42....
35d wt.

 

a: rt
«.4...

CW

.. . .s
Hwfiwu
$k D.

m

5....

.32.. 1.2:
.

an... .. L53...

3 .waf‘

”V
“V
n.
1

.
h

Wu

15¢ ‘ .
.a.” : Atathuvav‘u‘

Ostny

t... .l.‘ :I.

t‘ I'-
.\.?,: It: 4 ..
s. (autumn l

34:15" «i
1.41.1353}!
19503.! 1v

diur‘.
. fifty
«flung, 5

1....

1......3. .u.
374.... JP”!

5353‘

2v
:1:

in

 

 

LI DHHn I

 

Tina's-s Michigan State
3 University
202189

This is to certify that the
dissertation entitled

ASYMMETRIC SYNTHESIS WITH VAPOL DERIVATIVES
AND NOVEL CHIRAL THIOUREA ORGANOCATALYSTS

presented by

KONSTANTINOS RAMPALAKOS

has been accepted towards fulfillment
of the requirements for the

PhD. degree in CHEMISTRY

 

 

 
   

Major Professor’s Signature

01/7/67

Date

MSU is an afiinnative-action, equal-opportunity employer

~.-.—.—-.-a—n—A—-----—--n—--4------—.-:-.— - -

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

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

5/08 K:lProj/Acc8Pres/ClRC/DateDueindd

 

ASYMMETRIC SYNTHESIS WITH VAPOL DERIVATIVES AND NOVEL
CHIRAL THIOUREA ORGANOCATALYSTS

By

Konstantinos Rampalakos

A DISSERTATION

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

DOCTOR OF PHILOSOPHY
Department of Chemistry

2008

ABSTRACT

ASYMIVIETRIC SYNTHESIS WI'I'H VAPOL DERIVATIVES AND NOVEL
CHIRAL THIOUREA ORGANOCATALYSTS

By
Konstantinos Rampalakos

This thesis describes the investigation of the catalytic asymmetric imino—aldol
reaction using VAPOL-derived catalysts, and the development of novel chiral thiourea
catalysts for asymmetric transformations, specifically for the aza-Henry reaction and for
the Michael-type addition of nitroalkanes to nitroolefins.

Concerning the first project (Chapters 1 and 2), a derivative of the VAPOL ligand,
7,7’—dimethylVAPOL, was synthesized according to the method developed in our
laboratories for vaulted biaryls based on the Snieckus phenol synthesis. The new ligand
was designed based on CPK models for the suspected intermediates of the iminoaldol
reaction between imines with ketene silyl acetals catalyzed by a Zr-VAPOL catalyst.
Attempts to try the new ligand in reactions where VAPOL wasn’t very efficient afforded a
significant improvement on the enantioselectivities. These results not only demonstrate the
efficiency of the new ligand but also confirm the model that we proposed for the reaction.

The rest of the Thesis, and the major amount of work described in this disertation,
concerns the design, synthesis and application of novel chiral thiourea catalysts for
asymmetric synthesis. Concerning the aza-Henry reaction (Chapters 3 and 4) we evaluated
new potential catalysts with two thiourea moieties for simultaneous activation of both
reaction components via H-bonding. We identified a new promising catalyst based on
BINAM as chiral schaffold. Optimization of the reaction conditions for the aza-Henry

reaction led to the development of an efficient protocol for the reaction of N-Boc imines

with nitroalkanes. The yields for a variety of substituted imines were modest to good, and
the asymmetric inductions were high. To account for the enantioselective reaction of N-
Boc imines with nitroalkanes we proposed an intermediate where one thiourea moiety of
the catalyst is bound to the imine while the other one is hydrogen bonding to
nitromethane.

Chapter 5 describes the application of another novel thiourea catalyst to the
addition of nitroalkanes to nitroolefins. First, the unasymmetric addition was studied with
simple achiral aromatic thioureas as promoters. As a result, we developed the first
efficient protocol for diastereoselective synthesis of unasymmetric syn 1,3-dinitro
compounds. Subsequently, the asymmetric variant of the reaction was investigated. A
novel thiourea/DMAP bifunctional catalyst was made, based again on binaphthyl diamine.
The conjugate addition using this catalyst followed a unique selectivity trend according to
which, the less catalyst used, the higher the enantioselectivity of the reaction. Investigation
of this trend with a series of studies showed that the nitroalkane/catalyst ratio is the
determining factor for the selectivities. Optimization of the reaction conditions led to the
first highly efficient protocol for the organocatalytic direct conjugate addition of
nitroalkanes to nitroolefines with only 2% catalyst loading and ee’s in the mid-90’s for a
range of electron deficient and rich nitrostyrenes.

Finally, the last chapter (Chapter 6) discusses some efforts towards the synthesis of
new biaryl-based thiourea catalysts using vaulted biaryls as schaffolds, in an effort to
extend the scope and applicability of the thiourea catalysts that were described in the

previous chapters.

ACKNOWLEDGMENTS

I would like to thank Prof. Wulff for being a great advisor. His broad knowledge
and his valuable advice helped a lot in carrying out the reaserch presented herein.

It was very exciting to have Prof. Borhan around too. He is a very fun guy and has
provided a lot of help and support (despite the centuries-old Greek-Persian conflict).

I was lucky to meet great people in the group all these years. Manish Rawat and
Billy Mitchell were gret labmates who didn’t only gave valuable help in the beginning, but
also made life in the lab very amusing. The same is true for Glenn Phillips as well, and for
Zhengsheng Ding. Ding is probably one the funniest guys I’ve met, and also a very good
friend. Aman, Victor, Alex, and other people created nice atmosphere to work in.

I am also greatfull to my friends outside the lab; Kyoungsoo, Ninneta, Chryssoula,
Justas and Kit have been my dearest friends during the last five years, and I will miss their
great friendship.

Finally, I am greatfull to my family and my beloved fiancée for their love and

continous support.

iv

TABLE OF CONTENTS

LIST OF TABLES ................................................................................... vii
LIST OF FIGURES .................................................................................... x
LIST OF SCHEMES ................................................................................. xii
CHAPTER 1 .................................................................................................................... 1
THE CATALYTIC ASYMMETRIC IMINO ALDOL REACTION USING VAPOL-
DERIVED CATALYSTS ............................................................................................. l
1.1 INTRODUCTION ............................................................................................... l

1.2 THE IMINO ALDOL REACTION USING Zr-VAPOL CATALYST ................. 2

1.3 CAN A MODIFICATION ON THE CATALYST STRUCTURE EXPAND THE
SCOPE OF THE IMINO ALDOL REACTION AND PROBE ITS MECHANISM? 8

CHAPTER 2 .................................................................................................................. 12
SYNTHESIS OF (S)-7,7’-DIMETHYL VAPOL AND EVALUATION OF THE
LIGAND IN THE IMINO ALDOL REACTION ........................................................ 12
2.1 SYNTHESIS OF (S)-7,7’-DIMETHYL VAPOL ............................................... 12
2.1.1 Synthesis of 6-methyl-2-phenyl-phenanthrol .................................................. 12
2.1.2 Oxidative coupling of 6-methyl-2-phenyl phenanthrol and deracemization of
7,7’-dimethyl VAPOL. ........................................................................................... 16
2.2 EVALUATION OF 7,7’-DIMETHYL VAPOL IN THE IMINO ALDOL
REACTION AND COMPARISON WITH THE PARENT LIGAND ..................... 21
2.2.1 Reactions using the unsubstituted imine 2 ....................................................... 21
2.2.2 Reactions using the ketene acetal 3b ............................................................... 24
2.3 CONCLUSION ................................................................................................. 27
CHAPTER 3 .................................................................................................................. 28
THE CATALYTIC ASYMMETRIC AZA-HENRY REACTION .................................. 28
3.1 INTRODUCTION ............................................................................................. 28
3.2 METAL CATALYZED PROTOCOLS ............................................................. 29
3.3 ORGANOCATALYTIC APPROACHES ......................................................... 33
3.3.1 Chiral Proton Promoted Aza-Henry Reaction ................................................. 33
3.3.2 Cinchona alkaloid-based phase-transfer catalysts ............................................ 34
3.3.3 Chiral thiourea catalysts ................................................................................. 36
3.3.4 Bis-Thiourea Catalysts for the aza-Henry reaction and related transformations43
CHAPTER 4 .................................................................................................................. 45
A NOVEL BIS-THIOUREA ORGANOCATALYST FOR THE ASYMMETRIC AZA-
HENRY REACTION ................................................................................................. 45
4.1. IDENTIFICATION OF THE APPROPRIATE BIS-THIOUREA CATALYST
AND IMINE SUBSTRATE .................................................................................... 45
4.2. OPTIMIZATION OF REACTION CONDITIONS .......................................... 51

4.3. SYNTHESIS AND EVALUATION OF DERIVATIVES OF CATALYST 52.57

4.4. SUBSTRATE SCOPE ...................................................................................... 64

4.5. AN INSIGHT INTO THE MECHANISM ........................................................ 68
4.5. CONCLUSION ................................................................................................ 74
CHAPTER 5 .................................................................................................................. 75
THE FIRST HIGHLY ENANTIOSELECT IVE ORGANOCATALYTIC DIRECT
ADDITION OF NITROALKANES TO NITROOLEFINES ...................................... 75
5.1 . INTRODUCTION ............................................................................................ 75
5.2 PROTOCOLS FOR THE CONJUGATE ADDITION OF NITROALKANES TO
NITROOLEFINS .................................................................................................... 76
5.3. THE CONJUGATE ADDITION OF NOTROALKAN ES TO NITROOLEFINES
BY MEANS OF THIOUREA CATALYSIS ........................................................... 82
5.3.1. Development of the thiourea catalyzed non-asymmetric addition ................... 84
5.3.2 Development of the asymmetric conjugate addition of nitroalkanes to
nitroolefines ............................................................................................................ 95
5.3.2.1 Identification of the optimum chiral thiourea catalyst ................................... 97
5.3.2.2 Preliminary efforts for optimization ........................................................... 103
5.3.2.3 Effect of catalyst loading: discovery and investigation of a highly unusual
selectivity trend .................................................................................................... 107
5.3.2.4 Optimization of conditions and substrate scope .......................................... 124
5.3.2.5 Proposed Stereochemichal Model for the asymmetric addition using thiourea
82 ......................................................................................................................... 131
5.4 SUMMARY AND CONCLUSION ................................................................. 134
CHAPTER 6 ................................................................................................................ 135
TOWARDS VAULTED BIARYL-DERIEVED THIOUREAS AND THE SYNTHESIS
OF VANAM ............................................................................................................ 135
6.1 Examination of Buchwald amination as a key step in the synthesis of 91 ......... 136
6.2. Examination of a synthesis of 91 through oxidative coupling of arylamines.... 141
6.3. Attempts to synthesize 91 using the Bucherer reaction .................................... 147
6.4. Future Work ................................................................................................... 150
EXPERIMENTAL PROCEDURES ............................................................................. 153
Experimental procedures for Chapter 2 ......................................................................... 153
Experimental Procedures for Chapter 4 ........................................................................ 165
Experimental Procedures for Chapter 5 ........................................................................ 194
Experimental Procedures for Chapter 6 ........................................................................ 209
REFERENCES ............................................................................................................. 214

vi

LIST OF TABLES

Table 1. Reactions of Imine 2 with ketene acetal 3a using various catalysts ...................... 3
Table 2. Temperature dependence of the asymmetric induction in 10 ............................... 7
Table 3. Small scale oxidative dimerization of 20 and 26 ................................................ 17
Table 4. Examination of various reaction times for the oxidative coupling of 26 ............. 19
Table 5. Deracemization of 7,7’-dimethyl VAPOL ......................................................... 20

Table 6. Comparison of the efficiency of (S)-VAPOL and (S)-7,7’-dimethyl VAPOL
in the asymmetric induction in compound 4 .................................................................... 22

Table 7. Temperature dependence on the asymmetric induction of 27

using Zr-VAPOL catalyst ............................................................................................... 25
Table 8. Temperature dependence in the asymmetric induction of

27 using Zr-7,7’-dimethylVAPOL catalyst ..................................................................... 26
Table 9. Palomo’s approach using catalyst 45 ................................................................. 35
Table 10. Takemoto’s Aza-Henry reaction of phosphinoyl imines .................................. 38
Table l 1. Takemoto’s aza—Henry reaction with N-Boc imines ........................................ 39
Table 12. Jacobsen’s Aza-Henry reaction using 49 ......................................................... 41
Table 13. Ricci’s aza-Henry reaction using catalyst 50 ................................................... 42
Table 14. Evaluation of the bis—thioureas in the aza-Henry reaction

of N-Boc imine 41 .......................................................................................................... 49
Table 15. Solvent screening for the aza-Henry reaction with catalyst 52 ......................... 52
Table 16. Effect of the base on the reaction .................................................................... 53
Table 17. Optimization of temperature and catalyst loading ............................................ 56
Table 18. Synthesis of thioureas 56a-61a ....................................................................... 58
Table 19. Synthesis of aromatic N-Boc imines 41 ........................................................... 64

vii

Table 20. Reaction scope of the aza-Henry reaction of imines 41a-k using catalyst 52 66

Table 21. Aza-Henry reaction of imine 41 using catalyst 66 ........................................... 71
Table 22. Alcantara’s method for the synthesis of 1,3-dinitroalkanes .............................. 77
Table 23. The first asymmetric addition of nitroalkanes to nitroolefines ......................... 78
Table 24. Wang’s protocol using catalyst 74 ................................................................... 80
Table 25. Maruoka’s formal addition using catalyst 75 ................................................... 81
Table 26. The conjugate addition of nitroalkanes to nitrostyrene catalyzed

by thioureas .................................................................................................................... 88
Table 27. Optimization of conditions for the reaction using thiourea 47 .......................... 90

Table 28. Reaction scope for the thiourea catalyzed addition of

l-nitropropane to B-nitrostyrene ..................................................................................... 92
Table 29. Reactions of nitroalkanes with fl-nitrostyrene using catalysts 52 and 55 .......... 97
Table 30. Wang’s catalyst screening ............................................................................... 99
Table 31. Evaluation of the new catalyst 82 in the conjugate addition ........................... 103
Table 32. Preliminary optimization of conditions using catalyst 82 ............................... 105

Table 33. An unusual selectivity trend observed for the catalyst loading study for 82 109

Table 34. Concentration study of the reaction ............................................................... 114
Table 35. Cat. loading study with stable nitroalkane/catalyst ratio ................................ 116
Table 36. Effect of nitroalkane equivalents ................................................................... 117
Table 37. Survey of additives in the reaction with catalyst 82 ....................................... 121
Table 38. Optimization of reaction variables ................................................................. 125

Table 39. Substrate scope of the asymmetric addition of nitroalkanes to

nitrostyrenes using catalyst 82 ...................................................................................... 127
Table 40. Reaction with 2-nitropropane using catalyst 82 ............................................. 130
Table 41. Initial efforts for the synthesis of 96 .............................................................. 138

viii

Table 42. Bucherer reactions of VANOL and VAPOL ................................................. 149

ix

LIST OF FIGURES

Figure l. The VAPOL chiral ligand .................................................................................. 1
Figure 2.1minoaldol reaction between an imine and a ketene silyl acetal .......................... 2
Figure 3. CPK models of Intermediate 7a and of two imines complexed with it ................ 6
Figure 4. Structure of ketene acetal 3b and products 27a, 27b ........................................... 8

Figure 5. CPK models of complexes of VAPOL and 7,7’-dimethyl VAPOL
derived catalysts with imines 2 and 8 .............................................................................. 10

Figure 6. Steric strain between imine 2 and the 7-Me of dimethylVAPOL as
a possible explanation for the stereochemical outcome of the reactions in Table 6. ......... 24

Figure 7. Design of Takemoto’s catalyst for the activation of the nitro group in the

aza-Henry reaction. ........................................................................................................ 37
Figure 8. Jacobsen’s thiourea 49 ..................................................................................... 41
Figure 9. Ricci’s thiourea 50 ........................................................................................... 42
Figure 10. The concept of bis-thiourea catalysis ............................................................. 43
Figure 11. The chiral ligand BINAM ............................................................................. 45
Figure 12. Structures of bis-thiourea catalysts ................................................................. 46
Figure 13. X—ray structure of compound 53 .................................................................... 51
Figure 14. Schreiner’s thiourea 47 .................................................................................. 61
Figure 15. lH-NMR Study as a probe for imine-catalyst interaction ................................ 69
Figure 16. Postulated mechanism for the reaction using 66 with triethylamine ................ 72
Figure 17. Proposed mechanism for Du’s asymmetric addition ....................................... 79
Figure 18. Maruoka’s catalyst for the formal addition to nitroalkenes ............................. 81
Figure 19. The substituted aromatic thioureas used in this work ...................................... 84
Figure 20. Design of a novel bifunctional catalyst based on BINAM ............................ 100

Figure 21. Plot of the asymmetric induction for syn — 803 versus the catalyst loading .. 110

Figure 22. Possible scenarios to account for the catalyst loading effects ....................... 112
Figure 23. Plot of optical purity of 82 versus optical purity of syn-803 ......................... 119
Figure 24. Temperature study of the asymmetric addition ............................................. 123
Figure 25.Postu1ated stereochemical model ................................................................. 132
Figure 26. Vaulted biaryl thioureas and diamines ......................................................... 135

xi

LIST OF SCHEMES

Scheme 1. Proposed Mechanism for the Imino Aldol Reaction Using

Zr-VAPOL catalyst .......................................................................................................... 4
Scheme 2. Synthesis of naphthols through annelation of o—allyl benzamides ................... 12
Scheme 3. A tentative mechanistic rationalization for the anionic benzannelation ........... 13
Scheme 4. The synthesis of VAPOL using an anionic benzannelation as a key step ........ 14
Scheme 5. The Synthesis of 7-methyl-2-phenyl-4-phenanthrol 26 .................................. 15
Scheme 6. Oxidative dimerization of phenanthr0120 for the preparation of VAPOL ...... 17
Scheme 7. The Henry and the Aza-Henry reaction .......................................................... 28
Scheme 8. Useful transformations of fi-nitroamines ....................................................... 29
Scheme 9. Anderson’s aza-Henry reaction ...................................................................... 29
Scheme 10. Shibasaki’s protocol using the heterobimetallic complex 32 ........................ 30
Scheme 11. Jorgensen’s aza-Henry reaction ................................................................... 31

Scheme 12. Palomo’s aza-Henry reaction assisted by Zn(OTf)2
and N-methylephedrine .................................................................................................. 32

Scheme 13. Johnston’s aza-Henry reaction ..................................................................... 33

Scheme 14. Proposed reaction process for the Takemoto’s thiourea-catalyzed

aza-Henry reaction ......................................................................................................... 40
Scheme 15. Synthesis of bis-thiourea catalysts 52, 55 ..................................................... 47
Scheme 16. Synthesis of thioureas 53, 54 ....................................................................... 48

Scheme 17. A tentative mechanistic hypothesis that could explain the effect of
the base on the reaction .................................................................................................. 54

Scheme 18. Results for the screening of different thiourea derivatives in
the aza-Henry reaction at rt ............................................................................................. 60

Scheme 19. A possible mechanism to account for the high induction obtained with 52 62

xii

Scheme 20. Possible reaction intermediates .................................................................... 68
Scheme 21. Synthesis of thiourea 66 ............................................................................... 70
Scheme 22. A plausible stereochemical model for catalyst 52 ......................................... 73

Scheme 23. The conjugate addition of nitroalkanes to nitroolefines as a useful
route for the construction of valuable functionalities ....................................................... 76

Scheme 24. The conjugate addition of nitroalkanes to nitroolefines promoted

by thioureas .................................................................................................................... 83
Scheme 25. Synthesis of aromatic thioureas 47, 76-79 .................................................... 85
Scheme 26. Rate acelaration of the conjugate addition using thioureas 47 , 76, 77 ........... 86

Scheme 27. A possible explanation for the preference of the syn diastereomer

formation in the reaction catalyzed by thiourea 47 .......................................................... 94
Scheme 28. Bis-thiourea approach for the conjugate addition ......................................... 95
Scheme 29. Bifunctional thiourea approach for the conjugate addition ........................... 96
Scheme 30. 2-Aminopyridinium catalyst for nitroalkene activation .............................. 10]
Scheme 31. Synthesis of catalyst 82 ............................................................................. 102
Scheme 32. Reaction with trans-a -methyl-B -nitrostyrene ............................................. 129
Scheme 33. Synthesis and reaction of the aliphatic nitroolefine 88 ............................... 13]
Scheme 34. Buchwald amination of triflate 95 .............................................................. 136
Scheme 35. Proposed synthesis of 93 using Buchwald amination as a key step ............. 137
Scheme 36. Synthesis of VANOL triflate ..................................................................... 137
Scheme 37. An attempt for Buchwald amination of 95 using CSZCO3 ........................... 139
Scheme 38. Synthesis and amination of triflate 100 ...................................................... 140
Scheme 39. Oxidative coupling of 102 using CuCl2 ...................................................... 142

Scheme 40. Postulated mechanism for the CuCl2 mediated oxidative
coupling of substituted anilines ..................................................................................... 142

xiii

Scheme 41. Attempted oxidative coupling of 101 using CuCl2 ..................................... 144

Scheme 42. Synthesis and attempted oxidative coupling of 111 .................................... 145
Scheme 43. Attempted oxidation under neat conditions ................................................ 146
Scheme 44. The Bucherer reaction for the synthesis of 104 and 102 ............................. 147
Scheme 45 Bucherer reaction of BINOL ...................................................................... 148
Scheme 46. Cho’s synthesis of BINAM derivatives ...................................................... 150
Scheme 47 Thermal rearrangement of 105 for the synthesis of BINAM ........................ 151
Scheme 48. Proposed synthesis of 91 through hydrazide rearrangement ....................... 15]
Scheme 49. Proposed synthesis of 130 .......................................................................... 152

xiv

HAPTER 1

THE CATALYTIC ASYMMETRIC IMINO ALDOL REACTION

USING VAPOL-DERIVED CATALYSTS

1.1 INTRODUCTION

The development of stereoselective transformations for the creation of
functionalized optically active molecules is one of the ultimate goals in chemistry, and
hence the design of highly efficient catalytic asymmetric reactions has been a major focus
of organic synthesis during the last years.1 Consequently, the design and synthesis of new
effective ligands has been a great challenge in this field. The Cz-symmetric vaulted biaryl
ligand VAPOL2 (2,2’-diphenyl-[3,3’-biphenanthrene]-4,4’-diol, Figure 1) developed in
our laboratories has proven to be an excellent ligand for various important catalytic
asymmetric reactions, such as Diels Alder reactions’, aziridination reactions4 and imino

aldol reactionss.

Figure 1. The VAPOL chiral ligand

HO 0 Ph H0“ I P“
HO ‘ Ph H0 0 1’“

(S)-1 VAPOL (R)—1 VAPOL

The catalytic asymmetric imino aldol rection" (Figure 2) is one of the most
important carbon-carbon-bond forming reactions since it provides access to chiral B-amino
ketones or esters, which are versatile chiral building blocks for the synthesis of many
nitrogen containing biologically important compounds including B-arnino acids and
lactams.7 In 1997 Kobayashi reported the first asymmetric Manich type reaction between
an aldimine and a ketene silyl acetal using substoichiometric amounts of a Zr-BINOL
species as catalyst.8 Their method involves a catalyst generated from zirconium(IV) tert-
butoxide and two equivalents of (R)-6,6‘-dibromo BINOL (BINOL=1,1'-binaphth-2-ol).

In 2001 we developed an unusually robust Zr-VAPOL catalyst5 that could provide
even higher selectivities, and and which retained its selectivity over a range of
temperatures from 25 to 100°C, resulting in one of the most efficient protocols for the

name reaction to date.

Figure 2.1minoaldol reaction between an imine and a ketene silyl acetal

 

.Rg R
N 3 oms catalyst TMS.
_ NH 0
R 2' + >—<

1.2 THE IMINO ALDOL REACTION USING Zr-VAPOL CATALYST

Our interest in the synthesis of chiral amines led us to investigate the use of
VAPOL-derived catalysts2 (see Figure 1) for this reaction. The catalyst was prepared by
reaction of the ligand with 0.5 eq of zirconium tetraalkoxide in the presence of 0.6 eq of

N-methyl imidazole at room temperature for one hour. It was found that VAPOL and 6,6’-

dibromo BINOL are superior to BINOL at —45°C5. When the reaction was performed at
room temperature, the asymmetric induction for VAPOL remained unchanged, while that

for dibromo BINOL and BINOL decreased (Table 1).

Table 1. Reactions of Imine 2 with ketene acetal 33 using various catalysts

0H
OTMS ca,
MKFN + >_< “" NH 0

 

 

OMe
0” Ph OMe
2 3a 4
entry ligand mol% cat T(°C) t(h) solvent yield% ee%
1 S-VAPOL 20 -45 20 toluene 92 91
2 S-VAPOL 20 25 15 toluene 94 89
3 R-BINOL 20 -45 19 DCM 80 36
4 R-BINOL 20 25 4 DCM 100 28
R—6,6’-Br2-
5 10 -45 19 DCM 87 86
BINOL
R-6,6’-Br2-
6 10 25 4 DCM 87 48
BINOL

 

Catalyst prepared from Zr(OiPr)4/i-PrOH, chiral ligand (2.2 eq) and 1.1 eq of N-methyl
imidazole (NMI) in DCM or toluene at 25°C for 1 h. Reactions were performed with 1.2 eq of
ketene acetal and 0.125M in imine. (entry 2: 0.5M in imine)

The mechanism we proposed for the catalytic cycle involves a catalyst bearing two
VAPOL ligands on one zirconium and the coordination of the o-hydroxyphenylimine to

the zirconium as a bidentate ligands. It is clear from the examination of spacefilling CPK

models that it is possible to bind two VAPOL ligands to one zirconium atom only with a

facial arrangement of the four oxygen atoms as is illustratred by structure 73 in Scheme 1.
This is supported by lH-NMR experiments on a catalyst generated from zirconium
tetraisopropoxide and VAPOL in the presence of two equivalents of N-methyl imidiazole.
A clean spectrum is only observed with two equivalents of VAPOL relative to zirconium
and the spectrum is consistent with a single C2-symmetrical species which is tentatively
identified as structure 7b bearing mutually trans NMI ligands bound to the zirconium.“10
An approach of imine 2 to the open apical position in intermediate 73 is proposed to lead

to intermediate 5 in which a phenol exchange has occurred.

Scheme 1. Proposed Mechanism for the Imino Aldol Reaction Using Zr-VAPOL catalyst

Q OTMS
OH H—
? OMe

N\
2 Uh ON 3a
b (:H ”\WO)
NMI ph
5
X
s/ 0
Mi
O/|\ 0 M30 / HO
NMI C‘xzk/ D
7a: X=open V TMS/ Ph\
7b:X=NMl NMI 0
fi 6
OTMS
NH 0
Ph OMe

4
Reaction of species 5 with the ketene acetal would give intermediate 6 and then release of

the product would regenerate the unsaturated species 73 and complete the cycle.

4

A space-filling CPK model of intermediate 7a is shown in Figure 3 and illustrates
the binding cleft that is available for docking with imine 2. When imine 2 is bound to
zirconium un the cleft, CPK models reveal that there is some freedom of movement up
and down in the cleft. The figure depicts the imine rotated up and out of the cleft as far as
possible. This conformation should lead to reduced selectivity. The model in Figure 3
predicts that the re—face will be the more shielded face of the imine and this is consistent
with our observations. The CPK model indicates that imine 8 with a methyl group ortho to
the phenol function should be sufficient to push the imine down into the cleft, thus

increasing the selectivity of the reaction.

Figure 3. CPK models of Intermediate 7a and of two imines complexed with it

 

Intermediate 7a with Intermediate 7a with
CH3
A
P“ N oI H PhAN CH3
OH
2 8

Indeed, when the ortho position was methylated there was a profound increase in
the asymmetric induction.5 The rate of the reaction of imines with ketene acetals with the
VAPOL catalyst is slower with imines generated from substituted aminophenols,
however, as indicated in Table 2, this effect can be offset by performing the reaction at a
higher temperature where greater turnover numbers are observed. The induction with
imine 8a (R=phenyl) shows absolutely no temperature dependence over the range of 25 to

100°C.

Table 2. Temperature dependence of the asymmetric induction in 10

Qon

Q 0

/\ + >=<

R \ N OMe
OH

TMS Zr/(R)-VAPOL cat.

 

 

 

NH 0
toluene ’
R/><I\0Me
8 3a 1 0

entry R mol% cat. T(C°) yield% ee%
1 Ph 20 25 100 98
2 Ph 20 65 93 98
3 Ph 20 85 94 98
4 Ph 20 100 91 98
5 Ph 2 100 95 98
6 4-Cl-Ph 2 100 90 95
7 4-MeO-Ph 2 100 85 >99
8 3,4-(MeO)2 2 100 85 96
9 l -Naphthy1 2 100 83 93

 

Reaction time: 2-3 h. (Entry 1:
time 24h)

Furthermore, the reaction at 100°C can be performed with an order of magnitude
change in catalyst loading with no loss in induction. The electronic nature of the imine has

a small effect on the induction at this temperature with a slight increase noted for a p-

Reaction time 15 h, entry 5: 5h, entries 6, 7, 8, 9: reaction

methoxy substituent and a slight decrease for a p-chloro substituent.

1.3 CAN A MODIFICATION ON THE CATALYST STRUCTURE EXPAND THE

SCOPE OF THE IMINO ALDOL REACTION AND PROBE ITS MECHANISM?

For imine 2, that bears the non-substituted arene ring, selectivities are lower than
for imines 8. Additionally, the corresponding reactions with the unsubstituted ketene
acetal 3b derived from t-butyl thioacetate (Figure 4) mediated by the same catalyst are
slower and do not display the same temperature independences. For example, it was
reported that the reaction of imine 8 when R=1-naphthyl with 3b gives 27b in 91%
enantiomeric excess at 25°C (20 mol% catalyst) and in 76% enantiomeric excess at 100°C
(5 mol% catalyst). Similarly, the reaction of imine 8 when R=Ph with 3a gives 27a in 71%

ee at 25°C (20 mol% catalyst) and 60% ee at 100°C (2 mol% catalyst) .5

Figure 4. Structure of ketene acetal 3b and products 27a, 27b.

(1“ filo“
OTMS
=< NH 0 NH O
t , t
S Bu (DA/“\Stau O S Bu
27a 0 27b

3b
To improve the selectivities of these reactions, a slight modification on the catalyst
structure was envisaged. As it was described aboved (Figure 3), the source of selectivity is
the shielding of the re-face of the imine by the phenanthrene unit of the ligand, so that the
si-face is exposed for attack by the ketene silyl acetal. This model predicts that any
modification that would make the phenanthrene unit bulkier would result in more efficient
shelding and thus in higher selectivities. Accordingly, a methyl group on the 7 position of

the phenanthrene skeleton would increase the size of the phenanthrene unit and serve to

test this model. Figure 5 shows the space-filling CPK model of the complexes between the

8

imines 2 and 8 and the 7,7’-dimethylVAPOL (11) catalyst. The CPK model suggests that
there is a significant increase in the steric shielding provided by the phenenthrol of 7,7’-
dimethylVAPOL (7-methyl shown in black-red color) compared to that of VAPOL for
both imines 2 and 8, and that such a derivative of VAPOL would be expected to increase

the selectivities for the above transformations.

Figure 5. CPK models of complexes of VAPOL and 7,7’-dimethy1 VAPOL derived
catalysts with imines 2 and 8

.. 0 .
so“

11 (S)-7,7’-dimethyl VAPOL

   

Intermediate 711 with dimethylVAPOL
imine 2 catalyst with imine 2

   

Intermediate 7 a with dimethylVAPOL
imine 8 catalyst with imine 8

The next chapter details the synthesis of 7,7’-dimethylVAPOL 11 and its
evaluation in the asymmetric imino aldol reaction. Specifically, a comparison between
VAPOL and 7,7’-dimethy1VAPOL is described for the reactions between imine 2 and
acetal 3a, and between imines 8 and a ketene acetal 3b derived from t-butyl thioacetate.
The ultimate purpose of this project is on one hand the expansion of the scope and
synthetic utility of these reactions and on the other hand the justification of our model and

the proposed mechanism for the reaction.

11

CHAPTER 2

SYNTHESIS OF (S)-7,7’-DIMETHYL VAPOL AND EVALUATION

OF THE LIGAND IN THE IMINO ALDOL REACTION

2.1 SYNTHESIS OF (S)—7,7’-DIMETHYL VAPOL

2.1.1 Synthesis of 6-methyl-Z-phenyl-phenanthrol

In 1986, Snieckus and his coworkers reported that o-allylbenzamides undewent
MeLi induced cyclization to give l-naphthols". o—Allylbenzamides were prepared by the
directed ortho-metallation of tertiary benzamides followed by the reactions with allyl
bromide12 (Scheme 2). This chemistry has subsequently been used to prepare various

substituted naphtholsl 3.

Scheme 2. Synthesis of naphthols through annelation of o-allyl benzamides

 

 

NR2 1. s-BuLi NR2 OH
2- MgBrz MeLi
O 3. allyl bromide _ O : DC
71% \ 86%
12 13 14

The precursor 13 of the annelation product was synthesized using the extensively
reported directed ortho metallation methodology developed by Snieckus’ group.
According to that method —- which is focused on tertiary amides as effective ortho
metallation directors” — lithiation of aromatic benzamide 12 is directed ortho to the amide
group and subsequent reaction with MgBr2 affords an arylmagnesium bromide species as

the transmetallation reaction product. The lithium to magnesium transmetallation in the

12

first step of the sequence is necessary for the successful ortho allylation, since the lithiated
species failed to give the ortho substituted product”.

A mechanistic rational for the transformation is illustrated in Scheme 3. Formation
of the naphthol 14 may occur via intermediate 15 by two pathways. The first involves
direct cyclization from anion 16, while the second involves formation of

benzocyclobutane 17 followed by a [1,3]-sigmatropic rearrangement”.

Scheme 3. A tentative mechanistic rationalization for the anionic benzannelation.

NR2 LIO E
t
o, ,\ 2
7’: ‘—-_- /
Li +
16 17
+Li —o NH2
15

A few years ago, Dr Su Yu from our group developed a new strategy for the
synthesis of VAPOL that was inspired by the Snieckus’ naphthol synthesis. The synthesis
of 2-pheny1-4-phenanthrol 20 (the precursor to VAPOL) was realized from a MeLi
induced cyclization of diisopropyl 6-methyl-2-(2-phenylallyl)naphthamide 19, which was
obtained by a directed o-allylation of 1-naphthamide 18 (Scheme 4). This unprecedented
synthesis of a 4-phenanthrol using the Snieckus method established a new efficient

method for the synthesis of vaulted biphenanthrols.

13

The method Dr. Yu developed has a few modifications from Snieckus’ original
protocol. For the synthesis of naphthols, Snieckus used diethyl-benzamide, TMEDA as an
additive, and commercially available MgBrz. Dr Su Yu found that, for the synthesis of the
4—phenanthrol 20, diisopropyl naphthamide works better than diethyl naphthamide,
TMEDA is not necessary, and MgBr2 has to be prepared in situ in order for the
transformation to be most efficient. The in situ preparation of MgBr2 can be achieved by
reacting Mg turnings with 1,2-dibromoethane in EtzO. In this way higher and more

reproducible yields of compound 19 can be obtained.

Scheme 4. The synthesis of VAPOL using an anionic benzannelation as a key step

0 N1iPr>2 1)s-BuLi o N(iPr)2

2) Mg/C2H4Br2 0
00 3) Bromo Methyl Styrene A 00

75-80% 19
.. .. e
MeLi, THF air HO Ph

00 Ho P,
0
2° 00
1

 

18

 

Su Yu’s method for the synthesis of 2-phenyl-4-phenanthrol 20 was applied to the
synthesis of 7-Methyl-2-phenyl-4-phenanthrol 26, the precursor to 7,7’-dimethyl VAPOL.

The synthesis of this new phenanthrol is outlined below (Scheme 5).

l4

Scheme 5. The Synthesis of 7-methyl-2-phenyl-4-phenanthrol 26

COOH

 

 

 

 

 

o SOCI2
7 Benzene, DMF 0O
Me 00 M.
21 ” 5 /° 22 23
O N iPr) ,
. ( 2 1) s-BuLI o N(iPr)2
(IPr)2NH. EtsN _ 2) Mg/C2H4Br2 O
CH20|2, 0°C 00 3) Bromo Methyl Styrene , 00

74 % from 22 Me 24 ”75 % Me

25

“0 Ph
MeLi, THF
DO

80 % Me

 

26

The first step involves a formal Diels Alder reaction between toluene and furoic
acid to furnish 6-methylnaphthoic acid 22. The product was obtained in ~5% yield. The
majority of the mass balance for this reaction (as was described in the original protocol by
Price et a1.) is a mixture of high molecular weight acids of complex structure, formed by
the condensation of two or three molecules of toluene with one of furoic acid”. The larger
acids could be separated by extraction of the reaction mixture with aqueous sodium
bicarbonate (the high molecular weigh acids could not be extracted with this method). The
6-methylnaphthoic acid was obtained from the resulting mixture with barium hydroxide,

since the barium naphthoate of 22 (and furoate) dissolved in hot water while the high

15

molecular weigh acids formed an insoluble barium salt. Finally, after crystallization from
ethyl acetate, 8 grams of quite pure product were available for the next step.

Compound 22 was then treated with 1.1 eq of SOCl2 and 0.44 eq of DMF in
benzene to afford 6-methylnaphthoyl chloride 23. Assuming that the transformation was
quantitative, the crude product was taken on to the next step, where it was treated with 1.2
eq of (i—Pr)2NH and 1.2 eq of RN in DCM to give diisopropyl-6-methyl naphthamide 24
in 74% yield from the naphthoic acid, which is consistent with the parent ligand synthesis.

The stage was set for the key step of the sequence, the ortho-allylation /
cyclization reactions that would give the desired phenanthrol. Accordingly, following the
modified procedure of Dr. Yu, the 6-methyl naphthamide 24 was treated with 1.1 eq of s-
BuLi, 3 eq of freshly prepared MgBr2 and 2 eq of a-bromomethyl styrene to furnish
diisopropyl 6-methyl-2-(2-phenylallyl)naphthamide 25 in 75% yield.

Finally, the cyclization of the o-allyl naphthamide 25 proceeded smoothly upon
treatment with MeLi in THF at —78°C to afford the desired 7-methyl-2-phenyl-4-

phenanthrol 26 in 80 % yield.

2.1.2 Oxidative coupling of 6-methyl-2-phenyl phenanthrol and

deracemization of 7,7’-dimethyl VAPOL.

In 1996 we showed that racemic VAPOL can be prepared efficiently on a 10 g
scale by oxidative dimerization of 2-phenyl-4-phenanthrol 20, simply by heating the neat
solid at 190°C in a beaker for ~24h, provided that efficient stirring is supplied in order to

maintain sufficient contact of the solid with air). With this method, up to 89% yield of

16

VAPOL can be obtained after a simple extraction of the reaction mixture with ethyl

acetate (Scheme 6).

Scheme 6. Oxidative dimerization of phenanthrol 20 for the preparation of VAPOL

 

2° (2333.272, 00 1
Since the synthesis of compound 26 that was described in Scheme 5 afforded only
a few hundreds of milligrams of desired product, the oxidative coupling for the synthesis
of racemic 7,7’-dimethyl VAPOL had to be performed on a very small scale. Following
the procedure described by Bao, oxidative coupling of 2-phenyl-4-phenanthrol 20 and 7-

methyl-2-pheny1-4—phenanthrol 26 was examined on small scales (~0.3-0.5 g). The

preliminary results are shown below in Table 3.

Table 3. Small scale oxidative dimerization of 20 and 26
R O
HO Ph 0
O 185-190°C HO ph
00 air HO O P“

 

 

 

R
R 00
R scale t (h) yield %
H (20) 300 mg 39 48
Me (26) 490 mg 46 33

 

l7

It was found that the dimerization of 2-phenyl-4-phenanthrol on a 300 mg scale
gave a yield of 86% after extraction of the reaction mixture with EtOAc. Although the lH-
NMR was clean, TLC analysis showed some minor byproduct spots and baseline material,
probably polymeric products. The isolated yield after column chromatography was only
48% (Table 3, entry 1), indicating that the dimerization on 300 mg scale behaves
differently than on 10 g scale. This is probably due to insufficient stirring. As the
phenanthrol was slowly oxidized to VAPOL, the reaction mixture solidified and finally
the stirring stopped due to the high melting point of VAPOL(~250°C) and the small
amount of material that was present. In a similar fashion, the dimerization of 7-methyl-2-
pheny1-4-phenanthrol gave 33% of 7,7’dimethyl VAPOL after 46 h of heating.

Since the oxidative dimerization of the phenanthrols gave a significant amount of
byproducts, various reaction times were examined for the reaction of 7-methyl-2-phenyl-
4-phenanthrol in an effort to increase the yield. The results are summarized in the

following table.

18

Table 4. Examination of various reaction times for the oxidative coupling of 26

HO Ph I O
O 180-190°C HO Ph

 

 

26 11
entry t (h) dimer/monomer %yield 11

l 36 7/1 30
2 19 2.7/1 40
3 10 2.3/1 40

4 l 1 1 .5/1 51
5 8 1 .1/1 45

6‘ 10 nd 55

7" 10 nd 67'"

 

'Reaction performed in an Erlemeyer flask in order to increase the reactive surface.

"Reaction repeated using the recovered starting material from entry 6. "'Total yield from entries 6
and 7.

