MSU

LIBRARIES
m

».

 

 

RETURNING MATERIALS:
Place in book drop to
remove this checkout from
your record. FINES will
be charged if book is
returned after the date
stamped below.

 

 

 

 

 

 

ASYMMETRIC HYDROGENATION OF PROCHIRAL OLEFINS
WITH LAYERED SILICATE INTERCALATION CATALYSTS
AND SYNTHESIS AND CHARACTERIZATION OF NEW
FUNCTIONALIZED CHIRAL DIPHOSPHINE LIGANDS

By

Han-Min Chang

A DISSERTATION

Submitted to
Michigan State University
in partial fulfiilment of the requirements

for the degree of

DOCTOR OF PHILOSOPHY

Department of Chemistry

1982

ABSTRACT

ASYMMETRIC HYDROGENATION OF PROCHIRAL OLEFINS
WITH LAYERED SILICATE INTERCALATION CATALYSTS
AND SYNTHESIS AND CHARACTERIZATION OF NEW
FUNCTIONALIZED CHIRAL DIPHOSPHINE LIGANDS

By
Han-Min Chang

Cationic rhodium(I) complexes of the type Rh(diene)
(diphos*)+, where diphos* is a chiral diphosphine ligand,
(+)-2,3-0-isopropylidene-2,3-dihydroxy-l,4-bis(diphenyl-
phosphino)butane [DIOP(+)], SPIPHOS, (R)-l,2-bis(diphenyl-
phosphino)propane [(R)-Prophos] or (g)-l,2-bis(di-¢-
methylphenylphosphino)propane [(R)-4-Me-Prophos] and
where diene is norbornadiene (N80) or 1,5-cycloocta-
diene (COD), were intercalated into mica-type swelling
layer silicates such as Na-hectorite. The preparations
of the intercalation catalysts were readily achieved
through simple cationic exchange. The intercalated
rhodium(I) chiral diphosphine complexes were highly
effective asymmetric hydrogenation catalysts for the
hydrogenation of dehydroamino acids to the corresponding
amino acids at ambient conditions. The effect of inter-
calation on optical yields as a function of substrate

size, interlayer swelling and chelate ring size was studied.

HaneMin Chang

Seven amino acid precursors of alanine, phenylalanine,
tyrosine and DOPA were studied in the silicate and
homogeneous systems. For the intercalated [Rh(NBD)
DIOP(+)]+ catalyst, the optical yields of the corre-
sponding amino acids were comparable to the homogeneous
results. When the ligand was SPIPHOS, (R)-Prophos
or (R)-4-Me-Prophos, the optical yields were comparable
to or lower than those obtained with the homogeneous
analogues. The results obtained with the intercalated
catalysts are comparable to those reported by others
using rhodium diphos* complexes immobolized on polymers.
The new, functionalized diphos* ligand, 4-(l -
tetrahydropyranloxy)-l,2-(R)-bis(diphenylphosphino)
butane [THP-butaphos], 4-hydroxy -l,2-(B)-bis(diphenyl-
phosphino)butane [(3)-Hydroxylbutaphos] and 4-
{[(twrt-butyloxy)carbonyl]-isobutylamino}-l,2-(3)-
bis(di-4’-methylphenylphosphino)butane [N-BOC-butaphos]
were synthesized from L-malic acid. The N-BOC-butaphos
was highly effective for asymmetric hydrogenation of
dehydroamino acids.' The optical yields of the corres-
ponding amino acids with homogeneous [Rh(NBD)N-BOC-
butaphos]+Cl04 catalyst were in the range of 86-96%.
The homogeneous catalyst was more effective and stereo-

selective in THF than in 95% EtOH.

© 1982

HAN-MIN CHANG

All Rights Reserved

TO MY PARENTS AND MICHELE

ii

ACKNOWLEDGEMENTS

I would like to sincerely thank Professor
Thomas J. Pinnavaia for his guidance, support,
and valuable advice throughout my graduate study.
I would also like to extend my appreciation to
Professor Carl H. Brubaker for editorial assistance
as a second reader.

Thanks goes to my research committee and
Professor William H. Reusch for his generosity
in making laboratory equipment available.

I would like to acknowledge the financial
support of the National Science Foundation and
Michigan State University.

Special thanks to my parents and parents-in-

law, and to my wife.

1'11

Chapter

TABLE OF CONTENTS

LIST OF TABLES

LIST OF FIGURES

INTRODUCTION
1. The Historical Development of
Asymmetric Hydrogenation
II. Chiral Diphosphine Ligands and
Their Application to Asymmetric
Hydrogenation . . . . .
A. Rigid Ligands
l. Ligands Chiral at Carbon
Backbone . . . .
2. Ligand Chirality at Phosphorus.
B. DIOP and DIOP Analogues
C. Pyrrolidinephosphines
D. Ferrocenylphosphines
E. Atropisomeric Ligands
F. Bis-aminophosphines
G. Polymeric Chiral Diphosphines
III.

Asymmetric Hydrogenation Mechanism
Study . . . . . . .

A. Enamide Bonding Structure

B. Proposed Mechanisms and
Asymmetric Induction Steps

C. §-; Dehydroamino Acid
Isomerization . . . .

iv

Page
viii

XV

TO
10

l0
16
18
20
24
29
3O
35

4O
4T

4T

47

Chapter

IV.
V.

Catalyst Supports

Research Objectives

RESULTS AND DISCUSSION

1.

Asymmetric Hydrogenation of
Prochiral Olefins with Layered
Silicate Intercalation Catalysts

A. Diphosphine Ligands

B. General Method for the
Preparation of Cationic
Catalysts . .

C. Intercalation Catalysts and
Interlayer Spacings

D. Preparation of Dehydroamino
Acid Derivatives via Unsaturated
Azlactones . . .

E. Hydrogenation Conditions

F. Characterization of Dehydroamino
Acids and Their Corresponding
a-Amino Acids

G. Optical Yields

1. (3)-Prophos (6) and (R)-4-
Me-Pronhos (Q9) Systems

2. SPIPHOS (QB) System
3. DIOP(+) (g) System

'4. Solvent Effect on the
Optical Yields .

5. The Effect of Rhodium Complex
Precursor on Optical Yields

6. Asymmetric Hydrogenation with
Recycled Catalysts

Page
47
El

54

54

6O

62

63
64

67

71

72
77
79

82

83

84

Chapter

The Observed Hydrogenation Rates

A Comparison of Intercalated
Rh(I)-DIOP (R) System with
Other Rh(I)-DIOP Systems . . . . . . .

The Probable Asymmetric Hydrogenation
Mechanism 31P NMR Spectra of the
Solution Structures of (R)- 4- Me-
Prophos (49) and its Rhodium Complexes

II. Synthesis and Characterization of

New

Functionalized Chiral Diphosphine

Ligands

A.

EXPERIMENTAL
A.

Synthesis of the Key Synthetic
Precursor .

Synthesis of 4- (l -Tetrahydro-
pyranloxy)- l ,2-(R)- bis(diphenyl-
Phosphino)butane, [THP- -butaphos
(51)] and 4- Hydroxy -l ,2-(R)-
(diphenylphosphino)butane

[(R)- Hydroxylbutaphos (63)]

Characterization of THP-butaphos
(6]) and (R)-Hydroxylbutaphos (63)

Synthesis of 4- {[(tert- Butyloxy)
carbonle-isobutylamino}- l, 2- (R)-
bis(di- 4 methylphenylphosphino)
butane [N- BOC- butaphos (6})].

Characterization of 4- {[( tert-

Butyloxy)carbonyl]- -isobutylamino}
l, 2- (R)- bis(di- 4 -methylphenyl-
phosphino)butane [N- BOC- butaphos
53 . . . .

The Observed Hydrogenation Rates

and Optical Yields in the N- BOC-
butaphos (53) System . .

Suggested Future Work

General

vi

?_'Bage

87

Ill

119

T32

T35

137

139

141

I44

149
157

160

Chapter

N.
APPENDIX

Substrates

Identification of a-Amino Acid
Derivatives (Hydrogenated
Products) . . . . .

Preparation of (R)-Prophos (g)
and (R)-4-Me-Prophos (43) . .

Preparation of the Synthetic
Precursor, Acetonide-triol (68)

Synthesis of THF-butaphos (SJ)
and (R)-Hydroxylbutaphos (6})

Synthesis of N-BOC-butaphos (6})
Hectorite
Catalyst Precursors

X-ray Powder Diffraction Measure-
ments

Hydrogenation Procedure
Production Isolation
Chemical Conversions

Optical Yields

BIBLIOGRAPHY

vii

Page
162

165

166

170

170
175
180
180

183
184
185
185
186
187
205

LIST OF TABLES

Table Page

l Asymmetric Hydrogenation of

Prochiral Olefins Using a

Rh(I)-PMePh-n-Pr (1) Catalyst

Precursor . . . . . . . . . . . . . . . . . 5
2 Asymmetric Hydrogenation of I

Prochiral Olefins with

Rh (-)DIOP (3) Catalyst

Precursor . . . . . . . . . . . . . . . . . 7
3 Asymmetric Hydrogenation of

(;)-Dehydroamino Acids with

Rh(I)-ACMP (4) Catalyst

Precursor . . . . . . . . . . . . . . . . . 9
4 Asymmetric Hydrogenation of

Dehydroamino Acids Using

DIOP (R) and Carbocylic

Analogues . . . . . . . . . . . . . . . . 20
5 Asymmetric Hydrogenation of

(g)-Dehydroamino Acids with

In-situ Rh(I)-BPPM (2})

Complex at 20°C and 50 atm . . . . . . . . 22

viii

Table

10

11

12

Page
Optical Yields of (R)-(-)
Pantolactone by Using PPM (g6)
and its Analogues . . . . . . . . . . . . 24
Asymmetric Hydrogenation
Catalyzed by Rh(I)-(§)-(R)-
BPPFA (gj) Complex . . . . . . . . . . . 26
Asymmetric Hydrogenation
of Carbonyl Compounds

RI-co-R2

with (R)-(§)-BPPFOH

(29) or (R)-(S)-BPPFA (g8) . . . . . . . 27
Asymmetric Hydrogenation of

Aminomethyl Aryl Ketones with

(R)-(§)-BPPFOH (29) . . . . . . . . . . 29
Asymmetric Hydrogenation of

(;)-Dehydroamino Acid Derivatives . . . . 34
Asymmetric Hydrogenation of

Itaconic Acid and (L)-c-

Acetamidocinnamic Acid with

Polymer Supported Rh(I)-

4VB-PPM-HEMA (4]) Catalysts

and Homogeneous Analogues . . . . . . . 39
Spectra Parameters and Their

Physical Properties of (R)-

Prophos (g) and (R)-4-Me-

Prophos (49) . . . . . . . . . . . . . . 59

ix

Table Page
13 Preparation of (g)-Dehydro-
Amino Acids . . . . . . . . . . . . . . . . 55

14 ‘3

C NMR Chemical Shifts of

Dehydroamino Acids and Their

Corresponding a-Amino Acids

in 06-DMSO Solution at 20.03

MHz and 25°C . . . . . . . . . . . . . . . 53

15 ‘3

C NMR Chemical Shifts of

Dehydroamino Acids and Their

Corresponding a-Amino Acids

in 06-DMSO Solution at 20.03}

MHz and 25°C . . . . . . . . . . . . . . . . 59

13c NMR Chemical Shifts of

l6

Dopa Precursor and L-DOPA in

06-DMSO Solution at 20.05

MHz and 25°C . . . . . .L. . . . . . . . . . 7o
l7 Asymmetric Hydrogenation of

Dehydroamino Acids with

Intercalated and Homogeneous

[Rh(NBD)(R)-Prophos (§)]+

Catalysts . . . . . . . . . . . . . . . . . . 73
18 Asymmetric Hydrogenation of

Dehydroamino Acids with

Intercalated and Homogeneous

[Rh(NBD)(R)-4-Me-Prophos (4~9)]+
Catalysts . . . . . . . . . . . . . . . . . . 74

Table Page
19 Difference in Optical Yields
(YI - Y“) for Intercalated
and Homogeneous Cationic
[Rh(NBD)(diphos*)]+ Precursors . . . . . . 76
20 Asymmetric Hydrogenation of
Dehydroamino Acids with
Intercalated and Homogeneous
[Rh(NBD)SPIPHOS (453)]+
Catalysts . . . . . . . . . . . . . . . . . 78
21 Asymmetric Hydrogenation of
Dehydroamino Acids with
Intercalated and Homogeneous
[Rh(NBD)DIOP(+) (3)]+
Catalysts . . . . . . . . . . . . . . . . . 30
22 Solvent Effect on the
Optical Yields of Hydrogenation
Product of 1-Acetamino-
l-phenylethene with DIOP(+) (3) . . . . . . 81
23 Asymmetric Hydrogenation of
(g)-a-Benzamidocinnamic Acid
with Intercalated and Homogeneous
[Rh(NBD)(R)-Prophos (6~)]+ Catalysts . . . . 82
24 Asymmetric Hydrogenation of
Dehydroamino Acids with
Intercalated and Homogeneous

[Rh(diene)(R)-Prophos (6)]+ Catalysts . . . 85

xi

Table Page

25 Asymmetric Hydrogenation of

(Q-a-Acetamidocinnamic Acid

with Recycled Intercalated

[Rh(NBD)(diphos*)]+ Catalysts . . . . . . 86
26 Hydrogenation Rates for

Intercalated [Rh(NBD)(R)-Prophos

(6)]+ Catalyst and Its Homogeneous

Analogue . . . . . . . . . . . . . . . . . 100
27 Hydrogenation Rates for

Intercalated [Rh(NBD)(R)-

4-Me-Prophos (49)]+ Catalyst

and Its Homogeneous Analogue . . . . . . . 101
28 Hydrogenation Rates for

Intercalated [Rh(NBD)SPIPHOS

(48)]+ Catalyst and Its

Homogeneous Analogue . . . . . . . . . . 103
29 Hydrogenation Rates for

Intercalated [Rh(NBD)DIOP(+)

(§)]+ Catalyst and Its

Homogeneous Analogue . . . . . . . . . . 105
30 Hydrogenation Rates of (g)-

a-Acetamidocinnamic Acid with

In-aitu Rhodium Diphosphine

Catalysts . . . . . . . . . . . . . . . 107

Table Page

31 Asymmetric Hydrogenation of

Prochiral Olefins with

Polymer Supported Rh DIOP

Catalyst and Homogeneous

Analogue . . . . . . . . . . . . . . . . 112
32 Optical Yields for the

Intercalated [Rh(NBD)DIOP(+)

(3)]+ System Compared with

Other Rh(I)-DIOP Systems . . . . . . . . 115

33 31

P NMR Parameters of Chiral

Diphosphines (L) and Their

Rhodium(I) Complexes . . . . . . . '.° . 129
34 Hydrogenation Rates for .

Intercalated [Rh(NBD)-N-BOC-

butaphos (53)]+ Catalyst

and Its Homogeneous Analogue . . . . . . 152
35 Asymmetric Hydrogenation of

Dehydroamino Acids with

Homogeneous [Rh(NBD)-N-BOC-

butaphos (53)]+ Catalyst . . . . . . . . 153
36 Asymmetric Hydrogenation of

Dehydroamino Acids with

Intercalated and Homogeneous

[Rh(NBD)-N-BOC-butaphos (53)]+

Catalyst . . . . . . . . . . . . . . . . 155

xiii

Table Page

37 Basal Spacings of [Rh(NBD)diphos*]+-

hectorite . . . . . . . . . . . . . . . . . 184
38 Specific Rotations of Pure

Amino Acid Derivatives . . . . . . . . . . 186
39 High Resolution Mass Spectrum

of (R)-4-Me-Prophos (49) . . . . . . . . . . 802
40 High Resolution Mass Spectrum

of N-BOC-butaphos (53) . . . . . . . . . . 202

41 High Resolution Mass Spectrum
'of Benzyl-diol (59) . . . . . . . . . . . . 203
42 High Resolution Mass Spectrum

of (64) . . . . . . . . . . . . . . . . . . 203

43 High Resolution Mass Spectrum

of N-BOC-aminodiol (66) . . . . . . . . . . 204

xiv

Figure

LIST OF FIGURES

Page

X-ray structure of [Rh(DIPHOS)

(MAC)]+ . . . . . . . . . . . . . . . . . . 42
Hydrogenation catalytic cycle

by unsaturate route . . . . . . . . . . . . 43
Hydrogenation catalytic cycle

by the hydride route . . . . . . . . . . . . 46
Structure of the mica-type

swelling silicates such as

montmorillonite or hectorite . . . . . . . . 43
1H NMR spectrum of (R)-4-Me-

Prophos (49) in CDCl3 at

25°C (60 MHz) . . . . . . . . . . . . . . . 51
Hydrogen uptake plots for

reduction of 5 mmol of
N-acetyldehydrophenylalanine

with homogeneous and inter-

calated [Rh(NBD)(R)-Prophos

(6)]+ catalysts . . . . . . . . . . . . . . 33
Hydrogen uptake plots for

reduction of N-acetyldehydro-

phenylalanine with intercalated
[Rh(NBD)(R)-Prophos (2)]+

catalyst and recycled catalyst . . . . . . . 90

XV

 

 

Figure

10

11

12

Page

Hydrogen uptake plots for

reduction of N-acetyldehydro-

phenylalanine with homogeneous

and intercalated [Rh(NBD)

(R)-4-Me-Prophos (49)]+

catalyst . . . . . . . . . . . . . . . . . 92
Hydrogen uptake plots for

reduction of N-acetyldehydro-

phenylalanine with homogeneous

and intercalated [Rh(NBD)

SPIPHOS (45)]T catalyst . . . . . . . . . . 94
Hydrogen uptake plots for

reduction of N-acetyldehydro-

phenylalanine with homogeneous

and intercalated [Rh(NBD)

DIOP(+) (3)]+ catalyst . . . . . . . . . . 96
Hydrogen uptake plots for

reduction of N-acetyldehydro-

phenylalanine with intercalated

[Rh(NBD)DIOP(+) (33)]+ catalyst

and recycled catalyst . . . . . . . . . . . 98
Asymmetric hydrogenation

catalytic cycle using the

unsaturate route . . . . . . . . . . . . . 121

xvi

Figure

13

14

15

16

17

18

19

The probable mechanism of
asymmetric hydrogenation of
dehydroamino acids (enamides)
using cationic rhodium(I)

chiral diphosphine catalysts

31

The P NMR spectrum of (R)-

4-Me-Prophos (49) in CDC13

at 25°C

31

The P NMR spectrum of 0.01

M solution of [Rh(NBD)(R)-

4-Me-Prophos (4__9)]+C104

in CDC1 at -16°C .

31

3

The P NMR of 0.01 M of

[Rh(Z-ester)(R)-4-Me-Prophos

(43)]‘k1o4 in cn3oo at -30°c

31

The P NMR spectrum of THP-

butaphos (51) in CDC13, at

25°C
31P NMR spectrum of (R)-
Hydroxylbutaphos (52) in
EtOD, at 25°C

1H NMR spectrum of N-BOC-
butaphos (53) in CDC13, at

25°C (250 MHz)

xvii

Page

123

127

128

130

140

142

145

Figure

20

21

22

23

24

25

26

1

butaphos (63) in CDC1

at 25°C (250 MHz)
31

butaphos (63) in CDC1

at 0°C

Hydrogen uptake plots for
reduction of N-acetyldehydro-
phenylalanine with homogeneous

and intercalated [Rh(NBD)

H NMR spectrum of N-BOC-

39

P NMR spectrum ofIV-BOC-

3’

N-BOC-butaphos (§§)]+

catalyst .
1

H NMR spectrum of Acetonide-

triol (68) in CDC13, at

25°C (250 MHZ)
1

ether-ditosylate (69) in

H NMR spectrum of Benzyl-

CDC13, at 25°C (60 MHz)

1

H NMR spectrum of Acetonide-

toyslate (63) in CDC13,

at 25°C (250 MHZ)
1

butylamino-l,2-0-isopropy-

lidene-l,2-(S)-butanediol

H NMR spectrum of 4-150-

(64) in CDC13, at 25°C

(250 MHz)

xviii

Page

147

148

150

187

- 188

- 189

. 190

Figure

27

28

29

30

31

32

33

34

35

36

37

1H NMR spectrum of Aminodiol

(9?) in coc1
1

3 at 25°C (250 MHZ)

H NMR spectrum of N-BOC-
aminodiol (66) in CDC13, at
25°C (250 MHz)

1H NMR spectrum of N-BOC-

aminoditosylate (6]) in

CDC13 at 25°C (250 MHZ)
Mass spectrum of (R)-Prophos
(6)

Mass spectrum of (R)-4-Me-
Prophos (49)

Mass spectrum of N-BOC-
butaphos (63)

Mass spectrum of N-BOC-

aminodiol (66)

31P NMR spectrum of (R)-Prophos
(g) in CDC13 at 25°C
31

P NMR spectrum of [Rh(NBD)(R)-
Prophos (6)]C104 in CDC13 at -20°C

31p NMR spectrum of [Rh(NBD)SPIPHos (433)]+

C10 in CDC1 at 25°C

4 3
3‘P NMR spectrum of Rh SPIPHOS (4g)

PPh C1 in CDC1 at 410°C

3 3

Page

191

192

193

194

195

196

197

198

199-

200

201

LIST OF SYMBOLS AND ABBREVIATIONS

AIBN azobisisobutyronitril

Ar aromatic group

br broad

brs broad singlet

brm broad multiplet

COD 1,5-cyclooctadiene

d doublet

dd doublet of doublets
diphos* any chiral or achiraldiphosphine
DOPA B-3,4-dihydroxyphenyl-a-alanine
dt doublet of triplets

e.e. enantiomeric excess

m multiplet

M+ parent ion peak

Me methyl

MS mass spectrum

NBD norboradiene

NMR Nuclear Magnetic Resonance
O.Y. optical yield

Ph phenyl group

s singlet

S solvent

t triplet

THF tetrahydrofuran

TMS
(g) and (g)

tetramethylsilane

The prefix 1 (from zusammen,

German = together) and R

(entegen, German = opposite)

3V3 °\_/°'

b/__\b' b/_\a'
(z) . (g)

chemical shift

INTRODUCTION

Recently, asymmetric synthesis has become very im-
portant in organic synthesis, because a great number of
the new synthetic methods have been developed to control
the stereochemistry of reactions and produce highly optical-

ly pure products in high yields.1

New developments in this
field offer great opportunities for food, pharmaceutical
and agricultural industries to adopt these new methodolo-
gies to produce the pure optically active commercial
products. In many important chiral compounds only one
enantiomer is active for pharmaceuticals, food additives,
perfumes and insecticides; the other isomer is inactive
or even toxic. For example, (§)-asparagine is bitter,
whereas the (R)-isomer is sweet; 3-chloro-1,2-(R)-propane-
dial is toxic, but the (§)-enantiomer is under study as a
male antifertility agent. However, there are drawbacks
in most of the new methods of asymmetric synthesis. Op-
tically active starting materials are needed or stoichio-
metric chiral reagents are necessary to mediate the reac-
tions. It is also difficult to recycle the chiral reagents.
Asymmetric catalysis is the most effective way to
perform asymmetric synthesis, since only a catalytic amount
Of chiral reagent is needed to produce large quantities of

oD‘tically pure product without resolution. Asymmetric

catalysis with transition metal complexes has been an
exciting research area for the last ten years.' The
catalysts which contain chiral phosphines have proved to

be especially useful for a variety of catalytical asymmetric

reactions."2

In this introduction, the recent advances

in asymmetric hydrogenation catalysis are surveyed. The
focus of the study is asymmetric hydrogenation with rhodium
chiral phosphine complexes and synthetic development of chi-
ral phosphine ligands.

Asymmetric synthesis occurs as a result of a reagent
reacting with a substrate to form diastereomeric transi-
tion states. One of the reactants must have a chiral center.
The free energy difference between the diastereomeric
transition states AAGf determines the excess of one enan-
tiomer over the other. For example, AAGF of approximately
2 kcal of 0°C is considered essential to provide one of
the enantiomers at 80% excess (90:10 mixture).

la asymmetric

According to Morrison's classification,
synthesis can be carried out in one of two ways, either
by intramolecular or by intermolecular chiral mediation.
In intramolecular mediation, a second chiral center is
created in a molecule under the influence of an existing

chiral center in that same molecule (Scheme 1).

 

6h 6h Rh

('3=O 1) EtMgBr Ho—(ll—Et + Et—(f‘OH
H—<l:- OH 2) H20 7 H—(IZ- 0H 11-- cl-0H

Ph Ph Ph
(.3) " (') “331120111 (R33) (B$§)

Major Diastereomer

Scheme 1

In intermolecular chiral mediation, a chiral reagent
interacts with a prochiral substrate to mediate the

creation of a new chiral center (Scheme 2).

 

H
’/
Ow
N \
o \\ ’,th CH3
|| Li ‘31 1
CH 3— C —Ph 2 t H—(li—OH
chiral hydride Ph
reagent

§(-)-92% e.e.

Scheme 2

The definition of enantiomeric excess (e.eJ or optical
yield (O.Y.) is the ratio of actual specific rotationEa]
to the absolute rotation [a]abs, i.e., to the specific

rotation of the pure enantiomer.

e.e.= O.Y. = [“1 x 100%
“has

I. The Historical Development of Asymmetric Hydrogenation

In 1964, Wilkinson discovered that Rh(PPh3)3C1 and its
related derivatives were effective homogeneous catalysts
in hydrogenation for a variety of olefin substrates.

3 was the first to point out

Four years later, Horner
that modified Wilkinson catalysts which contained chiral
phosphine ligands should effect the asymmetric hydrogena-
tion of prochiral olefins. At the same time (1968) two

4 and Knowles5 independently confirmed that

groups, Horner
prediction: Horner and co-workers prepared a rhodium com-
plex in situ from the precursor [Rh(1,5-hexadiene)C1]2

and (§)-(+)-methy1-phenyl-n-propylphosphine4 (l) (69%
optical purity) to hydrogenate a-ethyl styrene and a-
methoxystyrene to (S)-(+)-2-phenylbutane (7-8% optical
purity) and (R)-(+)-l-methoxy1-1-phenylethane (3-4% opti-

cal purity) (Table l).

Me
P-Ph

n-Pr

(1)

‘V

Table 1. Asymmetric Hydrogenation of Prochiral Olefins
Using a Rh(I)-PMePh-n-Pr (1) Catalyst Precursor.

 

 

Substrate Optical yield (%)
c2115 7-8% (§)
C H
6 5
ocu3 3-4% (5)
::::§\
c6H5

 

Knowles' group used a similar catalystSto hydrogenate

a-phenylacrylic acid to (§)-hydratropic acid (Scheme 3).

C H CH

Céfisx. Hz 6 s\\. ,
. _ c"
Hooc’ Lg) -MePrPhP/Rh(I) HOOC / \ H

/

3

 

(61-15% e.e.

Scheme 3

6

Also, the hydrogenation rates and optical yields were
increased by using salts of the olefinic acids. This

was rationalized by more effective coordination of the
carboxylate. In 1971, Morrison and co-workers synthesized
a chiral neomenthyldiphenylphosphine, (+)-NMDPP6 (3),

from (-) menthol.

CH3/lll r '-i

\ PPh

2

NMDPP
(3)

With Rh(I)-NMDPP (2) as catalyst the optical yield was greatly

improved (Scheme 4)

C H

 

CH coon .6 5
3\ / H ,
/—\ 2 ; H—(f—cnzcoon
c6115 H NMDPP(3)/Rh(I) CH3

(§)-61% e.e.

Scheme 4

At the same time, Kagan and co-workers had synthesized
a bidentate diphosphine ligand, (-)DIOP (3), from tartar-
ic acid. A (-)DIOP(§)NUNI)catalyst precurser was obtained
in situ from [Rh(cyclo-octene)2Cl]2 and (-)DIOP (3) in

benzene-ethanol. The optical yields obtained from aswmnetric

hydrogenations of prochiral olefins are shown in Table 2.

 

P(C6H5)2

P(06H5)2

(-)-DIOP (3)

Table 2. Asymmetric Hydrogenation of Prochiral Olefins
with Rh/(-)DIOP (3) Catalyst Precursor.

 

 

Substrate Optical yield (%)
COOH
____///
63 (SJ
c6”5
,/COOCH3
____2 7 (R)
C H ~
6 5
H COOH
>_—_\‘/ 72 (B)
C H NHCOCH
6 5 3
H \. / COOH
.//—_—.\\\ 58 (R)
C6H5 ‘NHCOCH2C6H5

 

High optical yields observed using this catalyst were
accreditted to conformational rigidity of the chelating
diphosphine and the functionalized groups of the substrates
NMDPP (2) and (-)DIOP(3), were demonstrated to be effective
ligands, however, for both ligands the chirality was not
at the phosphorous atom. This new concept was a big break-
through for the design of new chiral phosphine ligands.
Another major development was the design of a new
chiral phosphine with a different synthetic path. A chiral
phosphine, ACMP8 (4) containing methyl, cyclohexyl, and
O-anisyl groups (95% optically pure) proved to be an ex-
cellent ligand for asymmetric hydrogenation of (;)-dehydro-

amino acids (Table 3).

:23
Me.

ACMP (4)

In this case, almost complete stereospecific hydrogena-,
tion could be achieved, since the oxygen atoms of thec7-
anisyl groups of the ligands could coordinate partially
with rhodium, resulting in the reduction of the conforma-
tion flexibility and the fixation of the chirality of

the complex. This successful development led Monsanto

to adopt the process for the manufacture of L-OOMA (a

9

drug used for the treatment of Parkinson's disease).

Table 3. Asymmetric Hydrogenation of (;)-Dehydroamino
Acids with Rh(I)-ACMP (4) Catalyst Precursor

 

 

Substrate Solvent Optical Yield (%)
”.c .,c00H 95% Ethanol 35
<E§§/___\\\NHCOCH3
“-c ,rcoon 95% Ethanol 85
<::;/——_\\\NHCOC6H5
'1 \ /.coou 'Isopropanol 88
NHCOCH3
OAc OCH3
H, / COOH Isopropanol 90
NHCOC6H5
OH ocn3

 

Since 1972 asymmetric hydrogenation has become one of the
most exciting research areas in chemistry. Many new chiral
phosphine ligands for asymmetric hydrogenation have been
SYnthesized and studied. Their catalytic applications for
other asymmetric reactions; hydroformylation, hydrosikfla-

tion, and alkylation also have begun to be explored.

10

II.‘ Chiral Diphosphine Ligands and Their Application to

Asymmetric Hydrogenation

A. Rigid Ligands

The optical yields from asymmetric hydroge-
nation of (g)-dehydroamino acids are extremely high when
chiral diphosphines form rigid five-membered chelating
rings.

