. .
«madnwwmmz

3.303.:

I Vur . . . . . . . .N. gr
. us;
9 t
. . I v.
’3‘ i s

Lani.» .
9.3%.; . ..
. .r _ .

. Lu . _ . u.;§«....bfi.niz

. _ gain.

. . .. do .
.. , in... .

.U

1.
5.2.3:...

1.21%.}!
2.1..
I}. . . . . . [PI 2" g .
t In. . II
, . . . . hwpafiaa §.fi§:aflfilu ibixio
. . 2311:... run :1”. iii...
u 21...): 4.1). “‘1‘: a, t\~..o
i1»? . 5.: 3.13m..- . Je
_ . . . Jinn. .. . ..
its... .
I: n
. 4
5.: (:3. 3
«.311 :5 v... ..
a.» . . 1.3.4.. .9...

. .. 23....
. . 2...»... 2.
2:. 2..
:29 .i Ln ..
I: a: 311W?
.
.2...»

.0100".

. . x
3....» . . . . Juan»... 3...)...
E. .2. .3. . . , ,

.GJW 5 .2 . , . 5.:

i . v.5.»

. sanmfimnr»
$3.59...“ “at; ,
Evans $.52 ..
.6 Ha
u. . 1.
54.9mm?»

I... b 9|...

305.. :25, .33.

.iifmt 1: 3'1“)».
My.“ tuft 3i! .

v til... 03. 1.

i5... va .

inn-ft...

1.1 23.9.

5.5.5\|'II
. 1A..)

\I.l 6.:
c. 93:}! 1. \ S:

 

.3
...:.
. 1.. alg§uu$¢
SI 2).;
S. i315». .2. l
. . :1}... by

.l.‘!.(\!..
. nix}?

)7... f. s... :5.

,fmmgfiklégdfi. V V . : . .H iapgfifi.

 

C

minim WWI

3

This is to certify that the
dissertation entitled
Preparation of Aluminum Enoiates and
Boron Enoiates Via Aluminum Amides
and Boron Phosphides

presented by

Robert Eighanian

has been accepted towards fulfillment
of the requirements for

PhoDo degreein ChemiStY‘y

Major professor

Date November 15, 1995

 

MS U is an Affirmative Action/Equal Opportunity Institution 0-12771

 

 

 

 

LIBRARY
Michigan State
University

 

 

 

PLACE N RETURN BOX to remove thin checkout from your record.

TO AVOID FINES Mum on or bdore date due.

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

MSU leAn Nfirmetive WM Opportunity lnetttwon
mm:

 

PREPARATION OF ALUMINUM ENOLATES
AND BORON ENOLATES VIA ALUMINUM
AMIDES AND BORON PHOSPI-IIDES

By

Robert Blghanian

A DISSERTATION

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

DOCTOR OF PHILOSOPHY

Department of Chemistry
1 995

ABSTRACT

PREPARATION OF ALUMINUM ENOLATES
AND BORON ENOLATES VIA ALUMINUM
AMIDES AND BORON PHOSPHIDES

By

Robert El ghanian

The main objective of this dissertation is to investigate the reactivity of aluminum
and boron bases to carbonyl substrates. For this purpose a series of aluminum amides,
boron amides, and boron phosphides were synthesised and tested to determine whether
enolate formation can be accomplished using these reagents. This work provides the first
observation of an aluminum enolate prepared using an aluminum amide. This work also
provides the first observation of a boron enolate prepared via a boron phosphide.

A series of dimeric and monomeric aluminum amides (compounds I-XV) were
synthesized and evaluated for their ability to react in an acid/base reaction using a variety
of ketones. Compounds I, II, and XV were unreactive to ketones. Compounds III-XIV
reacted with ketones to generate the corresponding aluminum enolates. Enolates
generated from the reaction of III with pinacolone and 3-pentanone were selected as
model systems and condensed with benzaldehyde. The stereoselectivity of the enolate
formation and the aldol reaction were investigated using lH-NMR spectroscopy.

Likewise, a series of boron amides and phosphides (II-III,V-XI) were
synthesized and evaluated for their ability to react in acid/base reactions using a variety of
carbonyl substrates. Boron amides (II-III) were unreactive to ketones. Boron
phosphides reacted to a variety of carbonyl substrates. The stereoselectivity of the aldol

condensation of 3-pentanone was determined to be the syn stereoisomer using 1H-NMR

spectroscopy. 31P-NMR spectroscopy was used to determine the extent of enolization

throughout this study.

To:My Parents

iv

Acknowledgments

I would like to express my deepest appreciation to Dr. Rathke for helping me and
guiding me throughout the course of this study. He is an example of a great teacher, a
great researcher, and a great advisor. I would also like to thank the other members of my
guidance committee, Drs. Reusch, Eick, Watson, Schwendeman, and Malechka. Thanks
to Dr. Azadnia for his friendship and many hours of conversation, chemistry as well as
Farsi. I would like to thank members of the Reusch and Baker labs for their friendship.
Many thanks to Dr. Long Le and Kermit Johnson for help with NMR experiments and
training. Thank you much to Lisa Dillingham and Diane Frost for their friendship and
excellent secretarial skills. I would like to thank Kyung II Kim (aka. Mr. Kim) and
family for their friendship. To Phil and Karen (Schultz & Inman), thanks for endless
hours of "fun".

I would like to thank my parents, Ruhollah and Roza Elghanian, my sisters Betty,
Iren, and Evelyn. Thank you to the parents of my wife, Frank and Jan Lill, for all their
support and many enjoyable visits we had here in Michigan.

Lastly, I want to thank my dear wife Debbie for all of her love, support, and
encouragement during the course of this endeavor. Debbie, it was not easy and I could

not have done it without you either.

Thank you all

Robert

TABLE of CONTENTS

List Of Tables vii
List Of Figures ix
List Of Schemes xi
List Of Abbreviations xii
Introduction 1
Literature Cited 5
Chapter I: Preparation of Boron Enolates From Boron Phosphides 6
Literature Review of Boron enolates 6
Preparation of Boron Amides 21
Synthesis of Boron Phosphides 27
Results and Discussion 29
Conclusions 37
Experimental 38
Literature Cited 50

Chapter IlzPr'eparation of Aluminum Enolates From Aluminum Amides 54

Literature review of Aluminum Enolates Structure, Properties, and Methods

of Synthesis 54
Literature review of Aluminum Amides Structure, Properties, and Methods

of Synthesis 59
Results and Discussion 63
Conclusions 86
Experimental 87

Literature Cited 97

vi

List of Tables

Chapter I

Table 1.1 Representative results for the reaction of alkylboranes with unsaturated

ketones after Matsumoto et al.

Table 2.1 Representative results for the reaction of boron sulfides with ketene
after Masamune et al.

Table 3.1 Representative results of the reaction of dibutylboron triflate
with silylenolethers after Kuwajima et al.

Table 4.1 Representative results for the reaction of dibutylborontriflate
with silylenolethers. After Wada et al.

Table 5.1 Representative results of the reaction of alkylboranes with
diazo ketones after Masamune et al.

Table 6.1 Representative results for the enolization of ketones and a thioester
after Evans et al.

Table 7.1 Representative results for the enolization of ketones with
B-X-9-BBN and Chxz-BX after Brown et al.

Table 8.1 Representative results for the enolization of ethylesters and
tertiaryethylamides after Brown et al.

Table 9.1. Reaction of Dicyclohexyl(di-t-butylphosphino)borane with carbonyl
substrates.

Table 10.1 Reaction of ChszPth with Carbonyl substrates

Table 11.1 Survey of the Reaction of heteroatom substituted boron phosphides
with pinacolone.

Chapter II

Table 1.2 Representative Results For The Reaction Of a-Haloketones With
Zinc and EtzAlCl After Maruoka et al.

Table 2.2 Representative results for reaction of an a—fluoroenolphosphonate
with LAI-I after Kurobushi et al.

vii

8

10

ll

12

14

15

17

33
35

36

57

Table 3.2 Representative results for reaction of trimethylaluminum with
ketones after Seebach et al.

Table 4.2 Scope Of Enolization, Stereoselectivity and Regioselectivity Of

[11. Experiments were carried out with SOymoles of III in benzene at 25C.

Table 5.2 Summary Of The Reaction Of Aluminum Amides Having
Non-Nitrogen Bridges With Pinacolone.

Table 6.2 Summary of Chemical Shift Data of Enolates Generated by

the reaction of aluminum amides having Non-nitrogen bridges.
Chemical shifts are reported in ppm.

viii

69

71

74

List of Figures

Introduction

Figure 1 X-Ray structure of LDA after Willard et al. 2
Figure 2 Generalized Transition State Structure For Metal Amides 3
Chapter I

Figure 1.1 S-trans and S-cis conformations of boron enolates after
Goodman et al. 6

Figure 2.1 31P-NMR spectrum of Dicyclohexyl(di-t-butylphosphino)borane 30

Figure 3.1 31P-NMR spectrum of Dicyclohexyl(di-t—butylphosphino)borane after

addition of MEK and standing at 25C for 1 hour 31
Figure 4.1 31P—NMR spectrum of Dicyclohexyl(di-t-butylphosphino)borane after
addition of diisopropyl ketone and standing at 25C for 24 hours 32
Chapter I]

Figure 1.2 1H-NMR spectrum of Dimethylalurninumdiethylamide after
addition of pinacolone and standing at 25C for 3.5 hours 76

Figure 2.2 1H-NMR spectrum of Dimethylaluminumdiphenylamide after
addition of one equivalent of pinacolone at 25C and immediate analysis 77

Figure 3.2 1H-NMR spectrum of Dimethylaluminumdiphenylamide after
addition of one equivalent of pinacolone at 25C and standing for 4 hours 78

Figure 4.2 ll—i-NMR spectrum of Dimethylaluminumdiphenylamide after
addition of one equivalent of 3-pentanone at 25C and standing for 15 minutes 79

Figure 5.2 1l-l-NMR spectrum of Dimethylaluminumdiphenylamide after

addition of one half equivalent of pinacolone at 25C and analysis 80
Figure 6.2 lH-NMR spectrum of Enolate 11 81
Figure 7.2 1H-NMR spectrum of the aldol condensation of Enolate II 82

ix

Figure 8.2 ll-I-NMR spectrum of the precipitate resulting from the reaction
of Compound X with pinacolone in hexane.

Figure 9.2 1l-I-NMR spectrum of the supematent of the reaction of X with
pinacolone.

Figure 10.2 1l-I-NMR spectrum of the exchange of Enolate IV with MezAlCl

Figure 11.2 Apparatus used in the synthesis of aluminum amides

Figure 12.2 Sample removal from flask A.

85

95

List of Schemes

Chapter 1

Scheme 1.1 Stereoselectivity of boron enolates 7
Scheme 2.1 Reactions of (CF3)zBN(CH3)2 18
Scheme 3.1 Proposed Mechanism for the rearrangement of boron enolate after
Burger et al. 19
Scheme 4.1 Carbon to oxygen rearrangement for alkylaminoborane 19
Scheme 5.1 Resonance stabilization of boron amides 20
Chapter II

Scheme 1.2 Proposed Mechanism For The Reaction Of 111 With one Equivalent of
Pinacolone 65

Scheme 2.2 Proposed Mechanism For The Reaction Of X With one Equivalent of
Pinacolone 73

xi

9—BBN

1 lB-NMR
i-Bu

t-Bu

13C NMR
Chex

ether

NON EE

List of Abbreviations

Borabicyclo[3 .3. 1 ]nonyl
Boron(1 13) NMR
iso-Butyl

Tertiary Butyl

Carbon (13C) NMR

cyclohexyl
doublet

chemical shift

Ethyl

equation

Diethyl ether

1 ,1 ,13,3,3-Hexamethyldisilazane
Proton (1H) NMR

hertz

Lithium aluminum hydride
Lithium diisopropylamide
Lithium isopropyl cyclohexylamide
Multiplet

Methyl

N—ethylethanolamine

N—silyl, O-silyl, N-ethylethanolamine

xii

NMR

31PNMR

c—pentyl

i-Pr

TMSCI
Tos

Methyl ethyl ketone

Nuclear Magnetic Resonance Spectroscopy

N-tosylethanolamine
Phosphorous (31p) NMR
cyclopentyl

Phenyl

Isopropyl

quartet

singlet

triplet

Tetramethyl pipridine
Trimethylchlorosilane

Tosyl

xiii

Introduction

Carbon-carbon bond formation is the fundamental step in the construction of
organic molecules in the laboratory. Many carbon-carbon bond forming processes
involve the reaction between a nucleophilic carbon species and an electrophilic carbon
species. The nucleophilic species generated upon deprotonation of the carbon alpha to a
carbonyl functional group is called an enolate. The most useful method for the
preparation of enolates is the enolization of the appropriate carbonyl compound via the

acid-base reaction of the carbonyl acid with a lithium dialkylamide base (eq.1). (1)

 

 

-’u\
o (i fRz on
I
/u\/H + LiNR3—> A /H -—>/§ + RzNH (1)
C
I

The resulting enolate in equation 1 is a lithium enolate. The regioselectivity and the
stereoselectivity of this reaction are rationalized by the six centered transition state I.(2)
There are several considerations in this acid-base reaction: a)-coordination of the metal
cation is known to enhance the acidity of the carbonyl ligand(3), b)-the amide anion is a
powerful base (pKa~40) c)- the concurrent coordination of oxygen and nitrogen atoms in
the cyclic transition state allows for stereoselective deprotonation of the substrate. This

rationale has scarcely been applied to the development of bases other than lithium and in

2
particular boron or aluminum bases have not been explored as potential bases for enolate

chemistry.

The chemical literature on the reactivity of metal bases analogous to the lithium
amides is rather scanty and little is known about their reactivity to carbonyl compounds.
However lithium dialkylamides have been extensively studied due to the prominent role
they play in organic synthesis. Studies based on the colligative properties(3a) and
solution NMR(3b)spectroscopy on the structure of lithium dialkyl amides have shown
that cyclic oligomers exist in both donor and non donor solvents. The structure of LDA

is believed to be the THF solvated dimer shown by Willard's X-ray structure in Figure
1(4)

 

Figure 1. X-Ray structure of LDA after Mllard et al.
The solution NMR studies have also revealed that for lithium isopropylcyclohexylamide
(LICA) an equilibrating set of stereoisomeric dimers (cis,trans) is present.(5) The latter
study suggests the presence of weak intermolecular interactions and their dynamic nature
in solution. Studies on other metal amides have demonstrated that magnesium, zinc and
aluminum amides are strongly associated in solution through nitrogen bridges.(6)
Magnesium and zinc amides are known to react only sluggishly(7) or incompletely (8)

with simple carbonyl compounds.

3
The rationale for our research centers on the consideration of transition states

analogous to I for a generalized metal amide, (R 2M-NR2) 11 shown in Figure 2.

 

 

r' R2 ‘
O"M\I:IR2
I .
XC/H
_ .1
11

Figure 2. Generalized Transition State Structure For Metal Amides

The objective of our study was to discover bases which can be used in an analogous
manner to lithium amides to prepare enolates of metals more electronegative than lithium.
The advantage of such bases would be that in contrast to lithium, aluminum and boron
are more powerful Lewis acids. From previous work performed in our laboratory it is
known that metal complexation with carbonyl compounds enhances the alpha proton

acidity; hence a weaker base can be effective (eq la, lb).(9)

 

 

1) MgClg,Et3N
EtOZCCHzcogEt > (EtOZC)3CHCOPh (1a)
2) PhCOCI
85%
l) EigN
EtOZCCHgCOZEt 7 (EtOQC)3CHCOPh (1b)
2) PhCOCl
0%

In our review of the literature on the preparation of boron enolates, we found the current
methodology for the preparation of boron enolates uses a dialkylboron halide (Lewis

acid), a hindered tertiary amine (base) and a carbonyl substrate to prepare boron enolates

4
with good success. This methodology has now been extensively studied by Brown et al.

(eq2).(10v ll)

1 R BCl
R’"\/ )_2_._.' R + R’K\/ (2)
2) E13N,0C
E Z

Our approach would simplify this procedure by incorporating both features of a base and
a Lewis acid into one molecule and reducing the number of reagents employed in the
reaction from three to two. In contrast to the general method available for the preparation
of boron enolates, we did not encounter an analogous method of preparation for
aluminum enolates. This prompted us to undertake our search for aluminum amides
hoping that we could emulate the success achieved with lithium amides.

We envision that in addition to the simplification of the current methodology, the
advantage of such bases may be that in contrast to lithium, aluminum and boron are
capable of accommodating ligands on the metal center due to their multivalency. This
property allows one to carry out double stereodifferentiation on the stereochemical
outcome of the aldol condensation reaction in the presence of appropriate chiral ligands
on the metal center. The other attractive feature of such metal amides is that if a direct
route to enolization of the carbonyl substrate can be devised then this route may be a

general and useful one for the preparation of enolates.

Literature Cited

1) (a) Seebach, D. Angew. Chem. Int. Ed. Eng. 1988, 27, 1624. (b) Stowell, J.C.
Carbanions in Organic Synthesis, Mley: New York, New York, 1979. (c) Asummetric
Synthesis; Morrison, J.D., Ed.; Academic Press: New York, 1983; Vols. 2 and 3. (d) d'
Angelo, J. Tetrahedron 1976, 32, 2979. (e) Heathcock, C.H. in Comprehensive
Carbanion Chemistry; Buncel, E.; Durst, T., Eds. ; Elsevier: New York, 1980; Vol B,
Chapter 4.

2) Ireland, R.E.; Mueller, R.H.; Willard, A.K. J. Am. Chem. Soc. 1976, 98, 2868.

3) Houghton, R.P.; Metal Complexes In Organic Chemistry, Cambridge University Press:
Cambridge, 1979. (a) Bauer, W.; Seebach, D. Helv. Chim. Acta. 1984, 67, 1972. (b)
Collum, D.B. Acc. Chem. Res. 1993, 26, 227.

4) Willard, P.G.; Salvino, J.M. J. Org. Chem. 1993, 58, 1.

5) De Pue, J.S.; Collum, D.B. J. Am. Chem. Soc. 1988, 110, 5518. ibid, 5524.

6) Lappert, M.E.; Power, P.P.; Sanger, A.R.; Skivastava, R.C.; Metal and Metalloid
Amides; John Wiley and Sons: New york, 1980.

7) Kraft, M.E.; Holton, R.A. Tetrahedron Lett. 1983, 24, 1345.

8) Hansen, M.M.; Bartlett, P.A.; Heathcock, C.H. Organometallics, 1987, 6, 2069.

9) Nowak, MA. Ph.D Dissertation, Michigan State University, 1983.

