IflES‘S * 4 A\

‘ LIBRARY
Michigan State

~31

  

 

 

 

This is to certify that the

dissertation entitled
I. ACYLATION OF TRIMETHYLSILYL ACETATE. A SYNTHETIC

ROUTE TO B-KETO ACIDS AND METHYL KETONES. .
II. ACYLATION OF CARBON ACIDS UNDER ESSENTIALLY |
NEUTRAL CONDITIONS. !
III. AN INTRODUCTORY STUDY OF THE CARBOMETHOXYLA-
TION OF KETONES USING WEAK BASES.

presented by

Patrick J. Cowan

 

has been accepted towards fulfillment
of the requirements for

Ph . D . degree in Chemistry

mm Queuw

Major professor

Date July 20. 1983

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

 

 

MSU
LIBRARIES
.—:—.

 

 

 

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

 

 

 

 

WW” I!“ am is;

 

 

 

CHAPTER I

ACYLATION OF TRIMETHYLSILYL ACETATE.
A SYNTHETIC ROUTE TO B'KETO ACIDS
AND METHYL KETONES.

CHAPTER II

ACYLATION OF CARBON ACIDS UNDER
ESSENTIALLY NEUTRAL CONDITIONS.

/§//-- P3?7

CHAPTER III

AN INTRODUCTORY STUDY OF THE CARBOMETHOXYLATION
OF KETONES USING WEAK BASES.

BY

Patrick J. Cowan

A DISSERTATION

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

DOCTOR OF PHILOSOPHY

Department of Chemistry

1983

ABSTRACT

CHAPTER I

ACYLATION OF TRIMETHYLSILYL ACETATE.
A SYNTHETIC ROUTE TO B-KETO ACIDS
AND METHYL KETONES.

CHAPTER II

ACYLATION OF CARBON ACIDS UNDER
ESSENTIALLY NEUTRAL CONDITIONS.

CHAPTER III

AN INTRODUCTORY STUDY OF THE CARBOMETHOXYLATION
OF KETONES USING WEAK BASES.

BY

Patrick J. Cowan

l. Trimethylsilyl acetate was acylated with a variety of
acid chlorides to give, after solvolysis and subsequent de-
carboxylation, excellent yields of methyl ketones. The
B-keto acids 3-oxo-3-phenylpropanoic acid and 4,4-dimethyl-
3-oxopentanoic acid were prepared in 97% and 81% yield,
respectively, by acylation of trimethylsilyl acetate with
benzoyl chloride and pivaloyl chloride. Acetylation of
trimethylsilyl butanoate gave, after hydrolysis and decar-

boxylation, 2—pentanone in 47% yield. Trimethylsilyl

2-methylpropanoate failed to react with acetyl chloride.

_l. Magnesium chloride and triethylamine were used to pro-
mote the acylation of diethyl malonate, ethyl acetoacetate,
and acetylacetone. Excellent yields of C-acylated products,
commonly known as triacylmethanes, were obtained in most
cases. Acylation of acetylacetone with isobutyryl chloride
gave a mixture of triacylmethanes resulting from transacyla-

tion reactions. Substituting collidine for triethylamine

was found to suppress the transacylation reactions.

ill. The carbomethoxylation of cyclohexanone and aceto-
phenone was accomplished with carbomethoxy imidazole in the
presence of triethylamine, magnesium chloride, and sodium
iodide to give 2-carbomethoxycyclohexanone and methyl ben-
zoylacetate, respectively, each in 64% yield. Carbomethoxyl-
ation of diethyl ketone gave a mixture of three B-dicarbonyl
compounds which were identified as dimethyl methylmalonate
(%7), methyl-3-oxo-2-methy1pentanoate, and 4-methyl-3,5-hep-
tanedione (%§). Compounds 1g and I] are thought to arise

from bis carbomethoxylation of diethyl ketone.

To my family and NO.

ii

ACKNOWLEDGMENTS

I wish to express my deep appreciation to Dr. Michael W.
Rathke for his guidance, encouragement, and, most importantly,
his friendship throughout the course of my graduate career.

I wish to thank my family, colleagues, and friends who
made my stay at MSU an enjoyable one. I especially wish to
thank Nancy, not only for typing and editing this disserta-
tion, but also for her companionship. "Thank you, Speedy, may
the end of this project mark the beginning of a long and
beautiful relationship.