1. Ligands Chiral at Carbon Backbone

The substituent at the carbon backbone of
chiral diphosphine can restrain the chirality of the
metal chelating ring and fix the orientation of the phenyl
rings on the phosphorous atoms. High stereoselection can
be achieved. The first highly effective ligand in this
category, (§,§)-Chiraphos (6), was initially developed by

9 The ligand was prepared from commercially

Bosnich.
available (R,R)-2,3-butanediol, by a simple synthetic
process (Scheme 5). With (g)-dehydroamino acids as sub-
strates, the optical yields were exceptionally high and
not very sensitive to the solvent. Leucine and phenyl-
alanine were obtained in optically pure forms. Bosnich
also developed (R)-Prophos10 (6) which was prepared from)
(§)-lactic acid (Scheme 6). This ligand only has a single
methyl group at a chiral center to constrain the chirality
of the chelate ring. The Rh(I)-(R)-Prophos (6) complex

is an efficient hydrogenation catalyst. High optical

CH

 

3 CH
H\ / 3 , mg /CH3
7C ”—' c\“‘“ —>‘ H_c——.c---H
0" OH OTs 015
CH /CH3
ii-iv> H--—C-—- c......H
(C6H5)2P P(C6H5)2

(§n§)-Chiraph05 (2)
(iii) Ni+2/Ncs'

Reagents: (i) TsCl/PY (ii) LiPth
(iv) CN
Scheme 5
CH
3 6:3
H“‘“C'—-*COOH ‘ . Hlf-i'C-——-CH2-OH
H //’ .
HO
(C6H5)2P
—ii:l—-‘ H 1V);‘——'CH2-P(C6H5)2
CH

3
(R)-Prophos (6)

Reagents: (i) LiAlH4 (ii) TsCl/PY (iii) LiPPh

2
(iv) Ni+2/NCS' (v) cn'

Scheme 6

12

yields were achieved for a variety of (;)-dehydroamino
acids (around 90% e.e). With Rh(I)-(R¥Prophos catalyst
the optical yields appear to be insensitive to the nature
of the substituent on the substrate. In comparison with
the Rh(I)-(§,§)-Chiraphos (5) catalyst, the optical yields
are rather substituent dependent. Moreover, the catalyst
is capable of breeding its own chirality, since a small

amount of (R)-Prophos (6) can produce large quantities

of itself (Scheme 7).

OAc i
:::i<;——-C02Et
ii-iv

 

(6)-Prophos

Reagents (i) Rh(I)-(R)-Proph05/ H2 (ii) LiAlH4

(iii) TsCl (iv) LiPPh2

Scheme 7

13

The product, lactate, can be converted into (§)-Prophos

easily by known procedures.

The (§,§)-Chiraphos (6) and (R)-Prophos (6) have
proved to be highly effective rigid ligands. ‘(R)-Prophos
(6), which has nonequivalent phosphorous atoms, provides
better structural information in the study of the reaction

mechanism by 31P NMR. These advantages led the other in-

vestigators to develop the new analogue ligands.
(§)-Phenylbis(diphenylphosphineo)ethanen (2) was
synthesized from (§)-mandelic acid by adopting the same

process10 which Bosnich used for the synthesis of (R)-

Prophos.
Ph

\

PPh
2 PPh2
(1)

The higher boiling point of the ligand allows easy cry-
stallization without tedious purification processing.
'The in situ Rh(I) catalygt was less active than the

Cationic Rh(I) catalyst and required high hydrogenation
The ligand (7) is not as efficient
It has

Pr‘essure for reaction.
‘35 (R)-Prophos (6) for asymmetric hydrogenation.

been successfully used for studying the reaction inter-

mediates in asymmetric hydrogenation through 31P NMR by

14

Brown.12

(R)-Cycphos (8) was prepared from (§)-(+)-mandelic
acid by Riiey‘3. The optical yields of (s)-a-amino acid
derivaties are generally above 90%. The synthetic scheme

is shown here (Scheme 8).

O
1

OH . _ H

i 1

1 i 1
.—i_COZH -——-- O‘f—COZH

H H

 

ii-iv I
>— i-——-CH2PPh2

(3)-Cycphos (g)

Reagents (i) Hz/Rh-A1203 (ii) LiAlH4/THF

(iii) .TsC1/PY (iv) 10% Excess Liph2

Scheme 8

 

15

The author rationalized that the high optical yields
were associated with the bulky cyclohexyl substituent
which could restrain the rigid chelate ring more effec-
tively than other structurally analogous ligands.

A series of new rigid ligands which have bulky chiral
carbon skeletons were prepared by Diels-Alder reactions
between the chiral dienes and P-C=C-P moieties such as
(41101112110514 (2), (s.§)-Phe11anphos15 (10) and (5,5)-
Nopaphos15 (Ll).

(C6H5)2P P(C6H5)2
P“36%)2 I
/b
c H -i
p(c6115)2 3 3 7
(-)NORPHOS (g) (s,§)-Phe11anphos (19)
P(C6H5)2

E ‘\ 1>(06115)2

(3,31-Nopaphos (1})

This is an alternative way of synthesizing the rigid type
ligands instead of using conventional methods, such as
the substitution of tosyl groups by diphenylphosphide.
Best results were found in asymmetric synthesis of a-

amino acid derivatives (80-95% e.e.)”-15

16

2. Ligand Chirality at Phosphorus
The highly efficient DIPAMP (12) was developed by

Knowles and his co-workers.16

Since the methyoxy groups
of ligand can partially bond with rhodium metal and fix
the phenyl rings' chirality. excellent stereoselection

can be achieved.

H

6 5

C
(: funCHz)2
I OCH3

DIPAMP (12)

High optical yields of up to 96% were obtained in the
hydrogenation of dehydroamino acids. These high optical
yields were not markedly sensitive to temperature and
pressure. Also, the ligand was effective for nonamide
substrates. Excellent selectivity (approximately 90% e.e)
was produced in reduction of the a-enol ester, (g)-ethy1-

2-acetyloxy-3-phenyl-2-propenoate (l3).

17

H . / COOEt

//’—_—“\\

CGH5 OCOCH

3

(13)

The ligand was synthesized by using a copper reagent to
induce the coupling reaction of chiral monophosphine oxide,

then it was reduced by HSiCl3 reagent (Scheme 9).

 

ll . . 11 ll
MePhArP l _ PhArPCH CH PArPh
* r ‘k 2 2*
fGHS
ii (: 1‘3---CH2———)2

DIDAMP (13)
Reagents (i) (i-C3H7)2NLi/CuC12 (ii) HSiCl3
Scheme 9

The drawback of the synthetic process is that it involves

long synthetic processing and the overall yield is low.

18

B. DIOP and DIOP Analogues

The DIOP (3) which was first synthesized by Kagan7

played a historically key role in the development of chiral
diphosphines for asymmetric hydrogenation. Also, the
DIOP (3) has proved to be useful for a variety of asym-

metric catalytic reactions.2b

Encouraging results obtained
by using DIOP (3)/Rh(I) catalysts led to the development

of a great number of new analogues, either for the eluci-
dation of the mechanism or for the improvement of optical
yields. The new DIOP analogues‘n’18 (14-22) were synthe-
sized either by varying the acetal substituents, replacing
the acetonide ring by a carbocycle, or by changing the
aromatic substituents. Modified acetal analogues (14,15)

did not affect the optical yields, since these groups

caused no perturbation on the chirality of the phenyl

rings.
H H
I .'
H3cj><:0 Fhrz Ré><0 ' P(C6H5)2
H C 0 1%r2 O P(C6H5)2
3 H H
Ar - c H (DIOP) R1 R2 - -(CH ) -
6 5 g 1 2 5

(16) Ar =

Q

$01.13
CH3
(16) Ar =:-€;:::;
3

(1]) Ar

CH

(131 (Ar)2 =

CH2-

CHZ-P(C6H5)2

19

1_ 2_
1‘
P(c6H5)2
P(C6H5)2
H
(23)

(CGHS)2P‘CH2

P(C H )
6 5 2
CHZ‘P(C6H5)2

<23)

(20)

~

The DIOP analogues, (16, 18, 19) which have different

substituents on the aromatic rings, generally give low

optical yields.

optical yield of DOPA was enhanced (90% e.e).

comparison with DIOP (g) with carbocylic analogues

Only ligand (1]) was exceptional.

The

In
18

(20, 2], 22), cyclobutane analogue (20) led to better

20

optical yields. Since the ligand (20) has a small trans-
fused carbocylic ring, it is more conformational rigid

and results in better asymmetric induction (Table 4).

Table 4. Asymmetric Hydrogenation of Dehydroamino Acids
Using DIOP (g) and Carbocylic Analogues18

 

Substrate ———— Optical Yield (%)
(,3) (29) (2,1) (23)

 

 

H coon '
\————/
82(R) 86(R) 63(R) 35(5)
06H5//‘ \NHcocn3 ” ~ ‘ 7
/’COOH
73(R) 72(R) 72 R 40
-___—\\‘NCHCOCH3 ~ ~ (~) (£)

 

C. Pyrrolidinephosphines

The (as, 4§)-N-butoxycarbonyl-4-dipheny1phosphino-
2-diphenylphosphinomethylpyrrolidine (BPPM)19 (26) was
first developed by K. Achiwa. The BPPM (26) has two kinds
of function groups, two phosphines as ligand for metal
complexion, and the amide group which is capable of being
modified by different substituents in order to optimize
optical yields. A series of pyrrolidinephosphines have

been synthesized by variation of the amide substituents

21

and have been successfully applied for a variety of

2b Interactions between

catalytic asymmetric reactions.
the amide group and substrate were crucial for highly asym-

metric inductions of some reactions.

(C6”5)2P
T (25) R = 11 PPM
R

The BPPM ligand (26) is effective for asymmetric

‘9 High optical yields were obtained with

hydrogenation.
(2)-dehydroamino acids by adding triethylamine (Table 5).
Rh(I)-BPPM catalyzed the hydrogenation of a-keto-
esters20 in dry benzene or THF (Scheme 10). High optical
yields of lactates were achieved in 65-75% enantiomeric
excess. The catalyst was more stereoselective than Rh(I)-

DIOP (6), which gave low optical yields (32-42%).

0 OH
I RhCl diphos* l
H C———C-——COOR e: H C-——CH———CO0R
3 H 3 *
2

 

diphos* = BPPM (23) or DIOP (3)

Scheme 10

22

Table 5. Asymmetric Hydrogenation of (1)-Dehydroamigo
Acids with In-situ Rh(I)-BPPM (23) Complex
at 20°C and 50 atm. ”

 

 

 

 

Substrate Solvent Optical Yield (%)
without NEt3 with NEt3
H\__/COOH MeOH 30 (B) 83 (B)
<::> ‘NHCOCH3 EtOH-HZO 2 (R) 48 (R)
- (2:1)
H~\ ,COOH EtOH 7 (s) 15 (B).
<::§_—\\NHCOCH3
”Vcoon EtOH 32 (g) 83 (g)
zO-§::§f—_\\NHCOCH3
CH2
\
0
H\__/coon EtOH — as (B)
NHCOCH3
OAc OCH3
”\___,zcoon EtOH —— 87 (R)
NHCOCH3
Ac

 

(a): In—Situ: O.5[Rh(l,5-hexadiene)C1]2 and BPPM (26)
as catalyst precursor _

23

The In-Situ Rh(I) catalysts which were prepared from
[Rh(diene)Cl]2 and PPM (26) or the analogues (26, 24)
were used for asymmetric synthesis of (R)-(-)-pantolactone2‘

(26) (Scheme 11).

0 CH CH

 

16 013 HO * 3
/, 3 26, 24, or 26 + H2 - CH3
0:;3\\0 , [Rh(diene)Cl]2 O;:\\O
(25)
Scheme 11

The hydrogenations were run at 30°-50°C and 50 atm hydro-
gen pressure. The best optical yield (87%) was obtained
by using Rh(I)-BPPM (26) catalyst and was run under op-
timum conditions. (Table 6). When N-substituents of

PPM (26) were varied, the optical yields were greatly
reduced. In the case of ligand PPM (26), even the ab-

solute configuration of pantolactone was reversed.

24

 

 

 

Table 6. Optical Yields of (R)-(-)-Pantolactone2] (26)
by Using PPM (26) and its Analogues T
Solvent Temperature ——— Optical Yields (%)
(23) (2,4) (15)

Benzene 30°C 86.7 (R)

50°C 84.8 (R) 59.2 (B) 15.4 (s)
Ethanol 50°C 32.1 (R) 35.9 (R) 8.5 (5)
THF 30°C 80.7 (R)
Chlorobenzene 50°C 63.5 (R)
Toluene 50°C 77.7 (R)

 

D. Ferrocenylphosphines

Ferrocenylphosphines which have planar chiraltiy and
various functional groups such as an amino or hydroxyl
group have been demonstrated to be effective ligands

(2]-29) for catalytic asymmetric reactions.2b

  

Pth

* (6)-(R)-BPPFA (27)
CHMeNMe2 ~

25

 

P(C
P(C6H

 

6H5)2
5)2

(Bl-(Sl-BPPFA (Z?) (3)-(§)-BPPF0H (29)

The (6)-(R1-BPPFA ligand (2]) is useful for asymmetric
hydrogenation.22 The catalyst, prepared in situ from
[Rh(l,5-hexadiene)Cl]2 and (6)-(R1-BPPFA (2]), has been
used to catalyze asymmetric hydrogenation of the (2)-
dehydroamino acids. The catalyst gave high optical yields
in an aqueous solution (Table 7). Ammonium-carboxylate
interactions between the amino group of (6)-(R)-BPPFA (2])
and the carboxyl group of substrate are crucial for high
asymmetric induction. By adding triethylamine the effect

of the interactions will be reduced.

26

Table 7. A54m?

( §ric Hydrogenation Catalyzed by Rh(I)-
s -

e
R -BPPFA (2]) Complex

 

 

Substrate Solvent OpticalYield (%)
5\ //COOH MeOH 93 (5)
NHCOCH3 HZO/EtOH (1/1) 92 (5)
HZO/MeOH (1/1) 89 (5)
H\\____,/C00H MeOH 8 (5)
NHCOCH3 Eto“ 33 (§)
H20/Me0H (1/3) 87 (5)
OAc
”\\ ,-CO0H EtOH 36 (5)
NHCOCH3 H20/MeOH (1/2) 86 (5
OAc OMe
H. .zCOOH HZO/MeOH (3/4) 52 (5)

-f;:§r___‘\\NHc0CH3
o
(‘0

A series of ketones and o-keto acids were reduced
by Rh(I)-(R)-(5)-BPPFA (26) and Rh(I)-(R)-(§)-BPPF0H (29)
cataiysts23 (Table 8). The high optical yields obtained

 

with (R)-(§)-BPPFOH (29) can probably be ascribed to
hydrogen bonding between the carbonyl of the substrate

and the hydroxyl groups of the ligand, which may increase

27

Table 8. Asymmetric Hydrogenation of Carbonyl Compounds
Rl- co- R2 with (R)- (5)- BPPFOH (29) or (R) (S)-

 

 

 

 

BPPFA (28)

Substrate _ Optical Yields (%)

Cationic Precursor In Situ Catalyst

(29) (2,33) (23)
H3C-C0-C6H5 40 (R) 15 (6) 35 (R)
H3C - C0-C4H9-t 43 (3) -- —“
H3C-C0-C00H 59 (R) 16 (6) 55 (R)
H3C-C0-C00H 83 (R)C -—» -—

 

(a) Cationic precursor: [Rh(COD)(dipho *)]C10
diphos*: (R)-(§)-BPPFOH (29) o3 (5)-(§)-
BPPFA (28)
(b) In Situ catalyst : 0.5 [Rh(1,5- hexadiene)Cl]2 and
(R)- -(§)-BPPFOH (29)

(c) By adding 1 equivalent of triethylamine

conformational rigidity in diasteromeric transitions re-
sulting in high stereoselectivity.

The (R)-(§)-BPPFOH-(26)-rhodium(I) complex also cata-
lyzed hydrogenation of aminomethy1.my1 ketones to give

corresponding 2-amino-l-ary1ethanols (Scheme 12).

28

 

1
R -CH2NHR ~HC1 g;

I
Rg;i:>>-C*-CH2NHR3-HC1
H

Scheme 12

High Optical and chemical yields were obtained24 (Table 9).
Conventional chiral hydride agents are not very efficient
with these prochiral carbonyl compounds because of the
presence of active hydrogen on the starting ketones and
the instability of the latter under base conditions. The
highest optical yield (95%) was achieved with dihydroxy-
phenyl methylaminomethyl ketone in comparison with other

Prochiral carbonyl derivatives.1a’ 25

29

Table 9. Asymmetric Hydrogenation of Aminomethyl Aryl
Ketones with (5)-(§)-BPPFOH (29)

 

 

 

Substrate Optical Yield (%)
1 Cationic precursora In Situ catalystb
R C-CHZNHR3-HC1 (2g) (29)
R1 R2 R3
OMe OMe H 90 (R) 92 (g)
H H H 60 (B) 57C(§)
H OH H 69 (B) -.

3 95 (B) '-

 

(a) Cationic precursor: [Rh(g)-(§)-BPPFOH-(§9) (NBD)]C104

(b) In situ catalyst : (g)-(§)-BPPFOH (a9) + 0.5 [Rh(l,5-
hexadiene)Cl]2

(c) Using enantiomeric (§)-(fi)-BPPFOH

E. Atropisomeric Ligands

Recently, a series of atropisomeric ligands, (19, 11,
3?), whose chirality was a result of atropisomerism due
to the presence of a chiral axis and not to the asymmetric
center on phosphorus or carbon, have been synthesized
and used for asymmetric hydrogenation of dehydroamino
acids. The (§)-(-)-NAPHos26 (99) and (§)-(-)-NAPHIN27 (91),
which were highly flexible ligands, generally gave low

optical yields.

30

 

(§)-(-)-NAPHOS (30) ' (§)-(-)-NAPHIN (2])

 

 

(§)-BIHAP (32)

The ligand (§)-BIHAP28 (32) is more rigid than the two
ligands (S9,Sl) mentioned above. High optical yields

were achieved under optimizing conditions (BO-100% with

low substrate concentrations and low pressure). The ligand
also was very effective for hydrogenation of configuration-
ally labile (§)-a-benzamidocinnamic acid. The optical

yield (87% e.e.) was among the highest ever reported.

F. Bis-aminophosphines

A series of bis-aminophosphines (33-39) were synthe-

Sized by Giongo.29 Rhodium(I) complexes with the chiral

31

chelating bis-aminophosphines are efficient catalysts for
the asymmetric hydrogenation of a-acetamidoacrylic acid
and (;)-e- acetamidocinnamic acid (Scheme 13). The optical

yields were around 70-94% with these bis-aminophosphines.

 

H \e ,/ COOH [Rh(COD)(PNNP)]ClO4
A 1*
R - NHCOCH3
Solvent: EtOH
R = CGHB or H
*
RCHZCHCOOH
NHCOCH3
70-94% e.e
Scheme 13
(CGHS)ZP
I
N

 

(33) (34)

32

 

(35)

(32) R=H
(:18) SR=CH3

(16)

f(°6”5)2
0 N
I.)
O ‘
P(Chi'sh

(39)

~

(13, 23)-Bis(diphenylphosphinamino)cyclohexane[(g,§)-

{9]30a, its enantiomer30a (§,§)-(§J) and its N,1Wdimethyl

derivative ligand30b.(Rfig)-(fig) were effective ligands

for asymmetric hydrogenation of (Z)-a-dehydroamino acid

derivatives30 (Table 10).

33

H - H
| 1
.ul
N"P(ces'l‘sM "N‘P(c5“5)2
"”T‘P(c6”5)2 T’P(c6”5)2
H H
(5.5)410) (gasp-(41)
CH3
N'P(C6”5)2

”MIN-P(CGH5)2

l

(3,E)-(€‘)

Inversion in the steranelectivity of asymmetric hydrogena-
tion of (g)-a-acetamidocinnamic acid and (Z)-a-benzamido-
cinnamic acid was found by using its m.N~dimethyl deriva-

tive 1igand3°b(R,R)-(qg).

34

Amwv-am.m

nee ~H_95ouvem m.o

hmwe-~m.w. eo hmwv-fim.mp,u emeseee . o_umemo;eeefioou egg

somgaumeq upcowumo

umzpmumo seem 2H Maw
a

 

mIcoouzzx, .xmzou
“we on ._"_v e:eu-=oem «Izouix xxee
“my om Amy we 5? .V m:eu-=oeu zeuouzz / \mzmu
Amy mm Amy me Amv me zoom =oou.\\iii;// I
mzuouzzx/(iii\\m=eo
va No Amy Na A. Fe exeo-=oem NIzou\\i= // I
Amy Ne Amv as A. PV ezeo-=oem mzuouzz,, mzeo
“my me Amy Fe Ame _e. zoom zoou\\ Ix:
Amwv-fimnwv AH«V-Amemv AQ«V-AMemV
aumxpopmu seem aw ugomgauoea uvcovumo
Auv ape?» _uuwuno ucm>~om mumgumnzm

 

 

 

mm>_ua>pema u—u< ospEmocuaguoiAmv we

cowamcomocua: uyguossam<

.op mpnmh

35

G. Polymeric Chiral Diphosphines

Asymmetric hydrogenation with immobilized homogeneous
catalysts have been studied for several years, but the op-

tical yields achieved from the initial attempts were rather

31’ 32 For example, in 1973, Kagan and his co-workers

32

low.
reported that a DIOP type ligand, 2,3-0-isopropylidene-
2,3-dihydroxy-l,4-bis(diphenylphosphino)butane (2}),
attached to cross-linked polystyrene, reacted with solu-

ble rhodium(I) complexes to give asymmetric hydrogenation.

CH '
/ 2\CH/

(43)

However, hydrogenation was possible only in benzene
and the optical yields were only one-tenth of those found
\Nith the homogeneous catalyst. In polar solutions of the

substrates, the beads collapsed and lost their catalytic

36

33 and his co-

activity. To overcome this problem, Stille
workers.successfully prepared a Rh(I)-DIOP ({5) catalyst

on a polar cross-linked copolymer of 2-hydroxyethyl-
methacrylate and p-styryl (DIOP) with a 92:8 mole ratio.
The preparation of this immobilized catalyst and asymmetric

hydrogenation results are shown in Scheme 14.

CH2 CH3
\ CHI2\I [RhCl (CZH “2142-. Catalyst
{iii . 0.92
0.08 C02 CH 2CH 20H
0 0

 

Ph P PPh

 

2 2
(43)
*
RCH CHCOOH

H \ , COOH H2 _> 2
R”/f___\\‘ NHCOCH3 catalyst NHCOCH3
R = Ph 86% e.e
R = H 52-60% e.e.

Scheme 14

37

Since this copolymer was essentially a hydroxyethyl
methacrylate gel, it could swell in polar solvents. a-
acetamidoacrylic acid and (£)-d-acetamidocinnamic acid
were hydrogenated in ethanol with this immobilized catalyst.
The optical yields of both products were similar to those
obtained in experiments with the homogeneous catalyst.

The immobilized catalyst could be re-used, but repeated
exposure to air during the filtering step led to low
optical yields.

Achiwa used a similar method to prepare new polymer
supported cationic and neutral chiral pyrrolidinephosphine
rhodium complexes. 4VB-PPM-HEMA34 (1]), which was pre-
pared by the copolymerization of 4-VB-PPM (Q5) with 2-
hydroxy-ethyl methacrylate at a l:24 mole ration, was
allowed to react with rhodium(I) precursors to afford
the new polymer supported catalysts (Scheme 15). Two
substrates, itaconic acid and (g)-a-acetamidocinnamic
acid were hydrogenated with these supported Rh(I)-4-
VB-PPM-HEMA (1]) catalysts. The optical yields with
polymer supported Rh(I)-4-VB-PPM-HEMA (fij) catalysts
depended on the conditions employed and whether triethyl-
amine was added or not. The homogeneous analogous
BZPPM (QC) -rhodium (I) complexes gave almost the same
optical yields (Table 11). Also, the polymer supported
cationic catalyst was more effective than the supported
neutral catalyst. In general, the supported Rh(I)-
4-VB-PPM-HEMA (Q?) catalysts gave lower optical yields

38

than those of their homogeneous analogues.

 

RPh2 \\ \\
) PPh2
H coc1 C0-N
CHZPPhZ
PPM (2s) 4-VB-Cl 4-VB-PPM (45)
CH3
CH CH I CH2
2 ,,,«”’ 2 /’
H CH C
CH3-C-C00CH2CH20H C=o
filial m I n
(”EMA) .. OCH CH CH
' PPh 2 2
CHZPth

4VB-PPM-HEMA (47) (m/n=l/24)

Scheme 15

~PPh2
O —-co-N

BZPPM (g9)

39

AMWV IIII-III-I>I ea Amwv IaaNI III ~H_UAIIQVIIH m.o “
ameIa_I eo_umfiaaaII_IVAooovIIH "tameaeeea e.eoeeao Aav

AmVIIIIIIeIZ .3 Am: :32

emspaeau seem ea any

 

 

 

III e.- ii Ame o.me .11
ii 11 Amy e.e“ .11.
.ii ii Awe e.e“ .11

“my m.oe :1 Amy m.~m ii

Amy o.m .1. Amy m.m~ ii
in Amy e.om .1- Amy I.Im
i: va m.mm .1: Amy m.~m
ANIV AmIV AMIV . Amwv

coax—on maoucmmoso; Losxpoa maomcomoso;
numapnueo :9»% ea meomezumem u_:o_umu

III mu_a.> _ae_eao

 

umpuaumm

Amemzv
FeeeIeI

vmpuxuom

Amemzv
PeeaIem

.eeagem

Amemzv
PoeeIeaI

_eeaIeaI

uca>~om

Inca

N

I ooqu

muogumnam

 

mozmopoc< maoocomoso:

ecu mumzpauau Amwv <zm=izaaim>e-AHvsm copeoaaam goeapoa gu_3 vpu<
uvsmccwuovvsuumu<iaiANv use n.0< owcououa we copumcmmoeu»: uvgumssam< .Fp «Fan»

40

Copolymer supported Rh(I)-PPM (25) catalysts also

35 The yields obtained

were synthesized by Stille and Baker.
from asymmetric hydrogenation of dehydroamino acids with

copolymer supported catalysts were the same as those of the
corresponding homogeneous results. The polymers containing
optically active alcohols and chiral phosphines were also

applied for asymmetric hydrogenation of dehydroamino acids.
The results indicated that the solvent-polymer interactions

dominated the effects of the additional chiral center.35

III. Asymmetric Hydrogenation Mechanism Study

Asymmetric hydrogenation occurs as a result of the
Chiral rhodium complex reacting with the substrate to form
diastereomeric transition states, which differ in energy.
AAGf determines the excess of one enantiomerower the other.
The chiral diphosphine ligands play a key role in catalytic
asymmetric hydrogenation with rhodium ions. The efficiency
of the enantioselectivity is dependent upon on how well
the chiral diphosphine ligand fits the substrate and how
capably the substrate chelates the rhodium metal from a
rigid complex in the transition states. Also, the solva-
tion effect is very important since it can maximize the

21.22

optical yields of the products. In the polar solvent,

solvent may form hydrogen bonds with functional groups

22

of substrates and ligands. The energy of such type

bonding is comparatively small, (2-5kal/mol); but it is

41

enough to alter AAGf and effect the optical yields.
A. Enamide Bonding Structure

Usually asymmetric hydrogenations of dehydroamino acids
(enamides) with rhodium(I) chiral diphosphine complexes
give high optical yields in comparison with other types of
substrates. The enamides are capable of coordinating with
rhodium via olefin and amide carbonyl residues to form
rigid transition states, leading to higher asymmetric in-
duction. The binding mode was first inferred from asym-

36

metric hydrogenation data by Kagan. The X-ray structure

evidence was found by Halpern and co-workers.37 The crystal
structure of [Rh(DIPHOS)(methyl-(Z)-a-acetamidocinnamate)]+ 37b
showed bonding by the olefin and bonding by amide carbonyl
in a square-planar complex (Figure 1). Also, Brown38 and
ChaJoner have proved that the same structure was maintained

31

in solution by the evidences of P and 13C NMR spectra.

8. Proposed Mechanisms and Asymmetric Induction Steps

Two kinds of mechanisms have been proposed for asym-
metric hydrogenation.
l. Mechanism I "unsaturate route". In this mechanism,
complexation of olefin proceeds the addition of hydrogen.

39 and co-workers for support

Evidence was found by Halpern
of the "unsaturate route" mechanism in the catalysis of
hydrogenation by rhodium chelate diphosphine complexes.

The catalytic cycle based upon his work is shown in Figure 2.

42

MINUS)

 

Figure 1. X-ray structure of [Rh(DIRH05)(MAC)]+

MAC:
/——_\
CGHS NHCOCH3

DIPHOS: 1,2-Bis(diphenylphosphino)ethane

The catalyst precursor is usually a cationic [Rh
(diene)diphos*]+ complex, where diphos* is any che-
lating chiral or achiral diphosphine. The complex
absorbs two moles of hydrogen to yield the highly active
unsaturated complex, [Rh(diphos*)J+. The olefin will
displace solvent from this complex tofonn a [Rh(diphos*)
olefin]+ complex then react with hydrogen, presumably
to form the dihydride-olefin intermediate. Following
the hydride transfer from the mean to coordinated
face of the olefin it forms a rhodium alkyl hydride
intermediate, which undergoes reductive elimination

to produce the hydrogenated product (RH) and

43

[Rh(diphos*)(diene)]+

[Rh(diphos*)]+

alkane olefin

. +
[RhH(d1phos*)alkyl] [Rh(diphos*)olefin]+

/"2

[RhH (diphos*)o1efin]+
2 ,

Figure 2. Hydrogenation catalytic cycle by unsaturate

route

44

regenerates the catalyst.
Catalytic hydrogenation is normally conducted with
100-200 fold excess of enamide substrate over catalyst.

3M. Usually

The catalyst concentration is lower than 10'
the rhodium(I) chiral diphosphine enamide complexes are
present at steady-rate equilibrium. Two diastereomeric
rhodium(I) enamide complexes can be observed with one
predominating in most cases. Sometimes, however, the

major diastereomeric Rh enamide complex does not correspond
with the preferred final product.

The asymmetric induction step can be either the dis-
placement of the solvent by enamide, or the addition of
hydrogen or ate-ligand migration, depending on the indivi-
dual case. The type of chiral phosphine, temperature and
pressure, are all factors which must be considered in the
determination of the asymmetric induction step.

In the case of L§,§)-Chirophos (5) system, Halpern, et al-
reported that asymmetric induction occurred with the addi-

40 The minor

tion of the hydrogen step at low pressure.
diastereomer Rh(I)-enamide complex (X-ray structure evidence)
wasthe true intermediate which addmdhydrogen much faster

than the major intermediate. With increasing pressure,

the asymmetric induction step shifted to the initial cata-
lyst-olefin adduct formation step. This led to lower

enantioselectivity, even with the possibility to reverse

the absolute configuration of the product.

45

with UIPAMP (12), Brown38 reported that the optical
yields observed by Knowles in hydrogenation of methyl-
(g)-a-acetamidocinnamate were higher than the diastereo-
selectivity in binding. This implies that the more stable
rhodium(I) enamide complex was hydrogenated faster than
the minor diastereomer. Moreover, the hydrogenation of
methyl-(Z)-o-acetamidocinnamate was reported to be pressure
dependent. At high pressure (27 atm) the optical yield
was reduced to 78%, the enantiomer ratio of 89:19 corre-
sponding closely with a diastereomer ratio of 91:9 which

31P NMR. This may imply that at very high

was observed by
pressures the initial bonding of the prochiral olefinic

substrate to the catalyst is the asymmetric induction step.