10) (a) Brown, H.C.; Dhar, R.K.; Bakshi, R.K.; Pandiarajan, P.k.; Singaram, B. J .Am.
Chem. Soc. 1989, II I, 3441. (b) Brown, H.C.; Dhar, R.K.; Ganesan, K.; Singaram, B J.
Org. Chem. 1992, 57, 499. (c) Brown, H.C.; Dhar, R.K.; Ganesan, K.; Singaram, B J.
Org. Chem. 1992, 57, 2716. (d) Brown, H.C.; Ganesan, K.; Dhar, R.K. J. Org. Chem.
1992, 57, 3767. (f) Ganesan, K.;Brown, H.C. J. Org. Chem. 1993, 58, 7162.

11) (a) Ganesan, K.; Brown, H.C. J. Org. Chem. 1994, 59, 2336. (b) Ganesan, K.;
Brown, H.C. J. Org. Chem. 1994, 59, 7346.

CHAPTER 1: Preparation of Boron Enolates From Boron Phosphides

Literature review of boron enolates

Several metal enolates have been prepared and studied in the past three decades
including B, A], Mg, Sn, Ti, Zr, Cu, and Zn. The studies on these enolates have revealed
that boron enolates are among the most stereoselective agents in organic synthesis“)
Goodman et al. have carried out theoretical calculations on the structure of boron
enolates.(2) S-trans structures for boron enolates having hydrogen ligands on boron were

found to be the more stable conformation and S-cis the less favored (Figure1.1).

H H

H H
S-trans S-cis
Figure 1.1 S-trans and S-cis conformations of boron enolates after Goodman et al.
These calculations have also revealed planar enolate structures resulting from the
conjugative delocalization of 4 pi electrons over the olefinic carbon, the enol oxygen, and
the coordinatively unsaturated trivalent boron. The most useful reaction of these enolates
is the aldol condensation of the enolate with an aldehyde. Evans and coworkers have
determined that the enolate stereochemistry and the resulting aldol product
stereochemistry are intimately related.(3) Evans has shown that the reaction of E boron
enolates produces anti aldol products, while Z enolates react to give syn aldol products

stereoselectively (Scheme 1.1).

OBRZ O O OH

R + HIJLH ——'—V M H1 I
E anti
OBR2 O O OH

R \ + Ph’u‘ H ———> /u\/\ Ph
Z syn

Scheme 1.1 Stereoselectivity of boron enolates

It was also found that the enolate stereochemistry is dependent on the structure of the
ketone. However, regardless of the structure of the ketone consistent correlation was
observed between the enolate stereochemistry and the aldol product stereochemistry
(Table 6). As a result of these studies and studies done by other workers in this area
several routes have become available for the synthesis of boron enolates.
l-Conjugate Addition

Suzuki and Brown et al. have prepared unsymmetrical ketones via the conjugate
addition of trialkylboranes to methyl vinyl ketone(4a)(eq2). It was later demonstrated
that this reaction proceeds via a free radical mechanism to the enolate.(4d) The
stereochemistry of the enolate depends on the enone substituents but without useful
trends in stereochemical control.(5)

CH3
' HoO
CHg=CHCOCH3 .3311, RCHgCHZC— 03122 _:_> R(CH2)2COCH3 (2)

Mukaiyma et al. have prepared boron enolates via the 1,4 addition of a dialkyl boron
thiolate to methyl vinyl ketone (eq3).(6) The aldol product was isolated in 92% yield in
reaction with benzaldehyde. The intermediate enolate was quenched with methanol and

the B—phenylthiobutan—Z-one was recovered in 86% yield.

8

CH3 RCHO
CH2: CHCOCH3 M PhS-CH3CH-= c: ———> Aldo] (3)
OBRg

Boldrini and Matsumoto et al. have utilized the 1,4 addition of borane reagents to (1,8

unsaturated ketones in order to prepare Z boron enolates using a variety of ligands on

 

boron (eq 4).(7)
H o
OBLo
'M LoBH
R R2 " a: R‘ / ,
L2=988N R-
Catechol

" PhCHO
———>

Aldol
(syn)

(4)

This is a particularly useful method since regioselectivity and stereoselectivity can be

achieved in absence of byproducts other than the ligands on boron. The stereoselectivity

of the enolate formation was reported to be greater than 99:1, in all cases in favor of the Z

enolate. Representitive results are shown in Table 1.1.

Table 1.1.Representative results for the reaction of alkylboranes with unsaturated ketones

 

 

after Matsumoto et al.
9BBN Catechol
3' 32 %Yield Z/E %Yield syn/anti
H Ph 97 96/4 98
Me Ph 99 99/1 98 75/25
Ph Ph 92 97/3 97 78/ 12

2)—13. Addition

Mukaiyama and Masamune et al. have prepared boron enolates of thioesters via

the acyl addition of an alkyl boron thiolate to a ketene (eq 5).(8,6) Representative results

9
are shown in Table 2.1. This method complements the stereoselective preparation of

tert-butyl thioester enolates available through enolization of the thioester by using boron

triflate reagents that are known to produce the E enolate stereoselectively. (8)

 

RzzB-SR3 R1 OBR2 RCHO
R2=c-Pentyl SR3 (syn)
R3=1-BU

Table 2.1. Representative results for the reaction of boron sulfides with ketene after
Masamune et al.

Aldehyde %Yield M

R=C6H5 78 7/93
C3H7 75 5/95
(CH2)2Ph 65 5/ 95

Ignsmetzlljation

This method takes advantage of the exchange of a previously formed metal or
metalloid enolate with the metal of choice in order to prepare the desired enolate.
Kuwajima and coworkers have shown that treatment of (Z)-trimethylsilylenol ethers with
dibutyl boryl trifluoromethane sulphonate generates the corresponding Z boron enolate
after a short time at low temperature. This enolate was then reacted with a variety of

aldehydes in order to produce aldol condensation products with excellent syn selectivity

 

(eq 6).(93)
R1 08M“ BuzBUI‘f R1 OBBU2
R2 \ > R2 \ fl Aldol (6)

(syn)

10
The disadvantage of this method is that trimethylsilyltriflate has to be removed prior to

the aldol reaction since it also participates in the reaction via a silicon mediated aldol.

Table 3.1 provides some representative examples.

Table 3.1 Representative results of the reaction of dibutylboron triflate with

silylenolethers after Kuwajima et al.

 

3‘ 32 E/Z R %aldol syn/anti
CH3 H 96/4 Ph 63 95/5
C4H9 H 95/5 Ph 74 94/6
Can H 91/9 Ph 80 91/9
C6HSCH2 H 95/5 Ph 82 95/5

Wada has compared the reaction of dialkylboron triflates and dialkylboron halides with

silylenolethers to give the corresponding boron enolates under mild conditions (eq 7). (9b)
OSlMC 3 OBRQ

BUQBX

 

4; + MegsiX
X=Cl,Br,OTf (7)

The reaction of the triflates was the faster reaction in the cases studied and produced

exellent yields and good selectivity. Representitive results are shown in Table 4.1.

11
Table 4.1 Representative results for the reaction of dibutylborontriflate with
silylenolethers. After Wada et al.

 

Silvlenol ether %Aldol syn/anti
081MB 3
90 25/75
OSiMe 3
88 70/30
Ph

Hoffmann and Froech generated the lithium enolate of acetaldehyde and transmetallated
it with chlorodimethoxyborane to give the enolborate (eq 8).“0) These enolborates were
reacted with aldehydes to generate a protected aldol-acetal product. These enolates are
not stable at room temperature and polymerize rapidly, and the aldol-acetal product was
then allylated with allylmagnesium bromide to prepare diols in 62-77% overall yield and

70/30 anti/syn selectivity.

on M 0.30 R OBOMeg RCHO
_ c - _
‘34 , >—< __, A... (8)

R H R H l

 

Diol
4)-1.2 Hydride or Alkyl Tmsfer
Transfer of a hydride or an alkyl group from a boron reagent to an alpha diazo
ketone in a 1,2 manner has also been utilized as a method to prepare enolates of boron.
Hooz, Brown, and Masamune et al. have prepared boron enolates by the reaction of a-
diazoketones with a variety of organoboron reagents (eq 9).(1 I) This method allows one
to prepare both E and Z stereoisomers of the enolate. This reaction proceeds with the

liberation of molecular nitrogen and no other byproduct. The enolate produced in this

12
method has the E stereochemistry and it can be isomerized to the Z enolate by addition of

a catalytic amount of pyridine or lithium phenoxide.

O

 

 

 

l
stR R OBX2 2
JL ~ 5: _ R CH0 _ Aldol (9)
X=Cl ,Alkyl R
Aryl E
Pyridineor
Lithium Phenoxide
R1 03x2 2
H R CH0
— > Aldol
R H
2

Representative results from the method developed by Masamune are shown in Table 5.1

Table 5.1 Representative results of the reaction of alkylboranes with diazo ketones after

Masamune et al.

isomer. 3 3‘ 32 w my!!!
E Me Bu Ph 92 1/3
PhCH2 88 l/4
Ph 86 1/3
Z Me Bu Ph 90 >20/ 1
PhCH2 84 >20/1
Ph 85 >20/1

Hooze has also modified this method in order to synthesize enolates of methyl ketones

regioselectively using dicyclohexyl borane (eq10).(1 1d)

l3

 

0 OBChX 2
o...“ ,g .
R’ILCHNZ : R (112 + N2 (10)
5)-_C__arbonyl Enolization

 

This method is probably the most common and convenient method used to
prepare boron enolates. Several reagents have been used with varying results. The

general equation for this transformation is shown in equation 11a.
0 03122

RoBX /\
H " (1 la)

Base

V

Fenzel and Koster used triethylborane and a catalytic amount of diethyl boronpivalate to
effect the enolization of a variety of ketones. This reaction proceeds with the liberation

of ethane under vigorous conditions (eql ”(123)

0 013512
5:33, 100C
RJK/ O > R/K/ _R_Ci‘_9_. Aldol (11)

JL (Syn)

t-Bu 085:2
(cat)

A milder method was introduced by Mukaiyama and coworkers using a dialkyl boron

triflate and a hindered tertiary amine (eq 12). ( 13)
033112

0
BlhBOTf,-78C
RJK/ ‘ - R’K/ fl Aldol (12)

 

r

i-PrgNEt (syn)
2

Several dialkylborontriflates have since been studied by other investigators.(14) Evans
has shown that a variety of factors affect the stereoselectivity of this reaction.(3) The
proposed mechanism of the reaction by Evans involves initial coordination of the boron
triflate to the ketone carbonyl with subsequent deprotonation by the amine in a cyclic six
centered transition state in analogy to the Zimmerman model. Evans observed consistent

correlation between enolate geometry and aldol product stereochemistry for acyclic

14

ketones. Simple esters and amides cannot be enolized by these reagents, but thioesters

are enolized readily. Representitive results of their experiments are shown in Table 6.1.

Table 6.1 Representative results for the enolization of ketones and a thioester after Evans

et al.
.8

Et
Ph
iPr

tBuS

Zfli

>99/ 1

>99/ l

>99/ 1

5/ 95

 

% Aldol syn/anti
77 >97/3
82 >97/3
82 >97/3
80 10/90

Another class of boron reagents used for the enolization of carbonyl compounds

has been the substituted boron halides. Preparation of various reagents with moderate

success has been reported in the literature.(12) However the most useful and versatile

method has been recently developed by Brown and coworkers. Brown has shown that

dicyclohexylchloroborane reacts with aldehydes, ketones, and carboxylic acids to

produce a near quantitative yield of boron enolates under mild conditions (eq 13). (16)

O

 

Jk/ Rng
R 7

EI3N

X=OTf
OMs
1, Br, Cl

OBRg

\

R

Of

08R:

(13)

R\

This method allows one to prepare either E or Z boron enolates of ketones by varying

either the leaving group on the boron atom of the reagent or the reaction conditions.

15

Representitive results are shown in Table 7.1 along with the reagents developed for

enolization.

Table 7.1 Representative results for the enolization of ketones with B-X-9-BBN and
Chexz-B-X after Brown et al.

 

B-X-9-BBN Chexz-B-X

Ketone X Z/E %Yield Z/E %Yield

R

i-Pr OT f 88/ 12 96 25/75 96
OMs 82/ 18 94 23/77 93
1 27/97 97 32/68 98
Br 57/43 96 1 1/89 95
C] 46/54 95 3/97 97

F1 OT f 97/3 97 80/20 96
OMs 93/3 95 80/ 20 93
I 97/3 97 56/44 96
Br 97/3 97 30/70 96
Cl 97/3 95 21/79 97

t-Bu OT f 10/90 90 3/97 85
OMS 3/97 87 3/97 66
I 97/3 95 97/3 96
Br 3/97 94 10/90 82
Cl 3/97 94 3/97 60

16
This method has been extended to the preparation of the enolates of esters and tertiary

amides by using dicyclohexylboron iodide and a tertiary amine (eq l4).(173vb) Brown
has demonstrated that the stereochemical correlation for the reaction of ketone enolates as
shown by Evans does not hold for amide enolates and in fact it is the exact opposite. For
the reaction of esters they observed a dependence of the stereochemistry of enolization on
the R substituent.( 17a) For example, bulky R groups produced anti aldols but methyl and
ethyl produced syn aldols stereoselectively. One unusual feature of enolization of amides
is the solvent dependence of enolization. Alkane solvents produced aldols with syn
stereochemistry whereas chlorinated or aromatic solvents produced aldols with anti
stereoselectivity.(17b). Representative results for the enolization of ethylesters and

tertiaryethylamides are shown in Table 8.1.

 

LR CheXQBl 0%ha OBChexg (14)
Av \ or
X El3N x X \
I R
x=-0R',NR 2

17
Table 8.1 Representative results for the enolization of ethylesters and tertiaryethylamides

after Brown et al.

 

 

Temperature aldol%
X R Solvent enol aflol syn/anti Yield
N(CH?)5 Me CC14 0 0 3/97 95
Hexane 0 -78 97/3 93
NEt2 CCI4 0 0 5/95 96
Hexane 0 -78 97/3 96
OEt Me CCl4 0 0 97/3 96
0 -78 94/6 96
25 25 88/ 12 97
Et 0 0 95/5 95
i-Pr 0 0 3/97 94
i-Bu 0 0 3/97 84
t-Bu 0 0 3/97 96

Structure, physical properties, and prepaLation of boron amides

Boron amides (also known as aminoboranes) are compounds in which boron is
attached to one or more nitrogen ligands. These compounds have been a focus of
attention in the chemical literature for the past thirty years and several reviews have
appeared on their structures and their physical properties.(18) Boron amides have been
used as ligands for transition metal catalysts(19), reducing agentsao), and precursors to
boron-nitrogen cage compounds.(21) One recent example of the reactivity of an unusual
nature is the compound dimethylaminobis(trifluoromethyl)borane reported by Burger et
al(22). This compound has been shown to display a variety of reactions not observed

before with boron amides.(23) Scheme 2.1 shows a compilation of reactions carried out

to date with this complex.

l8

  

‘3‘”: u
éF CH3 CF3 \ B. .N+
1.4 ('3: CH3
CF15 ‘ B' .N+ oCH3 3
' ‘CH E}, R
CFs 3 s
\b (92” I
go
‘5‘ «a CF3 ~ B °N+ cCHg
' ‘CH
O 0‘3 3
O H R
CFa'h '-. l+’CH3 __._ y; y
B N V _ (CF3)23=N(CH3)2

 

('33 ‘CH3

 

— R1 R1

 

CF3 ‘ ?- 'Ni' CH3 CF3 s B- .N:' CH3 CF3 \ B- 'hi+ CCH3
am aw am

Scheme 2.1 Reactions of (CF3)2BN(CH3)2

The reaction of this reagent with carbonyl compounds is noteworthy. Although this
aminoborane has been shown to react with ketones, boron-carbon bond formation is
favored over boron-oxygen bond formation in an equilibrium exchange involving the
enolate (Scheme 3.1). The rearrangement of the enolate produced as an intermediate has

been suggested as a possible pathway to the formation of the boron-carbon bond.

2 1
R COCHzR

MegNBrR

(CF3hBNM92
’

 

H g(CH2C02MC)

 

 

MegN

W;W<:'Q>=

R=Me, iPr, Ph

F

 

 

Scheme 3.1 Proposed mechanism for the rearrangement of boron enolate after Burger et

The alkyl boron products in this reaction are generally isolated in 70% to 90% yield.
Burger has suggested that the rearrangement of the enolate to the alkyl boronate is too
fast to be observed by NMR in the case of amides, esters, and methyl ketones. The only
enolate reported in this study was a diethyl ketone enolate of unknown stereochemistry.
In a related reaction reported by Paetzold and Bierrnann alpha mercury methylesters react
with boron amides bearing a bromine substituent to produce boron enolates through
transmetallation.(24) This reaction has been shown to be an equilibrium, the direction of

which is dominated by the alkyl substituent on the boron atom (scheme 4.1).

M€2N R

B’

R=Mc, iPr

Scheme 4.1 Carbon to oxygen rearrangement for alkylaminoborane

20
The general features of boron amides are: 1)-they are mostly monomeric and dimerize

slowly at room temperature and faster if heated,; 2)-they are weak Lewis Acids; 3)—boron
and nitrogen both possess trigonal geometry when monomeric; 4)-the boron-nitrogen
bond is isoelectronic with the carbon-carbon double bond of alkenes and regarded as
such; 5)-these compounds are air and moisture sensitive and are handled by using inert
atmosphere techniques.

The monomeric nature of boron amides requires special attention. Since these
compounds harbor both donor and acceptor sites within one molecular entity,
conjugation, and oligomerization both play a role as stabilizing effects in these

compounds (Scheme 5.1).

Q/ «1: a B N
\

>9—9/ ..__: ><N><

Scheme 5.1 Resonance stabilization of boron amides

 

 

 

 

 

The association of boron amides is highly dependent on the nature of the substituent on
the boron. For example, amino difluoroboranes dimerize rapidly, but the analogous
chlorides and bromides dimerize slowly.(25) Variable temperature NMR investigations
(13C,1H) have demonstrated that conjugation may be a dominant stabilizing force. The
rotation barrier around the boron nitrogen bond is in the order of 17-24 kcal/mol for
monoaminoboranes.(26) Of course, one cannot entirely rule out steric effects due to the

small size of boron. Electron diffraction data also support sp2 hybridization around the

21
boron nitrogen bond with bond angles of approximately 120 degree for a variety of boron

amides.(27)
Preparation of boron aLmides

There are five routes available for the preparation of boron amides. However, the
most useful and direct method for simple amides of boron is the reaction of amines with

boron trihalides. These methods generally produce boron amides in excess of 80% yield.

l)-Displacement reactions of haloboranes

Boron halides(28)or their amine complexesag) react in the presence of excess
amines to cleave boron halogen bonds (eqs 1 and 2). Reaction of amines with boron
trichloride is highly exothermic; hence, these reactions are often carried out at low
temperatures initially.