Finally, I would like to thank the Dow Chemical Company,
the National Science Foundation, and Michigan State Univer-

sity for my financial support.

iii

TABLE OF CONTENTS

List of Tables.. ............ ..... ...................... vi
List of Figures............................... ..... .... vii

Chapter I - ACYLATION OF TRIMETHYLSILYL ACETATE.
A SYNTHETIC ROUTE TO B-KETO ACIDS

AND METHYL KETONES.................. ....... 1
Introduction...................................... 1
Results and Discussion........... ............ ..... 6
Experimental ...... ................... ...... ....... 9

Materials.................................... 9
Methods of Analysis......... ...... ........... 9

General Procedure for the Acylation
of Trimethylsilyl Esters..................... 10

General Procedure for the Preparation
of Methyl Ketones.................. ........ .. 10

General Procedure for the Isolation
of B-Keto Acids................... ........ ... 11

Chapter II - ACYLATION OF CARBON ACIDS UNDER
ESSENTIALLY NEUTRAL CONDITIONS............ 12

Introduction...................................... 12
Results and Discussion ...... .. ...... .............. 19
Experimental...................................... 35
Materials....... ............ .. .............. . 35
Methods of Analysis...... ...... .............. 36

General Acylation Procedure Used to
Survey Various Lewis Acids................... 36

iv

General Procedure for the Acylation
of Diethyl Malonate....... ...................

1H NMR Study of the
Diethyl Malonate/Magnesium Chloride System...

Isolation of the Magnesium Enolate of
Diethyl Malonate ...... . .......... .. ..........

General Procedure for the Acylation
of Ethyl Acetoacetate.................. ......

General Procedure for the Acylation of
Acetylacetone.......... ....... ...... .........

GC Study of the Acylation of
Acetylacetone with Isobutyryl Chloride.. .....

Acylation of Acetylacetone with
Isobutyryl Imidazole...... ...................

Chapter III - AN INTRODUCTORY STUDY OF THE
CARBOMETHOXYLATION OF KETONES

USING WEAK BASES ........ . . . . .............

Introduction ......................................

ReSUlts and DiscuSSionOOOOOOOO......OOOOOOOOOOOOOO

Experimental. ....... . .............................
Materials ....... ........ ...... ...... .........
Methods of Analysis ...... ....... .............

General Procedure for the Carbomethoxylation
of Cyclohexanone.......... ..... . .............

Carbomethoxylation of Acetophenone...........

Carbomethoxylation of Diethyl Ketone .........

ReferenceSOOOOOOO......OOOOOOOOOOOOOO ......... O. .......

37

38

39

40

41

43

43

45
45
48
54
54

55

55

56
57

58

LI ST OF TABLES

Table Page
1 Acylation of Trimethylsilyl Acetate ...... .... 8
2 Survey of Metal Catalysts in the

Acylation of Diethyl Malonate ................ 20
3 Acylation of Diethyl Malonate.. .............. 21
4 Acylation of Ethyl Acetoacetate .............. 24
5 Acylation of Acetyl Acetone............. ..... 27
6 Reaction of Acetylacetone with Isobutyryl

Chloride, a Solvent Study...... ........ . ..... 30
7 Survey of Bases in the Acylation of

Acetylacetone with Isobutyryl Chloride ....... 32
8 Carbomethoxylation of Cyclohexanone..... ..... 51

vi

Figure

LIST OF FIGURES

Page
Metal Complexation to Carbon Acids ....... .... l3
Acylation of Magnesium Salts of
Malonate Half-Esters............. ..... ....... 18
Transacylation of Acetylacetone..... ....... .. 29

Carbomethoxylation of Diethyl Ketone......... 53

vii

Chapter I

ACYLATION OF TRIMETHYLSILYL ACETATE.
A SYNTHETIC ROUTE TO B-KETO ACIDS
AND METHYL KETONES.

Introduction

 

B-Keto acids are important intermediates for the prepara-

1,2

tion of ketones and for the synthesis of a variety of

natural products3’4. These compounds have also been used

extensively for the studies of the mechanism of decarboxyl-

ation5’6.

B-Keto acids have in general been synthesized by three

7,8

methods: acid or base9 catalyzed hydrolysis of the cor-

responding B-keto esters, carboxylation of the appropriate

enolate anionslo-12

2,13

, and acylation of dianions of carboxylic
acids

Although alkaline hydrolysis of B-keto esters has been
used successfully to prepare several of the corresponding

15,16

aromatic14 and aliphatic B-keto acids (eq 1), it has

proven to be unreliable as a general methodB'lo. Alkaline
hydrolysis is often complicated by competing attack of the
base at the ketone function, leading to retro Claisen

cleavage of the B-keto ester8'10

(eq 2).
Acid hydrolysis of methyl esters (eq 3) has been effec-

tive in the preparation of several long chain B-keto acidsB.

1

 

 

 

RCOCH2C02R1 OH >, Rcocuzcoz’ ——5—7> Rcocuzcozn (1)
H H ' fi fi
RCCRZCOR1 OH >= RCOH + -CR2COR1 (2)
RCOCH co CF HCl a4> RCOCH co H (3)
2 2 ‘3 HOAc '/ ‘ 2 2

Recently Logue10 has described the conversion of tert-butyl-

B-benzoylisobutyrates to the corresponding B-keto acids by

10

treatment with trifluoroacetic acid (eq 4). Acid catalyzed

thermal decomposition of the desired B-keto acid is often a

serious problem in the acid catalyzed hydrolysis of B-keto

5,6

esters (eq 5).