2. Mechanism II, "hydride route“ (favorable under high
hydrogen pressure). In this pathway, the dihydride Rh
complex forms preceeding the olefin coordination with

Rh metal.

This alternative mechanism was proposed by Ojima.41
Mechanism II has been widely accepted for hydrogenation

of olefins catalyzed with (Ph3P)3RhCl.42 According to
Ojima, Mechanism I favorably operated at low hydrogen
pressure, whereas Mechansim II became predominant at higher
pressures in the case of DIOP (3) and BPPM (23). In both
Mechanism I and II, the formation of preferential Rh(I)
enamide chiral diphosphine diastereomers are the asymmetric

41

induction steps. The stereochemistries of [Rh diphos*]+

46

[Rh(diphos*)(diene)]+

 

 

 

. + +
[Rh(d1phos*)] [Rh(diphos*)H2]
A ‘________
a
olefin
(r
product ¢ [Rh(diphos*)H2 olefin]+

fast
(a) rds = rate determing step

Figure 3. Hydrogenation catalytic cycle by the

hydride route

and [Rh sziphos*]+ dictate the orientation of prochiral
olefin upon coordination, the latter leads to inversion

in the configuration of the product. Hith the-addition

of triethylamine Mechanism II is greatly suppressed, i.e.,
triethylamine prevents pre-coordination of H2 at high
pressures as well as improving optical yields. There
needs to be more mechanistic work done to clarify the
asymmetric induction steps in different chiral diphosphine

systems.

47

C. E‘Z Dehydroamino Acids Isomerization

The degree of E'Z isomerization during hydrogenation
of (§)-dehydroamino acids with rhodium(I) chiral diphos-
phine complexes has been studied by NMR spectroscopy 38’43
and deuteriation studies.44.(E)-dehydroamino acids do
not bind strongly to rhodium(I) and convert into the
corresponding (g)-enamide complexes, but stable complexes

are formed in the presence of triethylamine.38

IV. Catalyst Supports

The swelling mica-type silicates, better known as
smectite clay minerals, have layer lattice structures,
in which rigid two-dimensional silicate sheets approxi-
mately 9.6 A thick are separated by interlayers of

45 The structure of the oxygen frame-

hydrated cations.
work of the silicate sheets and the position of inter-
layer cations are illustrated in Figure 4. The swelling
layered silicates consist of one octahedral sheet sand-
wiched between two tetrahedral sheets. In the tetrahedral
sheet, tetravalent Si is sometimes partly replaced by
trivalent Al. In the octahedral sheet, AlIIImay also be

I II, FeII III

replaced by Li , Mg , etc. As a result

, and Fe
of isomorphos substitution, the silicate sheet acquires

a negative charge, and cations are incorporated in the
spaces between the layers (interlayers) to preserve
electroneutrality. In the presence of water, the cations

on the layer surface may easily be exchanged by other

48

 

 

 

 

251, o the UHlor F)

.oclohedrol colions
(Al in montmorillonite. Hg in
hectorite) ‘

MONTMORRILLONITE: "30.66 X-HZO [A13.34Mgo.66] (51.8.0) 020(LOH)4

HECTORITzE: Names X-HZO [Mgs.34Lio.66] (518.0) 020(0H, F)4

Figure 4. Structure of the mica-type swelling silicates
such as montmorillonite or hectorite

49

cations when available in aqueous solution; hence they
are called "exchangeable cations". The cation exchange
sites in the layer silicates are found mainly in the
interlamellar spaces (Figure 4). Additional exchange
sites may exist at the edges and corners of silicate
sheets, however, their contribution to the total cation
exchange capacity is usually small.

Montmorillonite and hectorite are the two most im-
portant smectite minerals. In montmorillonite, which
has an idealized anhydrous unit cell formula of Na0.66
[M90.66A13.34] ($18.0)020(0H)4, two thirds of the octa-
hedral positions are occupied by Mg+2 and Al+3 and the
rest are vacant. In hectorite, "30.66[Li0.66M95.34]
(518.0)020(0H’ F)4, the octahedral sites are filled by

+2 1

Mg and Li+ , and sometimes fluorine atoms are substituted

for hydrogen atoms in the hydroxyl groups.
The properties of layered silicates are:

l. The internal surface area, which is calculated from
the unit-cell weight, 769 g/mole, and unit-cell sur-

.2
face area, 5.25 X 9.18 A, is about 750 m2/g.
2. The cation exchange capacity is around 70 meq/lOO g.

3. Interlayer swelling depends on the nature of the
solvent and the interlayer cations and the charge

density on the silicate framework.

50

The large internal surface area and the relatively.
high cation exchange capacity (70 meq/lOO g, which
compares favorably with the ion exchange capacities
of resins), allow appreciable amounts of large cata-
lytic cationic complexes to be intercalated. The cati-
onic metal complexes can be electrostatically bonded
between the silicate sheets. The intercalated catalysts
can be sufficiently swollen in the polar solvents so
that the active sites are accessible to the substrate.
Also, the reaction products can easily diffuse through
the layered silicates and out into the solution. In
this manner they free a place for new portions of the
substrate. Also, the intercalated catalysts are mechan-
ically stable, and they cannot be crushed during the
normal reaction process.

Studies of intercalates of the mica-type layered
silicates, montmorillonite and hectorite are particu-
larily appropriate at this time, since there has been
a renewed interest in the use of such materials as
catalysts or catalyst supports for highly specific
chemical conversions. For example, highly thermally
stable pillared interlayered clays (PILC) have been
recently synthesized by the exchange of inorganic mono-

and poly-nuclear cations.46 '{e.g., A11304(0H);Z, Si

- + . .
(acac)g [Zr4(0H)]4(H20)10] 2 and B16(OH)];6} into the
swelling layered silicates. These PILC clays offer

interesting possibilities for industrial application

51

in fcracking and hydro-cracking processes.47

Also.

Pinnavaia and his co-workers have proven that the

swelling silicates are promising catalyst supports

for the intercalation of a variety of cationic catalysts.48"51
The intercalated catalysts could easily be isolated

from reaction mixtures and reused. Several other ad-
vantages have also been reported for layered silicate
systems. Among them are the enhancement of hydrogenation

” 48

activity of the immobilized Rh(PPh3)n catalysts and

greater positional selectivity in both hydrogenationag’50

and hydroformylation.51

V. Research Objective

My research objective is asymmetric hydrogenation
of prochiral olefins with rhodium(I) chiral diphosphine
catalysts. This research can be divided into two

parts.

I. Asymmetric hydrogenation of prochiral olefins with

layered silicate intercalation catalysts.

II. Synthesis and characterization of new functionalized

chiral diphosphine ligands for asymmetric hydrogenation.

In the first part of my research the focus was on
the intercalation of different types of cationic chiral
rhodium diphosphine complexes. Also, the effect of

intercalation on the optical yields as a function

52

of substrate size, interlayer swelling and chelate ring
size was studied. Optimization of enantioselectivity
and a recycling of the catalysts were investigated.

In the second part of my research the emphasis
was on the synthesis and characterization of new function-
alized chiral diphosphine ligands:sud1as aminodiphos
and hydroxydiphos ligands for asymmetric hydrogenation.
The functionalized ligands were synthesized from natural
occurring L-malic acid. The design of the new chiral
diphosphine ligands was carried out keeping the following

ideas in mind;

1. Ligands which are conformationally rigid are capable

of producing high optical yields.

2. A third function group of the ligand can be modified
in order to optimize the interactions between the third
function group and substrate; resulting in enhanced

asymmetric induction efficiency.

3. Ligands can be attached to polymeric, inorganic
or organic carriers to recycle the precious catalysts
and chiral ligands for practical use, and even improve

optical yields.

4. Since Chiral ligands have nonequivalent phosphorus.

theycan provide sensitive probes for elucidation

53

of the mechanism of asymmetric hydrogenation by 3]

P NMR.
The results of the new ligands synthesis are
encouraging, and these ligands are able to meet most

of the requirements set forth above.

RESULTS AND DISCUSSION

1. Asymmetric HydrogenationirfProchiral Olefins with
Layered Silicate Intercalation Catalysts

A. Diphosphine Ligands

The diphosphine ligands, DIOP(+) (3), SPIPHOS (48),
and (3)-Prophos (Q) and its anlogue, (R)-4—Me-Prophos (4]),

have been studied in this work.

PPh2
3 jx<: RRh2
H
DIOP(+) (g) SPIPHOS (4g)
CH2'---'C‘CH3 (3)-Prophos Ar=C6H5 (Q)
pArz PAr2 (3)-4-Me-Prophos Ar=4-CH3C6H4
(49‘)

~

DIOP (3) has proved to be a flexible ligand quite useful

for asymmetric synthesis.]c’2b
52

SPIPHOS (48) is also a
highly flexible ligand; it was obtained as a gift from

Dr. OeVries. (3)-Prophos (6) and its analogues have been

54

55

shown to be rigid types of ligands. They are also effective

for asymmetric hydrogenation of prochiral amino acids.9'15

(R)-Prophos (g) and new (R)-4-Me-Prophos (49) were

1° The modified

prepared by modifying Bosnich's method.
method differed from Bosnich's method in the final phosphina-
tion step. The purification of the ligands also was sim-
plified. The synthetic precursor (§)-(+)-l,2-propanediol
di-p-toluenesulfonate (ditosylate) was synthesized according

10

to Bosnich's report (Scheme:16).

 

CH3 . CH3
Hlu'C——COOH LiA1H4 Hliu ——CH2——OH
/ —- /
HO HO
CH3
TsCl/PY H|““C——CH OTs
I. / 2
7' TsO
Oitosylate
Scheme 16

The reduction of the L-lactic acid with lithium aluminum
hydride in dry THF produced the (§)-(+)-l,2-propanediol

in 85% yield. The (§)-(+)-l,2-propanediol was converted in
ditosylate in 95% yield by reactingitwiflitoluenesulfonyl.

chloride in pyridine. The phosphination of the ditosylate

56

is the synthetic key step to form (R)-Prophos (g) and (R)-
4-Me-Prophos (49). In Bosnich's method the phosphination
reagent, LiPth, was generated by cleavage of triphenylphos-
phine with a stoichiometric amount of lithium metal in THF

(Scheme 17).

PPh + Li ; LiPPh + LiPh

 

2

Scheme 17

The by-product, phenyl lithium, was destroyed with a stoichio-
metric amount of tart-butyl chloride. 75% excess LiPPh2

was used for the phosphination reaction (assuming triphenyl-
phosphine was converted quantitatively to LiPth). Phosphina-
tion of the ditosylate with excess LiPPh2 in THF afforded

the crude oily product of (3)-Prophos (6) (Scheme 18).

 

(C H ) P
CH . 6 5 2 \
3 xs LlPth _
TsO CH3

Oitosylate (R)-Prophos (Q)

+ Unidentate—phosphines
Scheme 18
However, the crude product, (3)-Prophos (6). containing a

large amount of unidentate phosphine by-products, caused

a separation problem.

57

Bosnich used the following procedures to purify the

crude product, (R)-Prophos (6) (Scheme 19).

 

 

(R)-Prophos (6) .+2 -
~ N1 /NCS .
Unidentate-phos- 4_;. [N1(R)-Prophos)2NCS]NCS
phines
4; 2 6 5 2
2. Crystallization H3

(3)-Prophos (6)
Scheme 19

The addition of Ni(ClO4)2 and NaSCN to the reaction mixture
allowed the separation of the product as an insoluable Ni(II)
(Prophos)2 thiocyanate complex. The pure (3)-Prophos (Q)
was obtained by treating Ni(II) (Prophos)2 thiocyanate com-
plex with an excess of NaCN and crystallizing the ligand
from absolute ethanol.

With the new, modified method, the phosphide reagents
were generated from n-butyl lithium in hexane and diphenyl-
phosphine or di-(4-methylpheny1)phosphine in THF. Only
stoichiometric amounts of lithium phosphides were required
for the phosphination reactions (Scheme 20). After the
phosphination of the ditosylate with LiPth, the crude
(5)-Prophos (Q) was separated from the oily reaction mixture

by crystallization from absolute ethanol. A second

58

(C6H5)2P\\
H‘y9_CH2P(C6H5)2

HPth/anuLi
CH3

    
  

CH3

\ (R)-Prophos (6)
H 1'“ C-CHonS
/

~

T50 r2P\

H “H C‘CHszrz

A

Oitosylate
‘ HP(C6H4CH3-p)2/anuL1 C

I

3 _
Ar - 4-CH3C6H4

(R)-4-Me-Prophos (49)

Scheme 20
crystillization from the same solvent gave (R)-Prophos (g)
with the same specific rotation as Bosnich's product. Also,
(3)-4-Me-Prophos (49) could be easily separated from the
solid reaction mixture by crystallization from absolute
ethanol/CHZClZ. Further recrystillization did not change
the Optical rotation of (R)-4-Me-Prophos (49).

The advantages of the new, modified procedure lie in
its simplicity and in the avoidance of the use of excess
lithium-diphenylphosphideor lithium di-(4-methylphenyl)
phosphide and highly toxic NaCN.

Table 12 shows the spectral parameters and physical
properties of (Q) and (49). In using the modified phos-

phination method the melting point and optical rotation of

10

(R)-Prophos (g) were the same as Bosnich reported. Since

the four methyl groups of the 4-methylphenyl rings of (49)

Table 12.

59

The Spectra Parameters and Their Physical Properties of

(R)-Prophos (Q) and (R)-4-Me-Prophos (49)

(3)-Prophos (6)

(3)-4-Me-Prophos (4g)

 

Melting point (C°)

67.5-68.5°C

129.5-130°C

 

Optical rotation

[6123 186.0°‘

(c 1.0, acetone)a

[5123 l72.6°

(c 1.0, benzene)

 

 

 

(m/e)

370 (M--CH2CHCH3)+

_ '1”
303 (H. HPC6H5P )
262, P(C6H5)3
185, +P(C5H5)2
109, +HPC6H5

31p NMR 2.1 ppm, -20.2 ppm 0.3 ppm, -22.1 ppm
(85% H PO as exter- J _ =21 HZ J _ =21 Hz
nal stgndgrd P P P P
(CDC13, 25°C
1.26 (dd, J1=15 , 1.18 (dd, J1=15.8, J = 6.7
1H NMR Jz= 6.5 H2 Hz, 3H, CH3) 2
(CDC13, TMS) 3H CH ) 1.78 (m, 1H, CH)
6( - . 3 2.17, 2.21, 2.27
m, I2: I" EL?) (3. a,
. 111, , 2 9 9
7.25 (complex m, 5°95 (2°EPAE§ m: 1 "a
20 H, 4 C H ) 6 4
6 5
+ +
Mass spectra 412 (M ) 468 M )

426 M--CH3CHCH2)+
345 (M~-HRC6H4CH3)+
304, :P(C6H4CH3)3
213, +P(C6H4CH3)2
123, HPC6H4CH3

 

 

 

 

 

High resolution mass observed 468.21459
spectra calcd (C31H34P2) 468.21358
(m/e)
Chemical analysis found: C, 79.20; H, 7.36;
P, 13.47
calcd. g; 73.33, H, 7.31,

 

a c = g/lOO mL of solvent

 

60

are not identical,four methyl resonances are expected.
However, the two methyl groups adjacent to the achiral
carbon atom have the same chemical shifts, but the methyls
near the chiral carbon atom are distinguishable. The

1H NMR spectrum of the four methyl groups (49) showed three
distinguishable peaks in the region of 2.1-2.36.(Figure 5). Also,
the CH3 at the carbon backbone showed a doublet of doublets
due to coupling to the adjacent hydrogen and phophorus atoms.
The same characteristics are observed for (5)4Prophos'

(6). The 3‘

P NMR spectrum of (49) showed two sets of doublets
at -O.3, -22.1 ppm. The chemical shifts and the coupling
constant are similar to those of (3)-Prophos (6) (see Table
12). (3)-4-Me-Prophos (49) has a high melting point (129.5-
l30°C) in comparison with that of (R)-Prophos (6) (68.9°C),
and it is easily crystallized.

The mass spectrum of (3)-4-Me-Prophos (49) exhibits
the fragmentation pattern similar to the pattern of (R)-
Prophos (Q). The high resolution mass spectrum and the chemi-

cal analysis of (R)-4-Me-Prophos (49) are in good agreement

with the expected values.
B. General Method for the Preparation of Cationic Catalysts

[Rh(diene)Cl]2 was treated with silver perchlorate
and then with diphos* where diphos* was DIOP(+) (3), SPIPHOS
(4§). (3)-Prophos (6) or (R)-4-Me—Pr0phos (49) to form the

cationic precursor [Rh(diene)diphos*]ClO4 (Scheme 21).

61

ANII oev 00mm oa mpuoo e_ Amwv moIaoca-oI-I-Amv co eIeoooIm III I

P .m «game;

e S on S 4.. 2. 2a... on i, . S
IH)4 d) 1‘ 11 — T d 1 4 — d1 d 1 q I— )q J d i— d 1 d d) — I“ d, (d 1 _r d 1 4 ‘

a a 4 u

     

 

 

 

 

.ggpu—Q- con-

   

 

 

 

 

 

Amev moIaoeI-oI-I-Amv

 

62.

AgClO
[Rh(diene)Cl]2 4 > [Rh(diene)NflO4

diphos* [Rh(diene)diphos*]ClO4

‘
1

 

diene = NBD or COO

-Prophos (6)

diphos* = DIOP(+) 3 , (R)
(3)-4-Me-Prophos(49)

( )
SPIPHOS (48)m~

~

Scheme 21

The cationic precursor [Rh(diene)diphos*]ClO4 can be used
directly for homogeneous catalytic reactions, or it can be
exchanged with Na-hectorite to form [Rh(diene)diphos*]+-

hectorite for intercalated catalytic hydrogenation.
C. Intercalation Catalysts and Interlayer Spacings

The cationic exchange of [Rh(NBD)(diphos*)]+ complexes
in MeOH or 95% EtOH with Na*-hectorite led to the formation
of intercalation complexes in which 15-20% of the Na+ ions
have been replaced by the rhodium complex. The d(OOl)
X-ray spacings for the unsolvated intercalation complexes
were 16-18 A. Typically, oneinoerofOOl reflection was
observed. The observed spacings agree with the expected
values for monolayer coverage of the interlayer surfaces.
Solvating the intercalation complexes with methanol or
95% ethanol causes the d(OOl) spacings to increase to

19-20 A. Usually two orders of 001 reflection were observed.

63

Since the silicate sheets are 9.6 A thick, this means that
interlayers occupied by the rhodium complexes are around
0
9-10 A thick.
D. Preparation of Dehydroamino Acid Derivatives via Un-
saturated Azlactones
The condensation of N-acetyl-glycines or N—benzoyl-
glycines with aromatic aldehydes in the presence of acetic

anhydride and sodium acetate is the commonly used method

for the preparation of unsaturated azlactones53 (Scheme 22).

 

o o ,,
R-E-NHCHz-C-OH + ArCHO ACZO I ArCH = ,C-q
NaOAc "79’0

R = CH3 or C6H5 R

Scheme 22

The carbonyl group of the azlactones can undergo nucleo-

philic attack by OH' or OR' to form dehydroamino acid deri-

'vatives (Scheme 23).

 

0 H E Y
o -
ArCH= c—C\0 ”Y _r \__/
N§cx (v' - OH , 0R') Ar/ \ NHCOR
1
R

Scheme 23

64

The dehydroamino acid esters can be easily hydrolysized
to the corresponding carboxylic acids. Six dehydroamino
acid derivatives used in this work were prepared by adopt-

ing the above procedures (Table 13).
E. Hydrogenation Conditions

In general, the concentration of the substrate used
in the hydrogenation reactions were in the range of 0.13-
0.27 M. The hydrogenation of the L- DOPAprecursor, which
has limited solubility, was carried out by utilizing 1.17 g
of substrate in 30 mL of solvent. The molar ratio of sub-
strate to rhodium under both homogeneous and intercalation
catalyst conditions was 100 to 200. Methanol or 95%
ethanol was used as the solvating medium. The hydrogenations
were carried out at room temperature and 1 atmosphere
pressure. The concentrations of Rh(I)(diphos*) complexes

3 M. The cationic rhodium(I)(diphos*)

were around 10'
catalysts formed under hydrogenation conditions are highly
oxygen sensitive. Precautions were taken to exclude

oxygen, since traces of oxygen will deactivate the catalysts
easily and lead to low optical yields of the products.2C
The hydrogenation rates were monitored by following hydrogen
uptake as a function of time. The Rh(I)(diphos*) enamide
complexes, which formed as intermediates in the hydrooenatinn

remflflcns,exhibited a characteristic orange-red color.

 

65

. AW
mIUSII

mcoamuc\o~: :_ u_u< UPEaccwuxxoumwm mu\
mmmcoaumme mc_c=oammseou use we mwmzposuz: ieiou_5muwu<ioiANv :ooux

m o

I

<@

Ioom\eo_o=.0m Ioaz msooaea e_ m eeo<
smumm mcwccoammeeou on» mo mwm»_oeuxz mmmm u_Eccc_uov_EmNcwmimiAmv :oou\

IIQSII ©

 

 

m

mama—ou\:o_u:_om uuomz :_ mmucEc::_u .I\

mcopomPNo mcwvcoammsgou mgu mo mwmxpogoup< -oc_5c~cmmimiawv ngum “moou\

, @

N :uou:2/ \
o :\w:oumom cw c “:94 \H

mcouumFNc mc_ucoammseou an» we mwmxposu»: UFEcccwoou_5mumu<iwiAMV zoom

cacao: cowumemaosa mumeumnzm

 

muwo< oewsooeuzgoo-fiwv Io Iowuaeeamea «IF .m_ «Pack

mme>wue>Psee m.e_ee ePEe::_uee_se~:eeieiAmv
we eceueePNe ecu Lew eeeweemee eegues es» ea mc_eeeeee ee~_me;uczm we: eceueeFNe ache

 

mzo u<o

m
:uouzz
"w eceueee\o~: :_ e_e< e_se::_exxeueeei¢ /Hw\
emceuee_~e meweceemeseee one we mwmxpeee»: ixxegueeimiee_Eepee<ieiAMV :ooo\
:o
m
:uouzz \
cewuepem :oez z m.o :_ meeuee—Ne e_e< epseecwepxxeee»; JH
aspeceemeesee es» 4e mwmxpeeex; emem om -Iiee_seuee<ieiamv zoou\

 

 

eeguez cevueeeeeee eeeeumeem

 

eoIeeeeoe .m_ e.eae

67

This color persisted throughout the hydrogenation periods.
At the end of the reactions the color of the solutions
instantly changed to straw-yellow. The same color changes

were observed for the intercalated catalysts.

F. Characterization of Dehydroamino Acids and Their
Corresponding a-Amino Acids

13C NMR provided distinguishable features for the

dehydroamino acids and their corresponding hydrogenated

products. Table 14, 15 and 16 show the 13

C NMR chemical
shifts of selected carbons of the dehydroamino acids and
their corresponding a-amino acids. The Chemical shifts

of the amide carbons (NHEOR), carboxyl carbons (COOH),

and vinylic carbons (a position) were significantly altered
upon hydrogenation of the C = C bond. The chemical

shifts of selected carbons were assigned based upon

38,58. The 1

the literature H NMR spectra of the dehydro-
amino acid substrates and the corresponding hydrogenated
products also were different.

The chemical conversions were monitored by observing
the disappearanutof substrate and the appearance of product

1H NMR. The differences in chemical shifts of the a

by
and 8 hydrogens in the corresponding a-amino acid deriva-
tives allowed accurate estimation of the percent chemical
conversion in these systems. Chemical shift differences
between the aromatic hydrogens of tyrosines and DOPA

precursors and their corresponding hydrogenated products

68

.eepeeeeee :euese wee:

V

egueeem :< .eseeceum 353...? :e we m2» cu «>533 5.8.3 com—TE. Lee Seen. 5 emuseeee age «:2 u

 

 

m_a
AMIeev m e mu
I eeeIz T
~.Nm m.me_ N.eep ozeee \111
mIeeeeIz
e.e. m.ee e.em m.ee. ~._e_ _
mIeNIemeeImNIemIee
mIeeeeIz
_.I_ m.oe e.me_ m.eep m N N .
Ie Ie oe\\iiii
mIeeeeII
. . . . _
m em I ee_ m me_ IeoeInmIemIee
mIeeeeIz
m.ee_ m.ee.
Ieee_\
mam Ie mIewImo Im Iee< Ieee mIeeemII
aeamm III NI: me.e~ ea eoeeapom emze-ee e_ meeoI oeee<-e
me.eeoemoeeoe eeoIe eea mIPoI oe_e6oee»IoI Io moepIm Peo_eoee III e .I. IIIIP

mp

69

.ee—eaeeme :eaese wee: espeeem ~—< .egeeceam
peceeucw :e we,wzh ea «IIIeFeI Agnew cew__ps gee mueee :P eeugemue use eeuueem Izz um_
. .e

 

 

IIeeeIw
e.em m.- m.IeI I.MII IeeeIeNIeIIeeoI-I
a :o
mIeeeII
mIeeeII
I.mm m.- I.Ie_ m.mII IeeeIememIee
.4.
m
IeeeII / WW
I.- I.ee_ I.Ie_ Ieee \\111
:w IeeI meeeII Ieoe mIeewII
meomm III ~II me.e~ ea eo_ea_om emII-eI II me_o< oeIe<-e
IeIeeoImoeeoe IIoII III IIIoI oIIIIoIIIIoI Io mee_Im .aoIIoIe III e .m_ o_eae

mp

70

.eepezeeme :eeeee use: eeaeeem FF<

.eeeeceum.Pe:Ieu:_

:e we m5 3 3329.. ecu €53 5sz gen 3&3 E acetone..- 88 eeuemem «22 on—
e

_¢.om .I.NN

w.o~ m.-

IeemIm mImIeII

mfiIeI

I.mm m.Ie_
I.II_

I.mm .m.IeI
mIII eem II
III Ieee

N..ws~

v.mo_.

mIeewII

IIeeeII oIe eoI
I _
IeIeIm Ie
oII ooI
mIIIIII

:oou.\

 

m . euomm ecu ~22 mo.o~ we copuzpom
maze: a cw <eoD14 ecu Lemszumsm <aoQ we muwpsm peuwemso «:2 any .09 mpneh

71

provided additional information for estimating the chemi-
cal conversions of the products. Seven dehydroamino acids
were used for this work. The percent chemical yields

of the corresponding amino acids were larger than 97%,

1

as evidenced by the H NMR spectra.

G. Optical Yields

The optical yields were determined by dividing the
specific rotation of the product mixture by the absolute
rotation of the pure enantiomer determined under identical
conditions of temperature, concentration and wavelength.
The method assumes a linear relationship between rotation

and composition.1d

[9]
%optical yield = X 100% = %enantiomeric
[°]abs excess

 

It must be noted that the assumption is not always valid
as in the case of o-methyl-o-ethylsuccinic acid where the

59

plot deviated appreciably from linearity. The absolute

rotations of the Pure corresponding amino acid derivatives
were taken from the literature 9.

In this study optical rotations were recorded with a
Perkin Elmer 141 automatic polarimeter. Two measurements

were recorded for each sample by preparing two independent

solutions. The optical rotation method is the commonly

72

accepted method for estimating optical yields of asym—
metric hydrogenation reactions. Other direct methods
which are less sensitive_to impurities are also possible
such as NMR, GLC, or HPLC analysis of diastereomeric

mixtures.1c

l. (3)-Prophos (g) and (R)-4-Me-Prophos (49) Systems

The results of asymmetric hydrogenation of dehydro-
amino acids with intercalated and homogeneous [Rh(NBD)
(3)-Prophos (6)]+ catalysts are shown in Table 17. In
the homogeneoussystem the optical yields of the different
substrates were around 90%, in good agreement with the
results of Bosnichlo. The optical yields of the corres-
ponding o-amino acids with intercalation catalysts were
generally lower than those obtained with the homogeneous
catalyst, with one exception. The optical yield for
N-acetyl tyrosine was slightly higher for the intercalated
catalyst. In general, the deviations from solution yields
were smallest for the acetyl derivatives of L-phenyl-
alanine and DOPA (~5%). Larger deviations were observed
for the remaining precursors (~15-30%).

The results of asymmetric hydrogenation of dehydro-
amino acids with intercalated and homogeneous [Rh(NBD)
(3)-4Jm-Prophos (4§)]+catalysts are given in Table 18.

In the (R)-4-Me-Prophos (49). system, the optical yields
obtained with the homogeneous catalyst were similar to those

obtained with the homogeneous Rh(I)-(R)-Prophos (6) catalyst.

73

Table 17. Asymmetric Hydrogenation of Dehydroamino Acids
with Intercalated and Homogeneous [Rh(NBD)(R)-
Prophos (6)]Icatalysts

 

-——— Optical Yield (%)

 

 

 

 

Substrate Amino Acid Intercal. Homogen.
::<C°°” L-Alanine 73.3 93.7
NHCOCH3
_/COOH L-Phenyalanine 85.4 89.0

<:S_\NHCOCH3

coo" L-Phenyalanine 70.0 92.0
c°°c2”5 L-Phenyalanine 60.5 89.3

C00” L-Tyrosine 82.1 79.0

COOH L-Tyrosine 75.0 90.1

COOH L-DOPA 86.7 92.0

sg$\NHC0CH3

AcO OMe

 

aThe hydrogenations were carried out on 4-8 mmol of substrate in 30 mL
95% EtOH at 25°C and 740 torr. The amount of intercalated or homogeneous
catalyst. used for each reaction was 0.04 mol. All the corresponding
acids had (§)-configurations.

74

Table 18. Asymmetric Hydrogenation of Dehydroamino Acids
with Intercalated and Homogeneous [Rh(NBD)(R)-
4-Me-Prophos (49)]+ Catalysts

 

—— Optical Yield (%)9—

 

 

 

Substrate Amino acid Intercal. Homogen.
_d/COOH L-Alanine ' 76.5 88.0
—“\NHCOCH3
_J/COOH L-Phenyalanine 89.6 92.6
@5—\NHCOCH3

‘_/COOH L-Phenyalanine 75.7 94.5
@NHCO@

._/COOC2H5 L-Phenyalanine 75.5 93.6
@‘NHCO@

-_/COOH L-Tyrosine 78.5 72.0
Gig-\MCOCH3

H

._/CO0H L-Tyrosine 88.1 91.0
gHHCOCH3
Ac
_d/CO0H L-DOPA 95.1 95.3
NHCOCH3
Ac Me

 

aThe hydrogenations were carried out on 4-8 mmol of substrate in 30 mL
95% EtOH at 25°C and 740 torr. The amount of intercalated or homogeneous
catalyst’ used for each reaction was 0.03 mmol. All the corresponding
acids had (§)-configurations.