ClzB-NHzPh + SNHgPh _, 13(th + 3Pth~13Cl (1)
0313 + 6NH3Mc __, B(NHMe)3 + 3McNH3Cl (2)

Displacement of chlorides and bromides is most often used since displacement of fluoride
is more difficult.(30) Steric effects have been shown to prevent complete displacement of

halide in the case of hindered amines (eq 3).(31)

i-Pr NH
C138 —;y ClB(Ni-Pr3)3 + Ni-PrgHCl (3)
Excess

2)-Transamin2_1tion

Boron amides can be transaminated in the presence of amines. Transamination
reactions of boron amides are particularly prone to steric problems (eq 4).(32) The
displacement order follows the sequence NH2> NHR> NR2, and NMe2> NBu2>
NHtBu> Nin.(33, 3 lb)

22

EthNMCg + HN: 7____, 1313319 7 + MC,NH (4)

This method is also useful for the preparation of hindered unsymmetrical aminoboranes

(eq 5).(34)
B(NMe3)3 PhNH2 , PhNHB(NMe3)3 + MegNH (5)

3)—DisplaIcement reactions of hvdrido -ang/l -a_ll_<oxy-hydroxy or alkylthio-boranes
The general reaction for these transformations is depicted in equation (6). The

driving force for these reactions is the formation of the strong boron-nitrogen bond.

\ \
B-X + RzNH ——> B-NRg + HX (6)
/ /

XzR, OR, OH. SR, H

When a borane and an amine are reacted together at elevated temperatures, hydrogen

evolution is the result and the boron amide is prepared in good yield (eq 7).(35)

3H3 + NHM62 ' (HQBNMegh + 2H3 (7)

 

In contrast to the aluminum alkyls, elimination of alkane from boron alkyls is difficult
and requires severe conditions. However in the presence of coordinating functional

groups (i.e. carbonyl grouping of amides) elimination occurs under milder conditions (eq

8).(36)

o 0
R33 +HN: 7 __, RZBN 7 + RH (3)

Reaction of alkyl borates or alky thioborates with amines is an equilibrium reaction.
Removal of the leaving group from the reaction by using volatile or cheating ligands(37)

results in successful preparation of boron amides (eq 9,10).(28,38,37a)

 

B(OR)3 + 3RNH3 y B(NHR)3 + 3ROH (9)

B(SR)3 + 3RNH3 w B(NHR)3 + 3RSH (10)

 

The alternative to this method is the use of an alkali metal amide instead of the amine (eq

1 l).(39)

B(OMe)3 + 3NaNHBu % B(NHBu)3 + 3NaOMe (11)

 

4)-Disproportionfiation and metathetical exchange

Some unsymmetrical halogeno, alkoxy—, alkylthio-, or alkyl-aminoboranes are
prone to disproportionation when heated. As a result symmetrical trisubstituted boron
amides have been made inadvertently. For example bis(dibutylamino)fluoroborane

disproportionates upon distillation (eq 12)0(40,38)

2FB(NBU2)2 ———> B(NBuv_3)3 + FgBNBu;) (13)

Metathetical exchange has been used to prepare boron amides from a mixture of an
aluminum tris amide and an alkyl borane or phenyldiethylborate (eq 13,14).(41)

3R33 + AI(NM€2)3 ———y ZRQBNMeg + RgAlMeg (13)
PhB(OEt)2 + AI(NM02)3 _.___, PhB(NMe3)3 + (EtObAlNMcz (14)

A reaction analogous to eq 13 can be carried out with tris diethylaminoborane and triethyl
borane to prepare ethyl substituted diethyl boron amide (eq 15).(42) This reaction
requires the use of borane THF complex as a catalyst. The boron amide is obtained in
high yield.

BH.THF
Eth + swag), _3__,. 513319512 (15)

24

5)-lnsertion reactiog

Boron amides have been prepared by the insertion of isocyanates into boron-
nitrogen(38d), boron-chlorine(43), boron-carbon,(4338d) and boron-oxygen bonds (eq
16,17,18,19). The advantage of insertion reactions is the lack of any by products since

both reaction partners combine to form one product.
PrQBNHPh + PhNCO _—y PrgBNPhCONHPh (16)

PhBClg + 2PhNCO ———> PhCONPhB(Cl)NPhCOCl (17)

0. 0

Q: ,BB“ + Ptho ———>©: \BNPhCOBu (18)
O 0’
0‘ 0‘
’BOBu + PhNCO ————> BNPhCOgBu (19)
O 0’

Insertion of carbodiimides into boron-chloride and boron-carbon bonds has been reported

(eq 20).(44)

PhBClg + 2RN:C:NR —’ RN:CPhNRB(CI)NRC|C:NR (30)

Other examples of insertion reactions for the preparation of boron amides are the reaction

of borane(45) or a trialkylborane(46) or trialkylboronate(47) complexes with imines (eq

21,22,23).
3%, + 2PhN:CHAr———> 2H3BNPhCH3Ar (21)
BR3 + MchCHPh—y RgBN(Mc)CHRPh (22
LilBRgBHfl + 6ArlNzCHAr y 2B[N(ArI)CHgAr]3 (23)

 

Other examples of insertion reactions include carbon-oxygen or carbon-sulfur insertions
into the boron-nitrogen bond of a tris boron amide to produce oxygen(48) or sulfur(49)

substituted boron amides (eqs 24, 25).

25
B(NM92)3 + 2cos ————y [MegNC(S)O]ZBNMe2 (24)

XCNB(NMe3)2 + p—TolNCX ——y XNCB(NMe3[N(Tol-p)C(NI‘ol-p)NMe3] (25)
X=OorS
Hydroboration followed by amine insertion into the boron hydrogen bond provides a

route for the formation of aminoalkylhaloboranes (eq 26). (50)

NH2 NH
4 _ + BHZCI ————> I'm (26)

Dialkyl chloroboranes or trialkylboranes may be employed to prepare secondary amines
by the reaction of alkylazides in a manner analogous to hydroboration. However, the

aminoborane has not been isolated though it is believed to be an intermediate in this

reaction (eq 27).(51)
1

R N3 1? 1 Cl BNRRl N
o ————' - ‘— ., + 2
RBCl_ Cl [.3 NR __, _ 2 (7)

Cl N24-
NaOH
HNRRl

Reaction of secondary chloroamines with trialkylboranes produces boron amides with

elimination of alkylhalide (eq 28).(52)

8R3 CINR R0 + RgBNRg (28)

We began our initial studies by the design of boron amides that may act as bases.
Compounds I, II, and Ill were proposed as our first model compounds.

,Ph ’SiMe3
(Chex) zB-N (Chex) gB—N\ (Chex) gB—N‘

‘— Ph SiMe3

I II III

The purpose of this study was to determine the effect of bulky substituent and electron

deficient substituents on the back bonding of nitrogen to boron. The rationale behind this

26

initial proposal was to maximize steric interactions of the ligands on both boron and
nitrogen centers. This was accomplished by placing sterically demanding substituents on
boron (cyclohexyl), and nitrogen (i-Pr, TMS, Ph). At the same time we wished to
minimize the boron-nitrogen back bonding that may interfere with the reagent acting as a
base. Hence conjugating substituents on nitrogen were chosen for compounds 11 and III.
The proposed reagents were synthesized via salt elimination from the haloborane and
lithium amide precursors according to equation 29.

LiNR2
(Chcx)3BCl ’ (Chex)2BNR3 + LiCl (29)

 

These reagents were then tested by the addition of one equivalent of pinacolone to an
NMR tube and the recording of the 1H-NMR spectrum at room temperature. After three
hours at room temperature the methyl resonance (alpha) of pinacolone persisted and no
vinylic resonances were observed in the spectra of II and 111. It was then apparent to us
that minimizing the boron-nitrogen conjugation present in boron amides without
destroying either the Lewis acidity of boron or the basicity of the nitrogen ligand presents
a more formidable problem. We assume that either significant boron-nitrogen back
bonding is still present in these compounds (II-III) or else the nitrogen lone-pair is
rendered so weakly basic by the conjugating substituents that proton transfer is
unfavorable.

We hypothesized that if pi bonding of boron with the second-row element,
phosphorus, is unfavorable, then the use of boron phosphides (RgB-PRg) rather than
boron amides might provide acceptable base strength.

Boron phosphides (also known as phosphinoboranes) are compounds in which
boron is attached to one or more phosphorus ligands. Boron phosphides are normally
associated molecules, with phosphorus bridges.(53) The presence of donor and acceptor
sites in these molecules may provide conjugation and association as stabilizing
mechanisms for their structure. In contrast to boron amides these compounds are rarely

monomeric in the absence of bulky substituents or donor ligands such as nitrogen, and

27

dimerise rapidly. For example MezBPHz is dimeric at room temperature whereas
MezBNHz is only dimeric at low temperatures and a monomer at ambient temperature.
At the least this implies that for boron phosphides B-P pi bonding (present in the
monomer) is less favorable than B-P sigma bonding present in the oligomer.
Furthermore, it has been noted that boron phosphide-amides are monomeric and pi
bonding occurs from the nitrogen atom rather than the phosphorus atom.(54) Finally the
nature of the B-P bond in the rare monomeric boron phosphides has been examined by
spectral methods.(55) Power has concluded that while there is no inherent weakness in
the B-P pi bonds, the presence of a large inversion barrier often prohibits the necessary
planar configuration at phosphorous.(56) Other workers in this field have determined
that MezBP(TMS)2 is in a fast monomer-dimer equilibrium above 17 degrees
Celsius.(57) Encouraged by these considerations, we synthesized a number of boron
phosphides and examined their reaction with simple ketones. Our first model compounds
were compounds IV, V, and VI. These compounds were proposed with the same
considerations in mind as previously mentioned for corresponding amides compounds I,

II, and III.

I ’SiMe3
(Chex) gB—P\ (ChCX) zB—P\

(ChCX) gB—P .
W Ph SlMc 3

IV V VI

Synthesis Of Boron Phosphides

Several monomeric boron phosphides have been prepared and characterized in the
past five years.(58) Alkyl boron phosphides have previously been prepared for chemical
vapour deposition applications(59). There are three main routes available to the synthesis
of boron phosphides. These routes are generally elimination reactions and provide good

to excellent yields of the boron phosphide.

28
l)Elimination of hydrogen halide

Phosphines react with boron halides in the presence of a stoichiometric amount of tertiary

amine bases to produce boron phosphides (eq l).(60)

E13N
PH3 + MQBBI‘ $ MCzBPHz + EI3NHBI‘ ( 1)

 

Dimer

2)Elimination of trimethylsilvlhalide
Alkylsilyl phosphines react with boron halides to eliminate trimethylsilylchloride and
produce the corresponding boron phosphide. This reaction is a slow reaction and requires
heating for completion (eq 2).(60) The attractive feature of this method is the generation
of volatile trimethylsilylchloride.

BC13

MqSiPEtg y EtzpBClz + MC3SiCI (2)
120C

3)§lim_ination of metal sglg

Alkylhaloboranes react with alkali metal phosphides with elimination of alkali metal
halide (eq 3).(60) This method is probably the most convenient route since alkali metal
phosphides can be prepared from either deprotonation of phosphines or metallation of

halophosphines.

PthCl .
Li P512 47 E£21313th + LICI (3)

29

Results and Discussion

Our studies began with the preparation of dicyclohexyl(di-t-
butylphosphino)borane (IV). This compound has been previously prepared via the
elimination of lithium chloride from lithium di-t-butylphosphide and

dicyclohexylchloroborane and characterized by Noth et al . (eq l).(61)

0C
CheszCl+ LiPt-Bu2 p CheszPt-Buz + LiCl (l)

Hexane

 

We found 31P-NMR spectroscopy to be an extremely useful and invaluable method for
the direct analysis of the course of the reactions throughout our studies. Since the
evolution of dialkylphosphine can be directly monitored, it can be used as a means of
estimating the extent of enolization. We chose the non-donor solvent toluene to eliminate
any complexation of the solvent to the reagent. Addition of 100 ymol of methyl ethyl
ketone (MEK) to an NMR tube containing 100 ymol of IV (6 36.11ppm), (Figure 2.1) in
hexane resulted in complete consumption of IV at room temperature and appearance of
two new resonances at 64.69 ppm. and 21.17 ppm. in 8:1 ratio, (Figure 3.1).

We attribute these resonances to the addition product resulting from the
nucleophilic addition of the phosphide to the carbonyl carbon and enolization of the

carbonyl substrate to release di-t-butylphosphine, respectively (eq 2).

O OBChex2 OBChex2

2 o
ChexZBPt-BuZJLL, RI/tth-Buf R1 + t-BuzPH (~)

30

 

Ema—d N.— u. - .
m. 225 385:: a». U_o<o_o=ax<_€m-?c=¢._gem—5337083

31

 

1
Fl» lb l l Lll> l lit—ll I? b

umo who pmo p00 00 mo no mo 0 two ODS

 

ENE.» uh u:u.2?=~ €8ch 6.. 936—orox<_A&-~-c=Q_cram—6:53.538 3.8..
23an 0.. 7.5% 2:. «8.53m w. Nun 3.. H ~55.

32

 

L. E.

 

 

b 1" P b Fl ll
fi.11-11111411§1j1444144«4444_JJ«144‘4‘121411441.ra4d4-13.4141Jj.a. _ . . a .r- . .,d.-.
.30 HNO :3 .0 ma .0 no 0 3'

35:6 .9— w_1.2_<=~ €8.28 3. gown—cruxwxarrcegeromermaovgaaa ~52
33:0: 0.. E30963. .883 m2. «:5qu a. N8 m2 NA 72.3

33

Table 9.1 summarizes the results obtained with this reagent with a variety of carbonyl
compounds. The control experiment was the reaction of lithium di-t-butylphosphide with
pinacolone and MEK which produced di-t-butylphosphine as the sole product.

Table 9.1. Reaction of Dicyclohexyl(di-t-butylphosphino)borane with carbonyl substrates

 

31 R2 % addition % enolization Temperature Time
Me Et 89 11 25 1 hr
Me Et 78 20 -78 1 hr
‘Bu Me 3 19 25 1 hr
iPr iPr <1 >98 25 24 hr
Me OH 6 5 25 17 hr
Me NMeQ 25 52 25 4 hr

The results in Table 9.1 demonstrate the effect of bulky substituents on the ability
of the phosphide to partition between the deprotonation pathway and the addition
pathway. When the addition of the phosphide is suppressed, deprotonation is the prefered
pathway as demonstrated in the cases of pinacolone and diisopropyl ketone, (Figure 4.1).
Esters and amides appear to be less reactive due to their lower acidities.

We wished to verify the formation of enolate in these reactions by synthesizing
the enolate via a different route. For that purpose the lithium enolate of diisopropyl
ketone was prepared and transmetalated with dicyclohexylchloroborane to generate the

corresponding boron enolate (eq 3). BCh
I CK:
O O

1) LDA
>/u\< 4» W + LiCl (2)
2) ChngCl

 

34
Comparison of the 13C-NMR spectra obtained on the products generated via the two

different methods showed identical chemical shifts. This is the first observation of
enolization of a ketone with a boron phosphide.

Encouraged by the above observations, and in an effort to reduce the
nucleophilicity of the phosphorus ligand, the dicyclohexyl(diphenylphosphino)borane V
was synthesized. Addition of dicyclohexylchloroborane to a suspension of lithium
diphenylphosphide resulted in an insoluble and chalk like precipitate. 31P and “B NMR
analysis of the supernatant did not show any phosphorus or boron resonances.

We attribute this behaviour to the formation of the insoluble phosphinoborane
dimer in toluene. For this reason THF was chosen to carryout the synthesis of V, and
slow down the dimerization by serving as a Lewis base. A series of low temperature
NMR experiments were carried out in THF to verify the presence of one or more reactive
species in THF solutions of the boron phosphide. Addition of a THF solution of lithium
diphenyl phosphide (red solution) to dicyclohexylchloroborane resulted in the
disappearence of the red color upon each drop at -78 C. As the reaction mixture was
allowed to warm up a pale yellow solution containing a fine white precipitate (LiCl)
formed. Analysis of this solution by 31P NMR spectroscopy revealed the presence of
two resonances. Addition of pinacolone to the NMR tube resulted in the disappearence
of both resonances and appearence of diphenylphosphine over a period of one hour at
ambient temperature. Based on these observations a solution of the phosphinoborane was
then reacted with pinacolone, 3-pentanone, ethylacetate, and N,N-dimethylpropionamide

in four separate reactions. Results of these reactions are provided in Table 10.1.

35
Table. 10.1 Reaction of CheszPth with Carbonyl substrates

 

0
R11 R.
R} R2: Time Temperature % Phosphine
tBu Me 2hr 25 80
Et Et 2hr 25 92
E1 NMez 2hrs 25 46
4hrs 25 46
Me OEt 4hrs 25 48

We determined the stereoselectivity of this acid/base reaction with 3-pentanone.
The stereochemistry of enolate formation for 3—pentanone was investigated in an indirect
manner since 1H-NMR of the E and the Z enolates of boron bearing cyclohexyl
substituents are indistinguishable as previously reported by Brown et al.(2) Hence the
enolate of 3-pentanone was reacted with benzaldehyde at -78C as prescribed by Brown.

We determined the configuration of the enolate to be of Z configuration since the
aldol product isolate has the syn/anti ratio of 9: 1.

These results show that conjugating substituents on phosphorus reduce the
nucleophilicity of the phosphorus ligand, and due to the reduced basicity of the
phosphorus thermodynamic enolate formation is favored because of a slow reaction.

In another set of experiments the effects of donor ligand(s) on the efficiency of
enolate formation were examined. These experiments were carried out by substitution of
the cyclohexyl ligand for oxygen, nitrogen, electron dificient nitrogen, electron deficient
oxygens, and combinations of both characteristics. The diphenyl phosphide ligand was
chosen based on our previous observations that it is less nucleophilic than the di t-Butyl

phosphide ligand. Results of these studies are provided in Table 11.1.

