E.

.521
ram.
h

‘4
En

.3. Agto

5—.“ RUM...

.fio

 

Lunch an. ,

.2
.v‘évrd
: ,
n :34!
.fiiai {23.
.i. 3:;

n—«

“0'...

fig

Hannah?

v
n.

m:

‘t
.II.
b

m
V m
a

H..-J.I'.lolno¢..1.
. .I .v. ‘1‘

. .. . SUE
, 2i i.

‘1‘.....

“ I’li

‘2'; \l)‘ 1'

‘In'xt‘
1.15:! I ut‘l

 

 

This is to certify that the

dissertation entitled

ASYMMETRIC TRANSFORMATIONS OF C=O AND C=C DOUBLE BONDS
TO CHIRAL ALCOHOLS USING CHIRAL AUXILIARIES AND CATALYSTS

presented by

Gang Huang

has been acceptthowards fulfillment
of the requirements for

Ph . D . Chemistry

degree in

 

 

l

l {\
\OLKPL’ ‘ it fii \ (xxx-K"
Majoxzéfifessor ii

Date 107/1147?)

MSU is an Affirmative Action/Equal Opportunity Institution 0- 12771

 

LIBRARY
Michigan State
University

 

 

 

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

 

 

DATE DUE DATE DUE DATE DUE O

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

ASYMMETRIC TRANSFORMATIONS OF C=O AND C=C DOUBLE BONDS
TO CHIRAL ALCOHOLS USING CHIRAL AUXILIARIES AND CATALYSTS

By

Gang Huang

A THESIS
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of
DOCTOR OF PHILOSOPHY
Department of Chemistry

1999

 

ABSTRACT

ASYMMETRIC TRANSFORMATIONS OF C=O AND C=C DOUBLE BONDS
TO CHIRAL ALCOHOLS USING CHIRAL AUXILIARIES AND CATALYSTS

By

Gang Huang

The asymmetric transformation of C=C and C =0 double bond to chiral alcohols using
several readily available chiral molecules as chiral auxiliaries or phase-transfer catalysts
was investigated. In a key step, 2-OH of a D-glucose unit (chiral auxiliary) was used as
an intramolecular nucleophile for the oxymercuration of 2-alkenyl glucosides with high
to excellent stereochemical control. The absolute configuration of the new chiral center
was determined by which anomer (on or B) to start with. After removal of the auxiliary,
pure (R)- or (S)-alkane-l, Z-diols and other chiral intermediates were retrieved. The
chiral 1, 2-diols were also synthesized with good stereoselectivity via chiral reduction of
a-hydroxy ketones using D-glucose as the chiral auxiliary. The configuration of starting
anomer also determined the chirality of the major isomer. The complexation with
calcium (II) played an important role in the chiral induction. The use of hexadecyl
nicotinium iodide as a phase transfer catalyst for stereoselective reduction of 3-keto fatty
acids was also investigated, which only gave low enantioselectivity. The stereoselective
reduction of benzoyl formic acid to mandelic acid using nicotine as a complexing reagent
was also attempted and low to fair chiral induction was observed. In general, nicotine as
we employed it here is not a good chiral auxiliary for asymmetric reduction due to its
structural flexibility. The reductive cleavage of 3,4-dihydro-l(2H)-naphthalenespiro-2-

(5’-methyl-l ’,3’-dioxolan-4’-one) formed from (L)-lactic acid was achieved with good

stereoselectivity by using a combination of chlorotrimethylsilane, zinc borohydride and
lutidine. The optical purity of starting dioxolanone had no effect on the stereochemical
outcome of the product. The lactic acid unit was later removed successfully, but

racemization of l,2,3,4-tetrhydro-l-naphthol was observed. The investigation had shed

some lights on the synthesis of chiral alcohols using lactic acid as a chiral auxiliary.

To Mom and Dad

Thank you for everything

IOU"...
acwn
idliv
dlfii‘
and r
3d\ cr
lmpw
some

b€pfi

l alst
Bake
ibrii
310u

harp;

dig)
lllSig'
P351

Andi

ACKNOWLEDGMENTS

I am so glad that finally, I am finishing up my Ph.D. program here in MSU. It is a long
journey of hard work and I am so proud of myself. However, I certainly can not
accomplish it without the help of a number of people. First I would like to thank my
advisor, Dr. Rawle I. Hollingsworth, for your direction, patience and financial support
during the course of this work. This is really a unique lab where you will grow scienticly
and morally. I have learned how to be myself and how to deal with frustrations and
adversities which I will definitely encounter in the future. I have learned the very
importance of responsibility, honesty and fairness that are much more important than
some papers, patents, etc. Thank you, Rawle. You have made this lab the best place to

be prepared to face any obstacles in the future.

I also owe my great gratitude to the members of my guidance committee. Dr. Gregory
Baker, thank you for always being there to support me. Dr. Thomas Pinnavaia, thank you
for your honesty and frank opinion to re-switch my career path from MS to Ph.D.. It was
a tough call for me but I really appreciate your opinion. Dr. Stanley Crouch, I am just so

happy to have you in my committee and offered so much help along the way.

Also I would like to thank Dr. Maleczka and Dr. Reusch for their encouragement and
insightful discussions about synthesis problems. Thanks also go to Hollingsworth group,
past and present. To name a few, Yin Tang, Jianjun Wang, Ben Hummel, Wayne

Anderson, Jie Song, James Bradford and Hussen Mohammed. I really enjoyed the time

i: gt
of m}
pant;
prise.
{inst

Dips."

we spent together. Thank you so much for helping me going through some tough periods
of my life. It is only through those that you realize who your real friends are. I am
particularly grateful to Jie Song who has instilled much-needed strength and
perseverance in recent years. lie, I am very fortunate to know you and have you as my
close friend and colleague. It really means a lot. There are some friends in Chemistry

Department I would like to thank too, particularly Feng Geng and Baoyu Mi.

I always felt greatly indebted to my mom and brother, who have sacrificed dearly while I
am pursuing my Ph.D. here in USA. Mom, I will not be here without your guidance
through the years. I felt extremely guilty not being able to see you in the past six years
and I wanted to dedicate this thesis to you. Lastly, but definitely not the least, I want to
thank my beautiful wife Guangning and my lovely daughter Stephanie. Guangning,
thank you for your unconditional love, support and encouragement during these years.
Stephanie, what a joy you have brought to me. Thank you for giving me the “good
distraction” everyday and thank you for letting me view my favorite sports on TV! This

thesis can’t be done without you.

Thanks again,

Gang

vi

llSI

IJSI

LIST

lJSI

Cllt

lnno

Clix
TheS
(SH
0-61

CIII
Slere
as a I

TABLE OF CONTENTS

LIST OF FIGURES ................................................................................................. ix
LIST OF SCHEMES ............................................................................................... xi
LIST OF TABLES ................................................................................................... xiii
LIST OF ABBREVIATIONS ................................................................................. xiv
CHAPTER 1
Introduction .............................................................................................................. 1
References .......................................................................................................... 30
CHAPTER 2

The Stereoselective Conversion of 2-Alkenyl Alcohols to (R)— or
(S)-Alkane-1, 2-diols and Other Chiral Intermediates Using

D-Glucose as a Chiral Auxiliary ............................................................................. 36
Introduction ....................................................................................................... 37
Results and discussion ...................................................................................... 44
Experimental ..................................................................................................... 58
References .......................................................................................................... 68

CHAPTER 3

Stereoselective Reductions of Ketones Using D-Glucose

as a Chiral Auxiliary ................................................................................................ 69
Introduction .................................................. 70
Results and discussion ...................................................................................... 76
Experimental ..................................................................................................... 84
References .......................................................................................................... 94

CHAPTER 4

Stereoselective Reduction Using Nicotine and Its Derivative

as a Chiral Promoter ............................................................................................... 95
Introduction ....................................................................................................... 96
Results and discussion ...................................................................................... 105
Experimental ..................................................................................................... l 12
References .......................................................................................................... l 19

vii

(

 

CHAPTER 5
Stereoselective Reduction of 2,2-Disubstituted

S-Methyl-l,3-dioxolan-4-ones Formed from L-Lactic Acid ................................. 121
Introduction ....................................................................................................... 122
Results and discussion ...................................................................................... 125
Experimental .................................................. 133
References .......................................................................................................... 144

viii

 

 

 

fi;

fi;
II;
III

Hi

Fi«

Fi

Fit

n;

Fit

H.

n;
n,
n.

n,

LIST or FIGURES

Figure 1.1: Two enantiomers with drastically different effects ................................ 2

Figure 1.2: Some naturally and unnaturally occurring chiral compounds

synthesized from chiral alcohols ............................................................ 5
Figure 1.3: Structures of some phase transfer catalysts ............................................ 13
Figure 1.4: Structures of some cheap chiral molecules ............................................ 29
Figure 2.]: Ring formation to improve stereoselectivity .......................................... 40

Figure 2.2: lH-NMR and l3C-NMR of compound (2): 4,6-O-Benzylidene

- l ,2-O-[2-methyl-(R)- l ,2-ethanediyl]-a-D-glucopyranose .................... 46

Figure 2.3: lH-NMR and l3C-NMR of acetolysis intermediates .............................. 47
Figure 2.4: lH-NMR of crude products of oxymerCuration of allyl 4,6-0-

benzylidene-B-D-glucopyranoside ........................................................ 50

Figure 2.5: Comparison of steric hindrance in a- and B-isomers ............................. 52

Figure 2.6: 1H-NMR and l3C-NMR of compound ( l3): 4,6-O-Benzylidene-
1 ,2-O-[2-iodomethyl-(R)-l ,2-ethanediyl]-a-D-glucopyranose ............. 56

Figure 3.1: Working model for the stereochemical outcome of a-glycoside ........... 78

Figure 3.2: Working model to predict the stereochemical outcome
of B-glycoside ........................................................................................ 80

Figure 3.3: Some biologically active compounds containing C3 chiral synthons ....81
Figure 4.1: Structures of some phase transfer catalysts ............................................ 98
Figure 4.2: Structure of Nicotine .............................................................................. 101

Figure 4.3: Asymmetric reduction of a 3-keto fatty acid under
phase transfer catalytic condition ......................................................... 101

Figure 4.4: Reduction of benzoylformic acid with cyclodextrin
as the chiral promoter ............................................................................ 104

ix

 

Figure 4.5: Proposed complexation between benzoylformic acid and nicotine ....... 104

Figure 4.6: 1H-NMR spectra of nicotine and the nicotine/benzoylformic
acid complex .......................................................................................... 1 10

Figure 5.1: Structures of dioxolane and dioxolanone ............................................... 126

Figure 5.2: Possible pathways for reduction of dioxolanone using
Lewis acid and reducing agent ............................................................... 129

Sch
Sch
Sch
Sch
Sch
Sch

Sch

Sch
Sch
Sch
Sch
Sch
Sch
Sch

SCh.

SChl
SChc
SchE

Sche

Sche

LIST OF SCHEMES

Scheme 1.1: General approaches used in asymmetric synthesis ............................... 3

Scheme 1.2: Reduction of ketones with LiA1H(OR’)3-type reagents ....................... 11
Scheme 1.3: 1,2-Induction for stereoselective reduction .......................................... 16
Scheme 1.4: Reductions of chiral ketosulfoxides .................................................. 18
Scheme 1.5: Reductive cleavage of chiral acetal .................................... ' .................. 18
Scheme 1.6: An example demonstrating Prelog’s rule ............................................. 19

Scheme 1.7:

Scheme 1.8:

Scheme 1.9:

Scheme 1.10:
Scheme 1.11:
Scheme 1.12:
Scheme 1.13:
Scheme 1.14:

Scheme 1.15:

Scheme 1.16:
Scheme 1.17:
Scheme 2.1:

Scheme 2.2:

Scheme 2.3:

Synthesis of both enantiomers of mandelic acid
using a chiral auxiliary ....................................................................... l9

Reductions of oc-ketoesters with chiral alcohol auxiliaries ................ 19

Reduction of ketones with chiral acetal or oxathine in the vicinity 19

Diastereoselective reduction via 1,5-induction .................................. 20
Sharpless epoxidation of primary allyl alcohols ................................ 21
Control of absolute configuration of Sharpless epoxidation .............. 23
Asymmetric epoxidation using a chiral auxiliary ............................... 23
Chiral manganese complex for asymmetric epoxidation ................... 24

Prediction of the stereochemical outcome of Sharpless asymmetric
dihydroxylation of olefins ................................................................... 26

Sharpless asymmetric dihydroxylation .............................................. 27
Asymmetric oc-hydroxylation using SAMP as a chiral auxiliary ....... 28
Cyclopropanation using D-glucose as a chiral auxiliary .................... 38

Stereoselective oxidation of C=C double bond using
carbohydrates as chiral auxiliaries ..................................................... 38

Intramolecular iodolactonization showing high diastereoselectivity .40

xi

 

(I)

(1')

([7

In

In

I'l'I

In

In

In

In

Scheme 2.4: Experimental design for the synthesis of chiral diols ......................... 42
Scheme 2.5: Actual scheme used to make chiral diols ........................................... 42
Scheme 2.6: Mechanism of oxidative cleavage of chiral auxiliary ........................ 49
Scheme 2.7: Reaction scheme to synthesize (S)-propane-l,2-diol ......................... 49
Scheme 2.8: Synthesis of oc- and B-cis-2-pentenylglucosides ................................ 54
Scheme 2.9: Synthesis of (R)-3-hydroxybutyrolactone from

a 3-iodo substituted intermediate ........................................................ 55
Scheme 3.1: Reductions of carbonyl group using carbohydrate derivatives

as chiral auxiliaries ............................................................................ 71
Scheme 3.2: Stereselective reduction by internal and external hydride attack ....... 74
Scheme 3.3: Stereoselective reduction of 2-oxopropyl-a-D-glucopyranoside ....... 76
Scheme 3.4: Stereoselective reduction of benzoylmethyl-a-D-glucopyranoside...79
Scheme 3.5: Application of stereoselective reduction using D-glucose

as a chiral auxiliary ............................................................................. 83
Scheme 4.1: Asymmetric epoxidation catalyzed by nicotine ................................. 99
Scheme 4.2: Reduction of 3-oxo-tetradecanoic acid under

phase transfer catalytic condition ...................................................... 106
Scheme 5.1: Synthesis of chiral alcohols using mandelic acid as

a chiral auxiliary ................................................................................ 122
Scheme 5.2: Formation of 3,4-dihydro-l(2H)-naphthalenespiro-2-(5’-methyl-

1 ’,3 ’-dioxolan-4’-one) ....................................................................... 126
Scheme 5.3: Stereoselective reduction of dioxolanone ........................................... 126
Scheme 5.4: Oxonium formation to explain the stereochemical outcome .............. 127
Scheme 5.5: Attempted removal of auxiliary ......................................................... 132

xii

Iablc

Tablc

Table
Table
Table

Table

Table

Table

Table:

LIST OF TABLES

Table 1.1: Reduction of R.C(=O)R2 to R1C*H(OH)R2 using chiral
catalysts or reagents ................................................................................. 6

Table 1.2: Reduction of R.C(=O)R2 to R.C*H(OH)R2 using chiral metal

hydrides and boranes ............................................................................... 10
Table 1.3: Stereoselective reductions of ketones using chiral alkyl boranes ............ 15
Table 2.1: Synthesis of chiral diols using D-glucose as a chiral auxiliary ............... 54
Table 3.1: Diastereoselectivity obtained under different conditions ........................ 77

Table 4.1: Chiral induction of 3-oxo-tetradecanoic acid under various
phase transfer catalytic conditions ........................................................... 106

Table 4.2: Chiral reduction of benzoylformic acid complexed with
one equivalent of nicotine ........................................................................ 108

Table 5.1: Reduction of dioxolanone under different conditions ............................. 128

Table 5.2: Stereochemical outcome starting from substrates of different
optical purity ............................................................................................ 130

xiii

 

Ti

 

Ac

AQO

   

BICHI _
thlf
Bhuil
Bn
BPPFIH
Cat
CD

C(H)

 

lit-XIIR

LIST OF ABBREVIATIONS

Ac acetyl
AczO acetic anhydride
BICHEP (R)-2,2 ’-bis(dicyclohexylphosphino)-6,6’-dimethyl- 1 ,1 ’-biphenyl
BINAP (R)-2,2’-bis(diphenylphosphino)-l ’,l ’-binaphthyl
BINOL (R)-2,2’-dihydroxy- l ,l ’-binaphthyl .
Bn benzyl
BPPFOH (R)- l -[(S)-l ’,2-bis(diphenylphosphino)ferrocenyl]ethanol
Cat. catalyst
CD cyclodextrin
COD 1,5-cyclooctadiene
l3C-NMR carbon (”C) NMR
(1 doublet
d.e. diastereomeric excess
DHQ dihydroquinine
DHQN dihydroquinidine
DIBAH diisobutylaluminum hydride
DIBAL diisobutylaluminum hydride
DI OP (R,R)—2,3-o-isopropylidene-2,3-dihydroxy-l ,4-bis(diphenylphosphino)—
butane

DI PT diisopropyl tartrate

5 chemical shift

DMF N,N-dimethylformamide

xiv

56

sq.

E10
Hll
H0‘
H1

in It.
HS
IAH
IDA

III?

SICP
\lcO
lITP.
MTP,
NI-Io
.\'.\-1R

N0E3

PPTS

2D-NMR
e.e.

eq.

Eth
EtOH
HMPA
HOMO
Hz

in vacuo
'H-NMR
LAH
LDA
LUMO
m
MCPBA
MeOH
MTPA
MTPACl
NMO
NMR
NOESY
Np

PPTS

two dimensional NMR

enantiomeric excess

equivalent

diethyl ether

ethanol

hexamethylphosphoric amide

highest occupied molecular orbital

hertz

in vacuum oven

proton NMR

lithium aluminum hydride

lithium diisopropylamide

lowest unoccupied molecular orbital

multiplet

metachloroperbenzoic acid

methanol
methoxy-oc-(trifluoromethyl)phenylacetic acid
methoxy-oc-(trifluoromethyl)phenylacetic chloride
N-methylmorpholine N-oxide

nuclear magnetic resonance

nuclear overhauser and exchange spectroscopy
naphthyl

pyridinium paratoluenesulfonate

XV

Pic

SA

TB

TH

If

T_\

i-PrOH

PTC

SAMP

TBHP

THF

TFA

TMSCI

isopropanol

phase transfer catalyst
pyridine

quartet

room temperature
singlet

(S)-(-)- l -amino—2-(methoxymethyl)pyrrolidine
triplet

tert-butyl hydroperoxide
tetrahydrofuran ' '
trifloroacetic acid
trimethylchlorosilane

halogen atom

xvi

CHAPTER 1

Introduction

 

 

Cl

1:

Introduction

The importance of chirality in living systems is underscored by the fact that nearly all
natural products are chiral and that their physiological and pharmacological properties
are thought to depend upon their recognition by chiral receptors. These receptors interact
only with molecules of the proper absolute configuration. Thus, one enantiomer may act
as a very effective therapeutic drug whereas the other enantiomer is highly toxic. One
example is that the (R)-enantiomer of thalidomide (Figure 1.1) exhibited desirable
analgetic properties but the (S)-enantiomer caused fetal malformations or deaths if

administered to pregnant woman. Following this tragedy, the marketing regulations for

o g V0
N
N o H
0Q
' N
o

O
O
N
O
(S)-thalidomide (R)-thalidomide
teratogenetic anal getic

Figure 1.1
Two enantiomers with drastically different effects
synthetic drugs have become significantly more stringent. In order to commercialize a
racemic mixture, the activity of each enantiomer of the racemate must be carefully

evaluated and the drug is approved only if it can be shown that both enantiomers have

met
3??
Ilia
acc

tcr

similar potencies or that the nonpotent enantiomer is completely devoid of any side

effects. The marketing of single enantiomers is becoming the norm.

One of the important ways to obtain optically pure compounds is asymmetric synthesis.
The general approaches of asymmetric synthesis to create one or more chiral center(s)
include the use of chiral substrates (Scheme 1.1a), chiral auxiliaries (Scheme 1.1b),

chiral reagents (Scheme 1.1c) and chiral catalysts (Scheme 1.1d). The last three

a. s*+ R ——> P*
b. s + Ai’ __* S_A* _R___> P*_A* _:A*_> P*
c. s + R* ————> P*

Cat.*

d.S+R P*

(8* - Chiral Substrate; S — Achiral §ubstrate; R - Reagent; R* - Achiral Reagent;
A*-Chiral Auxiliary; P* - Chiral Product, A — Auxiliary; Cat.* - Chiral Catalyst)

Scheme 1.1
General approaches used in asymmetric synthesis
methods are currently being actively developed because of their flexibility and broader
applicability. The requirements for a useful asymmetric synthesis include high (greater
than 90%) regio-, diastereo- and enantioselectivities. Also important are the expense and
accessibility of the reagents that are involved, the conditions of the reaction (solvent,
temperature, pressure, etc.) and the ease of work-up and purification. The
stereoselectivity results from the energy difference between two (or more) diastereomeric
transition states. The success of using chiral substrates is based upon the availability of
the starting materials (chiral pool) and the ease of transformation of starting materials to

products. The use of chiral auxiliaries has the disadvantage of introducing two more

digs:
ffisii
in."
erit‘

0633

SEOL

steps for their attachment and removal, but its advantage lies in the formation of a pair of
diastereomers which are physically separable in many cases so that 100% pure
enantiomer can be obtained. Generally, the use of chiral auxiliaries requires that: l. The
transformation S-A* to P*-A* must be highly stereoselective (d.e. 3 90%), and the
purification of P*-A* should be straightforward. 2. The cleavage of P*-A* to P* should
occur under mild conditions, and no racemization at the newly introduced stereocenter
should occur. The approach of using chiral reagents and catalysts rather than chiral
starting materials or chiral auxiliaries affords greater flexibility. This makes the choice
of starting material much wider. The use of catalytic amounts of chiral catalysts also
reduces the consumption of chiral compounds, which can be very expensive. However,
chiral enrichment (e.g. via recrystallization) is often required to obtain the single isomer.
So far, there is no universal method for a highly stereoselective transformation. Different

methods have to be applied for each case.

A number of biologically significant chiral compounds are synthesized from optically
active functionalized alcohols. Figure 1.2 shows some naturally occurring and
unnaturally occurring chiral compounds that have been synthesized from alcoholic blocks
in recent years. Two major approaches to the synthesis of chiral alcohols are the
asymmetric reduction of C=O double bonds and the asymmetric oxidation of C=C double
bonds. Catalytic hydrogenation remains one of the most important reactions for the first
approach (Table 1.1). Both homogeneous and heterogeneous catalysts have been
developed. The homogeneous catalytic hydrogenation of prochiral ketones can be

carried out direct asymmetric hydrogenation across the double bond or by derivatization

TI

OH
°“ 3

0
j“ °"
H H
I
I
0" s
\/\ ,.
NH3
/ 5" °
OH C02"
0
H CaH17 O
o o o
O HO
H 0
OH
HO
WV 0
O

OH

O

@N/DIOM/vfll/OH

ml

Some naturally occurring and unnaturally occurring chiral compounds
synthesized from chiral alcohols

Figure 1.2

\..

C

P! FE; \mq- \--\. \I a:

 

Table 1.1

Reduction of R1C(=O)R2 to R.C*H(OH)R2
using chiral catalysts or reagents

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

¥

 

 

 

Tartaric acid, NaBr

 

R1 R2 e.e. (°/o) Reagent Ref.
Ph t-Bu 84 IrI[(S)-PPEI](COD) 1
Ph Me 43 ‘ (R,S)-BPPFOH-Ru 2
Ph Me 58 [erl(COD)]2-A 3
2-Naphthyl Me 63 [IrCl(COD)]2-A 3
n-Pr CHzCOzMe 98 RuzCl4[(S)-BINAP]2(NEt3) 4
n-Bu CHzCOzMe 97 RuzCl4[(S)-BINAP]2(NEt3) 4
n-C. 1H23 C H2C02Me 96 (R)-[Ru2Cl4(COD)2(MeCN)+BINAP 5
Me COzMe 87 (25,4S)-MCCPM-Ru 6
Ph CH2N(CH3)2 95 RuBr2[(S)-BINAP] 7
Me CHQOH 92 RuC12[(S)-BINAP] 7
Me C6H4-2-Br 92 RuBr2[(R)-BINAP] 8
Ph CHMeCOMe 99 RuC12[(S)-BINAP] 8
CHzO-n- CHzOCPh3 96 RuzCl4[(S)-BINAP]2.(NEt3) 9
C18H37
Ph CONH-t—Bu 88 Ru(OCOMe)2(BICHEP) 10
Ph Me 58 [Rh(COD)Cl]2,(+)-DIOP,l-NpPhSin 11
Me COOPr—n 85 [Rh(COD)Cl]2,(+)-DIOP, 1 -NpPhSiH2 12
CH3 CH2COOCH3 86 Raney Ni modified with (2R,3R)- 13
_ Tartaric acid, NaBr
CH3(CH2)3 CH2COOCH3 87 Raney Ni modified with (2R,3R)- 14

 

 

Table 1.1 (Cont.)

w

PPEI COD

©© P(C6HS)2
©© P(C6H5)2

A BINAP

© P(C"C6Hi 02 O

BICHEP mo"

 

BPPFOH
(C-CoHI 021’
PP
1.”
CONHMC
MCCPM

P(C6H5)2

O P(C'C6HI l)2 ><O\C
P(C6H5)2

 

substra
manna
mm
BIXAI
NCCOr
355mm
lfidUCl
ha\ln§
carbor
Ofprg
gfincn
B‘ C 2
all‘brc
lihlcl
of CI
inTPCI

NaBr

The

gene;

of the ketone prior to asymmetric hydrogenation. Chiral biphosphine ligands, like (R)-
BINAP, plays an essential role in such reactions. Hydrogenation of ketones with
phosphine-Rh complexes is slower and less stereoselective than the reaction with olefinic
substrates. So far, no Rh catalyst system can produce high enantioselectivity in a general
manner. In the past few years ruthenium has become a strong competitor to rhodium in
enantioselective ketone reductions. For example, the asymmetric reductions catalyzed by
BINAP-Ru(II) find a wide generality by converting a range of functionalized ketones to
the corresponding secondary alcohols with high enantiomeric purity [15]. However, the
asymmetric hydrogenation of simple ketones has not been very successful and chiral
induction is low in most cases. Functionalized ketones or carbonyl-containing substrates
having additional carbonyl, olefin, hydroxyl or ether functionalities oc- or [3- to the
carbonyl group generally afford high optical yields in the asymmetric catalytic reduction
of prochiral carbonyl compounds (75-98% e.e.). This is probably due to their ability to
generate simultaneous coordination of carbonyl oxygen and the functionality at the a- or
B- C atom with the metal atom. The formation of a five or six-membered chelate usually
affords a high degree of stereodifferentiation. Asymmetric hydrosilylation of ketones,
which produces the same result as hydrogenation, provides another way for the synthesis
of chiral alcohols. As for heterogeneous catalytic hydrogenation, one of the most
important catalysts developed is Raney nickel modified with (2R, 3R) tartaric acid and

NaBr. The substrates are mainly B-dicarbonyl compounds.