75

The optical yields of the corresponding o-amino acids

with the intercalated Rh(I)-(R)-4-Me-Prophos (4?) catalyst
precursor were lower than those obtained with the homogeneous
catalyst. However, the deviations from homogeneous opti-
cal yields were less for the (3)-4-Me-Prophos (49) system
than for the (3)-Pr0phos (6) system. This could be proven
by comparing the difference in optical yields (YI-YH)

for the (R)-4-Me-Pr0phos (49) system with those of the
(3)-Prophos (6) system; where YI = optical yield with
intercalated [Rh(NBD)diphos*]+catalyst, YH = optical

yield with homogeneous [Rh(NBD)diphos*]+ catalyst (Table
19).

The methyl groups at the para positions of the phenyl
rings of [Rh(NBD)(R)-4-Me-Prophos (43)]+ enhance the
catalyst's size, relative to [Rh(NBD)(R)-Pr0phos)]+. The
4-Me-Prophos complex is intercalated into the layered
silicates; the interlayers should be more solution-like
than when the (R)-Prophos occupies the interlayers.

Also, the optical yield (95%) of the most important amino
acid derivative investigated, L-DOPA. was the same for
intercalated and homogeneous [Rh(NBD)(R)-4-Me-Prophos
(49)]+ catalyst precursors. Thus, there appears to be
considerable potential for the practical application of

layered silicate intercalation catalysts.

76

Table 19. Difference in Optical Yields (Y -Y ) for
- Intercalated and Homogeneous Catio ic
[Rh(NBD)(diphos*)]+ Precursors.

 

 

 

 

 

 

YI-YHa
Substrate (R)-Prophos (6) 4-Me-(R)-Prophos (49)
_d/COOH -20.4 -11.5
-—\NHCOCH3
COOH - 3.6 - .
__/ ~ 3 0
<:jr_\NHCOCH3
._/COOH -22.0 -18.8
NHC0C5H5
of
‘_JC00C2H5 -28.8 -18.1
(:jr_\NHCOC6H5
_JLOOH +3.1 +6.5
/-\
O NHCOCH3
0H
-15.1 -
._/CO0H 2.9
<:> ‘NHCOCH3
/
ACO
_/CO0H -5.3 -0.2
€:§—\NHCOCH3

ACO OMe

 

aYI and YH are the optical yields for the products obtained
with the intercalated catalyst and the homogeneous catalyst,
respectively.

77

2. SPIPHOS (48) System

 

The SPIPHOS ligand (48), which had a high degree of con-

52 The cationic

formational mobility, was prepared by OeVries.
[Rh(COD)SPIPHOS (4,811+ precurosr and incite rhodium SPIPHOS
(48) catalyst, which was prepared from [Rh(cyclo-octene)2Cl]2
and SPIPHOS (48), were used for the asymmetric hydrogenation
of (%)-dehydroamino acids and non-enamide type substrates.
Both systems afforded moderate optical yields and showed
a strong base effect with the addition of triethylamine.52
In some cases, the absolute configuration of the products
were reversed by the addition of triethylamine. Variations
in temperature and pressure only slightly affected optical
yields.

An inspection of molecular models suggeSts that when
the ligand coordinates to rhodium, it is capable of forming
a ten-membered chelate ring. The complex is highly flexible.
Due to the large chelate ring size, the phosphorous atoms
of the ligand could trans-coordinate to rhodium. A trans
structure for the Rh(I)-SPIPHOS complex could lead to low
asymmetric induction and thus low optical yields in the asym-
metric hydrogenation of (;)-dehydroamino acids.

,In the layer silicate system, the flexibility of the
Rh(I)-SPIPHOS (48) complex might be restricted by the sili-
cate sheets. However, if the trans-coordinated structure
of the complex dominated in the interlayer regions, the

optical yields may not be improved by intercalation.

Table 20 shows the results of asymmetric hydrogenation of

78

.IIeIIIIIIIIIeeifimQ we; Ie_ee mcpeeeameggee an» __< .Peee
mo.o II: IeIaeeeI some gem eeI: IIIIFIuIe Iaeeeemese; Ie eeue_eegeuew we eczese och .egep oen ecu
uemm In Ion“ III Ie roe: Io Is on II eIIIIIIII me Peas I go I Ie uze eeIIIIe egez IIeIIIIIIeIeII egpe

 

mm
mm

mm
cu

me
em

me
we

.cemese:

IIIII_III IIoIIII 1111

cc
cw

mm
up

mm
me

me
mm

 

.peeeeucg

.oIIIIII
IIII III

IIIIIIII
IIII III

_IIIIIII
IIII III

IIIIIIII
IIII III

 

uIe>—em

<moo

ecwcepepacesm

eewcepe—xeese

eIIIIFepzceII

 

IIoI IIIII

4

e: e<
IIIIII
IIII\1.

m

 

euesumeem

 

use wean—eeeeacm cup: IeIe<_e:IEee

IIIIIIIII ”HMmIV IIIIIIIIIIIVIIH IIIIIIIoIoI
u

see we :ezeeemege»: ezueesxg . cm 023

79

(g)-dehydroamino acids with intercalated and homogeneous
[Rh(NBD)SPIPHOS (48)]I complexes. The lower optical yields
obtained with the intercalated [Rh(NBD)SPIPHOS (48)]+

catalyst appear to support this prediction.
3. DIOP(+) (3) System

As shown in Table 21, the optical yields obtained with
[Rh(NBD)DIOP (8)]+ intercalated in hectorite depend greatly
on the nature of the substrate. With the intercalated
[Rh(NBD)DIOP(+) (3)]+ catalyst, the optical yields of N-
acetyl .amino acids of alanine, phenylalanine and DOPA deviate
only slightly from those obtained with the homogeneous catalyst.
The optical yields of N-acetyltyrosine and N,0-diacetyl-
tyrosine were higher in comparison with the homogeneous re-
sults. The optical yield of the N-benzoylphenylalanine
was lower than the homogeneous result, whereas the reduction
of its corresponding ester gave zero optical yield with
the intercalated catalyst. Since the recycled catalyst ob-
tained from the reduction of the ester gave a 77% optical
yield for N,0-diacetyl-tyrosine reduction, the zero optical
yield obtained for the phenylalanine derivative was not due
to the loss of the catalyst chirality.

The differences observed between the homogeneous and
intercalated state are not surprising. DIOP(+) (8) forms
a seven-membered chelate ring when complexed to rhodium.

The optical yields obtained with such conformationally flexi-

ble ring systems are sensitive to solvation effects,36’6O

80

 

 

 

 

Table 21. Asymmetric Hydrogenation of Dehydroamino Acids
with Intercalated and Homogeneous [Rh(NBD)DIOP(+)
(8)]+ Catalysts
-—— Optical Yield (%)a
Substrate Amino acid Intercal. Homogen.
_/COOH L-Alanine 74.0 76.0
—\NHC0CH3
_/CO0H L-Phenylalanine 85. 84.8
@NHCOCHiS
_d/CO0H L-Phenylalanine 46. 58.0
@‘wnco@
_J/COOC2H5 L-Phenylalanine 42.8
6NHCO©
__/CO0H L-Tyrosine 82. 73.6
@rWHCOCH3
OH .
‘_/CO0H L-Tyros1ne 77. 72.4
és—WHCOCH3
ACO JCOOH L-DopA 86. 88. 0
9 NH COCH3
ACO OMe

 

aThe hydrogenations were carried out on 4-8 mmol of substrate in 30 mL

95% EtOH at 25°C and 740 torr.
catalyst used for each reaction was 0.04 mmol.
-configurations.

acids had (8)

The amount of intercalated or homogeneous
All the corresponding

81

and, in some cases, the absolute configurations of the pro-

ducts can be reversed by changing solvent.36’60

A strong
solvent effect was observed in the asymmetric hydrogenation
of l-acetamido-l-phenylethene with cationic [Rh(COD)OIOP
(+) (8)]1 precursor and in situ Rh(I)-DIOP(+) (3) catalyst

in different solvents (Table 22).36’60

Table 22. Solvent Effect on the Optical Yields of Hydrogenation Product
of l-Acetamino-l-phenylethene with DIOP(+) (8)6

 

 

 

Optical Yie1d

 

 

Substrate Solvent Cation precursorb In situ Catalyst:
__/NHCOCH3 EtOH 38.5 (3) 42.5 US)
“\

C6H5 C6H6 68.0 (3) 44.0 (§)

 

aData obtained from references 36 and 60.

bCationic precursor: [Rh(COD)OIOP(+) (3)]+C104

‘3In situ.catalyst: [RhCl(C2H4)2]2 and DIOP(+) (8) with DIOP/Rh ratio of
1.1.

Even the absolute configuration of N-acetylphenylethylamine
can be reversed by using in situ,Rh(I)-DIOP(+) (3) catalyst
in benzene.36’60 Also, in general, dehydroamino acids give
higher optical yields than the corresponding esters.6]
With the in situ Rh(I)-DIOP(+) (8) catalyst, N-acylamino-

cinnamic acids were hydrogenated to phenylalanine derivatives
with significantly higher optical yields (ZS-81%) than the

61

Corresponding esters (5-59%). It is reasonable, therefore,

to expect solvation differences also to exist between reaction

82

intermediates in the intercalated and homogeneous states.
Of course, other factors such as specific interactions of
the substrate with the silicate oxygens may also contribute

to the observed differences.
4. The Solvent Effect on the Optical Yields

(§)-Prophos (Q) is a conformationally rigid type of li-
gand. Wiflithe homogeneous Rh(I)-(3)-Prophos (8) catalyst
precursor the optical yields appear to be insensitive to the
nature of the substituent on the substrate and solvent.10
The solvent effect on the optical yields of one substrate
obtained with intercalated [Rh(NBD)(B)-Prophos (8)]+ is
illustrated in Table 23.

Table 23. The Asymmetric Hydrogenation of (;)-a-Benzamido-

cinnamic Acid with Intercalated and Homogeneous
[Rh(NBD)(5)-Prophos (6)]+ Catalysts

 

 

Optical Yield (ml——

 

Substrate Solvent Intercal. Homogen.
_d/COOH H20:Et0H (1:19) 70% 92%
d‘NHCOCGHS H20:EtOH (1 :3) 67% —
H20:EtOH (1:2) 63% ——
Methanol 71% 88%

 

aThe hydrogenations were carried out on 4 mmol of substrate in 30 mL

of solvent at 25°C and 740 torr. The amount of intercalated or homogeneous
catalyst used for each reaction was 0.04 mmol. The corresponding

spamino acid had an (§)-configuration.

83

In general, the intercalated catalyst afforded low optical
yields relative to the homogeneous catalyst. The optical
yields in EtOH/HZO decreased only slightly with increasing
water content. It appears that the interactions between
the layer silicates and the cationic Rh(I)-(R)-Prophos (g)

enamide complex are dominant.
5. The Effect of Rhodium Complex Precursor on 0ptica1 Yields

With cationic Rh(diene)diphos* catalysts, the rates
of hydrogenation with six- or seven-membered chelate rings
are somewhat faster than those observed for the five-
membered chelate ring analogues. Either COD or NBD rhodium
diphos* precursors which form six- or seven-membered chelate
rings can be used as effective asymmetric hydrogenation
catalysts for reduction of dehydroamino acids at ambient

60,62

conditions. However, with five-membered chelate chiral

diphosphine rhodium catalysts, COO precursors seemed to be

less active than NBD precursors.9’15

For some substrates,
high hydrogenation pressure was required to observe reactions
with coo derivatives.15 Bosnich9 observed that [Rh(NBD)
Chiraphos (5)]+ precursor was more active than cationic
[Rh(COD)Ohiraphos (5)]+ for asymmetric hydrogenation of
dehydroamino acids. The five-membered chelate [Rh(con)
(§,§)-Phellanphos(19)TPFg and [Rh(COD)(R,g)-Nopaphos (LJ)]+PFg
precursors were effective catalysts for asymmetric hydro-

genatfion of dehydroamino acids and non-enamide type substrates

in ethanol.15 The best results were obtained in asymmetric

84

synthesis of o-amino acid derivatives (BO-95% e.e.).
However, the (§,§)-Phellanphos (19) sometimes works under
one atmosphere of hydrogen. The (3)3):Nopaphos (1}) gives
a less active catalyst and high pressure (15 atm) is re-

15

quired for activating the catalyst. Also, Schrock and

Osborn63

found that the cationic [Rh(NBD)(PPh3)2]+ catalyst
reacts with hydrogen approximately 100 times faster than
the cationic [Rh(COD)(PPh3)2]+ catalyst.

The intercalated [Rh(NBD)(g)-Prophos (6)]+ catalyst
was found to be a more effective catalyst than intercalated
[Rh(COD)(B)-Prophos (8)]+ for the asymmetric hydrogenation
of the dehydroamino acids (Table 24). It is reasonable
to assume that there is a reaction activity differentiation
between the two cationic [Rh(diene)(R)—Prophos (6)]+ pre-
cursors (where diene is N80 or C00) in the layer silicate
system. These results suggest that some decomposition of
the [Rh(diene)(§)-Prophos-:(§)]+ species on the interlayer
surfaces may contribute to lower optical yields. But the
effect cannot be large with NBD precursor because of the

high optical yields obtained with [Rh(NBD)(R)-4-Me-Prophos
+
(1,9)] .

6. Asymmetric Hydrogenation with Recycled Catalysts

The results of asymmetric hydrogenation of o-aceta-
Sminocinnamic acid with recycled [Rh(NBD)DIOP(+) (8)]+
and [Rh(NBD)(R)-Prophos (§)]+ catalysts are shown in Table 25.

85

Table 24. Asymmetric Hydrogenation of Dehydroamino
Ac1ds w1th Intercalated an Homogeneous
[Rh(d1ene)(5)-Prophos (8)] Catalysts

 

Optical Yield (%)a

Substrate Diene Amino Acid Intercal.‘ Homogen.

COOH NBD L-Alanine 73% 94%
“‘\NHC0CH3 coo 50% ——
,COOCZHS NBD L-Phenyl- 61% 89%
A NHCOCGHS coo “am” 50% —

 

aThe hydrogenations were carried out on 8 mmol of aIacetamidoacrylic
acid or 4 mmol of ethyl (g)-oIBenzamidocinnamate in 30 mL 95% EtOH.
at 25°C and 740 torr. The amount of intercalated or homogeneous
catalyst used for each reaction was 0.04 mmol. The corresponding

. o-amino acid products had (§)-configurations.

86

eemeee exeue: :emeeexxe

II Pese\cms\~: II II eeugeeee men mean; PIIIIIHI
IeIIIIIIIIIeeude II em; ecwcepmpzcege
-qumemiz III .2 mio_ x mm._ ecaeem wee: Iqu—Igme eeuepmeeeucp Ie IceIIILpIeecee eII .Ieep

OII III comm “I 10““ III we 42 om II IIIIHIIII esp we Pose m :e use eeIIeIe mew; IIeIIIIemeIeII eIII

 

.cve mN mm &~.pw mpuau ugm mzoooxz/IWWV

 

 

 

 

 

.III III I.I II._I .III II II IN.II IIoIo III \1
Ieee
.III_ III I.IN II.II .III N_ III II.II I_oIo III
“IIIII IIIII I.>.I “IIIII IIIII I.>.e IIIIIIIII
Awe IIIIIII-IMI Awe I+VIIII
Imocawo

 

32,38 IWIIIIIIIIIIIIZIH IIIIIIIIIIII
empeaeem III: eIe< eIEIIIIeeIIEIuee<ieia v Ie ceIpeIemeeezz eIIuesEII< .IN epeeh

87

The rates of hydrogenation decreased by a factor of 4. with

the reuse of the intercalated catalysts, but even after

three cycles of the catalyst the optical yields of the pro-
duct were not affected more than 5%. It is possible that the
decrease in activity is due to loss of catalyst through
oxidation by trace amounts of oxygen. Also, clogging the

interlayers by adsorbed amino acid is possible.
H. The Observed Hydrogenation Rates

The hydrogen uptake plots for reduction of N—acetyl-
dehydrophenylalanine with homogeneous and intercalated
[Rh(NBD)diphbs*rcatalysts (where diphos* is (8)-Prophos (Q),
(8)-4-Me-Prophos (48), SPIPHOS (48), or DIOP(+) (8) are
shown in Figures 5 to 11.

The observed rates for hydrogenation of the dehydro-
amino acids with homogeneous and intercalated [Rh(NBD)
(jg-Prophos (8)]+ catalysts are shown in Table 26 . The
homogeneous [Rh(NBD)(8)-Prophos (8)]+ hydrogenation rates
were in the range of 19 to 330 (mL H2/min/mmol Rh). The
observed hydrogen rates were similar to those Bosnich found?)
In general, the rates of hydrogenation with intercalated
[Rh(NBD)(8)-Prophos (8)]+ catalyst were 0.13 - 0.45 as
fast as those with the homogeneous catalyst.

The hydrogenation rates for reduction of dehydroamino

acids with homogeneous and intercalated [Rh(NBD)(8)-4-Me-

Prophos (48)]+ catalysts are shown in Table 27. With

88

Figure 6. Hydrogen uptake plots for reduction of
5 mmol of N-acetyldehydrOphenylalanine
in 30 mL 95% ethanol at 25°C and 740 torr
with [Rh(NBD)(8)-Prophos (8)]+ as the
catalyst precursor. (A) Homogeneous
catalyst (B) Intercalated catalyst. In
each case the amount of rhodium complex
used was 0.04 mmol.

 

89

A:_Eveewp

_ «

oep
_

cup

cop
I

o eezmwm

_

om

 

 

H [9101

7v

("1‘“)

Figure 7.

90

Hydrogen uptake plots for reduction of

5 mmol of N—acetyldehydrophenylalanine

in 30 mL 95% ethanol at 25°C and 740 torr
with intercalated [Rh(NBD)(B)-Prophos
(8)]+ catalyst. (A) First cycle (8)
Second cycle. The amount of rhodium
complex used was 0.04 mmol.

91

35.8.: III
. _

  

mgzm_m
I. I

com
_

 

Figure 8.

92

Hydrogen uptake plots for reduction of

5 mmol of N-acetyldehydrophenylalanine

in 30 mL 95% ethanol at 25°C and 740 torr
with [Rh(NBD)(8)-4-Me-Prophos (48)]+ as
the catalyst precursor. (A) Homogeneous
catalyst (B)' Intercalated catalyst. In
each case the amount of rhodium complex
used was 0.03 mmol.

 

 

 

 

93

AIIIcmEII

QIP

om—

cop

w eezmII

 

I

 

om

ow

co

(1w)ZH 19401

cm

oop

 

Figure 9.

94

Hydrogen uptake plots for reduction of

5 mmol of N-acetyldehydrophenylalanine

in 30 mL 95% ethanol at 25°C and 740 torr
with [Rh(NBO)SPIPHOS (48)]+ as the
catalyst precursor. (A) Homogeneous
catalyst (B) Intercalated catalyst. In
each case the amount of rhodium complex
used was 0.03 mmol.

AcwEvamH.N— m meaawn—D
_ I _ H . I

 

 

 

95
(1w)zH 19101

 

Figure 10.

96

Hydrogen uptake plots for reduction of

5 mmol of N—acetyldehydrophenylalanine

in 30 mL 95% ethanol at 25°C and 740 torr
with [Rh(NBD)DIOP(+) (8)]+ as the
catalyst precursor. (A) Homogeneous
catalyst (B) Intercalated catalyst. In
each case the amount of rhodium complex
used was 0.04 mmol.

 

 

97

Acweveewp

NP

m

I_ IIIIII

m

 

A

 

I
1.II
1.II
I.
l O
m.
TIII.mm
w
mu
1.II
..II_

 

Figure 11.

98

Hydrogen uptake plots for reduction of

5 mmol of N-acetyldehydrophenylalanine

in 30 mL 95% ethanol at 25°C and 740 torr
with intercalated [Rh(NBD)DIOP(+) (8)]+
catalyst. (A) First cycle (B) Second
cycle. (C) Third cycle. The amount of
rhodium complex used was 0.04 mmol.

99

Time(min)

J
1 0

60Figure 11 80

 

40

20

 

0

 

)2

(1w H 19101

O

100

Table 25. Hydrogenation RateS'fiN‘Intercalated [Rh(NBD)(R)- -prophos (8)]
Catalyst and Its Homogeneous Analogue

 

Hydrogen Rates (ml H2/min/mmol Rh)a

 

Substrate Intercal. Homogen. 'Relative Rate
__/ COOH 128 328 0.39 ‘
\.. NHCOCH3
J00" 21.5 71 0.32,
@fViHCOCH3
' Jc00H 15.3 50 0.33
<§§_\11Hcoc6H5
JCOOCZHS 3.66 28.5 0.13
NHCOCSHS
COOH
J 8.60 19.1 0.45,
§CHHCOCH3 -.
_Icoo 11.3 60.0 0.19
0' NHCOCH3
OAc
COOH
J 13.3 51.3 0.26
NHCOCH3
OAc OCH3

 

a The reaction conditions are the same as those given in Table17 .

101

Table 27 . Hydrogenation Rates for Intercalated [Rh(NBD)(3)-4-Me-Prophos (1,911+
Catalyst and Its Homogeneous Analogue.

 

Hydrogen Rates (m1 Hzlmin/nlnol Rh)a

 

Substrate Intercal. Homogen. . . Relative Rate
_/COOH 333 537 0.62
_\HHCOCH3
COOH
J 30.7 7105 0.29
6NHCOCH3
JCOOH 45.6 80.0, 0.57
@‘NHCOC6H5
000 H
_/c 2 5 10.9 36.4 0.30
@‘NHCOCGHS
_/COOH
11.7 22.2 0.53
§(j‘\iiHCOCH3
H .
4°00” 11.8 53.3 0.22
HC0CH3
OAC
4cm" 30.7 46.0 0.67
HC0CH3
Ac OCH3

 

aThe reaction conditions are the same as those given in Table 18.

102

homogeneous [Rh(NBD)(8)-4-Me-Prophos (.§)]+ catalyst,

the hydrogenation rates for the reduction of dehydroamino
acids were in the range of 22 to 540 (mL Hz/min/mmol Rh).
The observed hydrogenation rates with homogeneous [Rh(NBD)
(8)-4-Me-Prophos (48)]+ catalyst were comparable to those
with the homogeneous [Rh(NBD)(8)-Prophos'(8)]+ catalyst.
The observed rates with intercalated [Rh(NBD)(R)-4-Me-
Prophos (48)]+ catalyst were 0.22 to 0.67 as fast as

those with the homogeneous catalyst.

The N-acetyldehydroalanine and N—acetyldehydrotyrosine
gave the highest and lowest rates, respectively, in the
homogeneous (8)-Prophos (8) and (8)-4-Me-Prophos (48)
systems. The same findings were observed by Bosnich with

[Rh(NBD)(8)-Prophos (8)]+catalyst.10

In the layer silicate
system, the ethyl N-benzoyldehydrophenylanaline ester gave
the lowest rates in the intercalated (8)-Prophos (8)

and (8)-4-Me-Prophos (48) systems.

The hydrogenation rates for reduction of dehydroamino
acids with homogeneous and intercalated [Rh(NBD)SPIPHOS
(48)]+ catalysts are shown in Table 28. With homogeneous
[Rh(NBD)SPIPHOS (48)]+ catalyst, the hydrogenation rates
for the reduction of dehydroamino acids were in the range
of 500 to 1600 (mL HZ/min/mmol Rh) in 95% ethanol and in
the range of 1000 to 2500 in methanol. Hydrogenation

rates are faster in methanol than in 95% ethanol. Ethyl

N-benzoyldehydrophenylalanine ester gave the highest and lowest

.ON mpnwh cw cm>_.m mmozu mm mEmm wc-H mgm mCOwHwUCOU cowuummg mnhm

 

103

 

 

 

III III
II.I III III IIII IIIIIII
IN.I III III IIII III IIII\\1
II.I IIIN . II. IIII IIIIIIII/,
II.I III, II IIII III IImuoeux
II.I IINI III IIII IIIeeIII/xthHv
II.I III II. IIII III IIII\\
. I @
I I III I II II II
I I I II I II /(1\\
IN.I III III IIII III IIII\\
wee; e>Iue~eI .cemeee: .pmeeeHIH uce>pem eumgumezm

IAII Pess\:_s\~: 45V Ieuem cemeee»:

 

2 enmepec< Izeeeemese: III
Ice umxpeumu HFAmIV mozememaom2vcmg eeaepmeeeecm Lew IeuII cevuecemege»: .mm epemh

104

rates in the homogeneous and intercalated system, respective-
ly. The observed rates with intercalated [Rh(NBD)SPIPHOS
(1L8)]+ catalyst were 0.05 to 0.40 as fast as those with

the homogeneous catalyst.

The hydrogenation rates for reduction of dehydroamino
acids with homogeneous and intercalated [Rh(NBD)DIOP(+) (§)]+
catalysts are shown in Table 29. The observed reaction rates
for homogeneous [Rh(NBD)DIOP(+) (3)]+ were in the range
of 230 to 680(mL Hz/min/mmol Rh). In the DIOP(+) (3) system,
the N-acetyldehydroalanine and [mpA precursor gave the
highest hydrogenation rates in the homogeneous and inter-
calated states. The N-benzoyldehydrophenylalanine and its
ester gave the lowest hydrogenation rates in the homogeneous
and intercalated states. The observed hydrogenation rates
in the intercalated state were 0-43 t0 0-94 as fast as
those observed in the homogeneous state.

The hydrogenation rates of dehydroamino acids (enamides)
with homogeneous [Rh(diene)diphos*]+ catalysts are de-
pendent on several factors, such as the type of the diene
precursor, the steric and electronic effects of substrate,
solvent, and the catalyst's chelate ring size. The hydro-
genations of dehydroamino acids with NBD precursor are
faster than the COD precursor in the five-membered chelate
ring's cationic Rh diphos* system. The steric and elec-
tronic effects of substrate are important. However, it
is difficult to predict these two effects precisely in each

substrate. The hydrogenation rates are faster in methanol

105

Table 29 . Hydrogenation Rates for Intercalated [Rh(NBD)DIOP(+) (3)]
Catalyst and Its Homogeneous Analogue

 

Hydrogen Rates (ml H2/min/mnol Rh)a

 

 

Substrate Intercal . Homogen. Relative Rate .
COOH
J
a 486 515 0.94
NHCOCH3
0H -
_/co 272 291 o .93
@gfirmcocu3
JCOOH 296 683 0.43
QS‘NHCOCH3
OAc OCH3
Jcoou 191 234 0.82
@‘NHCOCGH5
JCOOCZHS 125 243 0 - 5‘
@‘NHCOCGHS
1°00“ 237 375 0.63
gNHCOCHB
0H
_/CO0H 250 475 o .53
{Ci—\NHCOCst
0A3:

 

aThe reaction conditions are the same as those given in Table 2l .

l06

than 95% ethanol for reduction of dehydroamino acid with
Rh(NBD)diphos* catalyst.
A series of homogeneous in situ rhodium diphosphine

catalysts were prepared by Kagan64

to study the effect of
changing the catalyst's chelate ring size on the hydro-
genation rates of a-acetamidocinnamic acid. The results
indicated that as the chelate ring size of the catalyst
increased from 5 to 8, the reaction rates also increased
(see Table 30). However, when the catalyst's chelate

ring size increased to 9 the rate dropped again. DeVries52

found that the homogeneous ten-membered chelate ring's rhodium

SPIPHOS (fifi) catalyst, prepared from [RhCl(cyclooctene)2]2

and SPIPHOS (Q§), was inactive for asymmetric hydrogenation

of N-acetyldehydrophenylalanine in ethanol/toluene at am-

bient conditions. High pressure was required for activating

the catalyst. The finding was close to Kagan's prediction.64
With homogeneous cationic [Rh(NBD)diphos*]+ catalysts,

the rates of hydrogenation of dehydroamino acids were de-

pendent on the catalyst's chelate ring size.

SPIPHOS (4g) > DIOP(+) (g) > (R)-4-Me-Prophos (4 ) 2 (3)-Prophos (Q)

~

The hydrogenation rates of dehydroamino acids with the ten-

+
membered chelate ring's cationic [Rh(NBD)SPIPHOS (Q§)]

107

Table :H).a Hydrggenation Rates of (g)-a-Acetamidocinnamic

 

 

 

 

 

Acid with In situ Rhodium Diphosphine Catalystsc
Diphosphine Hydrogenation Rated Relative Rate
(mLH2/min)

PPh2(CH2)PPh2 O 0
PPh2(CH2)2PPh2 l.l 7O
PPh2(CH2)3PPh2 1.7 100
PPhZCHZOCHZPPh2 3.7 220
DIOP(+) (g) l2.5 740
PPh2(CH2)4PPh2 l7.0 1000
PPh2(CH2)5PPh2 17.0 lOOO
PPh2(CH2)6PPh2 l.5 90

aData obtained from reference 64.

COOH

b(pm-Acetamidocinnamic acid

__J/

CIn situ Catalyst: [RhCl(C2H4)2]2 and diphosphine with
P/Rh ratio of 2.10; [Rh] = 5.0 x 10'3M; substrate/
Rh = 100; P(Hz) z l.l atm; Solvent: C6H6/EtOH_
(l:2)

dMaxium initial rate.

108

catalyst are faster than those with the seven-membered
chelate cationic [Rh(NBD)DIOP(+) (3)]+ catalyst in

the homogeneous state at ambient conditions.

The results seem to contradict Kagan's prediction64.
However, the Rh(I)-SPIPHOS (4g) precursor was in cationic
form instead of in situ form. The catalytic reaction
path may be different with the cationic precursor, re-
sulting in a dramatic change in the catalyst's activity
at ambient conditions.

In the layer silicate system, the factors mentioned
in the homogeneous system are crucial in determining the
hydrogenation rates of dehydroamino acids with [Rh(diene)
diphos*]+ catalysts. Additional factors, viz., the diffu-
sion channel (depending on the polarity of solvent and the
size of the cationic rhodium diphos* catalyst) and the sub-
strate size.

The observed hydrogenation rates of dehydroamino acids
in the intercalated state were slower than those obtained
in the homogeneous state. The rates of the reactions might
be controlled by the rates of diffusion of substrates to
catalytically active sites. The polar solvents are suitable
for highly swelling and desolving the dehydroamino acids.
The interlayer spacing ( OOl spacing) is dependent on the
interlayer swelling and the size of the rhodium diphos*

complex. With the large size of the catalysts, polar solvent

l09

and appr0priate rhodium loading (15-20%) the exchanged
rhodium layer silicate sheets can be highly swollen and
provide appropriate diffusion channels. The substrates
are easily accessible to the active sites and the differ-
entiation of the hydrogenation rates in the homogeneous
and intercalated states can be reduced.

The effect of the catalyst and substrate size on the
hydrogenation rates were observed in the layer silicate
system. The largest size ethyl N-benzoyldehydrophenyl-
alanine ester gave lower hydrogenation rates with all the
intercalated [Rh(NBD)diphos*]+ catalysts. The smallest
size [Rh(NBD)(R)-Prophos (§)]+catalyst resulted in lower
interlayer spacing. The relative ratios of intercalated
to homogeneous rates with [Rh(NBD)(R)-Prophos (_6_)]+ were
lower than those of the DIOP(+) (g) and (3)-4-Me-Prophos
(Q9) systems.