It is obvious that by decreasing the time of the reaction, the yield of the desired
product increases. It was found that for heating times of >10 h a major byproduct with Rf
very similar to that of the desired product was formed (entries 1 and 2) and that compound
(not characterized) probably comes from decomposition of 7,7’-dimethyl VAPOL. By
decreasing the time of heating to 10-11 h, the decomposition can be prevented, and even if
the conversions are lower, the isolated yields are higher. The starting material for these

reactions can be recovered and used again. For example, as it is shown in entry 7, an

19

isolated yield of 67% can be obtained based on the recovery of starting material and re-
subjection in the reaction conditions.

Recently Yu et al. found that the deracemization of VAPOL can be readily
achieved in the presence of a copper complex of (-)-sparteine'°. The optimal procedure
involves the in situ generation of copper(II) and leads to the formation of (S)-VAPOL in
>99% ee from the racemate.

Yu’s conditions were used to deracemize 7,7’-dimethyl VAPOL. The results are

shown in the table below.

Table 5. Deracemization of 7,7’-dimethyl VAPOL

O CuCl, (-)—Sparteine 0
Ph

 

 

 

HO ultrasound _ HO ph
H0 Ph MeOH/DCM H0 P“
l 25°C ‘1
1 1 (3)4 1
entry purity of 11 yield% %ee
1 purified once 65 97.8
2 purified twice' 80 99.0

 

'A second column chromatography was performed before the reaction to ensure that all
yellow impurities were removed.

It was apparent from the above results that the purity of the ligand to be
deracemized is of crucial importance in order for the reaction to afford enantiopure

product (>99%ee).

20

It was also found that the enantioselectivity of the chiral ligand could be further
enriched by crystallization from methylene chloride/hexanes. Starting from material of
97.8% ee, the first crop of crystals (~50%) were found to have 99.6% ee, indicating that

the (S)-compound is less soluble than the racemic, in contrast to VAPOL.

2.2 EVALUATION OF 7,7’-DIMETHYL VAPOL IN THE IMINO ALDOL

REACTION AND COMPARISON WITH THE PARENT LIGAND

As was mentioned on chapter 1 (Figure 4), the purpose of synthesizing and
studying 7,7’-dimethyl VAPOL in the imino aldol reaction was to test weather the methyl
substitution on the 7 and 7’ positions can enhance the asymmetric induction of substrates
that didn’t afford excellent selectivities when VAPOL was used as a ligand.

The reactions that were chosen as models for the comparison were: (i) Reactions of
the ketene acetal 3a with the unsubstituted imine 2 derived from 2-aminonaphthol (Table
6), and (ii) Reactions using the acetal 3b derived from t-butyl thioacetate with imines 8

derived from 2-amino-4,6-dimethyl phenol (Tables 7 and 8).

2.2.1 Reactions using the unsubstituted imine 2

With (S)-7,7’-dimethyl VAPOL in hand, experiments for the comparison of the
new ligand with (S)-VAPOL were conducted for the reaction of imine 2 using ketene
acetal 3a. The results for the comparison of the two ligands at r.t. and at -45°C are

summarized Table 6.

21

Table 6. Comparison of the efficiency of (S)-VAPOL and (S)-7,7’-dimethy1 VAPOL in
the asymmetric induction in compound 4. (a)

CH
Q OTMS cat.
A + _ -—-—> NH 0
Ph N > (We

 

 

0“ Ph OMe
2 3a 4
entry ligand conditions yield% (b) ee% (c)
l (S)-VAPOL rt, 0.5M, Toluene 93 89
2 (S)-MeVAPOL rt, 0.5M, Toluened 94 75
3 (S)-MeVAPOL rt, 0.5M, DCM 100 52
4 (S)-VAPOL rt, 0.125M, Toluene 95 89
5 (S)-MeVAPOL rt, 0.125M, Toluene 97 86
6 (S)-VAPOL -45°C, 0.125M, Toluene 92 91
7 (S)-MeVAPOL -45°C, 0.125M, Toluene 90 91

 

(a) Catalyst prepared from Zr(OiPr),/i-PrOH, chiral ligand (2.2 eq) and 1.1 eq of N-methyl
imidazole (NMI) in DCM or toluene at 25°C for l h. Reactions were performed with 1.2 eq of
ketene acetal and were run for 24h. (b) Isolated yield after column chromatography. (c)
Determined throuh chiral HPLC analysis. ((1) The ligand was not completely soluble for the given
conditions.

In the first attempt to compare the two ligands at r.t., it was found that 7,7’-
dimethyl VAPOL was not completely soluble in toluene when the reaction was performed
at 0.5M in imine (entry 2), and presumably that’s the reason for the significant drop in the
ee (75%). The reaction in DCM removed the problem of solubility of the new ligand
(entry 3) but the reaction was probably too fast (considering that some evaporation of the

volatile solvent may took place) to be selective, affording an even bigger decrease in the

selectivity (entry 3, 52% ee).

22

It was then realized that high concentrations are not necessary for the reaction to
go to completion in toluene and that lower concentrations should allow the new catalyst to
dissolve completely, allowing for an accurate comparison with VAPOL. The optimum
conditions for the reactions — concerning catalyst solubility and reaction rate — were met
when 0.125M imine was used in toluene. The new catalyst was completely soluble in
0.125M toluene, but, as shown in entries 4 and 5, the enantiomeric excess was slightly
lower than that obtained using (S)-VAPOL. Finally, when the temperature was decreased
to -45°C, the stereochemical outcome of the two ligands was essentially the same (entries
6, 7).

This preliminary comparison shows that the new ligand can be as efficient as (S)-
VAPOL at low temperatures for this particular reaction, but at room temperature the
asymmetric induction was slightly lower. This is contrary to expectations since the design
of the ligand anticipated greater shielding of the imine and greater inductions (Chapter 1,
Figure 4). One possible explanation for this outcome is that, the methyl substituent at the 7
position of the phenanthrene ring is bulky enough to make close contacts with the imine,
preventing it from properly fitting down into the chiral pocket. In other words, steric
hindrance between the phenyl group of the imine and the 7-methyl group of the ligand
might be responsible for failure of the imine from properly binding with the catalyst with
the result that there is an increase in the exposure of the re-face of the C-N double bond to

attack by the ketene acetal leading to decreased selectivity.

23

Figure 6. Steric strain between imine 2 and the 7-Me of dimethylVAPOL as a possible
explanation for the stereochemical outcome of the reactions in Table 6.

    

steric

Q strain(?)

If the above arguments are true, then reactions with imine 8 should not encounter
this complication, since CPK models show that the methyl substituent ortho to the
hydroxyl group of the imine should push the imine down into the chiral pocket of the
catalyst. In this way, the o'-methyl group of the imine 8 should be expected to overcome
any problem of steric repulsion between the imine and the ligand. Thus, the rections with
imine 8 were chosen for evaluation with the new ligand. The ketene acetal 3b was chosen

because this substrate gave less than desired results with the VAPOL ligand.5

2.2.2 Reactions using the ketene aceta13b

Despite the high efficiency of the Zr-VAPOL catalyst for the imino aldol reactions
using ketene acetal 3a (Table 6), the corresponding reactions with the unsubstituted ketene
acetal 3b mediated by the same catalyst were slower and did not display the same
temperature independence. The reactions with imines 8a and 8b with ketene acetal 3b
with the Zr-(S)-VAPOL catalyst have been reported5 and were reexamined in the present
work as control experiments and the data in Table 7 are similar to those reported. The
imine 8c was also examined since it gave lower induction with ketene acetal 3 (Chapter 1,

Table 2). For imine 8a, in contrast wth the excellent enantioselection observed with ketene

24

acetal 3a at room temperature (98% ee, Chapter 1, Table 2) the reaction with 3b affords
only 83% ee (entry 1), while a slight drop is observed at 100°C (entry 2, 80% ee). For
imines 8b and 8c the enantioselectivites are also not very high at room temperature (88 %
ee and 81% ee respectively) and their decrease at high temperatures is even more

significant (80 % ee and 66 % ee, entries 4 and 6).

Table 7. Temperature dependence on the asymmetric induction of 27 using Zr-VAPOL

 

 

 

catalyst (a)
Zr-(S)-VAPOL
catalyst OH
N :
J OH OTMS NH O
A" (3b) 1
StBu Ar * SBu
Toluene
8 27
% s.m.
entry series Ar T°(C) % cat. %yield(b) %ee(c)
recovered
1 a Ph 25 20 65 23 83
2 a Ph 100 10 70 22 80
3 b l-Naphthyl 25 20 50 35 88
4 b l-Naphthyl 100 10 6O 30 80
5 c 4-Cl-C6H4 25 20 55 33 81
6 c 4-Cl-C6H4 100 10 6O 3O 66

 

(a) Catalyst prepared from Zr(OiPr)4/i-PrOH, chiral ligand (2.2 eq) and 1.1 eq of N-methyl
imidazole (NMI) in toluene at 25°C for 1 h. Reactions were performed with 1.2 eq of ketene acetal
and 0.125M in imine and were run for 24 hours. (b) Isolated yield. (c) Determined by HPLC
analysis.

25

The corresponding set of reactions using the catalyst obtained from 7,7’-

dimethylVAPOL is shown in Table 8.

Table 8. Temperature dependence in the asymmetric induction of 27 using Zr—7,7’-
dimethylVAPOL catalyst (a)

 

 

 

OH
Zr-(S)-7,7'-dimethy|VAPOL
N catalyst NH O
I >
Ar) OH OTMS Ar/I*\/U\StBu
(3b)
8 SIBU 27
Toluene
%yield %ee
entry series Ar T°(C) % cat. %s.m. recovered
(b) (C)
1 3 Ph 25 20 65 23 90
2 3 Ph 100 10 70 22 84
3 b l-Naphthyl 25 20 49 35 96
4 b l-Naphthyl 100 10 60 30 88
5 c 4-Cl-C6H4 25 20 56 32 91
6 c 4-Cl-C6H4 100 10 60 30 76

 

(a) See Table 7. (b) Isolated yield based on recovery of the starting material. (c)
Determined by HPLC analysis.

The very first attempts to evaluate the new ligand in the iminoaldol reaction of
ketene acetal 27 showed that there is a significant improvement in the enantioselectivity
compared to VAPOL. For imine 8a where Ar is phenyl the ee rises from 83% [(S)-
VAPOL] to 90% [(S)-7,7’-dimethyl VAPOL]. When the reaction is performed at 100°C
(Table 8, entry 2) the induction drops a little bit (84%), but it is still higher than that

afforded with (S)-VAPOL (80%, Table 5, entry 2). For the other imines these

26

improvements were even greater: For the naphthyl-substituted imine 7,7’—
dimethylVAPOL affords 8% higher ee than VAPOL both at rt and at 100°C (entries 3 and
4) while when R is 4-chlorophenyl the increase in the induction is 10% higher at both

temperatures (entries 5, 6).

2.3 CONCLUSION

It is obvious that the catalyst derived from 7,7’-dimethylVAPOL is quite efficient
for the imino aldol reaction with regard to the asymmetric inductions observed with both
the ketene acetals 3a and 3b.

In the first case where imine 2 was used the reason that dimethylVAPOL does not
afford higher selectivities than VAPOL is not very clear, although it is suspected that
steric strain between the imine and the 7-methyl group of the ligand may destabilize the
normal conformation of the imine in the complex.

The reactions of imines 8a-c with acetal 3b turned out to be less temperature
independent than the corresponding reactions with ketene acetal 33 (Chapter 1, Table 2),
even when 7,7’-dimethy1VAPOL was employed (Table 6), nonetheless it is exciting
though to observe that a modification on the ligand structure as small as a methyl group
can induce a remarkable enhancement in the enantioselectivity of these transformations.
Additionally, these results not only demonstrate the efficiency of the ligand but also help
to confirm the proposed model for the reaction. The enhancement of selectivities achieved
both through the modification of the imine (Chapter 1) and the modification of the catalyst

(Chapter 2) agree with our proposed catalyst structure and its mechanism.

27

HAPT ER

THE CATALYTIC ASYMMETRIC AZA-HENRY REACTION

3.1 INTRODUCTION

The development of stereoselective carbon—carbon bond forming reactions that
create contiguous stereogenic centers bearing heteroatom functionality can provide
valuable building blocks for organic synthesis. The aza-Henry (nitro-Mannich) reaction,
that is, nucleophilic addition of nitroalkanes to imines (Scheme 7), is such a useful
carbon—carbon bond-forming process. The resulting B-nitroamines can be transformed
into valuable compounds such as vicinal diamines and a-amino acids by reduction'7 and
Nef reactions'8 of the nitro moiety (Scheme 8), respectively. Vicinal diamines are of
particular interest owing to their broad utility in the synthesis of biologically active natural
products and drug candidates, and also more recently, as chiral ligands for asymmetric
reactions”. Although a variety of synthetic procedures have been devised to carry out the
Mannich reaction enantioselectively, until quite recently the enantioselective aza-Henry
reaction was unknown. Therefore, as an unsolved problem, considerable effort has been
directed toward the development of catalytic asymmetric aza—Henry reactions over the

past several years.

Scheme 7. The Henry and the Aza-Henry reaction

 

 

o R'ANOZ OH
J N02
R Lewis acid R
or base R'
11 ./\ 11
JNJR R No2 N, R
Lewis acid /I\( N02
R or base R R'

28

Scheme 8. Useful transformations of Iii-nitroamines

NHBoc NaNOz, ACOH, DMSO NHBOC

N102, NaBH4, MeOHA NHBoc
Ph COOH 70% Ph/K/ N02 *

NH2

 

 

95% Ph

This chapter is a review of the available methods that have been developed so far
for the catalystic asymmetric aza-Henry reaction, from the early metal-catalyzed examples

to the organocatalytic protocols that have appeared more recently.

3.2 METAL CATALYZED PROTOCOLS

To the best of our knowledge the first example of a stereoselective (although non-
asymmetric) aza-Henry reaction appeared in 1998 by Anderson’s group”. The method
they reported was an interesting synthesis of 1,2-diamines, through an addition of lithium
nitronates to imines that provided B-nitroamines 29. To avoid retro-addition — a common
complication in the synthesis of nitroamines - these intermediates were then reduced with
SrnIz, and after PMB cleavage the vicinal diamines 31 could be obtained in high yields
and with good diastereoselectivities).

Scheme 9. Anderson’s aza-Henry reaction

1) BuLi,THF,-78°C

 

 

 

”02 2) BnN=CHPh _ PMB‘NH Smlz PMB~NH

Et 3)THF, AcOH Ph/Kr Et THF/MeOH Ph/kr Et

-78 to 0°C N02 60 % NH
28 70% 29 30 2

CAN NH
: - 2
MeCN/HZO Ph/Y Et (anti / syn = 10 / 1)
9‘5 % NH2
31

29

Anderson’s group also reported preliminary studies in which imines and
nitroalkanes were allowed to react in the presence of Bronsted acids and later the reaction
was found to be catalyzed by Lewis acids like BF3 and Sc(OTf)32'. Thus, the stage was set
for developing the catalytic, asymmetric variants of this transformation-goals addressed
and achieved by the research groups led by Shibasaki and Jorgensen.

Shibasaki investigated the aza-Henry reaction conducted with the heterobimetallic
catalyst 32 (Scheme 10)”. Since 32 shows concomitant Bronsted-basic and Lewis-acidic
behavior, both the electrophile and the nucleophile can be activated. Under optimized
conditions 34 could be obtained in 90% yield (synzanti >6: 1) with an enantiomeric excess
of 80%. Among the disadvantages are the limitation to nitromethane, the restriction to N-
phosphinoylimines 33, and the large amount of catalyst required (20 mol%, which

corresponds to 40 mol% chiral ligand).

Scheme 10. Shibasaki’s protocol using the heterobimetallic complex 32

 

 

CIJtBu
DC I“ 00
o; 0‘“
o’ Al<o
CO CO
32
(I? 32 (20%) 9
N , PPh2 DCM, '4000 HN , P112 1)Sm12 NH2
J' Ara : R
NO 2)HCI, MeOH Ph
Ph 2 Ph then NaOH NH2
R N02
33 34 35

One point in favor of this methodology is that vicinal diamines 35 can be obtained easily

from 34 by reduction of the nitro group and cleavage of the phosphinoyl group.

30

The breakthrough in terms of handling and practicability was achieved by

Jorgensen et a1. (Scheme 11)23

. They showed that application of the C2 symmetrical chiral
Lewis acid 36 (20 mol%) to the addition of nitroalkanes to a-iminoester 37 can afford
aza-Henry products like 38 with high diastereo- and enantioselectivities.23b After reduction

of the nitro moiety in 38 very valuable building blocks like or,8-diaminocarboxylic acids

39 can be obtained (Scheme 11).

Scheme 11. Jorgensen’s aza-Henry reaction

 

Me Me
3° 0W0
(J l \2
_ N\ ,N
C C [PMB
,PMB Ph (091-02 Ph HN
N N02 20 0/
J'x + r ( o) T' OzNj/kco Et
H coza Et 1115131202.) E 2
0
ee=94%

O’>_‘,Ph
NH2‘ Me

- . \N O
H2,Raney Nl y /I\/C02Et Me T

 

 

 

I
T Et Cu‘ _O.
HN 0 \N 5”” x
39 plus H l NNrR
Ph P H
_ 40 ..

If silylated nitronates are employed, addition of base is not necessary for reaction
with imine 37.23“ Owing to the high reactivity of the silylated nitronates, the reaction
proceeds uncatalyzed even at —78°C. Reaction temperatures of -100°C were required to
achieve high diastereoselectivities (erythrozthreo >25:1) and high enantioselectivities
(>95%ee for the erythro product) in this case. The high selectivity of the addition in both

cases is attributed to the Zimmerman-Traxler transition state 40, in which both reaction

31

partners are coordinated to the chiral Lewis acid. If one assumes a rapid equilibration of
the E and the Z forms of the Copper nitronate, transition state 40 is formed exclusively,
which explains the preferred formation of the erythro product.

Very recently, one more important metal catalyzed variant of the name reaction
has been reported by Palomo’s group featuring the addition of nitromethane to N-Boc
protected imines 4124. The breakthrough in this case is that good to excellent
enantioselectivities of the desired products 42 can be obtained by exclusive use of

commercially available, inexpensive materials such as Zn(OTf)2 and N-methylephedrine.

Scheme 12. Palomo’s aza-Henry reaction assisted by Zn(OTf)2 and N-methylephedrine

 

, Boc 30% 2010102. , 300
IN 30% IPrgEtN HN
A J + CH3N02 = A )\/No2
' 45% (-)-NME '
_ o
4‘ 2° C 59-98%
87-99%ee

32

3.3 ORGANOCATALYTIC APPROACHES

The rapid development of organic catalysis25 during the last few years has led to
the appearance of several reports concerning the promotion of the aza-Henry reaction
through the activation of the imine and/or the nitroalkanes through metal-free catalytic
systems. The catalyst that have been used so far can be divided into three groups: (i)
Chiral proton catalysts, (ii) Cinchona-alkaloid-based phase-transfer catalysts and (iii)

chiral thiourea catalysts.

3.3.1 Chiral Proton Promoted Aza-Henry Reaction

In 2004 Johnston et a1. developed a chiral bisamidine triflate salt that effects the
diastereoselective addition of nitroethane to a range of electron-deficient N-Boc imines”.
The catalyst 44 is a protonated chiral diamine based on diaminocyclohexane as the chiral
schaffold. By using 44, neither a Bronsted base additive nor preactivation of the

nucleophile is necessary for the addition to occur.

Scheme 13. Johnston’s aza-Henry reaction

10% l+ OTI

 

NI-I HN
NOZSN H 448>ng
,B
J CC
+ /I\‘/NO2
Ar F1 -20°C Ar/Kl/
41 R
51-69% 43
d.r.=7:1 to 19:1
59'90‘7066

33

The exact nature of the stereochemical determining catalyst-substrate complex was
not clear, however, it was clear from the initial experiments that the proton on the
quinoline moiety plays a key role in both substrate activation and orientation leading to
asymmetric induction.

Johnston’s method is unique in demonstrating the use of a chiral proton (a polar
ionic hydrogen bond) alone as both the means of activation (function) and control
(structure) of absolute and relative stereochemistry. It also constitutes the first example of

a metal-free catalytic asymmetric aza-Henry reaction.

3.3.2 Cinchona alkaloid-based phase-transfer catalysts

Palomo’s group found recently that R-amidosulfones are appropriate in situ
precursors of enolizable aldehyde-derived azomethine compounds in the context of the
asymmetric aza—Heny reaction. Thus, a new asymmetric aza-Henry technology was
developed with broad substrate scope based on the use of R-amido sulfone substrates 44
and phase transfer catalysis (P’TC)27. Given that in situ generation of imines from R-amido
sulfones requires stoichiometric base”, the base-promoted, nonselective background aza-
Henry reaction constituted an initial obstacle”. It was postulated that phase transfer
conditions using chiral quaternary ammonium salts in combination with a nonsoluble base
would render the competitive undesired reaction marginal. Indeed, it was found that
catalytic quantities of 45 in combination with CsOHHzO (120-150 mol %) sufficed for the
reaction of nitromethane with a variety of R-amido sulfones 44. The enantiomeric
excesses are generally high for in-situ generated aryl-substituted imines irrespective of the
electronic nature of the aromatic ring: heteroaromatic imines being also tolerated. Most

remarkably, the aza-Henry reaction gave enantiomeric excesses regularly above 94% with

34

an array of enolizable aldehyde-derived imines. Linear as well as branched chain alkyl
amido sulfones 44 (R=alkyl) gave the corresponding adduct 46 in good yields and
enantiomeric excesses in the 94-98% range indicating no significant sensitivity to the size
of the alkyl group in 44. Some representative examples are summarized in the following

table.

Table 9. Palomo’s approach using catalyst 45

 

 

 

 

45
(12 mol%)
NHBoc \
NHBoc
F1 Sng-TOI R' /\ N 02 A R N02
44 CsOH.H20 (130%) 40 9'
toluene, -50°C, 44 h
entry R R’ yield % ee%(d.r.)
l ethyl H 80 96
2 iso-propyl H 8 l 95
3 c-hexyl H 77 98
4 Ph H 79 91
5 2-furyl H 72 84
6 Ph Me 88 94(93z7)

 

Almost at the same time, Ricci’s group reported on the same transformation using
the same catalyst 45 but employing KOH instead of CsOH.H203°. His method concerns
the addition of nitromethane to aliphatic and aromatic azomethine precursors 44 to afford

products 46 (R’=H) with very good yields and ee’s (73—99%).

35

3.3.3 Chiral thiourea catalysts

Electrophile activation by chiral small molecule H-bond donors has emerged as an
important paradigm for enantioselective catalysis, with new applications and
developments appearing at a rapidly increasing pace“. Particularly, the simultaneous
action of two hydrogen bond donors has proven to be a highly successful strategy, both in
enzymes and in synthetic catalyst systems. Several properties of the bifurcated hydrogen
bond may contribute to its utility in catalysis. Such interactions benefit from increased
strength and directionality relative to a single hydrogen bond. In an analogous manner,
two-point binding is an extremely powerful strategy for asymmetric catalysis with metal-
centered Lewis acids”. However, while the requirement for multidentate coordination to a
chiral Lewis acid often imposes limitations upon substrate structure, in principle any
Lewis base is capable of engaging in bifurcated hydrogen bonds. Electrophiles shown thus
so far to be activated by double H-bond donors include aldehydes, ketones, esters, imines,
N-acyliminium ions, and nitro compounds3 ‘.

Concerning the catalysts used in double H-bond activation, the synthesis of chiral
thioureas is facilitated from the ready availability of enantiopure chiral building blocks
bearing primary aminofunctionalities. Therefore given the excellent general stability, high
conformational rigidity33 and Lewis base binding proclivities of thiourea derivatives, it is
unsurprising that chiral analogues are rapidly emerging as versatile, functional group
tolerant and easily prepared and modified catalyst templates for the promotion of a wide

range of synthetically useful asymmetric carbon-carbon bond forming processes”.

36

Along these lines, given the fact that the nitro group and the imine functionality are
both capable of participating in H-bonding, it is not surprising that some protocols
concerning thiourea-catalyzed aza-Henry reactions have already appeared in the literature.

The first protocol for a thiourea catalyzed aza-Henry reaction was published in
2004 by Takemoto”. His design of the catalyst borrowed from Schreiner’s thiourea 473°,
which had been established a few years earlier for the promotion of Diels-Alder reactions.
Takemoto’s catalyst 48 was based on the electron withdrawing bis(trifluoromethyl
substituted) aryl group which ensures the necessary acidity for the thiourea N-H bonds -
and thus the capability in effective H-bond donation —— and incorporates a dimethylamine
group as a base for the activation of the nucleophile. Specifically, the corresponding
nitronate could be produced from the nitroalkane with the bifunctional catalyst via the
hydrogen-bonding activation with the thiourea moiety and subsequent deprotonation by

the neighboring tertiary amino group

Figure 7. Design of Takemoto’s catalyst for the activation of the nitro group in the aza-
Henry reaction.

CF3
CF3 or:3
. (1 IL
FC NJI\N or: i II 0 CF"
3 H H 3 /N\

47 48
CF3 CF3
Q i 0 CL JSL Q
,3 '3 I? _ \N: H 11 3
/..\ E_ E _ /+\H ,,,,, g E—
cho 0: IO

+
H/FIFR H/ILR

37

Using catalyst 48 at 10% catalyst loading Takemoto was able to promote the aza-
Henry reaction between phosphinoyl imines and nitroalkanes with good yields and modest

ee’s. The results for various aromatic and heteroaromatic imines are shown in Table 10.

Table 10. Takemoto’s Aza-Henry reaction of phosphinoyl imines

 

 

 

n 48 (10%) H
INJDPh2 DCM, rt HN,Ph2
Ar) '/ N02 Ar Ph
R N02
33 10 eq, 34
entry Ar R yield % %ee (dr)
1 Ph H 87 67
2 4-Me-C6H4 H 72 63
3 4-Cl-C6H4 H 76 67
4 2-naph H 78 70
5 2-furyl H 85 76
6 2-pyridyl H 91 68
7 2-thienyl H 57 64
8 trans-styryl H 68 65
9 Ph Me 83 67 (73/27)

 

The same group reported a second protocol in 2006 for the name reaction using the
same catalyst, but featuring N-Boc protected imines”. The reactions with N-Boc imines
turned out to be much more efficient in terms of yields and enantioselectivities and a
number of higher nitroalkanes could be successfully used affording syn products with high

diastereoselectivities as well. Representative examples are given in the Table 11.

38

Table 11. Takemoto’s aza-Henry reaction with N-Boc imines

 

 

 

jI’Boc N02 48 (10%) HN'BOC
Ar + 6 ; /I\rNo2
n DCM, 20°C A’ n
41 43
entry Ar R yield% % ee

1 Ph H 90 94
2 4-CF3C6H4 H 80 98
3 4-Me H 82 93
4 4-MeOC6H4 H 7 l 95
5 l-naphthyl H 85 95
6 3-pyridyl H 89 98
7 Ph Me 92 93(90/ 10)
8 Ph CHzPh 84 97(83/ 17)
9 Ph CHZOBn 80 95(86/14)

 

Takemoto’s protocol with N-Boc imines was the first report of a highly diastereo-
and enantioselective aza-Henry reaction with functionalized nitroalkanes to give N-
protected imines. To account for the highly stereoselectivity of the reaction, a ternary
complex C of catalyst 48, imine 41 and nitronate anion was proposed as a plausible
transition state, in which the imine is hydrogen-bonded to the thiourea moiety and the
amonium group of 48 is hydrogen bonded to the nitronate anion (Scheme 14). Ternary
complex C can be considered to be generated through two different ways. On one hand,
thiourea catalyst first activates the nitroalkane by hydrogen-bonding interaction which is
followed by intra- or intermolecular deprotonation by the amino group of 48 to generate

39

nitronate complex B. Subsequent binding of the imine to the thiourea moiety in place of
the nitronate produces complex C. On the other hand, complex C might be formed by the
successive interaction of the imine and the nitroalkane with thiourea catalyst 48 via
binding and deprotonation (D to C). In any event, the thiourea moiety of 48 is believed to
play a crucial role in activation of N-Boc imine in the nucleophilic addition step or
nitroalkane in the deprotonation step. Complex C correctly predicts both the

diastereomeric preference for the syn isomer as well as the absolute stereochemistry

observed.

Scheme 14. Proposed reaction process for the Takemoto’s thiourea-catalyzed aza-Henry

 

reaction
it s
-. Ar
’N ' -. ,A
2' H Iii/[LE r
—> \N E
/ ‘HWHF Z _
C .15: + (“’0
nitroalkane |+
/ H B ”A”
A
thiourea 48 1 imine
S H
\L’fis All. RCH2N02 _ Arm/IL”, M ‘1
C.) III III I 3" Boc
NVAr R ,N:—'...H-i"N\ HN’
Cir T 0‘5 s‘ 5 Me No
O£u H x If I N 2
H 0811 C major product

Another chiral thiourea catalyst successfully employed in the context of the aza-
Henry transformation is J acobsen’s thiourea 493°. Thiourea 49 can promote the addition of
nitroethane to N-Boc imines with good dr ratios, while the yields and enantioselectivities

for the resulting nitroamines are excellent.

40

Figure 8. Jacobsen’s thiourea 49

mg. :0

 

 

 

OPiv
Table 12. Jacobsen’s Aza—Henry reaction using 49
1"“ 123.1132. M“
Ar + I/ T Ar N02
Ft toluene, 4°C F1
41 (2.5 eq.) 43
entry R yield % d.r. % ee
1 Ph 96 15/1 92
2 4-Cl-C6H4 98 7/ l 95
3 3-Cl-C6H4 85 7/ 1 96
4 2-Cl-C6H4 99 2/ l 93
5 4-Me-C,5H4 90 12/ 1 96
6 2-Me-C6H4 99 9/ l 95
7 4-MeO-C6H4 95 16/ 1 96
8 3-pyridyl 79 7/ l 97
9 2-furyl 95 6/1 93
10 2-naphthyl 91 5/ 1 97

 

Finally, Ricci’s group has reported aza-Henry reactions of N-Boc imines with

nitromethane using the Cinchona alkaloid-derived thiourea 5039.

41

Figure 9. Ricci’s thiourea 50
01:3

8 6
H
N

HN—/<

50 HN.,_

/O CO

Catalyst 50 can promote the reaction of nitromethanes — but not of other higher
nitroalkanes — with N-Boc imines affording good yields and modest to excellent

enantioselectivities.

Table 13. Ricci’s aza-Henry reaction using catalyst 50

 

 

 

£00 .800
IN 50 (10%) ”N
J '1‘ M9N02 /I\/N02
Ar Ar
toluene, -24°C
41 (5 eq.) 42
entry Ar yield % % ee
1 l-naphthyl 87 88
2* 2-naphthyl 82 94
3 4-Cl-C6H4 77 94
4* 2-Br-C6H4 82 88
5 4-MeO-C6H4 65 82
6 2—thienyl 50 82
7* 2-furyl 58 63

 

*Reaction at -40°C

42

3.3.4 Bis-Thiourea Catalysts for the aza-Henry reaction and related transformations

In the protocols discussed so far, concerning the mechanistic details of the
transformations, it is clear that the imine and the nitroalkane are both capable of forming
H—bonds with the thiourea catalysts. Jacobsen’s thiourea 49 (Table 12) is not bifunctional
and external base is added for the deprotonation of the nitroalkane, so it is not clear
weather it is the imine or the nitroalkane that is interacting with the catalyst in the
transition state of the reaction. Takemoto’s and Ricci’s thioureas are bifunctional and so
they play a more definite role in the deprotonation step, but again the reaction features two
components (imine and nitroalkane) both of which are potential H-bond acceptors, while
the catalyst structures have only one thiourea moiety and thus it is not clear which
substrate is bound to the catalyst.

Our interest in thiourea catalyzed aza-Henry reactions led us to consider the
possibility of a catalyst that would incorporate two thiourea moieties within its structural
framework, and thus provide the potential for the simultaneous activation of both reaction

components, the imine and the nitroalkane.

Figure 10. The concept of bis-thiourea catalysis
S

JL

1’

o--::z
2.

’1
Z
\
II O+-I

L

 

 

a:

chiral

scaffold - . deprotonation

77

 

 

 

 

 

2D

 

 

/
)=cn

--IZ
2 ----:1:2

 

x40
/<20

43

A catalyst that would have two thiourea moieties should be able to interact with
the imine and the nitroalkane at the same time, activating the two components while
bringing them together in its chiral pocket. This could potentially result in well-defined
orientations and transition states that would promote high enantioselectivities.
Additionally, the concept of the bis-thiourea catalysis is interesting because it could be
applied to many other transformations that involve activation through H-bonding.

The next chapter details on the development of a novel bis— thiourea organocatalyst

for the asymmetric aza-Henry reaction.

44

HAPT ER 4

A NOVEL BIS-THIOUREA ORGANOCATALYST FOR THE

ASYMMETRIC AZA-HENRY REACTION

4.1. IDENTIFICATION OF THE APPROPRIATE BIS-THIOUREA CATALYST

AND IMINE SUBSTRATE

As was mentioned at the end of Chapter 3, the aim of the project presented in this
chapter is the identification of an efficient aza-Henry catalyst that would bear two thiourea
moieties in its structural framework, for the concurrent activation of the imine and the
nitroalkane substrates.

Concerning the chiral schaffold of the catalyst, we decided to focus our
investigation on l,l’-binaphthyl-2,2’-diamine 51 (BINAM, Figure 11), a chiral diamine
that is rarely used in organic catalysis. A search for BINAM in Scifinder gives dozens of
results for the use of the diamine as a ligand for metal catalysis, but its use in metal-free
catalysis and particularly in the field of chiral thioureas, by comparison, has been very rare
4°. Intrigued by the fact that thioureas based on BINAM are largely unexplored, we

designed and synthesized the new chiral thioureas 52 to 54.

Figure 11. The chiral ligand BINAM

45

Figure 12. Structures of bis-thiourea catalysts

CF3
QC 3 8 CO
NJLIN CFa 0C HJINII‘
com cow co
8 s
(FD-52 CF3 (R)-53

 

(FD-54 (Po-55 CF,

The design of catalyst 52 borrows from Schreiner’s3° and Takemoto’s35 thioureas
featuring the electron withdrawing bis-trifluoromethyl phenyl substituent for enhancing
the N—H acidity and thus H-bonding capability of the catalyst. Catalyst 53 was designed
based on the speculation that two binaphthyl moieties around the reactive site may create a
deeper and more efficient chiral pocket. Thiourea 54 has two thioureas and two histidines
that could act as bases, so it was designed as a potentially bifunctional acid/base catalyst.
The methylenes between the thioureas and the histidines in 54 would allow for the
necessary degrees of freedom of rotation so that an ethyl histidine moiety could fold,
approaching a nitroalkane molecule that would be bound to the thiourea and thus
deprotonating it, promoting its reaction with an imine, bound to the second thiourea.
Finally, the bis-thiourea catalyst 55, which is based on cyclohexyl diamine, is a known

catalyst that has been reported for an asymmetric Baylis-Hillman reaction“. Compound 55

46

is the only known bis-thiourea catalyst in the literature, and we decided to evaluate its
efficiency in the aza-Henry reaction in comparison with the BINAM—derived compounds.
The synthesis of compound 52 is straight-forward. Following the known
procedure“ for the synthesis of 55 illustrated in Scheme 1, the new thiourea can be
obtained from the reaction of BINAM with 3,5—bis(trifluoromethane)phenyl

isothiocyanate in an almost quantitative yield after crystallization.

Scheme 15. Synthesis of bis-thiourea catalysts 52, 55

 

2eq. ,CGS
F,C
00 Q 00 i G
H H
“”2 CF, N N CF,
MN”? .N N CF,
THF,O°Ctort 00‘ H\n/H
96% S
(Fi)-51 (R)-52 CF,
2eq. 04$
F,C N” CFa
JSL <1
<1“: a” i
H
NH2 THF, 0°C to rt “TN CF,
(Fi)-56 74% s O

The synthesis of catalysts 53 and 54 involves a simple two-step sequence through
the intermediacy of the BINAM-bis isothiocyanate 57 (Scheme 16). Compound 57 can be

obtained in a nearly quantitative yield from the reaction of (R)-BINAM with thiophosgene

47

under basic conditions.” Without any purification, 57 can be subjected to reaction with

another (R)-BINAM moiety furnishing 53 (55%), or with histamine, affording 54 (90%).

Scheme 16. Synthesis of thioureas 53, 54

(R)- -BlNAM H1“ 2:
3 0O :0

55%
(FD-BINAM

K2003 I H20 OO
90% histamine 00 i N’VN
(FU'57 900/0 N “W
.n N N
00 ”If H “I ,>
N

(Hi-54

The new bis-thioureas were evaluated in the aza-Henry reaction of nitromethane
with N-Boc imine 41. This particular imine was chosen as a model substrate since it has
been established in the literature as an imine that is amendable to catalysis by urea and
thiourea catalysts” (it is a bidentate imine since it features the N—carboxyl moiety that can
potentially participate in double H-bonding). The results of the preliminary screening are

shown in Table 14.

48

Table 14. Evaluation of the bis-thioureas in the aza-Henry reaction of N-Boc imine 41 (a)

 

 

 

N’ 3°C 20mol% catalyst HN .800
I 4;
P11) 10 eq MONOZ Ph *
toluene, rt N02
41 42
entry catalyst Et3N (eq.) conversion % (b) % ee (0)
1 (R)-52 - 0 _
2 - 0.2 20 -
3 (R)-52 0.4 100 (55) (d) 74
4 (R)-53 0.4 100 6
5 (e) (R)-54 - 65 5
6 (t) (R)-54 - 60 4
7 (R)-55 0.2 100 (50) (d) 8

 

(a) Reaction conducted with MeN02 (10 equiv), and several catalysts (20mol%) in
toluene at room temperature overnight (except entry 1, which was run for 2 hours) with 0.25M in
imine. (b) Determined by lH-NMR. (c) Enantiomeric excess was determined by HPLC analysis of
42 using a chiral column. (d) Isolated yield after column chromatography. (e) Reaction conducted
in DMSO. (f) Reaction conducted in MeNOz (120 eq. with respect to the imine).

As can be seen from the data in Table 14, catalyst 52 can’t promote the reaction
without the presence of a base (entry 1). On the other hand, the base alone (without the
presence of a catalyst) gives a very slow reaction (entry 2). On the contrary, when catalyst
52 is used (20 mol%) along with 40% Et3N at room temperature, a quantitative conversion
to the nitroamine 42 along with a very promising enantioselectivity (74%, entry 3) can be
obtained. The reaction was very fast at rt, taking only ~2h for complete reaction of the
imine. The selectivity of the catalyst 52 is superior to that of catalyst 55 (8% ee, entry 7),
indicating that the BINAM schaffold is transfering stereochemical communication to the

reaction center much more efficiently than its diamino cyclohexane analog. The isolated

49

yields for these reactions were found to be in the range of 50-55%, and the inconsistency
with the observed conversions can be attributed to possible decomposition of the sensitive
imine during the course of the reaction, along with the tight separation from the catalyst
and aromatic impurities presumably coming from decomposed catalyst in the SiO2 column
chromatography, in the case of entry 1 (52 decomposes but 55 does not).

Catalysts 53 and 54 turned out to be inefficient in inducing any stereocontrol,
affording very low ee’s. In the case of 53, the low asymmetric induction was in contrast to
our initial anticipation from the assumption that two BINAM moieties would create an
efficient chiral pocket for the reaction. The good conversion observed by lH-NMR for this
reaction, though, indicates that the catalyst interacts with the substrates, but obviously in a
way that lacks stereocontrol.

With a view towards understanding the structure of catalyst 53, we determined the
X-ray crystal structure of this compound (Figure 13). To our surprise the structure
indicated that the preferred conformation of the bis-thiourea is C2 symmetric only with
respect to the relationship between the two binphthyls. The skeleton of the molecule folds
in a way that the symmetry within each binaphthyl moiety “breaks” (which is in
agreement with the NMR data for compound 53, that indicates the presence of 12 aromatic
protons and 20 carbons). The crystal structure also shows that the thiourea moieties adopt
an s-trans, cis conformation instead of the expected s-cis, cis. The resulting structure is
expected to leave a single N-H bond available for coordination to substrates. This N-H
bond seems to be located outside the chiral cavity created by the binaphthyl moieties. The

second N-H bond is between the two binaphthyls that seem to Jr-stack, and is probably

less accessible. If the same conformation is favored in solution, that would explain the

50

inefficiency of catalyst 53 in exerting stereocontrol. On the other hand, the fact that
conversion to the product occurred is interesting because it indicates that even a single

thiourea N—H bond alone is enough for substrate activation and reaction promotion.

Figure 13. X—ray structure of compound 53

 

On the other hand, catalyst 53 was found to be highly polar and insoluble in most

common solvents, so polar solvents such as DMSO and nitromethane had to be used,
leading to almost racemic products (enties 5 and 6).

From the above study, catalyst 52 was chosen as the only bis-thiourea that was
promising for its reaction outcome, and the investigation was then focused on the

optimization of conditions.

4.2. OPTIMIZATION OF REACTION CONDITIONS

The search for the optimum conditions was based on four parameters: solvent, the

effect of the base, temperature and catalyst loadings.

51

Concerning the optimum solvent, it was anticipated that non polar solvents that are

not capable of participating in H-bonding with the substrates should promote the reaction

more efficiently, given that a solvent that is an H-bond acceptor can interact with the N-H

bonds of the thiourea and thereby “block” the reactive site. The results of the solvent

search were in agreement with this notion: As can be seen in Table 15, there is a

significant drop in the enantioselectivity when switching to THF (28% ee, entry 4), which

presumably interacts with the catalyst, which in addition slows down the rate of the

conversion. Benzene and dichloromethane gave satisfactory results, but the selectivity in

toluene was the optimum.