 

o
0"” CF3‘302H \ O OH
x

b

(4)
X:
H. CH3, OCH3, c1
1 H+ H+
RCOCHZCOZR ———+> Rcocnzcozn ———€> RCOCH3 + co2 (5)

3
The reaction of enolate anions with carbon dioxide also
leads to B-keto acids. Both aliphatic and aromatic B-keto
acids have been prepared by deprotonation of the appropriate

ketones with a suitable base in inert polar solvents, fol-
10-12

 

lowed by reaction with carbon dioxide (eq 6).
base - 1) CO2
RCOCH ————> RCOCH 9 RCOCH co H (6)
3 2 2) H+ 2 2

Stiles and Finkbeiner10 have demonstrated that methyl
magnesium carbonate (MMC) in dimethylformamide (DMF) solution
is capable of carboxylating ketones which have alpha hydro—

gens (eq 7). Although MMC gives good yields of B-keto acids

 

ii o’w‘o
1 CHgoMgOC OCH 11+ 0 0
RCOCH R J) O 4—9 (7)
2 R H OH
DMF , 1 2 0 °C 3' "1

in many cases, optimum yields require large excesses (5-20
fold) of the reagent.

18,19

Separate studies by Sakurai17 and Matsumura have

shown that complexes 1, 2, and 3 act effectively as carbon
dioxide carriers in the carboxylation of active methylene
compounds under mild conditions (eq 8). These procedures
give moderate to fair yields (44-65%) of B-keto acids.

Acylation of dianions of carboxylic acids has also been

. [:1
arMeO-‘vcfl [Congar Néucozuger "Ha/\T

(CI-I2)" 002M98r
2 , 3 26 L3

n:

1.

HCO H (8)

 

shown to be a viable route to B-keto acids. In 1971,
Ainsworth and Kuo2 reported a route to B-keto acids in which
intermediates formed in the reaction of carboxylic acid di-
anions with esters were trapped using trimethylchlorosilane
(TMCS). The resulting trimethylsilyl esters were solvolyzed

under neutral conditions to give good yields of B-keto acids

(eq 9).

chcoz' + R1C02CH3 ———> RICOCRZCOZ- L4C§—>

(9)

1 CH3OH 1
R COCRZCOZSi(CH3)3 ——-—> R COCRZCOZH

5
Recently van der Baan and coworkers20 have described the
preparation of B-keto acids in good yields by acylation and
subsequent hydrolysis and decarboxylation of the mono anion

of bis(trimethylsilyl) malonate (3, eq 10).

_ - H o
2 (Me3Si02C)2CH + RCOCl ——> (Me3Si02C)2C—COR —0‘3C——>

10

4 -C02 ( )

\
(HOZC) 2CHCOR / RCOCHZCOZH

 

Previous work in this laboratory has demonstrated that
enolates of monocarboxylic esters react with acid chlorides
to yield B-keto ester521. In fact, the lithium enolate of
trimethylsilyl acetate is a synthetic equivalent of enolate
%. Consequently, a less expensive and more direct route
towards the synthesis of B-keto acids would be the direct

acylation of a lithium enolate of a trimethylsilyl ester

followed by solvolysis to the desired compound (eq 11).

OLi (11)
| . 1
RCH=COSiMe3 + R coc1 ——9 R12COCIIHCO SiMe 3——> RJ‘C0:11CO2 H
R
Thus, we have studied the synthetic utility of the acylation

of trimethylsilyl esters for the formation of B-keto acids.

Results and Discussion

 

The lithium enolate of trimethylsilyl acetate [prepared
by reaction of trimethylsilyl acetate with two equivalents
of lithium diisopropylamide (LDA)] was reacted with one
equivalent of benzoyl chloride. Two equivalents of LDA
were necessary as the second equivalent is consumed by the

acidic intermediate, B-keto siloxy ester é, (eq 12). Due

 

M LIO o (12)
CH co SiMe 1) LDA \ osm ——)LDA “My".

3 2 3 2) PhCOCl/ Ph .3
E

to the susceptibility of B-keto acids to decarboxylation, it
was decided not to analyze the reaction by isolation of the
B-keto acid obtained by solvolysis of g. Instead, the re-
action yield was determined by solvolysis and subsequent
decarboxylation of 3! followed by GC analysis for the cor—

responding methyl ketone (eq 13). Following this procedure,

LIO O H+ O 0 -C02 0 (13)
PhMSIMoa —> ,. MMOH " 9 era/u\

R

acetophenone was obtained in 98% yield (GC).
To test the general applicability of the method, the
lithium enolate of trimethylsilyl acetate was reacted with a

variety of acid chlorides. For optimum yields, most cases

7
required use of 1.5 equivalents of trimethylsilyl acetate.
As can be seen in Table 1, the majority of the acid chlorides
tested gave the corresponding methyl ketone in approximately
90% yield.