The hydrogenation rates are faster in methanol than
in 95% ethanol in layer silicates. However, the catalysts
seem to be more stereoselective in 95% ethanol than in
methanol in the DIOP(+) (g), (3)-Prophos (Q) and (R)-4-
Me-Prophos (19) systems.

The hydrogenation rates are dependent on the chelate

ring size of the catalyst in the layer silicate system.

DIOP(+) (g) > SPIPHOS (fig) > (3)-4-Me-Prophos (49) > (3)-Prophos (Q)

llO

The rate order is different than that of the homogeneous
system. The rate order of SPIPHOS ({g) and DIOP(+) (g)
is reversed. The hydrogenation rates with intercalated
[Rh(NBD)SPIPHOS ({§)f catalyst are lower"than those ob-
served with intercalated [Rh(NBD)DIOP(+) (§)]+ catalyst.
[Rh(NBD)SPIPHOS .(4~8)]+is highly flexible due to the large
ten-membered chelate ring. The phosphorus atoms of the
ligand could possibly trans-coordinate with rhodium. In
the layer silicates, the flexibility of the Rh(I)-SPIPHOS
(4g) complex may be restricted by the silicate sheets.
The trans-coordinated structure of the complex could
dominate in the interlayer region. However, the trans-
coordination structure is unfavorable to asymmetric
hydrogenation and the bonding structure may need re-
arranging in the asymmetric hydrogenation transition states.
This may lead to low hydrogenation rates and even reverse

the rate order of the catalyst's ring size effect.

lll

I. A Comparison of Intercalated Rh(I)- DIOP (3) System
with Other Rh(I)- DIOP Systems

Kagan32

and his coworkers were the first to synthesize
the polymer attached Rh-DIOP catalyst for asymmetric hydro-
genation. The chiral DIOP ligand (43) was attached to a

cross-linked polystyrene (Merrifield resin).32

The synthetic
methods for the polymer supported Rh DIOP catalyst are

shown in Scheme 24.

OTs I"‘0TS
HICHO + 0\| T50“ “OE—cf 0W
H0 /;10TS\0/-H\/0Ts

 

 

LiPPh2 A <9 G “(OJ/Nam2 [Rh(02H4)2c‘]2 ;
V \o/gvpph:

(43)

0@ cu \R
‘\0: Pth /' \‘C

Il‘u

Scheme 24

llZ

The supported Rh-DIOP catalyst was used as an asymmetric
hydrogenation catalyst to hydrogenate a-ethylstyrene.and
a-phenylacrylic acid to (R)-(-)-2-phenylbutane and (S)-
(+)-methylhydratropate in benzene at 25°C and l atm H2.
The optical yields of (5)-(-)-2-phenylbutane and (§)-(+)-
methyl hydratropate with polymer supported catalyst were rather
low compared with the homogeneous results32 (Table 3l).

Table 3l. Asymmetric Hydrogenation of Prochiral Olefins

with Polymer Supported Rh DIOP Catalyst and
Homogeneous Analogue

 

 

 

 

Substrate Optical Yield (%)a
[Rh(C2H4)2Cl]2 + DIOP
Homogeneous Polymer Attached
__//c2”5
15 (g) 1.5 (3)
‘C H
6 5
__,/CO0CH
fie H 3 7 (g) 2.5 (5)”

_ '_675

 

a Data were obtained from reference 32.
b

27% Chemical conversion

The polymer supported catalyst only swells in nonpolar
solvents and is active for hydrogenation of nonpolar
olefins. However, in benzene-ethanol, the catalyst is

inactive toward the hydrogenation of a-acetaminocinnamic

ll3

acid. The ethanol cosolvent is requried to dissolve the
polar substrates. However, the beads collapsed in the
polar solvents, thus preventing the access of the substrates
to the catalyst sites.

To overcome this polymer swelling pr0blem, Stille
and his co-workers successfully prepared a Rh(I)-DIOP(-)
(if) catalyst on a polar cross-linked copolymer of 2-hy-
droxyethylmethacrylate and p-styryl (DIOP). The synthetic
processes of gel type copolymer supported Rh-DIOP(-) ($5)
catalyst are shown in Scheme 25.

.H\

CHZ\C H§\I 3
/ ‘i 0.92
c
8

 

 

 

‘CD . OZCHZCHZOH
_/ CH3 0.0
0 '\cOZCHzH20H 0 0
*i'-5 -"‘H l% AIBN H ‘”H
T50 OTS T50 0T5
CH Ditosylate copolymer
CH2 ”H2 3
‘CH \\\l ’
C
| 092
COZCHZCHZOH
XS NaPPh
2%: 0.08 [RhCl(C2H4)2]2'
0 =Catalyst
H— ”H
(43)
Ph2P Pth

Scheme 25

114

The ditosylate monomer was copolymerized with hydroxyethyl
methacrylate (HEMA) to form the ditosylate copolymer. The
ditosylate copolymer was phosphinated with excess NaPPh2
in a THF-dioxane mixture to afford phosphinated copolymer
(15). Since the chemical conversion of the phosphination
step was around 50% the final phosphinated copolymer con-
tained bidentate phosphine and monodentate phosphine.
Monodentate phosphine could coordinate to rhodium and lead
to low optical yields for asymmetric hydrogenation of
dehydroamino acids. A high phosphorus to rhodium ratio
(4:l) was required so that rhodium would be complexed mainly
by bidentate phosphines33 However, the complexation of
rhodium by monophosphine was still possible.

Stille and Masuda adopted a similar method to synthe-
size gel type chiral copolymer supported Rh-DIOP(-) (59)

catalyst containing chiral alcohol sites.65 (Scheme 26)

CH

 

{i 2
CH \\CH
*
H- --OH [RhCl(C H ) ]
m E n 2 4 2 2 Catalysts
CH3
\0 (B, or S)
H ouH
th Pth
(50)

Scheme 26

ll5

The ancillary asymmetric center was either the (R) or (S)
secondary alcohol. These catalysts were used to hydro-
genate a-acetamidoacrylic acid, u-acetamidocinnamic acid,
DOPA precursor and atropic acid in benzene/EtOH to the
corresponding (3)-amino acids or hydratropic acid.
The asymmetric hydrogenation products had the same config-
urations and comparable optical yields with those which
were obtained with the homogeneous Rh-DIOP(-) (3) complex.65
The ancillary center had little effect in benzene/EtOH.
Hydrogenation of a-acetamidoacrylic acid in THF gave varied
optical yields (24-40%) of the corresponding amino acid
depending on the configuration of the pendent alcohol group.
The results indicated that the solvent-polymer interactions
dominated the effects of the additional chiral center.65
Table 32 shows a comparison of the results of Rh(I)-
DIOP in the gel type copolymers, hectorite and homogeneous
systems. Stille and his co-workers develOped two gel type
c0polymer systems (i4, 51) )33’65 which proved to be
better polymer support systems for asymmetric hydrogenation
of dehydroamino acids than the cross-linked polystyrene
support system developed earlier by Kagan and co-workers.
With the first copolymer supported catalyst ({5)33 (without
chiral alcohol sites), only two prochiral amino acids were
used in the study. The optical yield of alanine was around
52-60%, less than the homogeneous result. The optical
yield of phenylalanine was better than the homogeneous

yield. In the second type of copolymer system,65 the

ll6

 

mzo u<o

 

 

 

 

 

 

 

. I
IIIIIIZ O
/1\
I.II I II II . II .58)
II
I 0v
IIIIIz/uu
II I II .II II .5qu
IIIIIIZ
III II II II II . IIII\.1
IIIIIIz
/I'
II-II IIII II II II IIII\
SEEISIIIIIBII as 383
smaxpoaoo FILIgu umguouu< gmeapoaou Izomcmmoso: IIIIPIULIIIH Izomcwmoeo: mumgumnam
IIIIII + IHIIIIIIIIIIII II +23 EIIIISIEIIT
Ismumxm

83-55. .855 5.; 8:38 53% +2.3 EIIIIIIIzES 3:39.35 2: III IEI; IIIBII .NI I22

ll7

. co wmgw>cou poo wEwgu RONU

.IIIIIIL3mwwcouuamv II;
mauzuoga Isa p—< .mm.%mmmapm> ILIIIIIIII III II_III IIIIIII och .ocpiom ucaogm II: IIIII Ezwuocg
.53. I

II muogumnsm III mcqu>poI II» II zo»m\ozoo sup: IIII :I_IIELII III: Iswumxm IIIFIIII mmmcpn

5832:328va III 3383 2: E .822
uczogm II: Ezwuozg on mumgumnzm Io ovum; LIIoE ugh .zoum III II: gomgzumgq Iwgu Low EIIIIE IIIII>IomI

 

.IIIISIII . NI I23

ll8

copolymer supported catalysts (39) contained chiral alcohol
sites. With the catalysts supported on the chiral copoly-
mers the optical yields of analine and phenylalanine were
slightly better than those obtained with the homogeneous
analogue in benzene and ethanol mixture solvent.

The optical yields of tyrosine and DOPA were lower than
those obtained with the homogeneous analogue. The chiral
sites copolymer supported catalysts (69) were sensitive to
solvent, in some cases leading to low conversions!55 In the
Rh(I)-DIOP(+) (3)-hectortie system, the optical yields

of four substrates deviated only slightly from those ob-
tained with the homogeneous analogue. Moreover, immoboli-
zation of the asymmetric hydrogenation catalyst was easily
achieved by a simple cationic exchange process.

In general, the hydrogenation rates of dehydroamino
acids with gel type copolymer supported Rh(I)-DIOP(-)
catalysts (435.5,?) were slower than those observed with the
homogeneous analogue.33’65 It took 2-24 hours to complete
the reactions with the substrate to rhodium ratio around
50--100.33’65 In the layer silicate system hydrogenation
rates of dehydroamino acids with the intercalated [Rh(NBD)
DIOP(+) (2)]+ catalyst were 0.40 to 0.90 as fast as those
observed with the homogeneous analogue. Most of the reactions

could be completed in 1 hr with the substrate to rhodium

ratio around 100-200.

119

From the discussion of the previous results it can be
concluded that the choice of the polymer matrix and the
synthesis of the catalyst site in the matrix are crucial
for the successful use of the polymer supported rhodium
complexes as efficient asymmetric hydrogenation catalysts.33"65
However, through simple cationic exchange, the preparation
of intercalation catalysts can be achieved easily without
long synthetic processing. Also, intercalation catalysts
are highly swollen in polar solvents, and substrates readily
are accessible to the catalytic sites. By using interca-
lated [Rh(NBD)DIOP(+) (3)]+ catalyst , one can obtain op-
tical yields of asymmetric hydrogenated products that are
comparable to the homogeneous results or even better. The-
hydrogenation rates of dehydroamino acids with intercalated
[Rh(NBD)DIOP(+) (3)]+ catalyst were faster than those
observed with gel type copolymer supported Rh(I)-DIOP(-)

catalysts.

J. The Probable Asymmetric Hydrogenation Mechanism
31P NMR Spectra of the Solution Structures of (R)-

4-Me-Prophos (43) and its Rhodium Complexes

There have been great advances in the elucidation of
asymmetric hydrogenation mechanism through the use of NMR
and X-ray structure studies. A knowledge of the structures
of the reaction intermediates is essential for the eluci-
dation of asymmetric hydrogenation mechanism and for the

improvement of catalytic systems for asymmetric hydrogenation.

120

Comprehensive studies on the mechanism of asymmetric

hydrogenation of dehydroamino acids with cationic rhodium
diphosphine catalysts have focused on the steps38’ 40’ 66
in the catalytic cycle and the origins of chiral dis-
crimination.62 The "unsaturate route" which was proposed

by Halpern 39

and co-workers is the most common asymmetric
hydrogenation mechanism for most cationic Rh(I)-chiral
diphosphine complex systems at low hydrogenation pressure.
The catalytic cycle of asymmetric hydrogenation by the
unsaturate route is shown in Figure l2. In this mechanism,
the complexation of olefin proceeds the addition of hydrogen.
Halpern and co-workers found definite evidence to support

39

it. The cationic [Rh(NBD)diphos*]+ complex was hydro-

genated in methanol to give a complex containing chelate

39 Similar evidence

diphosphine and solvent but no hydrogen.
was found by Slack67 and Baird using a variety 0f
chelating diphosphines. All the rhodium diphosphine com-
plexes formed solvent adducts rather than hydride complexes
on hydrogenation in polar solvents.67 The solvent adducts
of rhodium diphos* complexes react with dehydroamino acid
(enamide) to form cationic [thiphos*enamide]+ complexes.
The enamides are capable of coordinating with rhodium via
olefin and amide carbonyl residues to form a rigid chelate,
square-planar complex. The binding mode was first inferred

36 The X-ray

37

from asymmetric hydrogenation data by Kagan.

crystal structure evidence was found by Halpern and

121

 

[Rh(diphos*)(diene)]+
2H2 '
, ; [Rh(diphos*)] /"'\ a
R026 CH2R /R NHCOR
RI.
# 45’ l ,.‘.O==” .
(diphos1HRh\c’NH (diphos*) Rh.’
-. RC l:
1902cl 04:9 ozj R

-+
[RhH2(diphos*)enamide1

Figure l2. Asymmetric hydrogenation catalytic cycle
using the unsaturated route

122

co-workers. Also, Brown and Chaloner have proved that

31

the same structure was maintained in solution by P and

13C NMR spectroscopy38. The cationic [Rh(diphos*)enamide]+
complex reacts further with H2, presumably to form the
dihydride-enamide intermediate. The hydride transfers from
the rhodium to the coordinated face of the olefin, forming
a monohydrido-o-alkyl intermediate, which undergoes reduc-
tive elimination to produce the corresponding amino acid and
regenerate the catalyst. The monohydrido-o-alkyl inter-
mediate has been detected and its structure has been deduced
by NMR

The asymmetric induction step can be either the dis-
placement of the solvent by enamide, or the addition of
hydrogen or hydride insertion, depending on the individual
case. Thermodynamic and kinetic factors of asymmetric hy-
drogenation are important for determining the asymmetric
induction step. The probable mechanism of asymmetric
hydrogenation of dehydroamino acids using cationic rhodium
(I) chiral diphosphine catalysts is shown in Figure 13.
Usually two diastereomeric rhodium(I) enamide complexes
can be observed, with one predominating in most instances.

40’66’68 equilbrium of two diastere-

The attainment of the Kd
omers in most chiral diphosphine systems is rapid at ambient
conditions (FigurelB). However, Kd is kinetically irrele-
vant in deciding the optical yield. The optical yield is

decided by the different reaction rates of two diastereomeric

123

=C’R"

, o
(diphos*)Rhtq; J” . _g ”1:0 ‘Rh (diphos*)

“'02:ij i ”NRC“? R'

Kd IIR \

 

 

H . '
2 k1 k 1 H2
. . +
(RhH2(enam1de)d1phos*] [RhH2(enamide)diphos*]+
k k'
2 ‘2
’/fi§fi;fiu R"\C¢0\
* \
(diphos )liRh / Rh+H(diphos*)
\C/NH HN\C/
.4/ \\C
R‘20C CH2R RCH2/\COZR'
k3 k3
[Rh(diphos*)]+
H4 NHCOR" " H
'C’/ R OCNH ,\
. I \ /\c '
R0 0 CH R ‘~
2 2 . RCH c02R

(g) (3)

Figure N3. The probable mechanism of asymmetric hydrogenation of
dehydroamino acids (enamides) using cationic rhodium(I)
chiral diphosphine catalysts

124

rhodium enamide intermediates. These rates are determined
by the relative energies of two diastereomeric transition
states and are not reliant on the initial reactant energies‘5
(Curtin-Hammett Principle). Also, Bosnich 62believes that
the origins of the dissymmetric rate differentiation must
reside in the different dissymmetric interactions of the
chiral diphosphine ligand and the prochiral substrate in the
diastereomeric transition states of the reaction.

In the rigid-type diphosphine systems at ambient tem-
perature and pressure, the rate determining step is presuma-
bably the oxidative addition of hydrogen (k , ki). The
hydride insertion step (k2, ké) is assumed to be faster.

At low temperature (below -40°C) and ambient pressure, the
rate determining step would shift to reductive elimination
(k3. ké). The evidence for the monohydrido-o-alkyl inter-

66a and Brown.66b In

40

mediate has been obtained by Halpern
the case of (§,§)-Chiraphos (i), Halpern reported that
the minor diastereomeric Rh(I)-enamide (§,§)-Chiraphos (2)
complex is the true intermediate which added hydrogen much
faster than the major intermediate at ambient conditions.
The result indicates that the rate constant (kf) for
hydrogenation of the minor diastereomer is more than 3 orders
of magnitude faster than the rate constant (k]) for the
major diastereomer. 1

In the layer silicate system, at ambient conditions,

the rate determining step is expected to be the same as that

of the homogeneous state. This means that the oxidative

125

addition of hydrogen (kai) is the rate determining step.
Since the interactions between the silicate sheets and dias-
tereomeric [Rh(diphos*)enamide]+ complexes take place, the
relative energies of two diastereomeric transition states

in the intercalation state may vary from those of the homo-
geneous state, resulting in differentiation in the hydrogena-
tion rates of the two diastereomers. Also, changing the
ratio of two diastereomers is possible in the intercalation
state. However, changing the ratio of two diastereomers

may not necessarily alter the optical yield. The effect

on the optical yield in changing the rates of two diastereo-
mers may be the most important consideration in the inter-
calation state. Optical yield varation of the corresponding
amino acids obtained with Rh(diphos*) complex in the homo-

geneous and intercalation states should be expected.

There is still a lack of sufficient evidence to
determine the cause of the asymmetric induction step in
most diphosphine systems, however. Along with a considera-
tion of optical yields of the products and X-ray structures

31P NMR spectroscopy of the solution

of intermediates,
structuresof asymmetric hydrogenation intermediates in
each chiral diphosphine system can provide crucial
evidence for elucidating the induction step.
(R)-4-Me-Prophos ($9) is a new ligand. Studies of
31P NMR spectra of (3)-4-Me-Prophos (I?) may provide

useful information to clarify the reaction mechanism and

elucidate the asymmetric induction step. The free ligand

126

has two nonequivalent phosphorous atoms, and they are

31

coupled with each other. The P NMR spectrum shows

31

two sets of doublets (FigurelA). The P NMR specturm

of a 0.01 M solution of [Rh(NBD)(g)-4-Me-Prophos (Q9)]+Cl04

-n1<xm13at -16°C is shown in FigurelS. Norboradiene (N30)

is a symmetrical molecule, and two nonequivalent phos-

103

phorous atoms couple with each other and with Rh

nucleus (1 = I, natural abundance = 100%). The spectrum

shows two sets of four lines. It is an ABX type spectrum.

31

The P NMR spectrwnof [Rh(NBD)(R)-4-Me-Prophos (43)]+Cl0

31

4

has the same features as the P NMR spectrum of [Rh

(NBD)(R)-Prophos (6)]+C104 (Table33 ).

The 3]

P NMR spectrum of 0.0l MERh(I)(g-ester)(3)-
4-Me-Prophos (figfl+in EtOD in the presence of an 8 molar
excess of ethyl (;)-a-benzamodocinnamate (L-ester) is
shown in Figure16. The chemical shifts and coupling
constants of [Rh(g-ester)(3)-4-Me-Prophos (43)]+ are
shown in Table 33. Since the (3)-4-Me-Prophos (A?) and
(§)-Phenphos (Z) are analogues, both ligands are chiral.
There are four possible Rh enamide complexes for each
ligand. Either face of the enamide may be cis or trans

13

to P](Ar2PCH2). Brown and Chaloner. have used C

enriched methyl (;)-a-acetamidocinnamate (enriched amide
carbon) and dideuterated (§)-Phenphos (-CDZ-PAr2) to
study the solution structures of [Rh(g-ester)(§)-phenyl-

31 12

phos (1)]+ complexes by P NMR spectrosc0py. The

127

IIII II IIIII II ImIV IIIIII_-I2-I-IIV IIIIIIIIII III III III .IIIIIIII
IIIIIIIIIIV

(IIIIgI

Ema . Cal m7. 0.1 n1 o
1 _ I I _ .

 

 

 

 

Gum I.

IIII IIIIIII-II-I-IIV

128

 

 

 

 

 

 

 

 

. IIII- II IIIII II IIII
+HIIII IIIIIIATI2-I-IIVAIIZIIIH II IIIIIIII z _I.I IIISIIIIII Izz III III .IIIIIIII
. IIIIIIIIIIV

can. 9. II 9 II 9 II
11. 1 I _ _ I _ _ .

f, If; ‘25 I _ I a. E

I: III" I-III .

I.III

 

h.—v

IIII+IImII IIIIIII-II-I-ImIIIIzIIII

Table 33. 3‘

129

P NMR Parameters of Chira] Diphosphines (L)

and Their Rhodium(I) Compiexesa

 

Chemica] Shifts and Coupling Constants
(B)-Prophos (6)

(3)-4-Me-Prophos (49)

 

free ligand

[(NBD)Rh L]+

[g-ester Rh L]+b

2.1 ppm, -20.2 ppm,

JP_P=21 Hz
(CDC13, 25°C)

61.6 ppm (dd,
JRh_P=155 Hz),

42.9 ppm (dd,
JRh_P=155 Hz),
JP_P=34 Hz;

(CDC13, -20°C)

0
JP_P=21 Hz
(CDC13, 25°C)

60.4 ppm (dd,
JRh_P=155 Hz),

41.7 ppm (dd,
JRh_P=155 Hz),

JP_P=35 Hz;

(CDC13, -16°C)

74.9 ppm (dd.
JRh_P=162 Hz),

46.7 ppm (dd,
JRh_P=155 Hz),

J =46 Hz;

P-P
64.9 ppm (dd.
JRh-P=155 Hz),

59.7 ppm (dd,
JRh_P=158 Hz),

J =46 Hz;

P-P
(CH3OD,-30°C)

 

a72.88 MHz, 85% H P04 as externa] reference; downfieid

shifts positive.
g-ester = Etheyi-(g)-a-benzamidocinnamate

b

.3 ppm, -22.1 ppm

130

goon- um oomxu =_ cope

+5.: 8.32324-3324}: .3 z 8.9 to 5.2 a; 2: .223:

Eng 3 on 2 co 3 on «$32.53

§§ .. s .

 

 

 

~.m¢

 

 

 

 

 

 

 

 

 

 

n .1:
= H .¢ :3

 

a an

Wizouogu: .<
.o_o .=x~§ \.HHW«
O... .39 8:: m x

at» +~.<

 

“.mm m.¢o mews

copu+namwv mogqoga-m‘-¢-AmVAgupmm-mv;aH

131

spectrum pattern is the same as that of [Rh(g-ester)

(gl-Phenphos (1)]f complexlz.

It demonstrated that the
[Rh(L—ester)(fi)-4-Me-Prophos (49)]+ complex only has two
geometric isomers in the solution and that they are not
diastereomers. The result implies that [Rh(I)(3)-4-Me-
prophos (49l]+ complex is capable of enatioface differen-
tiation of enamide to form preferential Rh(I)-enamide
(3)-4-Me-Prophos (49) diastereomer. The major diastere-
omer is possibly the true intermediate. However, further

X-ray structure evidence of [Rh(I)-(g)-4-Me-Prophos (4?)

enamide]+ complex is needed to clarify this point.

132

11. Synthesis and Characterization of New Functionalized
Chiral Diphosphine Ligands

There have been significant advances in asymmetric

hydrogenation of prochiral olefins with chiral rhodium

phosphine catalysts"2

1963.3'5

since its original development in

Optical yields over 90% are possible with certain pro-

chiral substrates.2b

Chiral chelating phosphines play a
key role in asymmetric induction of the rhodium asymmetric
hydrogenation systems. The chiral diphosphine ligands
(g-Lg) which can coordinate with rhodium and form rigid
five-membered chelating ring complexes, usually give high

9-15

optical yields with (%)-dehydroamino acids. The chiral

diphosphine ligands which had the third functional groups

2b,22-24 A

were more useful for asymmetric synthesis,
series of pyrrolidinephosphines such as BPPM (Z}), PPPM (Z4),
PPM (g6) and their analogues which have third functional
groups (the amide substituents) have been successfully used

as ligands for a variety of rhodium catalyzed reactions2b

P(‘3cs”5)2

A
N
w

v
30

I

~ - C00C4H9

;l: (25) R
l
R

-tert BPPM

H PPM

133

The amide group of ligand (a3) was modified by different

substituents in order to optimize the optical yields2b

(see pages 20 to 24)to chemically clarify the reaction inter-

70 34

mediates, or to attach to the polymer . Ferrocenyl-

phosphines (17-19) which have planar chirality and various
functional groups such as amino or hydroxyl groups have

been demonstrated to be effective ligands for rhodium

catalyzed asymmetric hydrogenation of dehydroamino acids,22

ketones23

and imines24 (see pages 24 to 29 ). The inter-
actions between the substrates and the third functional
groups of the ligands in transition states are crucial for

highly asymmetric inductions of these catalytic reactions.”-24

(§)-(E)-BPPFA (g7)

 

 

(5)-(§)-BPPFA (2..) ,. (3)-(_§)-BPPFOH (g9)

134

The design of new rigid functionalized chiral
diphosphines may provide the following advantages. The
ligands, being rigid, can be highly effective for asymmetric
hydrogenation of dehydroamino acids. The third functional
group can be modified to optimize the optical yields and
can be attached to the polymeric, organic or inorganic
carriers. Also, the ligands have nonequivalent phosphorus.
They can be used as spectroscopic probes for studying
the reaction mechanism. The new THF-butaphos (1]) (3)-
hydroxylbutaphos (5g) and N—BOC-butaphos (5}) which were
synthesized from natural occuring L-malic acid are in this

catagory.

H
\
\
./
PAr2 PAr2

R = 0THP, Ar = C6H THF-butaphos (i1)

59
R = OH, Ar = C6H5, (3)-hydroxylbutaphos (6g)
R = NCHZCH(CH3)2, Ar = 4-CH3-C6H4

COO-C Hg-tert NeBOC-butaphos (sg)

4

In this part the focus was on the synthesis and char-
acterization of new functionalized chiral diphosphine ligands
(6], 5g, sg), and the application of the new, N-BOC-butaphos
(6g) ligand for asymmetric hydrogenation of (g)-dehydroamino

acids in homogeneous and intercalated systems.

l35
'A. Synthesis of the Key Synthetic Precursor

The key synthetic precursor, l,2-0-i50propylidene-
l,2,4-butanetriol [acetonide-triol (S§) 1 was prepared in
60% overall yield from L-malic acid.71The synthetic route
is shown in Scheme 27. Esterfication of L-malic acid
by stirring in 3% HCl-MeOH solution overnight gave (S)-
di-methyl malate (55) in 94% yield. Reaction of (§)-di-
methyl malate (S4) with l.l equivalent of dihydropyran
(DHP) and a catalytic amount of p-toluenesulfonic acid in
ether at room temperature for l6 hours formed the corre-
sponding THP ether (66) in 91% yield. Reducing the
corresponding THP-ether (66) with excess lithium aluminum
hydride in dry ether produced the diol THP ether (66) in
93% yield. Removal of the THP protecting group by stirring
with a catalytic amount of p-toluenesulfonic acid in
methanol for 16 hours afforded the triol (6]) in 97% yield.
Triol (6]) was converted to acetonide-triol (Sg) in 78%
yield by treatment with a catalytic amount of grtoluene-
sulfonic acid in dry acetone for two days.

The synthetic precursor, acetonide-triol (sg) had
three hydroxyl groups. The two hydroxyl groups at the 1,2
positions of acetonide-triol (sg) which were protected
by isopropylidene group could be transformed into diphos-
phine. The third hydroxyl group could be functionalized.
However, due to the limitation of the synthetic method,

it is crucial to follow closely the proposed reaction

136

 

H°><i 3% HCl/MeOH ”0 ,x“

H0 c co H = MeO c co Me
2 2 2 2
L-Malic acid (5,3)

DHP lether THPO 1H

(5)

~

 

 

 

 

LiA1H4/ether' : THPO ’H
H0 ’
0H
(5f)
MeOH/TsOH H0>yl
g: [I
H0 V\0H
(5])
Acetone/TsOH )Vo . ,H
‘7 I
0>\/\
OH
(5~8)

Scheme 27

137

sequences to achieve the synthetic goals.

B. Synthesis of 4- (l’-Tetrahydropyranloxy)- l ,2-(R)- bis(diphenyl-
phosphino)butane, [THP- butaphos (Sl)] and 4- -Hydroxy -l ,2 U0-
bis(diphenylphosphino)butane[(R)- -Hydroxylbutaphos (52)]

The l,2-0-i50propylidene-l,2-(§)-4-butanetriol [ace-
tonide-triol (68)] is a key synthetic precursor for syn-

thesis of THF-butaphos (6]) and (5)-Hydroxylbutaphos (62).

The synthetic route is shown in Scheme fli The acetonide-

triol (68) was reacted with NaH and benzyl chloride in

_toluene at 70°C and followed by acidic hydrolysis in the

mixted solvent (HAc:THF:H20) at 45°C to give benzyl-diol

(69) in 84% yield. The benzyl-diol (69) was converted

into benzylether ditosylate (69) in 9l% yield by reaction

with toluenesulfonyl chloride in pyridine. The benzyl-

ether ditosylate (69) was hydrogenolized over 10% Pd/C

to give hydroxylditosylate (63) in 98% yield. The hydroxyl-

ditosylate (63) was converted quantitatively into THP-ether-

ditosylate (62) by treatment with dihydropyran and a catalytic
amount of toluenesulfonic acid in ether. The THP-ether-
ditosylate (63) was treated with a stoichiometric amount

_ of LiP(C6H5)2 to afford the crude product of THP-butaphos

(6]). The final product was difficult to crystallize.

3‘? NMR,

The purity of the product, which was determined by
was larger than 90%. The (3)-Hydroxylbutaphos (62) was
obtained by acidic hydrolysis of THF-butaphos (61) with

a catalytic amount of toluenesulfonyl acid in ethanol.

TsC1/PY

OTs

TSO\X//\
0H

(6,1)

y
7

H

LiPth/THF

 

Eton/H+

V

138

H

1. C6H5CH2C1/NaH H 0\/H

 

2. Acidic hydrolysis. Q"’\V"~ocnz<::>

 

 

(69)
T50 015 ,’H l0%Pd/C _
(69)
OTs H
DHP/ether TsQ\\:&/’
: 0THP
(6;)
H
\\
/ \
Pth PPhZ

(6]) THF-butaphos

CH 0H

‘- ;CH2 2

{R
2
2

CH

/
.. PPh PPh2

(5 ) (3)-Hydroxylbutaphos

~

Scheme 28

139

The third hydroxyl group of the acetonide-triol (68)
was protected by benzy]ation. It was necessary to replace
the benzyl protecting group by the THP group before phos-
phination. If phosphination were to preceed debenzylation,
the final debenzylation reaction, which requires the use
of Pd/C as a catalyst, could permit reaction of the ligand
with palladium and form a palladium complex. This un-
desirable side reaction could cause stoppage of the reaction
or purification problems. However, the THP protecting group
is unaffected by the phosphide reagent and can be removed
easily by hydrolysis.