Table 15. Solvent screening for the aza-Henry reaction with catalyst 52 (a)

 

 

 

N,Boc 20mol°/o (FD-52‘ HN.Boc
I 7
Ph/l 10 sq MeN02 ph .
0.4 sq Et N
SOLVENT? rt ”02
41 42
entry solvent conversion% (b) %ee (c)
l toluene 100 (55") 74
2 benzene 100 62
3 DCM 82 59
4 THF 52 28

 

(a) Reaction conducted with MeN02 (10 equiv), catalyst 52 (20%), EN (0.4equiv) and
0.25M in imine using several solvents. Reactions run overnight. (b) Determined by 'H-NMR. (c)
Enantiomeric excess was determined by HPLC analysis of 42 using a chiral column. (c) Isolated
yield after column chromatography.

52

With toluene as the optimum solvent, the effect of the base was examined in order
to determine how the amount of base affects the yield and the selectivity. As can be seen
in Table 16, the amount of the base surprisingly has a determining effect on the
selectivity: the optimum induction (74% ee) is obtained when the level of Et3N is 40-50%.
Adding a whole equivalent of Et3N causes a significant drop in the cc, and the same
happens when only 20mol% Et3N is used. Stronger or weaker bases (entries 5, 6) were

found to give almost racemic products.

Table 16. Effect of the base on the reaction. (a)

 

 

 

N.Boc 20mol% (Po-52 HNIBOC
Ph/II 10 egAthNOZ r ph .
Toluene, rt N02
41 42
entry base (equiv) pKa yield % (b) % ee (0)
l Et3N (0.2) 9.0(DMSO) 30 60
2 Et3N (O .4) >> 50 74
3 Et3N (0.5) >> 50 74
4 EN ( l .0) >> 55 67
5 DBU (0.4) 12(H20) 100 (d) 2
6 N-Me-morpholine (0.4) 7.38 35 4

 

(a) Reaction conducted with MeNO, (10 equiv), catalyst 52 (20mol%) and 0.25M in imine
in toluene at room temperature overnight. (b) Isolated yield after column chromatography. (c)

Enantiomeric excess was determined by HPLC analysis of 42 using a chiral column. (d) Refers to
conversion by lH-NMR.

One could rationalize these results if a few assumptions were made about the

mechanism of the reaction. As shown in Scheme 17, it is reasonable to expect that

53

nitroalkane moieties bound to the thiourea will be much more acidic than the ones that do
not interact. One would expect that the asymmetric induction in 42 will depend on the
concentration of species II, if this is the reactive -and selective- species concerning the
catalyzed pathway. The concentration of II will have an upper limit that will be
determined not only by the amount of catalyst, but by the amount of the base as well. In
other words, species I will have a concentration that will depend on the catalyst itself. For
instance, 20% catalyst in a reaction of 0.25M substrate would afford a maximum
concentration of 0.050M for I assuming saturation in the binding of the substrates to the
catalyst. However, in order for 11 to reach the same level of concentration, the right

amount of base, or a base with the proper pr, will be needed.

Scheme 17. A tentative mechanistic hypothesis that could explain the effect of the base on
the reaction

 

 

 

 

 

 

 

 

S chiral S chiral
JL /scaffolg\ JL 3 scaffold S
RN N N NR JL / \ JL
1 1 1 1 RN N N NR
*1 t' +1 H —~ A .1 .1 l1
5gb :
— O N\ R -O\+,O : :
1 v N -
CH, .9: T; B——H CH2 OYNVR
(pKa=~10) + R'
I II
~3- H C NO H C NO
3 T 2 _, 2 = 2
\J B

111
The pKa of bound MeNO2 is expected to be around 10, similar to the pKa of the
compound in water, where it is H-bonded by water molecules in a similar fashion. The

pKa of unbound MeNO2 is expected to be about 17, as it is in DMSO (where it isn’t H-

54

bonded). Most of the nitroalkane in not bound, since there is only 20 mol% of the thiourea
catalyst and a large excess of nitromethane (10 eq). Based on these considerations, a base
with a pKa=~10 would be the right choice for promoting the conversion from I to II,
avoiding the formation of species 111. This is in accordance with Et3N being the optimum
base among the ones that were tried. It seems that DBU, which is a much stronger base
compared to Et3N, is able to promote the formation of III - and thus, the background
reaction - to a degree enough for causing a significant drop-off in the selectivity. On the
other hand, for the results in entries 1 to 4, it seems that 0.4-0.5 eq. triethylamine is the
necessary amount for maximizing the concentration of complex II, and that a whole
equivalent of the base will start promoting the background reaction, decreasing the
stereoinduction in 42. Concerning N-methylmorpholine, it seems that its lower basicity
makes it inefficient in deprotonating nitromethane to the desired degree, and in this case
the concentration of the “active catalyst” (or of II) is not very high (although one could
argue that this particular base might H-bond to the catalyst through its oxygen atom and
decrease the ee via catalyst deactivation).

Independent of the validity of the above assumptions and mechanism insights, one
thing that is very clear from the above set of data is that triethylamine in the amounts of
40-50% is the optimum for the reaction, and thus we decided to proceed with further
efforts to optimize the reaction by using 0.4 eq. triethylamine.

Next, the reaction was investigated at lower temperatures in an effort to enhance

the asymmetric induction of 42. The results are illustrated in Table 17.

55

Table 17. Optimization of temperature and catalyst loadings (a)

 

 

 

N. 3°C Xmol% (FD-52 HN’BOC
PhJ 10 eq MeNOz; ph .
0.:oelxc/IerE1I3N N02
41 42
entry cat.loading% solvent T°C conversion% (b) % ee (c)
l 20 toluene rt 100 (55) (d) 74
2 20 toluene -5 100 82
3 20 toluene -35 83 (55) (d) 86
4 20 toluene -55 50 84
5 20 mesitylene -35 86 84
6 20 Cl-benzene -35 93 84
7 10 toluene ~10 100 (50) (d) 80
8 10 mesitylene -10 90 82
9 10 toluene -35 53 78
10 5 toluene -10 66 (30) (d) 71

 

(a) Reactions conducted with MeNO2 (10 equiv), catalyst (R)-52, Et3N (0.4equiv) and
0.25M imine in several solvents. Reaction time was ~36h except for entry 1 (2h). (b) Determined
by 'H—NMR. (c) Enantiomeric excess was determined by HPLC analysis of 42 using a chiral
column. ((1) Isolated yield after column chromatography.

The temperature was gradually reduced from rt to -55°C with maintaining the
catalyst loading at 20mol%. As can be seen in entries 2 and 3 in Table 17, lowering the

tempeature to —5°C and -35°C resulted in a significant enhancement of the

enantioselectivity, to 82% and 86% respectively, without significantly affecting the

56

conversion of the imine. The reactivity of the catalyst was still satisfactory at —35°C,
affording the same isolated yield for 42 as for rt. The rate of the reaction at -35°C was
much slower than at rt (2h), but this could be overcome by using longer reaction times
(36h). A further decrease in the temperature caused a drop-off in the reactivity, so —35°C
was chosen as the optimum temperature. Other aromatic solvents such as mesitylene or
chlorobenzene were almost as efficient at -35°C as toluene (entries 5 and 6). Reducing the
catalyst loading to 10mol% still provided the chiral nitroamine with a high degree of
enantioselection (80 and 82% ee in toluene and mesitylene respectively at -10°C).

From the above data, the conditions in entry 3 (20 mol% catalyst and -35°C) were
chosen as the optimum balance between yield and asymmetric induction. It should be
noted here that the background reaction was tested at these conditions (using no catalyst

and 40% triethylamine) and was found to give a 9% yield of racemic 42.

4.3. SYNTHESIS AND EVALUATION OF DERIVATIVES OF CATALYST 52

Since catalyst 52 is readily obtained from the reaction of BINAM with 3,5-bis-
(trifluoromethane) phenyl isothiocyanate, we reasoned that a derivatization of the catalyst
with different functionalities on the aromatic substituents of the thiourea would be very
interesting, because it would allow for a practical and convenient, based on the large
amount of the commercially available isothiocyanates, comparison with the parent
catalyst, and it might lead to the identification of other promising catalysts as well.

We synthesized a series of differently substituted bis-thiourea derivatives with
variation of the electronic nature as well as on the steric bulkiness of the structures. The

synthesis of the derivatives is shown in Table 18.

57

Table 18. Synthesis of thioureas 56a-61a (a)

00 00
”.0112 THF I I ,

”in Ar +00 “infir
811/8
3

(NH

 

 

51 56-618 56-61b
entry Ar products T°C %yield (b) a %yield (b) b
1 Ph 56 rt 50 50
2 p-NO,-C,H4 57 rt 100 0
3 3,5-Clz-C6H, 58 rt 100 0
4 p-MeO-C,H4 59 rt 1 7 50
5 2,6—Me2-C,H4 60 rt 0 0
6 2,6-Me2-C,H4 60 70 0 50
7 l-naphthyl 61 rt 5 55
8 l-naphthyl 61 70 25 30

 

(a) Reactions run overnight except for entry 6, run for 2 days. (b) Isolated yield after column
chromatography.

The reactions for the synthesis of compounds 57a and 58a were fast and
quantitative, presumably because of the highly reactive electron-withdrawing
isothiocyanates employed. Catalyst 58a was obtained as a pure solid from the crude
reaction product that needed no further purification. The reaction of the rest of the
isothiocyanates turned out to be more complicated though, because the less reactive they
were, the greater the amount of the undesired monosubstituted byproducts 56-61b that was

obtained. As a result, for compound 61 the ratio of di- to monosubstituted products was

58

the least satisfactory, affording only a trace of 61a. This could be partially overcome by
increasing the temperature to 70°C which afforded a sufficient amount of the bis-thiourea
for testing (25%, entry 8). The most demanding reaction was the one in entry 5,
presumably because of the steric bulk at the 2,6-positions of the electrophile: compound
60a couldn’t be obtained even at elevated temperatures.

The new bis-thiourea derivatives were all tested in the aza-Henry reaction of N-

Boc imine 41 with nitromethane at rt. The results are illustrated in Scheme 18.

59

Scheme 18. Results for the screening of different thiourea derivatives in the aza-Henry
reaction at rt

 

Boc o
’ t l t . Bo
JN 20 /0 ca a ys H N c
P“ 10 eq MeN02 Fh/Ifi 42
41 0.4 eq E13N N02

Toluene, rt

“/1ng 0O “inflow
Ow N
III/”Q CO H\II/H\I©LOM9

100%(NMR) 36°/ NMR
568 5% ea 59' 0°/ /oe°e( )

N02
00 Hill/C la E G Hill/3
MN N
CO"‘”1”©°‘ CO “PO...

 

2

58a 73%(NMF1) CI 573 100%(NMR)
<5%ee 16%ee
5
0(1ng
Ohm/Hg
s
518
100%(NM11)D
4 %ee

As can be seen from Scheme 18, the only thiourea that didn’t afford a good

conversion was 593. This is reasonable, since the NH bonds of that thiourea are expected

to be the least acidic, and therefore the least efficient for H-bond interaction with the
substrates.

All the rest of the thioureas afforded good conversions, but the asymmetric
inductions of the product were disappointingly poor. One can expect that the H-bonding
ability of thioureas 57a and 58a should be similar to that of 52, but the inductions clearly
demonstrate that even if there is H-bonding interaction, the communication of
stereochemical information is ineffective. This difference might be associated with the
bulkiness of the 3,5—bis trifluoromethyl groups and / or the arrangement in space of 52
compared to the rest of the thioureas. A better understanding of this issue will require
further work — e.g. theoretical calculations or X—ray analysis of the catalysts.

In any case, the necessity of the 3,5—bis trifluoromethyl groups is not surprising on
the basis of the literature associated with thiourea catalysis. Indeed, this particular
substituent seems to dominate the structures of most of the thiourea catalysts that have
been used. For example, Schreiner has studied the acceleration effects of various aromatic

and aliphatic thioureas for a variety of non-asymmetric reactions and he has demonstrated
Figure 14. Schreiner’s thiourea 47

rigidifying S-H interaction
CF, fl CF,

F3C N N CF

H H

3

47

that catalyst 47 has been the most effective structure tested”. Schreiner has rationalized

his results in terms of the importance of entropic effects; specifically it was proposed that

61

the (computationally determined) rotational barrier of catalyst 47 is relatively high due to
an attractive interaction between the ortho-hydrogen atoms, which are polarized by the
adjacent electron withdrawing substituent, and the Lewis basic sulfur heteroatom (Figure
14). This rigidifying interaction would minimize entropy loss upon binding of the
substrate and thus facilitate catalysis“.

In the context of the catalysts 56-61a, the argument concerning the entropic effect
of the S-H interaction discussed above could still apply (Scheme 19). In a comparison of

catalyst 52 with 58a, for instance,

Scheme 19. A possible mechanism to account for the high induction obtained with 52
R

N)

CI
00 Nil,” “““ 02m — DO j‘I\N’I:j\01

—* ~13 .1.
I
0’ ”Q 90
O\ NVR
C c. c. r
1 11
n
z) CF.
N H
)c S'
S ...o R CO JL 3:1
N’ILN’H“ <— / II II CF"
, —» ‘5. H H
”‘1'
CO OYNVR
F,C CF, R
111 w

it could be argued that since the chloro- substituent is not as electronegative as the CF3

group, this attractive S-H interaction shouldbe diminished, and the thiourea moiety would

62

be more flexible in rotating and adopting different conformations, presumably less
selective ones. On the other hand, for catalyst 52 the S-H interaction should make
complex IV more abundant than III, resulting in bidentate binding and thus more well-

defined orientations.

63

4.4. SUBSTRATE SCOPE

We next turned our attention to examining the scope of the asymmetric aza-Henry
reaction for a series of differently substituted N-Boc imines and nitroalkanes.

For this purpose several imines were synthesized according to the standard
procedure”. Table 19 shows the two-step sequence for obtaining a variety of electron

withdrawing, donating or neutral aryl-substituted substrates 41.

Table 19. Synthesis of aromatic N-Boc imines 41

HZNBOC Boc

 

 

 

 

l PhSOgNa; 13: K2C03, Na2$O4 : IN,Boc

Ar Y,%%%n:fo Ar SOzPh THF reflux Ar

62 41
entry Ar series %yield 62 %yield 41

1 Ph a 100 98
2 p-Cl-C,H4 b 23 81
3 m-Cl-C,H4 c 77 99
4 o-Cl-C6H4 d 72 98
5 p-Br-C,H4 e 31 64
6 p-MeO-C,H4 f 84 77
7 o-MeO-C,H4 g 90 95
8 3-pyridyl h 68 98
9 p-Me-C,H4 i 74 97
10 l-naphthyl j 59 99

 

Most of the N-Boc imines were oils, and so they were used as crude products without any
purification. It should be noted here that N-Boc imines are very sensitive to moisture and
their synthesis and workup requires cautious handling. Additionally, it was found that the
degree of their purity greatly influences the outcome of the aza—Henry reaction with
catalyst 52, especially regarding the enantioselectivities obtained. For example, when the
reaction was performed with impure (partially hydrolyzed) imine 41a that contained some

benzaldehyde, product 42 was obtained in only 11% ee.

65

The results for the catalytic asymmetric aza-Henry reaction of the imines 41a-k

with nitromethane, nitroethane and nitropropane are summarized in Table 20.

Table 20. Reaction scope of the aza-Henry reaction of imines 41a-k using catalyst 52 (a)

 

 

 

N , Boc ”1:511:11“ .11 ,3...
Ar/II 10 sq RCHzNo: Ph ‘ N02
toluene, -35°C R
41 42, 438-k
entry Ar R time (h) adduct %yield (b) %ee (c) (d.r.) (d)
1 Ph H 36 42 55 86
2 p-Cl-C,H4 H 15 43a 62 85
3 m-Cl-C,H4 H 15 43b 55 91
4 o-Cl-C,H4 H 15 43c 61 74
5 p-Br-C,H4 H 24 43d 50 78
6 p-MeOC,H4 H 36 43c 50 89
7 o-MeO-C,H4 H 36 43f 40 65
8 3-pyridyl H 22 43g 63 8 1
9 p-Me-C,H4 H 36 43h 48 86
10 l-naphthyl H 36 431 65 85
11 Ph Me 36 43j 59 70 (e) (77/23)
12 Ph Et 36 43k 63 80 (e) (80/20)

 

(a) Reaction conducted with RCH2N02 (10 equiv), catalyst la (20mol%), Et3N (0.4equiv)

and 0.25M imine in toluene at -—35°C. (b) Isolated yield. (0) Enantiomeric excess was determined
by HPLC analysis of 7a—l using a chiral column. (d) Diastereomeric ratio was determined by
HPLC. (e) Major (syn) diastereomer.

66

As can be seen from the data in Table 20, N-Boc imines with either electron
donating or withdrawing groups can afford product 43 with moderate to good isolated
yields and overall high selectivities. N-Boc imines with chloro substituents seem to be
quite reactive, leading to quantitative conversions in short reaction times (15h) and to
good isolated yields of the product along with high enantioselectivities, up to 91% ee
(entries 2 to 4).

Electron donating groups slow down the reaction, but prolonged reaction times can
afford satisfactory conversions and yields, along with good asymmetric inductions. The
reason for the drop in the selectivity that is observed for the o-MeO substituted imine is
not very clear at this point. On the other hand, electron neutral substituents don’t have any
effect on the selectivity of the reaction affording ee’s always in the middle 80’s (entries
9,10).

Finally, for the reaction of imine 41 with nitroethane a drop in the
enantioselectivity is observed (70%ee, entry 11), while for the reaction with nitropropane,
the enantio- and diastereoselectivity are still in a synthetically usefull range (entry 12),
demonstrating that the asymmetric aza-henry reaction catalyzed by the bis-thiourea 52 is
not limited only to nitromethane, but can be applied also to higher nitroalkanes.

A disadvantage of the above methodology is that the catalyst can not be recovered,
since it partially decomposes during the course of the reaction and during column
chromatography. The decomposition material usually has an Rfvalue similar to that of the
product of the reaction, making the purification of some of the products tedious.
Nevertheless, satisfactory yields in the range of 40—65% can be obtained for the resulting

nitroamines.

67

4.5. AN INSIGHT INTO THE MECHANISM

A number of possible reaction intermediates can be envisioned for the reaction
with catalyst 52. It can be postulated that the two thiourea moieties can interact with two
nitro groups (A), or with two imines (B), or with a nitroalkane and an imine at the same

time (C), giving rise to three different species that interconvert with each other. These

possible intermediates are shown in Scheme 20.

Scheme 20. Possible reaction intermediates

 

 

 

 

 

 

 

 

. hiral
chiral S c s
JSL schaffold i JL /50hafi0“\ JL
RN N/ \N NH F"? 'i‘ '1' '1‘”
'1 t' 11 1.1 _ 2' '1 '9 1:
19,311 15,, $5 OYNVR CYNVR
Et N—H " +Ii‘ .
3 +3 CH2 CH3 R R'
Z + N62CH2
Ng/Fl A Et3NH

Oat

B
R' \
chiral
S schaffold S /

JL/ \JL

 

 

 

 

'1'.“ 1 1 11*
1.1 t' 1.1 H
_‘ + ‘
Et N—H‘P‘I'I’O- RVNYO
C

If the reaction occurred through intermediate A, the imine would have to be
activated through the protonated triethylamine (if the imine interacting with the complex
at all) in a “chiral nucleophile” mechanism.

Complex B seems less possible, because the nitro group has been documented in
the literature to be a very good thiourea coordinator”. It’s also reasonable to suggest that

MeNO2 is interacting with the thiourea, since Et,N is not basic enough to completely

68

deprotonate free MeNO2 (pka=17 in DMSO vs 10 in H20). On the basis of that,
mechanism B can be ruled out.

Finally, complex C describes our initial idea when designing the bis-thiourea
catalysts, that is the dual activation of both reaction components.

To distinguish between mechanism A and C, we performed 'H-NMR experiments
in order to probe the interaction between the imine and the catalyst. Specifically, we
prepared NMR samples of the imine alone and of imine / catalyst mixtures at different

ratios. The results of this experiment are summarized in Figure 15.

Figure 15. lH-NMR Study as a probe for imine-catalyst interaction

A .
8.88pm: 3 i
1 1'1 . 1
i 1, 1
l 'F l‘: I
1AA I .
I. 1’
I I n‘
F ,1. I A."
. \_ 1.-.... . .‘v , I’ I. \ .‘ “~"-\4 ..-.- —. our-w-“
7'7 . l r , I ~ - . TY ”Tr-”TI”. I I I I I I
9.2 9.0 8.8 8.5 8.4 8.1 8.0 7.8 7.5 74 7.2 7.0 DUI
1:
B 11 A
8.74pp1n ,’ 11‘ I'- .’ »
I 1' {I 2 I. I.) I 'II
A 312. Si .I‘A I , l:
1 TT‘T‘W -1 T 'TT"" 7 "T,
( I ‘T’. T'
92 90 88 86 8.8 82 00 78 76 74 7.2 709“
.3
i
l
I
f 1
C .
1 ' ‘
17.? 1
A r I ( - I .
i , “ ,1 A I,‘
A. ‘I, .A\ . I, g. 1 I}! :1
8.65m)!!! ' , g- . "I. ix
1 , .. '; A , ~ \... x- i 11>
- ~41 , A, _ ,,., _‘ . 7,. _. - ,..-.A_. ,, .-. , _ ..
9.2 9.0 I. O i '
A 8 0 6 0.8 8.2 8.0 7.8 7 6 7 4 7.2 7 0 pp

A: NMR of imine; B: leq. imine with 1 eq. catalyst 52; C: leq. imine with 2 eq. catalyst 52
69

As can be seen from the data in Figure 14, there is a significant change in the
NMR of the imine while adding first one equivalent (B) and then two equivalents (C) of
catalyst. The C(=N)H proton of the imine shifts from 8.88 ppm in the free imine to 8.65
ppm in the imine / 2 eq. catalyst mixture, strongly indicating that there is an interaction
between the two species. On the basis of this experiment, mechanism A seems to be less
possible than C. Subsequently, species C seems to be the actual intermediate of the
reaction, or at least expected to contribute more than A or B to the reaction outcome.

Additionally, the importance of both the thiourea moieties in the reaction
mechanism was probed. Catalyst 66, that bears only one thiourea in its structure, was
synthesized and evaluated in the aza-Henry reaction. The catalyst was accessed in 4 steps
from BIN AM 51, according to the procedure reported by Wang and coworkers40b (Scheme

21).

Scheme 21. Synthesis of thiourea 66

O
CO A_:2.__..O' AcOH formaldehyde; CO “k
NH2
O"\NH
..xNHz DCM 2 NaBH,CN "‘N\\
720/0 ACOH
THF
51

85% 64

 

4MHC|
30H

CF, 54%

01115‘ 32:9... co
:3 ‘32:

66 65

70

Catalyst 66 was evaluated in the reaction of imine 41 with nitromethane in toluene and

DCM and the data is shown in Table 21.

Table 21. Ala-Henry reaction of imine 41 using catalyst 66'

,Boc

 

 

 

N 20mo|°/o 66 , 300
| HN
Ph 10 eq MeN02 ph .
" N02
41 42
entry solvent Et3N eq. % yield ee%
1 toluene 0 O -
2 DCM O 0 -
3 DCM 0.4 50 4

 

‘Reactions run overnight, with 0.25M imine in DCM or toluene.

The reaction using catalyst 66 alone without an external base gave no product,
presumably because the dimethylamine functionality of 66 is not basic enough to
deprotonate nitromethane. When 0.4 eq. of triethylamine was added to the reaction
mixture a 50% yield of the product was obtained, but with almost no asymmetric
induction. This experiment shows that the presence of both thioureas is essential for the
reaction outcome in terms of stereoinduction. One would expect that the intermediate for
the reaction with 66 would feature nitromethane H—bonded to the thiourea, activated for
deprotonation, but with the imine probably not interacting directly with the thiourea
moiety (Figure 16). That would leave the imine without a well defined orientation,
resulting in the very low induction that was observed. These observations with catalyst 66

are in accordance with mechanism C in Scheme 20 for the reactions with catalyst 52.

71

Figure 16. Postulated mechanism for the reaction using 66 with triethylamine

 

I
R) imine orientation
not well defined

The absolute configuration of the products 42, 43 was determined by comparison
of the HPLC data and the optical rotations of the products with the data reported in the
literature. To account for the enantioselectivity and the diastereoselectivity of the
transformation we propose the model shown in Scheme 22. The imine can interact with
the thiourea adopting two different orienrations, one where the aryl group points to the
catalyst (complexes III and IV) and one where it points away from it (I and II). It is
proposed that the orientation of the imine with the aryl group away from the catalyst
should be more energetically favored, because this way the strain associated with this
interaction is minimized. This orientation exposes the si face of the imine towards
nucleophilic attack from the activated nitronate, and this is in accordance with the absolute
configuration of the stereogenic center bearing the N-Boc group.

Concerning the approach of the nitroalkane, it is clear that there are two different
orientations as well. Again, the determining factor is probably the minimization of the
steric strain of the alkyl chain of the RNO2 resulting from close contacts with the catalyst

structure. This hypothesis is in accordance with the data in Table 20, entries 11 and 12,

72

where a slightly higher selectivity is afforded when R=Et (product 43j) compared with
R=Me (product 43k). The triethylammonium ion which results from the nitroalkane

deprotonation might interact with the nitronate anion with hydrogen bonding, or with both

the nitronate ion and the imine, as in Takemoto’s model .35

Scheme 22. A plausible stereochemical model for catalyst 52

HN,Boc Boc
NO
a OBTAINED Ar . 2

A ENANTIOMER Ag

F3C

FA VORED
(LESS STRAIN)

   

73

Although the thiourea moieties in catalyst 52 presumably have some freedom of rotation
and other structures and models could be also drawn, the model in Scheme 22 is a working

hypothesis which can explain the observed selectivities.

4.5. CONCLUSION

In conclusion, a novel BINAM-based catalyst for the asymmetric aza-Henry
reaction has been developed. The bis-thiourea catalyst probably functions according to a
mechanism where both the components of the reaction are activated simultaneously by
hydrogen bonding, indicating a mode of action different from that of the existing protocols
and demonstrating the potential of bis-thioureas as an interesting class of unexplored
catalysts. The catalyst is very easy to make and modify, and affords the aza-Henry
products with modest to good yields, overall high enantioselectivities and synthetically

useful diastereoselectivities.

74

HAPT ER

THE FIRST HIGHLY ENANTIOSELECTIVE ORGANOCATALYTIC

DIRECT ADDITION OF NITROALKANES TO NITROOLEFINES

5.1. INTRODUCTION

The use of nitroalkanes as reactive nucleophiles42 and the use of nitroalkenes as
Michael acceptors43 have attracted significant interest in recent years in carbon-carbon
bond formation reactions. Because of the activating effect of the nitro group, as well as its
versatility as a masked functionality, nitro compounds have been quite useful in the
synthetic arena“?

The Michael addition of nitroalkanes to nitroolefins is particularly interesting
because the reaction products, 1,3-dinitro compounds (69), can be easily reduced by H2 to
1.3-diamines 7145, a valuable functionality in the synthesis of complex molecules - given
the number of natural products or pharmaceuticals that contain nitrogen — or other 1,3-
difunctionalized compounds, heterocycles“, and carbohydrate derivatives47 (Scheme 23).
Despite its synthetic value, this reaction is considerably overlooked and underdeveloped.
This is partly because of the difficulty of controlling the addition of a nitronate to a
nitroalkene. The initial product formed is the nitronate anion 68, which is sufficiently
reactive to also undergo conjugate addition to the nitroolefine 67, thus resulting in a
mixture of oligomerization products48 including 70 and its higher congeners (Scheme 23),
and in significantly reduced yields of the desired 1,3-dinitro products 69.

Perhaps because of these difficulties, very few protocols for the above reaction

have been published and are limited to one unasymmetric and two asymmetric methods

75

only. Thus, the addition of nitroalkanes to nitroolefins is one of the least developed and

most intriguing types of Michael addition processes.

Scheme 23. The conjugate addition of nitroalkanes to nitroolefines as a useful route for
the construction of valuable functionalities

R'V N02 R N02

L H+ R N02 H2 R NH?
—* . .. - L ----- - L
+ FAA/N02 R e 2 R, N02 WC n' NH2
67 53 69 71
‘R&NO2 E X
i x
i
R N02
R O R NH2
R' N02
. OH , OH
N02 R R
70

5.2 PROTOCOLS FOR THE CONJUGATE ADDITION OF NITROALKANES TO

NITROOLEFINS

Protocols for the non-asymmetric conjugate addition of nitroalkanes to nitroolefins
are extremely rare, although some early reports on the Michael addition of nitroalkanes —
or their alkali metal salts — to nitroolefines (either isolated or generated in situ) had been
described“9 in the late 40’s to early 50’s. The conditions for these reactions were harsh and
the yields were variable and often low. According to the best of our knowledge, the only
representative and relatively efficient method reported so far is reported by Alcantara’s
group in 1996”. According to that paper, reactions with the very reactive B-nitro esters as
nucleophiles are fast in hexane or benzene, but less reactive aliphatic nitroalkanes are
more sluggish, giving large amounts of polymers. Good yields of the desired 1,3-

dinitroalkane products from these nitroalkanes can be obtained in reasonable reaction

76

times only if a large excess of nitroalkane (~110 eq) is used in a neat reaction (Table 22).
The corresponding reactions in hexane or benzene are very slow, and in most cases the

reactions are not diastereoselective, leading to mixtures of isomers.

Table 22. Alcantara’s method for the synthesis of 1,3-dinitroalkanes

 

 

 

R3 N02
R1/\I/ N02 + R3 V N02 Eth L/ N02
92 ~1 1O eq. rt R1 R2
entry Rl R2 R3 time(h) solvent yield%

1 Gal H H 20 none 72
2 Gal H Me 24 none 68
3 Ph H H l .5 none 5 l
4 Ph H Me 7 none 95
5 Ph H COOMe 2 hexane 66
6 Ph Me H 26 none 98
7 4-MeO-C6H4 Me Me 24 none 80
8 furyl Me Me 120 none 72
9 isopropyl H COOMe 3 hexane 79

 

The breakthrough came in 2006 when Du’s group reported a highly
enantioselective version of the reaction using Cz-symmetric tridentate bis(oxazoline) and
bis(thiazoline) zinc complexes“. This group had used these catalysts before in the context
of Lewis acid catalyzed Henry reactions of nitromethane with a-keto esters“. The ability
of these zinc species to activate nitroalkanes was applied successfully in the conjugate

addition of the latter species to a series of substituted nitrostyrenes, affording the first

77

enantioselective protocol for the reaction with up to 95% ee for the resulting 1,3-dinitro

compounds (Table 23).

Table 23. The first asymmetric addition of nitroalkanes to nitroolefines

8?. 9

 

 

 

 

N
H
O N NL/JO S N N\_/_/S
\_j\Ph Pfi' Lkph Ph“
72 73
10 mol% 72 R
“~02 + RZYNOZ 25 mol% EtQZn ; R2 A. 3N02
R1 R 80 mol% Ti(OiPr)4 LNG
3 toluene/hexane. R1 2
4 eq. 3 days, rt
entry R1 R2 R3 yield% d.r. %ee

1* Ph Me H 87 1 1.7/1 95
2 4-MeC6H4 Me H 83 6.1/1 91
3 4-MeOC6H4 Me H 83 6.5/ l 89
4 2-MeOC6H4 Me H 90 5.8/1 91
5 3 ,4-(MeO)2C6H4 Me H 67 9 .6/ l 86
6* 4-FC6H4 Me H 79 9.3/ 1 92
7 2-FC6H4 Me H 87 6.2 90
8 2-C1C6H4 Me H 83 3 .8/ l 91
9 2-naphthyl Me H 83 3 .8/ l 88
10 2-furyl Me H 67 6.7/1 88
l 1 PhCHzCH2 Me H 80 3 .4/ l 72
12 Ph Et H 88 4.1/1 88
13 Ph Me Me O - -
*Ligand 73 was used.

78

As can be seen from the data in Table 23, Du’s method is highly efficient for the
synthesis of enantioenriched syn-1,3-dinitro products. The reaction is broad in scope since
either electron rich or electron deficient nitrostyrenes can be used, and additionally,
heteroaromatics (entry 10) and aliphatics (entry 11) can be also tolerated. Branched

nitroalkanes though turned out to be sluggish nucleophiles (entry 13).

Figure 17. Proposed mechanism for Du’s asymmetric addition

N

Ph 3 ,. S

0' \ 1" \\/Et/ 0
A/Nizn“.zn‘~6N\2

\

' O , s
0‘ ’0 \“ Ph
N k
a ‘2
R/
R1

The conjugate addition of nitroalkanes to nitroolefines is proposed to work through
the model depicted in Figure 17 (although the authors ignored the role of Ti), with the Zn
activating nitroethane for deprotonation and at the same time forming a complex with the
nitroolefin, which is oriented in the chiral pocket so that its Re face is exposed for
nucleophilic attack“.

The second protocol for the direct conjugate addition of nitroalkanes to
nitroolefines was reported by Wang’s group in the same year”. It described the addition of
branched nitroalkanes — mainly 2-nitropropane — to nitrostyrenes under neat conditions
using the modified C inchona alkaloid 74 to obtain 1,3-dinitro products in good yields and
modest to good enantioselectivities (67—85%ee, Table 24). This particular method has a
major drawback: the catalyst is not reactive enough to provide reasonable reaction times

even without the use of a solvent and under a large excess of the nitroalkane. Several days

79

are required for the reactions to go to completion (presumably the branched nitroalkane is

too hindered to be a reactive nucleophile).

Table 24. Wang’s protocol using catalyst 74

/\/ N02 + \r N02 10 mol% 74 £02
Ar ——’
neat, 0°C Ar N02

 

 

 

56 eq.
Ar t(days) %yield ee% Ar t(days) %yield ee%
Ph 6 79 78 2,6-C12C6H3 7 75 80
4-FC6H4 7 76 73 4-MeOC6H4 12 78 67
4-C1C6H4 8 70 83 2-MeOC6H4 8 73 84
2-CF3C6H4 9 78 83 2-BnOC.,H4 7 80 77
2-C1C6H4 10 73 8 1 2-thiophene 10 78 7O

 

 

Finally, Maruoka’s group has published the enantioselective addition of preformed
silyl nitronates to nitrostyrenes, a surrogate to the reaction of interest that requires the
separate preparation of the highly reactive silylated nitroalkanes for use as nucleophiles“.
Using catalyst 75, syn products can be obtained in excellent yields and good to excellent
enantioselectivities. As may be expected for the reactions of these highly reactive species,

very low temperatures (-78°C) have to be used in order to achive the maximum inductions.

80

Figure 18. Maruoka’s catalyst for the formal addition to nitroalkenes

F3C I CFS

 

Table 25. Maruoka’s formal addition using catalyst 75

 

r

93"“93 75 (0.5-2 mol%) H30+ EKG/02

 

 

Rz/VNOZ + F? [1L6 THF,-78°C,1h R2“ N02
‘2...
entry Rl R2 mol.% 75 yield% d.r. %ee
1 Et 4-MeOC.,H4 l 99 >95/5 91
2 >> 2-MeOC6H4 0.5 99 >> 87
3 >> 2-ClC6H4 0.5 98 >> 80
4 >> 2-naphthyl 0.5 95 >> 89
5 >> furyl 2 92 >> 93
6 >> thiophenyl 2 85 >> 91
7 >> cyclohexyl 1 99 >> 92
8 >> n-hexyl 1 93 >> 76
9 Pr Ph 1 99 >> 91
10 MeOCH2 Ph 1 99 82/18 83

 

81

5.3. THE CONJUGATE ADDITION OF NOTROALKANES TO

NITROOLEFINES BY MEANS OF THIOUREA CATALYSIS

It is clear from discussion in the previous section that the addition of nitroalkanes
to nitroalkenes is a reaction that is just starting to be developed, and that there is plenty of
room for improvement. There are a lot of aspects to address and improve, namely the
problem of oligomerization, the diastereoselectivity in the non-asymmetric reaction, and
the enantioselectivities, that are modest in the case of Wang’s organocatalyzed protocol”.
The development of an efficient protocol for a diastereoselective non-asymmetric
synthesis of 1,3-dinitro compounds, as well as the identification of catalyst structures —
either metal based or metal free - that can induce high asymmetric inductions in these
products would be highly desired.

Our interest in the addition of nitroalkanes to nitroalkenes — as well as in the field
of thiourea catalysis — focused our approach towards the development of an efficient
protocol for this reaction in an investigation of the use of thioureas as catalysts. It is
known that thioureas can activate nitrostyrene (and other nitroolefins) for the addition of a
vatiety of nucleophiles". On the other hand, the activation of nitroalkanes and the
enhancement of their acidity through their interaction with thioureas is also known, in
several reports where the successful employment of nitroalkanes as reactive nucleophiles
is described”. Despite all these reports, a thiourea promoted addition of nitroalkanes to
nitroolefines has never been reported in the literature, although it would be expected that
the application of thioureas could facilitate the above transformation and possibly control

the selectivity of the addition (Scheme 24). A careful review of the field of thiourea

catalysis reveals that, concerning the employment of thioureas as catalysts, this particular

82

transformation is probably one of the few reactions involving nitro compounds that has

not been addressed so f ‘3‘.

Scheme 24. The conjugate addition of nitroalkanes to nitroolefines promoted by thioureas

S 3 ft
Ft R R R FL ,R

‘ ’ ‘ ' N N

ANA H ”A H AA AA N02 N02

_= : ——* _= _: _= : —* R'
\+, \+, O,+,,O Fl"

0 ND 0 N O N
.cH ./”

WI
I=’\

Our approach is depicted in Scheme 24. It is expected that the use of thioureas for
the above reaction will activate the nitroalkane, enhancing its acidity, and activate the
nitroolefine, enhancing its reactivity of the C-C double bond. The presence of a catalytic
amount of a thiourea - for substrate activation - and the presence of a base - for nitroalkane
deprotonation — should be sufficient for reaction acceleration, and might lead to a
chemoselective and diastereoselective reaction outcome. The chemoselectivity is an
important issue since polymerization can cause a decrease in yield of the Michael adduct
or require large amounts of nitroalkane to maintain good yields.

The rest of the chapter describes the use of thioureas in the context of the
conjugate addition, in an investigation that was focused on i) the employment of simple
aromatic thioureas for the promotion of the unasymmetric reaction and ii) the
development of a novel chiral thiourea catalyst for a highly enantioselective conjugate

addition methodology.

83

5.3.1. Development of the thiourea catalyzed non-asymmetric addition

It was realized that an investigation of the thiourea promoted addition of
nitroalkanes to nitroolefines should be first focused on the non-asymmetric reaction, and
on the effects that simple achiral thioureas would have on the transformation, since as it
was mentioned, the employment of thioureas in this particular type of reaction has never
been described before in the literature.

The small symmetrically substituted aromatic thioureas 76-78 shown in Figure 19
have been reported before in the literature in various studies conducted by Schreiner’s
group and other research groups, for the acceleration of several transformations including
Diels-Alder, Bayllis-Hillman, or Claisen rearrangements“. Schreiner has established that
diaryl thioureas of this type with electron withdrawing substituents on the meta and/or
para positions are the most effective candidates for rate acceleration in a diverse set of
reactions343". Thus, it was judged that this type of thioureas should be the focus of our
study. Additionally, we decided to also make and evaluate the bis-thiourea 79, based on
the idea that two thioureas incorporated in the same structure might result in activation of

both reactants.

Figure 19. The substituted aromatic thioureas used in this work

F30 NYN cr=3 cr=3 “A,“
O s O S O Q s Q
JL F 77 F
N N CF3 Cl H n Cl
CF 1“ . 7! 7i? Q r O
or 78 cu

Scheme 25. Synthesis of aromatic thioureas 47, 76-79

thiophosgene/ H

H o
EI3N N N 47, 40 /o
—» A , \A 76, 60%
ArNH2 r \[r r 77’ 40%
THF 3 78, 96%
CF3

 

CENHz scn’ : ‘CF3
NH2 THF

1 00%

79

The synthesis of thioureas 47 and 76 to 79 was accomplished using standard
literature procedures .4" As shown in Scheme 25, thioureas 47, 76-78 can be accessed from
the reaction of commercially available substituted anilines with thiophosgene and
triethylamine in THF. On the other hand, thiourea 79 can be made from 1,2-
diaminobenzene with 3,5-bis-trifluoromethylphenyl isothiocyanate.

We started our investigation by looking at the effect that the thioureas 47 and 76-
79 have on the rate acceleration of the conjugate addition. The reaction between trans-B-
nitrostyrene and l-nitropropane was used as a model reaction in CDCl3, employing only 5
equivalents of nitropropane, a very small excess compared to Alcantara’s protocol, and
using 10 mol% thiourea and 10 mol% triethylamine. The rate of the reactions was
followed by lH-NMR using the catalyst as an internal standard. Thiourea 78 was not
soluble in CDCl3 and therefore was not evaluated. The results for reaction with thioureas
47 , 76 and 77 are shown in Scheme 26. As can be seen in Scheme 26, the reaction without
thiourea (using 10 mol% Et3N) is very slow, affording only a 5% yield of the desired 1,3-

dinitroalkane after 3h. On the other hand, the rate acceleration that thioureas 47, 76, and

85

77 can cause is very significant: Thioureas 76 and 77 can increase the relative rate of the
reaction by approximately 5 and 6 times, respectively. It is obvious that thiourea 47,
bearing the 3,5-bis-(trifluoromethyl) moiety is the most efficient catalyst, leading to
almost quantitative conversion of the starting nitrostyrene in 2 hours, and affording ~60%

yield of the desired 1,3-dinitroalkane.
Scheme 26. Rate accelaration of the conjugate addition using thioureas 47, 76, 77

\ N02 10 mol% cat. N02
10 mol°/oEth N02
+

——>

/\/N02 cool3
5 eq.