The non-catalytic reductions of saturated and unsaturated carbonyl compounds are

generally accomplished by using chiral tetracoordinated metal-hydride complexes (e.g.

3&0\}

mCOUI

0\1[;T

h} an

lhcrci‘
the c:
smith
Duni:
the re
ihoug

Step i

The
8min
3“ c1
3mm

hate

alkoxyaluminum hydrides or alkoxyborane hydrides bearing chiral substituents), chiral
tricoordinated alanes or boranes, or with achiral boranes in the presence of
oxazaborolidines (Table 1.2). Nucleophilic addition (H') to the C=O group ofien occurs
by an electrophilic assistance mechanism. Electrophilic assistance takes place when the
nonbonding electrons on a double-bonded oxygen atom coordinates with a Lewis acid,
thereby activating the double bond toward nucleophilic attack. The Lewis acid can be
the cation associated with the nucleophile or it. can be another compound added in
stoichiometric or catalytic quantities. Nucleophilic attack takes place along the Burgi-
Dunitz trajectory (109°). This minimizes the unfavorable LUMO-HOMO overlap and
the repulsive interaction between the nucleophile and the oxygen lone pairs. It is usually
thought that asymmetric induction takes place during the nucleophilic hydride transfer
step itself. With tricoordinated Lewis acidic reagents, the asymmetric induction can also
take place prior to the hydride transfer during the interaction of the reagent with the
carbonyl lone pair. Moreover, when the substrate bears a substituent prone to chelate
formation with either a cation associated with the reducing agent or with an added metal
salt, hydride transfer can take place on a rigid complex with both tricoordinated or

tetracoordinated reagents.

The various modifying reagents now used with LAH are chiral alcohols, amines and
amino alcohols, some of which are derived from chiral terpenes, carbohydrates, etc..
Several glucose derivatives have been used to effect stereoselective reductions [23-30].
Studies on the asymmetric reduction of ketones with LAH-monosaccharide complexes

have revealed that a 1:1 LAH-sugar complex affords maximum selectivities. This has

Table 1.2

Reduction of R;C(=O)R2 to R1C*H(OH)R2
using chiral metal hydrides and boranes

 

 

 

 

 

 

 

 

 

 

 

R1 R2 e.e. (%) reducing agent Ref.
Ph (CH2)2N(CH3)2 77 (-)-menthol-LAH 16
Ph Me 95 (S)-BINAL-H 17
Ph COOH 65 NaBH4-(L)-tartaric acid 18
n—C 5H1 1 CC H 95 NB-enantrane 19
HsCz CCC2H5 96 NB-enantrane 19
Ph C2H5 78 NaBH4, bovine serum albumin 20
2-naphthyl Me 66 NaBH4, bovine serum albumin 20
Ph Me 97 B 21
Ph t-butyl 32 NaBH4/benzquuinium chloride 22

 

 

 

 

 

/—OCH2Ph

FPr Ph

B/D H
\ H Ph

NB-enantrane B

10

 

 

 

([11

(D

r)

r)

been attributed to the formation of a cyclic complex between LAH and the hydroxyl
groups. The front runner among LAH-diol complexes is based on 2,2’-dihydroxy-1,l’-
binaphthyl, BINOL, discovered by Noyori et al.[31,32]. Incorporation of methanol or
ethanol as an additional ligand is necessary and reduction of a number of substrates gave
e.e. over 90%. The exceptionally high stereoselectivity achieved has been explained
according to the pathway shown in Scheme 1.2. The reaction is initiated by the
complexation of the Lewis acidic Li+ ion to the oxygen atom, which activates the
carbonyl group. Hydride transfer then occurs from Al to the carbonyl carbon by way of a
six-membered ring transition state. Success was also achieved with the LAH-(-)—N-
methylephedrine complex if 3,5-dimethylphenol or N-ethylaniline were added [33-38].
Reagents formed from sodium borohydride modified with various monosaccharide
derivatives have also been employed for the asymmetric reduction of prochiral ketones

[39]. A number of ketones have been reduced using such derivatives, affording the

 

 

R'O OR'
R' \ I
\ I x
R _LL .. _ R 'NH(0R')3 9 i'
0:0 V—_ LI---O—C —-> g n
\ \ LI~~ ,I’C-R
R R ‘o/ \
— R —l
H\ ’R
——-> AI(OR')3 + /c\ ——-> LiIRZCHOAKOR'h]
LiO R
Scheme 1.2

Reduction of ketones with LiAlH(OR’)3-type reagents

corresponding optically active alcohols. Up to 39% e.e. was thus obtained in the case of

propiophenone, whereas for other ketones the optical yields of the alcohols were between

11

l.9-l8.8% e.e.. The optical yields in the monosaccharide-modified NaBH4 reduction of
ketones can be enhanced by using NaBH4 treated with a carboxylic acid such as acetic

acid prior to the addition of the monosaccharide derivative [40-42].

The insolubility of sodium borohydride in non-aqueous media renders it less useful for
asymmetric reduction carried out in organic solvents. One way to accomplish such
reductions is to replace the sodium ion with chiral ammonium or phosphonium ions
under phase transfer conditions [43-45]. Alternatively, crown ethers having two to four
chiral centers have also been employed in phase transfer catalyzed reductions [46].
Figure 1.3 shows the structures of some chiral phase transfer catalysts. These ligands
have been employed with NaBH.; for the asymmetric reduction of ketones [43]. Thus,
the asymmetric reduction of phenyl t-butyl ketone yielded the corresponding alcohol in
32% e.e. by using NaBH4 with (I) as a phase-transfer catalyst. In similar reactions, the
optical yields were generally lower and ranged from 1.1 to 13.7% e.e. employing the
other ligands (C-H). Studies on asymmetric reduction under phase transfer conditions
have led to the conclusion that conformationally rigid ligands result in faster reductions
and higher optical yields, as exemplified by the quininium salt (I). It has also been found
that the hydroxy group in the ligand should be in the B-position to the onium function.
This may be important for its interaction with the carbonyl group of the substrate and to
favor one of the diastereomeric transition states which leads to the carbinol. A multipoint
interaction between catalyst and the substrate is essential for optical induction. A perusal
of the literature so far has shown that the selectivity in PTC reaction is dependent on: (i)

the structure of the phase transfer catalyst, (ii) reaction conditions such as temperature,

12

t 9H3
PhCHz—Ct—NLCHfir'
H C12st

C.R
D.R

OH
H

PhcmouqH—NYCHalzCF
CH3 CHZCHzOl-I

G

 

9H 9H3
—C-CH2—CI-I2— N1 CH3X'
H R1

Ph

E. R1=C12H25, X=Br
F. R1=CH2PII, X=CI

PhCqu*H—h|l*(CH3)2CI'
CH3 cnzcuzou

H

CH2C*HOHCH20H

/

NTBF
Ph-Ph

\

Figure 1.3

Structures of some phase transfer catalysts

polarity of the solvent, catalyst concentration, etc., and (iii) the structure of the substrate.
Of these, the most important factor was recognized to be the structure of the catalyst.
However, the use of chiral crown ethers hasn’t yielded many successes. Chiral
alkylboranes such as Alpine-borane and diisopinocampheylchloraborane (-)-Ipcz have
also been reported to reduce ketones with excellent optical induction. Table 1.3 shows

some examples.

In addition to the use of chiral catalysts or reagents, chiral auxiliaries can also be used for
the asymmetric reductions of ketones [50-55]. Although the necessary requirement for
diastereoselective reduction of a carbonyl group is that its faces should be diastereotopic,
a significant degree of selectivity is generally achieved only when the carbonyl carbon is
a member of a ring or when there is a nearby chiral center, preferably adjacent to the
carbonyl group. In that case, 1,2-asymmetric induction, or some other interaction, e.g. by
the metal complexation, can be established between the reaction center and the inducing
chiral center. One of the examples of 1,2-induction is shown in Scheme 1.3 in which
chiral oxathine in the vicinity of a carbonyl group directs the reduction of a ketone
toward a diastereoisomeric alcohol, whether or not chelation is operative. The LAH or
Li-s-Bu3BH reductions give predominantly the diastereoisomer predicted by the Felkin-
Ahn model while reductions with DIBAL or LAH/MgBrz give predominantly the

diastereomer predicted by the chelation model [56-59].

14

Table 1.3

Stereoselective reductions of ketones using chiral alkyl boranes

 

 

Substrate Reagent Product (%e.e.) Reference
0 H
,x /)28H
" 47
0 HO“. H
I23” "
O " 47
<1) 95 (S)
o . Ho H
i-P ('2' CH f ‘c"
I‘- - 3
‘ i-Pr/ \CH3 47
37(8)
H OH
\ .-"
(CH3);CHCOC ECH if ‘._:: fl) (CH3);CHCC ECH 48
99 (8)

HO :1
H5C20COCOCH3 f3" ‘ ‘33) . H5020C&CH3 49

89 (S)

 

SWN'Cme
~ ° ° ‘°'“°"° (ii/"W" G’ "8°“:
-——> +
OH \r

OH
DIBAH 9 91
Li s-Bu33H 91 9
Scheme 1.3

1.2-Induction for stereoselective reduction

1,3-Induction, i.e. the stereodirecting effect of a chiral center separated by two bonds
from the carbonyl group, plays an important role in C—C bond—forming reactions, but in
hydride transfer reactions the effect is generally too weak for practical application in the
absence of a polar group. A chiral sulfur atom is a moderately effective directing group,
which can then be removed by Raney nickel desulfiJrization to give an optically active
alcohol. For this purpose both sulfoxides and sulfoimides are useful, but sulfoxides are
more accessible and with the proper choice of reducing agent better selectivities and
control of relative configuration can be achieved. For example, the reductions of chiral
ketosulfoxides can yield either diastereoisomeric B-hydroxysulfoxide with a high
selectivity at —78°C [56,60], depending on whether DIBAH in THF is used alone or in the
presence of a stoichiometric or catalytic amount of Zan (Scheme 1.4). After Raney
nickel desulfurization, enantiomeric secondary alcohols are obtained [61,62]. The
method has been extended to enones [63,64], aryl-B-ketosulfoxides [65] and to
chloromethylketones [66]. Anti-1,2-diols can be prepared in a repeated relay-like 1,3-

induction process [67].

16

Another way to synthesize chiral alcohols relies on the reductive cleavage of chiral acetal
rings. The reduction of acetal by BrzAlH or EtgsiH/TiCl4 leads to secondary diol
monoethers, which are easily transformed into the corresponding alcohols (Scheme 1.5)
[68]. Such is the case with 1,3-dioxanes that prefer the conformation in which the large
group R- is equatorial. Selective coordination of 0-3 to the Lewis acid induces cleavage
of the C-2-O-3 bond [69,70]. This is followed by hydride transfer either on the same face
of the molecule (AlBrzH) or on the opposite face (Et3SiH). Depending on the reagent,
either the (R)- or (S)-alcohols could be produced with good enantioselectivity after
removal of the diol residue. The drawbacks are that the chiral auxiliary is too expensive

and its removal is also cumbersome.

Asymmetric induction can also be triggered by remote centers in acyclic compounds.
One example of 1,4-induction is demonstrated in Prelog’s rule. In 1953 Prelog reviewed
the accumulated data and formulated a predictionknown as Prelog’s rule, by which the
prevailing configuration of the product could be deduced from the constitution and
configuration of the chiral alcohol [71]. It operates by drawing the molecule in a fixed
conformation and classifying substituents at the inducing center as small, medium and
large. One example is shown in Scheme 1.6. The selectivity depends on reducing agent
and chiral auxiliary [72,73] (Scheme 1.7 and Scheme 1.8). Even control of
configuration may be possible [73]. To avoid racemization when the chiral amide
residue is hydrolyzed, easily cleavable amides must be used. 1,5-induction has also been

reported. Two examples are shown in Scheme 1.9 and Scheme 1.10 [74].

17

O
I,”

 

Me/s" Tol
1) LDA OH
2) Rcozst ; .....« ,
0 70-90% RA/Sxm
/U\/T ,,,,, ....: mam, THF d.e. > 90%
R S\ Tel
60-80% OH O
\ /T\/ I "mu" T
ZnCIz, R S‘r I
DIBAH, THF °
d.e. > 90%
R = n-C3H17, Ar, t-BuOCO(CH2)3, t-BUOCOCHz
Scheme 1.4
Reductions of chiral ketosulfoxides
R00 R" ~
R 0 2' i .. OH
X > 90% Wn
———> O ——>—-> A
R- o BrzAIH /L OH R R'
.. R R'
R e.e. 78-96%
R"
II m
85% o R A
+ —>—> '
Et38iH, TiCI4 I R R
/\ OH
R R'
e.e. 76-96
.LA.
36
I
R' R

R = Ph, Et, i- Bu, 6-C5H11
R' = Me, R'CC R" = Me, Ph

Scheme 1.5

Reductive cleavage of chiral acetal

pair at) 212+“ .A... °Y

H S
O m) ( ) 75% e.e.

Scheme 1.6

An example demonstrating Prelog’s rule

 

 

0" o 1) LiBH4, LiBr
h] 1) DIBAL, LiBr JLcou , THF, -78°C (2)"
c ‘ ’ /°\
66% e.e. 87% e.e.
Scheme 1.7

Synthesis of both enantiomers of mandelic acid using a chiral auxiliary

0 95% OH
Mejl/I'L 0”,... ———> A
(1) DIBAH Me CH20H
(2) LAH
O 98 83%
Scheme 1.8

Reductions of oc-ketoesters with chiral alcohol auxiliaries

"‘3 o X/
n-Pr ——> 6* n-Pr + G* n-Pr
Etzo

MON"... 0 o OH OH
LAH 17 83
LAHIMgBrz 91 9
Scheme 1.9

Reduction of ketones with chiral acetal or oxathine in the vicinity

l9

0 Ph
W NaBHoEtzBOMe’

 

1“who o o THF, 48% to RT
Me
OH OH
OH‘
-—> “0ch
Ph
100% d.e.
> 95% e.e.

Scheme 1.10

Diastereoselective reduction via 1,5-induction

The second approach to forming chiral alcohols or alcohol equivalents is the oxidation of
C=C double bonds with the aid of chiral substructures to differentiate the enantiotopic
faces of the double bond. Such reactions include asymmetric epoxidation,
dihydroxylation, oc-hydroxylation, etc.. The classic example is Sharpless epoxidation of
allylic alcohols. In the presence of titanium tetraisopropoxide and L-(+)- or D-(-)-diethy|
tartrate, allyl alcohol can be stereoselectively epoxidized by t-butylhydroperoxide under
stoichiometric or catalytic conditions (Scheme 1.11) [75]. Under catalytic conditions,
the enantiomeric excess of the product is usually lowered by 1-5% relative to the reaction
using 50-100 mol % of Ti isopropoxide; however, the catalytic version enlarges the scope
of the reaction, increases its efficiency, and results in a generally higher chemical yield.
In addition, the workup procedure is much simpler. The absolute configuration of the
epoxide formed depends on the optical configuration of the tartrate used and it can be

predicted from Scheme 1.12. The enantiomeric excess usually ranges from 70% to 98%

20

 

(R 50.90% g R
+ f

(R.Rl-DIPT CH20H
e.e. > 95%

R = H, Me, n-Pr, n-C14H29, PhCH20CH2

\— 70-90% ’ R
\CHZOH t-BuOOH,Ti(OR)4.

(R,R)-DIPT CHon
e.e. 92 - 98%

R = Me, Et, n-Pr, i-Pr, n-Csl-I", n-CgH17, CH2=CH, Ph

_ 70-90% a a

(R,R)-DIPT R CH20H
e.e. 85 - 95%

R = Me, Et. i-Bu, n-C1H15, n-Cgl-I19, Ph, PhCH20CH2

M 70-90% Me R
a ‘1. Z S f
(R,R)-DIPT CH20H

e.e. > 95%
R = Me, Et, Ph, PhCH20CH2, M92C=CHCH2

< 70-90% Me
a
(R,R)-DIPT R CHZOH
e.e. >90%

R = Mezc=CHCH2, Me2C=CHCHch2

 

 

 

 

Scheme 1.11

Sharpless epoxidation of primary allyl alcohols

21

depending on the substrate structure. In an application of the chiral auxiliary technique,
succindialdehyde has been transformed into mono-N-tosyloxazolidine. Treatment with
KOCl leads to an epoxyacid with high diastereoselectivity [76]. After nucleophilic ring
opening of the epoxide, the chiral auxiliary is recovered by treatment with ethanedithiol

(Scheme 1.13).

Compared to allyl alcohols, the epoxidation of unfunctionalized alkenes has yielded less
success. The most notable catalysts developed to give relatively high enantioselection
were Mn(III) complexes with C2 chiral salen ligands (Scheme 1.14) [77-82]. The 20-
93% e.e. obtained are not generally satisfactory but compared well with the highest
values obtained by other methods. C is disubstituted olefins give better enantioselectivity
than the trans olefins. Terminal olefins afford intermediate stereoselectivity. A related
reaction, dihydroxylation, is used to synthesize chiral vicinal diols. Dihydroxylation is
usually conducted with OsO4 either in stoichiometric or catalytic amounts in the presence
of cooxidants such as NMO. The asymmetric dihydroxylation can be performed in the
presence of chiral ligands of osmium or by reaction of 0504 or other reagents with
olefins bearing a chiral residue. Two classes of ligands have been successfully used in
asymmetric osmylation of prochiral olefins: chiral diamines and cinchona alkaloids. The
diamines have been developed mainly by Corey, Koga and Tomioka, Hirama and their
coworkers [83-87], and cinchona alkaloids have been designed and implemented by
Sharpless and coworkers [84-88]. The latter approach is more common because of its
convenience and high stereoselectivity. The reagent combination, AD-mix (0804 0.2

mol%, ligand, 1 mol% K3Fe(CN)6, K2CO3) has found many applications. Among the

22

D-(-)-diethyl tartrate (D-DET)
"O"

i

(f1 .32? Ti(o-i-c3H7), R2 R1

+ TBHP ’ E:
OH OH
R3 CH2C'2 R3

molecular sieves

 

 

70-90% yield
.90.. >900/0 e.e-
L-(+)-diethyl tartrate (L-DET)

cszooc...‘ OH . Czl-15000\[0H
Czl-I5OOC/[OH chsooc'“ OH
L-(+)-diethyl tartrate D-(-)-diethyl tartrate

[(R,R)- or L-DET] [(8,8)- or D-DET)

Scheme 1.12

Control of absolute configuration of Sharpless epoxidation

My)?“ Ken—s OWEW
3:893

Scheme 1.13

 

1) Nu'M‘, then H20
2) HS(CH2)2SH

Asymmetric epoxidation using a chiral auxiliary

23

 

\ + 1-8% MnL*
—____)
CH2CI2

 

 

73% Yield
_ __ 84% e.e.
Ph_ Ph *-
NM\ nN/ -
MnL*= 0M/ "O\ ”‘6
(CHalaSi

 

(CH3);Si ,
w + aq NaOCI 4 mol% MnL ’
\ CH2C12

 

 

 

[— ‘ + 65% yield
Hm» (trans: ds=5.2:1)
trans 98% e.e. (S,R)
cis 64% e.e.
mN/ -
MnL“ = O—:/ ”"0\ c1
Scheme 1.14

Chiral manganese complex for asymmetric epoxidation

24

tact:
and
En;

The

C

numerous ligands tested, those derived from DHQ, dihydroquinine (AD-mix-a) and
from DHQN, dihydroquinidine (AD-mix-B) often give rise to enantiomeric diols with a
high degree of facial discrimination. As an empirical rule, AD-mix-oc induces syn-
dihydroxylation of the bottom face of the olefin, while AD-mix-B induces
dihydroxylation on the top face (Scheme 1.15). The enantiomeric excess usually ranges

from 75% to 95% depending on the substrate structure (Scheme 1.16).

The asymmetric synthesis of a-hydroxycarbonyl compounds has also been achieved with
notable success. The synthetic utility of 2-(phenyl sulfonyl)-3-phenyl oxaziridine (B)
and analogs, a new class of aprotic and neutral oxidizing reagents, has been established in
asymmetric hydroxylations. B has been employed in the oxidation of chiral azaenolates
prepared from ketones and SAMP affording the ‘oc-hydroxylation product in 96% e.e..
This undergoes oxidative cleavage with ozone to yield the oc-hydroxy ketone in high

enantiomeric purity (Scheme 1.17) [89].

Over the years, a large number of chiral molecules have been used in asymmetric
synthesis applications. Their ability to induce high or excellent stereoselectivity is
certainly one of the most important factors in determining their utility in synthetic
chemistry. . However, other factors such as the cost of the chiral molecules and the
availability of the reagents in the process are also essential, especially in industry.
Nature has provided a rich and often inexpensive chiral pool of compounds such as

carbohydrates, amino acids, alkaloids, etc. Research on how to best utilize these readily

25

dihydroquinidine ester
"HO OH"

R2 R3 R2 R3
0.2-0.4% 0504
+ "MO 9 R1->——<
acetone-H2O

 

 

80-95% yield
T 40-95% e.e.

"HO OH"
dihydroquinine ester

NMO = N-methylmorpholine N-oxide

 

Cl

dihydroquinidine ester dihydroquinine ester
Scheme 1.15

Prediction of stereochemical outcome of Sharpless
asymmetric dihydroxylation of olefins

26

OH
80-90%
/ + \A
/\ R AD-mix, A ”O . R

e.e. 85-97%
R = n-C8H17. Ph, PhOCHz. Z'NPOCHz O'CsHs 0'08”". OC1H13

OH OH
R \/ 70-90% 3
I )
/\ R AD-mix, A R R' or RY\R'
OH OH

e.e. > 95%

 

 

R,R'=Et, n-Bu, n-05H11, Ph, PhCH=CH, COOMe, COOEt, CHzoAc, CH20H

R

OH
)\ 50-92% R
Ph * Ho *

AD-mix, A Ph

 

e.e. 78 - 97%

R = CHzcl, CHzBr, CHzoMe, CH20CH2Ph
Me

OH
80-90% Me
»
Me AD-mix, A ”0 * Me
R
R
e.e.‘ 85 -98%

 

R = n-C5H11, Ph

Scheme 1.16

Sharpless asymmetric dihydroxylation

27

ax

dc

19

9’

 

O OCH3 N/
Ph/u\, + —>
Ph OCH3

5“
Ph
N"'2 Ph
/N Ph
i) LDA N i) 03 ph
ii) a , OCH ii) Aczo * OCOCHa
Ph 0” 3 0
Ph
> 96% e.e. (R)
> 96% e.e.

B: PhSOzNQOHPh
Scheme 1.17

Asymmetric a-hydroxylation using SAMP as a chiral auxiliary
available chiral molecules has been intensive. However, there is still a lot of work to be
done. For instance, carbohydrates have been utiliied as chiral synthons extensively [90],
but their use as chiral auxiliaries is somewhat limited. Recently, a number of research
groups started investigations in this area and showed the great potential of carbohydrates
as chiral auxiliaries [91-93]. What we are interested is their use (particularly D-glucose
due to its low cost, even if you can’t recycle it) for the asymmetric conversion of C=C or
C=C double bonds to chiral alcohols. We would like to avoid the tedious protection and
deprotection which is typical in carbohydrate chemistry as much as possible. Oxidation
(e.g. oxymercuration) has been implemented on allyl glucosides and their analogs and
excellent stereoselectivity has been achieved after removal of the chiral auxiliary [Details

in Chapter 2]. Some other important chiral intermediates can also be produced using

28

similar procedures. In order to avoid the use of mercury salts and simplify the removal of
the glucose unit from the alcohol product, asymmetric reductions in which the substrate
is connected to the auxiliary by only the glycosidic bond have been conducted. Oxo-
alkyl or oxo-aryl glycosides were reduced via chelation control to produce chiral alcohols
[Details in Chapter 3]. Other than carbohydrates, the use of (S)-nicotine has also been
investigated both as a phase transfer catalyst afier derivatization and as a complexing
chiral base [Details in Chapter 4]. (S)-nicotine is famous for its adverse effect on human
health, however, its use in asymmetric synthesis is very limited even though it is readily
available. The other molecule we looked at is (L)-lactic acid [Details in Chapter 5].
The structures of these molecules are shown in Figure 1.16. In the following chapters, I
would like to report our endeavors to utilize these molecules as chiral promoters,

particularly as chiral auxiliaries.

 

0 HO 0
7 i HO OH
HO OH HO OH
(S)-nicotine (S)-lactic acid D-glucose

Figure 1.4

Structures of some cheap chiral molecules

29

10.

ll.

12.

13.

14.

15.

16.

References

. Zassinovich, G.; Bettella, R.; Mestroni, G.; Bresciani-Pahor, N.; Geremia, S.;

Randaccio, L. J. Organomet. Chem. 1989, 370, 187.

Hayashi, T.; Mise, T.; Kumada, M. Tetrahedron Lett. 1976, 4351.

Muller, D.; Umbricht, G.; Weber, B.; Pfaltz, A. Helv. Chim. Acta. 1991, 74, 232.
Taber, D.; Silverberg, L. Tetrahedron Lett. 1991, 32, 4227.

Heiser, B.; Broger, E.; Crameri, Y. Tetrahedron Asymm. 1991, 2, 51.

Takahashi, H.; Morimoto, T.; Achiwa, K. Chem. Lett. 1987, 855.

Kitamura, M.; Ohkuma, T.; Inoue, S.; Sayo, N.; Kumobayashi, H.; Akutagawa, S.;
Ohta, T.; Takaya, H.; Noyori, R. .1. Am. Chem. Soc. 1988, 110, 629.

Kitamura, M.; Okuma, T.; Inoue, 8.; Sayo, N.; Kumobayashi, H.; Akutagawa, S.;
Ohta, T.; Takaya, H.; Noyori, R. J. J. Am. Chem. Soc. 1988, 110, 629.

Cesarotti, 13.; Mauri, A.; Pallavicini, M.; Villa, L. Tetrahedron Lett. 1991, 32, 4381.
Miyashita, A.; Chiba, T.; Nohira, H.; Takaya, 11., Abstract of 38th Symposium on
Organometallic Chemistry, Kyoto, 1991, B216.

Dumont, W., Poulin, J., Dang, T., Kagan, H. J. Am. Chem. Soc. 1973, 95, 8295.
Ojima, 1.; Kogure, T.; Kumakai, M. J. Org. Chem. 1977, 42, 1671.

Tai, A.; Harada, T.;Hiraki, Y., Murakami, S. Bull. Chem. Soc. Jpn. 1984, 57, 1518.
Nakahata, M.; lmaida, M.; Ozaki, H.; Harada, T.; Tai, A. Bull. Chem. Soc. Jpn. 1982,
55, 2186.

Kitamura, M.; Ohkuma, T.; Inoue, S.; Sayo, N.; Kumobayashi, H.; Akutagawa, S.;
Ohta, T.; Takaya, H.; Noyori, R. J. Am. Chem. Soc. 1988, 110, 629.

Andrisano, R.; Angeloni, S.; Marzocchi, S. Tetrahedron 1973, 29, 913.