36
Table 11.1 Survey of the Reaction of heteroatom substituted boron diphenylphosphides
(RIRZBPPh2)with pinacolone

 

31 R2 % addition % enolization Temperature Time Solvent Compound

N‘Pr N‘Pr o 0 25 24hrs THF VII

Catechol >90 <10 25 15min THF VIII

NEE 25 75 25 30min Ether IX
34 66 25 30min THF

NTE 0.3 25 30min THF X

C4H8N Cl 0 l 25 30min Hexane XI

These results show that steric effects as well as electronic effects play a major role
in the activity of these reagents. The presence of two pi donors in the absence of any
steric effects increase the reactivity of the reagent towards a carbonyl substrate as evident
from the NEE phosphide (IX). In contrast the presence of one pi donor ligand and one
electon deficient ligand (X1) or two electron deficient pi donors (catechol ligand VIII)
diminish the activity of the boron complex towards deprotonation as evident from last
two enteries in the table (X, XII). The lack of reactivity in the latter case (XI) may be
attributed to significant back bonding contribution from phosphorus resulting in the
unavailability of the lone pair on the phosphorus. Another explanation in the latter case
may be the presence of strong boronphosphide aggregates (dimers) that are unreactive.
We attribute the lack of reactivity of the phosphide having two diisopropylamine ligands

(VII) to steric effects contributed from the diisopropylamido ligands.

37

Conclusions

A new procedure has been developed for the preparation of enolates of boron via
an acid/base reaction previously unknown for phosphinoboranes. Our results provide the
first example of a boron base capable of forming ketone enolates. We have demonstrated
that the reactivity of these complexes is very sensitive to both the nature of ligands on
boron as well as the nature of ligands on phosphorus. Bulky alkyl substituents on
phosphorus increase the nucleophilicity of the reagent. In contrast, phenyl substituents
decrease the nucleophilicity of the phosphorus center to the extent that an acid/base
reaction can occur efficiently when non-pi donor ligands are present on the boron center..
We have determined that the enolates prepared from the reaction of phosphinoboranes
with ketones are a thermodynamically favod product due to the slowness of the reaction
of these bases. Although the basicity of phosphide ligands used in this study is not
currently sufficient for the formation of a kinetic product, development of new phosphide

ligands in the future studies may overcome this limitation.

38

Experimental

All operations were carried out under an argon or nitrogen atmosphere.
Diphenylamine, diisopropylamine, 1,1,1,3,3,3-hexamethyldisilazane (HMDS), hexane,
chlorotrimethylsilane, N-ethylethanolamine, ethanolamine, and toluene were obtained
from Aldrich Chemical Company and distilled from calcium hydride prior to use.
Deuterated benzene, toluene, and chloroform were dried over calcium hydride and purged
with nitrogen. Ketones were obtained from Aldrich Chemical Company and dried over
calcium hydride and purified by distillation. n-Butyllithium was obtained from Aldrich
Chemical Company as a 1.6 M solution in hexane and was used directly.
Tetrahydrofuran and diethyl ether were distilled from sodium/benzophenone prior to use.
Lithium metal was obtained as rods from Aldrich Chemical Company and handled in a
dry box. D20 inserts were prepared by sealing the open end of a 12 cm thick walled
glass tube with 1mm internal diameter. To this tube was added approximately 100 pL of
D20 via a syringe and the glass tube was sealed over a bunsen burner. This tube served
as the source for the deuterium lock signal for the instrument. 1H-, 31P—, 13C-, 11B-
NMR data were recorded on a Varian (VXR) spectrometer operating at 300, 121.4, 75.4,
96.2 MHz respectively. Phosphorous and boron NMR data are reported with respect to
85% phosphoric acid and borontrifluoride etherate external standards. NMR tubes and
D20 inserts were dried at 150C in an electric oven, fitted with a septum, and purged with

nitrogen using a long needle for at least one hour prior to use.

Generi Method for Examingtion of Boron Phosphides

39
To an NMR tube fitted with a septum and a D20 insert was added 1 ml of a boron

phosphide solution. Ketone was added to the tube via a gas tight Hamilton syringe and

the spectrum recorded.

Preparation of Phosphines

Di-t-Butylphosphine (“P NMR 6 20.7 s) was prepared according to the
procedure by Timmer et al. by reduction of di-t-butylchlorophosphine with LAH(62).
Diphenylphosphine (31F NMR 6 —39.9 s) was prepared by lithiation of
triphenylphosphine followed by aqueous workup under nitrogen as described by
Wittenberg and Gilman(63).
figuration of Lithium Phosphides

Lithium di-t-Butylphosphide was prepared by deprotonation of the phosphine
precursor with n-Butyllithium. Lithium diphenylphosphide was also prepared by
deprotonation of the corresponding phosphine.
Prgvgaration of Chloroboranes

Bis-(diiopropylamino)chloroborane was prepared from borontrichloride and
excess diisopropylamine according to the procedure by Davies et al.(63)
Dicyclohexylchloroborane was prepared from dicyclohexyl borane and phosphorus
pentachloride according to the procedure by Brown et al.(64) Catecholchloroborane was
prepared by disproportionation of boron trichloride and 2,2'-o—Phenylenedioxybis(1,3,2-
benzodioxaborole) according to the procedure by Mannig and Noth et al.(65)

Pyrolidinodichloroborane was prepared according to the procedure by Musgrave.(66)

General Method for The Survey and Preparation of Phosphinoboranes
Phosphinoboranes were prepared by addition of a stoichiometric amount of the

desired chloroborane to a stoichiometric amount of freshly prepared lithium

diphenylphosphide (31F NMR THF 5 —24.6 s insoluble in hexane) or lithium di-t-

40
butylphosphide (31F NMR THF 6 36.71 s insoluble in hexane) at -78C. A precipitate

(LiCl) was observed in all cases upon warming to room temperature. The suspension was
allowed to settle, and cannula transfered into a 10 ml volumetric flask. The precipitate
was washed with two small portions (lml) of solvent and transfered to the volumetric
flask. The survey of the reagents prepared in this manner (typically 0.5 M) was
conducted based on the appearrance of phosphine upon addition of a stoichiometric
amount of pinacolone to a 0.7 mL (0.35 mmol) aliquot NMR sample. 31P—NMR of all
boron phosphides prepared in this manner showed the absence of the corresponding
lithium phosphide resonances and appearance of new phosphorous resonanance(s).
Percent yield of reagents are based on intergration of total phosphorous resonances.
Ergparation of N-Tosvl Ethanolamine (NTE)

To a 1L Erlenmeyer flask was added 200 mL of dichloromethane and 70 mL of
ethanolamine (1.2 mole). This solution was cooled using an ice bath. 91.1 grams of
tosylchloride (0.47 mole) was dissolved in 300 ml of dichloromethane and transfered to a
500 ml addition funnel. This solution was then added to the Erlenmeyer flask over an
hour and allowed to reach room temperature after the completion of addition. Aquous
workup with saturated bicarbonate solution and evaporation of dichloromethane resulted
in 80 grams (74% yield) of a clear thick liquid that crystallized upon standing. m.p.54—55
C
1H NMR(300MI-Iz;CDCl3) : o 2.42(s, 3H), 2.83(br s, 1H), 3.06(m, 2H), 3.67(br s, 2H),
5.57(s, 1H), 730(d, 2H), 7,75(d, 2H).
13C NMR(75.4MI-Iz; CDCl3) : 21.46, 45.16, 61.20, 127.05, 129.73, 136.25, 143.54.
Preparation of N-silvl.O-silvl. N-EthylethamLamine (NONEE)

Lithium metal 13.32 g (1.92 mole 20% excess) was shaved from lithium rods
inside a dry box and transfered to a three neck 1L round bottom flask. The flask was kept
under an atmosphere of argon and equipped with an addition funnel, a thermometer, and a

water cooled condenser. The flask was charged with 250 ml of dry ether and cooled to

41
-10C (internal temperature) with stirring using a C02/acetone bath. The addition funnel

was charged with 150 mL of ether, followed by 103 mL (0.96 mole) of n-butylbromide.
The butylbromide solution was added to the lithium suspension over a 30 minute period,
after which it was allowed to warm up to 10C and kept at this temperature for an hour.
The vessel was cooled down to OC and 39.0 mL (0.4 mole) of N-ethylethanolamine was
added dropwise through the addition funnel over 20 minutes while keeping the internal
temperature at 0C. After one hour 101.5 mL (0.8 mole) of TMSCl was added over an
hour while maintaining the reaction temperature at 0C. The reaction was allowed to
come to room temperature over an additional hour and filtered to give a clear solution.
Ether was distilled and 80 g of a liquid product (86%) was isolated by vaccum distillation
(bp. 62C, 0.] mm Hg).
1H NMR(300MHz;CDCl3) : 6 -0.01(s, 9H), 0.08(s, 9H), 0.95(t, J=6.6Hz, 3H), 2.75(q,
J=7.1Hz, 2H), 2.82(t, .I=7.0, 2H), 3.45(t, 1:7.0, 3H)
l3C NMR(75.4MI-Iz; CDCl3) : -0.51, 0.00, 16.10, 41.69, 48.77, and 62.30.
Preparation of N-Ethylethanolaminechloroborane

To a 250 mL round bottom flask was added 50 mL of a 1M solution (50 mmol) of
borontrichloride in hexane, and diluted with an additional 50 mL of hexane. The vessel
was cooled using an ice water bath. To this solution was added 11.65 grams of NONEE
(50 mmol) in a dropwise fashion, over a 10 minute period. Upon dropwise addition, a
white precipitate appeared. Removal of solvent and TMSCl by a water aspirator
produced 6.7 grams of a pale red solid. The solid was washed with two-10 mL portions
of hexane and residual hexane was removed under reduced pressure to produce 6.57 g of
a sticky solid (96% yield).

] 1B-NMR analysis in CCl4 showed presence of two singlets at 27.7 ppm and 8.70
ppm in 1:1 ratio. Heating the sample to 80C changes the ratio to 6:1. When the sample
was cooled to room temperature, a 1:1 ratio was re-established. This data suggests the

presence of a monomer and dimer in solution.

42
13C man/nu; CCI4, 80C) : 16.76, 41.69, 50.25, 67.63

Preparation of N-Tosvl-Ethanolaminechloroborane

To a 10 mL round bottom flask was added 1.07 g (5.0 mmol) of N-Tosyl—
Ethanolamine. 5 ml of dichloromethane was added and the solid dissolved. To this
solution was added 5 ml of a 1M borontrichloride solution (5 mmol) in dichloromethane.
After half an hour standing at room temperature the solvent was removed under a stream
of nitrogen leaving behind a white solid.( 1.27 g 97%)
1H NMR(300MHz;CDCl3) : 6 2.45(s, 3H), 3.80(ddd, .I=7.5Hz, 2H), 0.95(ddd, J=6.6Hz,
2H), 7.82(d, J=8.4Hz, 2H), 7.35(d, 1:8.7, 2H).
l3C NMR(75.4MHz; CDCl3) : 21.55, 47.74, 65.23, 127.15, 129.86, 136.01, 144.54.
1113 NMR(96.2MHz; CDCl3,) : 25.66
Preparation 0f N-Tosylethanolamine(diphenvlphosphino)boLa_rlta_X_

In a 25 mL round bottom flask 5 mmol of N-Tosylethanolaminechloroborane was
prepared. The solid was dissolved in 10 mL of THF at -78C. To this solution was added
5 mmols of a diphenylphosphine solution and the reaction was allowed to warm up to
room temperature. A pale yellow solution resulted. A 0.5 mL sample was transfered to a
NMR tube. 31P-NMR analysis showed presence of a major resonance at -61.2 ppm
(69%). Addition of pinacolone to the tube resulted in the release of 0.3% phosphine over
30 minutes.

Preparation of N—Ethylethanolamine(diphenylphosphino)borafl

To a 10 mL round bottom flask was added 0.51 mL of diphenyphosphine (3.0
mmol) followed by 2 mL of THF. 2 mL of a 1.5 M Butyllithium solution in hexane (3.0
mmol)was added at 0C in a dropwise manner. A deep orange solution resulted. To this
solution was added 4.3 mL of a 0.7 M solution of N-ethylethanolaminechloroborane (3.0
mmol). After half an hour a pale yellow solution resulted with appearance of a
precipitate. The solid was allowed to settle and the supernatant was transfered to a 10 ml

volumetric flask. The solid was washed with two portions (1 mL) of THF and the

43
resulting solution transfered to the volumetric flask. A one mL aliquot (0.3 mmol) of this

solution was transfered to a NMR tube. 31P NMR analysis of the sample showed a broad
singlet at -65 ppm that accounted for 73% with other resonances at -40 ppm (11%
diphenylphosphine), -19.4 ppm (13%), and -l4.4 ppm (3%). Addition of 35 pL of
pinacolone (0.3 mol) at room temperature resulted in disappearance of the resonance at
-65 ppm and increase in the amount of diphenylphosphine in addition to appearance of a
new resonance at 7.2 ppm (34%).
Examination of Catechol(diphenvlphosphino)borane VIII

A 5ml round bottom flask was charged with 0.8 ml of n-Butyllithium (1.25
mmol). 217 #1. of diphenylphosphine (1.25 mmol) was added via a syringe at zero
degrees. After standing for 10 minutes a white slury resulted. To this slurry was added 2
ml of a 0.6 M solution of catecholchloroborane (1.2 mmol) in benzene. The solution was
then allowed to settle and an aliquot was removed for NMR analysis. 31P NMR of the
sample showed a major broad singlet at -65.4 ppm which dissapeared upon addition of
pinacolone without the release of diphenylphosphine. A new singlet appeared at 7.1 ppm

which may account for addition of phosphorous to the ketone.

Examination of Pvrolidinechloro(Diphenylphosphino)bora_n£&

In a 5 ml round bottom flask 1.6 mmol of lithium diphenyl phosphide was
prepared by the addition of 1 ml of a 1.6 M n-Butyllithium to 278 14L of
diphenylphosphine in 2 ml of hexane at zero degrees. To this suspension was added 213
yL of pyrolidinedichloroborane (1.6 mmol). After 30 minutes a 0.5 ml sample was
removed for NMR analysis. 31P NMR showed presence of two new resonances at -35.5
ppm and -46.1 ppm. Upon addition of pinacolone no significant changes were observed.

Examination of bis-diisopropylaminegdiphenylphosphino)borane VII

1.05 mole of bis-diisopropylaminechloroborane (0.256 g) was transfered to a 5

 

ml round bottom flask and dissolved in 1 ml of THF. To this solution was added 1 ml of

44
a 1M lithium diphenylphosphide solution in THF in a dropwise manner at -78. The

cooling bath was removed and the reaction mixture was allowed to stirr at room
temperature for one hour. A pale yellow suspension was the result. An aliquot was
removed for NMR analysis. 31P NMR of the sample showed a single resonance at -36.5
ppm. Addition of pinacolone to this sample did not show any significant change.
Reexamination of the sample after 24 hours produced the same spectrum.
Examination of dicyclohevaoron(hexamethvldisiljazinmborane III

A 25 mL round bottom flask was charged with 8 mL of hexane and 2mL of a
1.6M n-butyllithium (3.2 mmol). The flask was then cooled to OC and 0.65 mL of
HMDS (3.2 mmol) was added dropwise. After 30 minutes a viscous liquid resulted. To
this suspension was added 680 pL (3.2 mmol) of dicyclohexylchloroborane in a dropwise
manner. During the addition of the boron chloride a salt precipitated and persisted. Upon
completion of addition the ice bath was removed. The mixture was stirred at room
temperature fo an hour and allowed to settle. 1H-NMR examination of a sample of the
supernatant after evaporation of hexane showed broad resonances between 0.8 and 1.8
ppm(ring resonances) and a singlet at 0.17 ppm (TMS resonances). Addition of
pinacolone to this sample did not result in any observable reaction (no change in the ratio
of TMS resonances and the pinacolone methyl resonance at 2.1 ppm).
Examination of dicyclohexy1(diphenylamino)borane II

A 25 mL round bottom flask was charged with 0.54 g of diphenylamine and 8 mL
of benzene. The flask was then cooled to OC and 2mL of a 1.6M n-butyllithium (3.2
mmol) was added dropwise. After 30 minutes a slurry resulted. To this suspension was
added 680 FL (3.2 mmol) of dicyclohexylchloroborane in a dropwise manner. During the
addition of the boron chloride a salt precipitate persisted. Upon completion of addition
the ice bath was removed. The mixture was stirred at room temperature for an hour and
allowed to settle. 1H-NMR examination of a sample of the supernatant after evaporation

of benzene showed broad resonances between 0.8 and 1.8 ppm(ring resonances) and a set

45
of multiplets above 7.0 ppm. Addition of pinacolone to this sample did not result in any

observable reaction (no change in the ratio of aromatic resonances and the pinacolone
methyl resonance at 2.1 ppm).
Madon of dicyclohexylboron enolate of diisfigpropvlfiketone

To a 25 mL round bottom flask was added 5 mL of hexane at 0C followed by 3.2
mL of a 1.6M n-Butyllithium (5mmol). Diisopropylamine (700 pL) was added dropwise
and the solution became more viscous. The vessel was allowed to come to room
temperature to complete the reaction. After 15 minutes the reaction vessel was cooled to
0 C and 708 ML of diisopropyl ketone (5.0 mmol) was added. Upon standing at room
temperature for half an hour a white precipitate appeared. The solvent was evaporated at
reduced pressure where a white solid remained. The solid was resuspended in 5 ml of
benzene, cooled to OC, and lmL of dicyclohexylchloroborane (5 mmol) was added
dropwise with stirring. The reaction was stirred at room temperature for an hour.
Benzene was evaporated at reduced pressure and replaced with CDC13.
13C NMR(75.4MHz; CDC13) : 18.06, 18.23, 20.14, 27.06, 27.41, 27.84, 28.09, 28.68,
29.49, 107.89, 149.38.
1113 NMR(96.2MHz; CDCl3,) : 49.32
Mon of Dicvclohexvl(di-t-butvlphosphino) borane With Diisopropvlketone

To a 10 mL round bottom flask was added 3.12 mL of a 1.6 M n-butyllithium
solution in hexane (5.0 mmol) followed by 0.9 mL of di-t-butyhlphosphine (5.0 mmol).
The mixture was then heated at 50C and after 30 minutes a slurry resulted. The flask was
then cooled with an ice bath. To this slurry was added 1.1 mL of
dicyclohexylchloroborane (5.0 mmol) at DC in a dropwise manner. After one hour a
sample was removed for NMR analysis. 31P--NMR showed a major resonance at 36.11
ppm which accounted for a 91% yield.