 

yield (as)

 

0 20 40 60 80 100 120
tlme (mln)

 

 

The results shown in Scheme 26 were very promising at that point, because it
showed the potency of thioureas in promoting the coupling between the two nitro-
components. The diastereoselectivity of the reaction was not determined — the yields in

Scheme 26 refer to that for the combined syn and anti products. Since good yields can be
86

obtained in only two hours these results prompted further investigation of these thioureas,
especially 47.

Next, the efficiency of thioureas 47 and 76-79 was investigated in regard to the
yield and the diastereoselectivity of the addition of various nitroalkanes. The results for
the reactions of nitroethane, l-nitropropane and l-nitrobutane with nitrostyrene are shown
in Table 26. Nitroethane affords a mixture of diastereomers (entry 1), however it was
delightful to observe that the control of the selectivity for the rest nitroalkanes is much
better. All thioureas can control the selectivity of the reaction in favor of the syn isomer,
furnishing the desired product in good yields (up to 83%) and affording d.r. ratios between
77/23 and 87/13. The background reaction without thiourea (10 mol% Et3N, entry 7) is
very slow, leading to a trace of the product after an overnight period. It was interesting to
observe that the background reaction even with a stoichiometric amount of base is still
inefficient in promoting the reaction and only furnishes a 39% yield with a slight
preference for the anti isomer, revealing that the catalytic and the background reaction
complement each other in terms of stereocontrol (Table 26, entry 8). Finally, the reaction
with nitrobutane (entry 9) is consistent with the rest, giving 82% yield of the syn-enriched

product.

87

Table 26. The conjugate addition of nitroalkanes to nitrostyrene catalyzed by thioureas (a)

\ N02 R N02
catalyst
©/\/ EtSN N02

#

+ Fl NO2 Toluene

 

 

v r.t. 80a
yield% d.r. (syn/anti)
entry R catalyst (mol%) mol% Et3N

(b) (C)
1 Me 47 (10) 10 85 55/45
2 Et 47 (10) 10 83 77/23
3 Et 76 (10) 10 66 78/22
4 Et 77 (10) 10 58 87/13
5 Et 78 (10) 10 75 80/20
6 Et 79 (10) 10 67 84/16
7 Et - 10 7 n.d.
8 Et - 100 39 40/60
9 n-Pr 47 (10) 10 82 77/23

 

(a) Reaction conducted with 0.22M with respect to nitrostyrene in toluene for 12h using
10 equivalents of nitroalkane. (b) Isolated yield of the mixture of diastereomers. (c) Determined by
HPLC analysis.

It was interesting to observe that in all of the reactions in Table 26 all of the
thioureas can promote the formation of the desired product and at the same time suppress
the production of polymeric material. The excess of nitroalkane used in the above
reactions was only 10 equivalents, much lower than in Alcantara’s protocol (~l 10 eq.).50

Nonetheless, the rate acceleration that thioureas afford in favor of the desired product is

88

apparently high enough to provide quite clean reactions (in terms of the amount of
polymer formed). The reaction catalyzed only by Et3N (entry 8) gave a higher amount of
polymeric material, as judged by the messy crude lH-NMR spectrum obtained.

Since catalyst 47 turned out to be the optimum for the reaction, the addition of l-
nitropropane to nitrostyrene catalyzed by thiourea 47 was next examined in different
solvents and using various catalyst loadings in order to optimize the conditions and the
results are summarized in Table 27. The yields of the products were determined by the 1H-
NMR spectrum of the crude reaction mixture with triphenylmethane as internal standard.
The catalyst loadings for the reactions in toluene were gradually reduced as indicated in
entries 1 to 4. The diastereoselectivity of the reaction remains intact when the catalyst is
reduced from 10 to to 2 mol%, while a small drop-off is observed in the rate of the
reaction (entry 3, 58%). This can be suppressed by adding more nitropropane and
increasing the reaction time. For example, by using 20 equivalents of nitroalkane and
keeping the catalyst loading at 2 mol%, a very good yield of the product can be obtained
(85%, entry 4) in 2 days reaction time and no decrease is observed in the syn to anti ratio.
The reaction in other solvents also afforded good yields and d.r.’s, with THF being the
only exeption, which is as expected for a solvent that can function as an H—bond acceptor

and decelerate the reaction (entry 8, 45%).

89

Table 27. Optimization of conditions for the reaction using thiourea 47 (a)

NO
\ N02 thiourea 47 2
————> N02

Et3N

+ /\/ N02 solvent, rt 80a

 

entry solvent mol% 47 mol% Et3N eq. n-prNO2 yield% (b) d.r. (c)

 

l toluene 10 10 10 83 77/23
2 toluene 4 4 10 63 80/20
3 toluene 2 2 10 58 ((1) 79/21
4 toluene 2 2 2O 85 77/23
5 - 2 2 20 82 73/27
6 benzene 10 10 10 89 68/32
7 DCM 10 10 10 78 75/25
8 THF 10 10 10 45 77/23

 

(a) Reactions conducted in the corresponding solvent, for l2-15h, except entries 2-5 which
were run for 2 days. Entries 1-5 were run with 0.22M with respect to nitrostyrene, entries 6-8 were
run with 0.1M with respect to nitrostyrene. (b) Determined by lH-NMR of the crude mixture. (c)
Determined by HPLC analysis. ((1) Incomplete conversion.

After optimization of the solvent and catalyst loading, the conditions on Table 27,
entry 4 were chosen as the optimum and the scope of the reaction was examined for a

series of differently substituted nitrostyrenes. As can be seen from the data in

90

Table 28, a number of differently substituted nitrostyrenes can undergo the thiourea
catalde addition to give the syn 1,3-dinitroadducts 80a-i in good yields. Electron rich
nitrostyrenes (entries 2 to 4) were less reactive and needed longer reaction times (40h) to
go to completion, while electron deficient substrates (entries 5 to 9) afforded complete
conversions in only 16 hours. In every case, the starting material underwent complete
conversion (90-100%) with only small amounts of polymer formed, and the reactions were
in general very clean compared to the background reaction (Table 26, entry 8). It seems
that the thiourea catalyst is accelerating the formation of the desired adducts suppressing
the overaddition process and affording very good turnovers. Additionally, the reactions

favor the formation of the syn diastereomer in good selectivities that range from 3/1 to 9/1.

91

Table 28. Reaction scope for the thiourea catalyzed addition of l-nitropropane to [3-
nitrostyrene (a)

 

 

 

2 mol%
H H
F30. ; ,N\n/N. ; ,cr-‘3
S
CF CF
23 I E N 3 N02
R/V N02 + /\/N02 mo /° t3 : lNOz
(O'ZZM) 20 eq toluene, rt
BOa-I
yield% syn
entry R adduct overall yield% d.r. (c)
(b)
1 Ph 80a 87 75 77/23
2 4-MeO-C6H4 80b 82 73 75/25
3 2-MeO-C6H4 80c 83 66 80/20
4 4-Me-C6H4 80d 72 56 77/23
5 2-Cl-C6H4 80c 80 68 85/ 15
6 4-Cl-C6H4 80f 73 58 80/20
7 4-Br-C6H4 80g 66 57 86/ 14
8 3-Br-C6H4 80h 75 69 92/8
9 2-Br-C6H4 80i 66 6O 90/ 10

 

(a) Reactions run overnight except entries 2-4 which were run for 40h. (b) Isolated yield
after column chromatography. (c) Determined from the ratio of isolated products except entries 1
and 2, which were determined by the 'H-NMR of the crude reaction mixture.

An explanation for the preferential formation of the syn isomer can be obtained
from the examination of the corresponding transition states for the two diastereomeric
dinitroalkanes (Scheme 27). It seems that the lowest energy transition states for both the

syn and the anti product formation would be the ones where the groups with the largest A

92

values would be anti to each other. That means that the phenyl group of the nitrostyrene
(A=3.l kcal/mol) and the ethyl group of the nitropropane (A=l .9 kcal/mol) have to adopt
an anti relationship (the A value of the NO2 group is 1.1 kcal/mol). In the TS A, which
gives the syn product, this requirement also results in the nitro groups being opposite to
each other. On the contrary, considering the TS B that gives the anti product, the anti
relationship of the phenyl and the ethyl group results in the nitro groups coming close to
each other and pointing to the same direction, creating a steric hindrance between the two
thioureas. The steric congestion of the TS B seems to be a plausible explanation for the
observed stereoselectivity, although more experiments should be done to verify this

hypothesis (e.g. rate studies to prove that the reaction is second order in catalyst).

93

Scheme 27 . A possible explanation for the preference of the syn diastereomer formation in
the reaction catalyzed by thiourea 47

F3C©CF3

N -, F30
8:2 9 ,CH3
F3C H‘O’NQH CF3 NO
”IO-HM): —’ N20
CF. m ‘O-HN S 2
CF
TSA (favored) F C 3
3
syn
F30 CFS
H c “N S CF3
3 (5 \_/
\ HN
.‘ '_ ’ A \NO
HYN O F C CFS 2
N—0 —> No2

at
F30 3 anti

TS B (unfavored)

94

5.3.2 Development of the asymmetric conjugate addition of nitroalkanes to
nitroolefines

The success of thiourea catalyst 47 in promoting the syn-selective non-asymmetric
addition of l-nitropropane to fi-nitrostyrene prompted the investigation of an
enantioselective variant using chiral thiourea catalysts. The reaction design for the
asymmetric variant is shown in Schemes 28 and 29 and includes two different approaches.

a) Bis-thiourea catalysts that can mutually activate both reaction components
(Scheme 28). In this approach, one thiourea moiety is used for H-bond activation of
nitrostyrene and the second thiourea moiety is activating the nitroalkane for deprotonation.
In this case an external base will have to be used for the deprotonation, leading to the
formation of a nitronate, in a complex that will feature the two components in close
proximity for reaction. This concept was the one that was used for the development of the

asymmetric aza-Henry reaction in Chapter 4.

Scheme 28. Bis-thiourea approach for the conjugate addition

 

 

 

 

 

 

 

 

JSL S
N N’Ar NJL N Ar
/ H H K H H
8.. a b g,-
chiral ‘ N ’ B - hiral ‘ N ’
. . deprotonation c
scaffold
R’kfi' / _ scaffold RJ
\ JSL \ S
N N,Ar NJLN Ar
'7 ti H H
5 i a a
.+, , ‘
O‘N O O‘N'O

b) A bifunctional thiourea catalyst that will incorporate the base in its structural
framework (Scheme 29). In this approach, that has been used by Takemoto and other
groups for other transformations”, the nitroalkane will be deprotonated by the basic
moiety of the catalyst probably through H—bonding activation from the thiourea. The
nitroolefin could then displace the nitronate and interact with the thiourea, while the
nitronate anion should be held by the positively charged basic moiety through H-bonding,
in a complex that would again feature the two reaction components in close proximity for
reaction.

The investigation to be described in this chapter was thus focused on identifying

efficient catalysts for both approaches.

Scheme 29. Bifunctional thiourea approach for the conjugate addition

 

 

 

 

 

 

 

 

 

 

 

 

it .
N N’Ar N’lLN Ar
/ H H K H H
chiral Q‘fi '0 deprotonation chiral O, N ,0 ‘
scaffold /L\—H ' scaffold I
\ ‘ “ r
.B. ) \BH
4.
IRrA/NOZ
it
N N'Ar
K H H
R N02 chiral 0°N’O
LNG ‘ _ scaffold RA
Ar 2 \ Fl'
BH
_ O
+ O‘Nl
M

96

5.3.2.1 Identification of the optimum chiral thiourea catalyst

Concerning the first approach, involving bis-thiourea catalysts, the bis-thioureas 52
and 55 were evaluated for the reaction of B-nitrostyrene with l-nitropropane and 1-

nitroethane. These reactions were studied in various conditions and the results are shown

in Table 29.

Table 29. Reactions of nitroalkanes with B-nitrostyrene using catalysts 52 and 55

\ N02 N02
W 20 mol% cat. N02
+ RVN02 solvent, temp.

(10 eq.)

 

entry R catalyst (%) Et3N (%) solvent T(°C) yield% d.r. (syn/anti) %ee

 

1 Et 52 (20%) 20 DCM 25 80 70/30 23
2 Et 52 (20%) 20 DCM -20 80 85/ 15 22
3 Et 52 (20%) 2O toluene -80 25 81/19 34
4 Et - 40 toluene 25 35 56/44 -
5 Et 52 (20%) 5 toluene 25 80 90/10 19
6 Et 55 (20%) 5 toluene 25 100 80/20 17
7 Et 52 (20%) 0 toluene 25 0 - -
8 Me 52 (20%) 20 DCM -20 100 63/37 0

 

97

As can be seen in from the data in Table 29, the reaction with catalyst 52 at rt
afforded a promising yield for the syn isomer but the enantioselectivity was disappointing
(23%, enatry 1). An effort was made to increase the induction by lowering the temperature
to -80°C (entries 2, 3) but with no result other than a dramatic drop-off in the rate of the
reaction. The background reaction was then measured to determine whether it is
responsible for the low inductions, and it was found that 40 mol% Et3N can afford ~35%
product at rt. The background reaction was slower than the catalyzed reaction, but
nevertheless it was significant. In order to diminish its effect, lower amounts of EtsN were
tried (5 mol% with catalysts 52 and 55) but these efforts did not improve the selectivity. In
the absence of Et3N no reaction was observed, which means that the catalyst itself is not
able to promote the deprotonation of the nitroalkane to the nitronate. Finally, the reaction
with nitroethane was unsuccessful as well, affording racemic product.

From the above results it became evident that the first approach that was to utilize
bis-thioureas as catalysts was not successful, since neither 52 nor 55 was able to afford
satisfactory inductions for the Michael addition product.

The investigation then switched gears to the second approach, that is to the

identification of efficient bifunctional catalysts.

98

A careful look into Wang’s development of an organocatalytic protocol for the
Michael addition of nitroalkanes to nitroolefins52 revealed that bifunctional thioureas could
actually be promising candidates for the promotion of the enantioselective conjugate
addition. A part of his screening of catalysts included the bifunctional catalysts 66 and 81
in a neat reaction without solvent. As can be seen in Table 30, catalyst 66 (which is a
catalyst Wang’s group has developed for other asymmetric reactions) affords a modest but
promising enantioselectivity but the reactivity of the catalyst is not sufficient. On the other
hand, catalyst 81 is obviously more reactive,but the increased reactivity does not translate

to increased selectivity.

Table 30. Wang’s catalyst screening

 

 

 

H
M80 S CF3
CD .
N02
10% t.
\ N02 + YNOZ ca N02
56 eq. neat, rt
catalyst yield% %ee
66 29 48
81 73 30

 

The above results that Wang’s group reported are interesting because they reveal
that a thiourea catalyst could be efficient for the addition of nitroalkanes to nitroolefines, if

it incorporated a base that would be strong enough to promote the nitroalkane

99

deprotonation. It is obvious that 81 is reactive enough and this may be attributed to the
strong basicity of its nitrogen. Wang's catalyst, on the other hand, suffers from the low
basicity of the nitrogen that is in conjugation with the binaphthyl ring.

The above considerations inspired an interesting idea: it was envisioned that the
reactivity problem associated with the low basicity in Wang’s catalyst (66) could be
solved if the catalyst was derivatized with a more reactive base. Along these lines, catalyst
82 was designed, that features a dimethylaminopyridine (DMAP) moiety in its structure.
Catalyst 82 should be more reactive than 66 simply on the basis of the higher basicity of

the DMAP moiety compared to that of the dimethylaniline in 66 (Figure 20).

Figure 20. Design of a novel bifunctional catalyst based on BINAM

Wang's catalyst: DMAP analog:
Low basrcrty higher basicity

zit/£1310 OHAHO
""" \NfipKa-w -12

p-Ka:‘52. 2 (H20)O)

_ + / \ /
R ,0" ’ ' HN _ N\ Aminopyridine moiety:
\=N\ deprotonation of nitroalkane/
+ o - - HN\ interaction via double H-bond
' H“

In addition to the greater basicity of catalyst 82, another unique feature of this

catalyst is that after deprotonation of nitroalkane the 2-aminopyridinium moiety would be

100

expected to interact with the resulting nitronate through double H-bonding, possibly
resulting in a well-defined orientation in the chiral pocket. The ability of 2-
aminopyridinium ions to interact with nitro groups is documented in recent literature
reports. For example, Takenaka’s group found in 2007 that 2-aminopyridinium ion 83 can
activate nitroalkenes through double H-bonding toward the addition of N-methylindole

and the Diels-Alder reaction with cyclopentadiene”.

Scheme 30. 2-Aminopyridinium catalyst for nitroalkene activation

 

 

83 (dos-0.5%) N
R“ N02 > )\/ N02
— Ar-H R
TFPB 990/
/ / \ up to o
‘N
83 (ODS-0.5%) R
83 liq/V N02 : /
Q N02
up 10 54°/o

endo/exo: up to 30/1

In the case of catalyst 82, the aminopyridinium ion of the catalyst could interact via
H-bonding either with the nitroolefin or the nitronate ion, but it is anticipated that it will
most likely form an ion pair with the latter, which is negatively charged and thus more
electron rich.

Catalyst 82 can be synthesized trough a simple two-step sequence starting from the
commercially available (R)-BINAM 51. The first step involves a Buchwald amination of
2-chloro-4-dimethylaminopyridine using Pd catalysis. The reaction using a 1:1
stoichiometry between (R)-BINAM and 2-chloro-DMAP 85 gives 60% isolated yield of

the desired monosubstituted product 84, along with a small amount of disubstituted adduct

101

that can be easily separated with column chromatography. 2-Chloro DMAP can be
obtained from 2,4-dichloropyridine through an a nucleophilic aromatic substitution using

dimethylamine.

Scheme 31. Synthesis of catalyst 82

 

F C NCS CF
szdbaa 3 3
:23 00 ‘
NaOtBu
H
”“2 35(1 90) CF3 “JLN CFa
Owl-12 ..N N
Toluene DCM, rt ' H l \
82:00 60°/o /
\N/
A \
0
5° C Cl N/
820/0 85

The second step involves the formation of the thiourea moiety from the reaction of 84 with
3,5-bis(trifluoromethyl)phenyl isothoocyanate. The product 82 can be obtained in 60%
isolated yield after chromatography purification.

The preliminary results from the evaluation of the new catalyst were very promising.
In terms of reactivity, the data in Table 31 present a comparison with catalyst 66: catalyst
66 showed no reactivity for the given set of conditions (10 equivalents nitroalkane in
dichloromethane at r.t.) while 82 provided the dinitro adduct 80a in an excellent 90%
yield as determined by NMR spectroscopy. Note that in the absence of solvent catalyst 66
can promote the addition of a nitroalkane to a nitroalkene to a small degree. A preference

for formation of the syn diastereomer was observed along with 78% ee, which at this point

102

was very satisfactory. Reactions in other solvents revealed that benzene is actually the

optimum in terms of the stereoinduction (80%ee, entry 4).

Table 31. Evaluation of the new catalyst 82 in the conjugate addition (a)

\ N02 NC)2
©/\/ 20 mol% cat. N 0
g 2

solvent, temp.

 

+ /\/N02
(10 9%)

80a

 

entry catalyst solvent yield% (b) 80a d.r. (c) (syn/anti) % ee (c)

 

1 66 DCM O - -

2 82 DCM 90 59/41 78
3 82 toluene 90 63/37 78
4 82 benzene 88 64/36 80

 

(a) Reactions run with 0.1M with respect to nitrostyrene for 12h. (b) Determined from the
crude lH-NMR spectrum using triphenylmethane as internal standard. (c) Determined by chiral
HPLC.

After this study, catalyst 82 was chosen as the optimum catalyst for the reaction
and an effort to improve the result through an investigation of various conditions was

conducted.

5.3.2.2 Preliminary efforts for optimization

The effort towards optimizing the conditions for the reaction of B-nitrostyrene with
nitroalkanes using catalyst 82 begun with variations in the temperature of the reaction with
l-nitropropane. Specifically, lowering the temperature to -25°C was attempted in order to
achieve higher selectivities. Unfortunately this had a dramatic effect on the rate of the
reaction which decreased significantly and resulted in only 31% yield (Table 32, entry 2).

The opposite would be expected when warming up the reaction, however, in the case of

103

this particular transformation there is one complicating factor, the formation of the
polynitrostyrene (discussed previously). It tumes out that increasing the temperature to
50°C causes an increase in the rate of the polymer formation which is apparently higher
than the increase in the rate of the formation of the desired product. This seems to be the
explanation for the result in entry 3. The reaction at 50°C actually gave complete
conversion of the starting material, but with a much lower yield of the desired product that
at room temperature. The decrease in the mass balance is assumed to be due to the
formation of polymer and this is consistent with the complicated NMR spectrum that was
observed for the crude reaction mixture. Unfortunately, temperature seemed to be a
limiting factor and it was obvious that any further optimization had to be conducted only

at room temperature .

104

Table 32. Preliminary optimization of conditions using catalyst 82 (a)

 

 

 

m N02 R N02
20 mol. °/o 82
= N02
+ Fl VN02 solvent, temperature
(10 eq.)

entry R solvent T(°C) yield%* d.r. %ee syn % ee anti
1 Et benzene rt 80 60/40 82 7O
2 Et toluene -25 31 n .d. n.d. n .d.
3 Et benzene 50 54 50/50 80 7O
4 Et neat rt 100 50/50 78 72
5 Et benzene, 4A M.S. rt 62 45/55 40 44
6 Me benzene rt 100 57/43 62 6O
7 Pr benzene rt 96 55/45 8 1 79
8 Et benzene rt 75 66/33 82 n.d.

 

(a) Reactions run with 0.1M with respect to nitrostyrene for 12h. *Entries 1-7: lH—-NMR
yield using triphenylmethane as internal standard. Entry 8: Isolated product after silica gel
chromatography.

Two more experiments were conducted with l-nitropropane: a neat reaction

without any solvent (entry 4), in order to increase the rate and the yield, and a reaction

using molecular sieves, in an effort to ensure dry conditions and removal of any traces of

moisture that would interact with the catalyst and/or the nitrocompounds (entry 5).

Unfortunately, neither change was helpful, since the neat reaction was too fast and

unselective in terms of the dr. ratio (entry 4), and the molecular sieves had a diminishing

105

effect on the selectivities (entry 5). The latter was not expected, and the effect of the
sieves on the reaction outcome is not clear.

Finally, a few more nitroalkanes were tried in the reaction at r.t. Nitroethane (entry
6) gave a lower but still respectable enantioselectivity (62% for the syn product) along
with excellent yield, while l-nitrobutane (entry 7) afforded results similar to those with l-
nitropropane, showing that the reaction is not very substrate sensitive (at least concerning
the nitroalkane) and can be applied to different nitroalkane nucleophiles. Among the last
two substrates, l-nitropropane was chosen for further study because of the easier
separation of the enantiomers in the chiral HPLC.

It should be noted here that for all the experiments in Table 32 the yield that is
reported is determined from the crude 'H-NMR spectrum using triphenylmethane as a
standard. Specifically, a known amount of triphenylmethane was added to the reaction
mixture at the end of every reaction and the yield was calculated by integration of
aliphatic protons of the product and the methine proton of the standard that has a
characteristic absorption at ~ 5.5 ppm. This accuracy of this method was checked by
performing a couple runs where the product was isolated by chromatography. The data in
entry 8 refers to isolated product and is the average of two runs. The consistency with the
data in entry I allowed for the reliable use of triphenylmethane as a standard in further
experimentation and optimization. In addition, the high polarity of the catalyst allowed for
a convenient determination of the ee’s and d.r. ratios without the need of column
purification: All that was needed in order to obtain this data was to simply filter the crude
reaction mixture through a short Pasteur pippete containing a small amount of silica gel.

The polymer and the catalyst stayed on the top of the silica, and product of very good

106

purity was eluted, which allowed for very clean HPLC spectra. This method proved to be
very useful in rapidly optimizing the reaction without the need to perform column
chromatography for each and every entry, thus saving a lot of time. The method described
here was applied to all the experiments that are discussed in the following sections, except
of course the section that deals with the substrate scope, where isolation and purification
of products was required.

A second point of importance is the reactivity and the chemoselectivity of the
catalyst 82. For the reactions conducted at r.t. the conversion is always nearly quantitative,
with only small amounts of polymer formed (<20%). The chiral thiourea seems to activate
the starting materials in favor of the formation of the desired adduct suppressing the
production of oligomers or polymers, just as the achiral thiourea 47 did for the non-

asymmetric addition (Table 28).

5.3.2.3 Effect of catalyst loading: discovery and investigation of a highly unusual

selectivity trend

The experiments discussed in the previous section led to the conclusion that the
optimum conditions for the maximum yield and selectivity of the reaction are met when
the reaction is conducted at room temperature in benzene. Up to this point the experiments
have been performed using 20 mol.% catalyst, along with 10 equivalents of l-
nitripropane. The next objective thus was to try to reduce the catalyst loading in order to
achieve a satisfactory balance between the amount of catalyst spent and the yields and
selectivities obtained.

The investigation was first focused on the effect of catalyst loading. A series of

experiments were conducted using 10 equivalents of l-nitropropane in benzene at room

107

temperature. All conditions were kept consistent and a number of different catalyst
loadings from 0.5% to 100% were screened. Analysis of the data led to a completely
unexpected result from the above study: It was intriguing to observe that a reduction of the
catalyst loading not only didn’t cause a drop in the selectivities, but on the contrary, it lead
to an unexpected enhancement of the selectivities! As can be seen from the data in Table
33 a stoichiometric amount of catalyst affords a preference for the anti isomer, with ee’s in
the 60’s and 70’s, while catalyst loadings from 20% and lower favor the syn isomer, with
the asymmetric induction increasing very significantly. The enantioselectivity of the syn
adduct spans a range from 70% (entry 1) to 95% where the catalyst is used in only 0.5
mol.% with respect to the nitroolefin. A significant increase in the diastereomeric ratio
was also observed for reduced loadings, with an increase to 86/14 for the lowest loading
(entry 7). This selectivity trend is obviously extremely unusual. Additionaly, the fact that
catalyst loadings as low as 0.5% can induce selectivities as high as 95% demonstrates that

thiourea 82 is a remarkably selective catalyst.

108

Table 33. An unusual selectivity trend observed for the catalyst loading study for 82 (a)

CF3

S Q
03 min CF,
.N N
©/\/ / NO2
(0.1M)
/N\

mo2

 

r

"' /\/ N02 Benzene,rt
(10 eq.) 80°

 

entry cat.load.% NMR yield% (b) d.r. (syn/anti) % ee (syn) % ee (anti)

 

l 100 60 39/61 70 63
2 20 80 60/40 82 70
3 10 40* 71/29 84 60
4 5 70* 76/24 90 70
5 2 43* 82/18 94 n.d.
6 l <30* 87/ 13 94 n.d.
7 0.5 18* 86/14 95 n.d.

 

(a) Reactions were run for 15h except entry 7 which was run for 40h. (b) Determined
using triphenylmethane as internal standard. *Incomplete conversion

It should be noted here that the rate of the reaction slowed down significantly
when the catalyst was used in a loading of 10 mol% or less. For entries 3 to 7, the
conversion was incomplete and a significant amount of starting material was unreacted.
Nevertheless, it is worth mentioning that for those reactions with reduced loadings the

crude 1H-NMR spectrum was generally cleaner with respect to polymer formation. In

109

general, the lower the catalyst loading that was used, the less polymer was formed and the
cleaner the reaction mixture that was obtained.

The trend that the enantioselectivity for the syn product followed with respect to
the catalyst loading was very smooth, and this is apparent when plotting the induction for

the product and the catalyst loading, as illustrated in Figure 21 .

Figure 21. Plot of the asymmetric induction for syn - 80a versus the catalyst loading

eo(syn)96
8

i 8
41 I

 

 

on
N

 

10 15 20 25
catkmdl%

O
01

It is obvious that 0.5-2 mol% seems to be the optimum catalyst loading with
respect to the selectivities, but unfortunately these amounts of catalyst are not practical
because the yields of the product are not acceptable. The solution to this problem came

with further optimization and will be described later on in the next section.

110

The investigation was then focused on rationalizing the above set of data in order
to shed light on the mechanism of this transformation. A careful look into the selectivity
trend shows that there are two factors that could possibly operate in the higher inductions
observed upon decrease of the catalyst loading: (a) the decrease of the catalyst
concentrations and / or (b) the increase of the nitropropane/catalyst ratios (since the
catalyst amount is reduced across the table but the substrate amount is kept constant).

Three different scenarions were envisioned in order to account for the unusual
effects that these two factors could have on the selectivities: A) stereochemichal
“scrambling” could happen if a nucleophile bound to one catalyst moiety reacts with an
electrophile bound to another catalyst when the catalyst concentration (or loading) is high
enough and catalyst moieties have a higher probability of coming close to each other in
solution. This hypothesis is illustrated in Figure 22 as pathway A. If it is presumed that
pathway D, which represents the design in the original model, is the more selective, then
the scrambling of stereochemistry at high catalyst concentrations (pathway A) would
occur because the nitroalkane would attack nitrostyrene in a stereochemically less defined
way, or even attack the opposite side of the C-C double bond rather than the one that is
exposed in D. According to this scenario, reducing the catalyst loading should disfavor

pathway A and result in higher enantioselectivity.

lll

Figure 22. Possible scenarios to account for the catalyst loading effects

   

A (Cross condensation)

l lower cat. oonoentratlon

F
3C CF3

Sj’NH .- O
,‘ /
NH___O,N\/\Ar

    

"""NH _ _ _ Ot—N
Jim” \
\N \ (+
/

D
(moraselecfivespecies)

lower cat. ooncentratlon/
higher RN02lcatalyst ratio higher HNOg/cat. ratio

CF3
C

Mi? F300 HQ’
HNYS S I

N .

a ,‘tN’ ,x’HN O

O b

O _

—N\ B

I

   

112

B) Intermolecular H-bond between different catalyst moieties that would lead to catalyst
self-aggregation (species B in Figure 22) when the catalyst loading is high and the
nitroalkane / catalyst ratio is small. This scenario seems reasonable on the basis of the
potential of the electron-rich sp2 nitrogen of the DMAP moiety of the catalyst to accept an
H-bond. As the nitroalkane/catalyst ratio increases (low cat. loadings) it is expected that
the interaction of the catalyst with the substrates will be more favorable, and this
interaction can break-up the aggregates and lead to species D. The magnitude of the
nitroalkane/catalyst ratio indeed ecomes very large when switching from 100 mol%
catalyst loading (ratio=10/1) to 20 mol% catalyst (ratio=50/1), to 5 mol% catalyst (200/1)
and then to 0.5 mol% catalyst (2000/1), and so at these very high ratios it is expected that
the catalyst would increasingly be more likely to associate with the substrates than with
itself. The observed trend on the selectivities then can be explained if the aggregated
species B can also catalyze the reaction but in a less enantioselective fashion than D.

C) Intramolecular H-bonding of the catalyst as indicated in species C would leave only
one thiourea N-H bond available for substrate activation, and this should presumably lead
to less defined substrate orientations and thus to lower asymmetric inductions. Species C
involves an intramolecular interaction and thus does not depend on the concentration of
the catalyst itself, but its formation should be clearly dependent on the nitroalkane/catalyst
ratio. Again, low catalyst loadings would translate to high nitroalkane/catalyst ratios,
which should promote the break-up of the intramolecular H-bond (just as in the case of B)
and the formation of the more selective species D. As with species B, the explanation of
the observed trend would require that species C be able to catalyze the reaction with lower

inductions than species D.

113

We next embarked on a series of studies in order to distinguish between scenarios
A, B and C, and to probe the mechanism of this highly enantioselective addition. First, it
is obvious that pathway A should be clearly concentration-dependent and that dilution of
the reaction mixture should disfavor the formation of A and lead to higher inductions. A
series of reactions were conducted where the same set of teagents and catalyst was
dissolved in different amounts of benzene so that the catalyst loading and the
nitroalkane/catalyst ratio were kept the same while the concentration of all the species
including the catalyst were varied. The results of these experiments are depicted in Table

34.

Table 34. Concentration study of the reaction (a)
20 mol% CF3

00 are.
..N N\
mmz 00 HEN? N02

 

 

 

k No2
+ /\/N02 Benzene,rt 7
(10 eq.) 80a
entry [catalyst], M yield%(NMR) (b) d.r. %ee (syn) %ee (anti)
1 0.040 77 58/42 84 70
2 0.010 71 64/36 85 70
3 0.005 55 52/48 78 64
4 0.003 45 55/45 67 48
5 0.002 31 55/45 70 42

 

(a) Reactions run for 48h. (b) Determined from the IH-NMR of the crude reaction mixture using
triphenylmethane as internal standard. All entries gave incomplete conversion of nitrostyrene.

ll4

The concentration study turned out to be very informative since the results were
the exact opposite of what would be expected if pathway A was operating. When the
reaction was performed at more diluted concentrations a very significant drop in the ee’s
of both the syn and the anti products was observed. This clearly rules out pathway A. The
dependence of pathway B on the concentration in not very clear at this point. An increase
of the concentration may promote the formation of the aggregated species B, but on the
other hand the increase in the concentration may allow for a more efficient interaction of
the catalyst with nitropropane, thus disfavoring the aggregation. Thus, these two effects
would offset each other.

The concentration study also led to an important observation: An increase of the
catalyst concentration is accompanied by higher inductions. This was not the case in the
catalyst loading study (Table 33) where increased loadings — and thus, catalyst
concentrations - gave lower inductions. This observation was the key in realizing that for
the catalyst loading study, the factor operating in the catalyst loading trend is probably not
the catalyst concentration itself, but rather, the nitroalkane/catalyst ratio. In other words,
the nitroalkane/catalyst ratio was responsible for the unusual selectivity trend in the data in
Table 33, and not the catalyst concentration itself.

Support for this idea was provided by a simple control experiment: two different
catalyst loadings from Table 33 were tested again, this time with the amount of 1-
nitropropane adjusted so that the nitropropane/catalyst ratio would remain the same. These
reactions gave the same enantioselectivity for the major (syn) product. These two catalyst
loadings had afforded a significant difference in the enantioselectivities under the

conditions in Table 33 (82% and 90% ee), and so the results here demonstrate clearly that

115

the nitropropane/catalyst ratio is the determining factor for the highly unusual selectivity

trend associated with 82.

Table 35. Cat. loading study with stable nitroalkane/catalyst ratio

\ N02 N02
X mol% cat. 82
——> N02

 

 

+ /\/N02 Benzene,rt
(x eq.) 80a
X n-prNOZ/catalyst ratio %ee (syn)
20 100 92
5 100 92

 

The idea that the nitroalkane/catalyst ratio is determining the selectivities was
further probed with a survey of the effects that the amount of l-nitropropane has on the
reaction outcome. It would be expected that increasing the amounts of nitroalkane for a
given catalyst loading should lead to increased inductions. Along these lines, a set of
reactions was undertaken where catalyst loading and concentration were kept constant and
the number of equivalents of nitroalkane was varied, from 2 to 30-fold excess with respect
to the nitroolefin. This study served in two purposes: it was not only a control experiment
for probing the importance of the nitroalkane/catalyst ratio, but also a survey for the
determination of the optimum amount of l-nitropropane that should be used in the

reaction. The results are presented in Table 36.

116

Table 36. Effect of nitroalkane equivalents (a)
20 mol% 01:3

QC 8 Q
H H
NJLN cr=3
g N\
w 00 p
(0.1M)
/N\

 

 

 

N02
+ >
/\/ N02 Benzene,rt 80a
equivalents NMR yield% d.r.
entry time(h) %ee syn %ee anti
n-PrNO,_ (b) (syn/anti)
1 2 60 36 46/54 82 70
2 5 60 43 46/54 83 74
3 10 15 85 60/40 82 70
4 15 15 100 63/37 83 71
5 20 15 85 65/35 85 72
6 30 15 100 66/33 87 76

 

(a) Entries 1-3 were run for 60h. Entries 4—6 were run for 15h. All reactions gave complete
conversions except entries 1,2. (b) Determined from the 'H-NMR of the crude reaction mixture

using triphenylmethane as internal standard.

The results from the study on the variation of the equivalents of nitroalkane were

interesting both in terms of the effect on the diastereoselectivity and the enantioselectivity

of the syn adduct. Concerning the diastereomeric ratio, there is a slight preference for the

anti isomer at small excesses of the reagent (enties 1 and 2) that inverts to a 2/1 selectivity

for the syn product when the reagent is used in more than 15 fold excess (entries 4 to 6).

Concerning the enantioselectivity of the syn adduct, the induction stays at 82—83% ee for

2-15 equivalents of nitropropane, but for larger excesses of 20 and 30 equivalents an

117

increase at 87% ee was seen. The ee of the anti product also went up from 70 to 76%, with
only the result in entry 2 showing a deviation from an otherwise smooth trend. The trends
in the selectivities in general agree with the notion about the nitroalkane/catalyst ratio is
directly proportional to the asymmetric induction.

As a bonus from the optimization of the reaction through the above experiments, it
was delighting to observe that the ee of the product can be increased by just adding more
nitropropane to the reaction mixture. On the other hand, considering about the rate and the
yield of the transformation, it became apparent that at least 10 equivalents of the
nucleophile should be used in order to obtain good yields in a reasonable amount of time.

The question of whether the catalyst concentration or the RNOzlcatalyst ratio is
responsible for the selectivity trend has been answered, however the question of whether
species B or C is operating is still pending, because the results of all these experiments to
this point are in accordance with the involvment of either species. In other words, the
RNOz/catalyst ratio should be expected to influence both species in the same direction, as
illustrated in Figure 22. To distinguish more clearly between species B and C and to}
determine whether the suspected self-association of the catalyst is intramolecular or
intermolecular, non-linear studies were conducted. In non-linear studies the relationship
between the enantiomerical purity of the product and that of the catalyst is surveyed.
Usually a non-linear relationship indicates the existence of catalyst aggregates, while a
linear one indicates that the catalyst functions as a monomer. For the purpose of this study,
three different samples of scalemic catalyst were prepared by mixing (R) and (S) -82 to
make catalyst samples that were 20%, 50% and 80% ee. The study was undertaken in

reactions using both 2% and 20% catalyst, such that the possibility of aggregation at

118

different concentrations of the catalyst can be tested. In the beginning of the discussion it
was mentioned that species B in Figure 22 should be favored at low RNOZ/catalyst ratios.
The equivalents of nitropropane used in these experiments for 2% and 20% catalyst were
30 and 15 respectively, such that the conditions wouldn’t disfavor the formation of the
suspected polymer in the second case. In other words, in case the high catalyst loading
(20%) created a catalyst aggregate, the RN02/catalyst ratio used here was much smaller
than in the reaction with 2% (75/1 versus 1500/1) so that it wouldn’t break up the
aggregate and give a misleading linear plot. The results of the study are illustrated in
Figure 23.

“s..‘.‘.:... 2% e.g., 5...... stair... " ”

Series 2: 20% cat" 15 eq. nitroalkane

100
90
80
70
60 <
50

 

-O- Seriesl
+ SerlesZ

30

20 «

10 .

o . - c . -

o 20 40 so so 100 120
%ee catalyst

%ee syn product

 

 

 

 

 

 

Figure 23. Plot of optical purity of 82 versus optical purity of syn-80a

The non-linear studies in Figure 23 show that there is a linear relationship between

the cc of the catalyst and the ee of the major diastereomer of the product 80a, both at 2%

119

and 20% catalyst loading. This indicates that even when the catalyst is used at high
concentrations or loadings, it still functions as a monomeric species, even if the
RNOz/catalyst ratio is much smaller. This study may not disprove the existence of
aggregated catalyst, but it is a strong indication that the active species is a monomeric
form of the catalyst 82. Consequently, the intramolecular H-bonded shown in Figure 22
as species C is the only scenario that fits all the data obtained so far. In case species C is
operating than the data obtained from the concentration study in Table 34 make sense if
the increase in the concentration leads to more efficient interaction between the catalyst
and the substrates, and helps to break-up the itramolecular H-bond in the catalyst.

Given the fact that excess l-nitropropropane can lead to increased asymmetric
induction that presumably results from altering the catalyst structure through disruption of
H-bonding leading to the conversion of species C to D (Figure 22), one could argue that
other compounds that could mimic the H-bonging ability of l-nitropropane should in
principle do the same. In other words, it would be expected that additives that could H-
bond with the catalyst and break-up the self-association should promote the formation of
species D just as nitropropane does, if our hypothesis is correct. This idea was intriguing
not only because it would give us one more probe to test our hypothesis, but also because
some additives might work even more efficiently than l-nitropropane affording even
higher inductions and the identification of such additives would furthermore contribute to
the optimization of the reaction.

Along these lines, a series of additives were tested in the reaction of nitrostyrene
and l-nitropropane using 20 mol% catalyst. The additives were chosen on the basis of

mimicking the ability of l-nitropropane in forming H-bonds with the catalyst, and thus

120

small molecules that can function as H—bond acceptors were chosen. The results of this

study are presented in Table 37.

Table 37. Survey of additives in the reaction with catalyst 82 (a)

20 mol% CFa

HJLH CF3
. H I

N N
\
\ No. CC /
No2
(0.16M) N

 

 

 

/ \ _ No2
+ N... 3233.... ' 3..
10 eq.
entry additive (eq.) yield% (b) d.r. ee% syn

l - 100 66/34 88

2 PhNO2 (5) 83 58/42 92

3 THF (5) 62 63/37 89

4 MeCN (5) 83 64/36 92

5 DMSO (5) trace nd nd

6 EtOAc (5) 67 57/43 89

 

(a) Reactions run for 12 h. (b) Determined from the Il~I-NMR of the crude reaction mixture
using triphenylmethane as internal standard.

