.
T m?
_ _ (E1 o
I m_.
«12 r om

_0-m.m
_0 O + ENmmE 5:598 cN

an. . 21
/\/=\

O

_0-m-m 95888 ,6 83:0QO 000m

349

 

Es 26:296.;
mt» mmv

~12
HTTm

u_ D + $322 53:33 :N

\JWZI

O

u- Tm 3:38:80 00 82:0me 000m

mum
1 CV.

T ONI

oz

{om

 

row

350

 

mnv

I151! _ E0 QC
2

 

3.5 568353
mmv mum
1 oo 7

I om...

E}

{om

 

NI I OOF

u..N.m
u. + 5392 5:398 cN

\jfiZI
O

u.-~-m 3:38:50 6 83:0QO 000m

351

A65 £mcm.m>m>>

 

 

 

wfl

mtg mmv mum
1 ow-
I om-
NEE _ (1x1?! 11111.1 0
T om
sz - ow

n..n-m
O + 5323 53:98 cN

m.
\52:
o

u-m-m nczanoo 8 E26QO 000m

352

 

mhv

F—

 

..._é-m

2.5 265.95;
mmv

«:2

n. O + 5322 5:598. cN
4%.}:
o

"TYm “0:38:50 6 83:0QO 000w

mum

 

11.1.

 

I

om..-

mm-

mn

om—

w<

353

 

Es £58652,

mt» mmw mum
1 can

T 2-

 

r I I _ 5] O
w<
umw

~12
u-m-m I on

..__ O + 58305335“; cN

i . z:
/\/=\

O

 

u-m-m nczanoo B Ezzowam 000m

354

mnv

2.5 582962,
mmv

~12

_ O + 53322598 5

\JWZI

O

_._...m

_- Tm chanoo 00 53:0wa 000m

mum

I

 

0v-

ON

ov

355

mt»

AES £65632,
mmw

£2
0 + .3me 53:98 cN

hm 33888 6 626QO 000m

mum

I

W

 

9».

ON

ov

w<

356

mhv

A85 £mcm.m>m>>

mmv

N_._z

an O :mNmmEccfeoa :N
2H “0

Ocn.

am 23888 00 £26me 000m

mum
n om-

I m—vl

w<

Imw

 

[cm

357

E5 5956225

 

 

mhv mmv mum
a 3.
I 8-

VIII- I I I o
I ON
I 0v

8 + 5395 5:398 cN

0
Q0

«0.22

cm 0:28:80 B 826QO 000m

w<

358

mt»

AES £05352,
mmv

N12
0 + .332; E3598 cN

5 3:89:00 ho 83:0me 000m

mnm
I om!

 

-om

359

A85 £000.90;
mtg mmv

«:2
mm
0 + .3002: 0:50.00 0N

Q0

002

mm 0:309:00 00 63:0QO 000m

mum
I 0.?!

I ONI

ION

 

.9»

360

A65 £00205;
mwv mmv

mm

N...z
O + 03002: £359.00 :N
Q0

eocm

mm 00000500 00 E26000 000m

mum
1 ON.

I OF!

04

low

 

row

361

m:

 

cm

3.5 £000.90;
mmv

N:2
O + .5305 £03900 cN

Q0
:02

0

cm 00300600 00 E38000, 000m

 

ON

ov

362

.85 £000.03;

 

 

mtg mmv mum

1 ON-
1 OP-

_ Er .

<f\.‘ Ii 0 w<
r 0..

N12

.. om

mm + .3002; 02>..an 0N

O

“on

.-zm

mm 00300800 .0 0.2.0000 000m.

363

mhv

 

.80. £000.90;
mmv mum
I om-

I le

3
I mm

 

Iom

«:2
E. O + .0303. 0.35900 0N
z

O

/

ms 00300800 00 82.00% 000m.

364

 

 

A55 5000.053

 

 

mnw mm.» mum
1 CV:
T. ONI
_|.I.1I| _ o
04
I om
I 00
N12
2. I 00
O +.0~00§:_.>c&00 0N
z 0

R 00300600 .0 Eszomam 000m.

365

 

A000. 0600.053
min mmv

on
+ 00302,. 0.30900 0N

W
RD

2. 003000000 00 8300000 000m.

 

mum

 

I owl

I om...

Iom

 

I00

366

 

000. 5000.90;

mt» va mum
0 mm-

 

367

00.02-.. +

 

N 6000.0 0.0.0900 00.~ 9 00.025. 00 00.0.02