30

17.

l8.

19.

20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

30.

31.

32.

33.

34.

35.

36.

Noyori, R.; Tomino, 1.; Tanimoto., Y.; Nishizawa, M. .1. Am. Chem. Soc. 1984, 106,
6709.

Masanobu, Y.; Tabashi, O. J. Chem. Soc. Perkin Trans. 1 1990, 1826.
Midland, M.; Kazubski, A. J. Org. Chem. 1982, 47, 2814.

Sugimoto, T.; Matsumura, Y.; Tanimoto, S.; Okano, M. J. Chem. Soc. Chem.
Commun. 1978, 926. .

Corey, E. J .; Link, .1. O. Tetrahedron Lett. 1989, 30, 6275.

Colonna, S ; Fomasier, R. J. Chem. Soc, Perkin Trans. 1 1978, 371-373.
Landor, S. R.; Miller, B. J.; Tatchell, A. R. J. Chem. Soc. 1964, 227.

Landor, S. R.; Miller, B. J.; Tatchell, A. R. J. Chem. Soc. C. 1964, 1822.
Landor, S. R.; Miller, B. J .; Tatchell, A. R., J. Chem. Soc. C. 1966, 2280.
Landor, S. R.; Miller, B. J.; Tatchell, A. R. J. Chem. Soc. C. 1967, 197.
Landor, S. R.; Miller, B. J.; Tatchell, A. R. J. Chem. Soc. C. 1971, 2339.
Landor, S. R.; Miller, B. J.; Tatchell, A. R. J. Chem. Soc. Chem, Commun. 1966,
585.

Landor, S. R.; Sonola, O. 0.; Tatchell, A. R. J. Perkin T ransJ. 1974, 1902.
Landor, S. R.; Miller, B. J.; Tatchell, A. R. J. Chem. Soc., C. 1978, 605.
Noyori, R.; Tomino, 1.; Tanimoto, Y. J. Am. Chem. Soc. 1979, 101, 3129.
Noyori, R. Pure Appl. Chem. 1981, 53, 2316.

Jacquet, I.; Vigneron, J .-P. Tetrahedron Lett. 1974, 2065.

Vigneron, J .-P.; Jacquet, l. Tetrahedron 1976, 32, 939.

Vigneron, J .-P.; Bloy, V. Tetrahedron Lett. 1979, 2683.

Yamada, S.; Kitamoto, M.; Terashima, S. Tetrahedron Lett. 1976, 3165.

31

37.

38.

39.

40.

41.

42.

43.

44.

45.

46.

47.

48.

49.

50.

51.

52.

Kitamoto, M.; Kameo, K.; Terasjo, S.; Yamada, S. Chem. Pharm. Bull. 1977, 25,
1273.

Terashima, S.; Tanno, N.; Koga, K. Chem. Lett. 1980, 981.

Hirao, A.; Mochizuki, H.; Nakahama, S.; Yamazaki, N. J. Org. Chem. 1979, 44,
1720.

Morrison, J. D.; Grandbois, E. R.; Howard, S. l. J. Org. Chem. 1980, 45, 4229
Hirao, A.; Nakahama, S.; Mochizuki, H.; Itsuno, S.; Yamazaki, N. J. Org. Chem.
1980, 45, 4231

Hirao, A.; Nakahama, S.; Mochizuki, H.; Itsuno, S.; Ohowa, M.; Yamazaki, N. J.
Chem. Soc. Chem. Chem. Commun. 1979, 807

Colonna, S ; Fomasier, R. J. Chem. Soc., Perkin Trans. 1 1978, 371-373.
Balcells, J .; Colonna, S.; F omasier, R. Synthesis 1976, 266.

Kinishi, R.; Nakajima, Y.; Oda, I.; lrouze, Y. Agric. Biol. Chem. 1978, 42, 869.

Shida, Y.; Onda, N.; Yamamoto, Y.; Oda, J .; lnouye, Y. Agric. Biol. Chem. 1979, 43,
1797.

Midland, M. M.; Greer, S.; Tramontano, A.; Zderic, S. A. J. Amer. Chem. Soc. 1979,

101, 2352.

Brown, H. C.; Mandal, A. K. J. Org. Chem. 1984, 49, 2558.

Brown, H. C.; Pai. G. G. J. Org. Chem. 1982, 47, 1608.

Soai, K.; Ookawa, A. J. Chem. Soc. Chem. Commun. 1986, 412.

Oppolzer, W.; Rodriguez, 1.; Starkemann, C.; Walther, E. Tetrahedron Lett. 1990, 31,
5019.

Ishihara, K.; Mori, A.; Arai, 1.; Yamamoto, H. Tetrahedron Lett. 1986, 26, 983.

32

53

54.

55.

56.

57.

58.

59.

60.

61.

62.

63.

64.

65.

66.

67.

68.

69.

70.

Mori, A.; Ishihara, K.; Arai, 1.; Yamamoto, H. Tetrahedron 1987, 43, 755.
Solladie-Cavallo, A.; Suffert, J. Tetrahedron Lett. 1985, 26, 429.

Soai, K.; Isoda, T.; Hasegawa, H.; Ishizaki, M. Chem. Lett. 1986, 1897.
Seyden-Penne, J. Reductions by the Alumino- and Borohydrides in Organic
Synthesis, Verlag-Chemie, New York, 1991.

Eliel, E. L.; Frye, S. V.; Hortelano, E. R.; Chen, X.; Bai, X. Pure and Appl. Chem.
1991, 63, 1591.

Alexakis, A.; Mangeney, P. Tetrahedron Asymm. 1990, I, 477.

Frye, S. V.; Elie], E. L. J. Am. Chem. Soc. 1988, 108, 484.

Carreno, M. C.; Rodriguez, J. H.; Rubio, A.; Sanchez, J .; Solladie, G. J. Org. Chem.
1990, 55, 2120.

Hua, D. H.; Bharathi, S. N.; Panangadan, J. A. K.; Tsujimoto, A. J. Org. Chem. 1991,
56, 6998.

Solladie, G.; Almario, A. Tetrahedron Lett. 1992, 33, 2477.

Solladie, G.; Hutt, J .; F rechou, F. Tetrahedron Lett. 1987, 28, 61.

Solladie, G.; Frechou, C.; Demailly, G. Tetrahedron Lett. 1986, 27, 2867.

Careno, M. C.; Ruano, J. L. G.; Martin, A. M.; Pedregal, C.; Rodriguez, J. H.; Rubio,
A.; Sanchez, J.; Solladie, G. J. Org. Chem. 1990, 55, 2120.

Solladie, G.; Hamdouchi, C .; Vicente, M. Tetrahedron Lett. 1988, 29, 5929.
Solladie, (3.; Ghiatou, N. Tetrahedron Lett. 1992, 33, 1065.

Alexakis, A.; Mangeney, P. Tetrahedron Asymm. 1990, I, 477.

Ishihara, K.; Mori, A.; Arii, 1.; Yamamoto, H. Tetrahedron Lett. 1986, 2 7, 983.

Yamamoto, K.; Ando, H.; Chikamatsu, H. J. Chem. Soc. Chem. Comm. 1987, 334.

33

71.

72.

73.

74.

75.

76.

77.

78.

79.

80.

81.

82.

83.

84.

85.

Prelog, V. Helv. Chim. Acta 1953, 36, 308.

Solladie-Cavallo, A.; Suffert, J. Tetrahedron Lett. 1985, 26, 429.

Soai, K.; Isoda, T.; Hasegawa, H.; Ishizaki, M. Chem. Lett. 1986, I897.

Reddy, G. B.; Minami, T.; Hanamoto, T.; Hiyama, T. J. Org. Chem. 1991, 56, 5752
Johnson, R. A.; Sharpless, K. B. in Comprehensive Organic Synthesis, Trost, B. M.
and Fleming, 1. Ed. Pergamon Press 1991, Vol. 7, Chap. 3.2.

Cardani, S.; Gennari, C.; Scolastico, C.; Villa, R. Tetrahedron 1989, 45, 7397.
Zhang, W.; Loebach, J. L.; Wilson, S. R.; Jacobsen, E. N. J. Am. Chem. Soc. 1990,
112, 2801.

Jacobsen, E. N.; Zhang, W.; Muci, A. R.; Ecker, J. R.; Deng. L. J. Am. Chem. Soc.
1991, 113, 7063.

Lee, N. H.; Jacobsen, E. N. Tetrahedron Lett. 1991, 32, 6533.

Hatayama, A.; Hosoya, N.; Irie, R.; Ito, Y.; Katsuki, T. Synlett. 1992, 407.

Collman, J. P.; Zhang, X.; Hembre, R. T.; Brauman, J. I. J. Am. Chem. Soc. 1990,
112, 5356.

Schurig, V.; Betschinger, F. Chem. Rev. 1992, 92, 873.

Tomioka, K. Synthesis 1990, 541.

Haines, A. H. in Comprehensive Organic Synthesis, Trost, B. M. and Fleming, 1. Ed.
Pergamon Press 1991, Vol. 7, Chap. 3.3.

Ogino, Y.; Chen, H.; Manoury, E.; Shibata, T.; Beller, M.; Lubben, D.; Sharpless, K.

B. Tetrahedron Lett. 1991, 32, 5761.

34

86.

87.

88.

89.

90.

91.

92.

93.

Sharpless, K. B.; Amberg, W.; Beller, M.; Chen, H.; Hartung, J.; Kawanami, Y.;
Lubben, D.; Manoury, E.; Ogino, Y.; Shibata, T.; Ukita, T. J. Org. Chem. 1991, 56,
4585.

Sharpless, K. B.; Amberg, W.; Bennani, Y. L.; Crispino, G. Hartung, J.; Jeong, K. S.;
Kwong, H. L.; Morikawa, K.; Wang, Z. M.; Xti, D., Zhang, X. L. J. Org. Chem.
1992, 57, 2768.

Johnson, R. A.; Sharpless, K. B. in Catalytic Asymmetric Synthesis, Ojima, 1. Ed.
VCH 1993, p. 227.

Enders, D.; Bhushan, V. Tetrahedron Lett. 1988, 29, 2437.

Hanessian, S. Total synthesis of natural products, the “Chiron ” approach, 1983,
Oxford, New York: Pergamon Press.

Kunz, H.; Ruck, K. Angew. Chem, Int. Ed. Engl. 1993, 32, 336-358.

Hale, K. J. Monosaccharides: Use in synthesis as chiral templates. In Second
Supplements to the 2nd Edition of Rodd ’s Chemistry of Carbon Compounds;
Sainsbury, M., Ed.; Elsevier: Amsterdam, 1993: Vol. I EFG, pp. 273-313.

Hultin, P. G.; Earle, M. A.; Sudharshan, M. Tetrahedron 1997, 53, 14823-14870.

35

CHAPTER 2
The Stereo-selective Conversion of 2-Alkenyl Alcohols to (R)- or (S)-Alkane-1, 2-

diols and Other Chiral Intermediates Using D-Glucose as a Chiral Auxiliary

36

Introduction

The enantio-selective functionalization of alkenes is one of the more important reactions
in synthetic organic chemistry and one for which much effort has already been expended.
The oxidation of the vinyl group of allylic alcohols is an important transformation and the
use of transition metal complexes with chiral ligands is a prominent way of achieving this
end [1,2]. Although the optical purity of the products is generally high for many
substrates, it is still not satisfactory for some allylic alcohols. There is still a need for
general methods for enantioselectively oxidizing the vinyl group of 2-alkenyl alcohols.
The use of carbohydrates (e.g. D-glucose) as chiral auxiliaries has a special appeal for
such applications because of their availability and intrinsic chirality. However, their
structural complexity and highly redundant functionality have hindered the development
of such applications [3]. The use of carbohydrates as chiral auxiliaries to functionalize 2-
alkenyl alcohols requires attachment at a strategically simple site such as the anomeric
position followed by transformation and removal. Although cyclopropanation of 2-
alkenyl glycosides has been reported [4] with excellent stereochemical control (Scheme
2.1), the more widely used oxidation reactions involving the addition of heteroatoms,
such as epoxidation [5], dihydroxylation [6], and dibromination [7] gave poor

selectivities (Scheme 2.2).

We felt the key to obtaining high stereoselectivity was to better utilize the free 2-hydroxy

group. A substrate, such as allyl alcohol, can be attached to the anomeric position of the

37

8110 0 R1 Etzzn BnO O
B" 0 R2 W Bn 0 R2
Bn 2 En
OH OH

>97%; >50:1
Scheme 2.1

Cyclopropanation using D-glucose as a chiral auxiliary

Bno O m-CPBA
B" 0
8n \/\\/ toluene OW

 

OH

80%; 9:1

Dihydroquinidine- OH

A6360 0 p-chloro benzoate Acsco 0 g
0 Ac 0304, NMO 0 Ac

Acetone, H20 .

95%; 3:1
CH3O HO CH3O

HO
0 Bufi‘Brgf O
CH CHzcl B H
cuao o 2 c" c1130 '
W O\></Br
quant.; 4:1

Scheme 2.2

Stereoselective oxidation of C=C double bond
using carbohydrates as chiral auxiliaries

38

glucose via traditional Fisher glycosidation or the Koenigs-Knorr reaction depending on
the isomer of interest. However, the 2-hydroxyl group is far away from the C=C double
bond we planned for the asymmetric transformation. There are many applications
showing that steric and electronic effects from an adjacent chiral center can lead to high
levels of asymmetric induction of C=C and C=O double bonds in acyclic compounds [8-
13]. But these effects usually decline rapidly in importance as the number of bonds
separating the chiral center from the prochiral double bond increases. One way of
imposing some conformational restraint is to carry out the reaction intramolecularly,
using one of the substituents on the chiral center as the attacking species. The
intramolecular reaction may be nucleophile-, electrophile- or radical-mediated, or it may
involve an intramolecular cycloaddition (Figure 2.1). The cyclic transition state sets up a
new set of steric or electronic interactions and these can give rise to high
diastereoselectivity. For example, cyclization takes place via nucleophilic attack on the
iodonium ion by the carbonate anion to give the 4,6-cis-disubstituted product
diastereoselectively (Scheme 2.3) [14]. Asymmetric induction is the result of a number
of factors including (a) cyclization via chair conformation, (b) preference of the
(CH2)zOBn side chain for an equatorial rather than an axial position on the chair, (c)
preference of the CH2 of the iodonium ion for an equatorial rather than an axial position
and (d) reversibility of iodonium ion formation. This tactic can be applied to the
diastereoselective addition to a double bond contained in or on a ring, the chiral center
being one of the ring atoms. This will further limit the freedom of the attacking species

so as to improve the diastereoselectivity.

39

a p a P

ch—Xn Mq b>r Xn

. a (‘9
Ian—kg.

Figure 2.1

(n = 1, 2 3,...)

distant chiral center

Ring formation to improve stereoselectivity

 

BnO 8:10
1. BuLi
2. co2 +
—-—>
yo I
H
o H H

69%
9 : 1 d.r.
Scheme 2.3

Intramolecular iodolactonization showing high diastereoselectivity

40

Our experimental design is to utilize the 2-hydroxy group as a nucleophile for
asymmetric oxidation of the prochiral C=C double bond. Once the substrate is attached
to the auxiliary (D-glucose), oxonium formation initiated by X+ followed by ring closure
via the 2-hydroxy group will form a six-membered ring and generate a new chiral center.
This process is highly stereoselective since it is an intramolecular cyclization and bulky
substituents tend to occupy the equitorial rather than axial position in the six-membered
ring. The anomeric Cl-Ol linkage can then be opened up to form the hemiacetal. Afler
cleaving Cz-O2 without scrambling the newly generated chiral center, the high
diastereoselectivity obtained in the cyclization is translated into high enantioselectivity in

the final product (Scheme 2.4).

In this study, the stereo-selective functionalization of the vinyl group of 2-alkenyl
alcohols is demonstrated by arranging for the nucleophilic addition of the 2-hydroxyl
group of D-glucose to the activated double bond. The glucose moiety is then removed to
give optically pure (R)- or (S)-alkane-I, 2-diols depending on the anomeric configuration
of the starting glycoside. In the present scheme, the double bond is activated by mercury
trifloroacetate and the alkoxy mercurated intermediate is reduced by borohydride. The
diol is recovered by acetolysis followed by oxidative removal of the glucose anomeric
carbon as formic acid using alkaline hydrogen peroxide (Scheme 2.5). A high degree of
stereoselectivity is ensured because the bulky alkyl group in the intermediate bicyclic
system has to be equatorial. Hence the a-glycoside should yield the (R)-diol and the B-
glycoside should yield the (S)-isomer. Because the absolute configuration of the diol is

determined by the configuration of the starting glycoside, enantiomeric purity can be

41

 

 

 

 

 

”a; 01H Substrate mags? Ring Cyclization

HO 4 3 ——>

HO

07104.0 ~|Subshate
D-glucose

 

 

 

 

OH
”Ho 4 HO O o OH H202IOH',70° it?"
——> “$0 \A’ ——> Substrate
OH

 

 

 

 

 

 

02
*1 I
,
Scheme 2.4

Experimental design for the synthesis of chiral diols

 

FF“
HO 0 R HO 0 1. Hg(OCOCF3)2
HO OH 2. NaBHJNaOI-l

OH

 

o 1. BF3, AcZO,TFA OH

”$94
H )
940 00 2. Hzolel-rJOO R\/'\/OH

 

 

 

 

Scheme 2.5

Actual scheme used to make chiral diols

42

ensured at the level of purity of the glycoside. The pure glycoside can be obtained via
enzymatic degradation in which a- or [3- glucosidase selectively catalyzes the hydrolysis
of the a- or B-glucoside. Alternatively, the a- and B-glucosides can be separated after
they are converted to a 4,6-benzylidene acetal or similar crystalline derivative. The
generality of the method was further demonstrated by converting cis-2-pentenol to the (R)
and (S)-pentane-1, 2-diols with excellent stereochemical control. Besides alkyl-1,2-diols,
other important chiral building blocks (e.g. hydroxylactones, aminoalcohols, etc.) can
also be obtained via the iodo intermediate by small modifications of the reaction

sequence.

43

Results and discussion

1. PREPARATION OF (R)—PROPANE-1,2-DIOL

The synthesis of (R)-proane-l,2-diol uses a scheme similar to that shown in Scheme 2.5.
The only difference is that the oxymercuration is performed directly on the allyl-O-4,6-
benzylidene-a-D-glucopyranoside and acid hydrolysis is used to remove the benzylidene
acetal before the acetolysis step. The derivatization to a1lyl-O-4,6-benzylidene-a-D-
glucopyranoside is simply for the convenience of separation since it can be easily
recrystallized and separated from the [3 form. Fisher glycosidation followed by
derivatization gave both isomers as allyl-O-4,6-benzylideneglucopyranoside (crude yield:
50%, a/B = 2.6/1). The 0t and B isomers were separated by recrystallization in methanol

and water. From the 1H-NMR, the a isomer showed a characteristic doublet for the

anomeric proton at 4.94 ppm (J = 4.9 Hz), while the B isomer gave the anomeric proton
signal at 4.45 ppm (J = 7.8 Hz). Oxymercuration was conducted by slowly adding a
CH3CN solution of Hg(OCOCF3)3 to the a-isomer in CH3CN. The use of Hg(OCOCF3)2
has the advantage over Hg(OCOCH3)2 since it improves both yield and selectivity. The
slow addition was crucial for obtaining a high percentage conversion of starting material.
Since intramolecular nucleophilic attack of the hydroxyl oxygen on the mercurinium ion
can only take place with inversion, the mercurinium ion must have the configuration
allowing that “anti” attack. However, there is no compelling reason why only one
diastereoface of the alkene should be attacked in the formation of the mercurinium ion,

and consequently attack on the C=C must be reversible if high yield of the final product

44

is to be obtained. This reversible process was effected by controlling the concentration of
Hg(OCOCF3)2. Slow addition helps to improve the efficiency of this reversible process.
The 1H-NMR and l3C-NMR spectrum (Figure 2.2) of the crude product 2 showed the
presence of predominantly one methyl group (1.21 ppm in lH-NMR; 16.0 ppm in 13C-
NMR), one benzylidene acetal signal (5.51 ppm in lH-NMR; 102.0 ppm in l3C-NMR)
and one anomeric position signal (4.96 ppm in IH-NMR; 94.6 ppm in l3C—NMR)

confirming that mainly one isomer was formed.

The major isomer was purified by column chromatography. Cleavage of the C'-Ol bond
by acid hydrolysis proved difficult. The common condition (2M TFA), which works well
for the cleavage of the methyl glycoside, only cleaved the benzylidene acetal of
compound 2. Raising the temperature and prolonging the reaction time did not help. A
number of other reaction conditions were tried. These included 80% formic acid, 2N
HCl, and methanolysis, but all resulted in failure. The stability of this glycosidic bond
stems from the bicyclic six-member ring system. Even if the bond did open up, the ring
would reform via an intramolecular process under acidic conditions. One way to get
around this problem would be to use a Lewis acid to activate the ring-opening and trap
the carbocation once it is formed. Acetolysis was chosen to solve this problem. First, the
benzylidene acetal was deprotected by hydrolysis in IN HCl (quantitative). The
acetolysis was conducted using BF3.OEt2, TFA and acetic anhydride (yield = 72.7%).
After workup, the product was analyzed by lH-NMR and l3C-NMR (Figure 2.3). In the
lH-NMR spectrum, the doublets at 5-6.5 ppm corresponded to -C'fl(OAc)(OR). Since

-OAc can trap the oxonium ion from either side of the ring, two isomers were obtained.

45

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

WW WW

W1117711‘IIIF‘VIUV'IYYV‘VlT—TTIVIVIYITIIVIVVIytyvlufYVITVVTIIIVIIijY'Y
1‘0 130 120 110 100 90 IO 70 ‘0 50 ‘0 30 20 m

 

Figure 2.2

lH-NMR and l3C-NMR of compound (2): 4,6-O-Benzylidene-1,2-O-
[2-methyl-(R)- l ,2-ethanediyl]-ot-D-glucopyranose

46

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

f l I I 1 l I 1 T

1 6 5 4 3 2 1 pp
A
A c
A

c
‘l.
160 100 120 100 00 60 00 20 pp
Figure 2.3

1H-NMR and 13C-NMR of acetolysis intermediates

47

Also the Cl-OS bond was cleaved during the acetolysis process to yield two other
isomers. The four isomers were not separated since the chiral segment of interest in all
four isomers remained the same. After oxidative cleavage of the auxiliary, the only diol
we obtained was (R)-propane-l,2-diol as expected. The mechanism of oxidative
cleavage is shown in Scheme 2.6. Peroxide anion attacks the carbonyl carbon atom to
form a tetrahedral intermediate, and then a lone pair of electrons from O2 is donated to
form an oxonium ion by eliminating forrnate and hydroxide. The oxonium ion is trapped
by hydroxide to form a hemiacetal. The unstable hemiacetal spontaneously falls apart to

give HOR*, which in this case is propane-1,2-diol.

The e.e.% was determined to be over 99% by chiral column gas chromatography of its
diacetate derivative. The absolute configuration was confirmed to be (R) afier comparing
the Rf values of both racemic propane-1,2-diol diacetate and (S)- propane-1,2-diol

diacetate.

2. PREPARATION OF (S)-PROPANE-1,2-DIOL

The same procedure (Scheme 2.7) was applied to allyl 4,6-O-benzylidene-B-D-
glucopyranoside. Chiral column gas chromatography confirmed that we obtained the
other enantiomer, (S)-propane-l,2-diol as the final product. The only difference was the
diastereoselectivity of the oxymercuration step. While nearly 82% d.e. was obtained
starting from the a-isomer, only about 50% d.e. was obtained for the B-isomer. Figure
2.4 shows the lH-NMR spectrum of the crude product. The doublets at 1.17 ppm and

1.41 ppm correspond to the methyl group signal of the two isomers. The reason for the

48

HO

ououov/'\. mono-0:3,.
tHfl”_,Ho

 

 

H
OH 012* 01-1 on“
-HC02' OH OH 'OH OH OH
—* HOMI ___) HOMO”
OH OH (O‘R‘r OH OR*
OH OH
—_’ HOR* + HO O
OH H
I repeat
5HCOz'
Scheme 2.6
Mechanism of oxidative cleavage of chiral auxiliary
1- =/—°“
HO O H‘ 1:1/V" O O 1-H9(°C0CF3)2
”0 O” i 910 Q 2 N BH on>
HO OH 2. PhCH(OMe)2 OH \ ' a 4"“
H

 

 

1. 1130*
u" $3.» 2mm w-
a
910 0,9 R\/K/OH

3. HZOzIOH',7O°

 

 

 

Scheme 2.7

Reaction scheme to synthesize (S)-propane-1,2-diol

49

 

 

 

 

 

 

 

 

 

 

.1

Figure 2.4

lH-NMR of crude products of oxymercuration of
allyl 4,6-O-benzylidene-B-D—glucopyranoside

50

selectivity can be explained by Figure 2.5. Starting from the a-isomer, if the
CHzHgOCOCFg, group occupies the axial position, it will experience strong steric
repulsion from H3. On the other hand, for the B-isomer, if the -CH2HgOCOCF3 occupies
the axial position, it will experience steric repulsion from H2. The closer the distance, the
stronger the interaction. Studies of models showed that the ratio of X (the distance of H3
to -QH2- in the a-isomer) to Y (the distance of H2 to -QH2- in the B-isomer) was about
3/5. That means that the axial orientation is much more restricted in the a-isomer than in
the B-isomer. Thus the (it-isomer gave excellent diastereselectivity but the B-isomer did

1101.

3. PREPARATION OF BOTH ISOMERS OF PENTANE-1,2-DIOL

Both isomers of pentane-l,2-diols can be obtained by a sequence similar to that shown in
Scheme 2.5. The a and B glycosides were synthesized as shown in Scheme 2.8. Since
Hg(II) does not complex with dialkyl alkenes as well as for terminal alkenes, a higher
temperature was needed to ensure high yield of the cyclization step. However, higher
diastereoselectivity was noticed due to the bulkiness of the side chain which is forced to

occupy the equatorial position exclusively (Table 2.1).

4. PREPARATION OF 4,6-O-BENZILIDENE-l,2-O-[2-lODOMETHYL-(R)-1,2-
ETHANEDIYLl-a-D-GLUCOPYRANOSE (11) AND B-
HYDROXYBUTYROLACTONE

The successful asymmetric synthesis of propane-1,2-diols and pentane-l,2-diols not only

opened ways to synthesize other alkyl and aryl diols, but also made it possible to

synthesize other chiral intermediates. The halo group is a very versatile group for

51

O O
O H 11
(eq.) (ax.)
HQOCOCF3
HQOCOCFa .
Favored Unfavored
a-isomer
HQOCOCF3
o o H
ween «zest»
HO HO
O
(9111-) (ax.)
HgOCOCF3
Favored I Unfavored
B-isomer

Figure 2.5

Comparison of steric hindrance in 01- and B-isomers

52

organic synthesis since it can be used to introduce a number of other important
functionalities. The 3-iodo intermediate (11) was synthesized with excellent
stereoselectivity as shown in Scheme 2.9. After OXymercuration, the mixture was treated
with excess NaCl and a catalytic amount of hexadodecyl trimethyl ammonium chloride to
form R*HgCI, which was then reacted with 12 to form R*I. Figure 2.6 shows its lH-
NMR and l3C-NMR spectrum. The signal at 3.13 and 3.05 ppm in IH-NMR spectrum
1.28 ppm correspond to two methylene protons and in the l3C-NMR spectrum
corresponds to the methylene carbon connected to I. The compound was also confirmed

by mass spectrometry.