C. Characterization of THP-butaphos (6]) and (3)-Hydroxyl-
butaphos (62)

Figure17 shows the 3]

P NMR spectrum of the THP-butaphos
(51) ligand. The ligand (6]) has two diastereomers and the
nonequivalent phosphorous atoms are spin-coupled. The
spectrum is supposed to show four sets of doublets, but

two sets of resonances overlap each other. Therefore, the

spectrum appears as two sets of triplets.

H .

\\ * *
CHZ— C‘CHZCH20<<;>
/ \

PPhZ .PPhZ (3.5 or R,§)

0".

THP-butaphos (6])

 

140

 

 

 

 

 

 

9:3 Z .388 5 :23 8583325352383 ”.22 a; 2: .22sz
MP :30
pan—Q cu...- n—I lopl ml 2: .W 3
.— 1 — 11 q d.
. \C‘. _'
£92”?qu
a ..
.an/ «in
a.
2.8 16.10....» .fo
-
I 1....
:1
4....
Q
In.
so?
.8?
.

:23 3.33325

141

Upon acidic hydrolysis, the THP protecting group was
removed, resulting in the destruction of two diastereomers,
and the conversion of the ligand (6]) into (R)-Hydroxyl-

31

butaphos (62). This was confirmed by P NMR spectra.

31P

(3)-Hydroxylbutaphos (68) only had one isomer. The
NMR of (3)-Hydroxylbutaphos (63) only showed two sets of
doublets (Figure18). The results confirmed the predictions.
The chemical shifts of (3)-Hydroxylbutaphos (68) were similar
to those of THP-butaphos (6]). The spectroscopic evidence of
(3)-Hydroxylbutaphos (68) proved that the THP-butaphos

(6]) had two diastereomers and no impurities.

- “tart-Butyloxy)carbonyl}isobutylamino}-I

D. Synthesis of 4 (
s 63-4-methylphenylphosphino)butane [N-BOC-

1,2-(R)-bi (
butaphos (68
l,2-0-Isopropylidene-l,2-(§)-4-butanetriol [acetonide-
triol (68)] is the key synthetic precursor for the synthesis

of N-BOC-butaphos (68). The synthetic route is shown in
Scheme 29.

/

The acetonide-triol (68) was converted into (6)-
tosylate (68) with p-toluenesulfonic chloride in pyridine
in 81% yields. The displacement of the (6)-tosy1ate (68)
in excess isobutylamine produced the corresponding amine
(64) in 93% yield. Acidichydrolysis of the corresponding
amine furnished the aminodiol (65) in 78% yield. The
aminodiol (68) was treated with di-tert-butyl dicarbonate

to form the N-BOC-aminodiol (68) in quantitative yield.

142

9.3 “a so: 5 Am: mofififiixcééAmv .co .5533 5;. a; .2233“.
T 123.253
. _

ran—a on!

Es$§g§§§§§a ,

NIoéunfl And :a

 

 

You...

n pl O—l

«can _ Foo
/ \

Iouzouzo I! my 1m: 0

.
.
I

Amwv mogampan—axoguxzuamv

   

143

 

 

 

°~\685~<;:L.\\ TsCl/PY o,‘3:><15’,\\ isobutylamine__
OTs AT
OH
(é?) (Q?)
0H
-(1)—o 1,H MeOH/H+ HO_ ”,H
V\ _:
NHCHZCH(CH3)2 [1012mm 312
(64) (5,5)
OH
‘ HO ,/H
. TsCl/PY
0(C02C4H9-t)24: HCHZCH(CH3)2 >_
(59
T50

312

‘
r

”0%;
' ’ LiP(C H CH
HCHZCH(CH3)2 6 4

(5,7)
H
CHé—DC“CH2CH2R R - I‘VCHZCH(CH3)2
/ \ COZC4H9-tert
PAr PAr
2 2 -
Ar - 4-CH3C6H4

.N-BOC-butaphos (68)
Scheme 29

144

N—BOC-aminodiol (66) was converted into N-BOC-amino-
ditosylate (6]) withID-toluenesulfonic chloride in pyridine
in 89% yield. Finally, phosphination of the N-BOC-amino-
ditosylate (6]) with a stoichiometric amount of lithium
bis(4-methylphenyl)phosphide in THF afforded the crude
oily product. Crystallization of the crude product from
95% EtOH produced pure N-BOC-butaphos (68).

The third unprotected hydroxyl group of acetonide-
triol (68) was first transformed into isobutylamino group.
After acidic hydrolysis of the acetonide group, the iso-
butylamino group was protected by the BOC group to prevent
tosylation and phosphination in the following reaction
sequences.

Also, BOC protecting group can be removed rapidly by
mildacid. Variation of amide substituent may alter op-
tical yields. The 80C functionality may also be useful
for attachment of the ligand to organic or inorganic
carriers.

E. Characterization of 4-“(tert-Butyloxy)carbonyU—isobutyl-
aminoyl,2-(B)-bis(di-4-methy1phenylphosphino)butane
[N-BOC-butaphos (68)]

The four methyl groups of 4-methy1pheny1 rings of
N-BOC-butaphos (68) are not in identical environments.

Two methyl groups, however, have the same 1H NMR chemical

shifts. The other two methyls gave different chemical

1

shifts. Thus, the H NMR spectrum (Figurel9) exhibits three

145

.ANIZ ommv 00mm “a .mpuou c. wav mogaapsn-oom-e co sztyomam as: I.
mzp . .

A59: 0 P N m e . m c m m

qdduddedd-—141Jddfi-<—qddqdquddddddifidddqdddddqdddd—ddddddddd—dudddqdufiddfiqdddddd‘u

.mp at=m_a

 

 

 

 

 

 

 

Acovpapow umuzppuv
Amwv mosaapsa-uom-z

 

146

distinguishable singlet peaks in the range of 2.1 - 2.5.
ppm. The spectrum of the methyl groups on the 4-methy1-
phenyl rings show the same characteristics as (3)-4-Me-
Prophos (48). The amide groups (N-BOC) of the ligand (68)
may not rotatefreely, resulting in conformational isomers.

This could be confirmed by 1H and 3‘

P NMR. When the con-
centration of N-BOC-butaphos (68) is increased, the three.
distinguisable methyl lines at 2.2 - 2.5 ppm split into three

sets of doublets (Figure an). The 3]

P NMR spectrum shows
two small shoulders at the low-field side of each major
doublet peak (-4.4 and -22.6 ppm) (Figure21 ). The compound
is difficult to crystallize from solvents other than 95%
ethanol.’ The compound was isolated as the monohydrate

The mass spectrum and high resolution mass spectrum gave

1 31

the expected parent peak. The H and P NMR spectra proved

that the compound was notcxnfiaminated by phosphine oxide.

1H

If the compound was oxidized to phosphine oxide, the
NMR spectra of the orduiand meta hydrogen peaks of the
phenyl groups of the ligand would split into two sets of

complex peaks (Figure20 ).

147

mze A~=z omNV comm we .mpuau :3 Away coeeepzb-uom-2 co Escpeeam m2: =_ . om ec.;weu
Asaav A. _ N m e a c A o . n

q—dddqdd4d4~+«444dfiddi—qddqq‘«ddddddqddddq‘dqddqfilus—d~4~dd~dd~<«1««did—444dqddddC—dqd—dqu—dI—Jd

I
P I I 1.11 1‘ I1 I- . I
I11 (11 ‘1 1| 1
‘1 ‘1 H

 

 

 

 

 

 

 

 

 

 

A=o_a:pom vauogucoucouv
Amwv mecaegsa-uom-z

 

148

.m

coo pm 58 5 2W3 mogamuaniuomiz mo .5533 ”.22 a; . 3 6.53...
m:
r Ema mm! on! 21 0.1 ml 0 2: L563
1 . H d - _ _

 

excunxule H .<

N IT NOW N153 H... . e.e. . q
fimxuvzufuzuz

 

 

 

 

 

 

Nh<m “—(l
a / u \
$5 1611:“
r
«*1
QNNI

Ami 8233-884

149

F. The Observed Hydrogenation Rates and Optical Yields

in the N-BOC-butaphos (68) System

The hydrogenation uptake plots for reduction of N-acetyl-
dehydrophenylalanine with homogeneous and intercalated
[Rh(NBO)N-BOC-butaphos (68)]+ catalysts are shown in figure
22 . .The observed rates for hydrogenation of dehydroamino
acids with homogeneous and intercalated [Rh(NBD)N—BOC-
butaphos (68)]+ catalysts are shown in Table 34. The hy-
drogenation rates with homogeneous catalyst in THF were
2 - 3.6 as fast as those observed in 95% ethanol. The hy-
drogenation rates with intercalated catalyst were 0.09 to
0.55 as fast as those with the homogeneous analogue in 95%EflHL

The results of asymmetric. hydrogenation of (g)-dehydro-
amino acids with homogeneous [Rh(NBD)-N-BOC-butaphos (68)]+
catalyst in different solvents at ambient temperature and
pressure are shown in Table 35. In the 95% EtOH solvent
system, the optical yields of the corresponding (6)-a-
amino acids obtained with this catalyst were around 85 -
96%. The optical yields of N-acetylphenylalanine, N-ben-
zoylphenylalanine and N,0~diacety1tyrosine and DOPA were
around 85%. The optical yield of N-benzoylphenylalanine
ethyl ester was 96%. In the dry THF solvent system the
optical yields of the corresponding (§)-a-amino acids were
around 90-96%. The optical yields of N-acetylphenylalanine
and N-benzoylphenylalanine ethyl ester were around 96%.

(The optical yields of N,0-diacetyltyrosine and DOPA were

Figure 22.

150

Hydrogen uptake plots for reduction of
5 mmol of N-acetyldehydrophenylalanine
in 30 mL solvent at 25°C and 740 torr
with [Rh(NBD) N-BOC-butaphos (53)]+

-as the catalyst precursor. (A) Homo-

geneous catalyst in THF (B) Homogeneous
catalyst in 95% ethanol (C) Intercalated
catalyst in 95% ethanol. In each case, the
amount of rhodium complex used was 0.03 mmol.

 

lSl

mcmevmerv oer

,

 

 

oer

152

.mm ucm mm mmpamp cw cm>_m omega mo mama on» «gm mcopuwucou cowuommg onhm .

 

 

 

mzuo u<
. mzoouzz
mo 0 «.mm o.m no.
zoou\1.
u<o
m on mzooo=z
111 . .111 mm_
=OOU\11.
. mzouoozz nu
mm o o.m_ m.o_ _.mm m N \1.
z 88
. mxououxz no
Fe o o.mm m.¢_ .1111
zoOU\11
Rm.o o no .mzoouzz
. o.m~ mm.
zoou\11
mama m>wumpmm .comoso: .—ougma:~ .zmmoso: ouogumnzm
noun Nmm uxp

 

 

cacm poss\:vs\~: paw mmumm :omogux:

 

mama . . . z
pmca mzoocmmcaox muH can “mapaumc +mfimmv mogaau331ccm15~Aamzvgmgcoua_augmuc_ upcmmuam =o_pa=omogux: .em «_nah

153

Table35 . Asymmetric Hydrogenation of Dehydroamino Acids
Wlth Homogeneous [Rh(NBD)N-BOC-butaphos (53)]+

Catalyst

 

Substrate

6NHCOC6H5

z<mooc2H5
6 NHCOCGHS

_/coou

@‘NHCOCH3

OAc

_/COOH

Qfiuucocg
Ac OCH

3

—— Optical Yield (%)
95% EtOH

Amino acid

 

L-Phenylalanine

L-Phenylalanine

L-Phenylalanine

L-Tyrosine

L- DOPA

86

86

94

85

86

a

lflE

95

96

9O

91

 

aThe hydrogenations were carried out in 30 mL of 95% EtOH or THF on 4-5
mmol of substrates at 25°C and 740 orr.
homogeneous catalyst were 1.0 X l0- M. All the corresponding amino
acids had (S)-configurations.

The concentrations of the

154

90%. The catalyst was more efficient and gave better op-
tical yields in THF than in 95% EtOH.

The results of asymmetric hydrogenation of (g)-dehydro-
amino acids with intercalated and homogeneous [Rh(NBD)-
N-BOC-butaphos (5})J1 catalysts in 95% EtOH are shown in
Table 36. The optical yields obtained with intercalated
catalyst were comparable to or even better than those
obtained with the homogeneous catalyst. With the inter-
calated [Rh(NBD)N-BOC-butaphos (553)]+ catalyst the optical
yields of N-acetylphenylalahine and L-DORA were better than
the homogeneous results. The optical yields of N-benzoyl-
phenylalanine and N-benzoylphenylalanine ethyl ester were
the same as the homogeneous results.

(3)-Prophos (g) has proved to be a highly effective
ligand for asymmetric hydrogenation of dehydroamino acids.10
The optical yields of the corresponding amino acids obtained
with homogeneous [Rh(NBD)(3)-Prophos (6)]+ catalyst were

10

around 90 i 3% in THF and 95% EtOH. The optical yields

appear to be less sensitive to the nature of the substi-

‘0 N-BOC-butaphos

tuent on the substrate and the solvent.
(5}) and (3)-Prophos (g) are analogues As with R-Prophos
(Q) , N-BOC-butaphos (5}) is an effective ligand, giving
high optical yields (BS-96%). However, the optical yields
obtained with [Rh(NBD)éW-BOC-butaphos (53)]+ catalyst are
varied by changing the solvent. In the protic solvent

(95% ethanol) the amide carbonyl group of the ligand

155

Tabler36. Asymmetric Hydrogenation of Dehydroamino Acids
with Intercalated and Homogeneous [Rh(NBD)N- BOC-
butaphos (5311+ Catalysts

 

 

—— 0ptica1 Yield(%)a

 

 

Substrate Amino acid Intercal. Homogen.
COOH
-" L-Phenylalanine 93 86
<iié‘rmcomg
COOH
‘4’ L-Phenylalanine 86 86
@gxmacocfiu5
_jcooczns
NHCOCGH5 L-Phenylalanine 94 94
COOH
-—/ L-Tyrosine —— 85
NHCOCH3
0Ac
COOH
_/' L-00PA 89 86
Sé’WHcocn3
OAc OMe

 

aThe hydrogenations were carried out in 30 mL of 95% EtOH on 4-5 mmol of
substrates at 25°C-and 740 torr. The concentrations of intercalated and
homogeneous catalysts were around 1.0 X 10'3 M. All the corresponding
amino acids had (§)-c0hfigurations.

156

may form intermolecular hydrogen bonds with the solvent

or intramolecular hydrogen bonds with the rhodium enamide
complex during the asymmetric hydrogenation cycle. The intra-
molecular hydrogen bonding interactions may lead to low
stereoselection of dehydroamino acids and reduce optical
yields. However, the possibilities of forming intramolecular
hydrogen bonding was low. The optical yields of the corre-
sponding amino acids were only 3 - 6% lower than those
obtained with homogeneous [Rh(NBD)(g)-Prophos (§)]+ catalyst
in 95% EtOH. Hith N-benzoyldehydrophenylalanine ethyl ester
intramolecular hydrogen bonding between the ligand

amide carbonyl and the substrate is not possible. A high
optical yield was obtained (96%). With the aprotic solvent
THF, intra- and inter-molecular hydrogen bonding are not
possible. The possibility of intramolecular interactions
between amide carbonyl and rhodium intermediates could be

eliminated and high optical yields could be achieved.

157

Suggested Future Work

 

Two flexible type chiral diphosphine ligands, DIOP (g)
and BPPM (8}) have been attached to gel type copolymers.33’34
In general, the optical yields afforded by the supported
catalysts were lower than their homogeneous analogues.

DIOP (g) and BPPM (8;) were also attached to modified gel
type copolymers containing chiral alcohol sites.35’651'he
findings indicated that the solvent-polymer interactions
dominated the effects of the additional chiral centers.35’65
The N-BOC-butaphos (5}) is a rigid type ligand and provides
high optical yields for asymmetric homogeneous hydrogenation
of (;)-dehydroamino acids. The ligand is less sensitive

to the solvent. In the light of these advantages, N—BOC-
butaphos (5g) may provide superior immobilization systems
for attaching the ligand to polymeric, inorganic or organic
carriers. Below are two possible routes for attachment

of the N-BOC-butaphos (5}) to polymer and glass. First,
removal of the protecting BOC group of N-BOC-butaphos (5;)
by treatment with CF3COOH should produce the 4risobutyl-
l,2-bis(di-4-methylphenylphosphino)butane [ aminobutaphos

(68)] Scheme30 .

 

H
H .
"~ __= CH CH NCH CH(CH )
CHz-(:\““CH2CH2NCH2CH(CH3)2 l._ _CF3COOH /CH2 c\’ 2 2:1 2 3 2
co (2 H --1: A,
pAr2 PAr2 .2 4 9 2. H20 PAr2 PArz
lV-BOC-butaphos (5}) (6y)

(Scheme 30 )

158

Scheme 3] proposes a route to attach the ligand to a polymer
support. The aminobutaphos (68) is treated with 4-vinyl-
benzoyl chloride to produce the monomer (69). The monomer
(69) is copolymerized with 2-hydroxyethyl methacrylate to
form the gel type copolymerized ligand (29). The co-
polymerized ligand (Zp) is allowed to react with rhodium

precursor to form the supported catalyst precursor.

 

 

4;
.H .
CH2_-C/CHZCH2HCHZCH(CH3)2 '
/ \ H mm / NEt3
PAr2 PAr2
(518)
*1 3 EH2
\ CH CH N-C-
CHZ— c\’ 2 2| @ CH3- C-COZCHZCHZOH COPOLYMER
c H «a
4 9 a: (7‘0)
PAr2 PAr2
(63)
Scheme 31

Scheme 32 shows the route for the attachment of theIV-BOC-
butaphos (58) to glass. The aminobutaphos (68) is coupled

with an acyl chloride functionalized glass to form glass

159

supported aminobutaphos (1]). The glass supported amino-
butaphos ([1) is allowed to react with rhodium precursor

to form the supported catalyst precursor. The glass support
has a covalently attached aliphatic group terminating in

an active carboxyl group. The 10 A length of the extension
arm should be long enough to permit immobilzation of bulky

aminobutaphos (68).

 

H
CH2_é,’CH2CH2HCHZCH(CH3)2 o 0
H 11 n
PAr2 PAr2 <:)-o Si-(CHZ )3 NH- c CHZCHZC-Cl
(Q§) (Acyl chloride support glass) ,
5 9 9
_&,!CH2 CHZN- -c- -CH2CH2 c-H-(CH2)351-o-(:)
//CH \ C4H9
PArz2 PAr2

Scheme 32

EXPERIMENTAL
A. General

Manipulations involving air-sensitive materials were
performed either under argon by Schlenk techniques or in
a nitrogen filled glove box. Melting points were deter-
mined in capillary tubes and were uncorrected. Optical
rotations were recorded with a Perkin-Elmer l4l automatic

1

polarimeter and a lO-cm path length micro-cell. H NMR

spectra were obtained with a Varian T60 or Bruker WH-ZSO

3]? NMR measurements were obtained with a

spectrometer.
Bruker WH-l80 spectrometer operating at a field of 4.23
Tesla and a frequency of 72.88 MHz in the pulsed Fourier

31

Transform mode. The P NMR spectra were calibrated in parts

per million (6) downfield from 85% H3P04 as an external

13c NMR

standard. All spectra were proton decoupled.
spectra were performed with a Varian CFT-ZO spectrometer
operating at a field of 1.868 Tesla and a frequency of 20

MHz. The 13

C NMR spectra were calibrated in parts per million
(6) downfield from tetramethylsilane as an internal standard.
All spectra were proton decoupled. Infrared spectra were
obtained with a Perkin-Elmer 237B spectrometer. The infrared
spectrometer was calibrated with polystyrene. Mass spectra

(MS) were obtained with a Finnigan 4000 GC/MS spectrometer.

160

151
"High resolution mass spectra were obtained with a Varian
CH5 double focusing mass spectrometer; NIH/M.S.U. mass
spectra facilities. Elemental analyses were done by the
Galbraith Laboratories Inc., Knoxville, Tennessee. All
solvents used were A.C.S. reagent grade. THF and ether
were purified by distillation from lithium aluminum
hydride under argon. Cyclohexane was distilled from
sodium metal under argon. Pyridine was distilled from
barium oxide under argon. Chlorodiphenylphosphine, l,5-
cyclooctadiene, 2.5-norboradiene, 4-hydroxybenzaldehyde,
vanillin, dihydropyran, benzylchloride, isobutylamine,
di-tert-butyl dicarbonate, Pd on activated carbon l0%
and n-butyl lithium were obtained from Aldrich Chemical
Company. (§)-Lactic acid, and L-malic acid were
purchased from Sigma Chemical Company. Tert-butyl alcohol
was obtained from Mallinckrodt, Inc. Lithium metal
wire l/8" diameter was purchased from Matheson, Colleman
& Bell. Sodium hydride (56.2% dispersion in mineral oil)
was obtained from Metal Hydrides Inc. Tris(4-methylphenyl)
phosphine was purchased from Strem Chemicals, Inc. Rhodium
trichloride hydrate was obtained from Engelhard Minerals
& Corporation. Hydrogen (99.999% purity) was purchased

from Matheson. SPIPHOS52

was a gift from Dr. R. A. OeVries.
(+)-2,3-0-Isopropylidene-2,3-dihydroxy-l,4-bis(diphenyl-
phosphino)butane [DIOP(+)] was purchased from Sigma

Chemical Company.

162

B. Substrates

l. 2-Acetamido-2epropenoic Acid (a—Acetamidoacrylic Acid)

 

This compound was purchased from Aldrich, and the

1

purity was checked by H NMR, MS and melting point. mp

l85-l86°C; MS m/e 129 (M+); 1H NMR (60 MHz, TMS, 0 -DMSO)

6
6 = 8.86 (brs, 1H, COOH) 6.17 (s, 1H, olefinic H), 5.63

(s, lH, olefinic H), 2.0 (s, lH, CH3).

2. gZ)-2-Acetamido-3-phenyl-2-propenoic Acid L(;)-a-Ace-
tamidocinnamic Acid]

This compound was prepared according to the procedure

of Herbst and Shemin.54

1

mp 192-193°c, MS m/e 205 (n+1;

H NMR (60 MHZ, TMS, D -DMSO) 6= 9.26 (s, 1H, COOH) 7.7-

6
6.9 (complex m, 6H, C H and olefinic H), 2.0 (s, 3H,

6 5
CH3); 136 NMR (20 MHz, TMS, 0 -DMSO) 6= 169.7 (NHCOCH3)

6
166.7 (COOH), 133.9, 131.6, 129.9, 129.4, l28.8, 127.5,
22.7 (NHcogH3).

3. (£)-2-Benzamido-3-phenyl-2-propenoic Acid [L;)-o-Benzami-
docinnamic Acid]

This compound was prepared by base hydrolysis of the
ethyl (%)-a-benzamidocinnamate according to the procedure
of Carter, et a1.55 mp 224-226°C; 1H NMR (60 MHz, TMS,
-DMSO) 6 = 9.92 (s, lH, COOH) 8.l5-7.8, 7.8-7.05 (complex

13 -DMSO) 5 = 166.5 (NHCOC6H5)1

D6

m, 12 H); 6
166.3 (COOH), 133.9, 133.8, 133.2, 131.9, 129.9, 129.4,

C NMR (20 MHZ, TMS, D

128.6, 127.8, 127.6.

163

4. Ethyl (;)-2-Benzamido-3-phenyl-2:propenoate [Ethyl(;)-
a-Benzamidocinnamate]

 

This compound was prepared by a slight modification

55 Azlactone

of the procedure described by Carter et a1.
(15 g) was suspended in 80 mL toluene and 10 mL of l N
sodium ethoxide solution was added in one portion. The
azlactone rapidly dissolved and ethyl (g)-o-benzamido-
cinnamate began to separate almost immediately. After

5 minutes an excess of dilute hydrochloric acid was

added and the mixture was vigorously shaken. The white
solid was collected by filtration, washed with water

and petroleum ether and then dried. The crude product

was recrystallized from EtOH/HZO to givea pure needle-like
crystalline product (16.6 g, 94%). mp 148.5-149°C;

MS m/e 295 (M+); 1 H NMR (60 MHz, TMS, coc13) 6 = 7.9 (s,
1H), 7.8-6.9 (complex m, 11H), 4.13 (q, J = 7H2, 2H, CH2),
1.23 (t, J=6Hz, 3H, CH3); 136 NMR (20 MHz, TMS, CDC13)

6 = 166.3 (NHCOCGHS), 165.0 (COOH), 133.6, 132.9, 132.4,
132.2, 131.9, 129.9, 129.6, 128.6, 127.8, 127.1, 60.9,
(QCHZCH3), 14.1 (CHZCH3).

p

a; )-2-Acetamido-3-(4’-hydroxyphenyl)-2-propenoic Acid
)-Acetamido-4-hydroxycinnamic Acid]

This compound was prepared according to the procedure
of 0akin56. mp 203-205°C; MS m/e 221 (M+); 1H NMR (60

MHz, TMS, D -DMSO) 6 = 9.13 (s, 1H,COOH) 7,23 and 5,57

6
(2d, J=8 Hz, 4H, arom. H), 7.1 (s, 1H, olefinic H), 1.96

(s, 3H, CH3). 136 NMR (20 MHz, TMS, 0 -DMSO) a = 169.9

6
(NHCOCH3), 167.1 (COOH), 159.0, 133.1, 132.2, 124.9,

164

124.2, 115.9, 22.9 (NHCOCH3).

6. )-2-Acetamido-3-(4’-acetoxyphenyl)-2-propenoic Acid
w146-Acetamido-4-acetoxycinnamic Acid]
The azlactone of (g)-a-acetamido-4-acetoxycinnamic acid

was prepared according to the procedure of Dakin.56

The
compound was obtained by hydrolysis of the azlactone in boil-
ing acetone/H20, according to the procedure of Herbst and
Shemin.54 mp 224-225°c; 1H NMR (60 MHz, TMS, 06-0MSO);

6 = 9.3 (s, H, COOH) 7.5 and 7.0 (2d, J=8 Hz, 4H, arom. H),
7.06 (s, 1H, o1efinic H), 2.13 (s, 3H, CH3C00), 1.96 (s,
3H, NHcocH3).

7. (z)-2-Acetamido-3-(3’-methoxy-4’-acetoxypheny1)-2-

propenoic Acid [(Z]-a-Acetamido-3-methoxy-4iacetoxy-
cinnamic Acid]

 

The azlactone of (g)-a-acetamido-3-methoxy-4-acetoxy-
cinnamic acid was synthesized according to the method

described for the azlactone of (g)«rbenzamidocinnamic acid's

derivative57. The compound was obtained by hydrolysis of

the azlactone in boiling aceuNm/HZO, according to the pro-

54

cedure of Herbst and Shemin. mp 206-207°C; MS m/e 294

(n+1), 293 (M+); 1H NMR (60 MHz, TMS, D -0MSO), 6 = 9.3

6
(s, 1H, COOH) 7.3-6.9 (complex m, 4H, arom. H and olefinic

H), 3.7 (s, 3H, 0CH3), 2.2 (s, 3H, CH3C00) 1.93 (s, 3H,

HHcocH3); 136 NMR (20 MHz, TMS, 06-DMSO) 6 = 169.4 (NHCOCH3)

168.5, 166.5 (COOH and CH3EOO), 150.7, 140.0, 132.6, 130.6,

127.5, 123.0, 122.7, 113.9, 55.8 (OCH 22.5 (NHCOQH3),

3),
20.4 (CH3C00).

165

C. Identification of a-Amino Acid Derivatives (Hydro-
genated Products)

The products obtained from the hydrogenation reactions

were identified by mass spectroscopy and 1H and 13C NMR

spectroscopy:

1. N-Acetyl-L§)-phenylanine

MS m/e 207 (n+1; 130 NMR (20 MHz, TMS, 0 -0MSO)

6
6 = 173.3 (NHCOCH3), 169.9 (COOH), 137.9, 129.2, 128.4,

126.6, 53.7 (ArCHZQH), 37.1, 22.5 (NHCOCH3).
2. N-Benzoyl-(§)-phehylalanine

MS m/e 225 (M -44)+; ‘3

C NMR (20 MHZ, TMS, 06-DMSO)
6 = 173.3 (NHEOC5H5)’ 155.5 (COOH), 138.3, 134.1, 131.4,
129.1, 128.3, 127.4, 125.4, 54.3 (AFCHZEH), 35.5

3. N-Benzoyl-(§)-phenylalanine ethyl ester

MS m/e 297 (M+); 130 NMR (20 MHz, TMS, 0 -0MSO)

6
6 = 171.7 (NHCOC6H5), 166.5 (COOH), 137.7, 133.8, 131.4
129.1, 128.2, 127.4, 126.4, 60.5 (09H20H3), 54.4 (ArCHng),
36.4, 14.0 (CHZCH3).

4. N-Acetyl-(;)-tyrosine
13

 

C NMR (20 MHZ, TMS, D -DMSO) 6 = 173.4 (NHCOCH3),

6
169.5 (COOH), 156.1, 130.1, 127.9, 115.2, 54.0 (ArCHZCH),
36.3, 22.5 (NHCOCH3).

5. N-Acetyl-3-(4-acetoxy-3-methoxyphenyl)-(§)-a1anine

130 NMR (20 MHz, TMS, 0 -DMSO) 6 = 173.2 (NHCOCH311

6
169.5, 168.6 (COOH and CH3COO), 150.5, 138.0, 136.7,

122.4, 121.1, 113.8, 55.7 (OCH3), 53.5 (ArCHZCH), 36.6,

166

22.4 (NHcogH3). 20.4 (EH3COO).
0. Preparation of (3)-Prophos (§)and (3)-4-Me-Prophos (49)

1. Preparation of-(§)-(-)-l,2-Propened1013DisE-toluene-
sfilfonate

 

(§)-Lactic acid (Sigma) was converted to the dito-
sylate derivative in 93% yield by the procedure previ-

ously described.10

mp 68-69°C; (11t.‘° mp 62°C) [6123
-20.6° (c 1.15, CHC13); MS m/e 384 (M+); 1H NMR (60 MHz,
TMS, 00013) 6~= 7.60 (dd, J,=8, 02=2 Hz, 4H, g>C6H4), 7.17
(d, J=9 Hz, 4H,,p-C6H4), 4.63 (m, 1H, CH), 3.90 (d, J=6 Hz,
2H, CH2), 2.43 (s, 6H, 2CH3), 1.23 (d, J=7 Hz, 3H, CH3).

2. Preparation of,( )-(+)-l,2-Bis(diphenylphosphinolpropane
[(3)-Prophos (6)

The HP(C6H5)2 could be prepared by reducing72 C1P(C6H5)2
with LiAlH4 in ether or c1eaving73 triphenylphosphine with
lithium metal in THF. This modified method for synthesis
of (3)-Prophos (6) was different from Bosnich's report.1O
Only a stoichiometric amount of LiP(C6H5)2 was used and the
pure product could be easily separated by crystallization
from absolute ethanol.