First of all, it was observed that all the additives slowed down the reaction, some to
a small extent (PhNOz, MeCN) and others more significantly (THF, EtOAc) while only a
trace of the product was obtained when DMSO was used. This is consistent with the

presumtion that all the additives interact with the catalyst. In particular DMSO, as maybe

121

expected, seems to completely block the catalyst preventing its interaction with the
substrates. Out of the five additives that were tested, both PhNO2 and MeCN afforded an
increase in the ee from 88% to 92%. This small but real change suggests that these two
additives interact with the thiourea of the catalyst shifting the equilibrium from the self
associated catalyst C to the more selective complex D, in accordance with the proposed
mechanism. It was also very interesting to observe that when acetonitrile is used as a
solvent the yield and the selectivities drop significantly (51%, 62% ee), as is typically the
case in thiourea catalysis when polar H-bonding acceptor solvents are employed. However
when acetonitrile is used as an additive in small amounts it contributes to the enhancement
of the ee.

Finally, the hypothesis of catalyst self-inhibition through species C was further
probed with temperature experiments. If, as proposed, catalyst self-inhibition is Operating
as a factor in regulating the reaction selectivities, it would be expected that this inhibition
should be favored at low temperatures, where the energy provided to the system is lower.
As the temperature rises one would expect that the increased energy of the system should
give to the molecules the necessary energy to overcome the entropic barrier of breaking an
intramolecular H-bond in the catalyst which can lead to intermolecular H-bonds between
the catalyst and the substrates. This scenario would lead to a counter inturtive outcome,
lowering the temperature would lead to decreased asymmetric inductions.

As a test of this hypothesis, the cc of the syn product was studied as a function of
temperature in a range between -40 and 70°C. Unfortunately, at high temperatures the
catalyst underwent decomposition according to the crude lH-NMR spectrum. Experiments

at low temperatures provided quite interesting results. Lowering the temperature from 5 to

122

-10° C led to a sharp decrease in the induction from 84% to 59% ee. This is in accordance
with internally H-bonded catalyst, because lowering the temperature, and thereby the
energy provided to the system, should otherwise lead to increased inductions. The trend
that was observed actually was not a gradual enantioselectivity decrease. As shown in
Figure 24, the ee starts rising again when going to -20 and -30°C and at -40°C the
induction is back to the initial levels. It seems that the internally H-bonded catalyst is

maximized at -5 to -10°C, and

Figure 24. Temperature study of the asymmetric addition

20% 82
N02 + N N02 N02
(01 M) (10 eq.) Benzene O 80

90
1,;

80*

7 .

60‘

 

{D

\

504
4o-

% 08 (SW!)

30‘
20-
10-

 

 

A
V r

o 20 4b so 80
"Q

at lower temperatures the increased selectivities from the reaction either through species D

or C takes over and almost recovers the induction.

123

In any case, the sharp drop in the selectivity when lowering the temperature is not
usually encountered in asymmetric catalysis when monomeric catalytic species is involved
in the reaction, and the result that was observed here is an indication for some sort of

intramolecular interaction, in agreement with the intramolecularly H—bonded species C.

5.3.2.4 Optimization of conditions and substrate scope

From the studies described in the previous section, the effect of most of the
reaction parameters has already been defined for catalyst 82 in the reaction of nitrostyrene
with l-nitropropane. The next objective is then to achieve the best combination of these
parameters so that the optimum balance between yields and selectivities can be identified.

In the case of the asymmetric addition with catalyst 82, it was fortunate to observe
that the unusual selectivity patterns described in the previous section were such that an
increase in both the concentration and the amount of l-nitropropane favor an increase not
only in the yields, but also in the diastereoselectivities and the enantioselectivities as well.

Thinking about catalyst loading, tha data in Table 33 suggests that in order to make
the reaction more selective small catalyst loadings have to be used. According to the the
data however, loadings smaller than 20% afford incomplete conversions and low yields.
Catalyst loading is apparently the only factor that has opposing effects on the yields and
selectivities. On the other hand, since an increase in equivalents of nitropropane and
higher concentrations favor the selectivities, it would be expected that using a large excess
of nitropropane and high concentrations would help to overcome this issue of sluggish
reactivity at 2 mol%, 1 mol% or even 0.5 mol% catalyst loadings. Therefore, further
optimization was focused on the objective of “pushing” these reactions to give higher

conversions and yields. The results are presented in Table 38.

124

Table 38. Optimization of reaction variables (a)
CF3

00 3.25..
[£:/\/No2
/N\

 

 

 

No2
+ ; No2
/\/ N02 benzene
[C] prNO2 yield% ee% (c)
entry‘ cat.load.% time (h) d.r. (c)
nitrostyrene equiv. (b) (syn)
1 2 0.2M 20 16 23 84/16 94
2 2 0.2M 30 40 90 84/ 16 94
3 2 0.3M 30 40 100 80/20 93
4 l 0 .2M 30 40 66 87/ 13 95
5 0 .5 0 .2M 30 40 40 88/ 12 96
6 0.5 0.4M (neat) 30 40 40 80/20 94

 

(a) Reactions were run for 12h. (b) Determined from the 1H-NMR of the crude reaction
mixture using triphenylmethane as intemal standard. (c) Determined by chiral HPLC.

As can be seen from the data in Table 38, using 2 mol% catalyst affords only 23%
yield of 80a and a lot of unreacted nitrostyrene was observed under the conditions in entry
1. However, this was solved by using more forcing conditions. Apparently increasing the
amount of nitropropane, the reaction time and the concentration leads to an excellent
NMR yield with the selectivities in the same high levels (entries 2 and 3). The same

strategy was attempted for the reactions with 1 mol% and 0.5 mol% catalyst loadings, but

125

in these cases it was not possible to push the yields to a satisfactory level, even when neat
conditions were employed (entries 4 to 6).

From the above study summarized in table 38, the conditions in entry 2 were
chosen as the optimum and then the scope of the asymmetric addition of l-nitropropane to
nitroolefins was studied using these parameters. Various differently substituted

nitrostyrenes were tested and the results are shown in Table 39.

126

Table 39. Substrate scope of the asymmetric addition of nitroalkanes to nitrostyrenes
using catalyst 82 (a)

 

 

 

NO
MN N02 + r=l\/No2 2% mol. 82‘ l 2
7 N02
(0.2M) 30 eq benzene, rt R
800-1
overall yield% %ee
entry Ar R adduct d.r. (0)
yield% (b) syn (b) (syn)
1 Ph Et 80a 80 60 84/16 95
2 4-MeO-C6H4 Et 80b 60 47 85/ 15 94
3 2-MeO-C6H4 Et 80c 55 45 80/20 94
4 2-MeO-C6H4 Et 80c 94 72 80/20 94
5 4-Me-C6H4 Et 80d 72 58 80/20 94
6 2-Cl-C6H4 Et 80c 84 62 74/26 92
7 4-Cl-C6H4 Et 80f 75 62 83/17 92
8 4-Br-C..,H4 Et 80g 75 65 87/ 13 93
9 3-Br-C6H4 Et 80h 69 62 90/ 10 94
10 2-Br-C6H4 Et 80i 78 58 75/25 92
1 1 Ph n-Pr 80j 78 70 90/ 10 91

 

(a) Reactions run for 40h except entries 7 and 8 which were 17h and entry 4 which was 60h. All
reactions were 0.2M in nitroolefin except enrty 4 which was 0.33M. (b) Isolated yields after
column chromatographic purifiction. (c) Determined by the weight of the separated diastereomers
except entries 1-4 which were determined by 'H-NMR.

As can be seen from the data in the Table 39, the reaction is highly efficient for a
range of substituted nitrostyrenes. Good isolated yields in the range of 69-94% of the
adducts 80a-i can be obtained with only 2 mol% catalyst, demonstrating a catalyst with

very efficient turnover ability. In the case of the electron rich methoxy-substituted styrenes

127

the rates of the reactions were slow when the optimum conditions were employed, and
isolated yields of 47% and 45% were obatained for the syn adducts for the 4-MeO and the
2-MeO substrates respectively. This problem could be suppressed in the case of the 2-
MeO adduct 80c (entry 4) by performing the reaction at a higher concentration (0.3M) for
a prolonged time (60h). In this way an excellent overall yield could be obtained (94%
overall) without any loss in the selectivities. The same didn’t work for the 4-MeO
substrate, though, since its conversion remained the same (~70%) despite the efforts to
push the reaction. Electron deficient chloro and bromo substituted nitrostyrenes reacted
faster and gave mostly complete conversions of the starting material, some of them in
short reaction times ( 17h for 4»Cl and 4-Br).

On the other hand, electron neutral, withdrawing or donating groups do not seem
to affect the selectivities. The diastereoselectivities are in the range of 74/26 tp 90/ 10 in
favor of i the syn adduct, while the asymmetric inductions are all between 92-95%,
demonstrating a highly efficient enantiodiscrimination with excellent consistency.

On the other hand, the reaction with l-nitrobutane (entry 11) is consistently
efficient, giving a slightly higher diastereomeric ratio along with 91% ee for the syn
isomer, indicating that the reaction is not restricted to l-nitropropane, but could be
potentially applied to higher nitroalkanes.

Switching from trans-B-nitrostyrene to the more demanding trans-or-methyl-B-
nitrostyrene proved disappointing. A reaction using 2 mol% catalyst was run for 2 days

and afforded no product but only unreacted starting material (Scheme 32).

128

Scheme 32. Reaction with trans-a-methyl-B-nitrostyrene

 

N
rah/Y N02 + /\/N02 2%: mol. 82 O2
: N
benzene, rt ph 02
2 days

0%, no reaction

An effort to extend the scope to branched nitroalkanes turned out to be more
promising. 2-Nitropropane features a secondary nitro group and thus is much less reactive
than l-nitropropane. Addition of 2-nitropropane to nitrostyrenes have only been reported
by Wang’s group52 and according to their protocol many (6-12) days are required for the
completion of these transformations. When the reaction of 2-nitropropane with
nitrostyrene was attempted using '10 mol% catalyst 82, a 55% yield of the adduct 83 was
obtained with 68% ee (Table 40). Lowering the catalyst loading to 5% afforded a better
induction (76%) which is in accordance with what has been seen for l-nitropropane,
however, the increased induction is at the expense of the yield, although in this case the

reaction was somewhat dilute (0.13M) and that might be a reason for the sluggishness.

129

Table 40. Reaction with 2-nitropropane using catalyst 82

Ph/V N02 + N02 Catalyst 82 32
\f benzene, rtr Ph N02

24h

 

 

entry %catalyst equiv. 2-nitropropane NMR yield% %ee

 

l (a) 10 27 55 68

2 (b) 5 55 27 76

 

(a) Reaction was run with 0.2M with respect to nitrostyrene. (b) Reaction was run with
0.13M with respect to nitrostyrene.

Finally, an effort was made to extend the scope of the reaction to aliphatic

nitroolefins. Nitroalkene 87 was synthesized according to a common procedure44b

from
butyraldehyde 85 through an addition of nitromethane to make 86 followed by an
elimination of water using CuCl and DCC (Scheme 33). Reaction of 87 with l-
nitropropane using 2 mol% catalyst 82 afforded a 66% conversion to the dinitro
compound 88 in a clean reaction, according to the crude lH-NMR spectrum, as a 58:42
mixture of diastereomers with 42% and 28% ee’s for the major and the minor

diastereomer respectively. The pruduct was not isolated though because most of it was lost

during purification because of its volatility.

130

Scheme 33. Synthesis and reaction of the aliphatic nitroolefine 88

 

 

 

? MeNOZ/NaOH 0“ DCC,CuCl
EtOH/AcOH 7 «BK/N02 320, n WNoz
87% 42% 87
catalyst 82 N02
WNOz + /\/N02 (2 mol%) ; 3L3“) as
87 (0.35M) 30 eq. Dengue. rt 2

66% (NMR), d.r.=58/42
ee=42% (major)
ee=28% (minor)

Although the result is not up to the level of the reactions with nitrostyrene, the good
conversion and the enantioselectivity of the major diastereomer suggest that a further
optimization and effort to extend the reaction to other aliphatic nitroalkenes would be

worthy.

5.3.2.5 Proposed Stereochemichal Model for the asymmetric addition using thiourea

82

According to the generalized mechanism for the role of bifunctional catalysts in
the conjugate addition of nitroalkanes to nitroolefins that was illustrated earlier in the
chapter (Scheme 29) the key reactive intermediate is believed to involve the DMAP
moiety of the catalyst interacting with the deprotonated l-nitropropane as a cation/anion
pair, along with the nitrostyrene which is held through H-bonding interaction from the
thiourea group. A postulated mechanism that accounts for the observed stereochemical

outcome with thiourea 82 is depicted in Figure 25.

131

Figure 25. Postulated stereochemical model

NO observed /1/
2 enantiomer ,.- N02 (28,3S)-808
(2R, 3Fl)-80a ©

ébfin ”’“o- Pb in ”k”

N _
. H 0'0

I’ N+

W5 /N\ F30 3.0 ”*3
N / NQQ /\

/ ‘N + \ %r Q
A': favored B' _N~
(less strain)

as S liq F30
Q @530. Q S 33 “J

N o _
F3C H H H H N O'NO
I ' N + H H
, . N F3C H
O. .0 / \ O O Q N + N
N / N a / \
/ —N\ ‘T \ \\ ’
DI Cl _N‘

\ N02
N02
(9%“ The. <28. 33......
(ZR, 3S)-803 ©

The absolute stereochemistry of the major enantiomer was determined by a

comparison with the HPLC data and the optical rotation values of the corresponding

compounds from the literature (for the known compounds 80a—c, 80c). The proposed

stereochemical model is similar to the one for the BINAM derived bis-thiourea catalyst 52

that was developed for the asymmetric aza-Henry reaction and described in chapter 4. The

major enantiomer (2R, 3R)—80a is believed to be derived via the intermediate A’, where

132

both components of the reaction orient themselves in the chiral pocket in such a way as to
minimize the steric interaction with the (R)—BINAM moiety of the catalyst. The opposite
enantiomer is obtained through B’, which is very unfavored due to the fact that both of the
substrates experience unfavorable contacts with the binaphthyl core. The expected energy
difference should be high, and it seems like a reasonable explanation for the excellent ee
and the very efficient enantiodiscrimination. for the syn adduct. The same factors make
intermediates C’ and D’, which afford the anti isomers, energertically unfavored and
explain the diastereoselectivity in favor of the syn isomer, as well. In other words, the anti
adduct is disfavored because both intermediates C’ and D’ involve steric interactions that
are absent in A’. This model is in accordance with the higher diastereoselectivity that was
observed for l-nitropropane and l-nitrobutane in comparison with nitroethane, since it
would be expected that bulkier nitroalkanes would impose a higher energy cost for
intermediate D’. Considering the cc of the minor diastereomer, which is determined by the
energy difference between C’ and D’, it is smaller than the ee of the syn product and thus
makes sense since C’ and D’ each has one negative steric interaction, and thus the
dicrimination between these two diastereomeric transition states shouldn’t be as high as
that for A’ versus B’.

Again, as in the case of the aza-Henry bis-thiourea catalyst 52 (chapter 4), the
above model is a working hypothesis that explains the stereochemical outcome. Given the
conformational freedom that both the thiourea moiety and the DMAP group can
experience other slightly different models could be proposed. The above model just
represents a straight-forward representation of the reaction according to the assumptions

described earlier.

133

5.4 SUMMARY AND CONCLUSION

In conclusion, the thiourea catalyzed conjugate addition of nitroalkanes to
nitroolefines has been studied both in its non-asymmetric and enantioselective version.

Concerning the non-asymmetric Michael reaction, it was found that using 2 mol%
of Schreiner’s thiourea 47 along with 2 mol% triethylamine can lead to the syn-selective
synthesis of 1,3-dinitroalkanes in good yields.

The successful employment of thiourea 47 to the addition of nitroalkanes to
nitroolefins prompted the investigation of an asymmetric variant. As a result, the first
highly efficient protocol for an organocatalytic direct addition of nitroalkanes to
nitroolefines has been developed. The reaction is promoted by a novel bifunctional
DMAP/thiourea catalyst and affords enantiopure syn-1,3—dinitro compounds with only 2%
cat. loading. The asymmetric inductions in the products are remarkably consistent (92-
95% ee for the addition of l-nitropropane to nitrostyrene) and no drop-off is observed
even when ortho-substituted styrenes are used.

A unique selectivity trend was discovered for this reaction according to which, the
less catalyst used, the higher the enantioselectivity of the reaction, this effect was
correlated with the nitroalkane/catalyst ratio. It was demonstrated that an increase in this
ratio leads to increasd inductions, presumably because the interaction of the catalyst with
nitroalkane changes its structure in solution to a more selective species. This idea can be
formulated as a self-deactivated catalyst, and a series of experiments were done to probe
this.

As a result of this unusual selectivity trend, ee’s up to 96% could be obtained with

only 0.5% catalyst loading, demonstrating a catalyst that is remarkably selective.

134

HAPTER

TOWARDS VAULTED BIARYL-DERIEVED THIOUREAS AND THE

SYNTHESIS OF VANAM

It was demonstrated in chapters 4 and 5 that 1,1’-binaphthyl-2,2’-diamine is an
efficient chiral schaffold for thiourea mediated catalysis. The success of BINAM in
inducing high enantioselectivities in the asymmetric addition of nitroalkanes to imines
using catalyst 52 and to nitroolefines using catalyst 82 showed the potential of biaryl-
based thiourea catalysts in a field dominated by cyclohexane diamine and Cinchona
alkaloid-based organocatalysts.

These results along with the work that has been done in our laboratory with the
ligands VANOL and VAPOL (see chapter 1) prompted the consideration of catalysts 89-

90 that are based on the vaulted binaphthyl structural unit.

Figure 26. Vaulted biaryl thioureas and diamines

I <1

IZ 23:
12 ZI
0
TI
on

O CF3

   

Catalysts 89 and 90 feature a vaulted biaryl unit that would be expected to

communicate stereochemical information closer to the reaction center, that is the thiourea

135

and/or DMAP moieties, and so their synthesis and application in these reactions — or other
related transformations - would be highly desirable. The chiral diamine 91, which would
be the starting material for the above catalysts, is not known in the literature. Therefore,
the key issue in the synthesis of these novel catalysts is the synthesis of the diamine-
analog of VANOL. Some preliminary efforts towards the synthesis of diamine 91 have

been undertaken, and this chapter details the results of thess studies.

6.1 Examination of Buchwald amination as a key step in the synthesis of 91

Concerning the synthesis of arylamines, one of the most powerful protocols
reported in the literature involve the reactions of aryl triflates (or bromides and chlorides)
with primary or secondary amines under Pd catalysis, described by Buchwald and
coworkers“. For example, the reaction of triflate 92 with benzylamine using a catalyst
formed in situ from Pd(OAc)2 and ligand 94 affords the secondary arylamine 93 in 81%

yield (Scheme 34)”.

Scheme 34. Buchwald amination of triflate 95

1% Pd(OAc)2

0T1 NHBn 0
2°/ 16
O + BnNH2 o > O PU‘Bulz
t-Bu NaOtBu 1.3”
Toluene, rt 94
92 81% 93

 

It was reasoned that this kind of amination reaction could be used as a key step in

the synthesis of 91 according to the sequence illustrated in Scheme 35.

136

 

Scheme 35. Proposed synthesis of 91 using Buchwald amination as a key step

0 Buchwald O I
m'nat' -
Ph OTf a ' '0” Ph NHBn deprotectlon

........ - Ph NH2

--------- ’

95 96 91

A number of different amines could be coupled with VANOL triflate 95 — which
should be easily obtained from VANOL — with benzyl amine being a good example of an
amine that could be subsequently deprotected easily to give 91.

The investigation of this stratgy started with the synthesis of triflate 95. The
reaction of racemic VANOL with triflic anhydride in the presence of pyridine afforded a
very good yield of the biaryl triflate 95, both on a small scale as well as on a 1g scale

reaction (Scheme 36).

Scheme 36. Synthesis of VANOL triflate

o 0
ph OH Ph OTf

 

Ph OH DCM Ph 0T1

116 I 95 I

150 mg scale: 91%
19 scale: 80%

137

Table 41. Initial efforts for the synthesis of 96

NHan O

10%. Pd(OAc)2 O
on "98nd Ph NHBn
OTf NaOtBu Ph NHBn
toluene O

96 I

 

 

 

ligand conditions result
94 rt, triflate added all at once 0% 96, triflate hydrolyzed
BINAP 80°C, triflate added over 1h 0% 96, triflate hydrolyzed

 

The first attempts to couple 95 with benzylamine using the conditions illustrated in
Table 41 were disappointing. The reaction using ligand 94 at rt and adding the triflate all
at once led to the hydrolysis of a big portion of 95 and the formation of free VANOL

(Table 41).

The hydrolysis of triflates during Pd catalyzed amination reactions is a common
problem according to the reports of Buchwald and Hartwig". This hydrolysis is attributed
to the rapid cleavage of the triflate by attack of NaOtBu (or other strong bases) on sulfur
leading to the corresponding phenoxide. It is usually associated either with the electronic
nature of the triflate (electron deficient triflates are prone to hydrolysis) or the strength of
the base that is used. In the case of compound 95 there is no issue with the electronic
nature of the substrate, but still it seems that the substrate is not stable under the given

conditions. Hartwig has reported that slow addition of the triflate to the reaction mixture

138

can prevent the undesired hydrolysis“. Along these lines, an experiment was performed
where 95 was added over a 1h period, but with no success (Table 41).

A second way to avoid triflate hydrolysis is using a relatively weak base for the
amine deprotonation step. Buchwald has reported an improved method using weaker bases
such as CSZCO3 for the amination of halides and triflates that affords higher yields and
improved functional group compatibility”. The use of C52CO3 was tested in the reaction of
benzylamine with 95. The hydrolysis of 95 was completely prevented this time and no free
VANOL was obtained, but the reaction didn’t afford any product either. All the starting

material remained unreacted under the conditions shown in Scheme 37.

Scheme 37 . An attempt for Buchwald amination of 95 using CszCO3

0 5% Pd(OAc)2 U
5.5% BINAP
Ph on
+ _
on BnNH2 7 Ph NHBn

052003 Ph NHBn
Toluene O
0 100002411

95 0%, NO REACTION 96 O

 

At this point, it was suspected that the bulkiness of the VANOL triflate might be
responsible for the failure of these attempts. Since 95 remains unreacted under forcing
conditions even when hydrolysis is prevented, this probably means that the triflate is
probably too bulky for the oxidative addition step, particularly when the ligand that is used

for Pd also is considerably bulky.

139

To test this hypothesis it was reasoned that the coupling should be tried with the
triflate 100 obtained from the mononer precursor of VANOL, 99. The result from the
coupling of 100 should give information about weather the bulkiness of VANOL triflate is
the reason for the failure of its coupling.

To test whether the bulkiness of VANOL is responsible for its unreactivity in the
Pd catalyzed coupling, triflate 100 was made in 80% yield from 99 and it was subjected in
coupling reactions with benzylamine using different bases. The results are shown in

Scheme 38.

Scheme 38. Synthesis and amination of triflate 100

O OH "20 W G OTf
O D
O O
99 1 00
5%catalyst
OTf base NHBn
O + BnN H2 t O
O Toluene O
100 101

 

 

 

 

entrty catalyst base T(°C) %yield 101
l Pd(OAc)2/BIN AP NaOtBu 80 0%
2 Pd(OAc)2/94 K3PO4 80 0%
3 Pd(OAc)2/BINAP CSZCO3 100 87%
4 Pd(OAc)2/94 CSZCO3 100 28%

 

140

The use of NaOtBu afforded the same result as for the attempted coupling of 95. A
big amount of 100 was hydrolyzed and 99 was obtained as a major product from the
reaction in entry 1. The use of K3PO4 afforded no reaction and recovery of unreacted 100.
The use of CszCO3 though afforded the desired product 101 in excellent yield when
BINAP was used as a ligand (entry 3) and with a lower yield when 94 was employed
(Scheme 34). These two last results in entries 3 and 4 confirm that the triflate 95 is
presumably too bulky for the Pd catalyzed coupling under the above conditions. The fact
that the monomer 100 can afford an excellent yield whereas the VANOL triflate gives no
reaction implies that the formation of the intermediate of the oxidative addition that
features Pd coordinated both by BINAP and VANOL triflate is probably disfavored due to

steric congestion.

6.2. Examination of a synthesis of 91 through oxidative coupling of arylamines

A second pathway that was examined for the synthesis of 91 involves the use of
oxidative coupling between substituted anilines.

The synthesis of 91 through oxidative diaryl coupling in fact has been attempted
before in the literature by Kocovsky’s group, but without success”. When Kocovsky
employed the free aniline 102 in an oxidative coupling using CuCl2 he obtained carbazole
103 as the only product in 25% yield. The driving force for the formation of compound
103 is the loss of ammonia. Kocovsky proposed that the loss of ammonia is not occurring
through the formation of 91 as an intermediate, but instead, via a competing pathway that

involves N-N coupling of 102 and a subsequent 3.3-rearrangement.

141

Scheme 39. Oxidative coupling of 102 using CuCl2

 

NFI

[Illa EEBG

ph l NH2 CUClg. 2H20 / BnNH2 Ph

0 MeOH IAnisole P“
rt

102 103

Specifically, through the oxidative coupling of a number of substituted aniline
derivatives and control experiments, he postulated the mechanism shown in Scheme 40 to

account for the formation of the resulting diamines and carbazoles.

Scheme 40. Postulated mechanism for the CuCl2 mediated oxidative coupling of

substituted anilines
so“ coco

104

\ 105

Cu(ll)

NH2 --/---+

Ii H
sllli] lVHg /lllii

107

142

According to this mechanism, naphthylamines like 104 can undergo a Cu(II)
mediated oxidative coupling either affording directly 5] (C-C coupling) or 105 through
coupling at nitrogen. Intermediate 105 can undergo a 3,3-sigmatropic rearrangement to
give 106, which subsequently can tautomerize to 51 or alternatively lead to the carbazole
precursor 107, that can release ammonia to afford 108. For 2-naphthylamine 104 the major
product was the diamine 51 while only a trace of carbazole was detected (~l%). For the
coupling of other aniline derivatives though the percentage of carbazole formation was
much higher and in the case of the coupling of 102, as mentioned before, only carbazole
was obtained. This mechanism was supported by the fact that the Bucherer-type reaction
of 2-naphthol with hydrazine that is expected to proceed through intermediate 105, was
found to readily afford 5159. Additionally, when the isolated diamines like 51 were
subjected to the condition of the oxidative coupling no carbazole was obtained, which
implies that the carbazole byproducts are not produced by loss of ammonia from the
diamine, but instead, via the alternative N-N coupling pathway. The same control
experiment was not performed with VANAM since no diamine was produced or isolated
from the coupling of 102, but the authors presume that its behavior should be similar to
that of 10458.

Concerning the synthesis of VANAM, it was reasoned that if compound 102 was
substituted at nitrogen then the formation of the carbazole byproduct should be disfavored
since the ammonia loss would not be an issue anymore. This way, even in the case of N-N
coupling the 3,3-rearrangement would hopefully result in the preferential formation of the
desired diamine. This synthetic route using the oxidative coupling as a key step would add

one more key step to the sequence, which is the resolution of the racemic diamine 96.

143

Along these lines, substrate 101 was tested in the oxidative coupling using CuCl2
and following the conditions reported by Kocovsky for the corresponding reaction with
102. The results, shown in Scheme 41, were rather disappointing. Koockovsky’s
conditions (r.t.) afforded no reaction and complete recovery of the starting material. An

attempt to push the reaction by heating at 100°C also did not produce a reaction.

ph NHBn CuClz, 2H20 I BnNH2 O

Scheme 41. Attempted oxidative coupling of 101 using CuCl2

Ph NHBn
MeOH IAnisole
101 96 !

 

 

temperature (°C) result
25 no reaction
100 no reaction

 

It was suspected that the failure of the above reaction might be a result of the
bulkiness of the substrate 101. The position of coupling is ortho, ortho- disubstituted by a
phenyl group and a benzyl group, and that might impose a considerable energy barrier for
the coupling of the two units. In this case, it was envisioned that the solution may be the

employment of substrates that bear smaller N-substituents.

144

Scheme 42. Synthesis and attempted oxidative coupling of 111

 

 

 

Pd(OAc) H

Ph on BINAP 2 Ph NA

CI + WNHQ CI
C52CO3
0 Toluene O
100 109 100°C 111
entry scale time % catalyst equiv. 109 % yield

1 300 mg 12h 5 1.2 14

2 300 mg 2d 10 2.4 60

3 1 g 2d 10 2.4 57

 

Ph A
O CUClz_ 2H20 / BHNHZ Ph

D ..
N
O MeOH/Anisole,rt V Ph 0 ”A

0%, No REACTION O
111 112

 

The allyl-protected substrate 111 was the compound of choice for testing this

hypothesis. Compound 111 was synthesized from a Pd catalyzed coupling of 100 with

allylamine 109. This amination was a little more problematic than the one with

benzylamine, since allylamine is a volatile compound and easily evaporates from the

reaction mixture when using high temperatures, thus affording low yields (entry 1). This

was solved when a larger excess of 102 was employed and higher catalyst loading was

used to increase the rate. In this way a good yield of the desired product was obtained

even when the reaction was scaled up to one gram of material.

145

It was disappointing to see that 111 did not give any product when its oxidative
coupling was attempted (Scheme 42). Similar to 101, the attempted reaction with 111
afforded a clean reaction mixture where surprisingly absolutely nothing had happened.

The failure of the CuCl2 oxidation method turned our attention to alternative, and
more forcing conditions for the oxidative coupling reaction. In specific, oxidative coupling
under neat conditions using high temperature and air was considered. This has been a
standard method for the synthesis of the chiral ligand VAPOL in our laboratories’, and we
thought it might turn out to be successful for the arylamine coupling as well. The coupling
under neat conditions was tried both for 101 and 111. The solids were put in a flask and
subjected to high temperature for 3.5 hours under the flow of air. The results are shown in
Scheme 43. Perhaps not surprisingly, the sensitive terminal double bond in 111 seems to
be responsible for the considerable amount of baseline material observed in the oxidation

of 111. Concerning 101, the only product observed after 3.5 h of heating was

Scheme 43. Attempted oxidation under neat conditions

H
P“ N\/\ neat, air 0 H
O \ Ph NV + basel'ne
1
18000 P" ”A material
3.5h O
1“ 0 112.0%

Ph NHBn . O
0 "eat” 8‘" Ph NHBn P“ vah
180°C Ph NHBn +
3.5 h 0 O
101 I

 

 

96. 0% 113, 32%

146

the imine 113 resulting from the oxidation of the benzyl moiety. These were the last

attempts concerning this particular approach.

6.3. Attempts to synthesize 91 using the Bucherer reaction

The last approach that was considered for the synthesis of 91 was the use of the
Bucherer reaction, that is, the direct displacement in one step of the hydroxy groups with
ammonia under forcing conditions”. The Bucherer reaction has been reported to transform
2—naphthol 110 and the VANOL monomer 99 to the arylamines 104 and 102

respectively”, via the mechanism shown in Scheme 44.
Scheme 44. The Bucherer reaction for the synthesis of 104 and 102
CH (NH4)2SO3.H20‘ NH2
””3 / H20
110 1500,1011 104
Ph 0H
0 P“ 0
NH3 / H20
200°C, 2d

 

 

 

99 42% 102
CH NaHSO3 O NH3 NH
= —— 0.
H20
11° soaNa 303Na

 

Z
I
M
l2
9)
I
U)
0
co
2
I
m

104 SOaNa

147

Despite the usefulness of the name reaction in the synthesis of simple naphthyl
amines, there is no precedent for the synthesis of 1,l’-biaryl-2,2’-diamines using the
above method. In fact, BINOL (120) has been reported to afford the monoaminated
product 121 when subjected to the reaction conditions, but no BINAM was produced“.
The difference in the reactivity of BINOL compared to its monomer 110 has been
explained on the basis of an intramolecular H-bonding that stabilizes the monoaminated
product 121 and “shuts down” any further reactivity, stopping the reaction and preventing

the second hydroxyl group from getting displaced by ammonia (Scheme 45).

Scheme 45. Bucherer reaction of BINOL

I I OH (NH412503.H2C3 I I OH
0” NH3 / H20 ”“2
200°, 5d

91 °/o
120 121

 

Despite the unfavorable precedent with BINOL it was decided that an effort to test
the VANOL (and VAPOL) ligand in the Bucherer conditions would be worthwile.
VANOL and VAPOL thus were put in an aqueous solution of ammonia and heated at high

temperatures in the presence of NaHSO3. The results are shown in Table 42.

148

Table 42. Bucherer reactions of VANOL and VAPOL

 

   

O \ ’3') I”) ’0
O aq. NH3
Ph OH NaH303
Ph OH +
O , 11" sq i,
\ ’3’) 3’). ,0.
116, VANOL 91
1, VAPOL 117

 

%yield %yield
entry ligand scale T(°C) time (h)
91 or 117 118 or 119

 

1 116 217mg 200 60 0 so
2 116 2g 200 40 0 25
3 1 270mg 220 24 0 0

 

As can be seen, the reaction of VANOL 116 on a small scale (entry 1) produced
compound 118 in 50% yield with no trace of the desired 91. When the reaction was
performed on a 2g scale though, the isolated yield dropped to 25% despite the almost
complete conversion of the starting material. Finally, for the VAPOL ligand (entry 3) no
reaction occurred at 220°C. Not surprisingly, the reactivity of VANOL resembles that of
BINOL and the suspected intramolecular H-bond controls the reaction outcome here as

well.

149

6.4. Future Work

In a recent report by Cho and coworkers it was shown that diaryl hydrazides 125-
127 prepared from the Pd(0) catalyzed coupling reactions between aryl bromides 122-124
and di—tert-butyl hydrazidoformate (BocNHNHBoc) can undergo acid-catalyzed

rearrangement to form 2,2’-diamino-l ,1’—binaphthyls, according to the sequence shown in

Scheme 46”.

Scheme 46. Cho’s synthesis of BINAM derivatives

BocNHNHBoc cat. HCl CO
Pd2(dba)3’xa"p"°: NBoc EtOH/80°C R NH2

 

122 (R=H) 125, 81% 51, 75°/
123 R=Br 125, 37% 128, 313/o
124 (R=Me) 127, 65% 129, 81%

This type of rearrangement has been known for a long time to proceed either
thermally or with acid catalysis for 105, giving BINAM as a major product. For example,
heating of 105 in boiling EtOH affords an 82% yield of 51 and a small amount (~15%) of
the carbazole 108 (Scheme 47)”. It seems that EtOH promotes the rearrangement by

acting as a general acid catalyst.

150

Scheme 47 . Thermal rearrangement of 105 for the synthesis of BINAM

”NH EtOH/80°C CO NH2

#

“.3

105

+ NH

108 (15%)

00

51 (82%)

It was envisioned that this synthetic sequence could be potentially used for the

synthesis of VAN AM 91. An N-protected diaryl hydrazide 131 could hopefully rearrange

to afford the desired product 91 without significant production of carbazole. If Kocovsky’s

proposed mechanism for the diamine versus carbazole formation is correct and the

carbazole is produced through ammonia loss from the 3,3-rearrangement pathway

intermediate 106, then one would expect that in case of 131 the ammonia loss should be

prevented if the rearrangement occurred before the deprotection of nitrogen. The

mechanism for the reaction in Scheme 47 is not clear in every detail, but it could be

argued that

Scheme 48. Proposed synthesis of 91 through hydrazide rearrangement

2
Ph

130

Br

BocNHNHBoc
Pd2(dba)3/xanphos

 

08003

Ph I NBoc EtOH/80°C Ph

Ph 0 NBoc 1h Ph i
O 9, O

E NH2

NH2

131

151

avoiding the use of HCl in the rearrangement of 131 and performing the reaction under
milder conditions (thermally, using only EtOH) might allow for the reaction to occur with
the Boc groups still on, thus eliminating the issue of ammonia loss.

Compound 130 could be obtained from the triflate 100 through the intermediacy of
a boronate ester 132 according to the method depicted in Scheme 49. This chemistry has
been developed by Huffman’s group and is known to afford very good yields of aryl
halides“. For example, 2-naphthyl and l-naphthyl bromides can be obtained in 77% and
84% overall yields from the starting triflate respectively. The reaction for the particular

triflate 100 is not known, but it is expected that it will hopefully show similar behavior.

Scheme 49 Proposed synthesis of 130

Ph OTf Ph 3' Ph Br
0 . 0 ‘° O

100 132 130

i. pinacolborane, PdClz, dppl, dioxane, NEt3; ii. CuBrz, MeOH, H20
In addition, the direct coupling of triflate 100 with BocNHNHBoc could be
examined as an alternative way of accesing 131 in a fewer number of steps. Cho’s method
only concerns naphthyl bromides, but an extension of the methodology to include aryl
triflates might be feasible. In this case the synthesis of 91 would be shorter and thus more

attractive .

152

EXPERIMENTAL PR ED RES

Experimental procedures for Chapter 2

6-Methylnapthoic acid 22

0 COOH
AIC|3 Me 00

21 ~ 5 % 22

 

2-Furoic acid (120 g, 1071 mmol, 1.0 equiv) was added portionwise to a 3-necked
flask containing toluene 21 (1000 mL), which was immersed in an ice bath and fitted with
a mechanical stirrer. A bubbler was attached before AlCl3 (300g, 2249 mmol, 2.1 equiv)
was added in spatula portions over a 2 h period, maintaining a steady evolution of gas.
The flask was removed from the ice and slowly heated to 60 °C and then stirred at this
temperature for 18 h. The solution was then poured into an ice-cold HCl solution (550 ml
of cone HCl in 2 litres of H20). This was then heated overnight (18 h) at 60 °C, to dissolve
any solids. The two resulting layers were separated, and the organic layer was washed
with H20 (2 x 300 ml) before being heated with a 1.0 M NaHCO3 solution (1000 mL) at
55 0C for 1 h. This was repeated once more and the combined aqueous layers (2000 mL)
were heated to 55 °C before Ba(OH)2.8 H20 (315.0 g, 1000 mmol) was added and the
resulting solution stirred for 2 h. The yellow precipitous solution was filtered and the
red/brown aqueous layer collected and split into 3 x 1 litre fractions. Using pH paper these

solutions were titrated with cone HCl to a pH of 7.0. The resultant yellow precipitate was

153

filtered and the green aqueous filtrate was further titrated with conc. HCl to a pH of 5.0-
6.0. A paler yellow precipitate was filtered and collected, TLC analysis indicated that it
was the desired product. The purity of the solid varied, with a solid orange contaminant
being present. The material was crystallized to purity from either hot benzene or EtOAc to
deliver bright yellow crystals of 6-methylnapthoic acid, (16.7 g, 85.6 mmol, 8%). 1H NMR
(300 MHz, MeOD) 6 2.39 (s, 3H), 7.30-7.38 (m, 2H), 7.58 (s, 1H), 7.85 (d, 1H, J = 8.5

Hz), 8.00 (dd, 1H, J: 7.0, 1.2 Hz), 8.68 (d, 1H,J= 9.0 Hz).

Diisopropyl-6-methynapthamide 24

 

 

O OH 0 CI 0 N(iPl’)2
socrz (iPr)2NH, Et3N _
Me 00 Benzene, DMF 00 CH2012. 0°C 00
Me 74 % from 20 Me
22 23 24

To a flask containing 6-methylnapthoic acid 22 (8.27 g, 44.46 mmol, 1.0 equiv)
under N2 was added benzene (18.5 mL, 2.4 M) and the resulting slurry stirred for 5 min.
SOCl2 (3 .5 mL, 48.11 mmol, 1.1 equiv) was added followed by the dr0pwise addition of
DMF (1.5 mL, 19.36 mmol, 0.44 equiv). The resulting clear red solution was allowed to
stir for 4 h, after which time the solution was concentrated on the rotary evaporator,
removing any excess SOClz. The benzene/DMF solution was taken on into the next step.

The above reaction was assumed to have progressed in quantitative yield.

154

CH2C12 (10 mL) was added to a flask containing the acid chloride under N2. This
solution was stirred for 5 min before being added via canula to a flask containing a stirring
solution of (i-Pr)2NH (7.48 mL, 53.38 mmol, 1.2 equiv) and NEt3 (7.44 mL, 53.38 mmol,
1.2 equiv) in CHZCI2 (70 mL, 0.5M in 23) under N2. After stirring overnight, H20 (50 mL)
was added to the reaction, the layers separated and the aqueous layer extracted with
CHzCl2 (2* 100 mL), and the combined organic fractions washed with 2M HCl (aq) (2*100
mL), sat NaHCO3 (aq) (2*100 mL) and dried with NaZSO4. Concentration under reduced
pressure and column chromatography (10% EtOAc2hexane) gave diisopropyl-6-
methynapthamide (32.9 mmol, 8.85 g, 74 % for the two steps) as a white solid; 1H NMR
(300 MHz, CDC],) 6 1.01 (d, 3H, J = 6.5 Hz), 1.05 (d, 3H, J = 6.5 Hz), 1.63 (d, 3H, J =
7.0 Hz), 1.69 (d, 3H, J = 7.0 Hz), 2.48 (s, 3H), 3.62-3.54 (m, 2H), 7.21-7.42 (m, 3 H), 7.60
(s, 1 H), 7.69-7.73 (m, 2 H); 13C NMR (300 MHz, CDCls) 5 20.6, 20.6, 20.8, 21.6, 45.9,
51.0, 121.2, 124.7, 125.3, 127.1, 127.4, 127.8, 128.9, 133.7, 136.0, 136.4, 170.0, 1 sp3 C
not located; IR (neat) 3053m, 2976m, 16225, 1332m, 12655, 710$, 704$ cm"; mass
spectrum m/z (% rel. intensity) 269 [M‘] (16), 169 (100), 141.0 (43), 115 (l 1), 88 (17), 84
(16), calcd for m/z 269.1780, meas 269.1777. Anal calcd for C18H230N: C, 80.26; H, 8.61;
N, 5.20. Found: C, 80.02; H, 8.73; N, 5.18; White solid (EtOAc) mp = 140-142°C;

R,=0.19 (10% EtOAczhexane).