To test the applicability of the method to the prepara-
tion of B-keto acids, we prepared and isolated 3-oxo-3-phenyl-
propanoic acid and 4,4-dimethyl-3-ox0pentanoic acid in 97%
and 81% yield, respectively. The lower yield obtained with
4,4-dimethyl-3-oxopentanoic acid was probably due to partial
decarboxylation of the sensitive B-keto acid.

We next attempted to extend this methodology to other
trimethylsilyl esters. Trimethylsilyl butanoate was reacted
with acetyl chloride to give, after hydrolysis and decarboxy-

lation, 2-pentanone in 47% yield (GC, eq 14). However, tri-

o (14)

2W CH 3COCl -CO2
|M93——> SIM03——>

methylsilyl 2-methylpropanoate failed to react with acetyl
chloride to give the expected 3-methyl-2-butanone (eq 15).
These results suggest that substitution at the alpha position
of trimethylsilyl acetate sterically hinders C-acylation of
the corresponding enolate. As the nucleophillic carbon of the
trimethylsilyl ester enolate becomes more hindered, O-acyla-

tion probably becomes the predominant reaction pathway.

Table 1. Acylation of Trimethylsilyl Acetate.

 

 

 

 

LDA (20 mmol) l) RCOCl (10 mmol)
CH3COZSi(CH3)3 > >1 RCOCH3
THF, -78°C 2) 1130+, A
19%}. M322m313_ Yield m a
C6H5C0C1 10 98
15 76
20 26
o-CH3C6H4COC1 10 70
15 88
(CH3)3CCOC1 15 94
20 6O
(CH3)2CHCOC1 15 99
n-C3H7COC1 10 49
15 77

 

a) Yields determined by GC, using an internal standard.

 

CH3COC1 O
I I x
2 SIM93 I I 7 [M03 9 fi/ (1 5)

In conclusion, this procedure offers a useful, high-
yield route to B-keto acids and methyl ketones by acylation
of trimethylsilyl acetate. This method, however, is not
directly applicable to the acylation of trimethylsilyl
esters other than trimethylsilyl acetate (with acid chlo-
rides). Perhaps the use of acylating agents less reactive

than acid chlorides would solve this problem22.

Experimental

 

Materials

 

Diis0propylamine was distilled from CaH2 prior to use.
THF was distilled from the sodium ketyl of benzophenone
prior to use. Trimethylsilyl esters were prepared from the
corresponding carboxylic acid and bis(trimethylsilyl) aceta-
mide by the method described by K1ebe23. Acid chlorides
(excluding o-CH3C6H4COC1)24 were obtained from Aldrich

Chemical Co. and distilled prior to use.

Methods of Analysis
1

 

H NMR data were obtained on a Varian T-60 spectrometer

at 60 MHz. Chemical shifts are reported in parts per million

10
on the delta scale relative to TMS internal standard. Gas
chromatographic analysis were performed with a Varian 920
chromatograph equipped with a 6 ft. by 0.25 in. stainless
steel column packed with Carbowax 20 M terephthalic acid on

acid washed Chromosorb P.

General Procedure for the Acylation of Trimethylsilyl Esters

 

A flame-dried 50 mL flask equipped with septum inlet,
magnetic stirrer, and mercury bubbler was flushed with argon
and immersed in an ice bath. 12.5 mL (20 mmol) of 1.6 M
M-butyllithium in hexane was injected followed by dropwise
addition of 2.82 mL (20 mmol) of diisopropylamine. The
solvent was removed under reduced pressure, leaving lithium
diisopropylamide as a white solid. The base was dissolved
in 10 mL THF and cooled to -78°C. The appropriate amount
of trimethylsilyl ester (see Table 1, 20 mmol for trimethyl-
silyl butanoate and trimethylsilyl 2-methylpropanoate) was
added dropwise and the resulting solution was stirred for
15 minutes at -78°C. After addition of 10 mmol of acid
chloride, the reaction was warmed to room temperature and
stirred for 15 minutes. The solvent was removed under re-

duced pressure leaving a slightly yellow solid.

General Procedure for the Preparation of Methyl Ketones
The solid obtained above was dissolved in 10 mL CH2C12.
3 mL 6 M HCl was added and the resulting mixture was refluxed

for 2 hours. The mixture was extracted with three 5 mL por-

tions of CHZClz, the combined organic layers were dried over

11

M980 and analyzed by GC (dodecane as an internal standard)

4
for methyl ketone.

General Procedure for the Isolation of B-Keto Acids
The slightly yellow solid isolated above was dissolved

in 10 mL of CH2C12 and cooled to 0°C. 6 M HCl was slowly

added to bring the pH of the solution to approximately 1.
The mixture was extracted with three 5 mL portions of
CH2C12, the combined organic layers were dried (M9804),
filtered, and the solvent was removed under reduced pressure
to give the crude B-keto acid. The crude product was dis-
solved in diethyl ether and washed with a small quantity of
water to remove minor amounts of triethylamine hydrochloride.
The organic layer was dried (MgSO4), and the solvent was
removed under reduced pressure leaving the B-keto acid as
a colorless solid. The product was recrystallized from

CHClB/pentane.