A 1 mL sample was removed from the flask and hexane was evaporated under

reduced pressure, 0.181 g of a clear pale yellow oil remained (0.56 mmol). Based on this

46
determination, 80 yL of diisopropyl ketone (0.56 mmol) was added to an NMR tube after

transfer of the sample in hexane. 31P—NMR analysis revealed slow release of di-t-
butylphosphine. After completion of the reaction (24 hours), the sample was evaporated
to remove excess phosphine and 13C NMR recorded.
l3C NMR(7S.4MHz; CDCB) : 18.07, 18.23, 20.13, 27.03, 27.38, 27.81, 27.94, 28.64,
29.48, 107.89, 149.36.
1113 NMR(96.2MI-Iz; CDC13,) : 49.32
Examination of the reaction of dicvclohexvl(di-t-butvlphosphino)bora£
N,N-dimethylpropionamide

A 0.5 M stock solution of dicyclohexyl(di-t-butylphosphino)borane was prepared
in d8 toluene in a 10 mL volumetric flask according to the general procedure. To an
NMR tube was added 700 yL of the stock solution (0.35 mmol). To this tube was added
38 pL of N,N-dimethylpropionamide via a syringe. Examination of the sample after 4
hours showed complete consumption of the starting material (31F 32.4 ppm) and
appearance of two major resonances at 18.1 ppm (52% phosphine) and 29.36ppm(25%).
Ethyheetate

A 0.5 M stock solution of dicyclohexyl(di-t-butylphosphino)borane was prepared
in d8 toluene in a 10 mL volumetric flask according to the general procedure. To an
NMR tube was added 700 pL of the stock solution (0.35 mmol). To this tube was added
28 pL of ethylacetate via a syringe. Examination of the sample after 17 hours showed
lack of consumption of the starting material (31P 32.4 ppm) and appearance of two minor
resonances at 18.1 ppm (5% phosphine) and 56.13 ppm(6%).
Pinacolone

A 0.5 M stock solution of dicyclohexyl(di-t-butylphosphino)borane was prepared
in hexane in a 10 mL volumetric flask according to the general procedure. To an NMR
tube was added 700 ”L of the stock solution (0.35 mmol). To this tube was added 43 pL

of pinacolone via a syringe. Examination of the sample after one hour showed slow

47
consumption of the starting material and appearance of two resonances at 21.1 ppm (19%

phosphine) and 61 .8ppm (3%).
MEK
A 0.5 M stock solution of dicyclohexyl(di-t-butylphosphino)borane was prepared in
hexane in a 10 mL volumetric flask according to the general procedure. To an NMR tube
was added 700 pL of the stock solution (0.35 mmol). To this tube was added 35 14L of
MEK via a syringe. Examination of the sample after 30 minutes showed complete
consumption of the starting material (31F 36.1ppm) and appearance of one minor
resonance at 21.2 ppm (2% phosphine) and a major resonance at 64.64 ppm(98%).
Examination of reaction of dicLLclohexv(diphenvlphosphino)borm
Diethylketone

A solution of dicyclohexylchloroborane in THF was prepared by addition of 1.25
mmoL (270 pL) of dicyclohexylchloroborane to 5 mL of THF at -78C in a 25 mL round
bottom flask. In a 10 mL round bottom flask 1.25 mmol of lithiumdiphenylphosphide
was prepared by the addition of 1.25 mmol (217 yL) of diphenylphosphine to 0.8 mL of
n-butyllithium in THF (5 mL) at -78C. A bright orange solution resulted and the flask
was allowed to warm up to room temperature. The phosphide solution was added to the
chloride solution via a syringe in three portions in a dropwise manner. Complete
discoloration of the lithium phosphide solution occured upon addition of each drop. The
reaction vessel was then allowed to warm up to DC where a white precipitate appeared
To this suspension was added 125 pl. of diethylketone (1.25 mmol) in a dropwise
manner. After 1.5 hours the flask was cooled to -78C and 127 yL of benzaldehyde was
added and the mixture was stirred for 2 hours. The reaction was worked up using 680 ”L
of 30% hydrogen peroxide after 3 hours. A 1 mL sample was removed and THF was
evaporated. This sample was extracted with hexane, washed with dilute l-lCl then water ,

and dried over anhydrous N32804. The product was analysed by 1H NMR to determine

48
the syn/anti ratio of 9:1 and the percent yield (92%).The percent yield is based on the

amount of aldehyde recovered.

N,N-Dimethylpmpionamide

A solution of dicyclohexylchloroborane in THF was prepared by addition of 1.25 mmoL
(270 yL) of dicyclohexylchloroborane to 5 mL of THF at -78C in a 25 mL round bottom
flask. In a 10 mL round bottom flask 1.25 mmol of lithiumdiphenylphosphide was
prepared by the addition of 1.25 mmol (217 pl.) of diphenylphosphine to 0.8 mL of n-
butyllithium in THF (5 mL) at -78C. A bright orange solution resulted and the flask was
allowed to warm up to room temperature. The phosphide solution was added to the
chloride solution via a syringe in three portions in a dropwise manner. Complete
discoloration of the lithium phosphide solution occured upon addition of each drop. The
reaction vessel was then allowed to warmed up to OC where a white precipitate appeared
To this suspension was added 137 pL of N,N-dimethylpropionamide (1.25 mmol) in a
dropwise manner. 1H NMR analysis after 2 hours showed presence of starting amide
(two methyl resonances at 2.92 and 3.0 ppm )and a singlet at 2.60 belonging to the
enolate (vinylic protons appeared at 4.85 ppm as 2 sets of overlaping quartets) which
accounted for 46% enolate formation. Re-examination of the reaction at 4 hours did not
show any changes.

Ethyhoetate

A solution of dicyclohexylchloroborane in THF was prepared by addition of 1.25 mmoL
(270 pl.) of dicyclohexylchloroborane to 5 mL of THF at -78C in a 25 mL round bottom
flask. In a 10 mL round bottom flask 1.25 mmol of lithiumdiphenylphosphide was
prepared by the addition of 1.25 mmol (217 pL) of diphenylphosphine to 0.8 mL of n-
butyllithium in THF (5 mL) at -78C. A bright orange solution resulted and the flask was
allowed to warm up to room temperature. The phosphide solution was added to the
chloride solution via a syringe in three portions in a dropwise manner. Complete

discoloration of the lithium phosphide solution occured upon addition of each dr0p. The

49

reaction vessel was then allowed to warmed up to OC where a white precipitate appeared
To this suspension was added 122 pl. of ethylacetate (1.25 mmol) in a dropwise manner.
31F analysis after 4 hours showed 48% release of diphenylphosphine and another major
resonance at 1.24 ppm (30%), the remainder was 9% starting material and two other
resonances at 17.6 and 14.2 ppm.

Pinacolone

A solution of dicyclohexylchloroborane in THF was prepared by addition of 1.25 mmoL
(270uL) of dicyclohexylchloroborane to 5 mL of THF at -78C in a 25 mL round bottom
flask. In a 10 ml. round bottom flask 1.25 mmol of lithiumdiphenylphosphide was
prepared by the addition of 1.25 mmol (217 pL) of diphenylphosphine to 0.8 mL of n—
butyllithium in THF (5 m1.) at —78C. A bright orange solution resulted and the flask was
allowed to warm up to room temperature. The phosphide solution was added to the
chloride solution via a syringe in three portions in a dropwise manner. Complete
discoloration of the lithium phosphide solution occured upon addition of each drop. The
reaction vessel was then allowed to warmed up to OC where a white precipitate appeared
To this suspension was added 156 pL of pinacolone (1.25 mmol) in a dropwise manner.
31F analysis after 2 hours showed 80% release of diphenylphosphine and other minor
resonances, the remainder was 14% starting material and two other resonances at 19.6

and 14.1 ppm.

Literature Cited

1) Mekelberger, H.B.; Wilcox, C.S.;Comprehensive Organic Synthesis, 2, Chapter1.4.
Pergamon Press: New York, 1991.

2) Goodman, J.M.; Paterson, I.; Khan, S.D.Tetrahedron Lett. 1987 28, 5212. Goodman,
J.M.; Khan, S.D.; Paterson, IJ. Org. Chem. 1990, 55, 3295.

3)_Evans , D.A.; Nelson, J.V.; Vogel, E; Taber, T.R. J. Am. Chem. Soc. 1981, 103, 3099.
Evans , D.A.; Nelson, J.V.; Taber, T.R. Top. Stereochem. 1982, 13, 1 .Evans , D.A.;
Vogel, E.; Taber, T.R. J. Am. Chem. Soc. 1979, 101, 6120.

4) (a) Suzuki, A.; Arase, A.; Matsumoto, H.; Itoh, M.; Brown, H.C.; Rogic, M.M.;
Rathke, M.W. J. Am. Chem. Soc. 1967, 89, 5708.(b) Brown, H.C.; Rogic, M.M.; Rathke,
M.W.; Kabalka, G.W. lbid. 1967, 89, 5709. (c) ibid. 1968, 90, 4165. (d) Kabalka,G.W.;
Brown, H.C.; Suzuki, A; Honma, S.; Arase, A.; Itoh, M. J. Am. Chem. Soc. 1970, 92,
710.

5) Fenzl, W.; Koster, R.; Zimmermann, I-IJ. Liebigs Ann. Chem. 1975, 2201.

6) Mukaiyama, T.; Inomata, K.; Muraki, M.J. Am. Chem. Soc. 1973, 95, 967. Inomata,
K.; Muraki, M; Mukaiyama, T. Bull. Chem. Soc. Jpn. 1973, 46 1807.

7) (a) Boldrini, G.P.; Mancini, F.; Tagliavini, E.; Trombini, C.; Umani-Ronchi, A. J.
Chem. Soc. Chem. Commun. 1990, 1680. (b) Boldrini, G.P.; Bartolotti, M.; Tagliavini, E.;
Trombini, C.; Umani-Ronchi, A. Tetrahedron Lett. 1991, 32, 1229. (c) Boldrini, G.P.;
Bartolotti, M.; Mancini, E; Tagliavini, E.; Trombini, C.; Umani-Ronchi, A. J. Org. Chem.
1991, 56, 5820. (d) Matsumoto,Y; I-layashi,T. Syn. Lett. 1991, 5, 349.

8) Hirama, M.; Masamune, S. Tetrahedron Lett. 1979, 24, 2225.

9) Kuwajima, I.; Kato, M.; Mori, A. Tetrahedron Lett. 1980, 21, 4291. (b) Wada, M.
Chem. Lett. 1981, 153.

10) Hoffmann, R.W.; Froech, S. Tetrahedron Lett. 1985, 26, 1643.

11) (a) Hooz, J.; Gunn, D.M. J. Am. Chem. Soc. 1969, 91, 6195. (b) Hooz, 1.; Bridson,
NJ.; Calzada, J.G.; Brown, H.C.; Midland, M.M.; Levy, A.B. J. Org. Chem. 1973, 38,
2574. (c) Masamune, S.; Mori, S.; Van Horn, D.E.; Brooks, D.W.Tetrahedron Lett. 1979,
9, 1665. ((1) H002, 1.; Oudenes, J.; Roberts, J.I..; Benderly, A. J. Org. Chem. 1987, 52,
1347.

12) (a) Fenzl, W.; Koster, R. Liebigs Ann. Chem. 1975,1322. (b) Fenzl, W.; Kosfeld, H.;
Koster, R. Liebigs Ann. Chem. 1976,1370.

13) Mukaiyama, T.; Inoue, T.Chem Lett. 1976, 559.

51
14)(a) Inoue,T.; Uchimaru,T.; Mukaiyama,T. Chem Lett. 1977 ,153. (b) Van Horn, D.E.;
Masamune, S.; Tetrahedron Lett. 1979, 24, 2229. (c) Inoue,T.; Mukaiyama, T. Bull.
Chem. Soc. Jpn. 1980, 53, 174. (d) Paterson, 1.; Lister, M.A.; McClure, C.K. Tetrahedron
Lett. 1986, 247 , 4787. (e) Masamune, S.; Kim, B.; Petersen, 1.S.; Sato, T.; Veenstra, 1.;
Imai, TJ. Am. Chem. Soc. 1985, 107, 4549.
15) (a) Sugasawa,T.; Toyoda,T.; Sasakura, K.; Syn. Commun.. 1979, 9, 583. (b) Genari,
S.; Colombo, L.; Scolastico, C.; Tetrahedron Lett. 1984, 25, 2283. (c) Hamana, H.;
Sasakura, K.; Sugasawa, T.Chem Lett. 1984, 1729. (d) Chow, H.F.; Seebach, D.;
Helvetica Chimica Acta. 1986, 69.
16) (a) Brown, H.C.; Dhar, R.K.; Bakshi, R.K.; Pandiarajan, P.k.; Singaram, B. J .Am.
Chem. Soc. 1989, 111, 3441. (b) Brown, H.C.; Dhar, R.K.; Ganesan, K.; Singaram, B J.
Org. Chem. 1992, 5 7, 499. (c) Brown, H.C.; Dhar, R.K.; Ganesan, K.; Singaram, B J.
Org. Chem. 1992, 5 7, 2716. ((1) Brown, H.C.; Ganesan, K.; Dhar, R.K. J. Org. Chem.
1992, 57, 3767. (f) Ganesan, K.;Brown, H.C. J. Org. Chem. 1993, 58, 7162.
17) (a) Ganesan, K.; Brown, H.C. J. Org. Chem. 1994, 59, 2336. (b) Ganesan, K.;
Brown, H.C. J. Org. Chem. 1994, 59, 7346.
18) Niedenzu, K; Dawson, 1.W. Boron Nitrogen Compounds, Springer Berlin 1964.also
metal and mettaloid amides Koster, R.; Methoden Org. Chem. (Houben-Weyl) 4th. Ed
1952- (Organoborverbindungen) Vol. 13/3a/3b/3c, Thieme, Stuttgart, 1982, 1983, 1984.
19) Cortwright, 1.; Hill, A.F. J. Organomet. Chem. 1992, 424, 229. Bloyce, P.E.;
Mazcetti, 1.; Rest, A.1 . J. Organomet. Chem. 1993, 444, 223. Herberich, G.E.; Hessner,
B.; Osht, H. J. Organomet. Chem. 1988, 348, 305.
20) Brown, 1.M.; Lloyd-Jones, G.C. Tetrahedron; Assymetry 1990; I, 869. Corey, E.1.;
Bakshi, R.K. Tetrahedron Lett. 1990, 31, 611. Corey, E.1.; Reichard, G.A. Tetrahedron
Lett. 1989, 30, 5207.
21) Thomas, P.C.; Paul, I.C. Chem. Comm. 1968, 1130.
22) Brauer, D.1.; Burger, H.; Dorrenbach, F; Pawelke, G.; Weuter, W. J. Organomet.
Chem. 1989, 378, 125.
23) Burger, H.; Hagen, T; Pawelke, G. Z. Naturforsch. 1993, 48b, 935. Ansorge, A.;
Brauer, D.1.; Burger, H.; Hagen, T; Pawelke, G. J. Organomet. Chem. 1994, 467, 1.
Burger, H.; Hagen, T; Pawelke, G. J. F iuorine Chemistry, 1991, 55, 323.
24) Paetzold, P.; Biermann, H.P. Chem. Ber., 1977, 110, 3678.
25) Lappert, M.F.; Majumdar, M.K; Tilley, B.P. J. Chem. Soc. (A), 1966, 1590.
26) Brown, C.; Cragg, H.; Miller, TJ.; Smith, D.O'N. J. Organometallic Chem. 1983,
244 209
27) Clippard, F.B.; Bartell, L.S. Inorg. Chem. 1970, 9, 2439.

52
28) Jones, RJ.; Kinney, C.R.; J. Am. Chem. Soc. 1939, 61, 1378.
29) Aubrey, D.W.; Lappert, M.F. J. Chem. Soc. 1959, 2927.
30) Niedenzu, K.; Dawson, 1.W.; Fritz, P.;1enne, H. Chem. Ber., 1965, 98, 3050.
31) (a) Aubrey, D.W.; Gerrard, W.; Mooney, E.F. J. Chem. Soc. 1962, 1786. (b) Aubrey,
D.W.; Lappert, M.F.; Majumdar, M.K. J. Chem. Soc. 1962, 4088.
32) (a) Niedenzu, K.; Blick, K.E.; Boenig, I.A.; Rothgerg, E.F. Z. Anorg. Chem. 1972,
387, 107. (b) Niedenzu, K; Boenig, I.A.; Rothgerg, E.F.Chem Ber., 1972, 105, 2258.
33) (a) Aubrey, D.W.; Lappert, M.F. Proc. Chem. Soc. 1960, 148. (b) English, W.D.; Mc
Closkey, A.L.; Steinberg, H. J. Am. Chem. Soc. 1961, 83, 2122. (c)1.appert, M.F.;
Majumdar, M.K. Proc. Chem. Soc. 1961, 425.
34) Brotherton, R.1.; Buckman, T. Inorg. Chem. 1963, 2, 424.
35) Wiberg, E.; Bolz, A.; Bucheit, P. Z. Anorg. Chem. 1948, 256, 285.
39) Dorokhov, V.A.; Lavrinovich, L.I.; Yakovlev, I.P.; Mikhailov, B.M. Zh. Obshch.
Khim. 1971, 41, 2501.
37) (a) Lappert, M.F. Ch. 2 in Developments in Inorganic Polymer Chemistry, Eds.
Lappert, M.F.; Leigh, G.1. Elsevier, Amsterdam 1962. (b) Cragg, R.H.; Weston, A.F.
Chem. Comm. 1972, 79.
38).(a) Brotherton, R.1.; Steinberg, H. J.Org. Chem. 1961, 26, 4632. (b) Gerrard, W.;
Lappert, M.F.; Pearce, CA. J. Chem. Soc. 1957, 381. (c) Shchegolera, T.A.; Shashkora,
E.M.; Mikhailov, B.M. Bull. Acad. Sci. U.S.S.R., Div. Chem. Sci, 1961, 848. (d) Cragg,
R.H.; Lappert, M.F.; Tilley, B.P. J. Chem. Soc. 1964, 2108.
39). Galchenko, G.L.; Brykina, E.P.; Shchegoleva, N .N.; Vasilev, L.S.; Mikhailov, B.M.
Izv. Akad. Nauk. SSr. Ser. Khim. 1973, 200.
40). Dornow, A.; Gehrt, H.H. Z. Anorg. Chem. 1958, 294, 81.
41). Ruff, 1.K. J. Org. Chem. 1962, 27, 1020.
42). Gupta, S.K. J. Organometallic Chem. 1978, 156, 95.
43). Lappert, M.F.; Prokai, B. Adv. Organometallic Chem. 1967, 5, 225.
44). Jefferson, R.; Prokai, B.; Lappert, M.F.; Tilley, B.P. J. Chem. Soc. (A) 1966, 1584.
45. Mikhailov, B.M.; Pavarov, L.S. Z. Obshch. Khim. 1971, 41, 1540.
46. Mellev, A.; Gerger, W. Monatsh. 1974, 105, 684.
47. (a) Aronovich, P.M.; Bochareva, N.M.; Mikhailov, B.M. Zh. Obshch. Khim.. 1971,
41, 1562.. (b) Aronovich, P.M.; Bogdanov, V.S.; Mikhailov, B.M. Izv. Akad. Nauk.
SSSR.Ser. Khim. 1970, 1682.
48. Meller, A.; Ossico, A., Monatsh. 1972, 103, 577.
49. Jefferson, R.; Lappert, M.F. Intro-Sci. Chem. Rept. 1973, 7, 123.
50. Wille, H.; Goubeau, F. J. Chem. Ber. 1972, 105, 2156.