With this iodo compound in hand, a number of chemical transformations can be
performed, such as cyanation, amination and chain elongation. In fact, (R)-3-
hydroxybutyrolactone was synthesized as shown in Scheme 2.9. The chiral 3-
hydroxybutyrolactone is a very versatile chiral building block and the (S)- isomer has
been used in the preparation of a chiral memberane probe [15], pH-sensitive liposomes

[16] and drug intermediates [17,18].

Reagents such as Brz, 12, NBS and NIS have also been tried to activate the C=C double
bond but with little success. This is not surprising since cyclization of alkenols activated
by a positive halogen source prefers to form five-member rings rather than six. However,
the smallest ring that can possibly form in the present case is six-membered. Other
activating reagents are worth searching for in the future in order to avoid the use of

mercury salt.

53

HgCNz, H90

 

MO 0
1.0%., + HOV
AGO OAC
Br

 

 

TiC|4
H910 4 a
0" \
V

 

Scheme 2.8

Synthesis of a- and B-cis-Z-pentenylglucosides

Table 2.1

Synthesis of chiral diols using D-glucose as a chiral auxiliary

 

 

 

 

 

 

 

 

 

Diolsa Diastereoselectivity Yield of desired e.e. (%)e
of cyclization diastereomer (%)d
(R)-l,2-propanediol 91 :9b 81.3 > 99
(S)-1,2-propanedio| 75:25b 58.1 > 99
(S)-l,2-pentanediol >99: 1° 60.3 > 99
(R)-l ,2-pentanediol >99: lc 65.8 > 99

 

a: Determined by comparing to standard alkane-1,2-diol using chiral column chromatography; b:
Determined by lH NMR of crude product; c: Determined by separation of crude product; d: Determined by
the weight of purified cyclized product; e: Measured by chiral column chromatography of 1,2—dihydroxy-
alkane after removal of the glucose moiety from the major diastereomer.

54

 

 

o o 1- H9(0C0CFa)2 xvo o NaCN
HO 2. NaCIIPTC HO pm:

 

 

 

 

 

0”o 3. l2 0 0
l
o
o o 1. H+
Ph/‘B/gA a O

HO 0 2. BF3, AczO,TFA

° 3. H OH',70° .r’

E ’02, HO
CN

(R)-3-hydroxybutyrolactone

Scheme 2.9

Synthesis of (R)-3-hydroxybutyrolactone from a 3-iodo substituted intermediate

55

NO 0
Ph 0
HO

 

 

 

 

 

 

T If I l I T' l I

I0 7! '10 6.5 ‘0 I! I. ll ‘0 3.3 30 —
av] v.1vyr:|v. .1. u ‘11,. IV. rr—v I v] War-WWW
“O Ila no 11° I." .9 .0 79 C. I. C. I. I. I. —

Figure 2.6

'H-NMR and '3C-NMR of compound (13): 4,6-O-Benzylidene-1,2-O-
[2-iodomethyl-(R)—l ,2-ethanediyl]-01—D—glucopyranose

56

The easy formation of Ot— and B-glucosides by 2-alkenols, the facile and stereoselective
cyclization of the 2-hydroxy group to the activated double bond, and a general method
for recovering the functionalized aglycon make the strategy described above for
functionalizing simple allylic alcohols attractive. The stereochemistry at the newly
formed chiral center can be chosen simply by deciding which anomer to start with. Both
Ot— and B-anomers can be synthesized through different routes and they are
interconvertable so that both enantiomers of diols can be prepared using the same chiral
auxiliary. This makes this method general for the selective hydroxylation of many simple
and complex allylic alcohols. Other chiral intermediates can also be easily produced via
the iodo compound, which makes this method even more versatile. Because different
anomers will lead to products of different configuration, optical purity must be ensured.
Two conventional ways to reach this end are enzymatic degradation using either 01- or [3-
glucosidase and separation after converting the mixture of on and B-glucosides to

crystalline derivatives.

57

Experimental

Optical rotaitions were measured using a Perkin polarimeter at 589 nm. 1H-NMR and
l3C-NMR spectra were measured in CDCI3 or D20 on a Varian-300 spectrometer
(300MHz). The chemical shifts are given in 5 values and are calibrated relative to the
chloroform line at 7.24 ppm for IH or 77.00 ppm for '3 C spectra. Silica gel flash column
chromatography was carried out on silica using Kieselgel 60 (Merck), 0.200-0.425 mm.
High resolution mass spectra was recorded on Chiral GC Column chromatography was
performed on HP 5890 on a bounded Betadex cyclodextron phase column (Supelco,
Bellefonte, PA) using helium as a carrier. Melting points were obtained with Fisher-

Johns Melting Point Apparatus.

Allyl 4,6-0-benzylidene-a-D—glucopyranoside (1)

D-glucose 18 g (0.1 mol), allyl alcohol 150 mL (2.2 mol) and dry Dowex-50wx-8 (H+)
resin (10 g) was stirred and heated under reflux (bath temperature: 120 °C) for 100 min.
After cooling, the resin was filtered off, washed with anhydrous ethanol (2 x 15 mL) and
the combined filtrate was concentrated. The residue was co-evaporated with a 1:1
mixture of benzene and anhydrous ethanol (2 x 30 mL), and the residual crude syrupy
allyl a-D-glucopyranoside was directly benzylidenated with 75 mL of benzaldehyde
dimethyl acetal (0.50 mol) in the presence of a catalytic amount of 4-toluenesulfonic acid
monohydrate by shaking for 15-20 h. The reaction mixture was diluted with n-hexane

(120 mL), then water (80 mL) and 10% aqueous sodium hydrogen carbonate (40 mL)

58

were added, the suspension was vigorously stirred for 15 min, and separated by filtration.
The solid was washed with 30 mL of hexane and dried, the crude yield was 15.4 g.
Recrystallized from methanol and water, the separated crystals of 1 (allyl 4,6-0-
benzylidene-a-D-gIucopyranoside) were isolated by filtration, washed with cold water
and dried. Yield: 11.2 g (37%). 1H NMR (300 MHz, CDC13)Z 5 3.47 (1H, dd, J = 9.21
and 9.21 Hz,), 3.60 (1H, dd, J = 9.21 and 3.91 Hz), 3.70 (1H, dd, J = 10.19 and 10.19
Hz), 3.81 (1H, m), 3.92 (1H, dd, J= 9.21 and 9.21 Hz), 4.02 (1H, m), 4.22 (1H, m), 4.25
(1H, dd, J = 9.91 and 4.75 Hz), 4.90 (1H, d, J = 4.2), 5.20-5.35 (2H, m), 5.50 (1H, s),
5.90 (1H, m), 7.30-7.50 (5H, m); 13C NMR (75 MHz, CDC13): 6 97.9, 101.9, 118.2,

133.3.

4,6-O-Benzylidene-1,2-0—[2-methyl-(R)-1,2-ethanediyll-a-D-glucopyranose (2)

1 (1.60 g, 5.2 mmol) in acetonitrile (200 mL) was added dropwise mercury trifluoracetate
(4.40 g, 10.4 mmol) in 100 mL CH3CN. The mixture was stirred at RT for 6 h after
which 10 mL of 3M NaOH solution was added and stirring was continued for a further 10
minutes. Sodium borohydride (0.79 g, 20.8 mmol) dissolved in 10 mL of 3M NaOH
solution was added and the mixture was stirred for 3 h. The suspension was then filtered
through celite and the clear filtrate was concentrated under vacuum at 30 °C to remove
acetonitrile. The white precipitate was recovered by filtration and purified by column
chromatography (hexane/ethyl acetate = 5: 8). Yield: 1.30 g (81.3%). mp (crystallized
spontaneously upon standing): 215.0-218.0 °C. IH NMR (300 MHz, CDCI3): 5 1.09 (3H,
d, J = 6.1 Hz), 3.46 (1H, dd, J = 11.9, 10.7 Hz), 3.49 (1H. dd, J = 9.4 Hz), 3.67 (1H, dd,

J = 10.3 Hz,), 3.71 (1H, dd, J = 9.4, 3.7 Hz), 3.89-4.00 (3H, m), 4.30 (1H, dd, J = 10.3,

59

5.2 Hz), 4.47 (1H, dd, J : 9.3 Hz), 4.95 (1H, d, J = 3.4 Hz), 5.50 (1H, s), 7.35 (2H, m),
7.48 (2H, m); I3c NMR (75 MHz, CDC13): a 16.07, 63.44, 64.08, 65.76, 68.75, 72.09,
75.37, 80.87, 94.69, 101.88, 126.29, 128.35, 129.29, 136.93; HRMS Exact mass: calcd

for C16H2006 M“, 308.1260, found 308.1262; [61020: +21° (c 2.08, CHC13).

(R)-Propane-1,2-diol (3)

To recover the chiral diol, 1.30 g, (4.2 mmol) of the product 2 was debenzylidenated by
treatment with 100 mL 1N HCI at 70 °C for 1 h. Benzaldehyde was removed by
extraction with chloroform. The aqueous layer was concentrated and acetolyzed with
acetic anhydride (3 mL), trifluoroacetic acid (2 mL) and boron trifluoride diethyletherate
(1.5 mL) at RT for 5 h. The diol was recovered by first diluting with 100 mL chloroform,
washing with water (30 mL) and saturated sodium bicarbonate (2x20 mL), concentrating
and heating with 1.1 g of NaOH and 9.0 g of H202 (30%) at 70 °C for 10 h. It was then
desalted on a Dowex MR-3 mixed bed ion exchange resin and evaporated under reduced
pressure to dryness. Yield: 0.30 g (95%). 1H NMR (300 MHz, CDCI3): 8 1.14 (3H, d, J
= 6.4 Hz), 3.38 (1H, dd, J = 11.2 and 7.8 Hz), 3.61 (1H, dd, J = 11.1 and 3.1 Hz), 3.89

(1H, m); l3C NMR (75 MHz, CDC13): a 18.66, 67.84, 68.28; e.e. > 99%.

Ally] 4,6-O-benzylidene-B-D-glucopyranoside (4)

4 was recrystallized from the mother liquor as the side product during the synthesis of 1.
Yield: 4.20 g (13.6%). mp (crystallized from methanol and water): 75.0-78.0 oC. lH
NMR (300 MHz, CDCI3): 6 3.46 (2H, m), 3.56 (1H, dd, J = 9.2 and 9.2 Hz), 3.78 (1H,

dd, J = 9.8 and 9.8 Hz), 3.82 (1H, dd, J = 8.9 and 8.9 Hz), 4.14 (1H, dd, J = 12.4 and 6.1

60

Hz), 4.34 (1H, dd, J = 10.5 and 4.7 Hz), 4.39 (1H, dd, 12.4 and 3.9 Hz), 4.45 (1H, d, J =
7.8 Hz), 5.25 (1H, dd, J = 10.6 and 1.4 Hz), 5.33 (1H, dd, J = 17.3 and 1.40 Hz), 5.53
(1H, s), 5.94 (1H, m), 7.36 (1H, m), 7.48 (2H, m); l3C NMR (75 MHz, CDC13): a 66.28,
68.58, 70.58, 73.04, 74.37, 80.46, 101.83, 102.07, 118.35, 126.24, 128.31, 129.24,

133.39, 136.89; HRMS Exact mass: calcd for CI6H2006 M“, 308.1260, found 308.1260;

[61],)”: -590" (c 2.3915, MeOH).

4,6-O-Benzylidene-1,2-0-[2-methyl-(S)-1,2-ethanediyll-B-D-glucopyranose (5)

By the same process used to convert 2 to 3, 3.32 g of 4 gave compound 5 after
purification. Yield: 1.93 g (58.1%). mp (crystallized from acetone): 189.0-192.0 °C. lH
NMR (300 MHz, CDCI3): 5 1.13 (3H, d, J = 6.1 Hz), 2.61 (1H, d, J = 2.2 Hz), 3.29 (1H,
dd, J = 7.8, 7.8 Hz), 3.48 (1H, dd, J = 10.5 and 10.5 Hz), 3.61 (1H, dd, J = 6.8 and 2.4
Hz), 3.61 (1H, t, J = 9.5 Hz), 3.87 (3H, m), 3.93 (1H, dd, J = 11.7 and 2.7 Hz), 4.36 (1H,
dd, J = 10.5 and 4.4 Hz), 4.39 (1H, d, J = 7.6 Hz), 5.53 (1H, s), 7.36 (3H, m), 7.48 (2H,
m); l3C NMR (75 MHz, CDC13) 8 68.04, 68.45, 70.85, 71.69, 72.09, 76.57, 77.41, 79.72,
81.30, 98.61, 102.16, 126.25, 128.36, 129.35, 136.73; HRMS Exact mass: calcd for

C16H2006 M“, 308.1260, found 308.1271; MD”: +10.27° (c 3.088, CHCI3).

(S)-Propane-1, 2-diol (6)

Starting from 5 (1.00 g), 6 was synthesized in the same way as 3. Yield: 0.24 g (96.1%).
lH NMR (300 MHz, CDC13) 6 1.14 (3H, d, J = 6.4 Hz), 3.38 (1H, dd, J = 11.2, 7.8 Hz),
3.61 (1H, dd, J = 11.1 and 3.1 Hz), 3.89 (1H, m); l3C NMR (75 MHz, CDC13) 5 18.66,

67.84, 68.28; e.e. > 99%.

61

Cis-Z-pentenyl-B-D-glucopyranoside (7)

To dry benzene (200 mL) was added Hg(II)CN2 (1 1.8 g, 46.7 mmol), HgO (2.00 g, 9.23
mmol), acetobromo-a-D-glucose (9.60 g, 23.3 mmol) and cis—2-penten-l-ol (10.00 g,
116.1 mmol). The mixture was stirred under N2 at RT for 20 h. The solid was filtered
off and the filtrate was concentrated using a rotavap. After dilution with CHC13 (300
mL), the solution was washed with 1N HCI (50 mL), sat. NaHC03 solution (50 mL), and
brine. The organic layer was dried with Na2SO4 and evaporated to give a colorless syrup,
to which was added NaOH (5.60 g) and H20 (100 mL). The mixture was heated at 70 °C
overnight to give a clear solution, which was acidified to pH = 4 by adding con. HCI.
After removal of the solvent the residue was extracted with acetone (3 x 200 mL). The
combined acetone portions were dried evaporated to dryness to give product 7. Yield:
4.92 g (84.9%); mp (crystallized from MeOH/acetone): 99.0-102.0 °C; ]H NMR (300
MHz, D20) 5 0.74 (3H, t, J = 7.6 Hz), 1.89 (2H, p, J = 7.5 Hz), 3.04 (1H, dd, J = 8.5 and
8.5 Hz), 3.15 (1H, dd, J = 8.6 and 8.6 Hz), 3.20 (1H, m), 3.25 (1H, dd, J = 8.6, 8.6 Hz),
3.49 (1H, dd, J = 12.3, 5.1 Hz), 3.69 (1H, dd, J = 12.3, 1.7 Hz), 4.15 (1H, m), 4.25 (1H,
d, J = 8.1 Hz), 5.33 (1H, m), 5.58 (1H, m); l3C NMR (75 MHz, D20): 5 8.92, 15.70,
56.10, 60.04, 64.97, 68.43, 71.22, 71.30, 96.06,. 118.28, 133.75; HRMS Exact mass:

calcd for C19H29010 (the peracetylated form of 7) [M+H] +', 417.1761, found 417.1766;

[611020; -28.4° (c 1.28, MeOH).

1,2-0-[2-propyl-(S)-1,2-ethanediyll-B-D-glucopyranose (8)
To 7 (0.63 g 2.5 mmol) in acetonitrile (100 mL) was added mercury trifluoracetate (2.17

g, 5.08 mmol). The mixture was stirred at room 50 °C for 18 h after which 5 mL of 3M

62

NaOH solution was added and stirring was continued for a further 10 minutes. Sodium
borohydride (0.19 g, 5.1 mmol) dissolved in 5 mL of 3M NaOH solution was added and
the mixture was stirred for 3 hour. The suspension was then filtered through Celite and
the clear filtrate was concentrated under vacuum at 30 °C to remove the acetonitrile. To
the residue was added 10 mL of H20 and the pH ‘was adjusted to 5 by adding con. HCl
dropwise. After removing the solvent by evaporation, the white solid was extracted with
acetone (3 x 100 mL). The combined acetone portions were evaporated to dryness to
give 0.58 g of crude product. Purification by column chromatography (CHC13/Me0H =
10/2.5) gave product 8. Yield: 0.38 g (60%). mp (crystallized spontaneously on
standing): 1010-1040 0C; IH NMR (300 MHz, D20) 8 0.74 (3H, t, J = 6.8 Hz), 1.25
(4H, m), 2.97 (1H, dd, J = 9.8 and 7.8 Hz), 3.30 (1H, dd, J = 10.1 and 8.8 Hz), 3.39 (1H,
dd, J = 10.5, 4.2 Hz), 3.41 (1H, m), 3.44 (1H, dd, J = 9.8 and 8.8 Hz), 3.57 (1H, m), 3.57
(1H, dd, J = 12.5 and 5.4 Hz), 3.72 (1H, dd, J = 12.5, and 2.0 Hz), 3.86 (1H, dd, J = 12.0
and 2.4 Hz), 4.28 (1H, d, J = 8.0 Hz); ”C NMR (75 MHz, D20): 8 8.69, 13.23, 27.36,
55.85, 65.32, 65.74, 68.19, 70.83, 72.93, 7409,9286; HRMS Exact mass: calcd for
C17H2609 (the peracetylated form of 8) [M+H] J" = 375.1655. Found 375.1667; [01]D20 =

+3592o (c 4.29, MeOH).

Recovery of (S)-1,2-dihydroxypentane (9)
Starting from 8 (0.35 g, 1.4 mmol), 9 was synthesized the same way (without the step of
removing the benzilidene acetal) as 3. Yield: 0.14 g (95% from 8). lH NMR (300 MHz,

CDCI3): 8 0.91 (3H, 1, J = 7.2 Hz), 1.37 (4H, m), 3.39 (1H, dd, J = 11.2, 7.8 Hz), 3.61

63

(1H, dd, J = 11.1, 3.2 Hz), 3.68 (1H, m); ”C NMR (75 MHz, CDC13) 8 14.03, 18.71,

35.21, 66.76, 72.02; e.e. > 99%.

Cis-Z-pentenyI-a-D-glucopyranoside (10)

A solution of TiCl4 (2.62 mL, 43.4 mmol) in 50 mL of dry CHC13 was added to 200 mL
of dry CHC13 solution of tetraacetyI-cis-pentenyl-B-D-glucopyranoside (9.04 g, 21.7
mmol). The solution was boiled gently for 4h at 65 °C. The solution was then poured
into cold water (250 mL) and CHC13 (250 mL). The organic layer was washed with
dilute NaHC03 solution (250 mL), twice with water and once with brine. The organic
layer was then dried with Na2SO4, and evaporated under rotavap. Purification by column
chromatography (hexane/ethyl acetate = 1/1) gave tetraacetyl-cis-penteny1-a-D-
glucopyranoside (3.67 g) to which was added NaOH (2.10 g, 52.5 mmol) and 100 mL of
H20. The mixture was heated at 70 °C overnight to give a clear solution, which was
acidified to pH = 4 by adding con. HCl. The solution was dried under vacuum and the
residue was extracted with acetone (3 x 200 mL). The combined acetone portions were
removed under vacuum to give product 10. Yield: 2.16 g (39.0%). mp (crystallized from
MeOH/acetone): 95.0-98.0 °C. lH NMR (300 MHz, D20): 5 0.79 (3H, t, J = 7.6 Hz),
1.93 (2H, m, J = 7.6 Hz), 3.23 (1H, dd, J = 9.3 and 9.3 Hz), 3.37 (1H, dd, J = 9.8 and 3.7
Hz), 3.50 (1H, m), 3.51 (1H, dd, J = 9.8 Hz), 3.58 (1H, dd, J = 12.2 and 4.6 Hz), 3.68
(1H, dd, J = 12.1 and 2.4 Hz), 3.97 (1H, dd,J = 11.9 and 7.3 Hz), 4.10 (1H, dd,J = 12.1
and 6.6 Hz), 4.77 (1H, d, J = 3.9 Hz), 5.40 (1H, m), 5.58 (1H, m); l3C NMR (75 MHz,

D20) 5 9.00, 15.78, 55.86, 58.34, 64.92, 66.58, 67.20, 68.51, 92.74, 118.71, 133.22;

64

HRMS Exact mass: calcd for C19H29010 (the peracetylated form of 10) [M+H] +°

417.1761, found 417.1766; [61030 = +151 .66° (c 3.87, MeOH).

1,2-0-[2-propyl-(R)-1,2-ethanediyll-a-D-glucopyranose (11)

Starting from 10 (1.20 g, 4.84 mmol), 11 was synthesized the same way as 8. Yield: 0.79
g (66%). mp (crystallized spontaneously on standing): 151.0-154.0 °C. 1H NMR (300
MHz, CDC13): 8 0.75 (3H, t, J = 6.8 Hz), 1.23 (4H, m), 3.31 (1H, dd, J = 9.3 and 9.3
Hz), 3.41 (1H, dd, J = 11.7 and 11.7 Hz), 3.47 (1H, dd, J = 9.9 and 3.4 Hz), 3.60 (1H,
m), 3.85 (1H, m), 3.87 (1H, dd, J = 12.0 and 2.4112), 4.12 (1H, dd, J = 9.5 and 9.5 Hz),
4.88 (1H, d, J = 3.4 Hz); 13C NMR (75 MHz, CDC13): 5 8.68, 13.19, 27.42, 55.59, 62.33,
62.83, 64.45, 65.95, 69.38, 69.86, 89.61; HRMS Exact mass: calcd for C17H2609 (the
peracetylated form of 11) [M+H]“ 375.1655, found 375.1660; [a]020 = +72.80 (c 1.89,

MeOH).

Recovery of (R)-1,2-dihydroxypentane (12)

Starting from 11 (0.60 g, 2.4 mmol), 12 was synthesized the same way as 9. Yield: 0.24
g (96%). lH NMR (300 MHz, CDC13): 8 3.68 (1H, m), 3.61 (1H, dd, J = 11.1 and 3.2
Hz), 3.39 (1H, dd, J = 11.2 and 7.82 Hz), 1.37 (4H, m), 0.91 (3H, t, J = 7.2 Hz); l3C

NMR (75 MHz, CDC13) 5 14.03, 18.71, 35.21, 66.76, 72.02; e.e. > 99%.

4,6-0-Benzylidene-1,2—O-[2-iod0methyl-(R)-1,2-ethanediyll-a-D—glucopyranose (13)
Compound 1 (1.60 g, 5.19 mmol) in acetonitrile (200 mL) was added dropwise mercury

trifluoracetate (4.40 g, 10.4 mmol) in 100 mL CH3CN. The mixture was stirred at RT for

65

6 h after which 1.50 g of NaCl and a catalytic amount of hexadecyltrimethylammonium
bromide were added and stirred for another 6 h. After filtration, 1.45 g of 12 was added to
the clear filtrate and the mixture was stirred at RT for 10 h. CH3CN was removed by
rotary evaporator and the residue was diluted with 200 mL CHC13. After filtration, the
filtrate was washed with saturated Na2S203 solution and subsequently with brine. The
product was purified by column chromatography (hexane/ethyl acetate=2:1). Yield:
1.32g (58.8%). mp: 204-206 0C. lH NMR (300 MHz, CDC13)I 6 3.05 (1H, dd, J = 10.62
and 6.22 Hz), 3.13 (1H, dd, J = 10.62 and 5.50 Hz), 3.51 (1H, dd, J = 9.28 and 9.52 Hz),
3.56 (1H, dd, J = 11.48 and 10.50 Hz), 3.68 (1H, dd, J = 10.25 and 10.25 Hz), 3.78 (1H,
dd, J = 9.52 and 3.42 Hz), 3.86 (1H, m), 3.96 (1H, m), 4.22 (1H, dd, J = 11.72 and 2.68
Hz), 4.31 (1H, dd, J = 10.38 and 4.96 Hz), 4.44 (1H, dd, J = 9.28 and 9.28 Hz), 4.97
(1H, d, J = 3.42 Hz), 5.51 (1H, s), 7.36 (3H, m), 7.48 (2H, m); l3C NMR (75 MHz,
CDC13): 6 1.28, 64.20, 65.94, 66.82, 68.69, 70.05, 75.64, 80.63, 94.69, 101.92, 126.29,

128.35, 129.32, 136.70; HRMS Exact mass: calcd for C16H19061 [M+H]“ 435.0305,

found 435.0318; [611,30 (c 1.25, DMF) = +27.4°.

III. 4,6-0-Benzylidene-1,2-0-[2-cyano-(R)-1,2-ethanediyll-a-D-glucopyranose (14)

1.00 g of 13 in dry DMF was added 2.05 g of NaCN and the mixture was heated at 50 °C
for 2 h. The bulk of the DMF was removed by rotary evaporation and xylene was added
to remove the residual DMF under reduced pressure. The residue was diluted with 50
mL CHCI3 and washed with 50 mL H20 and brine. After drying with Na2SO4, the
solvent was removed under reduce pressure. Yield: 0.62g (81%). mp: 245-248 °C. 1H

NMR (300 MHz, CDC13): 8 2.57 (1H, d, J = 1.2 Hz), 2.59 (1H, d, J = 1.2 Hz), 3.53 (1H,

66

 

dd, J = 9.6 Hz and 9.6 Hz), 3.70 (1H, dd, J = 10.50 and 10.50 Hz), 3.72 (1H, dd, J =
12.15 and 11.70 Hz), 3.84 (1H, dd, J = 9.6 and 3.6 Hz), 3.98 (1H, m), 4.11 (1H, dd, J =
12.3 and 2.7 Hz), 4.15 (1H, m), 4.33 (1H, dd, J = 10.35 and 4.8 Hz), 4.44 (1H, dd, J =
9.6 and 9.6 Hz), 5.02 (1H, d, J = 3.6 Hz), 5.53 (1H, s), 7.39 (3H, m), 7.43 (2H, m); ”C
NMR (75 MHz, CDCI3): 8 20.04, 62.87, 64.16, 65.75, 68.62, 69.45, 75.57, 80.60, 94.49,
101.96, 126.25, 128.39, 129.41, 134.9; HRMS Exact mass: calcd for C17H19096N

[M+H]“ = 334.1291, found 334.1288; [61020 = +14.7°.