3-oxo-3-phenylpr0panoic acid: Yield 1.57g (97%); m.p.

1

98-99°C (lit. 101-102°C)25; H NMR (coc1 a 4.12 singlet

3)!
(PhCOCHZCOZH), 5.72 singlet (-C=CMC02H), 7.25-8.05 multiplet

(C6H5-CO). Ketone/enol ratio = 3:1.

4,4-dimethy1-3-oxopentanoic acid: Yield 1.169 (81%);

1

m.p. 43-45°C (lit. 47-49°C)26; H NMR (CDCl 61.17 singlet

3):
[(CH3)3CCO], 3.60 singlet (-COCH COZH), 5.07 singlet

(-C=CM-C02H). Ketone/enol ratio = 14:1.

Chapter II

ACYLATION OF CARBON ACIDS UNDER
ESSENTIALLY NEUTRAL CONDITIONS

Introduction

 

A standard procedure for the C-acylation of carbon
acids, 1, is the formation of the conjugate base, 8, fol-
lowed by reaction with an acid chloride (or anhydride) in
a second, separate step (eq 16). A problem inherent in
this approach is that the acylation product, 9, is always a

stronger acid than 1 and thus may neutralize a portion of
21

 

 

anion 8 in a subsequent acid-base reaction (eq 17).
‘ Rlcoc1 1
122CH2 + base a RZCH > RZCHCOR (15.)
Z R 2
Q + g > Z + RZCCOR1 (17)

19

One solution to this problem is to effect a single-step
reaction of carbon acid 1 with an acid chloride in the pres-
ence of two equivalents of base, producing product enolate
10 stoichiometrically (eq 18). A critical requirement for
the success of this procedure is that the base must not
consume the acid chloride during the reaction. This re-
striction appears to have limited the application of the

12

13

Z + RICOCl base 9 2 base a 10 (18)

single-step procedure to those very strong carbon acids
(e.g. Meldrum's acid)27 whose anions can be generated almost
quantitatively by the weak but relatively tolerant tertiary
amine bases.

It is well known that complexation of a metal ion to a
ligand can enhance the acidity of that ligand28. According-
ly, we considered that one way to extend the single-step
acylation procedure to weaker carbon acids would be to use
metal complexation to enhance the acidity of carbon acids to

the point where useful concentrations of enolates could be

generated with tertiary amine bases (Figure 1). Clearly,

Figure 1. Metal Complexation to Carbon Acids.

 

 

O M... \ 9m) l I M+
I ll / I l
.__?___c_— ‘ _.?__. __
H H
' flu) I M... O-‘M +
- - '—- + :: + -
c c nan e—i >1: \ nan H

14

only the last step in Figure 1 is a proton exchange reaction
and the overall driving force of the reaction depends on the
stability of the metal-oxygen bond in enolate 11. Thus, our
objective was to find metal ions for which I} would be of
sufficient stability to be formed in useful concentrations
by weak bases but of sufficient reactivity to be attacked by
a variety of acylating agents.

We chose to use B-dicarbonyl compounds as the carbon
acids for our initial study of the use of metal chelation in
a one-step acylation procedure using weak bases. The pro-
ducts of such a reaction, commonly known as triacylmethanes

(L2, eq 19), have been found to be useful intermediates in

 

/jL\/ji\.1 base, M+
a 1 (19)

n 2
R coc1

\l/

the preparation of dyes, insecticides, and auxiliary agents

for lacquers and leatherszg. Triacylmethanes have also been
used as intermediates in the synthesis of B-keto acidsg'20
30-35

(eq 10) and B-keto esters

methane (L2, R=R1=R2

36

(eqs 20 and 21). Triacetyl-
=CH3) has been used as a heat stabilizer

and also as a food preservative37.

in plastics
Triacylmethanes are generally synthesized by reaction of

a preformed metal complex of a B-dicarbonyl compound with an

15

 

 

O 0 H30 0 o
1 \
n on _C02 7 n/‘K/‘Kon1 (20)
o
suufi
o o HO+ o o
R 8 -co 7 on (21)
o t 2
Bu

38-47

acylating agent (eq 22) For example, diethyl acetyl-

"I.
0 ’0

1
\
R/A§>/JK\R + R COCX i7

M = Mg, Na, Cu

 

(22)

 

malonate (L3) has been synthesized in 54% yield by treatment
of the magnesium enolate of diethyl malonate with the mixed

anhydride l4 (eq 23)38. Viscontini39

has described the syn-
thesis of ethyl-3—oxo-2-acety1butanoate in 73% yield by a
similar procedure (eq 24). Triacetylmethane has been syn-
thesized from acetylacetone and acetyl chloride (eq 25)40.
The formation of enolates of B-dicarbonyl compounds
using weak bases (e.g. tertiary amines) in the presence of

metal complexing agents should be relatively simple, not only

16

 