Into a flame-dried, argon-flushed, three-necked, 250-
mL flask equipped with a teflon-coated stirring, bar, a 50-
mL addition funnel and a reflux condenser, and capped with
rubber septums was placed dry THF (50 mL) followed by
(C6H5)2PH72 (9.33 mL, 53.6 mmol). The mixture was cooled
to -5°C and 34.7 mL of 1.6 M n-butyllithium (55.52 mmol)

in hexane was added by syringe. The resulting orange-red

167

solution was held at -5°C for 15 minutes, warmed to room
temperature for 1.5 hours, and cooled to ~5°C again.

A solution of (§)-(-)-1,2-pr0panediol-di-p-toluene-
sulfonate (10.38 g, 26.8 mmol) in 15 mL of dry THF was
added dropwise over a 45 minute period to the stirred phos-
phide solution. The resulting light yellow solution was
allowed to stir for 1 hr at room temperature. Deoxygenated
water (50 mL) was then added and most of the THF was removed
under reduced pressure with a vacuum pump. The aqueous mix-
ture was extracted three times with 40-mL portions of ether.
The combined ether extracts were washed with 30 mL water
and filtered. Ether was removed from the filtrate under
reduced pressure to give an oily residue. The oily residue
was taken up in absolute ethanol under argon at 50°C. The
solution was allowed to slowly cool to 25°C and then was
held at 5°C for 2 days. The crude product (5.7 g) was collected
and recrystallized from absolute ethanol to give 4.3 g of
pure (3)-Prophos (6) as small colorless prismatic crystals.
mp 67.5 - 68.5°C, [5123 186.0° (c 1.0, acetone); MS m/e
(relative intensity) 412 (M+, 3.51), 370 (7.96), 303 (7.19),
262 (16.8), 185 (59.6), 183 (100), 109 (16.5), 108 (23.3).
1H NMR (60 MHz, TMS, CDC13) 6 = 1.26 (dd, 01:15, 02=6.5 Hz,
3H, CH3), 1.84 (m, 1H, CH), 2.26 (m, 2H, CH2) 7.25 (complex
m, 20 H, 4C6H5), 3‘2 NMR (00013, 85% H3P04as external
standard, downfield shifts positive),6(ppm) 2.1, -20.2,

J =21 Hz.

P-P

168

3. Preparation of Bis(4-methy1phenyl)phosphihe

Bis(4-methy1pheny1)phosphine was obtained from tris-
(4-methy1phenyl)phosphine by Cleavage with lithium metal
in tetrahydrofuran, followed by protonation with NH4C1 and
distillation.

A 250-mL, three-necked flask, containing a magnetic
stirring bar and fitted with an argon inlet was charged with
tris(4-methylpheny1)phosphine (9.34 g, 30.7 mmol) and 100
mL of dry tetrahydrofuran. To the stirred solution was
added lithium (177.3 mmol, 36.1 cm of 1/8 inch diameter
wire) which was washed with hexane and dried carefully
with a paper towel. Lithium wire was added by cutting 3-5
mm segments directly into the center neck of the flask with
scissors. A slow argon flow was maintained throughout the
addition, which required about 5 minutes. The red solution
was allowed to stir for 4 hours and then filtered. The
filtrate was treated with 20 mL of deoxygenated water and
3.5 g of NH4C1. The mixture was allowed to stir for an
additional hour. During this treatment the solution be-
came clear and almost colorless. The solution was extracted
with two 25-mL portions of ether and the combined ether
layers were washed with two 15-mL portions of saturated
aqueous sodium chloride. The ether solution was dried
over magnesium sulfate for about 0.5 hours and filtered.
The solvent was pumped off under vacuum. The residue
was transferred under argon to a small flask for dis-

tillation under reduced pressure. The yield was 4.9 g

169

(74.6%) of bis(4-methy1phenyl)phosphine. bp 100-105°C
(0.3 torr). It was used within 24 hours without further
characterization.

4. Preparation of (B)-(+)-l,2-Bis(di-4’-methy1-phenylphos-
phino)propane, [(3)-4-Me-Prophos (49)]

To a solution of freshly distilled HP(C6H4CH3)2 (4.54 g,
21.2 mmol) dissolved in 35 mL dry THF and cooled to -5°C,
a 1.6 M hexane solution of n-butyl lithium (13.8 mL, 22.1
mmol) was added dropwise. The resulting orange-red solution
was allowed to stir at -5°C for 15 minutes. The solution
was warmed to room temperature and then allowed to stir for
another 1.0 hour. The mixture was cooled to -5°C again.
A solution of (§)-(-)-1,2-propanediol di-p-toluenesulfonate
(4.04 g, 10.5 mmol) in dry THF (7 mL) was added dropwise
over a 30-minute period to the stirred phosphide solution.
The resulting light yellow solution was allowed to stir for
1 hour at room temperature. Deoxygenated aqueous ammonium

chloride (3.5 9 NH C1 in 20 mL H20) was then added and most

4
of the THF was removed under reduced pressure. The aqueous
mixture was extracted with two 30-mL portions of CH2C12.

The combined CHZCl2 extracts were washed with water and dried
over anhydrous M9504 and filtered. Upon concentration of
the filtrate under vacuum a crude white precipitate was ob-
tained (3.9 g). The product was crystallized from absolute

ethanol/CH2C1 to give (2.3 g, 45%) of pure (R)-4-Me-Prophos

2
(49). mp 129.5-130°c; [6123 172.6° (c 1.0, benzene); MS

m/e (re1ative intensity) 468 (M+, 18.3) 426 (10.42), 345

170

(14.07), 304 (21.70), 213 (100), 123 (17.95); 1H NMR

(60 MHZ, TMS, CDC13) 6 = 1.18, (dd, J1=15.8, J =6.7 Hz,

2
3H, CH3), 1.78 (m, 1H, CH) 2.17, 2.21, 2.27 (35, 14H,

4CH3, CH2), 6.7-7.3 (complex m, 16H, 4C6H4); 3‘P NMR
(72.88 MHz, 85% H3P04 as external standard, downfield

shifts positive), 6(ppm) 0.3, -22.1 J =21 Hz, (CDC1

P-P 3’
25°C);

High resolution MS m/e 468.21459

Calcd for C31H34P2: 468.21358
Anal. Calcd for C31H34P2: C, 79.46; H, 7.31; P, 13.22
Found: C, 79.20; H, 7.36; P, 13.47

E. Preparation of the Synthetic Precursor
Preparation of 1,2-0-Isopropylidene-l,2-(§)-4-butanetriol
[Acetonide-triEl (58)]

‘Acetonide-triol (58) was prepared in 60% overall yield
from L-malic acid by known procedures.71 02% 0.600 (neat,
2:1 dm); 1H NMR (250 MHz, TMS, CDC13) 6 = 4.25 (m, J=6 Hz,
1H, CH2

6 Hz, 2H, CHZCHZOH), 3.58 (t, J=8 Hz, 1H, CHfl’CH-O), 3.37

Cfl-O), 4.09 (t, J=8 Hz, 1H, CHH’CHO), 3.74 (t, J=

(brs, 1H, 0H), 2.80 (q, 2H, CHZOH), 1.41 and 1.36 (25, 6H,
2CH3).

F. Synthesis of THP-butaphos (51) and (R)-Hydroxylbutaphos
(53)

1. Preparation of 4-0-Benzyl-l,2-(§)-4-butanetriol [Benzyl-
d101 (59)]

Sodium hydride (1.92 g, 45.0 mmol) as a 56.2 wt% mineral

oil dispersion, was washed three time with n-pentane to

171

remove the mineral oil, and suspended in 30 mL of toluene.
The mixture was allowed to stir under nitrogen while a toluene
solution of acetonide-triol (58) (5.48 g, 37.5 mmol) was
added dropwise. After the addition was complete (0.5 hours),
the mixture was stirred at room temperature for 0.5 hours
and benzyl chloride (5.7 g, 45 mmol) was added in one portion.
The mixture was heated at reflux for 8 hours, allowed to cool
to room temperature, and allowed to stir at room temperature
overnight. The mixture was filtered through a bed of Celite
which was then washed with ether, and the filtrate was evapora-
ted in vacuo ‘The residue was dissolved in 80 mL of a mixed
solution of HAC, THF and H20, (1:1:1) and the solution was
heated at 45°C for 2 hours. The solvent was removed in
vacuo and the residue was coevaporated with ethanol (4 x
20 mL). Kugelrohr distillation (145°C, 0.017 torr) gave
6.18 d (84%) of benzyl-dial (59). [6123 20.00 (c 1.0,
MeOH); MS m/e 196 (HT); 1H NMR (250 MHz, TMS, CDC13) 6 = 7.3
(s, 5H, C6H5), 4.47 (s, 2H, CflzAr), 4.05-3.75, 3.35-3.57
(br, 5H, CHO, 2 0H, and CHZO), 3.6 (br, 2H, CHZCHZO), 3.42
(br, 2H, CHZO), 1.7 (q, J=6 Hz, 2H, CHZCHZO).
Aflgl. High resolution MS m/e 196.10965

Calcd for C11H1603: 196.10995

2. Preparation of 4-0-Benzy1-1,2-di-0-(p-toluenesulfony1)-
1,2-(§)-4-butanetriol [Benzylether-ditosylate (50)]

A solution of benzyl-diol (59) (5.7 g, 29.1 mmol) in
5 mL pyridine was added dropwise to a stirred mixture

of p-toluenesulfonyl chloride (11.4 g, 59.8 mmol) and

172

pyridine (25 mL) at 0°C over a period of 0.5 hours. The
mixture was allowed to stir at 0°C for an additional hour
and then was kept in a refrigerator for 48 hours, with
occasional shaking. The reaction mixture was diluted with
100 mL ether and washed with cold 1 N HC1 aqueous solution
until the washings were acidic. It was then washed with
saturated aqueous NaHCO3 and H20. The organic layer was
dried over anhydrous MgS04. The filtrate was concentrated
to give 13.4 g (91%) of benzylether-ditosylate (59), which

1H NMR

was used for further reactions without purification.
(60 MHz, TMS, CDC13) 6= 7.7-6.9 (complex m, 13H, arom. H),
5.0-4.53 (m, 1H, CH), 4.24 (s, 2H, CflzAr), 4.05 (d, J=4 Hz,
2H, Cflz-OTs), 3.34 (t, J=5 Hz, 2H, CHZCHZO), 2.40 (s, 6H,
2CH3), 1.90 (q, J=6 Hz, 2H, CHCHZCHZ).

3. Preparation of 1,2-01-0-(p-to1uenesu1fony1)-1,2-(;)-4-
butanetriol [Hydroxylditosylate (61)]

Benzylether-ditosylate (59) (10.1 g, 20 mmol) was
dissolved in dioxane (70 mL) containing a few drOps of
concentrated hydrochloric acid. 10% Pd (0.35 g) on charcoal
was added, and the mixture was hydrogenated at 1 atmosphere
pressure and 25°C until the hydrogen consumption had ceased
(0.49 L). The catalyst was removed by filtration through
a bed of Celite. The filtrate was concentrated in vacuo
to give 8.1 g (98%) of hydroxylditosylate (5]). The product
was used in further experiments without purification.
1H NMR (250 MHz, TMS, CDC13) 6 = 7.74, 7.67 (2d, J=8 Hz,
4H, arom. H), 7.33 (d, J=8 Hz, 4H, arom. H), 4.89 (m, 1H,

173

CH-O), 4.04 (t, 2H, CHZCH), 3.65 (t, 2H, CHZCHZ), 2.45

(s, 6H, 2CH3), 1.84 (q, J= 6H2, CHZCHZOH).

4. Preparation of 4-0-(2’-Tetrahydropyrany1)-l,2-di-0-(p-
toluenesulfony1)-1,23(§)-4-butanetriol [THP étherifi1t05y-
late51621]

Dihydropyran (1.52 g, 18 mmol) and a catalytic amount

of p-TsOH (45 mg) were added .to a stirred solution

hydroxylditosylate (5}) (6.63 g, 16 mmol) in dry ether (100

mL). The reaction mixture was allowed to stir overnight at

room temperature. Then it was washed with aqueous Na2C03

solution and water. The ether layer was dried over anhydrous

M9504, filtered and concentrated in.vacuo at 40°C (0.1 torr)

overnight. A product (7.73 g, 97%) of THP ether-ditosylate

(52) was obtained and used for phosphination without further

purification. 1

H NMR (250 MHZ, TMS, CDC13) 6 = 7.8-7.6,
7.4-7.2 (2 complex peaks) 4.87 (br, 1H, CHOTs), 4.5 and 4.33

(2t, 1H, CHZCHOCHZ), 4.13 (m, 2H, CH OTHP), 3.74 (m, 2H,

2
052015), 3.6-3.2 (br, 2H, OCHOCHZ) 2.45 (s, 6H, 2CH3), 1.93
(m, 2H, CHZCHZOTHP), 1.82-1.23 (br, 6H, CHZCHZCHZCHz-O).

5. Preparation of 4-0’-Tetrah1dropyranyloxy)-l,2-(R)-bis
(diphenylphosphin01butahe THP-Butaphos (51)

 

 

To a solution of 2.79 g (15 mmol) of freshly distilled
HP(C6H5)2 dissolved in 15 mL of dry THF and cooled to -5°C
a 1.6 M hexane solution of n-butyl lithium (9.8 mL, 15.6
mmol) was added dropwise. The resulting orange-red solution
was stirred at -5°C for 15 minutes and warmed to room tempera-
ture and stirred for another 1.0 hour. The mixture was cooled
to -5°C again. A solution of THP ether-ditosylate (52)

(3.74 g, 7.5 mmol) in 5 mL dry THF was added dropwise for

174

over 20 minutes to the stirred phosphide solution. The
resulting light yellow solution was stirred for 1 hour at

room temperature. Deoxygenated water (15 mL) was then added
and most of the THF was removed under reduced pressure.
The aqueous mixture was extracted with 2 x 5 mL ether.
The combined ether extracts were washed with water and dried
over anhydrous M9504 and filtered. Upon concentration under
vacuum, a crude oil product of THP-butaphos (5]) was obtained.

31

The purity of the product was confirmed by P NMR and was

larger than 90%. 31

P NMR (72.88 MHz, 85% H3P04 as external
standard, downfield shifts positive), two diastereomers

6(ppm) -1.4, -20.1, J =27.6 Hz; 6(ppm) —1.8, -20.5,

P-P

J =27.6 Hz, (C001 25°C).

P-P 3’

6. Preparation of 4-Hydroxy-l,2-(g)-bis(dipheny1phosphino)
butane [(gI-Hydroxylbutaphos'(53)]

(3)-Hydroxylbutaphos (52) was obtained by acid hydrolysis
of THP-butaphos (51) with a catalytic amount of TsOH in
ethanol. THP-butaphos (51) (1.25 g) was dissolved in 10
mL ethanol with a catalytic amount of TsOH (10 mg). The solu-
tion was allowed to stir overnight under N2 atmosphere.

The solution was concentrated an vacuo to a small amount

and then diluted with ether (10 mL). The solution was washed
with aqueous NaHCO3 and H20. Upon concentration under vacuum,
a crude, oily product (1.1 g) of (R)-hydroxylbutaphos (52)

was obtained. 3‘P NMR (72.88 MHz, 85% H3P04 as external
standard, downfield shifts positive) 6(ppm) -l.5, -20.4,

J =22.6 Hz, (EtOD, 25°C).

P-P

175

6. Synthesis of N-BOC-butaphos (53)

1. Preparation of 4-0-(p-Toluenesulfony1)-1,2-0-isopropy-
lidene-l,2-(§)-4-butanetriol [Acetonide-tosylate (53)]
A solution of l,2-0-i50propylidene-1,2-(§)-4-butanetriol
(58) (11.7 g, 80.0 mmol) in 5 mL pyridine was added drop-
wise mo a stirred,mixture of p-toluehesulfonyl chloride
(15.9 g, 83.2 mmol) and pyridine (50 mL) at 0°C over a
period of 0.5 hours. The mixture was allowed to stir at
0°C for 1 additional hour, and then it was kept in a refri-
gerator for 24 hours. The reaction mixture was diluted with
100 mL ether and washed with cold 0.5 N aqueous HC1
until the aqueous wash solutions were acidic. It was then
washed with saturated aqueous NaHCO3 and H20. The organic
layer was dried over anhydrous M9504. The filtrate was
concentrated to give 19.5 g (81%) of 4-0-(p-toluenesulfony1)-
1,2-0-isopropylidene-1,2-(5)-4-butanetriol (53), which was
used without further purification for further reaction.
1H NMR (60 MHz, TMS, cpc13) 6 = 7.70 and 7.25 (A, A’, B, 8’,
4H, arom. H), 4.30-3.70 (m, 4H, 2H2C-O), 3.70-3.10 (m, 1H,
HC-O), 2.42 (s, 3H, CH3), 1.92 (dt, 2H, CH2), 1.32 (s, 3H,
CH3), 1.25 (s, 3H, CH3); 1H NMR (250 MHz, TMS, CDC13)
6 = 7.80 and 7.35 (A, A’, B, B’, 4H, C6H4)’ 4.30-4.0 (m,
3H, CH-o, C5201s), 4.0 (t, 01’1,=J]’2=8 Hz, CHH’-0), 3.5
=0

(t, J =8 Hz, CHfl’-0), 2.43 (s, 3H, CH3), 1.90

1,1’ 1,2
(brdt, 2H, CHZCHZOTS), 1.33 (S, 3H, CH3), 1.28 (s, 3H, CH3).

176

2. Preparation of 4-Isobutylamino-l,2-0-isopropylidene-

1,2-(§1:butanediol (54)
4-0-(p-lbluenesulfonyl)-1,2-0-is0propylidene-l,2-(§)-

 

4-butanetriol (5}) (19.5 g, 64.9 mmol) was dissolved in

120 mL isobutylamine (freshly distilled from CaHZ) and re-
fluxed at 60-70°C for 10 hours. It was allowed to stir over-
night at room temperature. The solvent was slowly evaporated
under vacuum until crystals formed. Once the crystals formed,
evaporation was discontinued. The solution was then diluted
with dry ether (60 mL) and filtered. The filtrate was con-

centrated and distilled at 80°C/O.l torr to yield 12.2 g

25
a D

H NMR (250 MHZ, TMS, CDC13) 6 = 4.17 (m, 1H, CH-O), 4.07

(93%) of a light yellow oil.
1

3.71 (neat, 2 = 1 dm);

(t, a, 1,=.11 2=8 Hz, 1H, CHH’-0), 3.55 (t, J 0 =8 Hz,

1,1’= 1,2

1H, CHfl’-O), 2.75 (m, 2H, CH CHZNH), 2.45 (d, J=Hz, 2H,

2

NHCHZCH), 2.26 (s, 1H, NH), 1.6-1 9 (brm, 3H, cg CH NH and

2 2
Cfl(CH3)2), 1.42 (s, 3H, CH3), 1.35 (s, 3H, CH3), 0.09 (d,
J=7 Hz, 6H, 2CH3).

Anal. High resolution MS m/e 201.17156

Calcd for C11H23N02: 201.17288

3. Preparation of 4-Isobutylamino-l,2-(§)-butanediol
[Aminodiol (55)]

4-Isobuty1amino-1,2-0-isopr0pylidene-1,2-(§)-butanediol
(54) (12.1 g, 60 mmol) was dissolved in 1 N HC1 (80 mL) and
120 mL MeOH. The solution was refluxed at 50-70°C for
4-5 hours and stirred overnight at 25°C. The solution was
made alkaline by the addition of NaOH (4 g, 100 mmol).

After the methanol was removed under vacuum, the aqueous

177

solution was extracted with ether (3 x 200 mL). The combined
ether extracts were dried over anhydrous M9504 and evaporated
under reduced pressure to yield 7.55 g (78%) of a light
yellow 011. MS m/e 162 (M+l)+; IR (neat) 3400 cm“; 1H NMR
(250 MHz, TMS, CDC13)’6=3.8-4.0 (m, 2H, 2 0H), 3.65-3.90

(m, 1H, CH-O), 3.45-3.65 (m, J=5 Hz, 2H, CHz-O), 2.73-3.00

(m, J=7 Hz, 2H, CHZCHZNH), 2.33-2.53 (m, J=7 Hz, 2H, NHCH CH),

2
1.60-1.84 (m, J=7 HZ, 1H, Cfl(CH3)2), 1.60-1.73 (m, J=6 HZ,
2H, CHZCHZNH), 0.92 (d, J=7 HZ, 6H, 2CH3).

4. Preparation of 4-{(tart-Butyloxy)carbony1]-isobutylaminqw
l-2-(§j-butanediolIF-BOC-aminodiol (55)]

To a solution of 4-isobuty1amino-1,2-(§)-butanediol
(55) (6.65 g, 41.24 mmol) in 50 mL of tert-butyl alcohol
at room temperature was added di-tert-butyldicarbonate
(10.36 g, 47.44 mmol). After the reaction mixture was allowed
to stir for 2 hours, excess di-tert-butyldicarbonate was
quenched by adding in one portion 5 mL H20 to the solution.
After one additional hour, the solution was concentrated
at 40°C and the residue was taken up in ether and recon-
centrated to give a light yellow crude product. Kugelrohr
distillation (145°C, 0.035 torr) gave 10.46 (97%) of N-BOC-
aminodiol (56); [a]23 13.5° (c 2.0, MeOH); MS m/e 261 (M+);
IR (neat) 3400, 1680 cm“;
High resolution MS m/e 261.19633
Calcd for C13H27N04: 261.19401
H N0

Anal. Calcd for C C, 59.74; H, 10.41; N, 5.36

13 27 4‘
Found: c, 59.84; H, 10.32; N, 5.29

178

5. Preparation of 4-{[(tart-Butyloxy)carbonyl]-isobut lamino}
-142-di-0-(p-toluepesulfony1)-l,2-(§)-butaneaiol .-BOC-
aminoditosylate5(57)]

. N-BOC-aminodiol (56) (5.33 g, 20.0 mmol) was dissolved
in 7 mL of dry pyridine and this solution was then added
dropwise over 0.5 hours to an ice cold solution containing
p-toluenesulfonyl chloride (7.85 g, 41.2 mmol) in 25 mL of
dry pyridine. The solution was allowed to stir at 0°C for
1 hour, and then stored in a refrigerator for 48 hours.

At this point, the solvent of the mixture was pumped off.

The residue was diluted with 100 mL of ether and washed with

cold H20 (15 mL x 2). The organic layer was dried over

M9504, and filtered. The filtrate was concentrated in vacuo

to give a chromatographically pure product (10.16 g, 89.1%)

of N-BOC-aminoditosylate (5]) which was used in further

1

experimentation without purification. H NMR (60 MHz, TMS,

C0013) 6= 7.55 and 7.17 (A, A’, B, 8’, 8H, 2C6fl4, 4.5 (m,
z-OTs), 3.0 (tr,

J=6 Hz, 2H, CHZCHZN), 2.83 (d, J=8 HZ, 2H, NCHZCH), 2.37

J=5 HZ, 1H, Cfl-OTS), 4.0 (d, J=4 Hz, 2H, CH

(S, 6H, 2CH3), 1.6-2.01 (m, 3H, Cfl(CH3)2 and CflZCHZN)’ 1.4
(s, 9H, 3CH3), 0.83 (d, 6H, 2CH3).
6. Preparation of 4-{L(tert-Buty1oxy)carbonyl]-isobutylamino}

-1,2-(B)-biS(di-4’-methylphenylphosph1n016utane [N-BOC-
butaphos (53)]

Lithium bis(4-methy1phenyl)phosphide was generated from

 

n-butyl lithium and bis(4-methy1phenyl)phosphine. The
latter compound was obtained from tris(4-methylphehyl)phos-
phine by cleavage with lithium metal in THF at room tempera-

ture, followed by protonation with NH4C1 and distillation.

179

The secondary phosphine had a boiling point of 100-105°C
(0.3 torr). It was used directly without further characteri-
zation.

To a solution of freshly distilled bis(4-methylphenyl)
phosphine (6.76 g, 31.57 mmol) in 30 mL dry THF and cooled
to -5°C a 1.6 M hexane solution of n-butyl lithium (20.5
mL, 32.8 mmol) was added dropwise. The resulting orange-
red solution was allowed to stir at -5°C for 15 minutes and
then was warmed to room temperature for 1.0 hour. The mix-
ture was cooled again to -5°C. A solution of N-BOC-amino-
ditosylate (67) (8.99 g, 15.78 mmol) in dry THF (8 mL) was
added dropwise over a period of 30 minutes to the stirred
phosphide solution. The resulting light yellow solution
was stirred for 1 hour at.rodntemperature. Deoxygenated
water (ZOHNJ then was added and most of the THF was removed
under reduced pressure. “Huaaqueous mixture was extracted
with ether (2 x50 mL), The combined ether extracts were
washed with water and dried over anhydrous M9504 and filtered.
Upon concentration under vacuum, a crude oil product was
obtained. The product was crystallized from 95% ethanol
to give 3.92 g of N-BOC-butaphos (53) as a monohydrate;
mp 89.5-90°C; [o]23 56.0° (c 1.0, acetone); MS m/e (relative
intensity) 653 (M+, 19.9), 426 (78.7), 304 (67.4), 213 (100),
122 (28.9) 57 (94.9); 1H NMR (250 MHz, TMS, CDC13) 6 = 7.0-
7.4 (complex m, 16H, 4C6H4), 3.4-3.15 (m, 2H), 2.8 (m, 2H,
CH2), 2.25-2.50 (35, 12H, 4 CH3Ar), 2.2 (m, 2H), 1.7-2.0
(m, 4H), 1.4 (s, 9H, C(CH3)3), 0.8 (d, 6H, 2CH3);

180

31P NMR (72.88 MHz, 85% H3PO4 as external standard, down-

field shifts positive) 6(ppm) -4.4, -22.6, JP_P=20.7 Hz,
(CDC13, 0°C);
High resolution MS m/e 653.35352
Calcd for C41H5302P2N: 653.35517
5331. Calcd for C41H5302P2N ‘ H20: C, 73.30; H, 8.25;
N, 2.08; P, 9.22
Found: C, 73.27; H, 8.40;

N, 2.06; P, 8.83

H. Hectorite

Naturally occuring Na+-hectorite (California) was
obtained from the Baroid Division of National Lead Company.
The hectorite was used in a freeze-dried form and with a
particle size < 2pm. Chemical exchange of Na+ in the native
mineral with 1.0 fl Cu(N03)2 and subsequent chemical analysis
of the mineral for Cu2+ indicated the cation exchange capacity

was 70.0 meg/100 g.
1. Catalyst Precursors

1. Preparation of Cationic [Rh(NBD)diphos*JClO, Solution

E 'phos* = DIOP(+) (5), (3)-Prophos ( ), (3)-4-Me-Prophos

D1
59), SPIPHOS (48) or N-BOC-butaphos 58)]

[Rh(NBD)Cl]274 (18.44 mg, 0.0.4 mmol) and AgC104

(16.60 mg, 0.08 mmol) were allowed to react in a mixture of

4 mL CH C1 and 0.5 mL 95% EtOH or MeOH (0.1% H20) for 15

2 2
minutes in a N2 glove box. The resulting pale-yellow

solution was filtered through a fritted funnel. The funnel

181

was rinsed with 3 mL CH2C12. Then the filtrate was added

to a solution of diphos* (0.06 mmol) in 4 mL of 95% EtOH

or MeOH (0.1% H20). After 30 minutes, the orange-red solution
was concentrated to a small amount (ca. 3 mL) and then-
diluted to 20 mL with 95% EtOH or MeOH ((0.1% H20). The
solution was divided into two parts for homogeneous hydro-

genation and for preparation of [Rh(NBD)diphos*]+-hectorite.

2. Preparation of Cationic [Rh(COD)(R)-Prophos (5)1+C104
Solution

The cationic [Rh(COD)(B)-Prophos (5)] solution was
prepared by adopting the same method previously described.
The precursor [Rh(COD)Cl]§5was prepared according to the
literature method.

3. Preparation of [Rh(diene)diphosj]f-hectorite

[Dienez C00 or NBD; 0iphos* = DIOP(+) (g), (3)-Prophos (
(§)-4-Me-Prophos (48), SPIPHOS (48) or N-BOC-butaphos (52

H
In a typical experiment, 0.3 g of freeze dried
Na+-hectorite was suspended for an hour in 10 mL 95% EtOH
or MeOH (0.1% H20) in a N2 glove box. To this slurry was
added a 10 mL solution of [Rh(diene)diphos*]+C104 (0 04
mmol) prepared above. After 1 to 2 hours, the mixture was
filtered and washed with the same solvent (3 mL x 4) to
remove unexchanged rhodium complex. The [Rh(NBD)diphos*]+
-hectorite was dried by suction.
Elemental analyses:

[Rh(NBD)DIOP(+) (8)]T-hector1te, Rh 1.1 wt%;
1'
[Rh(NBD)(3)-Prophos (5)]-hectorite, Rh 0.91 wt%.

182

4. Preparation of Cationic[Rh(NBD)diphos*]ClO, in Solid
FormpIDiphos* = (3)-Proph0515) or (fi)-4-Me3Prophos(4§)]

[Rh(H80)C1]274 (13.83 mg, 0.03 mmol) and AgClO

4
(12.45 mg, 0.06 mmol) were allowed to react in a mixture of

4 mL CHZCl2 and 0.5 mL 95% EtOH for 15 minutes in a N2

glove box. The resulting pale-yellow solution was filtered
through a fritted funnel, and the funnel was rinsed with 3

mL CH2C12. Then the filtrate was added to a solution of
diphos* (0.06 mmol) in 4 mL of 95% EtOH. After 30 minutes,
the solution was concentrated to a small amount 1under‘ reduced
pressure and 4 mL of 95% ethanol was added and the solution was filtered.
The fine yellow-orange precipitate was formed by slow evapor-
ation of the ethanol. The orange precipitate was filtered,
washed with water and dried in vacuo. The yields of [Rh

(NBD)(R)-Prophos (5)]C104 and [Rh(NBD)(B)-4-Me-Prophos (59)]

31

C104 were around 90%. P NMR (72.88 MHz, 85% H3PO4 as

external reference; downfield shifts positive) 6(ppm);
[Rh(NBD)(3)-Prophos (5)]C104: 6 = 61.6 (dd, JRh_P=155 Hz),

42.9 (dd, JRh_P=155 Hz).JP_P=34 Hz, (COCl -20°C). [Rh

3’

(NBD)(3)-4-Me-Prophos(48)]C104: 6 = 60.4 (dd, J =155 Hz,),

Rh-P

41.7 (dd, JRh-P=155 HZ), JP_P=35 Hz, (CDC1 -16°C).

39

5. Preparation of [L-ester Rh(R)-4-Me-Prophos (49)]+ Solution
for 3IP NMR

 

3[Rh(NBD)(E)-4-Me-Prophos (4__9)]C104 (31 mg, 0.04 mmol)
and an excess of ethyl (Z)-a-benzamidocinnamate (g-ester)
(95 mg, 0.32 mmol) were dissolved in EtOO (4 mL) in a'10
mm NMR tube fitted with a silicon rubber septum. The solution

was prepared under N2 atmosphere in a N2 glove box. The

183

tube was attached to a vacuum line, evacuated, and hydrogen
was admitted at -20°C. The tube was kept at positive H2
pressure until the solution became deep orange-red, then
hydrogen was removed by two further freeze-pump-thaw cycles.