(R)-3-hydroxybutyrolactone (15)

14 (0.50g) was synthesized using the same sequence as 3, namely, removal of the
benzylidene acetal followed by acetolysis and oxidative cleavage. After passage through
the mixed bed resin, the pH of the solution was adjusted to 2 and the solvent was
removed by rotary evaporator. Yield: 0.10 g (63%). lH-NMR (CDCI3): 5 2.51 (1H, dd, J
= 18.15 and 1.8 Hz), 2.75 (1H, dd, J = 18.15 and 6.31 Hz), 4.28 (1H, dd, J =10.5 and
2.7 Hz), 4.41 (1H, dd, J = 10.2 and 4.5 Hz), 4.70 (1H, m); l3C-NMR: 37.8, 67.5, 76.1,

176.5.

67

.95»

990.9954

11.
12.
13.
14.
15.
16.
17.
18.

References

. Finn, M. G., Sharpless, K. B. Comprehensive Organic Synthesis, Pergamon Press,

1991, 7, 389-436.

Finn, M. G., Sharpless, K. B. Asymmetric Synthesis, Academic Press, New York,
1985, 5, 247.

Kunz, H.; Riick, K. Angew. Int. Ed. Engl. 1993, 32, 336-358.

Charette, A. B.; Cote, B.; Marcoux, J. —F. J. Am. Chem. Soc. 1991, 113, 8166-8167.
Charette, A. B.; Cote, B. Tetrahedron: Asymm. 1993, 4, 2283-2286.

Gurjar, M. K.; Mainkar, A. S. Tetrahedron: Asymm. 1992, 3, 21-24.

Bellucci, G.; Chiappe, C., D’Andrea, F. Tetrahedron: Asymm. 1995, 6, 21-23.
Cherest, M.; Felkin, H.; Prudent, N. Tetrahedron Lett. 1968, 2199.

Koga, K.; Yamada, S. Chem. Pharm. Bull. 1972, 20, 526.

. Fleming, 1.; Sarkar, A. K.; Thomas, A. P. Chem. Commun. 1987, 157.

Midland, M. M.; Kwon, Y. C. J. Am. Chem. Soc. 1983, 105, 3725.

Nakada, M.; Urano, Y.; Kobayashi, S.; Ohno, M. J. Am. Chem. Soc. 1988, 110, 4826.
Barrett, A. G. M.; Flygare, J. J. Org. Chem. 1991, 56, 638.

Majewski, M.; Clive, D. L. J .; Anderson, P. C. Tetrahedron Lett. 1984, 25, 2101.
Huang, G. and Hollingsworth, R. I. Tetrahedron 1998, 54, 1355-1360.

Song, J. and Hollingsworth, R. I. J. Amer. Chem. Soc. 1999, 121, 1851-1861.

Wang, G. and Hollingsworth, R. I. J. Org. Chem. 1999, 64, 1036-1038.

Huang, G. and Hollingsworth, R. I. Tetrahedron Asymm. 1999, 9, 4113-41 15.

68

CHAPTER 3

Stereoselective Reductions of Ketones Using D-Glucose as a Chiral Auxiliary

69

Introduction

As shown in CHAPTER 2, we have succeeded in the stereoselective synthesis of (R)- or
(S)-alkane-1,2-diols and other chiral intermediates using D-glucose as a chiral auxiliary
[1]. Though the stereoselectivity of this methodology was excellent, the use of mercury
salts and the sequence used to remove the D-glucose auxiliary were drawbacks of this
method. One alternative approach is to reduce ketones stereoselectively after linking
them to D-glucose, followed by cleavage of the glycosidic bond to retrieve the chiral
diols. In order to accomplish this stereoselective reduction, 3 second point of interaction
of the reduction site with the chiral scaffold (auxiliary) has to be established or the

scaffold can be used to deliver the reagent stereoselectively.

The use of carbohydrates as chiral auxiliaries was not intensely investigated until
recently. Only a limited number of papers have been published regarding its use in
stereoselective reductions of ketones. In 1994 Akiyama found that either mandelate
stereoisomer could be obtained from the reduction of chiro-inositol phenylglyoxylate
ester (A) by simply altering the reaction conditions (Scheme 3.1, a) [2]. K-Selectride
selectively attacked the re face of A in 320 solution (Yield: 54%; B:C =4921), but the
reaction in THF in the presence of 18-crown-6 led to reduction at the si face (Yield: 66%;
B:C=1 :24). Selective 1,2-reduction of a-D-mannosyl enones D (Scheme 3.1, b) with a
very bulky hydride reagent gave (R) allylic alcohols E as the major components in 9-19:1

mixtures of diastereomers [3]. Higher stereoselectivity was obtained when the enone

70

0 O 0
Ph K-Selectride Ph K-Selectn'de Ph
R'O T R10 ? we (a)

H OH 0 18-cr-6 HO H
B A C
54%; 49:1 | 66%; 24:1

R* = 1L-3-O-(t-Butyldimefliylsilyl)-1,2:5,6-di-O-cyclohexylidene-chiro-inositol

CH H3 H3
81.6%g'tawmbtcn"ccflgkfltm
CH

 

   

R
D E
78-89% ; 9-19: 1
0
RI
1.1
. E O R.
R0110" -———-> ROI".- ....1 (C)
ZnClleaBH4 H o o
F H
R=CH3, Bn >90%; 6-28:1 80-82%; 72-93% e.e.
R'=CH3,C5H5

Scheme 3.]

Reductions of carbonyl group using carbohydrate derivatives as chiral auxiliaries

71

contained bulkier groups (Glycosidic enones are essentially masked a-diketones of a-
diketoesters). A remarkable level of remote stereochemical induction was obtained in
reduction of the y-ketoesters F, incorporating a bicyclic anhydro-D-glucose auxiliary
(Scheme 3.1, b) [4]. Reduction of the ketone with Zn(BH4)2 gave up to 28:1
diastereoselectivity. The resulting hydroxyesters C were hydrolyzed, and the products
isolated as lactones H with 72-93% e.e.. 0n the other hand, analogous B-ketoesters were
reduced with no more than 3.521 selectivity. The authors proposed that chelation of the
reagent was essential for obtaining selectivity, since NaBH4 was nonselective towards

either type of ketone.

In all these experiments good to excellent stereoselectivity was observed using modified
or fimctionalized carbohydrate derivatives as chiral auxiliaries, demonstrating the great
potential carbohydrates could have in this area. However, the use of otherwise
unfunctionalized carbohydrates as chiral auxiliaries, which will highly improve efficiency
in synthesis, is less investigated. We are therefore interested in the more direct approach
of stereoselective reduction of ketones using unprotected D-glucose as the chiral
auxiliary. In this approach, the substrate can be linked to glucose via the glycosidic bond.
Again, we think the key is to utilize the 2-hydroxy group to direct the attack of the

reducing agent.

It is well known that hydroxy groups could play a critical role in stereoselective reduction

of ketones. In 1983, it was discovered that NaBH(0Ac)3 can effect the reduction of

ketones bearing hydroxyl groups suitably positioned for interaction with the borohydride

72

anion [5-10]. The proposed mechanism involves two steps, a ligand exchange reaction at
boron by the substrate (ROH) to form NaBH(0Ac)20R followed by intramolecular
delivery of hydride to the carbonyl carbon. More recently it has been demonstrated that a
y-hydroxy group can deliver hydride to a carbonyl site as well. Similarly,
Me4NBH(0Ac)3 effected the reduction of acyclic B-hydroxy ketones, but ketones, B-keto
esters and simple B-diketones which lack a suitably disposed hydroxyl group are not
reduced under the standard conditions. Consideration of chair-like transition structures
provides a plausible rationale for the observed diastereoselectivities (Scheme 3.2. a) [11-
14]. The intramolecular delivery of hydride in Me4NBH(0Ac)3 reductions is supported
by the stereochemical outcome of the reaction of both cyclic and acyclic systems.
Generally, anti-1,3-diols are generated by this type of reaction. However, syn-1,3-diols
are obtained preferentially if reduction goes by chelate-controlled addition of hydride
reagents (Scheme 3.2. c). If the ketones bear 01- or B-heteroatoms, reducing agents such

as Zn(BH4)2 can also invoke the intramolecular reduction (Scheme 3.2. c) [15].

We started from a simple ketone with the glucose unit, 2-oxopropyl-a-D-
glucopyranoside. Unfortunately, the reduction which tried to utilize the free 2-OH group
with NaBH(0Ac)3 failed. This is not totally unexpected since the 2-hydroxy group is at
the 5-position of keto group, which is a little too far away. Our attention turned to the use
of reagents with counterions capable of forming multipoint contacts or complexes. The
proper complexing reagents will not only bind to oxygen atoms to limit the freedom of
the aglycon, but also direct the .H' attack on the carbonyl group. Our study showed good

diastereoselectivity could be achieved using Ca(BH4)2 (or CaC12 with NaBH.;) as a

73

 

o
R/[krow 21(13an
R.

 

 

 

-¢

 

 

H H OAc
. |_
R1 li‘o’j'BTNOAc
R2/ ‘H
Internal Hyd'ide Delively
r .. ‘ i
H'\ M
R1 \ 0"”; ——>
C\ ’
/ \O
R2
_ .1
External Hyd'ide Delivery
4’1
xi'zln‘
H
00R \

 

lntemal Hym'de Delivery

Scheme 3.2

Stereoselective reduction by internal and external hydride attack

74

(b)

F: 8.11.211:

 

reducing agent on 2-oxopropyl-a-D-glucopyranoside and benzoylmethyl-a- D-
glucopyranoside under certain reaction conditions. After the reduction, the glucose unit
can be removed to retrieve (R) diols as the major product without racemization. Using
the same scheme, 2-oxopropyl-B-D-glucopyranoside gave (S) diols as the major product.
A working hypothesis is proposed to explain these results and good to excellent

stereoselectivity is expected if similar reaction conditions are applied to other ketones.

75

1.;—m,

Results and discussions

2-0xopropyl-a-D-glucopyranoside, synthesized from allyl a-D-glucopyranoside using
the Wacker process, was reduced under a number of conditions. The glucose unit was
then cleaved (Scheme 3.3). Table 3.] lists the results of the key step, i.e. stereoselective
reduction of the keto group. Due to the polar nature of 2-oxopropyl-a-D-
glucopyranoside, the solvents that can be used were limited to H20, MeOH and EtOH.
Ethanol gave the best d.e. possibly because of its lower polarity which reduced the
tendency of the complex to dissociate. Isopropanol could only be used with the addition
of MeOH as a co-solvent and the reaction could not be conducted at low temperature due
to the solvent freezing. A divalent metal salt was added to effect complexation involving
the oxygen atoms, likely l0 and 20. Given the strong complexing power of Ca2+ with
carbohydrates [16,17], it is not surprising that CaCl2 turned out to be the best choice

compared to LiCl, MgCl2 and ZnCl2.

Cat. PdClz
> HO

O
HO OHo 1.4-Benzoquinone HO OH \/U\
W leTHF 0
OH
1
\/\

Ca(BH4)2 ”OH/3% OH 11*
EtOH ”0 01-1 § —" HO
-78°C °\/\
d.e. = 69% e.e. = 69%
Scheme 3.3

Stereoselective reduction of 2-oxopropyl-01-D-glucopyranoside

76

Table 3.1

Diastereoselectivity obtained under different conditions

 

 

 

 

 

 

 

 

 

 

 

 

Entry Reaction conditions Yield(%) d.e.(%)a
i H20/RT/NaBH4 96 <5
ii LiBH4/0 0C/EtOH 96 20
iii ZnCl2/NaBH4/MCOH/O 0C 89 <5
iv LiCl//NaBH4/-78 0C/EtOH 89 47
v MgCI2/NaBH4/-78 °C/EtOH 86 <5
vi CaC 12/NaBH4/-78 OC/EtOH 93 69

vii MeOH/—78 °C/Ca(BH4)2.2THF 96 60
viii EtOH/O °C/Ca(BH4)2.2THF 96 33
ix EtOH/-78 °C/Ca(BH4)2.2THF 93 69
x i-PrOH/MeOH( 10/1 )/ 86 52

 

-40 oC/Ca(BH.;)2.2THF

 

 

 

a: d.e. was calculated based on lH-NMR of crude reduction product

77

 

The reducing agents investigated were NaBH4, LiBH4 and Ca(BH4)2.2THF because they
were compatible with these solvent systems. The best condition found was
Ca(BH.;)2.2THF /Et0H/—78 0C, which gave a d.e. of 69% (d.e. was determined by
integration of methyl signal in the 300 MHz 1H-NMR spectra). Ca(BH4)2 played a dual
role as complexing reagent and reducing agent. In all cases, (R)-propane-l,2-diol was
obtained as a major product while the (S)-form was the minor product after removal of
the aglycon under acidic conditions (The absolute configuration and e.e. was determined
by comparing the chiral-GC spectra of final products and standard optically pure
propane-1,2-diols). This can be rationalized as shown in Figure 3.1. Conformer I is
more favorable than Conformer 11 because of the strain between methyl group and 2-OH
of glucose unit in Conformer B. Since Ca(II) is the counterion for borohydride, the
hydride will approach the carbonyl group from the same side as the Ca2+ giving mainly

(R)-propane- 1 ,2-diol.

H H H
°” 0 H ” 9135a» H
,Ca LCa
H o’ H 0-1
H H'
Ho MeH HO 0 H
HO H Ho H
I II
Figure3.l

Working model for the stereochemical outcome
of a-glycoside

If this model holds true, ketones with alkyl groups other than methyl will likely show

better selectivity because of the increase of steric overlap between the alkyl group and the

78

 

 

2-hydroxy group of the glucose unit. But what will happen with different substrates such
as aryl ketones? To answer this, benzoylmethyl-a-D-glucopyranoside was synthesized
(Scheme 3.4) using a method similar to the one published by Lubineau et al [18].
Ca(BH4)2.2THF was the reducing agent of choice based on its success in the earlier case.
However, using EtOH as the solvent gave only about 20% d.e. at 0 °C and 59% at —78 °C.
If the reduction was done in i-PrOH at —78 °C, the d.e. was improved to 62%. The
absolute configuration of the major product after cleavage of the auxiliary was
determined to be (R)—1-phenyl-ethane-1,2-diol. This result can be rationalized by the
same model shown in Figure 3.1. The slight decrease of d.e. could result from the size
difference between phenyl and methyl groups. Acid hydrolysis using strong mineral acid

to cleave the chiral auxiliary could not be applied here due to racemization at the

AcO O BuNH AcO 0
OAc 2 OH
AcO’gA/ ’ AcO’gA/
AGO ACO A60 A60

NaH, Bu4NI AcO O KCN HO O
> —>
CeHsCOCHzBr AC113:0 0 MeOl-l "30 OH 0

AcO
CH2CI2 O

HO O OH
C BH 1.NalO ;
-..a(—‘)1 “33% 0H _‘, Ho 1
IPI'OH OH 3 . \/\Ph

-78°C O\/\ Ph

 

d.e. = 52% e.e. = 62%
Scheme 3.4

Stereoselective reduction of benzoylmethyl-a-glucopyranoside

79

 

 

benzylic position. Sodium periodate was utilized to open up the sugar ring by oxidative
cleavage after which the pH was adjusted to l for one hour to release l-phenyl-1,2-
ethanediol.lf the steric overlap between the alkyl group and the 2-hydroxy group of the
glucose unit is the major factor in the stereochemical outcome, one may expect that the B
form glycoside will give mainly the other isomer based on Figure 3.2. The experiment
starting from 2-oxopropyl-B-D-glucopyranoside followed the prediction. 2-0xopropyl-
a-D-glucopyranoside and 2-oxopropyl-B-D-glucopyranoside was synthesized using
conditions similar to that reported by Lubineau et al [7] and the reduction was done under
the same conditions used for 2-oxopropyl-a-D-glucopyranoside. The decrease of d.e.
(33% compared to 67% for 2-oxopropyl-01-D-glucopyranoside) could be due to less
favorable complexation of Ca2+ with l0 and 20. Ca2+ complexes with cis —20H and —

lOR better than with trans —20H and —'OR.

H H'Me H H H H' O H
HO H 96L” Homoflfl
H0 H H 63" o H Ho H H 6H3?“ H
111 [V
Figure 3.2

Working model to predict the stereochemical outcome
of B-glycoside

Stereoselective reduction of ketones using D-glucose as a chiral auxiliary can lead to C3

chiral building blocks that are critical components of a number of biologically active

compounds (Figure 3.3). For instance, the Fisher glycosidation of D-glucose and acetol

80

 

 

 

I' ‘s H
O ,I \ N
.‘ : \l/ Adrenergic antagonist
\ I
(8)-Propanolol
/
0
0C16H33
O
0 I o ii
/ ’ ‘
N+ I “
//P—o/\/ \ :1 : Y
0 \ ‘\ It,
0' “ .... o’
Platelet activation factor (S)-Metoprolol

 

Captopril

Figure 3.3

Some biologically active compounds containing C3 chiral synthons

81

1

If

(hydroxyacetone) will produce both 0t and [3 forms of 2-oxopropyl-D-glucopyranoside.
Enzymatic degradation using or or [3 glucosidase will yield the pure [3 or or 2-oxopropyl-
D-glucopyranoside. After a-halogenation at the terminal position, stereoselective
reduction followed by removal of glucose will produce mainly one C3 chiral synthon.
Another application of 3-halo-2-oxopropyl-D-glucopyranoside is to make more complex

ketones before stereoselective reduction.

In summary, we have developed a general methodology to stereoselectively reduce alkyl
and aryl ketones using D-glucose as a chiral auxiliary. Unlike reports published so far,
our approach is more straightforward because we used unprotected D-glucose and
achieved good diastereoselectivity and enantioselectivity. Diols with the (R)
configuration are obtained as the major products starting from a-glycosides while (S)
diols can be obtained as the major product from B-glycosides. Our working hypothesis
successfully rationalized the stereochemical outcome for all the experiments and the
model implies that the stereoselectivity expected for other substrates should improve.
Other important chiral building blocks can also be easily synthesized under the current
scheme by changing R- to other functional groups (such as CHg-X). The versatility of
this method will be greatly enhanced by the use of glucosidase which will attach and

remove chiral auxiliary under milder conditions. This work is being pursued.

82

 

 

Hgg/ OH
HO
HO

 

OH Fisher glycosidation WHg%/O

 

 

Enzyme HO O HO O
degradation HO 0 a-halogenation HO 0
——> HO OH ———) O OH
([3 glucosidase) o 0
HO O /
HO OH
(3')“ ”0 0H
HOXW «— 0%
X
C3 chiral synthon *
HO O
HO 0
R
HOW O\/U\'
R
Scheme 3.5

Application of stereoselective reduction using D-glucose
as a chiral auxiliary

83

 

Experimental

lH-NMR and l3C-NMR spectra were measured in chloroform-d solutions on a Varian-
300 spectrometer (300 MHz) unless noted otherwise. The chemical shifts are given in 5
values and are calibrated with TMS as the internal standard or relative to the chloroform
line at 7.24 ppm for 'H or 77.0 ppm for '3 C spectra. Silica gel flash column
chromatography was carried out on silica using Kieselgel 60 (Merck), 0040-0063 mm.

The mixed bed ion exchange resin used in the experiment is AG 501-X8.

2-Oxopropyl a-D-glucopyranoside (1)

To 1.10 g of allyl glucoside was added 0.09g PdClz and 0.59 g of 1,4-benzoquuinone in
100 mL of H20 and 50 mL of THF. The mixture was stirred at RT for 4 h and
concentrated under reduced pressure. After going through mixed bed ion exchange resin,
the filtrate was filtered again to remove residual fine particles of PdClz. The solvent was
then removed and the crude product was recrystallized in MeOH/EtOH. Yield: 0.50 g
(42%). mp: 164-166 0C. lH-NMR (300 MHz, D20): 5 1.98 (3H, s), 3.25 (1H, dd, J =
9.60 and 9.60 Hz), 3.39 (1H, dd, J = 10.05 and 3.60 Hz), 3.50 (1H, m), 3.54-3.68 (3H,
m), 4.20 (1H, d, J = 18.30 Hz), 4.34 (1H, d, J = 18.30 Hz), 4.76 (1H, d, J = 3.9GHz);
l3C-NMR (75 MHz, CDC13): 8 21.09, 55.79, 64.77, 66.70, 67.43, 68.23, 93.62, 205.65;
HRMS Exact mass: calcd for Cqu607 [M+H]“ 237.0974, found 237.0978; [0.]020 =

+62.850 (c 3.8855, MeOH).

84

 

(2R)-2-Hydroxypropyl a-D-glucopyranoside (2)

Reduction of (l) was conducted under the conditions listed in i-x, and (2R)-2-hydroxyl-
a-D-glucopyranoside (2) was obtained as the major product and (ZS)-2-hydroxyl-a-D-
glucopyranoside as the minor product. lH-NMR (300 MHz, D20): 5 0.99 (3H, d, J = 6.6
Hz), 3.15 (1H, dd, J = 10.65 and 7.2 Hz), 3.22 (1H, dd, J = 9.6 and 9.6 Hz), 3.37 (1H,
dd, J = 9.9 and 3.9 Hz), 3.51-3.60 (4H, m), 3.67 (1H, dd, J = 11.85 and 2.4 Hz), 3.88
(1H, m), 4.74 (1H, d, J = 3.6); l3C-NMR (75 MHz, D30): 5 13.40, 55.85, 61.93, 64.89,

66.84, 67.10, 68.20, 68.40, 93.86.

i: To 30 mg of (1) was added 10 mL of H20 and 30 mg of NaBH4. The mixture was
stirred at RT for 4 h and was quenched with 0.5N HCl. The aqueous solution was then
passed through a short column of mixed bed ion exchange resin and dried under reduced

pressure below 40 °C. Yield: 29 mg (96%). d.e.% < 5.

ii: To 30 mg of(1) was added 10 mL of EtOH and 20 mg of LiBH4 at 0 °C. The mixture
was stirred at 0 °C for another 8 h and was quenched with 0.5N HCl. Ethanol was
removed by rotary evaporation and the aqueous solution was passed through a short
column of mixed bed ion exchange resin and dried under reduced pressure below 40 °C.

Yield: 29 mg (96%). d.e.% = 20.
iii: To 30 mg of (l) was added 10 mL of MeOH and 370 mg of ZnClz powder and the

mixture was stirred at 0 °C for 0.5 h. 30 mg of NaBH4 was added and the mixture was

stirred at 0 0C for another 8 h and was quenched with 0.5N HCl. Methanol was removed

85

 

 

.5,

 

by rotary evaporation and the aqueous solution was passed through a short column of
mixed bed ion exchange resin and dried under reduced pressure below 40 0C. Yield: 27

mg (89%). d.e.% < 5.

iv: To 30 mg of(1) was added 10 mL of EtOH and l 15 mg of LiCl and the mixture was
stirred at -78 0C for 0.5 h. 30 mg of NaBH4 was added and the mixture was stirred at

—78 0C for another 8 h and was quenched with 0.5N HCl. Ethanol was removed by rotary
evaporation and the aqueous solution was passed through a short column of mixed bed
ion exchange resin and dried under reduced pressure below 40 °C. Yield: 27 mg (89%).

d.e.% = 47.

v. To 30 mg of (1) was added 10 mL of EtOH and 260 mg of Mng and the mixture was
stirred at -78 °C for 0.5 h. 30 mg of NaBH4 was added and the mixture was stirred at

—78 0C for another 8 h and was quenched with 0.5N HCl. Ethanol was removed by rotary
evaporation and the aqueous solution was passed through a short column of mixed bed
ion exchange resin and dried under reduced pressure below 40 °C. Yield: 26 mg (86%).

d.e.% < 5.

vi. To 30 mg of (1) was added 10 mL of EtOH and 300 mg of CaClz powder and the
mixture was stirred at -78 0C for 0.5 h. 30 mg of NaBH4 was added and the mixture was
stirred at -78 0C for another 8 h and was quenched with 0.5N HCl. Ethanol was removed

by rotary evaporation and the aqueous solution was passed through a short column of

86

 

mixed bed ion exchange resin and dried under reduced pressure below 40 0C. Yield: 28

mg (93%). d.e.% = 69.

vii. To 30 mg of (1) was added 10 mL of MeOH and the mixture was stirred at -78 °C for
0.5 h. 54 mg of Ca(BH4)2.2THF was added and the mixture was stirred at —78 0C for
another 8 h and was quenched with 0.5N HCl. Methanol was removed by rotary
evaporation and the aqueous solution was passed through a short column of mixed bed
ion exchange resin and dried under reduced pressure below 40 0C. Yield: 29 mg (96%).

d.e.% = 60.

viii. To 30 mg of (1) was added 10 mL of EtOH and the mixture was stirred at 0 0C for
0.5 h. 54 mg of Ca(BH4)2.2THF was added and the mixture was stirred at 0 °C for
another 8 h and was quenched with 0.5N HCl. Ethanol was removed by rotary
evaporation and the aqueous solution was passed through a short column of mixed bed
ion exchange resin and dried under reduced pressure below 40 °C. Yield: 29 mg (96%)..

d.e.% = 33.

ix. To 30 mg of (l) was added 10 mL of EtOH and the mixture was stirred at -78 °C for
0.5 h. 54 mg of Ca(BH4)2.2THF was added and the mixture was stirred at -78 0C for
another 8 h and was quenched with 0.5N HCl. Ethanol was removed by rotary
evaporation and the aqueous solution was passed‘through a short column of mixed bed
ion exchange resin and dried under reduced pressure below 40 °C. Yield: 28 mg (93%)..

d.e.% = 69.

87

x. To 30 mg of (l) was added 10 mL of i-PrOH/MeOH (v/v=10/ l) and the mixture was
stirred at —40 °C for 0.5 h. 54 mg of Ca(BH4)2.2THF was added and the mixture was
stirred at 0 °C for another 8 h and was quenched with 0.5N HCl. Ethanol was removed
by rotary evaporation and the aqueous solution was passed through a short column of
mixed bed ion exchange resin and dried under reduced pressure below 40 0C. Yield: 26

mg (86%). d.e.% =52.

(R)-Propane—l,2-diol (3)

25 mg of 2 was added 10 mL of 1N HCl and heated at 85 OC overnight. The solvent was
removed by rotary evaporation and the residue was extracted with CHC13 (50 mL). The
organic portions were combined and evaporated to dryness to give 5 mg of propane-l ,2-
diol with 3 as the major component. Chiral GC analysis shows that the e.e.% is
consistent with d.e.% obtained in the reduction step. lH-NMR (300 MHz, CDCl3): 8 3.89
(1H, m), 3.61(1H, dd,J = 11.1, 3.1Hz), 3.38 (1H, dd,J =11.2, 7.8 Hz), 1.14 (3H, d,J =

6.4 Hz); l3C-NMR (75 MHz, CDC13): 5 18.66, 67.84, 68.28.

Benzoylmethyl a-D-glucopyranoside (4)

Benzoylmethyl 2,3,4,6-tetraacetyl-a-D-glucopyranoside was synthesized using a
procedure similar to that reported by Lubineau [8]. To 0.34 g of NaH in 10 mL of dry
CHzClz at —20 0C was added 4.09 g of Ble and 1.1 g of 2-bromoacetophenone. 1.93 g
of 2,3,4,6-tetraacetyl-D-glucose in 10 mL of dry CHzClz was added at this temperature
dropwise. The temperature was slowly raised to RT and the mixture was stirred for 16 h.