 

 

 

iii? (23’
Mg(OEt)2 \ CH3COCOEt (13)
CH2(C02Et)2 :7. EtOMgCH(C02Et)2 .>
CH3COCH(C02Et)2
L3
1) Mg(OEt)2
CH3COCH2C02Et >= (CH3CO)2CHC02Et (24)
2) CH3COC1
73%
(CH3CO)2CH2 1) NaH j} (CH3CO)3CH (25)
2) CH3COC1
18-30%

because of the high intrinsic acidity of the B-dicarbonyl
compounds, but also because their chelating nature should

favor the formation of a metal complex (15, eq 26). A

 

M
./ %
I
R R3N ( )
15

'b

variety of metal complexes, 15, have been known for more

than a century. Unfortunately, the same factors which make

17

15 easy to prepare also render it relatively inert in syn-
thetically useful reactions with electrOphiles. For example,
most transition metal complexes (15, M=transition metal) do
not react with electrophiles such as acid chlorides and alde-
hydes. On the other hand, the more reactive lithium and
sodium complexes (15, M=Li,Na) must usually be prepared with
relatively strong bases, such as the corresponding metal
alkoxides. The greatest chance for success would appear to
be with magnesium or zinc complexes (1?, M=Mg,Zn) where the
complexes are known to possess sufficient reactivity for re-
action with a variety of electrophiles28 and where the metal
ions have appreciable complexing power for oxygen ligands,
favoring the formation of complexes (15, M=Mg,Zn) in the
presence of weak bases. Magnesium enolates are also known
to have a high tendency to acylate at carbon rather than at
oxygen48.

Results reported by Masamune49 and Kobuke50 for the a-

cylation of magnesium salts of malonate half-esters provide

examples of these concepts (eq 27). Compound 16 (Figure 2)

O O C)
H
RC-Im + Mg (OZCCHZCOZRI) 2 ——9 “M031 (27)

Im = ._€:::?

18

Figure 2. Acylation of Magnesium Salts of Malonate Half-

 

 

 

Esters.
0.!“ o’w‘
0 ba 5 e \ No RCOX \
30 < 7 R /
H .

 

I
O u)“ ”9

is ideally structured for acidity enhancement by an efficient
internal metal complexation. Presumably the weak base nec~
essary for enolate formation is provided by imidazole (de-
liberately added by Kobuke and present as a by-product of
acyl imidazole formation in Masamune's reaction).

We also note that with very acidic dicarbonyl compounds,

27

such as Meldrum's acid (17, eq 28), the acylation reaction

 

O O O O
°)< + RCOX 91’ r > “To (2 8)
O O Oaf/’
17

can be conducted with weak bases in the absence of metal

ions, and that this represents an exceedingly useful

19

acylation procedure. In a sense, one of our goals was to
extend this procedure by means of metal complexation to

less acidic B-dicarbonyl compounds.

Results and Discussion

 

Initial experiments involved the reaction of diethyl
malonate with acetyl chloride in the presence of pyridine
and a variety of metal salts (MX, Table 2). Diethyl malonate
was chosen as the substrate for the initial study since it
is one of the least acidic members of the B-dicarbonyl
family of compounds and thus should provide the most strin-
gent test of this method.

In the absence of metal salt (Entry 1, Table 2), diethyl
malonate was recovered nearly quantitatively. In addition,
the reaction mixture turned black shortly after addition of
the acetyl chloride, presumably due to ketene formation.

When the reaction was repeated on a preparative scale (Entry
6, Table 3; note: triethylamine used as base), no C-acylated
product was formed in the absence of added metal salt. These
results indicate that diethyl malonate is too weak an acid

to form useful enolate concentrations in-the presence of
pyridine or triethylamine.

The reaction of diethyl malonate with acetyl chloride in
the presence of pyridine was repeated in the presence of
added metal salts (MX, Table 2). In the presence of LiCl,
CuCl ZnCl

or FeCl diethyl malonate was recovered nearly

2! 2! 3r

20

Table 2. Survey of Metal Catalysts in the Acylation of
Diethyl Malonate.

 

 

 

Mx, CH2C12
CH (CO Et) + CH COCl > CH COCH(CO Et)
2 2 2 3 3 2 2
2 pyr, 12 hr
Recovered
Entry 1425 S.M. (%)a
1 ——- 87
2 ZnCl2 93
3 CuCl2 99
4 FeCl3 63
5 LiCl 91
6 MgCl2 8
b
7 MgCl2 98

 

a) GC yields.

b) Reaction carried out in the absence of pyridine.

21

Table 3. Acylation of Diethyl Malonate.