31

Finally, the tube was filled with argon. P NMR (72.88

MHz, 85% H3PO4 as external reference; downfield shifts posi-

tive) 6(ppm), 74.9 (dd, JRh_P=l62 Hz), 46.7 (dd, JRh-P=155
Hz), JP_P=46 Hz; 64.9 (dd, 0Rh_P=155 Hz), 59.7 (dd, JRh_P=158
Hz), JP_P=46 Hz, (CH300, -30°C).

J. X-ray Powder Diffraction Measurements

A Phillips X-ray diffractometer within-filtered Cu-
Ko radiation was used to determine d(OOl) basal spacing.
Film samples of the [Rh(NBD)diphos*]+-hectorite were pre-
pared by placing solvent suspensions onto glass slides and
drying them at room temperature. Diffraction patterns under
different conditions of solvation were obtained by allowing
the film to equilibriate under the solvent for 30 minutes
and then keeping thefilm wetted while under the X-ray beam.
The basal spacings of [Rh(NBD)diphos*J-hectorite complexes
[where diphos* is DIOP(+) (8), (R)-Prophos (5) or (3)-4-

Me-Prophos (48)] are shown in Table 37.

184 .
Table 37. Basal Spacings of [Rh(NBD)diphos*]+-hectorite

 

0(001) (A)

 

 

Solvent DIOP(+) (3)-Prophos (3)-4-Me-Prophos
_l§1__. (5) (49)

Air dry 18.3 A 17.0 A 16.1 A

MeOH (0.1% H20) 19.2 A* 19.6 A* 19.2 A

95% EtOH 19.2 A* 19.6 A* 19.2 A*

 

*Two orders of reflection were observed.

K. Hydrogenation Procedure

The prochiral substrate (4-8 mmol) was weighed into
the hydrogenation vessel. The catalyst solution or [Rh
(NBD)diphos*]+-hectorite and the required 20-30 mL of oxy-
gen-free MeOH (0.1% H20) or 95% EtOH were added to the
hydrogenation vessel in a N2 glove box. The vessel was
attached to a glass vacuum manifold with gas inlets, gas
burette, mercury leveling bulb, and mercury manometer. The
vessel then was purged by filling and evacuating with
purified hydrogen. The hydrogen uptakes were monitored
at 25°C and a total pressure of 740 torr. The hydrogenation
rates for rhodium(I) (diphos*) catalysts with different
substrates were presented in Tables 26, 27, 28, 29 and 34

in the Results and Discussion, Chapter 2.

185

L. Products Isolation

1. General Work-uprrocedure in the Hectorite System

After the hydrogenation was completed, the rhodium-
hectorite was filtered off and washed thoroughly with 95%
ethanol. The filtrate was evaporated in vacuo to give the
product.
2. General Work-up Procedure in the Homogeneous System

A. Method I - 95% EtOH or MeOH as a Solvent

After the hydrogenation was completed, 1-2 9 of 200-
400 mesh Dowex 50W-X2 cation exchange resin in the H+ form,
or 0.3-0.5 g of Na-hectorite was added to the solution
(under N2). After the mixture was stirred for 15-30 minutes
(resin) or 1-2 hours (Na-hectorite) it was filtered and
the resin or clay was washed thoroughly with 95% ethanol.
The filtrate was evaporated 1%: vaaMDto give the product.

B. Method II - Tetrahydrofuran as a Solvent

After the hydrogenation was completed, the solvent
of the reaction mixture was removed under vacuum to give
an oily residue. 5 mL of deoxygenated 95% EtOH was added
to the residue and the solvent was removed again. This was
repeated two more times to ensure that none of the THF sol-
vent remained. The resulting yellow residue was dissolved
in 25 mL deoxygenated 95% EtOH. The work-up procedure was
then completed by following Method 1.

M. Chemical Conversions

1

The chemical conversions were estimated by H NMR.

The chemical conversions of the corresponding amino acids

186

were larger than 97%.

N. 0ptica1 Yields

All optical rotations of the hydrogenated products
were recorded with a Perkin-Elmer.l4l automatic polari-
meter Aand a 10 cm pathlength micro-cell. Two measurements
were recorded for each sample by preparing two independent
solutions. The optical yeilds were calculated byLmimgthe
values of specific rotations of pure products which are

listed on Table 38.

Table 38. Specific Rotations of Pure Amino Acid Deriva-

 

 

tives a
Amino Acid Derivative Specific Rotation
N-Acety1-(R)-a1anine [612° +66.3° (c 2. , H20)
N-Acetyl-(§)-phenylalanine [a]26

-40.3° (c l. , MeOH)
-42.7° (c l. , MeOH)

0

+46.0° (c 1.0, EtOH)
N-Benzoyl-(§)-phenylalanine [a]26 0
0

GOOD

N-Benzoyl-(§)-phenylalanine [o1]26
ethyl ester

.N-Acetyl-(§)-tyrosine [6127D +51.5° (c 1.0, MeOH)
0,N-Diacetyl-(§)-tyrosine [6127D +45.5° (c 1.5, MeOH)
.N-Acetyl-3-(4-acetoxy-3- [01200 +40.7° (c 1.0, MeOH)

methoxyphenyl)-(§)-alanine

 

aData were obtained from reference 9.

APPENDIX

187

Anzz ommv oomm aa .m_oou e_ Amwv _ewep-oeweeeoo< eeeseoooam as: I .mm 665525

F

 

4dd-—quddd4d-—4dedddqd—dqd-.dddddddddudduu—ddddqq«de—ufidddddqdddd

. j i 1
\

 

 

Amwv _o_ep-oe_eeeoo<

 

 

188

Anxz .MM-85mw,pa .mpuou e2 Amwv deep-moowe-eoeoo_sneom cogeepooem «:2 I

P

. emoeemwd

 

 

 

 

 

 

 

 

 

 

1 . 11 11.3.1.1! *0
.
. . .
. .

o

 

‘ ' '55-l'llo l-l.-.Al’u 1 1 1| 1|.

3epsaoeee-eoeAVmeeom.

9"! I.

 

 

 

 

 

 

 

 

.. .. . . . “ r111. ..
1—1 u a d . _ u :1 — d - q q u q q q q d d — 1 q q — J 1 — d u d 1 q
& o 1 .-
414 .5319}??? xi!)
2
- 111 11 1 .11 11.11.1111,! 111111135111 1 1.11.1-1 -1 1111-1111 1.111. 1.1-11.11.11:- 1--1-... 11 . ail-1:1-.. -
115.
1 _
L
n

189

.32 83 898 ea 888 E :43 8253182362 85588. «:2 E .3 88:

o _ a n v n .a o n m

dddqddudddAd‘uq«uddddd‘ddddddddd__deddddddedqdudquJdudd«dud—dddAdddd4J—dd4ddddq_—

4

 

 

 

Amwv mpepxmou1mo_:oumu<

 

 

 

190

.988 5 2m: _Eeoeaez-@-~.7meo23838.234.76553358,-- .55585 ”.22 z

.Anzz ommv 85mm ea
_
Ema o p a n v m o u a o

dq-uq—q-dd-ddquddququddd-.q-j-q—~uqddd<dd—d-ddqqdfidduqudddqqq—dfijdddd4_.qd1dd44q—1jqddqqd—u-4.

.13]

. 8 5.55.:

7 L
I]

 

 

 

 

\ A?
mIu quAm

_
54.101281831818110
£8

 

191

m

:2: ommv comm um Sou c.- Amzcv 3.59.22 $037302? «:2 IF . mm menu-p...

.eeo .. _ a w . -

u-d‘dqqdddd.q11444d‘q—4d.ddd4141.1qqudud1d|—J(dlu

lm

 

 

 

 

 

 

Amwv _o_eoe_e<

”Io :@
tofu... zazofozzolzo
£0.

192

.32 83 ”Comm 5 8.58 E 3.3 8855-88-55 e.e-:88 «:2 2 .mm 8:9...

_
End a _ a n v m

_q-d-udqq.__11-_ddd—d-dfiu4fidq—uu-dqudq«—#dddu‘ud—~Jqqqqq4-q

3 )1/\11é\1 l

N
00 I
51w _ Jo Jo .
:wfuz quurufu
nIo .

Amwv _e_eeeFEe-oom-z

A...uqJ4

 

 

 

 

 

 

 

 

 

 

193

.MZS E :13 ooelmofieoezaéom-z co 358% as: I

2:: 33 053 pa .8. 8:3:

.-

fidu—_fiuuddfidddd-c_duddddqdfidudufiddddqqdjdu—quddqdfidqduqquddddddddd--«d«~]<uqqqqdfid-qujlqd—

3 -

 

 

 

Amwv ope_smeu_eoe_e5-uom-e

 

194

Amy mocaoeaAmv wosznuuoqm 36: .8 9:5:

 

 

03 on. .1 Ion 3n ova can .1 'n can I).
h F D b h P b b h b p h I b h b an >
.4. 3fl n11. H. or u-
In
A v
A v
1 1.53
A
A
2 fl
'3 1 0...
Ion Gun 3. O: O: M. . 3 I. i).
1 n . u s
. . . . . . on
A a I n a. 8 t J a.
. .
.
A
A
g 1 Co.“
5
2 fl
0...

 

 

 

I _o 1 a .

an .53
.332 J; .o 43 :43. 88 5 coin... as}?!
6.. .U\‘ 34. a. 3kg oC-a :hug fig

195

.mmmm

Am:

mogaoéumsncudmv mo 58303. 36:

. E 933...

 

 

 

 

 

 

 

 

 

000 an“? 00' Oflm 00m m\r_
~|r-.—>P>P»’F>P>ppP»L{.b>P>—rPPP«r.rp>-.b.JFFL.—.pwprb
Rm 4
«.— «mm hum
mmv .
nvm
mm? ..
obm
10.0“
Y
fi0.0&n
OWN QON and 64¢ 00 ”\t
n r P r P > bIL‘lLJI piqu‘:><~ odp <0 4:- wfk F . pd »‘ 4P .qLfl r dbqq pup . -- J—.
m n n . mm mm a
.n _ 3
am: 1
N a
n
mm on r
10.0“
T
T
mfiu [6.9.1
new—8 I rtwnc
mumV.P3.JDZ UZTIQ 2:..le NW.“ ”Mira—gram
.mwmflw “UT”. Us 1.10 ”—410 Nvum ¢ Gounoumn wafi—x—O
mam Hui: NMIm hwy—s $.30 21.—IO . Zguhuwum mma...

196

AMWV mogamuzn-uom-z $o_azuuwam mmmz

. mm 933”.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

um“ ooh gum mom mum m.:
_P_Fp _._ kLF». .. . r .».Jf__. _rmp_._4P_L4r1._._w1_!r_.d
g1 _ “mm «mm mm» ,
M mmo H
A. j
1 1.odm
J r
L I
4 r .
l ..o co,
cam awe cow own can ¢ 3
. p _hmv _ 4 Mm? _ L1. p _ _ . p _ r p .4 . _ . L meu — .d b111(‘r. . _TL 4, f a .
‘ . . mm... :3 «mm «mm tn . RN 1
4 T
L Pmm -
4 f
1 ..avom
L ..
; com
4 -, «we fl
_ . cw“ . . own mm“ Um" mm mg:
_.. QJQLMJ4 «.4 . N . . _ . _= ”1.1. 44. 4 .4 . . T _: L,_1. .2 . _:.1_
L m8 8“ MW MW mm. 2.: a: J m: m? 84L ___ r8 _ .1
1 m _ m as -
1 _vw _ Fv -
1 198
u ,
mFN um ..c.ooF

197

 

.;u_.a_ “e.e

.5
ul-

Al.

.¥¢_uuzo

Popuocwsm-uom-z uS .5333 33.

 

 

vo.n .ng

m. 4¢u "_ng
mm; _nnm «zpza

 

‘h

.

 

 

m
u:-

 

 

 

.. ..;.mo_
P.-
w_a a m\.
:_: 2:: orzrw _nnm "u4;:xm
;:"_ . sou¢_"¢_ 3Q m;\_a

::;_gwgm ”ma:

198

9mm.

5

00mm um mpuou cw Amv Mogqogm-Amv mo sagpomgm azz m_m .qm mg=m_m

Acpm_w=zouv
San. ON] 2| OT. ml 0 n

1 . I q . H V

N: _Nua-¢a

 

Amy mongogauAmv

 

 

«can...

199

m

 

 

u omy um Hugo
c“ oHu.LflmV mocaonmyamVfiasz;¢_ mo coflusfiom 2 go. o mo anypoomm «:2 mam .mm 093w“;
Au_m_$=zouv
Emu av av w ow co
_ q r _ _ _ W *q
_ _¢.x_.:<.__fl§%h~ mg:, 72
E P 2;. ._ ; 2 g 5m- _
N: mm_u¢- gm”

 

¢;o

 

 

 

 

Topic; I 2: 3 m 2.7;..—
a.NV +

200

v .oomm um mfiuou

 

 

 

aw oHu _flva mozmugmflomzvnm_ mo cofipsHOm 2 Ho. o mo saauoomm «22 mHm .om oysmflm
m :30
.xma nu on Na Any. m$ av
35.; _ fl. _ éége
~I ~m_u¢-;me
udu
e

o_u+fifim«v mozammmfiomzvgm_

201

.UOGH pm Macao a“ Ho mama Amwv mozmHmm gm mo :owusHom z Hc.o mo enauoomm mzz a .BM chum“;

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Hm
Au_mwwczouv
.taa m— mm mm m?
_ _ q _ _ _ — _4
1 t
s {H
m . u
N: mm" ¢-ma
N: omu¢-¢q {\lm
; 3
1| {‘-
m .s o
ms<
flu mama mommHmm am
an: mmum-mw .Nm ovfiumygan Away sag m.m~ .m
flu: mmum-mw .Nz oefium.;mn .ucv emu ~.o~ .m
an: omum-ma .Nz mefinm-cmw .ucv an; w.w~ .a
an: omua-aa .Nz avfium-:mw .uwv Ema c.mm .u
an: mmum-¢w .Nz H¢~um-;mm .uuv .smn ”.mq .m
an: omum-mn .Nz Hmfium-;¢a .ocv sag m.eq .<

‘202

 

Table m9. High Resqution Mass Spectnnlof (5)-4-Me-Pr0phos
(i9)
' SYMBOL VALENCE MASS UL LL
C 4 12.00000000 ** O
H 1 1.00782522 ** O
P 3 30.97376300 ** 0

UL OF ## MEANS COMPUTER CALCULATES LIMIT.

 

MASS w 1 459.21459000 ACC 0.005
RINGS AND DOUBLE BONDS: -3.0 T0 18.0
MASS DIFF PPM RDB FORMUL:
“ 6 '2'3 -0.00101 2.15 15.0 c H
4 B. 1 59 31 34 2

 

 

 

Table 40. High Resolution Mass Spectrmnof N-BOC-butaphos (g?)

 

SYMBOL VALENCE MASS UL LL
C 4 12.00000000 50 38
H 1 1.00782522 60 40
O 2 15.99491410 4 O
P 3 30.97376300 4 O
N 3 14.00307440 3 0

UL OF 4* MEANS COMPUTER CALCULATES LIMIT.

-MASS # 1 653.35352000 ACC 0.005
RINGS AND DOUBLE BONDS. -3.0 TO 18.0

MASS DIFF PPM RDB FCRMULA
653.35083 -0;00259 4.12 17.5 c H 0 P N
- £0 50 4 1 2
553.35517 0.00155 2.52 17.0 c H 0 P N
41 53 2 2 1

 

 

 

 

Table 4l.

203

High Resolution Mass Spectrwnof Benzyl-diol (59)

 

 

 

 

 

SYMBOL VALENCE MASS UL LL
C 4 12.00000000 ** O
a 1 1.00782522 *4 0
Q 2 15.99491410 ** 0
UL OF 4* MEANS COMPUTER CALCULATES LIMIT.
MASS # i 196.10965000 ACC 0.005
RINGS AND DOUBLE BONDS; ~3.0 TO 18.0
MASS DIFF PPM BILL gm...
196.10995 0.00030' 1.51 4.0 C H O ‘
' 11 16 3
Table 42. High Resolution Mass Spectnmuof (Q5)
SYMBOL VALENCE MASS UL LL
C 4 12.00000000 ** O
H 1 1.00782522 ** O
N 3 14.00307440 ** O
O 2 15.99491410 ** 0
UL OF 4* MEANS COMPUTER CALCULATES LIMIT.
MASS # 1 201.17156000 ACC 0.005
RINGS AND DOUBLE BONDS: -3.0 T0 18.0
MASS DIFF PPM RDB FORMULA
201.17020 -0.00136 6.76 2.0 C H N
7 19 7
201.17154 -0.00002 0.09 1.5 C H N O
9 21 4..l
201.17288 0.00132 6.57 1.0 C H N O
11 23 1 2
201.16752 -0.00404 20.09 -2.5 C H N O
4 21 6 3
201.16886 -0.00270 13.42 -3.0 C H N O
6 23 3 4

 

 

 

 

 

2 0.4

Table 43. High Resolution Mass Spectrum of N-BOC-aminodiol (0'6)

 

SYMBOL VALENCE MASS UL LL
C 4 12.00000000 ** O
H 1 1.00782522 2* O
H 3 14.00307440 ** O
O 2 15.99491410 *2 O

.9

0; OF 2% MEANS COMPUTER CALCULATES LIMIT.

1 -e a
l 1 ~ ~ “'
1

A 2 261.19633000 ACC 0.005
SINPS AND DOUBLE BONDS: -3.0 TO 18.0

I“

1: LU
' LU

l4

 

 

 

MASS DIFF PPM RDB FORMULA
2,1 1953 -0.00098 3.74 6.0 c H N
14 23 5
2:1 20122 0 00489 18.73 2.5 c H N
. 6 21 12
261.19669 0.00036 1.39 5.5 c H N 0
16 25 2 1
“61 19720 0.00087 3.33 -1.5 c H N o
1 21 14 2
261.19267 -0.00366 14.01 1.5 c H N 0
11 25 4 3
261 19854 0.00221 8.46 -2.0 c H N 0
h A 3 23 11 3
251 19401 -o.00232 8.88 1.0 c H N 0 ~01
__ - _2__2 _13_ 22 1 4
261 19988 0.00355 13.60 -2.5 c H N o
_ . .- 5 25 8 4
261 20122 0.00489 18.73 —3.0 c H N 0
7 27 5 5

BIBLIOGRAPHY

10.

BIBLIOGRAPHY

(a) Morrison, J.D.; Mosher, H.S., "Asymmetric Organic
Reactions", Prentice-Hall, Englewood Cliffs, New
Jersey, 121}.

(b) Scott, J.“.; Valentine, D. Jr., Science, 1974.
184, 943-952. A’”

 

(c) Valentine, D. Jr.; Scott, J.w., Synthesis, 1978,
329-355. "”

(d) Apsimon, J.H.; Sequin, R.P., Tetrahedron, 1979,
fl, 2797-2842. m

(a) Morrison, J.D.; Masler, H.F.; Neuberg, M.K., Adv.
Cata1., 1976, 5, 81-124.

(b) Caplar, V.; Comisso, 6.; Sunjic, V., anthesis,
198‘]. 85-ll6.

(c) -Merri11, R.E.. CHEMTECH, 19 1. 118-127.

«(‘1

Horner, L; Buthe, H; Siegel, H., Tetrahedron Lett.,
Lg§§. 4023-4026.

Horner, L; Siegel, H; Buthe, H. Angew. Chem, Lg§§,
, 942.

 

-§Q, l034, Angew. Chem., Int. Ed. Engl., TEEB, _

Knowles, H.S.; Sabacky, M.J., J. Chem. Soc., Chem.

Commun., L369, 1445-1446.

Morrison, J.D.; Benett, R.E.; Aquiar, A.M.; Morrow,
C.J.; Phillips, C., J. Am. Chem. Soc., 121), 2;, l30l-
l303.

(a) Dang, T.P.; Kagan, H.B., J. Chem. Soc., Chem.(bmm,
197l, 48l.

m

(b) Kagan, H.B.; Dang. T.P., J. Am. Chem. Soc., 121?,
a, 6429 -6433.

Knowles, H.S.; Sabacky, M.J.; Vineyard, 3.0., J. Chem.
Soc., Chem. Comm., l972, lO.

Fryzuk, M.D.; Bosnich, B., J. Am. Chem. Soc., 121],
22. 6262-6267.

Fryzuk, M.D.; Bosnich, B., J. Am. Chem. Soc., 1916,
100, 549l-5494.

'205

11.

12.

13.

14.

15.

16.

17.

18.

19.
20.

21.

22.

23.

24.

25.

206

King, R.B.; Bakos, J; Hoff, C.D.; Marko, L” J. Org.
Chem” [319, 44, 1729-1731.

 

Brown, J.M.; Murrer, B.A” Tetrahedron Lett., 1 79,
4858-4862. “’”

 

Riley, D.P.; Shumate, R.E” J. Org: Chem. 1980, 4_,
5187-5193. “’“

 

Brunner, H; Pieronczyk, N” Angew. Chem., 1913, 91,
655-656; Angew. Chem., Int. Ed. Engl., 1219, 18, 620-
621.

 

 

 

Samuel, 0; Couffignal, R.; Lauer, M; Zhang, S.Y.;
Kagan, H.Ba Nouv. J. Chim., 138), 5, 15-20.

Vineyard, B.D.; Knowles, w.s.; Sabacky, M.J.;
Bachman, G.L.; Neinkauff, D.J” J. Am. Chem Soc.,
121], 22, 5946-5952.

 

 

Dang, T.P.; Poulin, J.C.; Kagan, H.B” J. Qrganomet.
Chem., 1215,,21, 105-115.

Glaser, R.; Twaik, M.; Geresh, S.; Blumenfeld, J,
Tetrahedron Lett., L917, 4635-4638.

 

 

Achiwa, K., J. Am. Chem. Soc., 1916, 28, 8265-8266.

Ojima, I.; Achiwa, K.,J. Chem. Soc., Chem. Commun.,
1,2/U, 428‘430.

 

(a) Achiwa, K.; Kogure, T.; Ojima I., Tetrahedron Lett.
191]. 4431-4432.

(b) Ojima, I; Kogure, T.; Terasaki, T.; Achiwa, K.,
J. Org, Chem., 1218, 48,-3444-3446.

 

Hayashi, T.; Mise, T.; Mitachi, S.; Yamamoto, K2;
Kumada M., Tetrahedron Lett., 1216, 1133-1134.

 

 

Hayashi, T.; Mise T.; Kumada M., Tetrahedron Lett.,
121§. 4351-4354. .

Hayashi, T.; Katsumura, A.; Konishi, M.; Kumada, M.,
Tetrahedron Lett., [218, 425-428.

Ojima, I.; Kogure, T.; Achiwa, K.,J. Chem. Soc.,
Chem. Commun., 121], 428-430.

 

 

26.

27.

28.

29.

30.

31.

32.
33.

34.
35.

36.

37.

38.

207

Tamao, K.; Yamamoto, H.; Matsumoto, H.; Miyake, N.;
Hayashi, T.; Kumada, M., Tetrahedron Lett., 1977,
1389-1392. "”

Grubbs, R.H.; DeVries, R.A., Tetrahedron Lett, 1977,
1879-1880. “’”

 

Miyashita, A.; Yasuda, A.; Takaya, H.; Toriumi, K.;
Ito, T.; Souchi, T.; Noyori, R., J. Am. Chem. Soc.,
1289. 102, 7932-7934.

Fiorini, M.; Giongo, G.M., J. Mol. Catal., 1979, 5,
303-310. 2’”

(a) Onuma, K.; Ito, T.; Nakamura, A. Tetrahedron
Lett., 1919. 3163-3166.

(b) Hanaki, K.; Kashiwabara, K4 Fujita, J., Chem.
Lett., 121?, 489-490.

Babievske, K.K.; Latov, V.K.,Us;. Kh'm., 1 9, 38
1009, Inch, T.D., Synthesis, 2,», 4665%g3.
Izumi, Y., Angew. Chem., 121}, 83, .

 

Dumont, H.; Poulin, J.C.; Dang, T.P.; Kagan, H.B.,
J. Am. Chem. Soc., 121}, 95, 8295-8299.

 

Takaishi, M.; Imai, H.; Bertelo, C.A.; Stille, J.K.,

 

J. Am. Chem. Soc., 1918, 100, 264-268.
Achiwa, K., Chem. Lett, 1918, 905-908.

Baker, G.L., Dissertation, Dept. of Chem., Colorado
State University, 1 O, 180 pp.; Dissertation Abstracts
International, Vol. 4 , No.11, May 1281. Order No.
8110793.

Kagan, H.B.; Langlois, N.; Dang, T.P., J. Organomet.
Chem., 1915, 90, 353-365.

(a), Halpern, J., Trans. Am. Crystallogr. Assoc.
1339 l_4_9 59-70. I

(b) Chan, A.S.C.; Pluth, J.J.; Halpern, J” Inorg.
Chim. Acta., 121?, 37, L477-L479.

(c) Chan, A.S.C.; Pluth, J.J.; Halpern, J., J. Am.
Chem. Soc., 1980, 1 2, 5952-5954.

M—

 

 

Brown, J.M.; Chaloner, P.A., J. Am. Chem. Soc., 1289,
102, 3040-3048.

39.

40.

41.

42.

43.

44.

45.

46.

47.

48.

49.

.208

Halpern, J. ; Riley, D.P.; Chan, A.C.S.; Pluth, J.J.,
J. Am. Chem. Soc., 1911,,99, 8055-8057.

Chan, A.C.S.; Pluth, J.J.; Halpern, J., J. Am. Chem.
5°C., 1%, 102’ $952-$954.

 

 

(a) Ojima, I.; Kogure, T.; Yoda, N., Chem. Lett.,
1919. 495-498. ’

(b) Ojima, I.; Kogure, T.; Yoda, N., J. Org. Chem.,
19§9. 5;. 4728-4739.

Halpern, J.; Okamoto, T.; Zakhariev, A., J. Mol. Catal.,
191], ;, 65-68.

Koenig, K.E.; Knowles, N.S.; J. Am. Chem. Soc., 1978,
199. 7561-7564. "”

 

Detellier, C.; Gelbard, G.; Kagan, H.B., J. Am. Chem.
Soc., 1918, 100, 7556-7561.

 

Grim, R.E., "Clay Mineralogy", 2nd Ed., McGraw-Hi11
Book Company, New York, NY, 1968 , pp. 77-92.

(a) Lahav, N.; Shani, V.; Shabtai, J., Clays and
Clay Minerals, 1918, 99, 107-115.

(b) Endo, T.; Mortland, M.M.; Pinnavaia, T.J., Clays
and Clay Minerals, 1989, 28, 105-110.

(c) Yamanaka, S.; Brindley, G.N., Clays and Clay
Minerals, 1919,‘91, 119-124.

(d) Yamanaka, S.; Brindley.G.N.;Hattori, M., Clays
and Clay Minerals, 1999, 293 281-284.

Lussier, R.J.;Magee, J.S.; Vaughan, D.E.W.,

(a) Pinnavaia, T.J-; Nelty, P.K., J. Am. Chem. Soc.,
191§. 21, 3819-3820.

(b) Pinnavaia, T.J.; Nelty, P.K.; Hoffman, J.F., in
"Proceedings of the International Clay Conference",
191§, Applied Publishing Ltd., Wilmette, Illinois,
pp. 373-381.

(c) Quayle, W.H.; Pinnavaia, T.J., Inorg. Chem.,1919,
1g. 2840-2847.

Pinnavaia, T.J.; Raythatha, R.; Lee, J.G.S.; Halloran,
J.; Hoffman, J.F., J. Am. Chem. Soc., 1919, 101, 6891-
6897.

50.

51.

52.

53.
54.

55.

56.
57.

58.

59.

60.

61.

62.

63.

64.

65.

66.

‘159, 255-262.

209

Raythatha, R.H.; M.S. Thesis, Dept. of Chem., Mich-
igan State University, 1978.

Farzaneh, F.; Ph.D. Dissertation, Dept. of Chem.,
Michigan State University, 198).

DeVries, R.A.; Ph.D. Dissertation, Dept. of Chem.,
Michigan State University, 1980,

Baltazzi, E.Q., Rev., Chem. Soc., 1985, 9, 150-173.

Herbst, R.H.; Shemin, 0., "Organic Syntheses", Collect.
Vol. 1; Wiley; New York, NY, 1943, p. 1-3.

Carter, H.E.; Risser, N.C., J. Biol. Chem., 1941,

 

Dakin, H.D., J. Biol. Chem., 1989, 82, 439-446.

 

Harrington, C.R.; McCartney, N., Biochem. J., 1927,
852-856. “’V

 

Levy, G.C.; Nelson, G.L., "Carbon-13 Nuclear Magnetic
Resonance for Organic Chemists", Niley-Interscience,
N.Y., 1972.

Horeau, A., Tetrahedron Lett., 1989, 3121-3124.

 

Sinou, D.; Kagan, H.B.; J. Organomet. Chem., 1976, 114,
325-337. “"

Gelbard, G.; Kagan, H.B.; Stern, R., Tetrahedron, 1976,
88, 233-237. "‘

 

MacNeil, P.A.; Roberts, N.K.; Bosnich, B., J. Am. Chem.
Soc., 1981, 103, 2273-2280.

“-

 

Schrock, R.
2

.; Osborn, J.A., J. Am. Chem. Soc., 1916,
98, 2134- l

R

43.
Poulin, J.C.; Dang, T.P.; Kagan, H.B., J. Organomet.
Chem., 1915, 81, 87-92.

Masuda, T.; Stille J.K., J. Am. Chem. Soc., 1918, 10
268-272.

 

 

(a) Chan, A.S.C.; Halpern, J., J. Am. Chem. Soc.,
1989, 102, 838-840.

(b) Brown, J.M.; Chaloner, P.A., J. Chem. Soc., Chem.
Commun., 1989, 344-346.

67.

68.

69.

70.

71.

72.

73.

74.

75.

210

Slack, D.A.; Baird, M.C., J. Organomet. Chem., 1977,
142, C69-C72. "’V

Brown, J.M.; Chaloner, P.A., J. Am. Chem. Soc., 1978,
100, 4307-4309. “’”

 

Eliel, E.L., "Stereochemistry of Carbon Compounds",
McGraw-Hill: New York, 1988; Chapter 8.

Achiwa, K.; Ohga, Y.; Iitaka, Y., Tetrahedron Lett.,
1918, 4683-4686.

 

Mori, K.; Takigawa, T.; Matsuo, T., Tetrahedron, 1979,
99, 933-940. “’V

 

Morrison, J.D.; Masler, w.F., J. Org. Chem., 1974,
270-272. “’”

 

Bianco, V.D.; Doronzo, 5., "Inorganic Syntheses",
Vol. XVI; McGraw-Hill Book Company, New York, NY,
1918, P. 161-163.

Abel, E.N.; Bennett, M.A.; Wilkinson, 6., J. Chem.
Soc., 1989, 3178-3182. __—

Chatt, J.; Venanzi, L.M., J. Chem. Soc. A., 1957,
4735-4741. -‘ “’V