53
51. Brown, H.C.; Midland, M.M.; Levy, A.B. J. Am. Chem. Soc. 1972, 94, 2114, ibid.
3662., ibid. 95, 2394.
52. Davies, A.G.; Hook, S.C.W.; Roberts, B.P. J. Organometallic Chem. 1970, 23, C11.
53) Corbridge, D.C.E.; Studies in inorganic chemistry, 10, Elsevier Science Publishers:
New York 1990.
54) Arif, A.M.; Cowley, M.; Pakulski, M.; Power, 1.M. J. Chem. Soc. Chem. Commun.
1986, 889.
55) Pestana, DC; Power, P.P. J. Am. Chem. Soc. 1991, 113, 8426.
56) Power, P.P. Angew. Chem. Int. Ed. Eng. 1990, 29, 449.
57) Groshens, T.1 .; Hi ga, K.T.; Nissan, R.; Butcher, R.1.; Freyer, A.1. Organometallics,
1993, 12, 2904.
58) Power, P.P.; Moezzi, A.; Pestana, D.C.; Petrie, M.A.; Shoner, S.C.; Waggoner, K.M.
Pure Appl. Chem. 1991, 63, 859. Power, P.P. Angew. Chem, Int. Ed. Engl. 1990, 29, 449.
59) Groshens, T.1 .; Johnson, C.E. J. Organometallic Chem. 1994, 480, 11.
60) Corbridge, D.C.E.; Studies in inorganic chemistry, 10, Elsevier Science Publishers:
New York 1990.
61) Noth, H.; Staude, S.; Thomann, M.; Paine, R.T. Chem. Ber. 1993, 126, 611.
62) Timmer, K.;Harry, D.; Thewissen, M.W.; Marsman, 1.W. Reel. Trav. Chim. Pays-Bas
1988, 107, 248.
63) Wittenberg, D.; Gilman, H. J. Org. Chem. 1958, 23, 1063.
64) Davies, P.A.; Turchi, 1.1.; Greely, D.N. J. Org. Chem. 1971, 36, 1300.
65) Brown, H.C.; Kulkami, S.U. J. Organomet. Chem . 1979, I68, 281.
66) Manni g, D.; Noth, H. J. Chem. Soc. Dalton Trans. 1985, 1689.
67) Musgrave, G.C. J. Chem. Soc. 1956, 4305.

Chapter II:Preparation of Aluminum Enolates From Aluminum Amides

Literature review of Aluminum Enolates Structure, Properties, and Methods
of Synthesis

A review of the literature shows that descriptions of aluminum enolates are rare,
and no general method is available for their preparation. The best example for the
structure of aluminum enolates is the work by Jeffery et al. According to molecular
weight determinations in benzene, aluminum enolates tend to be associated in solution
with oxygen bridges.(10)

Work in this area during the past three decades has resulted in the preparation of
aluminum enolates via several indirect routes. These routes are in many ways analogous
to the routes to boron enolates described in chapter I.
l)-Coniu2§te addition.

Conjugate addition of trimethylaluminum to mesityloxide in the presence of a
catalytic amount of nickel acetoacetonate has been reported by Jeffery.(]b) This reaction
generates the internal enolate with 4:1 selectivity in favor of the Z stereoisomer in

excellent yield (eq 1).

 

A1M€3 ’OAIMC 3
MegC= CH-czo fir t-BuCH=C. (1)
Mc Mow): Me
15:2
1:4

ltoh, and Sazaki et al. have reported the 1,4 addition of dimethylbenzenethiolate,
dimethylaluminummethy] selenide, and diisobutylaluminumphenyl telluorolate to (1,8-
unsaturated carbonyl compounds as a route to the preparation of internal aluminum
enolates (eq 2).(2) No attempt was made to isolate or characterize the enolate in this

study.

\ .
/C=C RzAlX \c-C', RCHO Aldol (2)
,c=o ———> 4% C-OAle _——'
R X=SeMe R
SPh
TePh

Tsuda et al.. have reported the conjugate addition of DIBA-H to a,fi—unsaturated
carbonyl compounds in the presence of catalytic amounts of methyl copper(I) (eq3).(33)
In this study the presence of enolate was verified by quenching the reaction mixture with

water and the recovery of ketone.

C=c i-BmAlH \ - Hoo
\ ~ C" C \ " Ketone (3)
/ C=0 —M—C—" 4; C-OAli-Bu2 -——r
C U

Taniguchi and coworkers have suggested the allenolate intermediate generated by the
addition of diethylaluminum iodide to a,B-acetylinic ketones (eq4).(4) Subsequent
reaction of this intermediate with an aldehyde provided the expected aldol product in

moderate yield and selectivity.

 

OAlEl’)
~ OH
Rl’l\\ EtoAll MK RCHO O
\ ——"’ R . I _'—_’Rl R (4)
H I
z
R Z/lf 0 0H
1
Ph 64/16 R R
Me 54/22 I
E

2)-Transmetallation
Zakharkin and Savina have reported the reaction of DIBA-H with silylenolethers

of simple aldehydes as a means of generating the diisobutyl aluminum enolates of

56
quenching the reaction mixture with water and recovering the aldehydes in 80% yield

(eq5). '
B3SiOCH=CHR l-BUQAIH; i-BuzAlOCH=CHR + Et3SiH (5)

 

R=H or Me

Maruoka and coworkers have reported that simultaneous addition of aldehydes and or-
haloketones at low temperature to a suspension of zinc dust and a catalytic amount of
copper bromide and diethylaluminum chloride produced excellent yields of the expected
aldol products (eq 6).(6) The involvement of a zinc enolate that may be the result of the
kinetic Reformatsky reaction of 4—bromocrotonate makes the above results questionable

as presented by the authors. Representative results are provided in Table 1.2.
O OH

0
R2 EtoAIClZn
R"U\r + RCHO ~ > R‘JH/LR (6)

x CuBr(caL) R2

 

Table 1.2. Representative Results For The Reaction Of a-Haloketones With Zinc and
Et2A1Cl After Maruoka et al.

 

Ketone R syn/anti %Yield
2-bromocyclohexanone Ph 1/ 1 97
2-Bromo-2-metylcyclohexanone Ph 4/3 100

Methyl bromocrotonate Ph 5/4 100

Kurobushi has suggested the presence of an aluminum enolate generated from the
reaction of an a-fluoroenolphosphonate with LAH in the presence of copper (II) bromide

at low temperature. The reaction of this mixture with aldehydes produces the aldol

57

condensation product in moderate to good yields (eq 7).(7) Representative results are

provided in Table 2.2.
II
1 ,O-P(OEt)3 0 A14 2
R FC-QR CUB”? , R‘Fczq fl, Aldol (7)

LAH R

Table 2.2 Representative results for reaction of an a-fluoroenolphosphonate with LAH

 

after Kurobushi et al.
31 R R2 sun/anti %Yield
CF3CF2 C-CGH l 1 Ph 1.5/1 51
(E)—CH3CH(CH3) “0.8 38
CF3(CF2)5 CH3(CH2)2 Et l/0.8 84

3)-1,2 Addition of Metal Alkvls

The 1,2 addition of organometallic compounds to ketene is a well established
method for the preparation of metal enolates.(10) Jeffrey and Meisters examined the
addition of trimethylaluminum to diphenyl ketene in order to obtain an internal aluminum
enolate (eq 8).(1 1) Ebulometric determination of the molecular weight for this enolate in

benzene revealed the presence of dimeric structure in solution.

AIM 'OA'(MC)3 HCI Ph
— , ’3
Ph3C2C=O——C3—-> thC—QM ~ Me (8)
C

‘1

(d1 mcr)

58
4)-C§rbonyl Enolization

These methods take advantage of the acidity of the a-proton of the carbonyl
compound and the use of a strong base to effect the enolization. Jeffrey and Meisters
have prepared enolates of sterically hindered ketones by the reaction of
trimethylaluminum with triphenylmethyl ketone or pinacolone (eq 9,10).(831
Etras and Seebach have prepared the enolate of ethyl trityl ketone in an analogous
manner.(8b) Table 3.2 provides a sample of the Seebach et al. results with representative

aldehydes in an aldol reaction after 40 hours.

P113C AIM63 PhaC RCHO
to ———> 2—OAIMC3 _, Aldol (9)
CH; CH3 (dimer)
\ /

A1M63 Al
0 I (AF + l/2CH4 (10)
Me / \

Table 3.2 Representative results for reaction of trimethylaluminum with ketones after
Seebach et al.

 

R syn/anti % Aldol
Ph 5/95 96
p—CH3OPh 7/93 95
p—NMezPh 8/92 92
p—CNPh 1/99 93

p-NOzPh 1/99 95

59

Tsuji et al. found diisobutyl aluminum aryloxide in combination with pyridine to be an
effective agent for the regioselective aldol condensation of methyl ketones at the methyl

side (eq 11).(9)

O O

)L i-BuzAlOPh >\/U\
(CH215CH3 *7 CH3(CH2)5 \ (CH2) 5CH3 (H)

Pyridine

 

85%

To our knowledge the latter examples represent the only attempted enolizations of

carbonyl compounds via an aluminum-heteroatom base and trimethylaluminumuz).

Literature review of Aluminum Amides Structure, Properties, and Methods
of Synthesis

Aluminum amides are compounds in which aluminum is bonded to one or more
nitrogen ligands. These compounds have been the subject of study in the past thirty years
and several reviews are available on their structures and properties.(13a) They are
generally air and moisture sensitive and are handled under an inert atmosphere. Their
uses include reducing agents(13), polymerization catalysts(14), Diels-Alder catalysts(15),
amidation agent(16), precursors to aluminum cage compounds,(17) aluminum nitride
coatings for ceramics(18), reagents for the isomerisation of epoxides to allylic
alcohols(19)’ and a reagent in the Fisher indole synthesis.(201 In contrast to boron
amides, there are only a few examples of aluminum amides in which the metal and
nitrogen are both tri—coordinate. These include the sterically hindered aminoalanes, such
as Al[N(i-Pr2)3], Al[N(SiMe3)2]3, and a new monomeric azaaluminatrane A1(tert-
BuMe28iNCH2CH2)3N synthesized by Verkade et al.(21) Other aluminum amides are
dimers, oligomers, or polymers, formed as a result of intermolecular dative bonding
between monomer molecules which have donor (N) as well as acceptor sites (Al). The

degree of association of these amides depends on the ligands on the aluminum as well as

60
the ligands on the nitrogen atom. For example Me2AlNHMe is trimeric(22) whereas

MezAlNMez is dimeric(23) as determined by X-ray crystallography.
Prgaration of Aluminum Amides

There are several routes available to the synthesis of aluminum amides. These
methods generally produce aluminum amides in excess of 90% yield. The most useful
routes are the alkane elimination(24) from a trialkylaluminum precursor or hydrogen
eliminationas) from an aluminum hydride precursor. Examples of these methods are
demonstrated in equations 1 and 2. The reaction of aluminum hydrides provides a milder

method since aluminum hydrides are more reactive than aluminum alkyls.

NHM62
Mc3Al _____' MegAlNMcz 4. CH4 (1)

100C

NHMcz g
MCzAlH ._______,, MCgAlNMCZ + H,’ (_)
25C
These two methods are by far the most efficient methods since the byproducts of the
reactions are gases and are easily removed from the reaction vessel. The other attractive
feature of these methods is the measurement of the amount of gas produced provides a
means of monitoring the progress of the reaction.

A direct method for the preparation of aluminum amides is reaction of aluminum

metal, hydrogen gas, and the amine in a sealed reaction vessel at 4000 psi in benzene

(eq3).(26)

3NHE12
Al + 30.11; , A1(NEI2)3 + 3H2 (3)

4000 psi
Salt elimination from the reaction of LAH and an amine hydrochloride has been

used to prepare aluminum alanes (eq4).(27)

61

L1A1H4 + ElzNHCI 3E12NH $ AlNEl3 + L1C1 + 3H2 (4)

 

Metathetical exchange between trialkyl boranes and tris aluminum amides

provides another method for the synthesis of dialkylaluminum amides (eq5).(28)

2R3B + A1(NMeg)3 , 2RgBNMc2 + RgAlNMeg (5)

 

Hydride insertion into the nitrogen carbon double bond of aromatic amines by

DIBAL is a route to the synthesis of aluminum amides (eq6).(29)

. N \ .B I \
' , —___' ' UzA -N
\ /

We used compounds I and II for our initial studies for the following reasons:
l)-Compounds I and II are easily prepared by elimination of methane from the reaction
of trimethylaluminum with diethyl- and diisopropylamine.
2)-Sterically hindered amines are less nucleophilic hence carbonyl addition reactions can
be avoided.
3)-Methyl ligands on aluminum avoid a possible B-hydride elimination reaction
commonly observed with ethyl and isobutyl ligands.
4)-Analogous lithium bases have proved to be useful for enolate chemistry.
5)-Due to the symmetry of the molecules NMR spectral interpretation would be simple.
6)-Methy1 substituents on the aluminum center are converted to methane upon protic

workup.

62

/...\
N\ /N H
\ /\
J \N/Jl

/.\
N\ /N
\ /\
_ \N/

63

Results and Discussion

Our studies began with the preparation of compounds I
(Dimethylaluminumdiethylamide) and II (Dimethylaluminumdiisopropylamide). These
compounds were prepared by the reaction of trimethyl aluminum with a stoichiometric

amount of the dialkylamine as previously described by Thomas et al . (eqs 1, 2).(3O)

 

 

CH3 CH3
EQNH \\
A1 — CH3 ’ A1 — NEIZ + CH4 (eql)
CH Toluene,Reflux CH
3 Quantitative 3
I
CH3 i CH3
PrZNH .
A1 —CH3 ? Al —N'Prz + CH; (e142)
Toluene Reflux
CH ’ .
3 Quantitative CH3
II

We found 1H-NMR spectroscopy to be a powerful method for the direct analysis of the
course of reaction throughout our studies. The nondonor solvent deuterated benzene was
chosen to prevent complexation of the solvent to the reagent. At a later point in our
studies we found deuterated chloroform to be just as useful and less costly. Hence 100
pmoles of pinacolone was injected into an NMR tube containing 100 pmoles of
compound I in d6-benzene at room temperature. Immediate analysis of the reaction
mixture at ambient temperature showed no evidence of reaction. Upon standing at room
temperature for 3.5 hours the sample was reanalysed and no change was observed,
(Figure 1.2). In an analogous fashion compound II was tested and no reaction was

observed after 3.5 hours.

64
We attribute the lack of reactivity of these complexes to the formation of strong

aggregates in solution. The steric contributions of the diisopropyl ligands do not seem to
have a significant effect on the reactivity of compound II to a carbonyl substrate. We
reason that since aluminum is a strong Lewis acid and nitrogen is a strong Lewis base,
association effectively neutralizes both the Lewis acidity of the aluminum and basicity of
nitrogen to the extent that proton transfer to the nitrogen is prohibited on thermodynamic
grounds.

We then prepared compound Ill (Dimethylaluminumdiphenylamide) to see if
conjugating substituents on nitrogen can reduce the strong association of aluminum
amides we previously encountered with I and II and whether enolate formation can be
effected. Hence compound III was reacted with pinacolone at room temperature. Upon
addition of 100 pmol of pinacolone to an NMR tube containing 100 pmole of compound
III at room temperature a burst of yellow color was observed. Immediate NMR analysis
showed the presence of new vinylic resonances and a complex mixture of other
resonances in the aliphatic region with release of amine, (Figure 2.2). This experiment is
the first observation of an aluminum enolate prepared via an aluminum amide.
Reexamination of the NMR tube after 3.5 hours showed a very simple spectrum that had
only a singlet in the aliphatic region as its major component and vinylic protons in
addition to the aromatic resonances of the diphenylamine and methyl resonances on
aluminum, (Figure 3.2).

We attribute these observations to the initial formation of a nitrogen bridged
enolate that is highly reactive and condenses with remaining ketone in solution. The
condensed aldol product is itself unstable and hence in equilibrium with the starting
Enolate I that dissociates to an enolate dimer, Enolate II (scheme 1.2). Collum et al.

have spectroscopically detected (NMR) the presence of limited concentrations of mixed

65

Ph\ /Ph
PhNPh\/

MegAlNth )ak qu \Aj/\N/\Al/____' \Al/NA1\ /
CDCl3,..rt w/\/\‘_/\OA{I/o\

X8 W

Enolate I Aldol Complex

3%

/ /
\/\

O

 

Z/O

\
/Al
Enolate IV

Scheme 1.2 Proposed Mechanism For The Reaction Of 111 With one Equivalent of

Pinacolone.

66
aggregates similar to Enolate I in their investigations of the reaction of LDA with
pinacolone(32). Similar observations were made with addition of one equivalent of
diethyl ketone to 111. However in the case of diethyl ketone, four individual vinylic

resonances were present which may be attributable to the diastereomeric enolates and
their aggregates, (figure 4.2).

When a half equivalent of pinacolone is used the reaction can be cleanly halted at
the Enolate I stage without any detectable amounts of the aldol complex (eq3), (Figure

5.2).

>5BK PK ”1
CH3
\Al—NPhg seq? \ Al/ NA1\ /+ PhoNH (3)

CH3/ Benzene,r.L M/ \ ()/Al \+

 

Enolate 1

A series of carbonyl substrates were then tested to determine the scope of enolate
formation with this reagent. The results of these experiments are shown in Table 4.2.
These results show the formation of the kinetic enolate in all cases studied with moderate
to excellent stereoselectivity and regioselectivity. The stereochemistry of the enolate of
3-pentanone was determined by nOe experiments. Irradiation of the methylene protons
afforded enhancement (8.56%) at the vinylic methyl protons, whereas irradiation of the
vinylic hydrogen did not show an enhancement at the methylene protons. All reactions
showed complete consumption of the ketone and release of diphenylamine. The NH
proton was used as an internal standard to measure the extent of the reaction. In contrast
to our results reaction of esters with aluminum amides has previously been shown to

produce the carboxylic acid amides by Weinreb et al. (eq 3a).(31)

67

CH3C02(CH2)3CH3
MegAlNth : CH3CONHPh (3a)

benzene, 40 hrs 78%

Since enolate formation had been accomplished, we wished to examine the aldol
condensation reaction of the Enolate I with benzaldehyde using lH-NMR to follow the
course of the reaction. Addition of benzaldehyde to this enolate resulted in the
consumption of the aldehyde but the enolate resonances remained. The conclusion from
this experiment was that the bridging diphenyl amido ligand in Enolate I interferes with
the aldol reaction by possibly adding to the aldehyde. Hence the diphenylamido ligand in
Enolate l was replaced with t-butylalcohol by addition of a stoichiometric amount of t—
butyl alcohol to generate Enolate II (eq 4), (Figure 6.2). Other alcohols, such as

methanol and ethanol reacted with Enolate I to produce pinacolone.