 

 

 

Mgc12, 2 Et3N
CH2(C02Et)2 + RCOCl > RCOCH(C02Et)2
CH3CN, 12 hr
Entry RCOCl Yield (as)a

1 CH3COCl 85

2 c6Hscoc1 89

3 (CH3)2CHCOC1 92

4 (CH3)3CCOC1 9o

5 n-C3H7COC1 86

6b CH3COC1 o

 

a) Isolated yields.

b) Reaction carried out in the absence of magnesium chloride.

22

quantitatively. Also, the reaction mixture turned black

shortly after the addition of acetyl chloride. With MgClZ,

however, the reaction mixture remained colorless and only 8%

of diethyl malonate was recovered. Attempts to isolate di-

ethyl acetylmalonate (12, R=R=OEt, R2=CH3), however, gave

relatively low yields, 61%. Replacement of pyridine (pKa

E 5.0)51 with the stronger base, triethylamine (pKa E 10-0)5

increased the isolated yield of diethyl acetylmalonate to
85%. Excellent yields of C-acylated product were also ob-
tained for a variety of acid chlorides using triethylamine
in the presence of MgCl2 (Table 3). These results indicate
that the MgCl2 is in some way (presumably by chelating to
diethyl malonate) enhancing the acidity of diethyl malonate
to the point where its enolate may be generated by the weak
base triethylamine.

A 1H NMR study was conducted to verify that the enolate
of diethyl malonate was in fact formed on treatment with

triethylamine in the presence of MgClz. One equivalent of

triethylamine was added to a 0.5 M solution of diethyl
1

malonate in CD CN, and H NMR analysis revealed no change in

3
the methylene singlet of diethyl malonate (for a more com-

1H NMR, see the experimental sec-

plete description of the
tion). On addition of one equivalent of magnesium chloride
to the triethylamine-diethyl malonate system, a precipitate
formed. 1H NMR analysis showed that the methylene singlet
of diethyl malonate had disappeared and the methylene

quartet of triethylamine had shifted downfield by approxi-

mately 0.5 ppm. This observation may be attributed to

1

 

23

enolate formation with rapid proton exchange between diethyl

malonate and triethylamine52 (eq 29). Finally, enolate 18

\./' l
1‘9

1 ‘Mg
\\‘ I §
f‘ __. <3.“ \1
l H
EtOC-CH-COEt + Et3N é—— EtOC-CH=COEt + Et3N-HC1 (29)
I
H 18

was isolated in approximately 40% yield by reaction of a

0.5 M CH CN solution of diethyl malonate with triethylamine

3

and MgCl We interpret this result as an enhancement of

2.
the acidity of diethyl malonate by complexation with MgClZ.
To test the generality of this procedure, the acylation
reaction using MgCl2 was examined with ethyl acetoacetate as
the substrate. As in the case of diethyl malonate, no
C-acylated product was isolated when ethyl acetoacetate was
reacted with acetyl chloride in the absence of MgCl2 (Entry
10. Table 4). Reaction of the ethyl acetoacetate with
acetyl chloride in the presence of MgCl2 and triethylamine
in CHBCN gave only a 10% yield of the expected C-acylated
product (Entry 1, Table 4). Changing the solvent to methy-
lene chloride (CH2C12) had almost no effect on the yield of
C-acylated product. However, when the reaction was repeated
using pyridine as a base in CH3CN, the yield of C-acylated
product increased to 73% (Entry 3, Table 4). Changing the
solvent to CHZCl2 with pyridine as the base further increased
the yield to 91% (Entry 4, Table 4).

The low yield of C-acylated product obtained when

24

Table 4. Acylation of Ethyl Acetoacetate.

MgClz, base

 

 

 

CH3COCH2C02Et + RCOCl > CH3COCH(COR)C02Et
1 hr
m ggo_c1 gage: Solvent Yield (%) b
1 CH3COC1 Et3N CHBCN 10
2 CH3COC1 Et3N CH2C12 13
3 CH3COC1 pyr CHBCN 73
4 CH3COC1 pyr CH2Cl2 91
5 C6H5COC1 pyr CH2C12 81
6 (CH3)2CHCOC1 pyr CH2C12 77
7 (CH3)3CCOCl pyr CH2C12 18
ac (CH3)3CCOC1 pyr CH2C12 75
9 n-C3H7COC1 pyr CH2C12 78
10d CH3COC1 pyr CH2C12 0

 

a) 2 Equivalents of base used in all cases.
b) Isolated yields.
c) Reaction stirred for 12 hr.

d) Reaction carried out in the absence of magnesium chloride.

25

triethylamine was used as the base may be due to side re-
actions of triethylamine with acetyl chloride to form ketene.
This side reaction would be expected to be more prevalent
when using ethyl acetoacetate as the substrate rather than
diethyl malonate since the magnesium enolate of the stronger
acid, ethyl acetoacetate (pKa E 10)51, is expected to be
less nucleophillic and to react at a slower rate with acetyl
chloride than the magnesium enolate of the weaker acid, di-
ethyl malonate (pKa 2 14)51.