155

Diisopropyl 6-methyl-2-(2-phenylallyl)naphthalen-l-amide 25.

o N(iPr)2

 

1) s-BuLi O N(iPr)2
2) Mg/C2H4Br2 O
00 3) Bromo Methyl Styrene
~75 °/ 0‘
Me 0 Me
24 25

s-BuLi (6.43 mL, 8.99 mmol, 1.1 equiv) was added dropwise over 1 min to a
solution of diisopropyl-6-methylnapthamide (2.2 g, 8.17 mmol, 1.0 equiv) in THF (80 mL,
0.08 M in 24) under N2 at —78 °C. The resulting orange solution was stirred for l h before
the canular addition of the ethereal MgBr2 solution, which was prepared in situ as
described below.

Dibromoethane (2.1 mL, 24.51 mmol, 3.0 equiv) was added over a 1 h period to a
flask, fitted with a condenser, containing magnesium tumings under N2. The resulting
reaction was stirred for 2 h before being added via canular to a solution containing the
ortho-lithiated napthamide. The colour of the solution changed from orange to pale green
and it was then stirred for a further 15 min before being allowed to warm to rt over a l h
period. The solution was then cooled to —78 °C, followed by the dr0pwise addition of or-
bromomethyl styrene and then allowed to warm to room temperature overnight. The
reaction was quenched with sat NH4C1 (100 mL), the layers separated and the aqueous
fraction extracted with EtOAc (2*100 mL), washed with brine (2*50 mL) and dried with
NaZSO4. Concentration under reduced pressure and column chromatography (10%
EtOAczhexane) gave the desired diisopropyl 6-methy1-2-(2-phenylallyl )naphthalene-l -

amide (2.35 g, 6.12 mmol, 75%) as a pale yellow foam. 'H NMR (300 MHz, CDCl,) 6

156

1.03 (d, 3H, J = 6.5 Hz), 1.11 (d, 3H, J = 6.5 Hz), 1.69 (d, 3H, J = 6.5 Hz), 1.81 (d, 3H, J
= 6.5 Hz), 2.51 (s, 3H), 3.54—3.72 (m, 2H), 3.88 (d, 1H,J= 16.5 Hz), 4.12 (d, 1H, J: 16.5
Hz), 5.07 (d, 1H, J = 1.5 Hz), 5.64 (s, 1H), 7.26-7.75 (m, 10H); 13C NMR (300 MHz,
CDC13) 5 20.5, 20.6, 21.0, 21.3, 21.5, 38.0, 46.1, 50.9, 115.8, 124.7, 126.0, 126.9, 127.0,
127.3, 127.5, 128.0, 128.3, 128.6, 131.1, 132.3, 134.7, 135.3. 140.6. 145.9, 169.4; IR
(CDC13) 3055, 2986, 2936, 1622, 1265 cm" ; mass spectrum m/z (% rel. intensity) 385
[M’] (10), 285 (25), 269 (30), 249 (30), 226 (15), 169 (70), 141 (30), 88 (100), 71 (75), 57

(65), 51 (100), calcd for m/z 385.2406, meas 385.2416.R,=0.45 (20% EtOAc:hexane).

7-Methyl-phenylphenanthren-4-ol 26

o N(iPr)2 ”0 Ph
0 MeLi. THF
0Q 00

Me so % Me
25 25

 

MeLi (5.52 mL, 5.52 mmol, 2.2 equiv) was added dropwise over a 5 min period, to
a solution of diisopropyl 6—methyl-2-(2-phenylallyl)naphthalen-l-amide (0.966 g, 2.509
mmol, 1.0 equiv) in THF (50 mL, 0.05 M) at -—78 °C, followed by stirring for l h before
allowing the reaction to warm to room temperature overnight. The reaction was quenched
with sat NH4Cl (60 ml), the layers separated and the aqueous layer extracted with EtOAc
(3 x 60 mL). The combined organic fractions were washed with brine (2 x 30 mL) and
dried with NaQSO4. Concentration under reduced pressure and column chromatography

(10% EtOAc:hexane), resulted in isolation of the desired 3-phenyl-8-methylphenanthrol

157

as a brown solid (0.602 g, 2.00 mmol, 80%). 1H NMR (300 MHz, CDCl,) 6 7.23 (d, 1H, J
= 1.5 Hz), 7.28 (s, 1H), 7.42 (d, 1H, J = 7.5 Hz), 7.51 (t, 3H, J = 7.5 Hz), 7.73 (t, 6H, J =
10 Hz), 9.53 (d, 1H, J: 8.5 Hz); l3C NMR (300 MHz, CDCl,) 6 21.3, 112.0, 118.6, 119.7,
127.1, 127.2, 127.4, 127.8, 128.1, 128.2, 128.3, 128.8, 132.7, 134.9, 135.6, 138.7, 140.2,
154.4, 1 sp2 C not located; IR (CDC13) 3566, 3055, 2988, 1205, 710, 706 cm"; mass
spectrum m/z (% rel intensity) 284 M+ (100), 241 (10), 184 (50), 141 (60), 115 (20), 57
(30), calcd for m/z 284.1201, meas 284.1197. Brown solid (hexane), mp = 146-147°C,

R,=0.36 (20% EtOAc:hexane).

7 ,7 ’-Dimethyl-2,2’-diphenyl-(3,3’-biphenanthrene)-4,4’-diol 11

Ho Ph 0
0 180-190°C _ H0 Ph
DO air HO 0 Ph

26 00 11

 

A round bottom flask containing 3-phenyl-8-methylphenanthrol (0.49 g, 1.33
mmol) was fitted with condenser. The flasks were then heated in an oil bath to 185 °C for
42 h. At 20 h interval, the black solid that was forming was dissolved with EtOAc, to
ensure good mixing, and the polymer was filtered through celite, before the mixture was
heated again. The reaction was monitored by TLC and lH-NMR. Work-up of the reaction
involved dilution with EtOAc, filtration of insoluble polymeric material and column

chromatography (4% EtOAc:pentane) leading to the isolation of name ( 164 mg, 0.438

158

mmol, 33.5%) as a slightly impure orange solid. Some starting material and a third
compound was recovered - not characterized. 'H NMR (300 MHz, CDC13) 6 2.58 (s, 6H),
6.65-7.64 (overlapping m, 22H), 9.63 (d, 2H, J = 8.5 Hz); ”C NMR (300 MHz, CDC13) 6
21.3, 115.7, 118.1, 123.1, 126.7, 127.0, 127.5,12803, 128.08, 128.68, 128.70, 128.8,
128.9, 133.0, 134.9, 136.1, 139.8, 141.1, 153.2; IR (neat) 3484w, 3026w, 2916w; mass
spectrum m/z (% rel intensity) 566 M” (100), 384 (13), 283 (83), 244 (28), 154 (18), 105
(12), calcd for m/z 566.2246, meas 566.2245.

NOTE: The reaction using air flow is slower and generates more byproducts

(S)-7 ,7 ’-Dimethyl-2,2’-diphenyl-(3,3’-biphenanthrene)-4,4’-diol 11

O CuCl, (-)-Sparteine O

HO Ph ultrasound HO Ph

“0 P“ MeOH/DCM HO Ph
0 o
(

‘

11 S)-11

(Typical procedure for deracemization of 7,7’-dimethyl VAPOL)

To a 50 mL round bottom flask was added 12 mL of MeOH, 64.7 g of CuCl (0.65
mmol) and 0.30 g of (-)-sparteine (1.30 mmol). The flask was sonicated in an ultrasonic
bath with air sparged into the solution. The temperature was maintained at or below 25 0C
by the addition of ice to the water in the bath. The flask was sealed with a septum after 45

minutes and then purged with argon for 60 minutes in the ultrasonic bath. The green

159

Cu(II)-sparteine solution was transferred via cannula to the solution of racemic 7,7’-
dimethyl VAPOL (0.22 g, 0.38 mmol) in 50 ml of CH2C12 in a 250 m1 round bottom flask
which was had previously been purged with argon and sonciated for 60 minutes. The
combined solution was sonicated for 10 minutes and then the flask was removed from
ultrasonic bath, covered with aluminum foil and stirred at room temperature for 6 hours
under a positive pressure of argon maintained by slow flow through the head space of the
flask. The reaction was quenched by adding 5 mL of concentrated HCl and then stirring
for 10 minutes. The two layers were separated. The lower CHZCl2 layer was collected, the
upper MeOH/1120 layer was washed with 3*30 mL of CHZClz. The combined CHzCl2 layer
was dried over MgSO4. Upon removal of solvent, the crude reaction mixture was loaded to
silica gel column and eluted with a 1:1 mixture of CHzClzl hexane to give 0.17 g (0.30
mmol, 79% yield) 26 as a white solid. The optical purity of 26 was found to be of 2 99 %
by HPLC on a Pirkle D—Phenylglycine column with a 75:25 mixture of
hexane/isopropanol (260 nm, flow rate: 2 mL/min). Under these conditions the retention

time of (R)-25 was 14.8 min and that of (S)-25 was 19.8 min.

160

A typical experimental procedure for the reaction of aldimine with silyl ketene
acetals.

To VAPOL (0.063 g, 0.11 mmol) and Zr(O-i-Pr),.iPrOH (0.05 mmol) in toluene
(0.9 mL) was added l-methylimidazole (0.06 mmol) in toluene (0.1 mL) at room
temperature. The mixture was stirred for l h at the same temperature. An aldimine (0.25
mmol in 1 mL) toluene solution was added to the catalyst and the mixture was stirred for
an additional 5 min. Then the silyl ketene acetal (0.3 mmol) was added. The mixture was
stirred for 15 h at room temperature. Aqueous N aHCO3 was added to quench the reaction.
The mixture was extracted with CH2C12 (2 x 10 mL). The combined organic layer was
concentrated to give the crude product. The crude product was treated with THF and
aqueous 1N HCl (10:1) at 0°C for 30 min. Then the reaction was quenched with aqueous
NaHCO3 and extracted with ethyl acetate (2 x 10 mL). The combined organic layer was
washed with brine and concentrated. The desired product was obtained by

chromatography on silica gel. The optical purity was determined by chiral HPLC analysis.

Methyl 3-(2-hydroxy-phenylamino)-3-phenyl-2,2-dimethyl-propionate 4.

OH
PhAN + >_<0Me —’ NH 0
OH

Ph OMe
2 3 4
The spectral data for this compound are the same as those previously reported for
this compound (ref. 8). 'H NMR (300 MHz, CDCl,): 6 = 1.20 (s, 3H), 1.23 (s, 3H), 3.67
(s, 3H), 4.53 (brs, 1H), 4.85 (brs, 1H), 5.29 (s, 3H), 6.37 (dd, J = 7.8, 1.1 Hz, 1H), 6.52

(dt, J = 7.0, 1.2 Hz, 1H), 6.59 (dt, J = 7.8, 1.1 Hz, 1H), 6.70 (d, J = 7.6 Hz, 1H), 7.20 -
161

7.27 (m, 5H). The optical purity was determined by chiral HPLC analysis (Daicel
Chiralpak AD, Hexanes/i-PrOH = 9/1, flow rate 1.0 mL/min): R, = 9.56 min (major

enantiomer), R, = 12.71 min (minor enantiomer).

tert-Butyl 3-(2-hydroxy-3,5-dimethylphenylamino)-3-phenyl-propionthioate 27a.
31"“

NH O

* StBu

27a

White solid; Rf = 0.32 (9/1 hexanes/ethyl acetate); mp 90 — 92°C; lH NMR (300

MHz, CDC13): 6 = 1.41 (s, 9H), 2.05 (s, 3H), 2.17 (s, 3H), 2.85 (dd, J = 4.9, 14.8 Hz, 1H),
2.96 (dd, J = 8.8, 15.1 Hz, 1H), 4.17 (s, 1H), 4.68 (dd, J = 4.9, 8.8 Hz, 1H), 5.36 (s, 1H),
6.17 (s, 1H), 6.39 (s, 1H), 7.27—7.30 (m, 5H); 13C NMR (75 MHz, CDCl,): 6 = 15.8, 20.9,
29.6, 48.5, 51.7, 56.7, 115.0, 122.2, 122.7, 126.3,127.2,128.5,129.3,134.2,l41.8,141.9,
198.6; MS (EI) m/z (relative intensity): 327 (9) [M’], 227 (18), 226 (100), 148 (8), 136(5),
131(2), 91 (12), 77 (5); Anal. Calcd. for C20H25NO3: C, 73.37; H, 7.69; N, 4.28. Found:
C, 71.19; H, 7.63; N, 4.09 (failed); HPLC (Daicel Chiralpak AD, hexanes/i-Pr= 9/1, flow

rate = 1.0 mL/min): Rt = 11.7 min (minor enantiomer), R1 = 15.5 min (major enantiomer),

(5)-VAPOL used as ligand.

162

tert-Butyl-3-(2-hydroxy-3,5-dimethylphenylamino)-3-(1-naphthyl)-propionthioate

fl...
NH O
O * S‘Bu
Q

White solid; R, = 0.31 (8/2 hexanes/ethyl acetate); mp 122 — 124°C; 'H NMR (300

27b.

MHz, CDCl,): 6 = 1.45 (s, 9H), 2.00 (s, 3H), 2.21 (s, 3H), 2.95 (dd, J = 9.3, 14.8 Hz, 1H),
3.12 (dd, J = 4.1 , 14.8 Hz, 1H), 4.74 (br, 1H), 5.09 (br, 1H), 5.64 (dd,J = 4.1, 9.3 Hz,1H),
6.13 (s, 1H), 6.35 (s, 1H), 7.38 — 7.65 (m, 4H), 7.76 (d, J = 8.2 Hz, 1H), 7.90 (dd,J = 1.4,
8.2 Hz, 1H), 8.23 (d, J = 8.4 Hz, 1H); 13C NMR (300 MHz, CDC13): 6 = 15.8, 20.9, 29.7,
48.6, 51.2, 52.3, 113.3, 121.1, 122.3, 122.4, 123.1, 125.4, 125.5, 126.3, 127.8, 129.0.
129.8, 130.4, 133.9, 134.8, 137.1, 140.7, 198.4; HPLC (Daicel Chiralpak AD, hexanes/i-
PrOH = 9/1, flow rate = 1.0 mL/min): R, = 12.9 min (minor enantiomer), R, = 16.5 min

(major enantiomer), (S)-VAPOL used as ligand.

1.63

tert-Butyl 3-(2-hydroxy-3,5-dimethylphenylamino)-3-(4-chlorophenyl)-propionthioate

fl...
NH O

* StBu

14c.

(31 27c

Yellow oil; Rf = 0.09 (9/1 hexanes/ethyl acetate); 1H NMR (300 MHz, CDC13): 6 =
1.48 (s, 9H), 2.13 (s, 3H), 2.24 (s, 3H), 2.89 (dd,J = 5.4, 15.0 Hz, 1H), 2.99 (dd, J = 8.4,
15.0 Hz, 1H), 4.75 (m, 1H), 6.19 (s, 1H), 6.45 (s, 1H), 7.30 — 7.34 (m, 6H); l3C NMR (300
MHz, CDC13) 15.7, 20.8, 29.6, 29.7, 48.7, 51.6, 55.9, 114.6, 122.2, 122.7, 127.8, 128.7,
129.6, 132.9, 134.1, 140.6, 141.6,198.3 ppm; IR (neat): 3410, 2964, 2922, 1672, 1601,
1516, 1491, 1456, 1385, 1199, 1151, 1091, 1030, 1014, 985, 827, 783 cm"; LRMS
(FAB’) m/z (relative intensity): 393 (33, 37Cl) [NF], 39] (76, 35Cl) [M’], 260 (20), 197
(20), 135 (60), 109 (24), 95 (40), 83 (40), 69 (60), 57 (100); HRMS (FAB+) Calcd for
C2,H26NOZSC14391.1373, found 391.1375; HPLC (Daicel Chiralpak AD, hexanes/i-PrOH
= 9/1, flow rate = 1.0 mL/min): R, = 17.3 min (minor enantiomer), R, = 26.2 min (major

enantiomer), (S -VAPOL as ligand).

164

Experimental Procedures for Chapter 4

General procedure for the preparation of the catalysts 52, 55.
3,5-Bis-trifluoromethylphenyl isothiocyanate (8.12 mmol, 1.48 mL) was added to
a solution of the appropriate diamine (4.06 mmol) in 7 mL THF at rt and the mixture was

stirred overnight. The volatiles were evaporated and the crude mixtures were purified by

 

crystallization.
Catalyst 52
043
1152311310
OW”? .N 001::
THF, 0°C to rt
96%
(FD-51 (Fl)- 52 CF3

(R)-BINAM was used (1.26 g, 4.00 mmol). Purified by crystallization from
dichloromethane/hexanes to afford 3.22g (3.89 mmol, 96%) isolated yield. White crystals,
m.p. = 120-122°C. 1H NMR (CDC13, 300 MHz): 6 = 7.09 (d, J = 8.7 Hz, 2H), 7.26 (t, J:
6.9 Hz, 2H), 7.48 (m, 4H), 7.54 (m, 2H), 7.67 (br s, 4H), 7.81 (d, J = 8.7 Hz, 2H), 7.95 (d,
J = 8.3 Hz, 2H), 8.10 (d, J = 8.7 Hz, 2H) ppm; 13C NMR (CDCl3, 300 MHz): 6 = 119.2,
122.8 (q,J= 272.9 Hz), 124.6, 125.3, 125.6, 126.7, 127.5, 127.6,128.5,129.7,131.7 (q,J
= 33.7 Hz), 132.3, 132.8, 134.4, 139.1, 180.4 ppm; IR (neat) v = 3265, 1688, 1498, 980
cm"; HRMS (ESI) calcd for C38H23N4FUSZ: 827.1179, found 827.1172; [alzsD = +103.8 (c

= 1.0, acetone).

165

Catalyst 55

 

2 eq. C48
F30 N” CF3
© 1 Q
_,.NH2 01:3 ‘ ..u u CF3
E l E l H
NH2 THF, 0°C to n “TN cps
(5'56“ 74% s
(S.S)-55

Synthesized according to the general procedure using 468 mg (4.10 mmol) (S,S)-
1,2-cyclohexyldiamine, purified by crystallization from EtOAc/hexanes to afford 1.99 g
white crystals (3.03 mmol, 74%). IH NMR (300 MHz, CDC13) 6 = 1.35 (br s, 4H), 1.81
(br s, 2H), 2.20 (br s, 2H), 4.38 (br s, 2H), 7.07 (br s, 2H), 7.69 (s, 2H), 7.81 (s, 4H), 8.12
(br s, 2H); 13C NMR (300 MHz, CDCl,) 6 = 24.4, 31.7, 59.4, 119.6, 122.7, 124.0, 132.9,

138.59, 180.54. The data agreed with the reported literature 3“"

Catalyst 53

 

 

S 00 0C 3 CO
H .-
(R) BINAM Cl/u\0| / CHCla N:(C:=: (9)-BIN AM “A“
- 3 .‘N= : \N N
choa I H20 00 550/0 . H \n/H
90% ICC S 0
(m-57 (Rum-53

Synthesized using 365 mg (1 mmol) (R)-2,2’-bis(isothiocyonato)-2,2’-binaphthyl
diamine” and 281 mg (1 mmole) (R)-l,l’-binaphthyl-2,2’-diamine to afford 351 mg
(55%) of white crystals (acetone/hexanes). lH-NMR (CDCl3, 300 MHz): 6 = 6.68 (d, J =
8.7 Hz, 2H), 6.82 (s, 4H), 6.87 (s, 2H), 7.02 (d, J = 8.4 Hz, 2H), 7.15 (t, J = 6.9 Hz, 2H),
7.40 (q, J = 6.9 Hz, 6.9Hz, 4H), 7.57 (s, 2H), 7.67 (t, J = 6.9 Hz, 2H), 7.76 (d, J = 8.1Hz,

166

2H), 7.90 (d,J = 8.1Hz, 2H), 8.02 (d, J = 9.3 Hz, 2H), 9.41 (d, J = 9.3 Hz, 2H) ppm; l3C-
NMR (CDC13, 300 MHz): 6 = 119.8, 120.8, 123.1, 125.4, 125.5, 125.9, 127.1, 127.2,
127.8, 128.2, 128.6, 128.8, 129.3, 130.2, 131.1, 132.2, 132.3, 132.5, 133.7, 135.7, 177.6
ppm. IR (KBr) 3154, 1595 cm"; MS (FAB’): m/z (%): 653 [M+1-I]+ (18), 619 (4), 460 (5),

307 (40), 252 (45) 154 (100), 136 (84), 102 (38); mp. 230-231°C.

Catalyst 54
\
WN
N=C =3 histamine fill
OxNzc :8 O
flN
H)-57 ( H)-54

Synthesized according to the general procedure using 0.5 g (1.3 mmol) (R)-2,2’-
bis(isothiocyonato)-2,2’-binaphthyl and 302 mg (2.71 mmol) histamine to afford 540 mg
(0.91 mmol, 100% from 57) of an off-white solid. lH-NMR (DMSO-d6, 300 MHz): 6 =
2.55 (br. s, 4H), 3.47 (br. s, 4H), 6.62 (br. s, 2H), 7.07-7.17 (m, 4H), 7.40 (m, 4H), 7.73
(m, 4H), 7.95 (t, J = 9.3, 4H), 8.61 (br. s, 2H), 11.71 (br. s, 2H); l3C-NMR (DMSO-d6, 300
MHz): 6 = 27.0, 44.5, 55.4,126.1,126.2,126.7,127.2,128.3, 128.7, 131.9, 133.1,135.1,
136.4, 181.3; two carbons are not located. MS (FAB’): m/z: 591 [M+H]” (17), 307 (14); IR
(neat) 3150, 1595, 1490 cm"; HRMS (FAB’) calcd for C32H3INSSZ: 591.2113, found

591.2117.

167

*— ‘M.~‘.p- -, -.

General procedure for the preparation of catalysts 56-61a

Following the general procedure for 52 and 55, the appropriate isothiocyanate (2
equiv.) was added to a solution of (R)-BINAM in THF (0.6M in BINAM) and the mixture
was stirred overnight at rt. The volatiles were evaporated and the crude mixtures were

purified by column chromatography on silica gel (always using 30% acetone in hexanes).

Catalyst 56a, compound 56b.

   

(Po-56a (50%) (H)-56b (50%)

Catalyst 56a

Synthesized from (R)—BINAM (300 mg, 1.05 mmol) and isolated as a white solid
(277 mg, 0.50 mmol, 50%), mp. = 128-130°C; 'H-NMR (CDC13, 300 MHz): 6 = 6.12 (d, J
= 7.2 Hz), 6.92 (t, J = 7.2 Hz, 4H), 7.03 (t, J = 7.5 Hz, 2H), 7.35-7.28 (m, 4H), 7.58-7.53
(m, 4H), 7.89 (d, J = 8.7 Hz, 2H), 8.0l(d, J = 2.7 Hz, 4H), 8.04 (d, J = 2.7 Hz, 4H), 8.30
(br. s, 2H); three exchangeable H’s do not show up. l3C-NMR (CDC13, 300 MHz): 6 =
125.30, 125.33, 126.4, 127.23, 127.29, 128.0, 128.1, 129.2, 129.5, 132.31, 132.33, 135.0,
135.6, 179.7; one C not located. IR (neat) 3159, 1593, 1535 cm"; MS (FAB*): m/z (%):
555 [M+H]*, (26), 462 (17), 307 (30), 154 (100), 136 (60); (1250 = +600 (c = 1.0,

dichloromethane).

168

Compound 56b. White solid, 50% isolated (210 mg, 0.5 mmol) , m.p. = 108-
110°C; lH-NMR (CDCl3, 300 MHz): 6 = 6.46 (d, J = 9MHz, 2H), 6.83 (d, J = 7.8 Hz, 1H),
7.07-6.92 (m, 4H), 7.31-7.24 (m, 3H), 7.13 (td, J = 1.5, 6.9 Hz, 1H), 7.51-7.41 (m, 1H),
7.52 (br s, 1H), 7.85 (d, J = 8.7 Hz, 2H), 7.96 (d,J = 8.4 MHz, 1H), 8.36 (d, J = 9 Hz, 1
H), 8.40 (br s, 1H); three exchangeable H’s do not show up. '3C-NMR (CDC13, 300 MHz):
6 = 111.5, 118.3, 122.5, 122.9, 124.5, 125.6, 125.8, 126.0, 126.7, 126.7, 126.9, 127.2,
127.4,127.9,128.1,129.6,132.3,132.4,133.4,135.2,135.3, 142.7,179.0 ppm; 2 CS not
located. MS (FAB+): m/z (%): 420 [M+H]+ (40), 307 (30), 154 (100), 136 (65); IR (neat)

3352, 3186, 1626, 1510; a251,: +24.2 (c = 1.0, dichloromethane).

Catalyst 57a.

ONOZ 3 N02
00 .0
CO NH2 SCN 11%
,,NH _,,N
so 2 T so “do
N02

(FD-57a
(1 00%)

Synthesized from (R)—BINAM (300 mg, 1.05 mmol) and isolated as a white solid
(644 mg, 1.00 mmol, 100%). 'H-NMR (CDCl3, 300 MHz): 6 = 6.69 (d, J = 9 Hz, 4H),
7.18 (d, J = 8.7 Hz, 2H), 7.31 (t, J = 8.0 Hz, 2H), 7.53 (t, J = 8.0Hz, 2H), 7.82 (d, J = 9
Hz, 4H), 7.98 (d, J = 8.1 Hz, 2H), 8.06 (s, 4H) ppm; 4 exchangeable H do not show up.
13C-NMR (CDC13, 300 MHz): 6 = 112.8, 122.1, 124.7, 126.3, 126.6, 126.9, 127.1, 127.6,
128.5, 132.3, 133.1, 136.4, 142.9, 145.9, 180.0 ppm; MS (FAB*): m/z: 645 [M+H]", 507.
460, 307, 154, 136; IR (neat) v: 3410, 1702, 1594, 1500, 1388, 1268 cm“; m.p. = 134-

136°C; [a]25D = -56.7 (c = 1.0, acetone).

169

Catalyst 58a

Cl

 

(3)538 Cl
(1 00%)

Synthesized from (R)-BINAM (300 mg, 1.05 mmol) and isolated as a white solid
(692 mg, 1.00 mmol, 100%), mp. = 116-120°C; lH-NMR (CDC13, 300 MHz): 6 = 6.50 (s,
4H), 7.04-7.03 (m, 2H), 7.30-7.22 (m, 4H), 7.55-7.48 (m, 2H), 7.74 (s, 2H), 7.88 (d, J =
9Hz, 2H), 8.00 (t, J = 9 Hz, 4H), 8.57 (br. s, 2H); ”C-NMR (CDC13, 300 MHz): 6 =179.8,
137.5, 135.2, 134.9, 132.12, 132.11, 128.9, 128.5, 128.2, 127.2, 127.0, 126.6, 126.3,
125.1, 123.6; IR (neat) 3434, 1644 cm"; MS (FAB‘): m/z (%): 693 [M+H]+ (l9, 3*35C1,
1*37Cl ), 691 [M+H]+ (8, 4*”Cl), 530 (1 l), 154 (100), 136 (60); [a]25D = +106.8 (c = 1.0,

dichloromethane).

Catalyst 59a, compound 59b.

OMe

 

Catalyst 59a. Synthesized from (R)—BINAM (300 mg, 1.05 mmol) and isolated as a
white solid (104 mg, 0.17 mmol, 17%). lH-NMR (CDC13, 300 MHz): 6 = 3.77 (s, 6H),
6.18 (d, 6.9 Hz, 4H), 6.47 (d, 6.9 Hz, 8H), 7.27-7.56 (m, 8H), 8.02-8.15 (m, 4H) ppm; 13C-

NMR (CDCl3, 300 MHz): 6 = 55.5,114.9,125.2,126.4,127.3,127.6,127.9,128.1,128.2,

170

132.3, 135.9, 158.9, 180.7 ppm; 3 CS not located. MS (FAB‘): m/z: 615 [M+H]*, 492,
307, 154; mp. = 124-125°C; IR (neat) v = 3163, 2953, 1508, 1244, 860 cm "; [alzso =

+15.3 (c = 1.0, acetone).

Compound 59b. Synthesized from (R)-BINAM (300 mg, 1.05 mmol) and isolated
as a white solid (224 mg, 0.50 mmol, 50%); mp. = 114-116°C; lH-NMR (CDC13, 300
MHz): 6 = 3.68 (s, 3H), 6.45-6.38 (m, 4H), 6.81 (d, J = 8.4 Hz, 1H), 7.15-7.09 (m, 1H),
7.29-7.22 (m, 3H), 7.37 (br. s, 1H), 7.50-7.45 (m, 1H), 7.70 (d, J = 8.7 Hz, 1H), 7.84 (d, J
= 8.7 Hz, 2H), 7.94 (d, J = 8.4 Hz, 1H), 8.02 (d, J: 9.0 Hz, 1H), 8.15 (br. s, 1H), 8.49 (d,
J = 8.7 Hz, 1H) ppm; 2 exchangeable H’s do not show up; '3C-NMR (CDC13, 300 MHz): 6
= 55.6, 111.6, 115.5, 123.3, 125.6, 125.9, 126.3, 127.0, 127.1, 127.6, 127.8, 128.3,
128.43, 128.47, 128.5, 130.0, 132.6, 132.7, 133.7, 133.8, 135.8, 143.2, 158.7, 179.9,
122.7, 118.7 ppm; IR (neat) 3350, 3184, 3055, 2961, 2837, 1620, 1531 cm"; MS (FAB*):
m/z (%): 450 [M+H]+ (98), 416 (20), 284, (100), 154 (70), 136 (50); [31250 = +662 (0 =

1.0, dichloromethane).

Compounds 61a, 61b.

NH2 1-Naphthyl-NCS
..~NH2

 

THF, rt

   

(H)-61b
(55%)

 

Compound 61a. Synthesized from (R)-BINAM (300 mg, 1.05 mmol) and isolated

as a white solid (32 mg, 0.05 mmol, 5%), not pure enough to characterize completely. IH-

171

NMR (CDC13, 300 MHz): 6 = 5.91 (d, J = 7.2 MHz, 2H), 6.85 (t, J = 8.1 MHz, 2H), 7.59-
7.14 (m, 16 H), 7.73 (dd, J = 8.7, 17.4 MHz, 4H), 7.88 (t, J = 8.4 MHz, 4 H); 2
exchangeable H’s do not show up. R, = 0.17 (30% acetone in hexanes); MS (FAB"): m/z

(%): 655 [M+H]+ (40), 512 (37), 436 (20), 391 (18).

Compound 61b. Synthesized from (R)-BINAM (300 mg, 1.05 mmol) and isolated
as a white solid (259 mg, 0.55 mmol, 55%); mp. = 126-128°C; ‘H-NMR (CDC13, 300
MHz): 6 = 3.1 (br. s, 2H), 6.42 (d, J = 7.2 Hz, 1H), 6.71 (d, J = 8.4 Hz, 1H), 6.76 (d, J =
8.7 Hz, 1H), 6.92 (t, J = 7.8 Hz, 3H), 7.01 (t, J = 7.2 Hz, 1H), 7.21 (t, J = 8.1 Hz, 3H),
7.56-7.43 (m, 3H), 7.63 (d, J = 6.0 Hz, 1H), 7.66 (d, J = 9.0 Hz, 1H), 7.76 (d, J = 8.1Hz,
1H), 7.81(d,J= 8.1Hz, 1H), 7.93 (d, J: 8.1 Hz, 1H), 8.01 (d,J= Hz, 1H), 8.53 (d, J = 9
Hz, 1H), 8.63 (br. s, 1H) ppm; l3C-NMR (CDC13, 300 MHz): 6 = 110.9, 117.8, 122.0,
122.3, 122.7, 124.3, 125.00, 125.05, 125.5, 125.8, 126.5, 126.6, 126.9, 127.1, 127.6,
127.9, 128.03, 128.05, 128.4, 129.4, 129.5, 130.8, 132.1, 132.2, 133.0, 134.1, 135.3,
142.3, 179.9 ppm; 2 C’s not located. MS (FAB+): m/z (%): 470 (60) [M+H]", 307 (20),

284 (40); IR (neat) 3350. 2959, 1620, 1531 cm"; [a]25D = +4.7 (c = 1.0, dichloromethane).

(R)-N-(l-(2-Aminonaphthalen-l-yl)naphthalen-2-y1)acetamide 63.

CO CO °
NH A020, ACOH: “k

2

_,\NH2 DCM .\‘NH2
00

(FD-51 (FD-63

 

To a solution of (R)-(+)-1,1’-Binaphthyl-2,2'-diamine (284 mg, 1.0 mmol) and

AcOH (0.6 mL, 10 mmol) in 10 mL of dried CH2C12 was added acetic anhydride (104 111.,

172

1.0 mmol) at 0 °C under N. The resulting solution was stirred overnight at room
temperature, then 2N NaOH aqueous solution was added until pH z 7. The reaction
mixture was extracted by CH2C12 (3 x 50 mL) and the combined organic phases were
washed with saturated brine and dried over MgSO4. The solvent was removed under
reduced pressure and the crude product was purified by flash chromatography (ethyl
acetate/hexane = 2/1) to afford a colorless oil in 77% yield (0.25 g, 0.77 mmol). lH-NMR
(CDC13, 300 MHz): 6 = 1.85 (s, 3H), 6.91-7.42 (m, 8H), 7.81‘8.03 (m, 4H), 8.58 (d, J =

9.0 Hz, 2H). The data agreed with the reported literature‘wb.

(R)-N-(l-(2-(Dimethylamino)naphthalen-l-yl)naphthalen-Z-yl)acetamide 64

O O

CO “k formaldehyde CO “k
ACOH
THF

(FD-63 85% (F064

 

N-(l-(2-Aminonaphthalen-1-yl)naphthalen-2-yl)acetamide 63 (0.25 g, 0.77 mmol)
and aqueous formaldehyde (37%, 0.75mL, 9.0 mmol) were combined in 10 mL of THF
and stirred for 15 min. NaBH3CN (200 mg, 5.3 mmol) was added, followed 15 min later
by AcOH (1.0 mL). The resulting solution was stirred for 4 h at room temperature, then
1N NaOH aqueous solution was added until pH z 7. The reaction mixture was extracted
by CHzCl2 (3 x 50 mL) and the combined organic phases were washed with saturated
brine and dried over MgSO4. The solvent was removed under reduced pressure and the
crude product was purified by flash chromatography (ethyl acetate/hexane = 1/5) to afford

a brown powder in quantitative yield (272 mg, 0.77 mmol). 1H-NMR (CDC13, 300 MHz):

173

6 = 1.88 (s, 3H), 2.58 (s, 6H), 6.95 (d, J = 8.7 Hz, 1H), 7.12-7.55 (m, 6H), 7.84-7.80 (m,
4H), 8.49 (d, J = 9.0 Hz, 1H). 1 exchangeable H does not show up. The data agreed with

the reported literature”.

(R)-1-(2-(Dimethylamino)naphthalen-l-yl)naphthalen-2-amine 65

o
“k 4MHC|
EtOH NH2
..N\ \ _
0O \ 00"“

(FD-64 (H)-65

 

To a solution of N-(1-(2-(dimethylamino)naphthalen-1-yl)naphthalen-2-
yl)acetamide 64 (0.18 g, 0.51 mmol) in 15 mL of EtOH was added 4M HCl (6 mL). The
resulting solution was stirred for overnight at room temperature, then 1N NaOH aqueous
solution was added until pH 2: 7. The reaction mixture was extracted by CHZCI2 (3 x 50
mL) and the combined organic phases were washed with saturated brine and dried over
MgSO4. The solvent was removed under reduced pressure and the crude product was
purified by flash chromatography (ethyl acetate/hexane = 1/10) to afford a colorless oil in
93% yield (148 mg, 0.47 mmol). 'H-NMR (CDC13, 300 MHz): 6 = 2.59 (s, 2Me), 7.0-7.29

(m, 7H), 7.47 (d, J = 9.0 Hz, 1H), 7.74-7.91 (m, 4H).

174

(R)-l-(3,5-Bis(trifluoromethyl)phenyl)-3-(l-(2-(dimethylamino)naphthalen-1-
yl)aaphthalen-Z-yl)thiourea 66

CF13 CF3

so we go . (j
NH2 CF3 “in CF3
.\N_ t
CC

(FD-65 (me

 

To a solution of 1-(2-(dimethylamino)naphthalen-1-yl)naphthalen-2-amine 65 (36
mg, 0.12 mmol) in 2 mL of dried CHZCl2 was added 3,5-bis(trifluoromethyl)phenyl
isothiocyanate (22 mg, 0.132 mmol) at 0 °C under N2. The resulting solution was stirred
for overnight at room temperature. The reaction was concentrated in vacuo and the residue
was purified by flash chromatography on silica gel (ethylacetate/hexane = 1/ 10) to afford a
slight yellow solid in 91% yield (64 mg, 0.11 mmol); 1H NMR (300 MHz, CDC13): 6 =
2.59 (s, 6H), 6.90 (d, 1H, J = 7.5 Hz), 7.09 (t, 1H,] = 7.5 Hz), 7.26 (m, 2H), 7.36 (s, 2H),
7.41 (s, 1H), 7.56-7.50 (m, 4H), 7.71 (d, 1H, J = 8.5 Hz), 7.82 (d, 1H, J = 8.0 Hz), 7.98 (d,
2H, J = 9.0 Hz), 8.06 (d, 1H, J = 8.5 Hz), 8.37 (s, 1H); MS (FAB*): m/z (%): 355 (100),

252 (30), 157 (37), 140 (47), 123 (37), 73 (60). The data matched the reported literature‘o".

175

General procedure for the asymmetric aza-Henry reaction (Table 20):

 

20% catalyst (FD-52
N’B°° 40%EtaN HN’BOC
| = NO
Ar) 10 eq RCHZNOZ Ph/yxr 2
toluene, -35°C R
41 42, 433-1:

A flame—dried round bottom flask was loaded with 0.2 equiv. of catalyst 52 (0.2
mmoles, 160 mg) and 1 equiv of imine 41a-k (1 mmol). The compounds were dissolved in
4 m1 of toluene. MeNOZ was added (10 equiv., 0.52 mL) at -35°C and after 5 minutes
Et3N was added (0.4 equiv, 56 uL). The mixture was stirred at —35°C for 17-36 hours and
then the volatiles were evaporated. The crude mixture was purified by column
chromatography on silica gel (always 20% acetone in hexanes) to afford products 42, 433-
R.

For all the optimization studies of the aza-Henry reaction detailed in Chapter 4, the
experimental procedures are the same as for the optimized reaction (described before),
with slight modifications that are clarified in each Table. Whenever a % conversion is
mentioned, that means the product was not purified, and the conversion was calculated
from the integration of product peaks versus the imine C(=N)H proton, given that the only
species observed in the crude lH-NMR was the catalyst, the desired product and imine,

whenever some of it remained unreacted.

tert-Butyl (R)-2-nitro-l-phenylethylcarbamate 42

NHBoc
N02

(FD-42

176

According to the typical procedure, imine 41a (1 mmol, 205 mg) and MeNO2 were
stirred for 36 h and converted to the product 42 (146 mg, 0.55 mmol, 55%) as a white
solid. HPLC analysis (Chiralpak OJ-H, hexane/2-PrOH 95/5, flow rate = 1.0 mL/min, l =
210 nm): tr (minor) = 32.3, tr (major) = 36.8 min; [a]259 = -18.0 (86% ee, 0 = 1.0, CHC13);
mp. 107-108 °C; 1H-NMR (300 MHz, CDCl3): 6 = 1.40 ppm (s, 9H); 4.64-4.81 (m, 2 H),
4.80 (br s, 1H), 5.34 (br s, 1 H), 7.37-7.23 (m, 5 H); l3C-NMR (300 MHz, CDC13): 6 =
28.3, 52.9, 78.9, 80.7, 126.4, 128.7, 129.2, 137.0,154.9 ppm.

Palomo’s group (ref. 27) reported the same absolute configuration (R) for 42 with
a negative Optical rotation, while Takemoto’s group (ref. 37) assigned the same absolute
configuration (R) to a positive optical rotation. While it is not clear which assignment is
right, we assigned the absolute stereochemistry of 42 following Palomo’s data (which
agrees with the majority of the literature reports), and subsequently assigned the

stereochemistry of the rest products (including 43k and 431) using 42 as a reference.

1-(p-Chlorophenyl)-2-nitroethyl carbamic acid t-butyl ester 43a

NHBoc
N02

CI
(R)-43a

Following the general procedure 239 mg imine were used (1 mmol) and
compound 43a was obtained as a white solid in 62% yield (0.62 mmol, 182 mg). The ee of
the product was determined by HPLC using a Daicel Chiralpak OJ-H column (n-hexane/i-
PrOH = 97/3, flow rate = 1 mL/min, tr (minor)=68 .6 min; tr (major)=74.l min, [alzsD = -

44.0 (c = 1.0, CHClg); 85% ee; lH-NMR (300 MHz, CDCl3): 6 = 1.47 (s, 9H), 4.72—4.90

177

(m, 1H), 4.73 (dd, J = 5.0, 12.6 Hz, 1H), 5.40 (br s, 2H), 7.28—7.30 (m, 2H), 7.34—7.41 (m,

2H); l3C-NMR (300 MHz, CDC13): 6 = 28.2, 52.3, 78.7, 80.9, 127.8, 129.4, 134.6, 135.6,

154.8. The data agreed with the previously reported literature”.

1-(m-Chlorophenyl)-2.nitroethyl carbamic acid t-butyl ester 43b.