Ph\ N/Ph I
\,/\N/ m,\,/\,/
/W\/\smn /“\/“\

Quantitatii c

(4)

 

Enolate 11

Addition of one equivalent of benzaldehyde to Enolate II resulted in complete

consumption of the aldehyde with concurrent consumption of the enolate in ten minutes

(eq 5), (“gm 7.2).

+

O\N/ PhCHO \ Al/ O\Al / (5)
/ \ B. M/ \ .Ai/ .\

0 Quantitative

 

/\0

\Al
/

Enolate II

Four valuable conclusions were made from these sets of experiments:
1) Presence of conjugating substituents such as phenyl on the amido ligand increase the
reactivity of an aluminum amide to carbonyl substrates presumably because the Lewis
acidity of the metal complex is greater.
2) Enolate formation can be accomplished efficiently and cleanly with half equivalent of
ketone.
3) The self-aldol reaction following addition of stoichiometric amount of ketone is
reversible.
4) The bridging amido ligand can be replaced by a sterically hindered alkoxide ligand
without quenching of the enolate.
Rglgents Inert To Enolate II

Reactions of Enolate II with a variety of electrophiles at room temperature were
examined by addition of the electrophile to NMR samples prepared from a stock standard
solution of Enolate II. The electrophiles studied were t-butylchloride, styrene oxide,
benzoylchloride, acetic anhydride, and ethylchloroformate. In all cases no reaction was

observed after 24 hours at room temperature.

69

Table 4.2 Scope Of Enolization, Stereoselectivity and Regioselectivity Of III.
Experiments were carried out with 50pmoles of III in benzene at 25C.

Ph\ N/Ph

\N/N\N/Km

 

/“\/“\
thph/\

Ph\N/Ph
\Al/\N/
W/\/\

«K

 

 

R
Ketone Regioselectivity Stereoselectivity % Conversion
Less Subs. More Subs. E Z
W >98 100
O
M 4.8 1 100
O
100
mk/ 50 50
0
fly >67q <1a 100
O
>/l1\ 1.9 l 100

a- Due to the overlap of the ring resonances an accurate measure of the regioselectivity
was not possible. Based on the amount of amine released and amount of the less
substituted enolate generated an estimate of the more substituted regioisomer was made.

70
The lack of reactivity of Enolate II is attributed to the coordination of the two

aluminum centers to the oxygen of the enolate. It is evident that coordination of the
oxygen of the enolate to two aluminum centers reduces the nucleophilic character of the
enolate towards electrophiles.

The stereoselectivity of the aldol condensation reaction was studied with the
enolate of diethyl ketone prepared in an analogous manner to that used for Enolate II and
benzaldehyde. A low temperature NMR study of the reaction mixture showed complete
consumption of the enolate at -20C with the appearance of the aldol carbinol doublets in a
4:1 ratio in favor of the anti aldol product. Further examination after 24 hours of
standing at room temperature showed isomerization of the aldol product to a 1:6 ratio in
favor of the syn diastereomer.

The studies conducted on Enohte I] led to the development of aluminum amides
that would incorporate one basic component in a dimeric aluminum amide. At the same
time we were interested in increasing the reactivity of the enolate by increasing the Lewis
acidity of dimeric complexes analogous to Enolate II. For that purpose a series of mixed

aluminum amides of the general type MezAlXNRzAlMez were synthesized (eq 5).

W M\/\N/
Toluene M/R \ N/Al R\

 

Me3Al + MegAlX +CH4 (6)

X=CH3
F
Cl
R=Ph

TNHS
Et
LPr
TNH’

Reactions of these amides with pinacolone were studied to evaluate the efficiency of the
acid/base reaction vs the self aldol previously observed with the symmetrical complex

III.

71

Table 5.2 provides the compilation of results of these amides with pinacolone. It is
evident that the bridging atom plays an important role in the reactivity of these mixed

dimers towards pinacolone.

Table 5.2. Summary Of The Reaction Of Aluminum Amides Having Non-Nitrogen
Bridges With Pinacolone.

 

X R %Enolate %Aldol Compound
Cl Ph 72 28 IV

CH3 Ph 60 40 v

Cl TMS 6O 40 VI

F TMS >95 VII

F in >95 VIII

C1 in 60 40 IX

Cl Et >95 X

In the cases of compounds IV and X a precipitate was observed during the enolization
reaction when carried out in toluene or hexane. The precipitate was isolated and
dissolved in d-chloroform.1H-NMR analysis showed the absence of enolate in this
solution, (Figure 8.2). Analysis of the supernatant showed the enolate presence in that
fraction, (Figure 9.2). In the case of the diethylamido complex, this precipitate is formed
rapidly upon addition of the ketone and persists from the beginning of the reaction only
when the reaction is carried out in hexane. The latter observation may be due to the lack
of solubility of the amine complex as the driving force for rapid disproportionation of the
chlorine bridged enolate. We propose that upon formation of the bridging enolate
disproportionation of this complex occurs, followed by the formation of an enolate dimer.
The insoluble precipitate that forms is the amine complex of dimethyl aluminum chloride

according to Scheme 2.2.

72

Since Enolate IV could be prepared amine free, we proceeded to prepare a
halogen-bridged enolate by the reaction of Enolate IV with an equivalent of
dimethylaluminum chloride (eq7). The disproportionation reaction was complete after 24
hours at ambient temperature, (figure 10.2) or by heating at 80C for half an hour. These
two enolate complexes display distinct chemical shifts in their NMR spectra. Acylation
of both enolates was attempted in separate experiments. In contrast to Enolate IV,
Enolate III can be acylated with two and a half equivalents of ethyl chloroformate to

produce the ketoester of pinacolone and pinacolone in 2.3:] ratio.

 

O\2/O

Al/ 2Me3AlCl \ /Cl\ /
A'\ cm, / \ / \

Enolate IV Enolate III

\NC'N/\/ ><qum\/\AH/
/AlNAl\/\ Hexane “IO/“M

/ NB\ >(K
Enolate III

N/ + AlMezClNHEtz
\

NHEIQ

 

\Al:
/

/\O
O\/0

i

Enolate IV

Scheme 2.2 Proposed Mechanism For The Reaction Of X With one Equivalent of

Pinacolone.

74

A series of other mixed amides were prepared and their reactions with 2-

pentanone and 3-pentanone were studied in order to determine the generality of this

methodology. Two or more sets of vinylic resonances were observed at times in most

cases which may be due to disproportionation of these complexes. A compilation of

these results is provided in Table 6.2

Table 6.2 Summary of Chemical Shift Data of Enolates Generated by the reaction of

aluminum amides having Non-nitrogen bridges. Chemical shifts are reported in ppm.

 

X R 3-Pentanone 2 Pentanone Pinacolone Compound

Cl 4.9(q) 4.3, 4.5(s) 4.4, 4.6(s) IV
5.0(t)

CH3 Ph 4.4, 4.7(s) V

Cl TMS 4.4, 4.5(s) VI

F TMS 4.9(q) 4.2, 4.4(s) 4.5, 4.8(s) VII

5.1(q) 4.7(t)

F in 4.5,4.5(s) VIII

CI in 4.6, 4.8(s) IX

Cl B 4.6, 4.8(q) 4.1, 4.3(s) 43, 45(5) X
4.51(t)

F Et 4.1, 4.3(s) 4.4, 4.5(s) XI
4.8(t)

F TMP 4.7,4.9(q) 4.4, 4.5(s) XII

Cl TMP 4.6, 4.8(s) XIII

Examination Of The Reaction Of Monomeric Aluminum Amides With Ketones

Two monomeric aluminum amides were prepared and tested to determine their

reactivity in an acid/base reaction with ketones. Tris[bis(trimethylsilyl)amino]aluminum

(XV) was inert to the ketones tested. In contrast tris(diisopropylamino)aluminum (XIV)

75

reacted with pinacolone efficiently to produce the enolate at ambient temperature.

Attempts to selectively substitute other bulky ligands for one of the amido ligands failed.

76

 

o
#500
0.79::
Xmas
K

 

—l.7312

 

 

III-Fir

 
 

 

.. . Lb-

114441d%44ddd1ddd—ddddddiddd—J¢1¢dddddd4dddddqdd-4441«dd«1.44d1-ddd1-111q44+4—

m m A u w u c In

Ema—d FN 5-2—5w $85.3 a. 9393;»—:BE:3&2=<§BE« “52
23an an E30225 man «822% m" Nun «on mm 79.3

1- I 2187
9.9335
£462

 

 

 

 

  

 

 

 

U1 . .
a. .3 m
. o
A _
fl 4
I\II \II \H y
'I xr HI» I I; :ll‘fli P I
d‘qd “in“ 1.11! dd.‘ 17“. cddd dqddd‘Jl-‘ddi qqddquId—dddil-‘idd‘ddfifidqqqddi—lqdddddu‘dl
.15. . m . m . ._ a. m . Pi q.
. r1... 3 0.. Po _ .o
m... we 9. Z. 2.... human";

mama-=6 N.» _:-22=~ 383.3 cm 9323.5.camacaamvroafifizaa ”:2
2550: 0.. can 355:2: om 1:80—25 a: nun Ea 533:5 maeqam

14.1241

 

 

 

 

 

7
2
a.
fi
_
3 7MB
main“
wamm...mm . m
27777 71& 1R 1 fl
7 _ 59 0
Q. \ 4‘” _
‘
xx x P
> b —llr b .— LF
qqquqqqqu—qjqu‘JQfiA-dqd-diqdd.dqddduddq-dddudddqd—udddjjd—ddfiuddddu—quaddeJd‘quddd-ddd‘
w m m A u m n can»: a.
..IL._.r1. .11... r... .t r. .1.
3b; Mk f~ ab mu.» Nb 3..

Ema—d u.» _:-Z?=~ $005.3 0.. 9335.».camazaamvraatmaaa ~22
2530: an 2.0 3:332: 2. 3383.5 a. N8 2:. «335m 3.. A 30:8

 

 

 

 

 

 

 

 

 

 

 

I d dilWl‘ d. 1 1 d d 1 1 1 - 1 d 1 d — d d d d - d d 1 4 q 1 dild 4 d ‘l‘ d d d. d 1 4 - d d 1 d d ‘4 4 d
m. mb 9m 9o uh uh m. ~.o _.m —.o 20:
wild-ll. {Jr-ILA...
u~o 0.2 ..3
ubm PS
14‘1dfi11q1-4dud-1dudqlulfidfl-114lqdwlql-111-dddi1-dddd—dewI-Gil-ddquddqdddddfidddddddddd‘dddd‘.‘
u m m A u m a o 20: up

30...... a.» 5.230 €8.23 2. 9323.».50230303533... .0».
000500 0.. 000 0930—00” 0*. 900080000 an Mun 000 080015 3.. 3 000500

 

 

 

 

ENE—.0 m.» 02.23% 00000.00. 00. 93003.0.03000300000000.0000 00.8..
000500 00. 000 00:. 00:78.00. 0.. 000000—000 0. NMO 000 000:.me

81

10/ \3. _
O:

 

 

 

 

30.5. a.» 5.22.. 0.0855. o. 0.5....» ..

 

 

 

 

 

 

 

 

Emana a.» .:-Z.<=~ 00000.03 0.. 0.0 0.00. 00000000000 0.. M0058 ~—

 

 

- 4.6747

-I.Z759

-1.1699

 

 

 

 

ErLEr

q d d 4 - 4 l! d 1 d d d 1 d .— < l4 1 1 d d u d d

Jfid-4<«-<qd<~<diq—qq<d—.uqudqfiqu—aqq<u
m 0 w m . 0002

i

Ema—.0 a.» .I-ZZ..~ 00003:: 0.. 0.0 0.00.0208 4000.05 3.0:. 0.0 300000
0.. 00.000000 N :0... 0.008.000 .0 00.300.

1 0821
1.0780

p0 0003
x -0.7315

\WWB'T

m.

0. ya
‘ 1 7
63 38 6
7 1 “ .
mam... \00u .7
" 0 0 01. r

1.0 1100

L ./ I t.

 

-7.2632
l-—— —4.4522

——l

 

 

 

mm.
2......
r : Lr . l. :1:

dfiiiddqddd—duqu—1114d<411dl1444dlfiddd—ddqud‘dufid‘fiqdifiddddd‘dddldjq<jld~ldddJinqudddldqddfidad

~ 0 m . 0 m N . o .. 00x

 

 

 

 

0.05.0 9N .:-Z.<=~ 00005:: 0.. 0.0 0000.00.00. 0.. 0.0 4000000 0.. N 0.....
0.008.000.

 

 

 

 

 

 

 

0... mm. mm m...
4 44.4 0. 4... 4,
x J I. [.
m. m
.. 4 .m.
M
4..
J
0... 0... 0.~ 0.0. 0..... 0.0 001
. , 0. _ Kr. E...

 

 

 

.4144 .qua-<j4_4.4+uqqaqdq<d.11111d01404-0.W Ajjfiqq.—ulq._jajdjl1334fii
. 0 m . . 0...... -.

305.0 5.» .I-Z.<=~ 00000.50 0.. 0.0 00000000 0.. 0000—08 —< 0..... 7.03.0.

86

Conclusions

A new procedure has been developed for the preparation of aluminum enolates via
an acid/base reaction previously unknown for aluminum amides. Our results provide the
first observation of an aluminum enolate prepared via an aluminum amide base. The
reactivities of these bases are greatly influenced by the nature of ligands on the aluminum
center as well as on the nitrogen. We have demonstrated that use of conjugating
substituents on the amido ligand can overcome the strong association present in
aluminum amides bearing alkyl substituents on nitrogen. We have shown that weakly
bridging ligands in mixed dimeric complexes provide these complexes with enough
reactivity so that an acid/base reaction is favorable. We have also shown that monomeric
aluminum amides react with pinacolone. Our results also provide a new mechanistic

insight into the reactivity of these new bases in solution

Experimental

All operations were carried out under an argon or nitrogen atmosphere.
Diphenylamine, diisopropylamine, diethylamine, 1,1,1,3,3,3-hexamethyldisilazane
(HMDS), tetramethylpipridine (TMP), hexane, and toluene, were obtained from Aldrich
Chemical Company and distilled from calcium hydride prior to use. Deuterated benzene,
toluene, and chloroform were dried over calcium hydride and purged with nitrogen.
Ketones were obtained from Aldrich Chemical Company and dried over calcium hydride
and purified by distillation. Trimethylaluminum, and dimethylaluminum chloride were
obtained from Aldrich Chemical Company as neat substances and stock solutions were
prepared as needed inside a glove box. Sodium fluoride was obtained from Aldrich
Chemical Company. lH-NMR data were recorded on a Varian spectrometer operating at
300 MHz. NMR tubes were dried at 150C in an electric oven, fitted with a septum, and
purged with nitrogen using along needle for at least one hour prior to use.

Apparatus Used For Prglaration Of Aluminum Amides

Due to the pyrophoric nature of the reagents and their air and moisture sensitivity,
a one step preparation of aluminum amides was developed. This method minimizes the
number of transfers of the reagents and allows for a one pot synthesis of the amides as a
standard stock solution. The apparatus used is shown in Figure 11. This apparatus also
allows one to monitor the course of the reaction by measuring the amount of gas
produced.The volumetric flask A (10 mL) was modified by the MSU glass blowing
facility to incorporate a 14/20 ground glass joint and a sample inlet having 8 mm
diameter for removal of sample aliquots. Sample removal was performed through the

sample port as shown in Figune 12.

88

Preparation of Aluminum Amides

Preparation of compounds I, II, 111,80) V,(33) XV, XIV(21) have been
previously reported by several investigators. Compounds IV, X, IX, XIII can be
prepared via two procedures. Procedure A is a direct method and does not require the
preparation of amine hydrochloride salts. In procedure B an amine hydrochloride salt is
reacted with trimethylaluminum. Both procedures produce identical results except in the
case of compound VI, where reaction of I-ICl gas with HMDS cannot be used to prepare
the amine hydrochloride salt of this amine.
Procedure A

Trimethylaluminum, 10 mmoles (0.72g) and dimethylaluminum chloride 10
mmoles (0.92g) were weighed directly into the volumetric flask (A) in a dry box. The
mixture was protected from air using a septum and cooled to -78C. Argon was
introduced through the sample inlet (D). The condensor (B) and the gas burette (C) were
assembled and the entire apparatus was purged with argon for 20 minutes. 10 mmoles of
the desired amine was introduced through the sample inlet at -78C and the cooling bath
was removed. The apparatus was then allowed to come to room temperature. The
reaction vessel was immersed in an oil bath at 110C and gas burette was zeroed. Slow
evolution of methane started. After 12 hours of heating gas evolution ceased (230 mL).
The apparatus was disassembled under a stream of argon and the volumetric flask was
fitted with a septum. The solution level was adjusted to the mark by addition of solvent.
In the case of compound VI reaction was completed after 3 hours at room temperature.
Procedure B

The desired amine hydrochloride (5 mmoles) was weighed directly into the
volumetric flask (A) inside the dry box. To this solid was added 3 mL of dry toluene and
the suspension was cooled to DC in an ice bath. The apparatus in Figure 11 was
assembled and purged with argon for 20 minutes. 5 mL of a 2M trimethylaluminum in

toluene (10 mmoles) was introduced in a dropwise manner via a syringe. Gas evolution

89
started and 5 mmoles (115 mL) of methane was produced after one hour with

disappearance of the solid. The reaction vessel was then allowed to come to room
temperature and immersed in an oil bath at 110C. The gas burette was zeroed after a 10
minute equilibration period, and the mixture was heated for 12 hours after which gas
evolution seased. At the end of the heating period an additional 5 mmoles of methane
was generated. Near quantitative amounts of methane evolved in all cases.
Gene‘rai Procedure For Preparation Of Fluoro-Substitiuted Aluminum Amides

Dimethylaluminum fluoride was prepared in situ according to the procedure by
Laubengayer et al.(34) The volumetric flask (A) in Figure 11 was charged with 10
mmoles (0.42g) of sodium fluoride and 10 mmoles (0.92g) of dimethylaluminum chloride
inside a dry box. Toluene was added (3mL) and the apparatus in Figure 11 was
assembled under a stream of argon. This suspension was heated for 16 hours at reflux
after which a fine white precipitate was present. The 1H-NMR analysis of a IOuL sample
after evaporation of toluene under vaccuo and resuspension in d6 benzene showed a
singlet at -O.56 ppm.(syrupy clear liquid). The vessel was then charged with 10 mmoles
of a 2M trimethylaluminum solution in toluene (5 mL) at room temperature. The vessel
was immersed in an oil bath at 110C, allowed to equilibrate, and gas burette was zeroed.
The desired amine (10 mmol) was added. Slow evolution of methane started and ceased
after 1.5—8 hours. Near quantitative amounts of methane evolved in all cases. In the case
of compound VIII the reaction was complete after 1.5 hours.
General procedure for NMR Tube Reactions

An aliquot (50 }4L) of the aluminum amide in toluene was injected into an NMR
tube that had been previously purged with nitrogen. Toluene was evaporated under
vaccuo, 0.7 mL of deuterated solvent was introduced via a syringe and the tube was
shaken to dissolve the aluminum amide. 50 umoles of the desired ketone was injected

directly into the tube, the tube was shaken and spectra recorded.