The reaction was quenched with 2 mL acetic acid and partitioned between 100 mL

88

 

Ci

"Q

C 0

I-J

CHzClz and 50 mL H20. The organic layer was washed with brine and dried with
NazSO4. Purification by column chromatography (hexane/ethyl acetate = 1/1) gave 2.07
g of benzoylmethyl 2,3,4,6-tetraacetyl-a-D-glucopyranoside (mixed with a little [3-
conformer). Methanol (50 mL) and 0.2 g of KCN were added to the mixture and after
stirring at RT overnight, the solvent was removed under reduced pressure and diluted
with 5 mL of H20. The aqueous solution was passed through a short column of mixed

bed ion exchange resin and the resin was washed with 20 mL H20. The aqueous portions
were combined and dried under reduced pressure to give 1.26 g of benzoylmethyl a-D-
glucopyranoside (mixed with a little B-conformer). 50 mL of acetate buffer (pH = 5) was
added together with 2 mg of B-glucosidase and the mixture was incubated at 37 °C for 24
h before purification by column chromatography (CHC13/MeOH = 10/3.5), 1.13 g of
benzoylmethyl a-D-glucopyranoside was obtained. Yield: 1.13 g (68%); mp 148-151 0C.
1H-NMR (300 MHz, D20): 6 3.28 (1H, dd, J = 9.3 and 9.3 Hz), 3.41 (1H, dd, J = 10.2
and 3.6 Hz), 3.49 (1H, m), 3.57 (1H, m), 3.67 (1H, dd, J = 9.3 and 9.3 Hz), 4.75 (1H, d, J
= 17.7 Hz), 4.84 (1H, d, J = 3.6 Hz), 4.88 (1H, d, J =17.7 Hz), 5.13 (1H, d, J = 18.0
Hz), 7.31 (2H, dd, J = 7.8 and 7.8 Hz), 7.47 (1H, t, J = 7.5 Hz), 7.67 (2H, d, J = 7.5 Hz);
l3C-NMR (75 MHz, D30): 8 57.17, 66.21, 66.95, 68.29, 69.04, 69.74, 95.60, 124.64,
125.76, 130.56, 131.40, 195.78; HRMS Exact mass: calcd for C|4H|807 [M+H]“

299.1131,found 299.1135; [(11030 +164.10 (c 2.588, MeOH).

(2R)-2-Hydroxy-2-phenylethyl a-D-glucopyranoside (5)
The reduction was done under three different conditions, xi-xiii described below. (2R)—2-

Hydroxy-2-pheny1ethyl a-D-glucopyranoside (5) was obtained as the major component

89

(~11

 

Cl.)
'¢ J

68.

3C

dr

xi

31

Ex

xi
0.

31‘.

(the other component was (ZS)-2-hydroxy-2-phenylethyl a-D-glucopyranoside). 1H-
NMR (300 MHz, D30) 3.19-3.21 (2H, m), 3.33 (1H, dd, J = 9.6 and 3.6 Hz), 3.42-3.56
(4H, m), 3.65 (1H, dd, J = 10.2 and 4.5 Hz), 4.75 (1H, d, J = {3.6 Hz), 4.78 (1H, dd, J =
8.3 and 4.5 Hz), 7.23-7.26 Hz (5H, m); I3C—NMR (75 MHz, D20) 5 57.12, 66.15, 68.26,
68.61, 69.11, 69.34, 69.83, 95.35, 123.28, 125.16, 125.55, 137.05; HRMS Exact mass:

calcd for C14H2007 [M+H]“ 301.1287, found 301.1295.

xi. To 70 mg of (4) was added 10 mL of EtOH and the mixture was stirred at 0 °C for 0.5
h. 55 mg of Ca(BH4)2.2THF was added and the mixture was stirred at 0 0C for another 8
h and was quenched with 0.5N HCl. Ethanol was removed by rotary evaporation and the

aqueous solution was passed through a short column of mixed bed ion exchange resin and

dried under reduced pressure below 40 0C. Yield: 67 mg (95%). d.e.% = 20.

xii. To 70 mg of (4) was added 10 mL of EtOH and the mixture was stirred at -78 0C for
0.5 h. 55 mg of Ca(BH4)2.2THF was added and the mixture was stirred at 0 0C for
another 8 h and was quenched with 0.5N HCl. Ethanol was removed by rotary
evaporation and the aqueous solution was passed through a short column of mixed bed
ion exchange resin and dried under reduced pressure below 40 °C. Yield: 65 mg (92%).

d.e.% = 59.
xiii. To 70 mg of (4) was added 10 mL of PrOH and the mixture was stirred at 0 0C for

0.5 h. 55 mg of Ca(BH4)2.2THF was added and the mixture was stirred at 0 0C for

another 8 h and was quenched with 0.5N HCl. i-PrOH was removed by rotary

90

 

 

CV

ior

d.e

(R)

T0

at?

5,...

si

evaporation and the aqueous solution was passed'through a short column of mixed bed
ion exchange resin and dried under reduced pressure below 40 OC. Yield: 65 mg (92%).

d.e.% = 62.

(Ryl-PhenyI-ethane-l,2-diol (6)

To 50 mg of 5 was added 80 mg of NaIO4 in 30 mL H20. The mixture was stirred for 1h
at RT before 1 g of mixed bed ion-exchange resin was added and stirred for 20 minutes.
The suspension was filtered and several drops of formic acid was added to the filtrate to
adjust the pH to 1. After stirring for 3 h at RT, the solution was lyophilized and the
product was extracted by CHC13. (R)-1-Phenyl-ethane-l,2-diol (6) is the major
component with e.e. equal to the d.e. obtained from the reduction step. lH-NMR (300
MHz, CDC13): 5 3.66 (1H, dd, J = 11.4 and 8.1 Hz), 3.76 (1H, dd, J = 11.4 and 2.7 Hz),
3.82 (1H, J = 8.1 and 3.6 Hz), 7.36 (5H, s); l3C-NMR (75 MHz, CDC13): 6 68.02, 74.67,

126.03, 127.95, 128.50, 140.43.

2-Oxopropyl B-D-glucopyranoside (7)

2-Oxopropyl 2,3,4,6-tetraacetyl-0t-D-g1ucopyranoside was synthesized using a procedure
similar to that reported by Lubineau [8]. To 1.12 g of NaH in 50 mL of dry CH2C12 at -
20 0C was added 3 mL of bromoacetone [19], followed by the dropwise addition of 4.87
g of 2,3,4,6-tetraacetyl-D-glucose in 50 mL of dry CH2C12. The temperature was slowly
raised to RT and the mixture was stirred for 36 h. The reaction was quenched with 6 mL
acetic acid and partitioned between 200 mL CHzClz and 100 mL H20. The organic layer

was washed with brine and dried with NazSO4. Purification by column chromatography

91

(hexane/ethyl acetate=l/1) gave 2.94 g of 2-oxopropyl B-D-glucopyranoside (mixed with
some a-conformer, (it/13:1 :2). Methanol (100 mL) MeOH and 0.6 g of KCN were added
and the mixture was stirred at RT overnight afier which the solvent was removed under
reduced pressure and the residue diluted with 10 mL of H20. The aqueous solution was
passed through a short column of mixed bed ion exchange resin and the resin was washed
with 40 mL H20. The aqueous portions were combined and dried under reduced pressure
to give 1.63 g of 2-oxopropy1 B-D-glucopyranoside (mixed with a-conformer). 50 mL of
K3PO4 buffer (pH = 7) was added together with 1 mg of maltase and the mixture was
incubated at 37 °C for 24 h before being purified by column chromatography
(CHC13/MeOH = 10/3.5). 2-Ox0propyl B-D-glucopyranoside was obtained. Yield: 1.09
g (33%). 'H-NMR (300 MHz, D30): 6 2.02 (3H, s), 3.20 (1H, dd, J = 8.4 Hz), 3.26 (1H,
dd, J = 14.4 and 8.4 Hz), 3.33 (1H, dd, J = 8.4 and 8.4 Hz), 3.55 (1H, dd, J = 12.3 and
5.4 Hz), 3.73 (1H, dd, J =12.3 Hz and 2.1Hz), 4.31(1H, d, J = 7.8 Hz), 4.36 (1H, d, J =
18.0 Hz), 4.49 (1H, d, J = 18.0 Hz); '3C-NMR (75 MHz, D20): 6 22.41, 57.45, 66.34,
69.80, 70.62, 72.40, 72.82, 98.83, 206.87; HRMS Exact mass: calcd for C9H|607

[M+H]“ 237.0974, found 237.0974; [(11030 —48.1 (c 0.988, MeOH).

(ZS)-2-Hydroxypropyl B-D-glucopyranoside (8)

To 30 mg of ( 1) was added 10 mL of EtOH and the mixture was stirred at -78 0C for 0.5
h. 54 mg of Ca(BH4)2.2THF was added and the mixture was stirred at -78 °C for another
8 h and was quenched with 0.5N HCl. Ethanol was removed by rotary evaporation and
the aqueous solution was passed through a short column of mixed bed ion exchange resin

and dried under reduced pressure below 40 °C. Yield: 27 mg (89%). d.e.% = 33. IH-

92

NMR (300 MHz, D20) 5 0.98 (3H, d, J = 6.6 Hz), 3.12 (1H, dd, J = 9.00 Hz), 3.19 (1H,
dd, J = 9.00 Hz), 3.22-3.35 (3H, m), 3.49 (1H, dd, J: 10.95 and 3.30 Hz), 3.53 (1H, dd,
J = 12.15 and 5.7 Hz), 3.58 (1H, dd, J = 10.80 and 6.9 Hz), 3.73 (1H, dd, J = 10.80 and
2.4 Hz), 3.85 (1H, m), 4.28 (1H, d, J = 7.8 Hz); 130mm (75 MHz, D30) 5 14.73,

57.51, 63.38, 66.44, 70.03, 71.94, 72.41, 72.69, 99.52.

(S)-Propane-1,2-diol (9)

 

25 mg of 8 was added to 10 mL of 1N HC1 and heated at 85 0C overnight. The solvent
was removed under reduced pressure and the residue was extracted with CHC13 (2 x 50
mL). The organic portions were combined and evaporated to dryness to give 5 mg of
propane-1,2-diol with 3 as the major component. Chiral GC analysis shows that the
e.e.% is consistent with d.e.% obtained in the reduction step. lH—NMR (300 MHz,
CDC13): 5 1.14 (3H, d, J = 6.4 Hz), 3.38 (1H, dd, J = 11.2, 7.8 Hz), 3.61 (1H, dd, J =

11.1, 3.1 Hz), 3.89 (1H, m); l3C-NMR (75 MHz, CIDC13): 5 18.66, 67.84, 68.28.

93

 

10.

11.

12.

13.

l4.

16.

References

. Huang, G; Hollingsworth, R. I. Tetrahedron Lett. 1999, 40, 581-584.

Akiyama, T.; Nishimoto, H.; Kuwata, T.; Ozaki, S. Bull. Chem. Soc. Jpn. 1994, 67,
180-188.

DiCesare, J. C. Carbohydrates as chiral auxiliaries in organometallic reactions.
Georgia Institute of Technology 1992.

Nair, V.; Prabhakaran, J. J. Chem. Soc, Perkin Trans. 1 1996, 593-594.

Saksena, A. K.; Mangiaracina, P. Tetrahedron Lett. 1983, 24, 273-276.

Tumbull, M. D.; Hatter, G.; Ldgewood, D. E. Tetrahedron Lett. 1984, 25, 5449-5452.
Hughes, M. J .; Thomas, E. J .; Tumbull, M. D.;xJones, M. D.; Jones, R. H.; Warner, R.
E. J. Chem. Soc, Chem. Commun. 1985, 755-758.

Adams, J.; Poupart, M.-A.; Grenier, L. Tetrahedron Lett. 1989, 30, 1753-1756.
Evans, D. A.; Chapman, K. T. Tetrahedron Lett. 1986, 27, 5939-5942.

Evans, D. A.; Chapman, K. T.; Carreira, E. M. J. Am. Chem. Soc. 1988, 110, 3560-
3578.

Evans, D. A.; Hoveyda, A. H. J. Am. Chem. Soc. 1990, 112, 6447-6449.

Anwar, S.; Davis, A. P. J Chem. Soc, Chem. Commun. 1986, 831-832.

Anwar, S.; Davis, A. P. Tetrahedron 1988, 44, 3761-3770.

Fujita, M.; Hiyama, T. J. Org. Chem. 1988,53, 5405-5415.

. Oishi, T.; Nakata, T. Acc. Chem. Res. 1984, 17, 338-344.

Angyal, S. J. Tetrahedron 1974, 30, 1695-1702.

94

l7. Angyal, S. J; Bethell, G. 8.; Beveridge, R., J. Carbohydrate Research 1979, 73, 9-18.
18. Lubineau, A.; Escher, S.; Alais, J .; Bonnaffe, D. Tetrahedron Lett. 1997, 38, 4087-
4090.

19. Levene, P. A. Organic Synthesis 1930, 10, 12.

 

95

CHAPTER 4

Stereoselective Reduction Using Nicotine and Its Derivative as a Chiral Promoter

96

 

Introduction

Some organic reactions can be accomplished by using two-layer systems in which phase-
transfer catalysts play an important role [1,2]. The phase-transfer reaction proceeds via
ion pairs, and asymmetric induction is expected to emerge when chiral quaternary
ammonium salts are used. The ion-pair interaction, however, is usually not strong
enough to control the absolute stereochemistry of the reaction [3]. Numerous trials have
resulted in low or only moderate stereoselectivity, probably because of the loose
orientation of the ion-paired intermediates or transition states. These reactions include,
but are not limited to, carbene addition to alkenes, reaction of sulfiir ylides and
aldehydes, nucleophilic substitution of secondary alkyl halides, the Darzens reaction,
chlorination of alkenes, aldehyde cyanohydrin formation, and borohydride reduction of
prochiral ketones [4-16]. With some special phase-transfer catalysts and under
appropriate reaction conditions, however, respectable enantioselectivities can be
achieved. Figure 4.1 shows the structures of some chiral phase transfer catalysts used in
sodium borohydride reduction of prochiral ketones. The asymmetric reduction of phenyl
t-butyl ketone yielded the corresponding alcohol in 32% e.e. by using NaBH4 with (G) as
a phase-transfer catalyst. Employing ligands A-F, the optical yields were generally lower
for the reduction of ketones, ranging from 1.1-13.7% e.e.. Studies on asymmetric
reduction under phase transfer conditions have led to the conclusion that
conformationally rigid ligands result in faster reductions and high optical yields, as

exemplified by the quininium salt (G). It has also been found that the hydroxyl group in

97

 

F: 9H3 0"

 

CH3

I I
PhCHz—c'r—ntcmar Ph—c-CHZ—CHz—HtCHzx-

H (31sz i H R1

(A) R=OH (C) R1=C12H25, X=Br

(B) R=H (D) R1=CH2Ph, x=c1

Phc*H0Hq*H_I:I*(CH3)zcr PhCqu*H-t|4*(CH3)2CF

CH3 CHZCH20H CH3 0*!chon

(E) (F)

CH2C*HOHCH20H

 

Ph-Ph

(“1

Figure 4.]

Structures of some chiral phase transfer catalysts

98

the ligand should be in the B-position to the onium fiJnction. This may be important for
its interaction with the carbonyl group of the substrate and to favor one of the
diastereomeric transition states which leads to the carbinol. A multipoint interaction

between catalyst and the substrate is essential for optical induction.

(S)-Nicotine can be obtained on a large scale from the tobacco industry and is a relatively
cheap chiral source. Although its biological effects have been investigated extensively,
little work has been done on its applications in organic synthesis, especially asymmetric
synthesis. Only one report, from a Japanese group in 1981, describes the use of nicotine
in enantioface-differentiating epoxidation [17]. Under catalysis by nicotine, (l-
phenylpropylidene)-malononitrile was epoxidized with oxygen to give 2,2-dicyano-3-

ethyl-3-phenyloxirane. Under the most favorable condition, only 6% optical purity was

 

obtained (Scheme 4.1).
Pb CN nicotine, 02 j, Ph CN
R: CN 20°C, DMF R: :0; CN
R = Et : 6.0 ee%
R = n-Pr: 5.5 ee

Scheme 4.]-

Asymmetric epoxydation catalyzed by nicotine

Our interest in the asymmetric synthesis of chiral 3-hydroxy fatty acids has prompted us
to develop a phase transfer catalyst based on (S)-nicotine to initiate stereoselective

reduction of 3-keto fatty acids. Chiral 3-hydroxy fatty acids are a group of compounds

99

with important biological significance. Some naturally occurring long-chain 3-
hydroxyalkanoic acids include (-)-3-hydroxydecanoic acid (found in secretions of the
leaf-cutting ant), (R)-3-hydroxytetradecanoic acid (a constituent of lipid A in endotoxin),
(S)-3-hydroxy-tetradecanoic acid (in the fish toxin) and (R)-3-hydroxydecanoic acid (a
constituent of extracellular glycolipids from the red yeast Rhodotorula). There are
several well documented methods in the literature for the synthesis of chiral 3-hydroxy
fatty acids by asymmetric reduction. Noyori and co-workers have demonstrated that
chiral RuX2[BlNAP] complexes catalyze hydrogenation of 3-keto esters with >96% yield
and 97-100% e.e. [18]. Baker’s yeast can also yield chiral 3-hydroxy esters with
excellent e.e.% but low to moderate yields [19]. However, the direct chiral reduction of

3-keto fatty acids is less investigated.

A literature review showed that nitrogen compounds, such as ephedrine alkaloid salts and
cinchona alkaloid salts, were among the most successful phase transfer catalysts for
asymmetric reduction. The successes were attributed to the rigidity of the catalyst and a
B-hydroxyl functionality which enables H-bonding to the prochiral ketone. (S)-Nicotine
has two nitrogen atoms as shown in Figure 4.2. Studies have shown that the nitrogen
atom of the pyridine ring is more nucleophilic, while that of the pyrrolidine ring is more
basic. In other words, the former would be alkylated first and the latter would be
protonated first. In order to generate multi-point interactions with certain substrates such
as 3-keto fatty acids, both nitrogen atoms should be utilized properly. A long alkyl chain,
which can generate van der Waals interactions with the hydrophobic part of the fatty

acid, can be installed on the nitrogen atom of the pyridine ring (Figure 4.3). The

100

 

 

 

 

“D
/ "" N
I I ‘ more basic
\

CH3

more nucleophilic

Figure 4.2

Structure of nicotine

aqueous phase

 

organic phase
‘/ electrostatic interaction

0

 

‘_—)
CH3 (CH2)12

van der Waals

CH3

Figure 4.3

Asymmetric reduction of a 3-keto fatty acid under phase transfer catalytic condition

101

 

pyrrolidine nitrogen atom can act as an organic base to complex with the head group (-
COOH) of the fatty acid via electrostatic interaction. If both forces are able to rigidify
the complex, the two sides of the keto group could be differentiated. Borohydride, which
is likely delivered from the pyridinium group, will then effect stereoselective reduction.
In this experiment, hexadecyl nicotinium iodide was synthesized and used as a phase-
transfer catalyst for the reduction of 3-keto tetradecanoic acid and enantioselectivities

ranging fi'om 1.2 to 8.2 were recorded.

In 1992 an interesting method reported was used to effect the chiral reduction of
benzoylformic acid with NaBH4 in the presence of 6-deoxy-6-amino-[3-cyclodextrin in
aqueous media [20]. The mole ratio between benzoylformic acid and 6-deoxy-6-amino-
B-cyclodextrin was 1:1, which is supposed to lead to a rigid complex. The existence of
three interactions was proposed to explain the asymmetric selectivity (Figure 4.4). First,
the aromatic ring of the substrate should be included in the cyclodextrin cavity
(hydrophobic interaction). Second, a hydrogen bond between the hydroxyl group and the
carbonyl group of benzoylformic acid can be presumed. And third, there is an ionic
interaction between an amino group of the cyclodextrin and a carboxyl group in
benzoylformic acid. Even though the e.e.% was low (32%), it is an example that shows
that the two enantiotopic faces can be differentiated by the ionic interaction and hydrogen

bonding between substrate and chiral promoter.

This example prompted our approach of using (S)-nicotine as an organic base. Nicotine

will form an ion pair with an organic acid in a nonpolar system. The electrostatic

102

 

interaction itself is not likely to differentiate well between the two faces of the carbonyl
group of the organic acid, so other forces need to be included. Benzoylformic acid was
chosen as our substrates because its benzene ring can possibly interact with pyridine ring
of nicotine through n-n interactions. Figure 4.5 shows the proposed electrostatic
interaction between the pyrrolidine ammonium of nicotine and the carboxylate of
benzoylformic acid, and the alignment of the pyridine ring of nicotine and phenyl rings of
benzoylformic acid that could result in favorable n-stacking interaction. A series of
experiments under different conditions were conducted. Unfortunately, the highest e.e.
obtained was 32.6 %. 1D and 2D-NMR studies were done to try to obtain the reason for

low chiral induction.

103

 

electrostatic interaction

  
 
   

 

H-bonding

Hydrophobic
interaction

Figure 4.4

Reduction of benzoylformic acid with cyclodextrin as the chiral promoter

electrostatic

3\ interaction
0'

O

Ir-stacking

Figure 4.5

Proposed complexation between benzoylformic acid and nicotine

104

 

Results and discussion

Nicotine was alkylated by l-iodohexadecane (1 eq.) at 80 0C for 24 h in THF. The 3-keto
fatty acid was synthesized according to a known method [21]. Scheme 4.2 shows the
general scheme for the reduction of 3-oxo-tetradecanoic acid. Though a number of
reaction conditions (solvent, temperature, reducing agent) were employed (Table 4.1),
none showed significant chiral induction (optical purity was determined by
corresponding MTPA ester [22]). This may be due to the formation of a loose ion pair
between the substrate and phase transfer catalytic, which failed to differentiate the two
sides of the keto group. Even though the electrostatic interaction binds 3-keto acid to
nicotine, the keto group is still relatively free since it is two C-C bond-length away from
the carboxylate group. Van der Waals interactions between the two long alkyl chains did

not appear to be strong enough to rigidify the keto group.

Benzoylformic acid was then chosen as a substrate since it is an a-keto acid, whose
carbonyl group is closer to the carboxylate compared to that of a B-keto acid. We
expected that the keto group will be less labile once benzoylformic acid complexed with
nicotine by electrostatic interaction. In addition, n-n stacking between the phenyl and
pyridine ring can also help rigidify the complexation between nicotine and benzoylformic
acid so as to differentiate the two sides of the keto group. Since benzoylformic acid is

very soluble in H30, the reduction can’t be done under phase

105

O O OH O

/U\/U\ reducing agent /I\/U\
+ *
*
CH3(CH2)1o OH PTC CHdCHzlio

PTc*: derived from nicotin

 

OH

Scheme 4.2

Reduction of 3-oxo-tetradecanoic acid under phase transfer catalytic condition

Table 4.1

Asymmetric reduction of 3-oxo-tetradecanoic acid
under various phase transfer catalytic conditions

 

 

 

 

 

 

 

 

 

Entry Reaction conditions e.e. (%)
l H20/CH2C12, NaBH4 (4 eq.), PTC (10%), RT 4.4
2 H20/C1CH2CH2CI, NaBH.. (4 eq.), PTC (20%), 0 °C 1.4
3 H30/CC14, NaBH4 (4 eq.), PTC (20%), 0°C 3.2
4 H20, NaBH.; (4 eq.), PTC (20%), 0 °C 1.2
5 H20/toluene, NaCNBH3 (4 eq.), CF3COOH (1 eq.), 2.3
PTC (20%), 0 0C

6 H20/CH2C12, NaCNBH3 (4 eq.), CF3COOH (1 eq.), 8.2
PTC (20%), RT

7 H20/toluene, NaCNBH3 (4 eq.), CF3COOH (1 eq.), 1.6
PTC (20%), 0 °C

 

 

 

106

 

transfer catalytic conditions. A number of reductions were performed using a 1:1
mixture (mole ratio) of benzoylformic acid and nicotine (Table 4.2). The best optical
yield obtained was 32.6%. The reducing agent has a strong effect on chiral induction.
Cations like Zn2+ and Li" gave better results possibly because they can chelate the keto
oxygen with the carboxylate and reduce the freedom of the keto group. The optical yield
was an indication of differentiation of the enantiotopic faces of the keto group, which is
directly related to the rigidity of the benzoylformic acid/nicotine complex. 1D and 2D-

NMR (NOESY) in CDC13 were used to study this complex.

By comparing the lH-NMR spectrum of the complex (Figure 4.6) with the spectra of
pure nicotine (Figure 4.6) and benzoylformic acid, we noticed the down field shift of H9
(2.20 ppm to 2.60 ppm), H8b (3.24 ppm to 4.0 ppm), H8a (2.38 ppm to 3.0 ppm), and H5
(3.10 ppm to 4.0 ppm). But the chemical shifis of H', H2, H3, and H4 of the pyridine ring
are only shifted down field by a small amount, indicating that protonation occurred at the
nitrogen atom of the pyrrolidine. So we can expect a strong electrostatic interaction
between ammonium and carboxylate. However, the complex is not rigid since three sets
of signals were seen for the phenyl ring as before, which means the phenyl ring rotates
freely. In order to get a 3D-picture of the complex, NOESY was conducted to determine
which atoms were close to each other. According to the crosspeaks of the NOESY
spectrum, there are several intermolecular interactions between nicotine and
benzoylformic acid. The cross peaks show the proximity between HI and HP, H4 and
H", Hil and H2, H4 and H3, H6 and H', H6 and H2. Based on these relationships, several

constraints (distance) were applied using the BlOGRAF program and the computer-

107

Table 4.2 .

Chiral reduction of benzoylformic acid complexed with one equivalent of nicotine

 

 

 

 

 

 

 

 

 

 

Entry Reaction conditions e.e. (%)
8 CHC13, NaBH3CN (4 eq.), HC1 (1 eq.) 1.9
9 THF, NaBH3CN (4 eq.), O 0C, 20 h 10.6
10 THF, NaBH3CN (4 eq.),-78 oC,20 h 3.0
11 CHzClz, NaCNBH3 (2 eq.), 0°C 10.6
12 Benzene, NaCNBH3 (2 eq.), 0 0C, catalytic amount of 19.9
BU4N+BI’
l3 THF, 2 eq. NaBH4, traces of Bu4N+Br’, 0 0C 1.4
14 THF, Zn(BH4)2 (3 eq.), RT, 4 h 28.1
15 THF, LiA1(OC(CH3)3)3H (4 eq.), 0°C,l h 32.6

 

 

 

108

 

simulated conformation is shown in Figure 4.7. ‘ The crosspeaks corresponding to the
close proximity of H6 and H1, and H6 and H2 indicated that the pyridine ring was not
fixed in one conformation, but instead resides in two conformers. The rotation of
pyridine ring of nicotine, as well as the phenyl ring of benzoylformic acid, indicated a

low rigidity of the complex, leading to low chiral induction.