NHBoc
No2

Cl (H)-43b
Following the general procedure, 239 mg imine were used (1 mmol) and
compound 43b was obtained as a white solid in 53% yield (161 mg, 0.53 mmol). The cc
of the product was determined by HPLC using a Daicel Chiralpak AD column (n-
hexane/i-PrOH = 97/3, flow rate = 1 mL/min, tr (minor) = 23.2 min; tr (major) = 33.6 min,
[31250 = -24.4 (c=1.0, acetone); 91% ee; lH-NMR (300 MHz, CDCl;,): 6 = 1.46 (s, 9H),
4.80-4.90 (br s, 1H), 4.72 (m, 1H), 5.42 (br s, 1H), 5.70 (br s, 1H), 7.21—7.35 (m, 4H);
l3C—NMR (300 MHz, CDC13): 6 = 28.2, 52.2, 78.6, 80.9, 124.6, 126.7, 128.8, 130.4,
135.0, 139.2, 154.9. MS (FAB‘): m/z (%): 303 [M+H]* (3, 37C1), 301 [M+H]+ (10, ”Cl),
245 (90), 184 (90) 154 (82); IR (neat): v: 3370, 2980, 1686, 1556, 1368, 1165 cm"; m.p.

98—100°C.

1-(o-Chlorophenyl)-2-nitroethyl carbamic acid t-butyl ester 43c.

NHBoc
N02

Cl

(9)-43C

178

Following the general procedure, 239 mg imine were used (1 mmol) and
compound 7d was obtained as a white solid in 61% yield (179 mg, 0.61 mmol). The cc of
the product was determined by HPLC using a Daicel Chiralpak AD column (n—hexane/i-
PrOH = 99/1, flow rate = 1 mL/min, tr (major) = 69.1 min; tr (minor) = 92.3 min, [a]25D 2
+190 (0 = 1.0, acetone); 74% ee; lH-NMR (300 MHz, CDC13): 6 = 1.47 (s, 9H), 4.84 (m,
2H), 5.78 (m, 2H), 7.31-7.46 (m, 4H); ”C-NMR (300 MHz, CDCl3): 6 = 28.2, 50.6, 77.5,
80.8, 127.5, 128.0, 129.9, 130.3, 132.6, 134.4, 154.6. MS (FAB*): m/z (%): 303 [M+H]+
(3, 37C1), 301 [M+H]+ (10, 35C1), 245 (90), 184 (90) 154 (82); IR (neat): 3355, 2982, 2935,

1712, 1685, 1555, 1367, 1253 cm"; mp. 106-108°C.

1-(p-Bromophenyl)-2-nitroethyl carbamic acid t-butyl ester 43d.

NHBoc
N02

Br
(FD-43d

Following the general procedure, compound 43d was obtained as a white solid in
50% yield (172 mg). The cc of the product was determined by HPLC using a Daicel
Chiralpak OJ-H column (n-hexane/i-PrOH = 95/5, flow rate = 1 mL/min, tr (minor) = 51.1
min; tr (major) = 54.9 min, 78% ee; 1H-NMR (300 MHz, CDC13): 6 = 1.47 (s, 9H), 4.75
(m, 1H), 4.76—4.90 (br s, 1H), 5.40 (br s, 2H), 7.23 (d, J = 6.9 Hz, 2H), 7.53 (d, J = 6.9
Hz, 2H); 13C-NMR (300 MHz, CDC13): 6 = 28.2, 52.2, 78.6, 80.9, 122.7, 128.1, 132.3,
136.1, 154.7; MS (FAB‘): m/z (%): 347 [M+H]“ (10, 8'Br), 345 [M+H]+ (10, 79Br), 289
(60), 228 (65), 154 (100); IR (neat): 3338, 2981 , 2935, 2363, 1687, 1527, 1166 cm"; m.p.

= 140-142.

179

tert-Butyl (R)-2-nitro-l-(p-methoxyphenyl)ethylcarbamate 43c.

NHBoc
N02

(FD-43:9

According to the typical procedure, 235 mg imine were used (1 mmol) and the
product 43c was obtained in 49% (145 mg, 0.49 mmol) as a white solid. HPLC analysis
(Chiralpak OJ-H, hexane/EtOH 95/5, flow rate=l .0 mL/min, 1:210 nm): tr (minor) = 65.8,
tr (major) = 72.0 min; [a]25D = -31.0 (89% ee, c = 1.00, CHCl3); 1H-NMR (300 MHz,
CDC13): 6 = 1.47 ppm (s, 9H), 3.82 (s, 3 H), 4.69 (dd, J = 12.0, 5.3 Hz, 1 H), 5.36 (m, 1H),
4.83 (s, 1H), 5.42 (s, 1 H), 6.92 (d, J = 8.6 Hz, 2 H), 7.25 (d, J = 8.6 Hz, 2H) l3C-NMR
(300 MHz, CDC13): 6 = 28.3, 52.4, 55.3, 78.9, 80.6, 114.5, 127.7, 129.0, 154.9, 159.8

ppm. The data agreed with the previously reported literature”.

tert-Butyl (R)-2-nitro-1-(o-methoxyphenyl)ethylcarbamate 43f.

NHBoc
N02

0
I (H)-43f

According to the typical procedure, 235 mg imine were used (1 mmol) and the
product 43f was obtained in 40% yield (116 mg, 0.4 mmol) as a white solid. HPLC
analysis (Chiralpak OJ-H, hexane/2-PrOH= 90/10, flow rate=1.0 mL/min,): t, (major):
17.2 min, tr (minor): 20.8.0 min; 65% ee; 'H-NMR (300 MHz, CDCl3): 6 = 1.48 (s, 9H),
3.94 (s, 3 H), 4.72-4.78 (m, 1H), 4.84 (dd, J = 7.5 Hz, J = 7.2 Hz, 1H), 5.63 (br s, 1 H),

5.76 (br s, 1H), 6.95-7.00 (m, 2H), 7.28-7.38 (m, 2H); l3C-NMR (300 MHz, CDC13): 6 =

180

28.3, 51.0, 55.5, 77.9, 80.3, 111.0, 121.19, 129.1, 130.9, 154.8, 156.8 ppm; MS (FAB+):
m/z (%): 297 [M+H]+ (5), 252 (100), 140 (100); IR (neat): v: 3433, 2358, 2105, 1646 cm'

I; mp. 138-140°C.

tert-Butyl (R)-2-nitro-1-(4-methylphenyl)ethylcarbamate 43h.

NHBoc
N02

H3C
(H)-43h

According to the typical procedure, 219 mg imine were used (1 mmol) the product
43h was obtained in 45% yield (134 mg) as a white solid. HPLC analysis (Chiralpak AD,
hexane/2-PrOH 98/2, flow rate = 1.0 mL/min, 1 = 210 nm): tr (minor)=50.5, tr (minor):
59.0 min; [a]25D = -26.0 (86% ee, c = 1.00, CHC13); 1H-NMR (300 MHz, CDC13): 6 =1.48
'(s, 9H), 2.37 (s, 3H), 4.70 (dd, J = 12.4, 5.7 Hz, 1H), 4.83 (s, 1H),5.40 (br s, 1 H), 5.55 (m,
1H), 7.22 (s, 4 H); 13C-NMR (300 MHz, CDC13): 6 = 21.1, 28.3, 52.7, 79.0, 80.5, 126.3,

129.8, 133.1, 138.5, 154.9 ppm. The data agreed with the previously reported literature”.

tert-Butyl (R)-2-nitro-l-(l-naphthyl)ethylcarbamate 431.

NHBoc
l N02

(Hi-431
According to the typical procedure 255 mg imine were used (1 mmol) and the
product 431 was obtained in 65% yield (165 mg, 0.65 mmol) as a white solid. HPLC
analysis (Chiralpak OJ-H, hexane/2-PrOH 95/5, flow rate = 1.0 mL/min, 1 = 210 nm): tr

(minor) = 35.9, tr (major) = 50.7 min; [a]25D = -8.1 (85% ee, c = 0.5, CHC13); lH-NMR
181

(300 MHz, CDC13): 6 = 1.43 (s, 9H), 4.88 (br s, 2 H), 5.25 (m, 1H), 6.27 (m, 1H), 7.45 (m,
2H), 7.54 (t, J = 7.5 Hz, 1 H), 7.61 (t,J = 7.3 Hz, 1H), 7.84 (dd, J = 5.7, 3.5 Hz,1H), 7.90
(d, J = 8.2 Hz, 1 H), 8.12 (d, J = 8.2 Hz, 1H); l3C-NMR (300 MHz, CDC13): 6 = 28.3,
49.3, 78.3, 80.7, 122.2, 123.3, 125.2, 126.3, 127.3, 129.3, 129.5, 130.3, 132.6, 134.1,

154.7 ppm; The rest of the data agreed with the previously reported literature”.

tert-Butyl (R)-2-nitro-1-(3-pyridy1)ethylcarbamate 43g.

NHBoc

\ N02

N/ (FD-439

According to the typical procedure, 206 mg imine were used (1 mmol) and the
product 43g was obtained in 63% yield (168 mg, 0.63 mmol) as a white solid. HPLC
analysis (Chiralpak AD, hexane/2-PrOH 90/ 10, flow rate = 1.0 mL/min, l = 210 nm): t,
(major) = 16.7, tr (minor) = 17.9 min; [a]25D = -335 (81% ee, c = 1.00, acetone); 'H—NMR
(300 MHz, CDC13): 6 = 1.42 (s, 9H), 4.76 (d, J=8.9 Hz, 1H), 4.90 (s, 1 H), 5.40 (s, 2H),
7.32 (dd, J = 7.5, 4.7 Hz, 1 H), 7.65 (d, J = 7.9 Hz, 2H), 8.62 (d, J = 21.4 Hz, 2H); 13C-
NMR (300 MHz, CDC13): 6 = 28.2, 50.7, 78.3, 81.0, 123.8, 133.1, 134.2, 148.1, 149.9.
154.8 ppm; The data agreed with the previously reported literature” except for the sign of

the optical rotation (see discussion on data for 42).

182

tert-Butyl (1R,ZS)-2-nitro-l-phenylpropylcarbamate 43j.
NHBoc
N02
(1R, 23)-43]

According to the typical procedure, 205 mg imine were used (1 mmol) and the
product 43j was obtained in 59% yield (165 mg, 0.59 mmol) as a white solid and a 77/23
mixture of diastereomers by lH-NMR. The major diastereomer was determined to have
70% ee by chiral HPLC analysis and identification of enantiomeric pairs by UV
absorption spectrum (Chiralpak AD, hexane/iPrOH 92/8 0.8 mL/min, 1 = 210 nm): Syn
isomer: tr (minor) = 13.8 min, tr (major) 15.9 min; anti isomer: t, (major) = 17.8 min, tr
(minor) = 21.8 min;); 1H-NMR (300 MHz, CDC13): 6 = 1.40 (s, 9H), 1.50 (d, J = 6.7 Hz,
3H), 4.91 (br s, 1 H), 5.18 (dd, J = 8.8, 5.8 Hz, 1 H), 5.32 (br s, 1H), 7.18—7.24 (m, 2H),
7.37-7.34 (m, 3 H); ”C-NMR (300 MHz, CDCl,): 6 = 154.9, 136.7, 129.0, 128.7, 128.5,
126.8, 126.4, 85.8, 80.0, 57.7, 28.3, 15.2 ppm. The data agreed with the previously

reported literature”.

tert-Butyl (1R,28)-2-nitro-l-phenylbutylcarbamate 43k.

NHBoc
No2
(1 R, 25)-43k
According to the typical procedure, 205 mg imine were used (1 mmol) the product
431 was obtained as white solid in 63% yield (185 mg, 0.62 mmol) and a 80/20 mixture of

diastereomers by HPLC analysis (enantiomeric pairs identified by comparison with a

183

racemic sample). The cc of major diastereomer 7k was determined to be 80% by chiral
HPLC analysis (Chiralpak OJ-H, hexane/2-PrOH 97/3, 0.8 mL/min, 1 = 210 nm). Syn
isomer: t, (major) = 28.1 min, tr (minor) = 50.1 min; 1H-NMR (300 MHz, CDC13): 6 =
0.96—1.03 (t, J = 3.5 Hz, 3H), 1.43 (s, 9 H), 1.84—1.92 (m, 2H), 5.10—5.14 (m, 1 H), 4.74
(br s, 1H), 5.14—5.20 (br s, 1H), 7.22-7.26 (m, 2 H), 7.30—7.40 (m, 3H); l3C-NMR (300
MHz, CDC13): 6 = 10.4, 24.8, 28.2, 56.8, 80.0, 93.0, 126.9, 128.7, 129.0, 136.7, 154.9

ppm. The data agreed with the previously reported literature”.

General procedure for the synthesis of aminosulfones 62 and imines 41.

 

 

NHBoc
JO PhSOzNa HN ' Boc K2C03, Na2804 N , Boc
; > I
A' MeOH/1420 A' 302% THF reflux Ar
HCOOH 62 41

The appropriate aldehyde (1.2 eq, 20.48 mmol) was added to a mixture of 1 eq (2g,
17.07 mmol) of NHBoc and 2.0 eq. (5.6 g, 34.1 mmol) of PhSOzNa in 50 ml of water and
MeOH (2:1 v/v) and stirred at rt for 3 days, after which time the resultant white
suspension was filtered, washed with water and diethyl ether and then triturated with
diethyl ether overnight. The product 62 was dried under vacuum.

Subsequently, 1 eq. of 62 (2 mmoles) was refluxed in THF for an overnight period
in the presence of 1.60 g (12 mmol) K2C03 and ~2g Na2S04 (drying agent). The resulting
mixture was filtered through a clean white frit funnel and the volatiles were evaporated.
The resulting imine was transferred carefully under nitrogen to the pump where it was
dried under high vacuum. Cautious handling of the procedure is required for preventing

decomposition and hydrolysis of the imines, which were used in the aza-Henry reaction

184

 

without further purification. Imines 41a, e-f and h-j are known in the literature and their

preparation and data has been reported before.28

N-(ten-butoxycarbonyl)-a-(phenylsulfonyl)benzylamine 62a.

NHBoc
SOgPh
62a

Synthesized using (2g, 17.07 mmol) of NHBoc to afford 5.89 g (17 mmol, 100%)
of 623 as a white solid. 'H-NMR (300 MHz, CDC13): 6 = 1.24 (9H, s), 5.72 (1H, br d, J =
10.3 Hz), 7.49-7.38 (5H, m), 5.90 (1H, br (1, J = 11.0 Hz), 7.52 (2H, m), 7.62-7.60 (1H,

m), 7.89 (2H, d, J = 7.3 Hz).

N-(tert-butoxycarbonyl)-a-(phenylsulfonyl)-4—chlorobenzylamine 62b.

NHBoc
SOzPh

Cl
621)

Synthesized using 2g (17.07 mmol) of NHBoc to afford 1.48 g (3.91 mmol, 23%)
of 62b as a white solid. lH-NMR (300 MHz, CDC13): 6 = 1.25 (9H, s), 5.63 (1H, br (1, J =
10.1 Hz), 5.86 (1H, br d, J = 10.1 Hz), 7.36 (4H, s), 7.52 (2H, t, J: 7.8 Hz), 7.63 (1H, t,J

= 7.5 Hz), 7.88 (2H, d, J = 8.4 Hz).

185

N-(tert-butoxycarbonyl)-a-(phenylsulfonyl)-3-chlorobenzylamine 62c.

NHBoc
SOzPh

Cl
62c

Synthesized using 2g (17.07 mmol) of NHBoc to afford 4.98 g (13.9 mmol, 77%)
of 62c as a white solid. 1H-NMR (300 MHz, CDCl3): 6 21.23 (9H, s), 5.69 (1H, br (1, J =
10.5 Hz), 5.87 (1H, br (1, J = 10.5 Hz),7.32-7.41 (4H, m), 7.54 (2H, t, J = 7.5 Hz), 7.62

(1H, t, J: 7.2 Hz), 7.90 (2H, d, J: 7.2 Hz).

N -(ten-butoxycarbonyl)-a-(phenylsulfonyl)-2-chlorobenzylamine 62d.

NHBoc
SOzPh
Cl

62d

Synthesized using 2g (17.07 mmol) of NHBoc to afford 4.66 g (12.24 mmol, 77%)
of 62d as a white solid. 'H-NMR (300 MHz, CDC13): 6 = 1.32 (9H, s), 5.85 (1H, br (1, J =
10.2 Hz), 6.65 (1H, br (1, J = 10.2 Hz), 7.37-7.48 (3H, m), 7.55-7.58 (3H, m), 7.69 (1H, t,

J = 6.9 Hz), 7.97 (2H, d,J = 7.5 Hz)

N-(ten-butoxycarbonyl)-a-(phenylsulfonyl)-4-bromobenzylamine 62e.
NHBOC
SOzPh

Br
62e

186

Synthesized using 2g (17.07 mmol) of NHBoc to afford 2.24 g (5.27 mmol, 31%)
of 62e as a white solid. 1H-NMR (300 MHz, CDC13): 6 = 1.25 (9H, s), 5.70 (1H, br d, J =
10.7 Hz), 5.89 (1H, br d, J = 10.7 Hz), 7.28 (2H, d, J = 8.2 Hz), 7.58-7.51 (4H, m), 7.72-

7.62 (1H, m), 7.91 (2H, d,J = 7.6 Hz).

N-(tert-butoxycarbonyl)-a-(4-methylphenylsulfonyl)-4-methoxybenzylamine 62f.

NHBoc
SOzPh

62f
Synthesized using 2g (17.07 mmol) of NHBoc to afford 5.83 g (14.28 mmol, 84%)
of 62f as a white solid. 1H-NMR (300 MHz, CDCl,): 6 = 1.25 (9H, s), 2.41 (3H, s), 3.81
(3H, s), 5.79 (1H, br d, J = 10.6 Hz), 5.85 (1H, br d, J = 11.0 Hz), 6.92 (2H, d, J = 8.8

Hz), 7.31 (2H, d, J: 8.1 Hz), 7.36 (2H, d, J: 8.8 Hz), 7.78 (2H, d, J: 7.7 Hz).

N-(tert-butoxycarbonyl)-or-(4-methylphenylsulfonyl)-2-methoxybenzylamine 62g.

NHBoc

SOgPh

O/

629
Synthesized using 2g (17.07 mmol) of NHBoc to afford 5.76 g (15.3 mmol, 90%)
of 62g as a white solid. 1H-NMR (300 MHz, CDC13): 6 = 1.28 (9H, s), 2.37 (3H, s), 3.70
(3H, s), 6.16-6.27 (2H, m), 6.84 (1H, d, J = 8.1 Hz), 6.96 (1H, t, J = 7.2 Hz), 7.22-7.35
(4H, m), 7.68 (1H, d, J = 8.1 Hz); l3C-NMR (300 MHz, CDCl,): 6 = 21.6, 28.1, 55.9, 71.1,

80.8, 111.5, 118.9, 120.9, 129.4, 129.5, 130.3, 131.1, 134.6, 144.6, 154.5, 157.9; MS

187

 

(FAB*): m/z (%): 307 (10), 236 (100), 180 (100), 136 (90); IR (CHC13): v = 3344, 2978,

2358,1704, 1493, 1141; mp 160-162.

N-(tert-butoxycarbonylya-(phenylsulfonyl)-4-methylbenzylamine 62i.

NHBoc

/©/‘\802Ph

621
Synthesized using 2g (17.07 mmol) of NHBoc to afford 4.54 g (12.58 mmol, 74%)
of 62i as a white solid. 1H-NMR (300 MHz, CDC13): 6 = 1.28 (9H, s), 2.40 (3H, s), 5.77
(1H, d, J = 10.6 Hz), 5.90 (1H, br d, J =10.6 Hz), 7.27 (2H, d,J = 8.0 Hz), 7.34 (2H, d, J

= 7.7 Hz), 7.53 (2H, t,J = 7.6 Hz), 7.64 (1H, t,J = 7.0 Hz), 7.93 (2H, d, J = Hz).

N-(ten-butoxycarbonyl)-a-(phenylsu1fonyl)-C-naphthalen-l-yl-methylamine 62k.

0 NHBoc
O SOgPh

62k
Synthesized using 2g (17.07 mmol) of NHBoc to afford 3.98 g (10.03 mmol, 59%)
of 62k as a white solid. 'H-NMR (300 MHz, CDC13): 6 = 1.27 (9H, s), 5.98 (1H, br (1, J =
10.6 Hz), 6.88 (1H, br (1, J = 10.6 Hz), 7.66-7.48 (6H, m), 7.79 (1H, d, J = 7.0 Hz), 7.88
(1H, d,J: 8.1 Hz), 7.94 (1H, d, J: 8.1Hz,), 7.99 (2H, d,J: 7.7 Hz), 8.14 (1H, d, J: 8.4

Hz).

N-(tert-butoxycarbonyl)-a-(phenylsulfonyl)-C-pyridin-3-yl-methylamine 62h.

188

NHBoc

/
N
62h
Synthesized using 2g (17.07 mmol) of NHBoc to afford 4.02 g (11.56 mmol, 68%)
of 62h as a white solid. 'H-NMR (300 MHz, CDCl,): 6 = 1.25 (9H, s), 5.97 (2H, br), 7.39-
7.35 (1H, m), 7.57 (2H, t, J = 7.5 Hz), 7.68 (1H, t, J = 7.0 Hz), 7.86-7.83 (1H, m), 7.92

(2H, d, J = 8.3 Hz), 8.70-8.63 (2H, m).

Benzaldehyde N-(tert-butoxycarbonyl)imine 4la.

N,Boc

d...

Synthesized from 5.89 g (17 mmol) of 62a to afford 4.40 g (16.66 mmol, 98%) of

41a as a colorless liquid. 'H-NMR (300 MHz, CDC13): 6 = 1.59 (9H, s), 7.49-7.43 (2H,

m), 7.57-7.54 (1H, m), 7.93-7.90 (2H, m), 8.88 (1H, s).

p-Chlorobenzaldehyde N-(tert-butoxycarbonyl)imine 41b.

.80c

.04"

42b
Synthesized from 1.48 g (3 .91 mmol) of 62b to afford 0.91 g (3.83 mmol, 98%) of
41a as a colorless liquid. IH-NMR (300 MHz, CDC13): 6 = 1.57 (9H, s), 7.42 (2H, d, J =

8.4 Hz), 7.85 (2H, d, J: 8.4 Hz), 8.81 (1H, s).

189

m-Chlorobenzaldehyde N-(tert-butoxycarbonyl)imine 41c.

N,Boc

CI
41c

Synthesized using 4.98 g (13.9 mmol) of 62c to afford 3.32 g (13.76, 99%) 41c. as
a clear oil. 'H-NMR (300 MHz, CDC13): 6 = 1.56 (9H, s) 7.39 (1H, t, J = 7.5 Hz), 7.48-

7.52 (1 H, m), 7.73 (1H, dd, J=1.2 Hz, J = 7.5 Hz), 8.77 (1H, s), 7.93 (1H, s).

o-Chlorobenzaldehyde N-(ted-butcxycarbonylfimine 41d.

Synthesized using 4.66 g (12.24 mmol) of 62d to afford 2.86 g (11.99 mmol, 99%)
41d as a clear oil. 1H-NMR (300 MHz, CDC13): 6 = 1.63 (9H, s), 7.37 (1H, t, J = 6.0 Hz),

7.47-7.50 (2H, m), 8.22 (1H, d, J :60 Hz), 9.31 (1H, s).

p-Bromobenzaldehyde N-(tert-butoxycarbonyl)imine 41c.

190

Synthesized using 2.24 g (5 .27 mmol) of 62e to afford 0.95 g (3.37 mmol, 64%)
41c as a white solid. 1H-NMR (300 MHz, CDC13): 6 = 1.57 (9H, s). 7.60 (2H, d, J = 8.4

Hz), 7.77 (2H, d, J: 8.4 Hz), 8.81 (1H, s).

p-Methoxybenzaldehyde N-(tert-butoxycarbonyl)imine 41f.

Synthesized using 5.83 g (14.28 mmol) of 62f to afford 2.58 g (10.99 mmol, 77%)
41f as a clear oil. 1H-NMR (300 MHz, CDC13): 6 = 1.56 (9H, s), 3.86 (3H, s), 6.96 (2H, d,

J = 8.8 Hz), 7.88 (2H, d,J: 8.8 Hz), 8.88 (1H, s).

o-Methoxybenzaldehyde N-(tert-butoxycarbonyl)imine 41g.

Synthesized using 5.76 g (15.3 mmol) of 62g to afford 3.41 g (14.53 mmol, 95%)
41g as a clear oil. 1H-NMR (300 MHz, CDC13): 6 = 1.56 (9H, s), 3.88 (3H, s), 6.90-7.00
(2H, m), 7.49 (1H, t, J = 9.0 Hz), 8.09 (1H, d, J = 7.8 Hz), 9.35 (1H, 5); ”GM (300
MHz, CDC13): 6 = 27.9, 55.6, 81.9, 111.27, 120.77, 122.6, 128.3, 135.2, 160.9, 165.8,
163.2; MS (FAB*): m/z (%): 236 [M+H]+ (60), 180 (100), 136 (95); IR (CHC13): v=2978,

1711,1600,1237,1153.

191

p-Tolualdehyde N-(tert-butoxycarbonyl)imine 4li.

N,Boc

fig
411
Synthesized using 4.54 g (12.58 mmol) of 62i to afford 2.67 g (12.20 mmol, 97%)
4li as a white solid. 1H-NMR (300 MHz, CDC13): 6 = 1.58 (9H, s), 2.41 (3H, s), 7.26 (2H,

d,J: 7.7 Hz), 7.81(2H, d,J = 8.1 Hz), 8.87 (1H, s).

l-Naphthaldehyde N-(ten-butoxycarbonymmine 41k.

0 IN’BOC

41k
Synthesized using 3.98 g (10.03 mmol) of 62k to afford 2.55 g (10.00 mmol, 99%)
41k as a white solid. 'H-NMR (300 MHz, CDC13): 6 = 1.61 (9H, s), 7.60-7.54 (2H, m),
7.68-7.63 (1H, m), 7.90 (1H, d, J = 7.7 Hz), 8.05 (1H, d, J = 8.4 Hz), 8.17 (1H, d, J = 7.0

Hz), 8.92 (1H, d, J: 8.4 Hz), 9.53 (1H, s).

3-pyridinecarboxaldehyde N-(tert-butoxycarbonyl)imine 41h.

192

Synthesized using 4.02 g (11.56 mmol) of 62h to afford 2.32 g 41h (11.3 mmol,
98%) as a clear oil. 1H-NMR (300 MHz, CDCl3): 6 = 1.58 (9H, s), 7.40 (1H, dd, J = 4.8,
8.1 Hz), 8.25-8.28 (1H, dt, J = 1.8, 8.1 Hz), 8.76 (1H, dd, J = 1.83, 4.8 Hz), 8.87 (1H, s),

9.00 (1H,d,J= 2.2 Hz).

 

193

Experimental Procedures for Chapter 5

General procedure for the non-asymmetric addition of l-nitropropane to
nitrostyrene using catalyst 47 (Table 28).

2%

 

FC. ; ii ixi ; .CF
3 \n/ 3
S
R/VNOZ CF3 CF3
N
(022M) 2% EtaN 02
> R N02
+ /\/N02 toluene, rt
20 eq 803-1

A flame-dried round bottom flask was charged with 1 mmol of the appropriate
nitrostyrene and 10 mg (0.02 mmol) of catalyst 47 and the solids were dissolved in 3 mL
of toluene. 1-Nitropropane (1.6 ml, 20 mmol) was added followed by 3uL of EN. The
mixture was stirred at r.t. for the time indicated in the Table. The volatiles were
evaporated and the crude product was purified with silica gel column chromatography to
afford the racemate products 80a-i as separated diastereomers. The syn/anti ratio was
measured by the weight of the isolated products except in the case of 80a, 80b where it

was determined by lH-NMR analysis.

1. 94

General procedure for the asymmetric addition of l-nitropropane to

nitrostyrene using catalyst (R)-82 (Table 39).

 

NO
R “~02 + /\/N02 2% mol. 82 2
> N02
(0.2M) 30 eq benzene, rt R
BOa-l

A flame-dried round bottom flask was charged with 1 mmol of the appropriate
nitrostyrene and 13.5 mg (0.02 mmol) of catalyst 82 and the solids were dissolved in 2.5
mL of benzene. 1-Nitropropane (2.5 mL, 30 mmol) was added and the mixture was
stirred at r.t. for the time indicated in the Table 39. The volatiles were evaporated and the
crude product was purified with silica gel column chromatography to afford the products
80a-i as separated diastereomers. The syn/anti ratio was measured by the weight of the
isolated products except in the case of 80a, 80b, 80c, 80d where it was determined by 1H-

NMR analysis.

2-Phenyl-1,3-dinitropentane 803

N02
No2

(2H, 3F0-80a
Prepared following the general procedures described above using 150 mg (1.00
mmol) nitrostyrene. Purified with silica gel column using 30% EtOAc/hexanes to afford
144 mg (60%, 0.60 mmol) of syn adduct as a clear oil and 47 mg (0.20 mmol, 20%) anti
adduct from the asymmetric addition (Table 39). Syn adduct: 'H-NMR (300MHz), CDCl3
6 =1.00 (t, 3H, J = 7.1 Hz), 1.90-1.82 (m, 1H), 2.02-1.94 (m, 1H), 4.02 (q, 1H, J = 6.7
Hz), 4.78-4.71 (m, 2H), 4.86 (dd, 1H, J = 6.3, 13.9 Hz), 7.14-7.11 (m, 2H), 7.33-7.29 (m,

195

3H); l3C-NMR (300MHz), CDC], 6 = 10.3, 24.4, 46.6, 76.4, 91.2, 127.9, 128.1, 129.3,
133.9 ppm; R, = 0.30 (30% EtOAc/hexanes); HPLC analysis: Chiralpac OJ—H, hexane/2-
PrOH 80/20, 1.5 mL/min, 95% ee, (1”,, = +7.2 (c=1.0, dichloromethane), tr = 31.0 min
(major), tr = 42.0 min (minor). The data agreed with the reported literature“.

The non-asymmetric addition (Table 28) afforded 178 mg (75%) of syn-80a

adduct and 28 mg (0.12 mmol, 12%) anti adduct.

2-(4-Methoxyphenyl)-1,3-dinitropentane 80b

NO2
N02

MeO
(2R, 3H)-80b

Prepared following the general procedures described above using 179 mg (1.00
mmol) of the nitrostyrene. Purified with silica gel column using 20% EtOAc/hexanes to
afford 126 mg (47%, 0.47 mmol) of syn adduct as a clear yellow oil and 34 mg (13%, 0.13
mmol) anti adduct from the asymmetric addition (Table 39). Syn adduct: lH-NMR
(300MHz), CDC], 6 = 0.99 (t, 3H, J = 7.1 Hz), 187-1.1.80 (m, 1H), 2.02-1.93 (m, 1H),
3.75 (s, 3H), 3.95 (q, 1H, J = 6.7 Hz), 4.72-4.71 (m, 2H), 4.83 (dd, 1H, J: 6.3, 13.9 Hz),
7.04 (d, J = 8.7 Hz, 2H), 7.84 (d, J = 8.7 Hz, 2H); l3C-NMR (300MHz), CDC], 6 =10.5,
24.6, 46.2, 55.5, 76.8, 91.5, 114.9, 125.8, 129.3, 160.2; R, = 0.08 (20% EtOAc/hexanes);
HPLC analysis: Chiralpac OD-H, hexane/2-PrOH 70/30, 0.5 mL/min, 94% ee, (1”,, =
+220 (c=1.0, dichloromethane), tr = 15.6 min (major), tr = 31.1 min (minor). The data

agreed with the reported literature“.

196

The non-asymmetric addition (Table 28) afforded 197 mg (0.73 mmol, 73%) of

syn-80a adduct and 24 mg (0.09 mmol, 9%) of the anti adduct.

2-(2-Methoxyphenyl)-1,3-dinitropentane 80c

No2
NO2

OMe
(2R, 3H)-80c

Prepared following the general procedures described above using 179 mg (1.00
mmol) of the nitrostyrene. Purified with silica gel column using 20% EtOAc/hexanes to
afford 193 mg (72%, 0.72 mmol) of syn adduct as a clear oil and 58 mg (0.22 mmol, 22%)
of the anti adduct from the asymmetric addition (Table 39). Syn adduct: lH-NMR
(300MHz), CDC1,6 = 0.98 (t, 3H,J= 7.1 Hz), 1.93-1.82 (m, 1H), 2.03-1.94 (m, 1H), 3.85
(s, 3H), 4.24 (q, 1H, J = 7.5 Hz), 4.84 (dd, 2H, J = 5.1, 7.5 Hz), 5.00 (m, 1H), 6.88-6.86
(m, 2H), 7.04 (dd, J = 1.6, 7.5 Hz, 1H), 7.28-7.25 (m, 1H). l3C-NMR (300MHz), CDC], 6
= 10.6, 24.8, 43.9, 55.7, 75.6, 90.2, 111.6, 121.4, 122.4, 130.4, 130.5, 157.5 ppm; R, =
0.20 (20% EtOAc/hexanes); HPLC analysis: Chiralpac OD-H, hexane/Z-PrOH 80/20, 0.5
mL/min, 94% ee, a251, = +8.0 (c=1.0, dichloromethane), tr = 11.9 min (major), tr = 23.3
min (minor). The data agreed with the reported literature“.

The non-asymmetric addition (Table 28) afforded 178 mg (0.66 mmol, 66%) of

syn-80c adduct and 45 mg (0.17 mmol, 17%) of the anti adduct.

197

4-Methyl-1,3-dinitropentane 80d

N02
N02

H3C
(29, 3H)-80d

Prepared following the general procedure described above using 163 mg (1.00
mmol) of the nitrostyrene. Purified with silica gel column using 20% EtOAc/hexanes to
afford 146 mg (58%, 0.58 mmol) of syn adduct as a clear oil and 35 mg (0.14 mmol, 14%)
from the asymmetric addition (Table 39). Syn adduct: 'H-NMR (300MHz), CDC], 6 =
0.98 (t, J = 7.2 Hz, 3H), 1.80-1.84 (m, 1H), 2.02-1.96 (m, 1H), 2.29 (s, 3H), 3.97 (q, J =
8.1 Hz, 1H), 4.75-4.68 (m, 2H), 4.80 (dd, J = 13.5, 6.3 Hz, 1H), 7.10 (d, J = 8.1 Hz, 2H),
7.12 (d, J = 8.1 Hz, 2H); l3C-NMR (300MHz), CDC], 6 = 10.3, 21.1, 24.4, 46.2, 76.5,
91.2, 127.8, 129.9, 130.8, 138.9 ppm; R, = 0.17 (20% EtOAc/hexanes); MS (FAB'): m/z
(%): 251 [M-H]' (60), 239 (25), 204 (60); IR (neat) 3028, 2978, 2926, 2883, 1552, 1510,
1460, 1435, 1377, 1318, 1122 cm"; HPLC analysis: Chiralpac OD-H, hexane/2-PrOH
70/30, 0.5 mL/min, 94% ee, 0”,, = +11.0 (c=1.0, dichloromethane), tr = 14.8 min (major),
tr = 45.7 min (minor).

The non-asymmetric addition (Table 28) afforded 142 mg (0.56 mmol, 56%) of

syn-80d adduct and 40 mg (0.16 mmol, 16%) of the anti adduct.

198

2-(2-Chlorophenyl)-1,3-dinitropentane 80c

no2
N02

Cl
(23, 3H)-BOe

Prepared following the general procedure described above using 183 mg (1.00
mmol) of the nitrostyrene. Purified with silica gel column using 15% EtOAc/hexanes to
afford 175 mg (0.62 mmol, 62%) of syn adduct as a clear oil and 60 mg (0.22 mmol, 22%)
of the anti adduct from the asymmetric addition (Table 39). Syn adduct: 'H-NMR
(300MHz), CDC], 6 = 1.06 (t, J = 7.2 Hz, 3H), 2.08-2.04 (m, 2H), 4.70-4.67 (m, 1H), 4.90
(dd, J = 5.9, 8.3 Hz, 2H), 5.02-5.07 (m, 1H), 7.17-7.20 (m, 1H), 7.29-7.45 (m, 2H), 7.49-
7.46 (m, 1H); l3C-NMR (300MHz), CDC], 6 = 10.6, 24.3, 43.0, 75.4, 90.2, 127.9, 128.8,
130.4, 130.9, 132.0, 134.5; R, = 0.22 (15% EtOAc/hexanes); HPLC analysis: Chiralpac
OJ-H, hexane/2-PrOH 70/30, 0.5 mL/min, 92% ee, (1”,, = +6.7 (c=1.0, dichloromethane),
tr = 48.4 min (major), tr = 74.9 min (minor). The data agreed with the reported literature48
except for the sign of the optical rotation.

The non-asymmetric addition (Table 28) afforded 185 mg (68%) of syn-80c adduct
and 32 mg (0.12 mmol, 12%) of the anti adduct.

Data for anti-80c: lH-NMR (300MHz), CDC], 6 = 0.93 (t, J = 7.5 Hz, 3H), 1.61-
1.69 (m, 1H), 1.95-2.00 (m, 1H), 4.55-4.62 (m, 1H), 4.69 (dd, J = 3.9 Hz, 13.5 Hz, 1H),
4.80-5.00 (m, 1H), 7.14-7.44 (m, 4H); l3C-NMR (300MHz), CDC], 6 = 10.2, 25.4, 42.6,
75.0, 90.76, 103.9, 127.9, 130.2, 130.9, 131.9; one C not located; a251, = +8.0 (c=0.8,

dichloromethane).

199

2-(4-Chlorophenyl)-1,3-dinitropentane 80f

no2
N02

Cl
(2 H, 33)-80f

Prepared following the general procedure described above using 183 mg (1.00
mmol) of the nitrostyrene. Purified with silica gel column using 20% EtOAc/hexanes to
afford 168 mg (0.62 mmol, 62%) of syn adduct as a clear oil and 35 mg (0.13 mmol, 13%)
anti adduct from the asymmetric addition (Table 39). Syn adduct: lH-NMR (300MHz),
CDC],6 = 0.99 (t, J = 7.2 Hz, 3H), 1.80-1.90 (m, 1H), 1.96-1.99 (m, 1H), 4.00 (q, J: 7.8
Hz, 1H), 4.75-4.67 (m, 2H), 4.83 (dd, J = 13.5, 6.0 Hz, 1H), 7.08 (dd, J = 6.6, 2.4 Hz,
2H), 7.31 (dd, J = 6.6, 2.4 Hz, 2H). l3C-NMR (300MHz), CDC],6 = 10.2, 24.5, 45.9, 76.2,
90.9, 129.3, 129.5, 132.3, 135.]; MS (FAB'): m/z (%): 273 [M-H]' (27, 37Cl), 271 [M-H]'
(80, 35Cl) 224 (100); IR (neat) 3406 (H20), 2980, 1554, 1493, 1460, 1435, 1377, 1095,
1014 cm"; HPLC analysis: Chiralpac OD-H, hexane/Z-PrOH 70/30, 0.5 mL/min, 92% cc,
0125,, = +10.9 (c=1.0, dichloromethane), tr = 13.7 min (major), tr = 24.8 min (minor). For
the anti adduct (x250: +23.2 (c=1.0, dichloromethane).

The non-asymmetric addition (Table 28) afforded 157 mg (58%) of syn-80a

adduct and 40 mg (0.15 mmol, 15%) of the anti adduct.

2-(4-Bromophenyl)-1,3-dinitropentane 80g.

NO2
N02

Br
(2R, 3H)-809

200

 

Prepared following the general procedure described above using 228 mg (1.00
mmol) of the nitrostyrene. Purified with silica gel column using 20% EtOAc/hexanes to
afford 207 mg (0.65 mmol, 65%) of syn adduct as a clear oil and 31 mg (0.10 mmol, 10%)
of the anti adduct from the asymmetric addition (Table 28). Syn adduct: lH-NMR
(300MHz), CDC],6 = 1.00 (t, J = 7.2 Hz, 3H), 1.88-1.81 (m, 1H), 2.03-1.96 (m, 1H), 3.99
(q, 6 Hz, 1H), 4.67-4.76 (m, 2H), 4.83 (dd, J = 19.5, 6 Hz, 1H), 7.02 (d, J = 9 Hz, 2H),
7.46 (d, J = 9 Hz, 2H). I3C-NMR (300MHz), CDC], 6 = 10.2, 24.4, 46.0, 76.0, 90.8, 123.3,
129.6, 132.5, 132.8 ppm; R, = 0.11 (20% EtOAc/hexanes); IR (neat) 3451(H20), 2112,
1653, 1558, 1375; MS (FAB'): m/z (%): 317 [M-H]' (98, S'Br), 315 [M-H]' (97, 7S’Br), 268
(100); HPLC analysis: Chiralpac OD-H, hexane/Z-PrOH 70/30, 0.5 mL/min, tT = 14.8 min
(major), tr = 31.8 min (minor), 93% ee, 0,st = +10.2 (c=1.0, dichloromethane).

The non-asymmetric addition (Table 28) afforded 180 mg (0.57 mmol, 57%) of

syn-80a adduct and 28 mg (0.09 mmol, 9%) of the anti adduct.

2-(3-Bromophenyl)-l,3-dinitropentane 80h.