90
Preparation of u-Diphenylamido, u-Chloro, Tetrametlyldialuminum Compound I!

The vessel described in Figure 11 was assembled and purged with argon for 15
minutes. 10 mmoles of trimethylaluminum (SmL of a 2M solution in toluene) was added
slowly at DC to 5 mmoles (1.02g) of diphenylamine hydrochloride. Upon completion of
addition a clear solution resulted with evolution of 114 ml. of gas. The ice bath was
removed and the reaction was allowed to come to room temperature. After stirring at
room temperature for 12 hours gas evolution ceased with an additional evolution of 119
ml. of methane. An aliquot (100 pL) was removed from the vessel through the sample
inlet, toluene was evaporated and remaining white crystalline solid was dissolved in
deutero chloroform.
lH NMR(300MHz;CDCl3) : 6 —O.62(s, 6H), 7.17(m, 2H), 7.28(m, 4H), 7.35(m, 4H)
Preparation of u-HMDS, u-Chloro, Tetramethyldialuminum Compound 3;!

The vessel in Figure 11 was assembled and purged with argon for 15 minutes. A
room temperature water bath was used to cool the vessel. 5.0 mmoles of
trimethylaluminum (2.5 mL of a 2M solution in toluene) and 5.0 mmoles of
dimethylaluminum chloride (2.5 mL of a 2M solution in toluene) were introduced
through the sample inlet via a syringe. 5.0 moles of HMDS (1.0 mL) was introduced
via the sample inlet in a dropwise manner. Slow gas evolution was evident. After 3.5
hours 120 mL of gas was produced. An aliquot (100 yL) was removed from the vessel
through the sample inlet, toluene was evaporated and remaining clear syrupy liquid was
dissolved in deutero chloroform.
1H NMR(300MHz;C6D6) : 6 -O.19(s, 12H), O.21(s, 18H)

Prefipgrgtion of u-Diethylafldoir-Chloro, Tetramethvldialumintuln compound};

10 mmoles of trimethylaluminum(0.72g) and 10 mmoles of dimethylaluminum
chloride(0.92g) were weighed directly into the volumetric flask (A) in the dry box. The
mixture was protected from air using a septum and cooled to -78 degrees. Argon was

introduced through the sample inlet (D). The condenser (B) and the gas burette (C) were

91
assembled and the entire apparatus was purged with argon for 20 minutes. 10 moles of

diethyl amine (1.2 mL) was introduced through the sample inlet at -78C and the cooling
bath was removed. The apparatus was then allowed to come to room temperature. The
reaction vessel was immersed in an oil bath at 110C and gas burette was zeroed. Slow
evolution of methane started. After 12 hours of heating gas evolution ceased (231 mL).
Stock solution was prepared by adding toluene to the mark. An aliquot (100uL) was
removed from the vessel through the sample inlet, toluene was evaporated and remaining
clear syrupy liquid was dissolved in deutero benzene.

1H NMR(3OOMHZ;C6D6) : 6 -O.15(s, 12H), 0.78(t, J=7.1 Hz, 6H), 2.75(q, J=7.1 Hz, 4H)
Prepfiagm of p-DiisopropylamidoLu-Chloro, Tetramethyldiajyminum compou_rgl_I_X_

10 mmoles of trimethylaluminum(0.72g) and 10 mmoles of dimethylaluminum
chloride(0.92g) were weighed directly into the volumetric flask (A)in the dry box. The
mixture was protected from air using a septum and cooled to -78C. Argon was
introduced through the sample inlet (D). The condenser (B) and the gas burette (C) were
assembled and the entire apparatus was purged with argon for 20 minutes. 10 mmoles of
diisopropyl amine (1.4 mL) was introduced through the sample inlet at -78 and the
cooling bath was removed. The apparatus was then allowed to come to room
temperature. The reaction vessel was immersed in an oil bath at 110C and gas burette
was zeroed. Slow evolution of methane started. After 12 hours of heating gas evolution
ceased (230 mL). Stock solution was made by adding toluene to the mark. An aliquot
(100uL) was removed from the vessel through the sample inlet, toluene was evaporated

and remaining clear syrupy liquid was dissolved in deutero benzene.

1H NMR(3OOMHZ;C6D6) : 6 -0.02(s, 12H), l.18(d, J=6.7 Hz, 12H), 3.55(m, J=6.7 Hz,
2“)

92
Preparation of u-Tetramethvlpipridido, u-Chloro, Tetramethvldialuminum compound
El;

10 mmoles of trimethylaluminum(0.72g) and 10 mmoles of dimethylaluminum
chloride(0.92g) were weighed directly into the volumetric flask (A) in the dry box. The
mixture was protected from air using a septum and cooled to -78 degrees. Argon was
introduced through the sample inlet (D). The condenser (B) and the gas burette (C) were
assembled and the entire apparatus was purged with argon for 20 minutes. 10 mmoles of
tetramethylpipridine (1.68 mL) was introduced through the sample inlet at -78C and the
cooling bath was removed. The apparatus was then allowed to come to room
temperature. The reaction vessel was immersed in an oil bath at 110C and gas burette
was zeroed. Slow evolution of methane started. After 12 hours of heating gas evolution
ceased (238 mL). Stock solution was made by adding toluene to the mark. An aliquot
(100 id.) was removed from the vessel through the sample inlet, toluene was evaporated
and remaining clear syrupy liquid was dissolved in deutero benzene.
lH NMR(300MHz;CDCl3) : 6 -O.40(s), 1.44(s), 156(5), 1.9-1.2(br m)

Manon OfJL-I‘IMDSLIl-FIUOI'O, Tetramethyldialuminum Compound VII

Dimethylaluminum fluoride was prepared in situ according to the procedure by
Laubengayer et al.(4) Hence the volumetric flask (A) in Figure 11 was charged with 10
mmoles (0.42g) of sodium fluoride and 10 mmoles (0.92g) of dimethylaluminum chloride
inside a dry box. Dry toluene (3 mL) was added and the apparatus in Figure 11 was
assembled under a stream of argon. This suspension was heated for 16 hours at reflux
after which a fine white precipitate was present. The 1H NMR analysis of a lOuL sample
after evaporation of toluene under vaccuo and resuspension in d6 benzene showed a
singlet at -O.56 ppm.(syrupy clear liquid). The vessel was then charged with 10 mmoles
of a 2M trimethylaluminum solution in toluene (5 mL) at room temperature. The vessel

was immersed in an oil bath at 110C, allowed to equilibrate, and gas burette was zeroed.

93
10 mmoles of HMDS (2.1 mL) was added. Rapid evolution of methane started and

seased afte 1.5 hours. Near quantitative amounts of methane evolved
1H NMR(300MHz;CDCl3) : a -0560, 1H), 0.360, 1.7H),
Preparation of u-Diisopropvla_rr_Iido.u-Fluoro. Tetramethyldialuminum Commund VIII
Dimethylaluminum fluoride was prepared in situ according to the procedure by
Laubengayer et al.(4) Hence the volumetric flask (A) in Figure 11 was charged with 10
mmoles (0.42g) of sodium fluoride and 10 mmoles (0.92g) of dimethylaluminum chloride
inside a dry box. Dry toluene (3 mL) was added and the apparatus in Figure 11 was
assembled under a stream of argon. This suspension was heated for 16 hours at reflux
after which a fine white precipitate was present. The lH-NMR analysis of a IOuL sample
after evaporation of toluene under vaccuo and resuspension in d6 benzene showed a
singlet at -0.56 ppm.(syrupy clear liquid). The vessel was then charged with 10 moles
of a 2M trimethylaluminum solution in toluene (5 mL) at room temperature. The vessel
was immersed in an oil bath at 110C, allowed to equilibrate, and gas burette was zeroed.
10 mmoles of diisopropylamine (1.4 mL)was added. Slow evolution of methane started
and seased after 3.5 hours. Near quantitative amounts of methane evolved.
1H NMR(300MHz;CDCl3) : 6 -O.62(br s, 6.8H), 1.3 1(d, J=9.0 Hz, 5.8H), 3.57(m, 129.0
Hz, 1H).
Preparation of u-Diethvlamido. u-FluoroLTetramethvldialuminumComeund XI
Dimethylaluminum fluoride was prepared in situ according to the procedure by
Laubengayer et al.(4) The volumetric flask (A) in Figure 11 was charged with 10
mmoles (0.42g) of sodium fluoride and 10 mmoles (0.92g) of dimethylaluminum chloride
inside a dry box. Dry toluene (3 mL) was added and the apparatus in Figure 11 was
assembled under a stream of argon. This suspension was heated for 16 hours at reflux
after which a fine white precipitate was present. The 1H NMR analysis of a 10 yL
sample after evaporation of toluene under vaccuo and resuspension in d6 benzene showed

a singlet at -O.56 ppm.(syrupy clear liquid). The vessel was then charged with 10 moles

94
of a 2M trimethylaluminum solution in toluene (5 mL) at room temperature. The vessel

was immersed in an oil bath at 110C, allowed to equilibrate, and gas burette was zeroed.
10 mmoles of diethylamine (1.0 mL) was added. Slow evolution of methane started and
seased after 3.5 hours. Near quantitative amounts of methane evolved.
lH NMR(300MHz;CDCl3) : 6 -O.71(br s, 2.4H), 1.04(t, J=7.02 Hz, 1H), 2.86(q, J=7.02
Hz, 1H).
Preparation of u—Tetramethylpipridido, u-Fluoro, Tetramethvldialuminum Compound
_X_II

Dimethylaluminum fluoride was prepared in situ according to the procedure by
Laubengayer et al.(4) Hence the volumetric flask (A) in Figure 11 was charged with 10
mmoles (0.42g) of sodium fluoride and 10 mmoles (0.92g) of dimethylaluminum chloride
inside a dry box. Dry toluene (3 mL) was added and the apparatus in Figure 11 was
assembled under a stream of argon. This suspension was heated for 16 hours at reflux
after which a fine white precipitate was present. The lH-NMR analysis of a 10 pL
sample after evaporation of toluene under vaccuo and resuspension in d6 benzene showed
a singlet at -0.56 ppm.(syrupy clear liquid). The vessel was then charged with 10 moles
of a 2M trimethylaluminum solution in toluene (5 mL) at room temperature. The vessel
was immersed in an oil bath at 110C, allowed to equilibrate, and gas burette was zeroed.
10 mmoles of tetramethylpipridine (1.7 mL) was added. Slow evolution of methane

started and ceased after 8 hours. Near quantitative amounts of methane evolved.

1H NMR(300MHz;CDCl3) : a -O.7l(br s), 1.37(br s), 1.42(br s), 1.6-2.0(br m).

 

 

 

B—->

 

 

 

 

 

+—D

 

 

A: lOmL volumetric flask
B= condenser

C= gas burette

D= 8mm sample inlet

E: drying tube

F: reservoir

 

95

 

 

 

 

 

250

 

 

F

 

 

/"!'l

 

 

 

 

 

figure 11 Apparatus used in the synthais ofaluminum amides

 

 

“8m 12- Sample removal from flak A

Literature Cited

1) (a) Maruoka, K.; Nozaki, H.; Oshima, K.; Osawa, T.; Osawa, S. Chem. Lett. 1979 ,
379. (b)Jeffery, E.A.; Meisters,A.; Mole, T. J. Organomet. Chem. 1974, 74, 365.

2) (a) Itoh, A.; Ozawa, S.; Oshima, K.; Nozaki, H. Tetrahedron Lett. 1980, 2], 361. (b)
Itoh, A.; Ozawa, S.; Oshima, K.; Nozaki, H. Bull. Chem. Soc. Jpn. 1981, 54, 274. Sasaki,
K.; Aso, Y.; Otsubo, T; Ogura, F. Chem. Lett. 1989 , 607.

3) (a) Tsuda, T.; Hayashi, T.; Satomi, H.; Kawamoto, T.; Saegusa, T. J. Org. Chem.
1986, 51, 537. (b) Tsuda, T.; Satomi, H.; Hayashi, T.; Saegusa, T. J. Org. Chem. 1987 ,
52, 439. (c) Daniewski, A.R.; Keigiel, Syn. Commun. 1988, 18, 115.

4) Taniguchi, M.; Hino, T.; Kishi, Y. Tetrahedron Lett. 1986, 27, 4767.

5) Zakharkin, L.I.; Savina, LA. Bull. Acad. Sci. Ussr. Engl. Trans]. 1961, 345.

6) Maruoka, K; Hashimoto,S.; Kitagawa, Y.; Yamamoto, H.; Nozaki,H. J. Am. Chem.
Soc. 1977, 99, 7705. Maruoka, K; Hashimoto,S.; Kitagawa, Y.; Yamamoto, H.;
Nozaki,H. Bull. Chem. Soc. Jpn. 1980, 53, 3301.

7) Kuroboshi, M.; Okada, Y.; lshihara, T.; Ando, T. Tetrahedron Lett. 1987 , 28, 3501.
8) (a) Jeffery, E.A.; Meisters, A. J. Organomet. Chem. 1974, 83, 307. (b) Etras, M.;
Seebach, D. Helv. Chim. Acta. 1985, 68, 961.

9) Tsuji, J.; Yamada, T.; Kaito, M. ; Mandi, T. Tetrahedron Lett. 1979, No.24, 2257.
Tsuji, J.; Yamada, T.; Kaito, M. ; Mandi, T. Bull. Chem. Soc. Jpn. 1980, 53, 1417.

10) Baukov, Y.I.; Lutsenko, I.F. Organometal, Chem. Rev. Sect.A. 1970, 6, 335.

11) Jeffery, E.A.; Meisters, A. J. Organomet. Chem. 1974, 82, 315.

12) Nozaki, H; Oshima, K.;Takai, K; Ozawa,S. Chem. Lett. 1979, 379. .Nozaki, has
reported enolate formation with a mixture of lithium tetramethyl pipridide (LTMP) and
diethylaluminum chloride. We examined this reaction and concluded that LTMP is the
actual base and not an aluminum amide.

13) (a) Lappert, M.F.; Power, P.P.; Sanger, A.R.; Skivastava, R.C.; Metal and Metalloid
Amides; John Wiley and Sons: New york, 1980. Cucinella, S.lnorg. Chim. Acta,Rev.
1970, 4, 51.(b) Ashby, H.C.; Lin, J.J. Tetrahedron Lett. 1976, 3865. Giongo, G.M.; Di
Gregorio, F.; Palladino, N.; Marconi, W. Tetrahedron Lett. 1973, 3195.

14) Mazzei, A.; Cucinella, S.;Marconi, W. Inorg. Chim. Acta, 1968, 2, 305. Mazzei, A.;
Cucinella, S.;Marconi, W. Makromol. Chem. 1969, 122, 168.

15) Corey, E.J.; Imwinkelreied, R.; Pikul, S.; Xiang, Y.B. J. Am. Chem. Soc. 1989, I I I ,
5493.

98
16) Basha, A.; Lipton, M.; Winreb, S. Tetrahedron Lett. 1979, 4174. Sidler, D.R.;
Lovelace, T.C.; Mc Namara, J.M.; Reider, P]. J. Org. Chem. 1994, 59, 1231.
17) Veith, M. Chem. Rev. 1990, 90, 3.
l8) Sauls, P.C.; Interrante, L.V. Coordination Chem. Rev. 1993, 128, 193.
19) Yamamoto, H.; Nozaki, H. Angew. Chem. Int. Ed. Eng. 1978, I 7, 169
20) Maruoka, K.; Oishi, M.; Yamamoto, H. J. Org. Chem. 1993, 58, 7638.
21) Pinkas, J.; Wang, T.; Jacobson, R.A.; Verkade, J .G. Inorg. Chem. 1994, 33, 4202.
22) Gosling, K.; Mc Laughlin, G.M.; Sim, G.A.; Smith, J.D. Chem. Comm. 1970, 1617.
23) Hess, H.; Hindere,A.; Steinhauser, S. Z.Anorg. Chem. 1970, 377, 1.
24) Davidson, N.; Brown, H.C. J. Am. Chem. Soc. 1942, 64, 316.
25) Ashby, E.C.; Kovar, R.A. J. Organomet. Chem. 1970, 22, C34.
26) Ashby, E.C.; Kovar, R.A. Inorg. Chem. 1971, 10, 893.
27) Ruff, J.I(.; Hawthorne, M.F. J. Am. Chem. Soc. 1960, 82, 2141. ibid.l961, 83, 535.
28) Ruff, J.I(. J. Am. Chem. Soc. 1961, 83, 2835.
29) Neumann, W.P. Ann. 1963, 667, 1.
30) Thomas, C.J.; Krannich, L.l(.; Watkins, C.L. Polyhedron, 1993, 12, 389.
31) Lipton, M.F.; Basha, A.; Weinreb, S. Org. Synth. 1979, 59, 49.
32) Galiano-Roth, A.S.; Kim, Y.J.; Gilchrist, J .H.; Harrison, A.T.; Fuller, D.J.; Collum,
D.B. J. Am. Chem. Soc. 1991, 113, 5053.
33) Rie, J.E.; Oliver, J.P. J. Organomet. Chem. 1977, 133, 147.
34) Laubengayer, A.W.; Lengnick, G.F. Inorg. Chem. 1966, 5, 503.

 

HICHIGRN STQTE UNIV. LIBRRRIES
lllllllllllIIHIIHIIlllllllllllllllllllllllllllllllllllllll
31293014009017