In summary, the use of (S)-nicotine in asymmetric reduction of keto group was explored.
The low chiral induction of 3-keto fatty acid and benzoylformic acid was due to the
structural flexibility of nicotine, which indicated that (S)-nicotine was not a good choice

unless some structural modification could be made.

109

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Figure 4.6

1H-NMR spectra of nicotine and the nicotine/benzoylformic acid complex

110

 

nicotine

———___—

benzoylformic acid

Figure 4.7

3D picture of complexation between nicotine and benzoylformic acid

111

Experimental

1H-NMR and l3C-NMR spectra were measured in chloroform-d solutions on a Varian-
300 spectrometer (300 MHz) unless noted otherwise. The chemical shifts are given as 5
values and are calibrated with TMS as the internal standard or relative to the chloroform
line at 7.24 ppm for 1H or 77.0 ppm for I3C spectra. Silica gel flash column

chromatography was carried out on silica using Kieselgel 60 (Merck), 0.040-0.063 mm.

1. REDUCTION OF 3-OXO-TETRADECANOIC ACID
3-(1-Methyl-pyrrolidin-2-y1)-1-hexadecyI-pyridinium iodide

To 0.57 g of (S)-(-)-nicotine was added 1.02 g of l-iodohexadecane (1 eq.) and 3 mL
THF. The mixture was heated at 80 °C for 24 h,‘ after which the solvent was removed
under reduced pressure and the product was purified by column chromatography
(CHCl3/MeOH=10/2). Yield: 1.44 g (80%). lH-NMR (CDCl3, 300MHz): 5 0.87 (3H, t,
J = 6.60 Hz), 1.24 (25H, m), 1.91 (2H, m), 2.26 (3H, s), 2.45 (2H, m), 3.25 (2H, m), 3.60
(1H, t, J = 8.1 Hz), 4.70 (2H, t, J = 7.5 Hz), 7.99 (1H, dd, J = 8.10 and 6.00 Hz), 8.37
(1H, d, J = 8.10 Hz), 9.02 (1H, s), 9.09 (1H, d, J = 6.00 Hz); l3C-NMR (CDC13,
75MHz): 5 13.64, 22.20, 22.73, 25.58, 28.57, 28.87, 28.92, 29.04, 29.17, 29.20, 31.38,
31.43, 35.23, 40.08, 49.62, 56.29, 61.66, 66.38, 128.16, 142.67, 143.03, 143.76, 145.99;
0t=-46.70 (0 2.4400, CHC13); HRMS Exact mass: calcd for C36H47N2 [M-l]+', 387.3742,

found 387.3756.

112

3-Oxo-tetradecanoic acid

To 1.6 g of methyl 3-oxo-tetradecanoate was added 0.50 g NaOH in 30 mL wet ethanol.
After stirring at RT overnight, the ethanol was removed under reduced pressure. 30 mL
of 1N HC1 was added and the suspension was extracted with two portions of 50 mL
CHC13. The combined organic layers were washed with brine and dried over NaZSO4.
Yield: 1.48 g (98%). lH-NMR (CDC13, 300MHz): 5 0.85 (3H, t, J = 6.60 Hz), 1.20-1.30
(16H, m,), 1.56 (2H, m), 2.53 (2H, t, J = 7.5 Hz), 3.48 (2H, s); l3C-NMR (CDC13,
75MHz): 5 14.09, 22.65, 23.35, 28.92, 29.29, 29.37, 29.56, 31.87, 43.24, 47.58, 170.38,

204.94.

3-Hydroxytetradecanoic acid

3-Hydroxy-tetradecanoic acid was synthesized by reduction of 3-hydroxytetradecanoic
acid under following reduction conditions listed below (PTC refers to the use of 3-(1-
methyl-pyrrolidin-Z-yl)-l-hexadecyl-pyridinium iodide). lH-NMR (CDC13, 300MHz): 5
0.85 (3H, t, J = 6.60 Hz,), 1.10-1.50 (20H, m), 2.44 (1H, dd, J = 16.65 and 9.00 Hz),
2.55 (1H, dd, J = 16.55 and 3.30 Hz), 4.00 (1H, m); l3C-NMR (CDC13, 75MHz): 5 14.01,

22.67, 25.43, 29.32, 29.46, 29.54, 29.56, 29.61, 29.62, 31.90, 36.48, 40.98, 68.00, 177.34.

The optical purity of 3-hydroxytetradecanoic acid was determined by forming the MTPA
ester [22] from 3-hydroxytetradecanoic acid methyl ester. 3-Hydroxytetradecanoic acid
methyl ester was synthesized by mixing 40 mg of the acid with 2 mL of MeOH and 0.02
mL of concentrated HC1 and heating at 72 0C for 4 h. The MTPA ester was made by

allowing 3-hydroxytetradecanoic acid methyl ester (42 mg), distilled MTPA-C1 (37.1

113

mg), and pyridine (0.5 mL) to stand for 1 h. Water was added to the cooled mixture
which was then extracted with ether. The ether extract was washed successively with
dilute HC1 and dilute sodium carbonate solution, dried over MgSO4, and evaporated to

driness to give the MTPA ester.

Reduction condition

Entry 1: To 60 mg of 3-oxo-tetradecanoic acid in 5 mL of CH3C13 was added 0.01 g of
PTC (0.1 eq.) and 0.04 g of NaBH4 in 5 mL of H30. The mixture was stirred at RT for
12 h and then quenched with 1N HC1. 20 mL of CH3C13 was added and after extraction,
the organic layer was washed with brine and dried with Na3SO4. Afier it was
concentrated, the crude product was purified by column chromatography. Yield: 58 mg

(96%).

Entry 2: To 60 mg of 3-oxo-tetradecanoic acid in 5 mL of CICH3CH3C1 was added 0.02
g of PTC (0.2 eq.) and 0.04 g of NaBH4 in 5 mL of H30. The mixture was stirred at 0 °C
for 12 h and then quenched with 1N HC1. 20 mL of CH3C13 was added and after
extraction, the organic layer was washed with brine and dried with Na3SO4. Afier it was
concentrated, the crude product was purified by column chromatography. Yield: 30 mg

(50%).

Entry 3: To 60 mg of 3-oxo-tetradecanoic acid in 5 m1. of CC14 was added 0.02 g of
PTC (0.2 eq.) and 0.04 g of NaBH4 in 5 m1. of H30. The mixture was stirred at 0 0C for

12 h and then quenched with 1N HC1. 20 ml. of CH3C 13 was added and after extraction,

114

the organic layer was washed with brine and dried with Na3SO4. After it was
concentrated, the crude product was purified by column chromatography. Yield: 35 mg

(60%).

Entry 4: To 60 mg of 3-oxo-tetradecanoic acid in 5 mL of H30 was added 0.02 g of PTC
(0.2 eq.) and 0.04 g of NaBH4 in 5 mL of H30. The mixture was stirred at 0 °C for 12 h
and then quenched with 1N HC1. 20 mL of CH3C13 was added and after extraction, the
organic layer was washed with brine and dried with Na3SO4. After it was concentrated,

the crude product was purified by column chromatography. Yield: 56 mg (93%).

Entry 5: To 60 mg of 3-oxo-tetradecanoic acid in 5 mL of toluene was added 0.02 g of
PTC (0.2 eq.) and 0.04 g of NaBH4 in 5 mL of H30. The mixture was stirred at 0 °C for
12 h and then quenched with 1N HC1. 20 mL of CH3C13 was added and after extraction,
the organic layer was washed with brine and dried with Na3SO4. After it was
concentrated, the crude product was purified by column chromatography. Yield: 52 mg

(86%).

Entry 6: To 60 mg of 3-oxo-tetradecanoic acid in 5 mL of CH3C13 was added 0.02 g of
PTC (0.2 eq.), 0.06 g of NaCNBH3, in 5 mL of H30 and 25 mg of TFA. The mixture was
stirred at 0 °C for 12 h and then quenched with 1N HC1. 20 mL of CH3C13 was added and
after extraction, the organic layer was washed with brine and dried with Na3SO4. After it
was concentrated, the crude product was purified by column chromatography. Yield: 54

mg (89%).

115

Entry 7: To 60 mg of 3-oxo-tetradecanoic acid in 5 mL of toluene was added 0.02 g of
PTC (0.2 eq.), 0.06 g of NaCNBH3, in 5 mL of H30 and 25 mg of TFA. The mixture was
stirred at 0 0C for 12 h and then quenched with 1N HC1. 20 mL of CH3C13 was added and
after extraction, the organic layer was washed with brine and dried with Na3SO4. After it
was concentrated, the crude product was purified by column chromatography. Yield: 53

mg (88%).

II. REDUCTION OF BENZOYLFORMIC ACID

lH-NMR (CDC13) data for the reduction product (mandelic acid) are: 55.20 (1 H, s), 7.31-
7.39 (5H, m). The optical purity of mandelic acid was determined from 1H-NMR spectra
of methyl mandelate using the chiral shifting reagent europium tris[3-(heptafluoropropyl-
hydroxymethylene)]-(+)-camphorate. The methyl mandelate was made by mixing the
reaction product (mandelic acid) with 2 mL of MeOH and 0.02 mL of concentrated HC1

and heating at 72 °C for 4 h, after which it was blown to dryness.

Reduction conditions

Entry 8: To 30 mg of BFA and 32.4 mg of nicotine in 6 mL of CHC13 was added 0.016
mL of concentrated HC1 and 50 mg of NaCNBH3 and stirred for 4 h at RT. The reaction
was quenched with 1N HC1. CHC13 was removed under reduced pressure and the
aqueous residue was passed through a strong cation exchange resin and dried under

rotary evaporation. Yield: 25 mg (83%).

116

Entry 9: To 30 mg of BFA and 32.4 mg of nicotine in 2.5 mL of THF was added 50 mg
of NaCNBH3 in 2.5 mL of THF. The mixture was stirred at 0 °C for 20 h and was
quenched with 1N HC1. THF was removed under reduced pressure and the aqueous
residue was passed through strong cation exchange resin and dried under rotary

evaporation. Yield: 27 mg (90%).

Entry 10: To 30 mg of BFA and 32.4 mg of nicotine in 2.5 mL of THF was added 50 mg
of NaCNBH3 in 2.5 mL of THF. The mixture was stirred at -78 0C for 20 h and quenched
with 1N HC1. THF was removed under reduced pressure and the aqueous residue was
passed through strong cation exchange resin and dried under rotary evaporator. Yield: 24

mg (80%).

Entry 11: To 30 mg of BFA and 32.4 mg of nicotine in 5 mL of CH3Cl3 was cooled to 0
0C and added 50 mg of NaCNBH3. The mixture was stirred at 0 °C for 20 h and
quenched with 1N HC1. CH3C13 was removed under reduced pressure and the aqueous
residue was passed through strong cation exchange resin and dried under rotary

evaporation. Yield: 28 mg (93%).

Entry 12: To 30 mg of BF A and 32.4 mg of nicotine in 5 mL of benzene was cooled to 0
0C and added 50 mg of NaCNBH3. Also added was several grains of Bu4N+BrZ The
mixture was stirred at O 0C for 20 h and quenched with 1N HC1. Benzene was removed
under reduced pressure and the aqueous residue was passed through strong cation

exchange resin and dried under rotary evaporator. Yield: 23 mg (77%).

117

Entry 13: To 30 mg of BF A and 32.4 mg of nicotine in 2.5 mL of THF was cooled to 0
0C and added 30 mg of NaBH4. The mixture was stirred at 0 0C for 20 h and quenched
with 1N HC1. THF was removed under reduced pressure and the aqueous residue was

passed through strong cation exchange resin and dried under rotary evaporator. Yield: 28

mg (93%).

Entry 14: To NaBH4 (9.2 mg) dissolved in 5 mL of THF was added ZnC13 (10.8 mg).
After stirring for 3.5 h at 25 0C, the solution was cooled down to 0 °C. Then an ice-
cooled 5 mL of THF solution of 30 mg of BFA and 32.4 mg of nicotine was added
dropwise. The mixture was stirred at 0 0C for 20 h and quenched with 1N HC1. THF was
removed under reduced pressure and the aqueous residue was passed through strong

cation exchange resin and dried under rotary evaporation. Yield: 29 mg (97%).

Entry 15: To 30 mg of BFA in 5 mL of THF was added 32.4 mg of nicotine. The
mixture was cooled to 0 0C in ice-bath. 101.7 mg of LiAl(OC(CH3)3)3H dissolved in 5
mL of THF was added under N3. The mixture was stirred at 0 °C for 1.5 h and then was
quenched with 1N HC1. THF was removed under reduced pressure and the aqueous
residue was passed through strong cation exchange resin and dried under rotary

evaporation. Yield: 29 mg (97%).

118

10.

ll.

12.

13.

14.

15.

16.

17.

References

. Weber, W. P.; Gokel, G. W.; Phase Transfer Catalysis in Organic Synthesis,

Springer-Verlag, Berlin, 1977.

Dehmlow, E. V.; Dehmlow, S. S. “Phase Transfer Catalysis” in H. F. Ebel, ed.,
Monographs in Modern Chemistry, Vol. 11, Verlag Chemie, Weinheim, 1983.
Pochapsky, T.C.; Stone, P. M.; Pochapsky, S. S. J. Am. Chem. Soc. 1991, 113, 1460.
Hiyama, T.; Sawada, H.; Tsukanaka, M.; Nozaki, H. Tetrahedron Lett. 1975, 3013.
Hiyama, T.; Mishima, T.; Sawada, H.; Nozaki, H. J. Am. Chem. Soc. 1975, 97, 1626.
Hiyama, T.; Mishima, T.; Sawada, H.; Nozaki, H. J. Am. Chem. Soc. 1976, 98, 641.
Chiellini, E.; Solaro, R. J. Chem. Soc, Chem. Commun. 1977, 231.

Julia, 8.; Ginebreda, A.; Guixer, J. J. Chem. Soc, Chem. Commun. 1978, 742.
Colonna, S ; Fomasier, R; Pfeiffer, U. J. Chem. Soc, Perkin Trans. 1 1978, 8.

Julia, 3.; Ginebreda, A. Tetrahedron Lett. 1979, 2171.

Julia, 8.; Ginebreda, A.; Guixer, J .; Tomas, A. Tetrahedron Lett. 1980, 3709.

Masse, J.P.; Parayre, E. R. J. Chem. Soc., Chem. Commun. 1976, 438.

Colonna, S ; Fomasier, R. J. Chem. Soc, Perkin Trans. 1 1978, 371-373.

Balcells, J .; Colonna, S.; Fomasier, R. Synthesis 1976, 266.

Kinishi, R.; Nakajima, Y.; Oda, 1.; lrouze, Y. Agric. Biol. Chem. 1978, 42, 869.

Julia, S.; Ginebreda, A.; Guixer, J.; Masana, J.; Tomas, A.; Colonna, S. J. Chem.
Soc, Perkin Trans. 1 1981, 574.

Nanjo, K.; Suzuki, K.; Sekiya, M. Chem. Pharm. Bull. 1981, 29(2), 336-343.

119

18.

19.

20.

21.

22.

Noyori, R.; Ohkuma, T.; Kitamura, M.; Takaya, H.; Sayo, N.; Kumobayashi, H.;
Akutagawa, S. J. Am. Chem. Soc. 1987, 109, 5856.

Utaka, M.; Watabu, H.; Higashi, H.; Sakai, T.; Tsuboi, S.; Torii, S. J. Org. Chem.
1990, 55, 3917-3921.

Hattori, K.; Takahashi, K.; Sakai, N. Bull. Chem. Soc. Jpn. 1992, 65, 2690.
Utaka, M.; Watabu, H.; Higashi, H.; Sakai, T.; Tsuboi, S.; Torii, S. J. Org. Chem.
1990, 55, 3917-3921.

Dale, J. A.; Dull, D. L.; Mosher, H. S. J. Org. Chem. 1969, 34, 2543.

120

CHAPTER 5
Stereoselective Reduction of 2,2-Disubstituted 5-Methyl-1,3-dioxolan-4-ones

Formed from (L)-Lactic acid

121

Introduction

a-Hydroxy acids, such as mandelic acid, have been successfully used as a chiral auxiliary
for the synthesis of chiral alcohols (Scheme 5.1) [1,2]. The chiral auxiliary was attached
to the substrate (an aldehyde or ketone) via the formation of 1,3-dioxolan-4-ones.
Nucleophilic opening by silyl enol ether catalyzed by Lewis acid or magnesium-copper
reagents followed by removal of chiral auxiliary afforded chiral alcohols with good to

excellent optical purity. (L)-Lactic acid, one of the most common 01- hydroxyacids,

Ph6/—-< 0

“1"”313 Bonlcszlz 0%

Ff .
0x0 28"" orSnCl4 Rfif 1:31:94”

d.e.= 26% - 87%

P" o 1)RC(Mng) P" H .
.._ ' u r "a, .
H or R'zCuMgBr /L COM HO>/R
° ° 2) CHzNz ) O>-/R. H cit-M
H>{"'o,/\/\ H ..""/\/\

d.e. = 54% - 92%

 

Scheme 5.1
Synthesis of chiral alcohols using mandelic acid as a chiral auxiliary
has been used extensively as a chiral building block [3-8]. However, its use as a chiral
auxiliary is somewhat limited [9—11]. It is most commonly employed in the Diels-Alder

reaction. It has been reported that lactic acid can form 2,2-disubstituted 5-methyl-l,3-

122

dioxolan-4-ones with good to excellent diastereoselectivity [12]. Related dioxolanones
have been widely used recently for several applications [1, 2, 13-18], but their ring
opening by hydride attack at the C3 position has never been reported. This reaction is
very important since it can open ways to synthesize chiral alcohols using lactic acid as a
chiral auxiliary, similar to the synthesis of chiral alcohols using chiral diols as chiral
auxiliaries [19]. The latter method generally gave high to excellent diastereoselectivity,

but the auxiliary was too expensive to be used on large scale.

a-Tetralone was chosen as our substrate. The asymmetric reduction of a-tetralone can
yield a-tetralol, which is a precursor to chiral aminotetralins. Aminotetralins are known
to have high potency in the treatment of central nervous system disorders such as
Parkinson’s disease. The synthesis of chiral ‘ a-tetralol is typically achieved by
asymmetric hydrogenation [20,21], the use of chiral reducing agents [22,23], or kinetic

resolution by enzymes [24,25]. Chiral a-tetralol is still very expensive, nearly $200/g.

In our study, 3,4-dihydro-1(2H)-naphthalenespiro-2-(5’-methyl-1’,3’-dioxolan-4’-one)
was synthesized according to the known method [12]. As in the case of the reduction of
chiral acetals, Lewis acids are expected to activate the oxygen atom of the carbonyl
group to initiate ring-opening. The resulting oxonium species can then be reduced by a
reducing agent. A number of reagent combinations were investigated, among which the
combination of Zn(BH4)3, Me3SiCl and lutidine gave the best result (46% yield, 64%
d.e.). The optical purity of the starting material has little effect on the outcome of

product, which is in accordance with the findings by Mashraqui and Kellogg in 1984 [1].

123

The removal of lactic acid unit of O-[l,2,3,4-tetrahydro-(l)-naphthyl]-lactic acid proved
difficult. One of the approaches was to install a hydroxyl group at the 01- position, which
would give an unstable hemiacetal intermediate leading to the release of the chiral
auxiliary. The attempted a-hydroxylation of 0-[1,2,3,4-tetrahydro-(l)-naphthyl]-lactic
acid only caused decomposition of starting material and no reaction was observed by
starting from corresponding methyl ester. Two other approaches, the oxidative cleavage
of a-alkoxy aldehyde and the reductive elimination of B-haloalkoxy compound
successfully removed the chiral auxiliary, but racemization of product occurred during

the process.

In summary, the successful reductive opening of 1,3-dioxolan-4-ones, as well as the

unsuccessful removal of the lactic acid unit, will shed some light on attempts to use lactic

acid as a chiral auxiliary to generate chiral alcohols.

124

Results and discussion

3,4-Dihydro-1(2H)-naphthalenespiro-2-(5 ’-methyl-l ’,3 ’-dioxolan-4’-one) was
synthesized according to the known method (Scheme 5.2) [12]. The diastereoselectivity
obtained was 95:5. The reductive ring-opening was performed directly on the mixture of
cis and trans isomers unless otherwise specified. ‘Since the structure of the dioxolanone
was very similar to dioxolanes formed between ketones and diols (Figure 5.1), we
expected that reagents used for the reduction of chiral acetals (generally a combination of
a Lewis acid and a reducing agent) may also work for the reduction of dioxolanone. The
reaction of 3 ,4-dihydro- 1 ( 2H)-naphthalenespiro-2-( 5 ’-methyl-l ’,3 ’-dioxolan-4’-one)
turned out to be more complicated than expected, mainly due to the stability of the
benzylic carbocation. If not properly trapped by hydride from the reducing agent, a
number of other side products would be obtained. Table 5.1 summarized the results
using different reaction conditions, and Figure 5.2 shows the possible reaction pathways
which would lead to those products. So far, the reaction conditions for Entry 12

provided the best results with a yield of 46% and a d.e. of 64%.

The ring opened at the carboxylate end since Lewis acids prefer to complex with the
carboxylate oxygen (high electron density). Generally, decomposition of dioxolanone to
a-tetralone (and subsequently, reduction to the corresponding alcohol) was predominant
if protic acids (e.g. HC1, TFA) were used. This trend is shown in Entries 1-4. The use

of proton scavengers like lutidine can significantly inhibit the decomposition

125

.. V—i

 

H300 OCH3 o o
HC(OMe)3 dry lactic acid
——-> >
pst, 70°C Toluene,95°C

cis:trans= 95:5
Scheme 5.2

Formation of 3,4-dihydro-1(2H)-naphthalenespiro-2-(5 ’-methyl-1 ’,3 ’-dioxolan-4’-one)

l < l E 0
0x0 0x0
R1 R2 R1 R2
dioxolane dioxolanone

Figure 5.1 -

Structures of dioxolane and dioxolanone

 

O
‘ O
m
o o o H

Zn(BH4)3, Me38iCI
Lutidine 7

cis:trans = 95:5 Yield = 46%

do = 64%

Scheme 5.3

Stereoselective reduction of dioxolanone

126

process. To form the desired compound 1, the oxonium ion has to be trapped by hydride.
Trapping of the oxonium ion by moisture was extremely detrimental, leading to the
formation of side products 2 and 3. Et3SiH and NaBH4 proved to be less efficient
hydride donors than Zn(BH4)3 (Entries 4, 5, 7, 8 and 12). Lutidine was a better proton
scavenger than K3CO3 (Entries 10, 11 and 12). The yield of product was better if the
reaction was done in dichloromethane and at a lower temperature. The other interesting
finding was that the stereochemical outcome was not related to the optical purity of the
starting dioxolanone. These results are shown in Table 5.2. This observation was in
accordance with the findings by Mashraqui and Kellogg in 1984 [1]. In their work, they
made 1,3-dioxolan-4-ones by condensation of (S)-(+)- or (R)-(-)-mandelic acid with
aldehydes. Then they performed nucleophilic substitution using silyl enol ethers or
allylsilanes (Scheme 5.1). Since the same intermediate (oxonium) will result from both
diastereomers, the optical purity of the substrate was not a factor (Scheme 5.4). The
diastereoselectivity will be only determined by the steric hindrance when the nucleophile

approaches this oxonium ion (trans to the phenyl group).

Ph_ OBFs' Ph_ OBF3
major 0 O *—— +0 O
\ ‘ H >— R
R \ minor H
Scheme 5.4

Oxonium formation to explain the stereochemical outcome

127

Table 5.]

Reduction of dioxolanone under different conditions

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Entry Reaction conditions Solvent Product (8)
1 Con. HC1 (2 eq.), NaCNBH; (4 eq.), MeOH 5 (major), 2 (minor)
RT for 12 h
2 HC1 (ether) (10 eq.), NaCNBH3 (4 eq.), THF 2
RT for 12 h
3 TFA (10 eq.), NaCNBH3 (4 eq.), RT for 6 h THF 4,2
4 TFA (30 eq.), Et3SiH (1.2 eq.), 50 °C for 8 h THF 3
5 Lutidine (0.2 eq.), Et3SiH (10 eq.), TiCl4 (1.1 CH3C13 1 (major), 3 (minor)
eq.), CH3C13. —78 °C for '/2 h, then warm up to
RT for another 12 h
6 Me3SiC1 (1 eq.), Zn(BH4)3 (0.5 eq.), B30 4 (major), 5,3 (Minor)
0°C for 12 h
7 Me3SiCl (1 eq.), Zn(BH4)3 (0.5 eq.), Lutidine B30 1
(0.5 eq.), 0°C for 2 h
8 Me3SiCl (1 eq.), Zn(BH4)3 (0.5 eq.), Lutidine Et3O 4
(0.5 eq.), RT for 12 h
9 Lutidine (0.3 eq.), ZnC13 (4 eq.), NaBH4 (4 eq.), Et30 isomerization
ether, 0 °C, then gradually rise to RT for another
36 h
10 MegsiCl (1 eq.), Zn(BH4)3 (0.5 eq.), K3C03 (3 CH3C13 4,2
eq.), RT for 12 h
11 BF3.Et30 (1 eq.), Zn(BH4)3 (0.5 eq), K2CO3 (3 CH3C13 2 (major), I (trace)
eq.), RT for 12 h
12 Me3SiCl (1 eq.), Zn(BH4)3 (0.5 eq.), Lutidine CH3C13 l

 

(0.5 eq.), 0 °C for 4h, gradually raise temp. to
RT for another 10 h

 

 

 

128

 

O—l-

O

"y we j.--
8%
ob” ‘ " :Lofb

Figure 5.2

Possible pathways for reduction of dioxolanone using Lewis acid and reducing agent

Table 5.2

Stereochemical outcome starting from substrates of different optical purity

 

Entry starting mat. Reaction conditions solvent product

 

12 cis:trans = 95:5 Me3SiCl, Zn(BH4)3, Lutidine, 0°C, CH3C13 1 (de = 64%)
gradual temp. rise to RT

 

13 cis Me3SiCl, Zn(BH4)3, Lutidine, 0°C, CH3C13 1 (de = 64%)
gradual temp. rise to RT

 

 

l4 trans Me3SiCl, Zn(BH4)3, Lutidine, 0°C, CH3C13 1 (de = 64%)
gradual temp. rise to RT

 

 

 

 

 

Lewis acids can catalyze the isomerization as shown in Figure 5.2. A positive charge
generated on the benzyl position is stable enough to allow formation of the same planar
structure from both the cis and trans dioxolanones. Hydride will attack from the less
hindered side and since the difference in steric hindrance is, for the most part, due to the

methyl group of the lactic acid part, the diastereoselectivity was not excellent.