N02
N02
Br
(21?, 3Fi)-80h

Prepared following the general procedure described above using 228 mg (1.00
mmol) of the nitrostyrene. Purified with silica gel column using 20% EtOAc/hexanes to
afford 197 mg (0.62 mmol, 62%) of syn adduct as a clear yellow oil and 22 mg (0.07

mmol, 7%) of the anti adduct from the asymmetric addition (Table 39). Syn adduct: 1H-

201

NMR (300MHz), CDC], 6 = 1.00 (t, J = 7.5 Hz, 3H), 1.88-1.81 (m, 1H), 2.03-1.94 (m,
1H), 3.97 (q, 6 Hz, 1H), 4.77-4.70 (m, 2H), 4.83 (dd, J = 13.5, 6 Hz, 1H), 7.07 (d, J = 8.1
Hz, 1H), 7.21 (t, J = 7.8 Hz, 1H), 7.30 (t, 1.8 Hz, 1H), 7.47 (dq, J = 8.1 Hz, J = 1.2 Hz,
1H). l3C-NMR (300MHz), CDC], 6 = 10.2, 24.5, 46.1, 76.0, 90.9, 123.2, 126.5, 130.9,
131.2, 132.3, 136.3; MS (FAB'): m/z (%): 317 [M-H]' (98, 8'Br), 315 [M-H]‘ (97, 7S’Br),
268 (100); IR (neat) 3451 (H20, 2112, 1653, 1558, 1375, 1010; HPLC analysis: Chiralpac
OD—H, hexane/2-PrOH 70/30, 0.5 mL/min, 94% ee, 0251) = +4.3 (c=1 .0, dichloromethane),
tr = 14.5 min (major), tr = 69.6 min (minor).

The non-asymmetric addition (Table 28) afforded 217 mg (0.69 mmol, 69%) of

syn-80h adduct and 19 mg (0.06 mmol, 6%) of the anti adduct.

2-(2-Bromophenyl)1,3-dinitropentane 80i

NO2
N02

Br

(21?, {Sm-80]

Prepared following the general procedure described above using 228 mg (1.00
mmol) of the nitrostyrene. Purified with silica gel column using 20% EtOAc/hexanes to
afford 185 mg (0.58 mmol, 58%) of syn adduct as a clear oil and 63 mg (0.20 mmol, 20%)
of the anti adduct from the asymmetric addition (Table 39). Syn adduct: 'H-NMR
(300MHz), CDC], 6 = 1.00 (t, J = 7.5 Hz, 3H), 1.94-2.04 (m, 2H), 4.62-4.70 (m, 1H),
4.82-4.91 (m, 2H), 4.95-5.04 (m, 1H), 7.11-7.21 (m, 2H), 7.29 (td, J = 7.5, 1.2 Hz, 1H),
7.61 (dd, J = 8.1, 1.5 Hz, 1H); l3C-NMR (300MHz), CDC], 6 = 10.4, 24.0, 44.8, 75.3,

90.0, 128.3, 130.4, 133.4, 134.]; MS (FAB'): m/z (%): 317 [M-H]' (98, S'Br), 315 [M-H]'
202

(97, 79Br), 268 (100); IR (neat) 3451(H20), 2112, 1653, 1558, 1375, 1010; HPLC analysis:
Chiralpac OD—H, hexane/2-PrOH 70/30, 0.5 mllmin, t, = 11.8 min (major), tr = 38.5 min
(minor), 92% ee, (1”,, = -6.8 (0:10, dichloromethane). For the anti adduct 0:25., = +6.6
(c=1 .0, dichloromethane).

The non-asymmetric addition (Table 28) afforded 191 mg (60%) of syn-80i adduct

and 19 mg (0.06 mmol, 6%) of the anti adduct.

2-Phenyl-1 ,3-dinitropentane 80j

N02
N02

enamem

Prepared following the general procedure described above, using 50 mg (0.33
mmol) of nitrostyrene. Purified with silica gel column using 20% EtOAc/hexanes to afford
63 mg (0.23 mmol, 70%) of syn adduct as a white solid and 20 mg (0.08 mmol, 8%) of the
anti adduct. Syn adduct: 'H-NMR (300MHz), CDC], 6 = 0.94 (t, J = 7.2Hz, 3H), 1.34 (tq,
J = 7.2 Hz. 2H), 1.74-1.68 (m, 1H),2.00—1.91 (m, 1H), 4.00 (t, J = 6.9 Hz, 1H), 4.92-4.72
(m, 3H), 7.15-7.11 (m, 2H), 7.34-7.31 (m, 3H) ppm; 13C-NMR (300MHz), CDC], 6 =
13.3, 19.1, 32.7, 46.7, 76.2, 89.3, 127.8, 129.0, 129.2, 133.7 ppm; HPLC analysis:
Chiralpac OJ-H, hexane/Z-PrOH 80/20, 1.5 mllmin, tr = 26.2 min (major), tr = 31.9 min
(minor), 91% cc, (1251, = --(c=1.0, dichloromethane); The data agreed with the reported

literature“.

203

Thioureas 47 and 76-78 were prepared according to literature procedures“c and

their spectroscopic data matched those of the literature.

Thiourea 79.

CF3 CF3

©:NH2 scr‘1/<j\cr:3 guinflcfit
THF

NH2 100% ”Tl/ti

S

 

CF3

Synthesized using the same procedure as for catalysts 52, 55 (Chapter 4). 162 mg
(1.50 mmol) o-phenylene diamine and 0.54 mL (813 mg, 3.00 mmol) 3,5-
trifluoromethylphenyl isothiocyanate were used, to afford 79 as a red solid in 100% yield
(975 mg, 1.50 mmol); 'H-NMR (300MHz, CD,CN) 6 = 7.60 (s, 2H), 7.60-7.48 (m, 4H)
8.13 (s, 4H), 8.65 (s, 2H), 8.74 (s, 2H); IR (neat): 3240, 2993, 1705, 1539, 1471, 1381,
1280, 1180, 1136, 1045, 983, 889, 700, 682; R, = 0.42 (50% EtOAc in hexanes), mp =
176-178°C; MS (FAB"): m/z (%): 651 (50) [M+H]*, 617 (20), 378 (20), 362 (24), 346

(100), 272 (24), 151(44), 119 (17), 109 (26).

General experimental procedure for the rate acceleration study using
thioureas 47 , 76, 77 (Scheme 26). trans-B-Nitrostyrene (20 mg, 0.13 mmoles) and the
appropriate thiourea (10 mol.%) were dissolved in 1 mL CDC], at r.t. and l-nitropropane

was added (0.060 mL, 5 eq.). The mixture was put in an NMR tube and was let stand for a

204

2h period, during which several lH-NMR spectras were obtained. The yield of the product
80a was determined by integration of the quartet of the product at 4.02 ppm and the

aromatic peaks of the thiourea catalysts, that were used as internal standards.

General procedure for the study of the non-asymmetric addition of 1-
nitropropane to nitrostyrene using thiouras 47 , 76-79 (Table 26). trans-B-Nitrostyrene
(100 mg, 0.66 mmoles) and thiourea 47 were put in a flame-dried flask and dissolved in 2
mL toluene. The appropriate nitroalkane was added (10 eq.) followed by triethylamine (10
mol%, 0.010 ml) and the mixture was stirred overnight at r.t. followed by evaporation of
the volatiles. The crude mixture was subjected to column chromatography purification
(20% ethyl acetate in hexanes) and the product 80a was obtained as a mixture of
diastereomers. The diastereomeric ratio of the product was determined by injection of the
crude reaction mixture (passed through a pippete with silica gel) to HPLC, using Chiracel
OJ—H, hexane/2-PrOH 80/20, 1.5 mL/min. Retention times of syn adduct: 31.5 min, 44.4

min. Retention times of the anti adduct: 20.0 min, 26.4 min.

General procedure for the optimization of the non-asymmetric addition of 1-
nitropropane to nitrostyrene using thiourea 47 (Table 27). N itrostyrene and thiourea
47 were put in a flame-dried flask and dissolved in various solvents. The appropriate
nitroalkane was added followed by triethylamine and the mixture was stirred overnight at
r.t. followed addition of 50 mol.% triphenylmethane (internal standard) and evaporation of
the volatiles. The yield was determined from the crude 1H-NMR spectrum by integration

of the quartet of the product at 4.02 ppm and the singlet of triphenylmethane at 5.6 ppm.

205

The diastereomeric ratio was determined by injection of the crude reaction mixture
(passed through a pippete with silica gel) to HPLC, using Chiracel OJ-H, hexane/2-PrOH
80/20, 1.5 mL/min. Retention times of syn adduct: 31.5 min, 44.4 min. Retention times of

the anti adduct: 20.0 min, 26.4 min.

For the rest of the tables in Chapter 5, the experimental procedures were the same
as for the ones described before for the optimum conditions, with slight variations that are

clarified in each table.

Preparation of catalyst 82.
Synthesis of compound 84.

szdbaa

DPPP

NaOtBu
NH2 85(1eq.) NH2
,NH ,N N
" 2 Toluene "H l \

80°C /

60% .
(W51 (FD-84 / N \

 

Compound 51 (1.077 g, 3.79 mmol) was put in a Schlenck flask along with
Pd2dba, (262 mg, 0.28 mmol), diphenylphosphinopropane (DPPP) (238 mg, 0.57 mmol),
NaOtBu (549 mg, 5.71 mmol). Toluene (30 mL) and 1 equivalent of 2-chloro-4-
dimethylaminopyridine 85 (595 mg, 3.79 mmol) were added and the mixture was freeze-
thawed twice. The flask was then put in an oil bath at 80°C and stirred at that temperature
for 36h. Filtration of the reaction mixture through Celite and evaporation of toluene under
reduced pressure followed. The dark residue was extracted with dichloromethane and
water and the dichloromethane layer was dried with MgSO4. The crude product was

206

purified by column chromatography (SiO,) using 50% ethylacetate in hexanes to afford
982 mg 84 as a yellow solid (2.27 mmol, 60%). lH-NMR (300MHz), CDC], 6 =2.88 (s,
6H), 3.69 (br s, 2H), 5.87 (d, J = 1.2 Hz, 1H), 6.07 (dd, J = 6.0, 0.9 Hz, 1H), 6.24 (br s,
1H), 7.44-7.01 (m, 7H), 7.85-7.75 (m, 3H), 7.91 (d, J = 9.3Hz, 1H), 8.24 (d, J = 9.0 Hz,
1H); 1 exchangeable H did not show up. I3C-NMR (300MHz), CDC], 6 = 39.2, 91.1,
101.2, 111.9, 118.3, 119.1, 120.4, 122.3, 123.8, 123.9, 124.9, 126.6, 126.8, 128.0, 128.1,
128.3, 128.8, 129.7, 130.0, 133.4, 133.9, 138.1, 142.7, 147.8, 155.9, 156.1 ppm; R,= 0.36
(5% Et,N/EtOAc); MS (FAB+): m/z (%): 405 (100) [M+H]", 387 (40), 267 (18); IR (neat)

3389, 3194, 3053, 2924, 1607, 1487, 1290; (1251,: +759 (c=1.0, dichloromethane).

Catalyst 82:

 

_,.N N =
com: gist“:
/ 60%
/ N\ /N \
(FD-34 (”1'32

In a typical procedure, 982 mg (2.43 mmol) of compound 84 was put in a round
bottom flask and dissolved in THF (10 mL). 3,5-Bis(trifluoromethyl)
phenylisothiocyanate (2.43 mmol, 0.443 mL) was added and the mixture was stirred
overnight at room temperature. The solvent was evaporated under reduced pressure and
the crude mixture was subjected to silica gel column chromatography purification (50%
EtOAc in hexanes followed by 10% MeOH in dichloromethane) to afford 986 mg (60%)

of the catalyst 82 as a yellow solid, mp. = 130°C; R, = 0.47 (10%

207

MeOH/dichloromethane); 'H-NMR (300MHz), CDC], 6 = 2.82 (s, 6H), 5.89 (d, J = 5.7
Hz, 1H), 6.00 (s, 1H), 6.96 (d, J = 8.7 Hz, 1H), 7.40-7.18 (m, 4H), 7.54-7.49 (m, 3H), 7.63
(br s, 2H), 7.72 (d,J = 8.4 Hz, 1H), 7.80 (d, J: 9.0 Hz, 1H), 7.88 (d,J = 8.1 Hz, 1H), 7.93
(d, J = 8.7 Hz, 2H), 8.01 (d, J = 8.7 Hz, 1H); 3 exchangeable H’s did not show up. ”C-
NMR (300MHz), CDC],6 = 39.1, 91.7, 101.6, 117.6, 117.7, 121.2, 121.4, 122.6, 124.7,
124.8, 124.9, 125.5, 125.9, 126.4, 127.1, 127.33, 128.3, 128.34, 128.7, 129.3, 129.5,
130.3, 130.9, 131.4, 132.5, 133.2, 133.5, 134.2, 137.9, 140.3, 146.4, 155.6, 156.], 179.8
ppm; IR (neat) 3308, 1610, 1489, 1385, 1277, 1132 MS (FAB*): m/z (%): 676 (100)

[M+H]’, 642 (40), 447 (30), 388 (10), 267 (7); 035,, = +129.9 (c=1.0, dichloromethane).

208

Experimental Procedures for Chapter 6

Preparation of VANOL triflate 95

TfZOrpy 0
pr, 0H Ph on

Ph ] OH DCM Ph 0 OTf
(rac)-116 O 95 0

In a typical experiment, 150 mg (0.34 mmol) of (rac)-116 were put in a flame-

 

dried flask and were dissolved in 3 mL of dichloromethane. Dry pyridine was added (82.9
uL, 1.03 mmol) followed by triflic anhydride (126 uL, 0.75 mmol). The mixture was
stirred overnight at room temperature and then washed with 10 mL 2N HCl, 4 mL
saturated NaHCO, and 4 mL water. The organic layer was filtered through silica gel and
the silica was washed with dichloromethane/EtOAc (10/1). The volatiles were evaporated
to afford 220 mg (0.30 mmol) 95 as a yellow solid (91%), mp. = 228-230°C; 1H-NMR
(300 MHz), CDC], 6 = 6.57 (d, J = 7.8 Hz, 2H), 6.89 (t, J = 7.5 Hz, 2H), 7.06 (t, J = 7.2
Hz, 2H), 7.74-7.61 (m, 6H), 7.88 (d, J = 8.1 Hz, 4H), 8.29 (d, J = 8.4 Hz, 4H)ppm; 13C-
NMR (300 MHz), CDC],6 =115.8,120.1,121.8,124.3,126.5,126.8,127.9,128.3,128.5.
128.9, 130.4, 134.6, 139.0, 140.5, 144.7 ppm; IR (neat): 3057, 2928, 1414, 1217, 1134,
1010, 812, 765, 702; MS (FAB’): m/z (%): 702 (90) [M’], 570 (42), 436 (78), 402 (100),

154 (45), HRMS (FAB’): calcd for C,,,H20F,,O,S2 m/z 702.0605, meas 702.0602.

209

3-Phenyl-1 -naphthyl triflate 100

0%,-'00.“ T120 PY %Tf
80%
99 1 00

The procedure was identical to the one for preparing VANOL triflate 95 . In a

 

typical experiment 1.50 g VANOL (3.40 mmol) were employed affording 1.90 g (2.72
mmol) of 100 (80%) as a brown oil (turned to a solid at 4°C). lH—NMR (300 MHz), CDC],
6 = 7.96-7.93 (m, 1H), 7.44-7.39 (m, 1H), 8.07-8.05 (m, 2H), 7.52-7.47 (m, 2H), 7.71-
7.60 (m, 5H); l3C-NMR (300MHz), CDC], 6 = 117.5, 120.6, 120.9, 125.2, 125.9, 127.2,
127.71, 127.73, 128.1, 128.3, 129.1, 134.9, 138.4, 139.1, 145.9 ppm; IR (neat): 3063,
3036, 1637, 1599, 1496, 1421, 1213, 1142, 1062, 1010, 924, 879, 819, 783, 696, 650 cm’l
MS (FAB+): m/z (%): 352 (100) [M’], 219 (40), 154 (23), HRMS (FAB‘): calcd for

C,,H,,F,o,s m/z 352.0381, meas 352.0378.

3-Phenyl-l-(N-benzyl) naphthylamine 101

5°/oPd(OAc) O
O on 10% BINAP? NHBn
O + BnNH2 O
C82003
0 Toluene, O

1 00°C 101
87%

 

100

A flame-dried Schlenk flask was charged with 150 mg (0.42 mmol) 100, 4.7 mg
(0.021 mmol) Pd(OAc)2, 14.5 mg (0.023 mmoles) BINAP and 193.8 mg (0.59 mmol)

Cs,CO,. The solids were dissolved in 3 m1 toluene and 55.7 u] (0 .51 mmoles) BnNH2 was

added. The resulting mixture was deoxygenated by the freeze-thaw method three times

210

and put in an oil bath at 100°C. The mixture was stirred at that temperature for 17h and
then the volatiles were evaporated under reduced pressure. The crude product was purified
by silica gel column chromatography using 10% EtOAc in hexanes to afford 114 mg (0.36
mmol, 87%) of 101 as a yellow solid, m.p.=1 10-1 12°C; IH-NMR (300MHz), CDC], 6 =
4.55 (s, 2H), 4.75 (br s, 1H), 6.96 (s, 1H), 7.54-7.35 (m, 11H), 7.74 (d, J = 6.0 Hz, 2H),
7.84 (d, J = 6.9 Hz, 1H), 7.92 (d, J = 8.1 Hz, 1H). '3C-NMR (300 MHz), CDC],6 = 48.6,
104.4, 115.8, 119.8, 122.6, 124.7, 126.1, 127.1, 127.3, 127.4, 127.8, 128.64, 128.67,
128.7, 128.9, 134.5, 138.9, 139.3, 141.9, 143.5 ppm; IR (neat): 3445, 3061, 3030, 1626,
1593, 1581, 1525, 1495, 1412, 1257, 775, 698 cm"; MS (FAB’): m/z (%): 309 (100) [M’],
252 (75), 140 (70), 123 (50); HRMS (FAB’): calcd for m/z C,,H,9N 309.1517, meas

309.1515.

3-Pheny1-l-(N-allyl) naphthylamine 111

1 0°/o Pd (OAC)2

H
P“ 0“ 11% BINAP P“ ”A
+ Wm, H O

2.8 eq. Cszco,
Toluene

100 109 100°C 111

 

The procedure was the same as for the synthesis of 101, only more catalyst (10
mol%) was used. For 1g scale reaction (2.8 mmol 100) 416 mg (1.59 mmol, 57%) of 111
were obtained as a yellow oil after column chromatography purification on silica gel using
10% EtOAc/hexanes. lH-N1\'£R(3001v11-lz), CDC], 6 = 4.02 (d, J = 4.8 Hz, 2H), 4.54 (br s,
1H), 5.26 (d, J = 10.5 Hz, 1H), 5.41 (d, J = 16.8 Hz, 1H), 6.87 (s, 1H), 6.20-6.05 (m,
1H), 7.49-7.35 (m, 6H), 7.73-7.68 (m, 2H), 7.84 (t, J = 7.2 Hz, 2H). l3C-NMR (300MHz),

CDC], 6 = 46.6, 104.2, 115.5, 116.4, 119.4, 122.4, 124.4, 125.8, 126.8, 127.1, 128.3,

211

128.7, 134.2, 134.7, 139.0, 141.7, 143.]; IR (neat): 3443, 3061, 1626, 1583, 1525, 1496,
1414, 1235, 920, 839, 761, 698 crn'l ; MS (FAB’): m/z (%): 259 (100) [M’], 232(8), 219

(10), 191 (5); HRMS: calcd for C.,H,,N m/z 259.1361, meas 259.1363.

Benzaldehyde N-(3-phenyI-l-naphthyl) imine 113

Ph NHBn . O
O neat, aIr Ph NHBn Ph NV Ph
180°C Ph NHBn 1‘
3.5 h 0 0
101 O

96, 0% 113, 32%

Obtained as a major product from the attempted synthesis of 96 from 101 through
oxidative coupling under neat conditions. Compound 101 was heated at 180°C in an oil
bth for 6 h, after which time it was let to cool down. The product 113 was purified with
column chromtography on silica gel using 5% EtOAc/hexanes, and in this way 50 mg
(0.16 mmol, 32%) 113 were obtained 3 a white solid. 1H-NMR (300 MHz), CDC], 6 =
7.24 (s, 1H), 7.32 (d, J = 7.2 Hz, 1H), 7.50-7.40 (m, 7H), 7.68 (d, J = 7.2 Hz, 2H), 7.85-
7.82 (m, 2H), 7.96 (br s, 2H), 8.26 (d, J = 7.5 Hz, 1H), 8.54 (s, 1H); l3C-NMR (300MHz),
CDC], 6 = 112.5, 123.6, 123.9, 125.8, 126.9, 127.4, 127.5, 128.0, 128.9, 129.1, 131.6,
134.2, 136.4, 138.9, 141.1, 149.9, 160.7; two C’s not located; MS: m/z (%): 307 (100)

[M+], 230 (25), 202 (85).

212

Compound 118

0 0
Ph OH NaHSOa Ph OH

Ph OH Ph I NH2

116 118

 

VANOL 116 (2.00 g, 4.56 mmol) was put in a Parr reactor with 11.5 g (110 mmol)
NaJ-ISO, and 87 mL of 30% aq. NH, was added. The mixture was heated at 200°C and
stirred with a mechanical stirrer for 40h. The resulting mixture was cooled down and
extracted with dichloromethane, the volatiles were evaporated under reduced pressure and
the crude product was purified by column chromatography on silica gel using 10%
EtOAc/hexanes to afford 498 mg (0.91 mmol, 25%) of 118 as a light brown solid; m.p. =
220-224°C; lH-NMR (300MHz), CDC], 6 = 4.45 (br. s, 2H), 6.00 (br. s, 1H), 6.53 (d, J =
7.5 Hz, 2H), 6.62 (d, J = 7.5 Hz, 2H), 7.04-6.84 (m, 6H), 7.10 (s, 1H), 7.22 (s, 1H), 7.48-
7.44 (m, 4H), 7.73-7.68 (m, 2H), 7.87-7.84 (m, 1H), 8.34-8.26 (m, 1H). l3C-NMR (300
MHz), CDC], 6 = 116.3, 120.5, 121.2, 121.6, 122.1, 122.7, 123.0, 125.3, 125.4, 125.5,
126.3, 126.4, 126.8, 126.9, 127.2, 127.3, 127.6, 128.1, 128.4, 128.8, 128.9, 134.2, 140.6,
140.7, 140.9, 141.5, 149.7 ppm; one C not located; IR (neat): 3491, 3383, 3053, 3028,
1608, 1588, 1493, 1383,1265, 758, 738, 702 cm"; MS: m/z (%): 437 (100) [M’], 307 (10),

219 (20), 154 (40), 136 (22), HRMS: calcd for C,,H,,NO m/z 437.1779, meas 437.1777.

213

REFERENCES

1. a) Comprehensive Asymmetric Catalysis; Jacobsen, E. N., Pfaltz, A.,
Yamamoto, H., Eds.; Springer-Verlag: Heidelberg, 1999. b) Noyori, R. Asymmetric
Catalysis in Asymmetric in Organic Synthesis; John Wiley & Sons: New York, 1994. c)
Ojima, 1., Ed. Catalytic Asymmetric Synthesis, 2nd ed; Wiley-VCH: New York, 2000.

2. Bao, J.; Wulff, W. D.; Dominy, J. B.; Fumo, M. J.; Grant E. B.; Rob, A. C.;
Whitcomb, A. C.; Yeung, S. M.; Ostrander, R. L.; Rheingold, A. L. J. Am. Chem. Soc.
1996, 118, 3392.

3. a) Bao, J.; Wulff, W. D.; Rheingold, A. L. J. Am. Chem. Soc. 1993, 115, 3814.
b) Bao, J .; Wulff, W. D. Tetrahedron Lett. 1995, 36, 3321. c) Heller, D. P.; Goldberg, D.
R.; Wulff, W. D. J. Am. Chem. Soc. 1997, 119, 10551.

4. a) Antilla, J. C.; Wulff, W. D. J. Am. Chem. Soc. 1999, 121, 6403. b) Antilla, J.
C.; Wulff, W. D. Angew. Chem. Int. Ed. 2000, 39, 4518. c) Loncaric, C.; Wulff, W. D.
Org. Lett. 2001, 3, 3675.

5. Xue, S.; Yu, S.; Deng, Y.; Wulff, W. D. Angew. Chem. Int. Ed. 2001, 40, 2271.

6. For recent examples of the catalytic enantioselective Mannich reactions using
preformed metal enolates: a) Fujieda, H.; Kanai, M.; Kambara, T.; Iida, A.; Tomioka, K.
J. Am. Chem. Soc. 1997, 119, 2060; b) Ishitani, H.; Ueno, M.; Kobayashi, S. J. Am. Chem.
Soc. 1997, 119, 7153; c) Ferraris, D.; Young, B.; Cox, C.; Dudding, T.; Drury HI, W. J.;
Ryzhkov, L.; Taggi, A. E.; Lectka, T. J. Am. Chem. Soc. 2002, 124, 67; d) Kobayashi, S.;
Hamada, T.; Manabe, K. J. Am. Chem. Soc. 2002, 124, 5640; e) Wenzel, A. G.; Jacobsen,
E. N. J. Am. Chem. Soc. 2002, 124, 12964; f) Kobayashi, S.; Matsubara, R.; Nakamura,
Y.; Kitagawa, H.; Sugiura, M. J. Am. Chem. Soc. 2003, 125, 2507; g) Josephsohn, N. S.;
Snapper, M. L.; Hoveyda, A. H. J. Am. Chem. Soc. 2004, 126, 3734; h) Akiyama, T.; Itoh,
J.; Yokota, K.; Fuchibe, K. Angew. Chem. 2004, 116, 1592; Angew. Chem. Int. Ed. 2004,
43, 1566; i) Kobayashi, S.; Ueno, M.; Saito, S.; Mizuki, Y.; Ishitani, H.; Yamashita, Y.;
Proc. Natl. Acad. Sci. USA 2004, 101, 5476.

Recent examples of direct catalytic asymmetric Mannich reactions: a) Yamasaki,
S.; Iida, T.; Shibasaki, M. Tetrahedron 1999, 55, 8857; b) Juhl, K.; Gathergood, N.;
J¢rgensen, K. A. Angew. Chem. 2001, 113, 3083; Angew. Chem. Int. Ed. 2001, 40, 2995;
c) List, B.; Pojarliev, P.; Biller, W. T.; Martin, H. J. J. Am. Chem. Soc. 2002, 124, 827; d)
Cordova, A.; Notz, W.; Zhong, G.; Betancort, J. M.; Barbas III,C. F. J. Am. Chem. Soc.
2002, 124, 1842; e) Notz, W.; Tanaka, F.; Watanabe, S.; Chowdari, N. S.; Turner, J. M.;
Thayumanavan, R.; Barbas III, C. F. J. Org. Chem. 2003, 68, 9624; f) Itoh, T.; Yokoya,
M.; Miyauchi, K.; Nagata, K.; Ohsawa, A. Org. Lett. 2003, 5, 4301; g) Hayashi, Y.;
Tsuboi, W.; Ashimine, I.; Urushima, T.; Shoji, M.; Sakai, K. Angew. Chem. 2003, 115,
3805; Angew. Chem. Int. Ed. 2003, 42, 3677; h) Bemardi, L.; Gothelf, A. S.; Hazel], R.
G.; J¢rgensen, K. A. J. Org. Chem. 2003, 68, 2583; i) Trost, B. M.; Terrell, L. R. J. Am.
Chem. Soc. 2003, 125, 338; j) Matsunaga, S.; Yoshida, T.; Morimoto, H.; Kumagai, N.;

214

Shibasaki, M. J. Am. Chem. Soc. 2004, 126, 8777; k) Uraguchi, D.; Terada, M. J. Am.
Chem. Soc. 2004, 126, 5356; l) Zhuang, W.; Saaby, S.; Jorgensen, K. A. Angew. Chem.
2004, 116, 4576; Angew. Chem. Int. Ed. 2004, 43, 4476; m) Cobb, A. J. A.; Shaw, D. M.;
Ley, S. V. Synlett 2004, 558; n) Cordova, A. Chem. Eur. J. 2004, 10, 1987; see also, 0)
Cordova, A. Acc. Chem. Res. 2004, 37, 102.

7. For leading references, see: E. Juaristi, Enantioselective Synthesis of b- Amino
Acids, Wiley-VCH, New York, 1997.

8. Ishitani, H.; Ueno, M.; Kobayashi, S. J. Am. Chem. Soc. 1997, 119, 7153.

9. As was reported in previous work (ref. 5), the reaction of VAPOL with 0.5
equivalents of zirconium tetraisopropoxide in the presence of two equivalents of NMI
produces the clean formation of a single Cz-symmetrical species which is tentatively
assigned by 1H NMR spectroscopy as 8: IH- NMR (C6D5CD,): 6 1.60 (br s, 6 H), 4.37 (br
s, 2H), 5.57 (br s, 2 H), 5.87 (br s, 2 H), 6.79 - 6.84 (m, 12H), 6.98 - 7.01 (m, 8H), 7.05 (s,
4 H), 7.14 (d, 4H,J= 8.7 Hz), 7.19 (d,4H,J= 8.7 Hz), 7.22 (td, 4H, J: 8.1, 1.1 Hz), 7.41
(dd, 4H, J: 8.1,1.2 Hz), 7.84 (td, 4H,] = 81,15 Hz), 11.33 ((1, 4H, J = 8.4 Hz). The 'H-
NMR spectrum shows that a number of species are generated when a 1:1 stoichiometry of
VAPOL to zirconium is employed.

10. As reported with the BINOL - Zr catalysts, the induction falls off with a
catalyst prepared with a 1:1 stoichiometry of zirconium to VAPOL (to 60%ee for the
reaction at room temperature in toluene).

11. Sibi, M. P.; Dankwardt, J. W.; Snieckus, V. J. Org Chem. 1986, 51, 271.
12. Sibi, M. P.; Jalil Miah, M. J.; Snieckus, V. J. Org. Chem. 1984, 49, 737.
13. Snieckus, V. Chem. Rev. 1990, 90, 879.

14. Beak, P.; Brown, R. J. Org Chem. 1982, 47, 34.

15. Price, C. C.; Chapin, E. C.; Goldman, A.; Krebs, B.; Shafer, H. M. J. Am.
Chem. Soc. 1941, 63, 1857.

16. Zhang, Y.; Yeung, S.-M.; Wu, H.; Heller, D. P.; Wu, C.; Wulff, W. D. Org.
Lett. 2003, 5, 1813.

17. a) Sarma, D. N.; Sharma, R.P.; Tetrahedron Lett. 1985, 26, 371; b) Lloyd, D.
H.; Nichols, D. E.; J. Org. Chem. 1986, 51, 4294; c) Barrett, A. G. M.; Spilling, C. D.;
Tetrahedron Lett. 1988, 29, 5733; d) Kende, A. S.; Mendoza, J .S.; Tetrahedron Lett. 1991 ,
32, 1699; e) Sturgess, M. S.; Yarberry, D. J .; Tetrahedron Lett. 1993, 34, 4743; f) Adams,
H.; Anderson, J. C.; Peace, S.; Pennell, A. M. K.; J. Org. Chem. 1998, 63, 9932.
18. a) Pinnick, H. W.; Org. React. 1990, 38, 655; b) Matt, C.; Wagner, A.; Moiskowski,
C.; J. Org. Chem. 1997, 62, 234; c) R. Ballini, M. Petrini, Tetrahedron 2004, 60, 1017.

215

19. Lucet, D.; Le Gall, T.; Mioskowski, C. Angew. Chem. 1998, 110, 2724; Angew.
Chem. Int. Ed. 1998, 37, 2580.

20. Adams, H.; Anderson, J. C.; Peace, S.; Pennell, A. M. K. J. Org. Chem. 1998,
63, 9932.

21. Anderson, J. C.; Peace, S.; Pih, S. Synlett 2000, 850.

22. a) Yamada, K.; Harwood, S. J .; Groger, H.; Shibasaki, M. Angew. Chem. Int.
Ed. 1999, 38, 3504; b) Yamada, K.; Moll, G.; Shibasaki, M. Synlett 2001, 980; c)
Tsuritani, N.; Yamada, K.; Yoshikawa, N.; Shibasaki, M. Chem. Lett. 2002, 276.

23. a) Knudsen, K. R.; Risgaard, T.; Nishiwaki, N.; Gothelf, K. V.; Jorgensen, K.
A. J. Am. Chem. Soc. 2001, 123, 5843; b) Nishiwaki, N.; Knudsen, K. R.; Gothelf, K. V.;
Jorgensen, K. A. Angew. Chem. Int. Ed. 2001, 40, 2992.

24. Palomo, C.; Oiarbide, M.; Halder, R.; Laso, A.; Pez, R. L. Angew. Chem. Int.
Ed. 2006.45, 117.

25. For general reviews on asymmetric organocatalysis, see: a) Dalko, P. 1.;
Moisan, L. Angew. Chem, Int. Ed. 2001, 40, 3726; b) Dalko, P. I.; Moisan, L. Angew.
Chem. Int. Ed. 2004, 43, 5138; c) Seayad, J.; List, B. Org. Biomol. Chem. 2005, 3, 719.

26. Nugent, B.M.; Yoder, R, A.; Johnston, J. N.; J. Am. Chem. Soc. 2004, 126,
3418.

27. Palomo, C.; Oiarbide, M.; Laso, A.; Lopez, R., J. Am. Chem. Soc. 2005, 127,
17622.

28. Aromatic N-acyl imines can be obtained as isolable, relatively stable
compounds upon treatment with stoichiometric potassium carbonate. See: a) Kanazawa,
A. M.; Denis, J.-N.; Greene, A. E. J. Org. Chem. 1994, 59, 1238; b) Wenzel, A. G.;
Jacobsen, E. N. J. Am. Chem. Soc. 2002, 124, 12964.

29. Ballini, R.; Petrini, M. Tetrahedron Lett. 1999, 40, 4449.

30. Fini, F.; Sgarzani, V.; Pettersen, D.; Herrera, R. P.; Bemardi, L.; Ricci, A.,
Angew. Chem. Int. Ed. 2005, 44, 7975.

31. Jacobsen, B.; Taylor, M. S. Angew. Chem Int. Ed. 2006, 45, 1520.
32. Johnson, J. S.; Evans, D. A. Acc. Chem. Res. 2000, 33, 325.

33. Haushalter, K. A.; Lau, J.; Roberts, J. D. J. Am. Chem. Soc. 1996, 118, 8891.

216

34. Connon, S. Chem. Eur. J. 2006, 12, 5418.
35. Okino, T.; Nakamura, S.; Furukawa, T.; Takemoto, Y., Org. Lett. 2004, 6, 625.

36. a) Schreiner, P. R.; Wittkopp, A. Org. Lett., 2002, 4, 217; b) Wittkopp, A. ;
Schreiner, P. R. Chem-Eur. J ., 2003, 9, 407.

37. Xu, X.; Furukawa, T.; Okino, T.; Miyabe, H.; Takemoto, Y., Chem. Eur. J.
2006, 12, 466.

38. Yoon, T. P.; Jacobsen, E. N., Angew. Chem. Int. Ed. 2005, 44, 466.

39. Bemardi, L.; Fini, F.; Herrera, R. P.; Ricci, A.; Sgarzani, V., Tetrahedron
2006, 62, 375.

40. a) Wang, J.; Li, H.; Duan, W.; Zu, L.; Wang, W.; Org. Lett., 2005, 7, 4713; b)
Wang, J.; Li, H.; Yu, X.; Zu, L.; Wang, W, Org. Lett. 2005, 7, 4293; c) While we were
working on this project, Connon’s group reported on a Fridel-Crafts type reaction using
catalyst 52, yet affording only modest ee’s (12-50 %): Tetrahedron Lett. 2006, 47, 7037.

41. a) Sohtome, Y.; Tanatani, A.; Hashimoto, Y.; Nagasawa, K.; Tetrahedron
Letters, 2004, 45 , 5589. b) Atzrodt, J .; Beckert, R.; Darsen, A. Tetrahedron: Asymmetry
1997, 8, 2257. c) Wilcox, C. S.; Kim, B.; Romano, D.; Kuo, L. H.; Burt, A. L.; Curran, D.
P.; Tetrahedron Lett. 1995, 51 , 621.

42. Ballini, R.; Bosica, G.; Fiorini, D.; Palmieri, A.; Petrini, M. Chem. ReV. 2005,
105, 933.

43. Bemer, O. M.; Tedeschi, L.; Enders, D. Eur. J. Org. Chem. 2002, 1877.

44. a) Ballini, R. In Studies in Natural Products Chemistry; Atta-ur-Rahman, Ed.,
Elsevier: Amsterdam, 1997; Vol. 19, p 117. b) Lucet, D.; Sabelle, S.; Kostelitz, 0.; Le
Gall, T.; Mioskowski, C. Eur. J. Org. Chem. 1999, (10), 2583.

45. Lu, S. F.; Du, D. M.; Xu, J.; Zhang, S.W. J. Am. Chem. Soc. 2006, 128, 7418.

46. Escribano, F. C.; Alcantara, M. D.; Gomez-Sanchez, A. Tetrahedron Lett.
1988, 29, 6001.

47. Pham-Huu, D. P.; Petrusova, M.; BeMiller, J. N.; Petrus, L. Tetrahedron Lett.
1999, 40, 3053.

48. 001, T.; Takada, S.; Doda, K.; Maruoka, K.; Angew. Chem. Int. Ed. 2006, 45,
7606.

217

49. a) Lambert, A.; Piggot, H. A.; J. Chem Soc. 1947, 1489. b) Bahner, C. T.; Kite,
H. T.; J. Am. Chem. Soc. 1949, 71, 3597. c) Snyder, H. R.; Hamlin, W. E. Am. Chem. Soc.
1950, 72, 5082. d) Shoemaker, G. L.; Keovin, R. W. Am. Chem. Soc. 1954, 76, 6374.

50. Alcantara, M. D.; Escribano, F. C.; Gomez-Sanchez, A.; Dia’nez, M. J.;
Estrada, M. D.; Lopez-Castro, A.; Perez-Garrido, S. Synthesis 1996, 64.

51. a) Lu, 8. F.; Du, D. M.; Zhang, S. W.; Xu, J. Tetrahedron: Asymmetry 2004,
15, 3433. b) Du, D. M.; Lu, S. F.; Fang, T.; Xu, J.J. Org. Chem. 2005, 70, 3712.

52. Wang, J.; Hao, L.; Zu, L.; Jiang, W.; Wang, W. Adv. Synth. Catal. 2006, 348,
2047.

53. Takenaka, N.; Sarangthem, R. S.; Seerla, S. K. Org. Lett. 2007, 9, 2819. Gobel
et a]. have also reported the catalysis of Diels-Alder reaction with amidinium ions:
Schuster, T.; Kurtz, M.; Gobel, M. W. J. Org. Chem. 2000, 65, 1697.

54. For important reviews of palladium— and nickel-mediated aryl amination, see:
(a) Wolfe, J. P.; Wagaw, S.; Marcoux, J. F.; Buchwald, S. L. Acc. Chem. Res. 1998, 31,
805-818. (b) Hartwig, G. F. Acc. Chem. Res. 1998, 31, 852-860. (c) Hartwig, J. F. Angew.
Chem. Int. Ed. Engl. 1998, 37, 2046-2067. (d) Hartwig, J. F. Synlett 1997, 329-340. (e)
Yang, B. H.; Buchwald, S. L. J. Organomet. Chem. 1999, 576, 125-146.

55. Wolfe, J.P.; Tomori, H.; Sadighi, J.P.; Yin, J.; Buchwald, S. L. J. Org. Chem.
2000, 65,1158.

56. Louis, J.; Driver, M. S.; Hamman, B. C.; Hartwig, J. F. J. Org. Chem. 1997,
62. 1268.

57. Ahman, J .; Buchwald, S. L. Tetrahedron Letters, 1997, 38, 6363.

58. Vyskocil, S.; Smrcina, M.; Lorenc, M.; Tislerova, 1.; Brooks, R. D.;
Kulagowski, J .; Langer, V.; Farrugia, L. J .; Kocovsky, P. J. Org. Chem. 2001, 66, 1359.

59. a) Miyano, S.; Nawa, M.; Mori, A.; Hashimoto, S. Bull. Chem. Soc. Jpn. 1984,
57, 2171; b) Shine, H. J .; Gruszecka, E.; Subotkowski, W.; Brownawell, M.; San Fillipo,
J., Jr. J. Am. Chem. Soc. 1985, 107, 3218; c) Shine, H. J.; Trisler, J. C. J. Am. Chem. Soc.
1960, 82, 4054; d) Shine, H. J.; Huang, F.-T.; Snell, R. L. J. Org. Chem. 1961, 26, 380. e)
Banthorpe, D. V.; Hughes, B. D.; Ingold, C. J. Chem. Soc. 1962, 2418.

60. Bucherer, H. T. J. Prakt. Chem. [2] 1904, 69, 49.

61. Korber, K.; Tang, W.; Hu, X.; Zhang, X. Tetrahedron Letters 2002, 43, 7163.

62. Lim, Y. K.; Jung, J. W.; Lee, H.; Cho, C. G. J. Org. Chem. 2004,69, 5778.

218

63. Thompson, A. L. S.; Kabalka, G. W.; Akula, M. R.; Huffman, J. W. Synthesis
2005, 547.

64. Rowland, G. B.; Zhang, H.; Rowland, E. B.; Channamadhavuni, S.; Wang, Y.;
Antilla, J. C.; J. Am. Chem. Soc. 2005, 127, 15696.

219

MICHIGAN STATE UNIVERSITY LIBRARIES
lllllllllllllllllllllllllllllllllllllllllllllllllllll
I
3 1293 02956 5870