The next step is to remove lactic acid unit of O-[l,2,3,4-tetrahydro-(1)-naphthyl]-lactic
acid without racemization at the newly generated chiral center. This turned out to be
very difficult. Several reaction schemes were investigated and the results were shown in
Scheme 5.5. a-Hydroxylation of O-[l,2,3,4-tetrahydro-(1)-naphthyl]-lactic acid using
LDA followed by oxygenation led to decomposition and the methyl ester was used
unreactive. The oxidative cleavage using H303/NaOH successfully removed the

auxiliary, but the chiral center epimerized to give the racemic tetralol. Radical

130

 

abstraction of the benzylic proton was suspected. The or-alkoxy methyl ester was also
unreactive toward reductive cleavage using Sml3. Zn powder in refluxing ethanol
smoothly yielded the alcohol of interest from the iodide, but racemization occurred again

and the reason is not clear.

In summary, we have successfully achieved the stereoselective reduction of 3,4-dihydro-
1(2H)-naphthalenespiro-2-(5’-methyl-1’,3’-dioxolan-4’-one) derived from (L)-lactic
acid. The good diastereoselectivity obtained has little to do with the optical purity of the
starting dioxolanone. This trend is expected with substrates which will generate benzylic
or allylic carbocations during the reductive ring-opening process. The
diastereoselectivity will be influenced by the size difference of the two substituents of the
carbocation. Since it is the methyl group of lactic acid that differentiates the two sides of
the oxonium, a more bulky group at this center will improve the selectivity. The direct
modification of lactic acid is not as practical as hydroxylation of some common amino
acids such as L-valine. Although the sequences for removing lactic acid failed to give
chiral alcohol, they may work on those that would not yield the benzylic or allylic
alcohols. Certainly, an alternative way to remove the lactic acid unit to reach chiral

tetralol is still waiting to be discovered.

131

Smlle HF
0C HMPA
”3 ———-> No reaction

 

 

 

o H
Decomposition Eb _.
A d 649 1) LDA
.e. = 2
3 33A 10 o )0: No reaction
K2603
(11119012502
98%
O
\
o H o H o H
BH3.THF Swem
THF oxidation
———-> >
100% 96%
d.e. = 64% d.e. = 64% d.e. = 64%
1 TSCI, Py. 6 7
97%
H303, OH'

82%
OTs l
O H o H
Nal anZnCl-Acat.)
Acetone EtOH, reflux
——-> ‘ —9—8——->
95%

d.e. = 64% d.e. = 64% Mom: 00
8 9 2

 

 

 

 

Scheme 5.5

Attempted removal of auxiliary

132

Experimental

]H-NMR and l3C-NMR spectra were measured in chloroform-d solutions on a Varian-
300 spectrometer (300 MHz) unless noted otherwise. The chemical shifts are given in 5
values and are calibrated with TMS as the internal standard or relative to the chloroform
line at 7.24 ppm for lH or 77.0 ppm for 13C spectra. Silica gel flash column

chromatography was carried out on silica using Kieselgel 60 (Merck), 0.040-0.063 mm.

3,4-Dihydro-1(2H)—naphthalenespiro-Z-(5’-methyl-1’,3’-dioxolan-4’-one)

(S)-(+)-lactic acid (4.05 g, 45 mmol) dissolved in ethyl acetate (20 mL) was dried by
reaction with methyl orthofonnate (3.5 g) overnight. This mixture was added dropwise
over 1 h to a hot solution (90 °C) of 6.75 g (40 mmol) of a-tetralone dimethyl acetal [26]
in 70 mL of PhCH3. The resulting mixture was heated for an additional 1h at 95 °C. The
total volume of volatile solvents removed in the Dean-Stark trap during the reaction was
about 12 mL. The solution was then neutralized with 5 mL of pyridine and cooled to RT.
To the resulting mixture was added 70 mL of saturated aqueous NaHC03 solution. The
aqueous layer was extracted with 3 x 70 mL of B30, and the combined organic layers
were washed with brine, dried over Na3SO4, filtered and concentrated in vacuo. The
crude product was purified by column chromatography (hexane/ethyl acetate=10:2).
Yield: 6.62 g (43%) as a 95:5 mixture of isomers. Major isomer: lH-NMR: 5 1.57 (3H,

d, J = 6.9 Hz), 1.56-2.16 (4H, m), 2.78 (2H, t, J = 6.3 Hz), 4.58 (1H, q, J = 6.9 Hz), 7.07-

133

7.46 (4H. m); l3C-NMR: 173.64, 138.73, 129.65, 128.61, 126.39, 126.35, 33.78, 28.57,

19.85, 16.94.

Zn(BH4)3

To ZnCl3 (4.0 g, 0.029 mol) was added 50 mL of ether. The mixture was boiled till most
of the solid had dissolved. The mixture was allowed to stand, and the supernatant liquid
was carefully decanted from insoluble material. The ethereal zinc chloride solution was
added dropwise at RT to a stirred suspension of 2.8 g of NaBH4 in 160 mL of absolute
ether. Stirring was continued overnight and the Zn(BH4)3 ethereal solution was decanted

and stored at 5 °C. The concentration of Zn(BH4)3 was estimated to be about 0.13 M.

O-ll,2,3,4-tetrahydro-(1)-naphthy|]-Iactic acid (Entry 12)

To 1.0 g of 3,4-dihydro-l(2H)-naphthalenespiro-2-(5’-methyl-1’,3’-dioxolan-4’-one)
(cis:trans = 95:5) dissolved in 20 mL of CH3C13 was added 0.30 mL of lutidine. While
stirring at 0 °C, 20 mL of Zn(BH4)3 and 0.6 mL of Me3SiCl were added sequentially. The
mixture was stirred at 0 °C for 4 hours, after which the reaction temperature was
gradually raised to RT and the mixture was stirred for another 10 hours. Then mixture
was then washed with dilute HC1 and brine. The organic layer was then dried over
Na3SO4 and concentrated in vacuo. Yield: 0.46 g (46%) as a 82:18 mixture of two
isomers (de = 64%). Major isomer: lH-NMR: 51.49 (3H, d, J = 7.0), 1.77 (1H, m), 1.92-
2.05 (3H, m), 2.80 (2H, m), 4.21 (1H, q, J = 6.9 Hz), 4.59 (1H, t, J = 4.5 Hz), 7.06-7.40
(4H, m); l3C-NMR: 18.46, 28.48, 28.76, 72.25, 75.44, 125.76, 127.99, 129.14, 129.23,

135.33, 137.76, 176.58; Minor isomer: 5 1.48 (3H, d, J = 7.0 Hz), 1.77 (1H, m), 1.92-

134

2.05 (3H, m), 2.80 (2H, m), 4.28 (1H, q, J = 7.0 Hz), 4.52 (1H, t, J = 4.7 Hz), 7.06-7.40

(4H. m).

Other reduction conditions

Entry 1: To 40 mg of 3,4-dihydro-1(2H)-naphthalenespiro-2-(5’-methy1-l’,3’-dioxolan-
4’-one) (cis:trans = 95:5) in 5 mL of MeOH was successively added 45 mg of NaCNCH3
and 0.03 mL of concentrated HC1. The mixture was stirred at RT for 12 hours. The
reaction was quenched with 1N HC1 and MeOH was removed under reduced pressure.
The residue was partitioned between 20 mL of CHC13 and 20 mL of saturated NaHCO3.
The organic layer was washed with brine and dried with Na3SO4. After removal of
CHC13, the residue was analyzed by NMR in CDC13 which shows 5 as a major product

and 2 as a minor one.

Entry 2: To 40 mg of 3,4-dihydro-l(2H)-naphthalenespiro-2-(5’-methyl-1’,3’-dioxolan-
4’-one) (cis:trans = 95 :5) in 5 mL of THF was successively added 45 mg of NaCNBH3
and 1.8 mL of 1M HC1 in ether. The mixture was stirred at RT for 12 hours. The
reaction was quenched with 1N HC1 and THF was removed under reduced pressure. The
residue was partitioned between 20 mL of CHC13 and 20 mL of saturated NaHCOg. The
organic layer was washed with brine and dried with Na3SO4. After removal of CHC13,

the residue was analyzed by NMR in CDC 1 3 which shows 2 as the product.

Entry 3: To 40 mg of 3,4-dihydro-1(2H)-naphthalenespiro-2-(5’-methyl-1’,3’-dioxolan-

4’-one) (cis:trans = 95:5) in 5 mL of THF was successively added 45 mg of NaCNCH3

135

and 0.2 g of TF A. The mixture was stirred at RT for 12 hours. The reaction was
quenched with 1N HC1 and THF was removed under reduced pressure. The residue was
partitioned between 20 mL of CHC13 and 20 mL of saturated NaHCO3. The organic layer
was washed with brine and dried with Na3SO4. After removal of CHC13, the residue was

analyzed by NMR in CDC13 which shows 4 and 2 as the products.

Entry 4: To 40 mg of 3,4-dihydro-1(2H)-naphthalenespiro-2-(5’-methyl—1’,3’-dioxolan-
4’-one) (cis:trans = 95:5) in 5mL THF was added 26 mg of Et3SiH. The mixture was
heated at 50 °C and 0.62 g of TFA was added dropwise. After stirring at the same
temperature for 8 hours, the reaction was quenched with 1N HC1. The mixture was then
partitioned between 30 mL of CHC13 and 30 mL of H30. The organic layer was washed
with 20 mL of saturated NaHCO3 and brine and dried with Na3804. After removal of

CHC13, the residue was analyzed by NMR in CDC13 which showed 3 as the product.

Entry 5: To 40 mg of 3,4-dihydro-1(2H)-naphthalenespiro-2-(5’-methyl-l ’,3’-dioxolan-
4’-one) (cis:trans = 95:5) in 5 mL of CH3C13 at -78 °C was successively added 3.9 mg of
lutidine, 220 mg of Et3SiH, and 38 mg of TiC14. The mixture was stirred at -78 °C for 0.5
hour and warmed up to RT and stirred for 12 hours. The reaction was quenched with 1N
HC1 and was partitioned between 20 mL of CHC13 and 20 mL of H30. The organic layer
was washed with saturated NaHCO3 and brine and dried with Na3SO4. After
concentration, the residue was analyzed by NMR in CDC13 which showed mainly the

starting material and 3 as the minor product.

136

Entry 6: To 40 mg of 3,4-dihydro-l(2H)-naphthalenespiro-2-(5’-methyl-1’,3’-dioxolan-
4’-one) (cis:trans = 95:5) in 5 mL of 330 at 0 °C was successively added 20 mg of
Me3SiCl and 0.8 mL of Zn(BH4)3 (0.13M). The mixture was stirred at 0 °C for 12 hour.
The reaction was quenched with 1N HC1 and was] partitioned between 20 mL of CHC13
and 20 mL of H30. The organic layer was washed with saturated NaHC03 and brine and
dried with Na3SO4. After concentration, the residue was analyzed by NMR in CDC13

which showed 4 as a major product and 3, 5 as the minor products.

Entry 7: To 40 mg of 3,4-dihydro-l(2H)-naphthalenespiro-2-(5’-methyl-1’,3’-dioxolan-
4’-one) (cis:trans = 95:5) in 5 mL of Et30 at 0 °C was successively added 9.8 mg of
lutidine, 20 mg of MegsiCl and 0.8 mL of Zn(BH4)3 (0.13M). The mixture was stirred at
0 °C for 2 hour. The reaction was quenched with 1N HC1 and was partitioned between 20
mL of CHC13 and 20 mL of H30. The organic layer was washed with saturated NaHCO3
and brine and dried with Na3SO4. After concentration, the residue was analyzed by NMR

in CDC13 which showed the starting material.

Entry 8: To 40 mg of 3,4-dihydro-l(2H)-naphthalenespiro-2-(5’-methyl-1’,3’-dioxolan-
4’-one) (cis:trans = 95:5) in 5 mL of 330 at RT was successively added 9.8 mg of
lutidine, 20 mg of Me3SiCl and 0.8 mL of Zn(BH4)3 (0.13M). The mixture was stirred at
RT for 12 hour. The reaction was quenched with 1N HC1 and was partitioned between
20 mL of CHC13 and 20 mL of H30. The organic layer was washed with saturated
NaHCO3 and brine and dried with Na3SO4. After concentration, the residue was

analyzed by NMR in CDC13 which showed 4 as the products.

137

Entry 9: To 40 mg of 3,4-dihydro-1(2H)-naphthalenespiro-2-(5’-methyl-l’,3’-dioxolan-
4’-one) (cis:trans = 95:5) in 5 mL of Et30 at 0 0C was successively added 5.9 mg of
lutidine, 98 mg of ZnCl3 and 27 mg of NaBH4. The mixture was stirred at RT for 36
hours. The reaction was quenched with 1N HC1 and was partitioned between 20 mL of
CHC13 and 20 mL of H30. The organic layer was washed with saturated NaHCO3 and
brine and dried with Na3S04. After concentration, the residue was analyzed by NMR in

CDC13 which showed mainly the starting material (cis:trans = 82: 18).

Entry 10: To 40 mg of 3,4-dihydro-1(2H)-naphthalenespiro-2-(5’-methyl-1’,3’-
dioxolan-4’-one) (cis:trans = 95:5) in 5 mL of CH3C13 at RT was successively added 75
mg of K3CO3, 20 mg of Me3SiCl and 0.8 mL of Zn(BH4)3 (0.13M). The mixture was
stirred at RT for 12 hour. The reaction was quenched with 1N HC1 and was partitioned
between 20 mL of CHC13 and 20 mL of H30. The organic layer was washed with
saturated NaHCO3 and brine and dried with Na3SO4. After concentration, the residue

was analyzed by NMR in C DC 13 which showed 4 and 2 as the products.

Entry 11: To 40 mg of 3,4-dihydro-l(2H)-naphthalenespiro-2-(5’-methyl-1’,3’-
dioxolan-4’-one) (cis:trans = 95:5) in 5 mL of CH3C13 at RT was successively added 75
mg of K3C03, 26 mg of BF3.0Et3 and 0.8 mL of Zn(BH4)3 (0.13M). The mixture was
stirred at RT for 12 hour. The reaction was quenched with 1N HC1 and was partitioned
between 20 mL of CHC13 and 20 mL of H30. The organic layer was washed with

saturated NaHC03 and brine and dried with Na3SO4. After concentration, the residue

138

was analyzed by NMR in CDC13 which showed 3 as the major product and 1 as a minor

product.

Entry 13: To 40 mg of 3,4-dihydro-1(2H)-naphthalenespiro-2-(5’-methyl-1’,3’-
dioxolan-4’-one) (cis) in 5 mL of CH3C13 at 0 °C was successively added 9.8 mg of
lutidine, 20 mg of Me3SiCl and 0.8 mL of Zn(BH4)3 (0.13M). The mixture was stirred at
0 °C for 4 hours, after which the reaction temperature was gradually raised to RT and the
mixture was stirred for another 10 hours. The reaction was quenched with 1N HC1 and
was partitioned between 20 mL of CHC13 and 20 mL of H30. The organic layer was
washed with saturated NaHC03 and brine and dried with Na3SO4. After concentration,

the residue was analyzed by NMR in CDC13 which showed 1 as the product.

Entry 14: To 40 mg of 3,4-dihydro-l(2H)-naphthalenespiro-2-(5’-methyl-1’,3’-
dioxOlan-4’-one) (trans) in 5 mL of CH3C13 at 0 °C was successively added 9.8 mg of
lutidine, 20 mg of Me3SiCl and 0.8 mL of Zn(BH4)3 (0.13M). The mixture was stirred at
0 °C for 4 hours, after which the reaction temperature was gradually raised to RT and the
mixture was stirred for another 10 hours. The reaction was quenched with 1N HC1 and
was partitioned between 20 mL of CHC13 and 20 mL of H30. The organic layer was
washed with saturated NaHCO3 and brine and dried with Na3SO4. After concentration,

the residue was analyzed by NMR in CDC 13 which showed 1 as the product.

Attempted Removal of Chiral Auxiliary:

139

Compound 6

To 2.0 g of compound 1 was added 50 mL of dry THF and the mixture was stirred at 0 °C
for 20 min. 18.2 mL of 1M solution of BH3 in THF was then added dropwise, after
which the reaction temperature was gradually raised to RT and stirred for l h. The
reaction was quenched with 1:1 water-acetic acid with stirring until no more gas
evolution occurred. Most of the THF was removed by rotary evaporation and the residue
was partitioned between 100 mL of ethyl acetate and 50 mL of sat. NaHCO3 solution.
The aqueous layer was extracted with 50 mL of ethyl acetate and the organic layers were
combined and washed with brine. The organic layer was then dried over Na3SO4 and
concentrated in vacuo to give compound 6. Yield: 1.9 g (100%) as a 82:18 mixture of
two isomers (d.e. = 64%). Major isomer: 'H NMR (300 MHz, CDC13) a 1.23 (3H, d,
J=6.0 Hz), 1.73-2.02 (4H, m), 2.66-2.88 (2H, m), 3.48 (1H, dd, J=11.7 and 7.8 Hz), 3.56
(1H, dd, J=l 1.7 and 3.6 Hz), 3.77-3.88 (1H, m,), 4.52 (1H, t, J=4.5 Hz), 7.08-7.37 (5H,
m); l3C NMR (75 MHz, CDC13) 5 16.19, 18.88, 28.17, 29.11, 66.63, 72.42, 73.42,
125.88, 127.42, 128.86, 129.18, 136.81, 137.43; HRMS Exact mass: calcd for C13HI7O3

[M-H] H, 205.1229, found 205.1235.

Compound 7

To a stirred solution of 0.15 mL of (COC1)3 in 15 mL of dry CH3C13 was added dropwise
0.36 mL of Me3SO. The mixture was stirred 15 min. at 78 °C. A solution of 0.25 g of
compound 6 in 10 mL of CH3C13 was added dropwise via a syringe. After 15 min.
additional stirring, 1.6 mL of i-Pr3NH was added and the mixture was allowed to warm

up to RT. After 1h, the reaction was poured into 20 mL of sat. NaHC03 solution and the

140

aqueous layer was extracted with 50 mL of CH3C13. The organic layer was washed with
brine and after removal of the solvent the residue was dissolved in 50 mL of hexane and
washed with 30 mL of water. The organic layer was then dried over Na3SO4 and
concentrated in vacuo. The crude product was further purified by column
chromatography to yield compoun 7. Yield: 0.24 g (96%) as a 82:18 mixture of two
isomers (d.e. = 64%). Major isomer: 'H NMR (300 MHz, CDCI3) 5 1.33 (3H, d, J=6.9
Hz), 1.70-2.08 (4H, m), 2.65-2.89 (2H, m), 4.10 (1H, q, J=6.9 Hz), 4.55 (1H, t, J=4.8
Hz), 7.08-7.32 (5H, m), 9.69 (1H, d, J=1.5 Hz); l3C NMR (75 MHz, CDC13) 5 15.76,
18.54, 28.81, 28.84, 75.36, 78.54, 125.80, 127.78, 129.02, 129.10, 125.98, 127.64,

203.94.

Compound 2

0.20 g of compound 7 was added 0.86 g of 30% H303 and 5 mL of H30. The pH was
maintained between 8~9 by adding just enough 0.5N NaOH solution. The mixture was
heated at 60 °C for 6 h and then cooled. It was partitioned between 50 mL of ethyl
acetate and 30 mL of water. The organic layer was washed with brine, dried over
Na3SO4 and concentrated in vacuo. The crude product was further purified by column
chromatography to yield compound 2. Yield: 0.12 g (82%). lH NMR (300 MHz,
CDC13) 5 1.78-2.02 (4H, m), 2.70-2.89 (2H, m), 4.78 (1H, t, J=5.1 Hz), 7.10-7.45 (5H,
m); l3C NMR (75 MHz, CDC13) 5 19.03, 29.48, 32.50, 68.36, 126.41, 127.81, 128.91,

129.26, 137.36, 139.04; [(1][)20 = 0° (c 1.84, CHC13).

141

Compound 8

To 1.0 g of compound 6 in 20 mL of dry pyridine was added 1.1 1 g og p-toluenesulfonic
chloride. The mixture was stirred at RT overnight. The pyridine was removed by rotary
evaporation and 10 mL of toluene was added to the residue. The solvent was removed
and the residue was diluted with 100 mL of ethyl acetate and washed with water and
brine. The organic layer was dried over Na3SO4 and concentrated in vacuo. The crude
product was further purified by column chromatography to yield compound 8. Yield:
1.70 g (97%) as a 82:18 mixture of two isomers (d.e. = 64%). Major isomer: IH NMR
(300 MHz, CDC13) 5 1.21 (3H, d, J=6.0 Hz), 1.66-1.93 (4H, m), 2.41 (3H, s), 2.60-2.77
(2H, m), 3.91-4.02 (2H, m), 4.43 (1H, t, J=4.8 Hz), 7.05-7.29 (6H, m), 7.77 (2H, d,
J=8.4 Hz); ”C NMR (75 MHz, CDC13) 5 17.05, 18.42, 21.51, 28.20, 28.93, 70.81, 72.96,
73.73, 125.69, 127.42, 127.84, 128.83, 129.14, 129.67, 137.48, 144.66; HRMS Exact

mass: calcd for C30H3304S [M-H]+' = 359.1318, found 359.1332.

Compound 9

To 1.0 g of compound 8 was added 50 mL of acetone and 4.16 g of Nal. The mixture
was refluxed for 24 h. Acetone was removed by rotavap and the residue was partitioned
between 50 mL of diethyl ether and 30 mL H3O. The organic layer was subsequently
washed with water and brine. It was then dried over Na3SO4 and concentrated in vacuo
to yield compound 9. Yield: 0.83 g (95%) as a 82:18 mixture of two isomers (d.e. =
64%). Major isomer: 1H NMR (300 MHz, CDC13) 5 1.39 (3H, d, J=6.0 Hz), 1.71-2.12
(4H, m), 2.68-2.90 (4H, m), 3.23 (1H, d, 2.1 Hz), 3.25 (1H, d, J=2.1 Hz), 3.77 (1H, m),

4.47 (1H, t, J=4.8 Hz), 7.08-7.34 (5H, m); 13C NMR (75 MHz, CDC13) 5 11.78, 18.82,

142

20.42, 28.65, 29.00, 73.40, 73.52, 125.79, 127.46, 128.86, 128.92, 136.68, 137.55;

HRMS Exact mass: calcd for C13H|601 [M-H]+' = 315.0247, found 315.0264.

Compound 2

To 0.37 g of compound 9 was added 50 mL of absolute EtOH, 0.39 g of zinc powder and
0.02 g of ZnC13. The mixture was refluxed for 6 h and cooled. It was then filtered
through Celite and the EtOH was removed by rotary evaporation. The residue was
partitioned between 100 mL of ether and 50 mL of water. The organic layer was washed
with brine, dried over Na3SO4 and concentrated in vacuo to yield Compound 2. Yield:

0.17 g (98%). [(11020 = 0 0 (c 2.12, CHC13).

Compound 10

To 1.00 g of compound 1 in 50 mL of acetone was added 0.94 g of K3C03 and 0.50 mL
(MeO)3SO3. The mixture was refluxed for 3 h. Acetone was removed by rotavap and the
residue was partitioned between 100 mL of ethyl acetate and 50 mL of 0.5 N acetic acid.
The organic layer was washed with water and brine. It was then dried over Na3SO4 and
concentrated in vacuo. Pure compound 10 was obtained after column chromatography.
Yield: 1.04 g (98%) as a 82:18 mixture of two isomers (d.e. = 64%). Major isomer: lH
NMR (300 MHz, CDC13) 5 1.41 (3H, d, J=6.9 Hz), 1.67-2.10 (4H, m), 2.64-2.88 (2H,
m), 3.76 (3H, s), 4.18 (1H, q, J=6.9 Hz), 4.47 (1H, t, J=4.5 Hz), 7.07-7.26 (5H, m); 13C
NMR (75 MHz, CDC13) 5 18.71, 19.28, 28.70, 28.90, 73.24, 75.32, 75.94, 126.03,
128.43, 128.99, 129.62, 135.80, 137.74, 174.26. HRMS Exact mass: calcd for C14H1703

[M-H] fl, 233.1178, found 233.1176.

143

10.

ll.

12.

13.

14.

15.

16.

References

. Mashraqui, S.; Kellogg, R. J. Org. Chem. 1984, 49, 2513-2516.

Heckmann, B.; Mioskowski, 0; Yu, J .; Falck, J. R. Tetrahedron Lett. 1992, 33, 5201-
5204.

Vicart, N.; Ortholand, J .; Emeric, G.; Greiner, A. Tetrahedron Lett. 1994, 35, 3917-
3918.

Rodriguez, G.; Biellmann, J. J. Org. Chem. 1996, 61, 1822-1824.

Keck, G. E.; Savin, K. A.; Weglarz, M. A.; Cressman, N. K. Tetrahedron Lett. 1996,
37, 3291-3294.

Enders, D.; Jandeleit, B. Synthesis 1994, 1327-1330.

Cheng, Z.; Hamada, Y.; Shioiri, T. Synlett 1997, 109-110.

Matsuo, K.; Sakaguchi, Y. Chem. Pharm. Bull. 1997, 45, 1620-1625.

Hartmann, H.; Hady, A.; Sartor, K.; Weetman, J.; Helmchen, G. Angew. Chem. Int.
Ed. Engl. 1987, 26, 1143.

Charlton, J. and Pham, C., Tetrahedron Lett. 1992, 33, 6073-6076.

Faunce, J. A.; Grisso, B. A.; Mackenzie, P. B. J. Am. Chem. Soc. 1991, 113, 3418-
3426.

Ortholand, J.; Vicart, N.; Greiner, A. J. Org. Chem. 1995, 60, 1880-1884.

Seebach, D.; Imwinkelried, R; Stucky, G. Helv. Chim. Acta. 1987, 70, 448-457.
Reddy, V.; Franck, R. J. Org. Chem. 1993, 58, 691 1.

Beckwith, A.; Chai, C. Tetrahedron 1993, 49, 7871.

Kneer, 0.; Mattay, J. Tetrahedron Lett. 1992, 33, 8051-8054.

144

17.

18.

19.

20.

21.

22.

23.

24.

25.

26.

Roush, W.; Brown, B. J. Org. Chem. 1992, 57,3380-3387.

Krysan, D.; Mackenzie, P. J. Am. Chem. Soc. 1988, 110, 6273-6274.

Alexakis, A. Mangeney, P. Tetrahedron Asymm. 1990, 1, 477.

Zhang, X.; Taketomi, T.; Yoshizumi, T.; Kumobayashi, H.; Akutagawa, S. J. Amer.
Chem. Soc. 1993, 115, 3318-3319.

Noyori, R.; Tomino, 1.; Tanimoto, Y.; Nishizawa, M. J. Amer. Chem. Soc. 1984, 106,
6709-6716.

Midland, M.; McLoughlin, J. J. Org. Chem. 1984, 49, 1316-1317.

Hong, Y.; Gao, Y.; Nie, X.; Zepp, C. Tetrahedron Lett. 1994, 35, 6631-6634.

Lin, G.; Liu, H. Tetrahedron Lett. 1995, 36, 6067-6068.

Laumen, K.; Schneider, M. J. Chem. Soc. Chem. Commun. 1988, 598-600.

Napolitano, E.; Fiaschi, R.; Mastrorilli, E. Synthesis 1986, 122.

145

   

”11111111111111lllllllflllllllljlllis