An alternative explanation for the low yield of C-acy-
lated product obtained with triethylamine is illustrated in

equation 30. Triacylmethanes are known to be readily

Mg;:
\ +
Et3N=JH~ ’5 P)- ; M +Et3N-HC1+ (30)

I +2
o“&M9

19

CH2=C=O

cleaved by base with cleavage normally occurring at an acetyl
function42. Thus, triethylamine may react with the C-acy-
lated product 19 to form ketene and the magnesium enolate of
ethyl acetoacetate. This cleavage would be expected to be
much slower with the weaker base, pyridine.

Under the standard conditions of 2 equivalents of pyri-
dine and 1 equivalent of MgCl2 in CHZClZ, ethyl acetoacetate

was reacted with a variety of acid chlorides. With the ex-

ception of pivaloyl chloride, all the acid chlorides

26

examined gave excellent yields of C-acylated product after
a reaction time of one hour (Table 4). Pivaloyl chloride
gave only an 18% yield of C-acylated product after one hour,
but after a 12 hour reaction time the yield increased to 75%.

Acylation of acetylacetone, using the procedure devel-
oped for ethyl acetoacetate, gave excellent yields of the
corresponding C-acylated product, provided that the acid
chloride was relatively unhindered (Table 5). In cases
where the acid chloride has no alpha protons available for
ketene formation, yields were increased using triethylamine
as the base and lengthening the reaction time from 1 to 12
hours (Entries 5 and 6, Table 5).

Acylation of acetylacetone with relatively hindered
acid chlorides (i.e. isobutyryl and pivaloyl chloride) led
to a mixture of triacylmethanes. Reaction of acetylacetone
with pivaloyl chloride gave not only the expected diacetyl

pivaloylmethane (230%, 29, eq 31) but also a small amount

0 0 MgCl 0 o o o
+ 2 \ + (3 1)
Cl Et3N ’
CH 3CN o o

2.0 2.1

 

(21% by GC) of triacetylmethane (2}, eq 31). Use of iso—
butyryl chloride as the acylating agent gave not only the
expected diacetyl isobutyrylmethane (540%, 23) but also

small amounts of triacetylmethane (215%, 2}) and

27

Table 5. Acylation of Acetylacetone.

MgClz, base

 

 

 

CH2(COCH3)2 + RCOCl > RCOCH(COCH3)2
1 hr
Entry RCOCl Basea Solvent Yield(%)b
1 CH3COC1 pyr CH2C12 83
2 CGHSCOCl pyr CHZCl2 78
c
3 C6H5COC1 pyr CH2C12 79
4c C6H5COC1 pyr CH3CN 79
c
5 C6H5COC1 Et3N CH3CN 98
6 (CH3)3CCOC1 pyr CH2C12 5
7° (CH3)3CCOC1 Et3N CH3CN 30d
e
8 (CH3)2CHCOC1 pyr CH2C12 40
f
9 CH3COC1 pyr CH2C12 g

 

a)
b)
C)
d)
e)
f)

9)

2 Equivalents of base used in all cases.

Isolated yields.

Reaction stirred for 12 hr.

21% (GC) triacetylmethane.

215% triacetylmethane, 25% diisobutyryl acetylmethane.

Reaction carried out in the absence of magnesium chloride.

37% Isolated yield of O-acylated acetylacetone.

28

diisobutyryl acetylmethane (25%, 13, eq 32). A possible

+
\T/JK\Cl——_> 2 pyr
CH2C12

%}

explanation for the occurrence of the anomalous triacyl-
methanes (11, eq 31; 11 and 13, eq 32) is offered in Fig-
ure 3. Acetylacetone is acylated to give the expected tri-
acylmethane 15. Previous studies have shown that if R is a
bulky group, the triacylmethane 15 exists almost exclusively
in the keto form53. If 15 is in the keto form, it can
act as an acylating agent and react with 14 to give tri-
acetylmethane (11) and enolate 1634. Enolate 16 can then
react with remaining acid chloride to give the disubstituted
triacylmethane 17.

In order to determine Optimum conditions for the re-
action of acetylacetone with isobutyryl chloride, the re-
action was run under a variety of conditions and the pro-

ducts determined by GC. As can be seen in Table 6, although

the reaction is faster in CH2C12 solution, less of the

29

Figure 3. Transacylation of Acetylacetone.

O 0'

//u\v/*\‘ RCOCl \
// .7
25

 

RCOCl

1o ‘0

\
/

19

 

\/

30

Table 6. Reaction of Acetylacetone with Isobutyryl Chloride,
a Solvent Study.

 

 

 

 

 

MgCl2
CH2 (COCH3) 2 + (CH3)2CHCOC1 2 Et N > 12 + 11

3

Reaction Yield (%)a

Entry Solvent Time (hr) 12 11

1 CH3CN 0.25 11 ——

2 CH3CN 0.50 55 1

3 CH3CN 1.0 55 5

4 CH2C12 0.25 50 1

5 CH2C12 1.0 56 10

6 CH2C12 12.0 28 15

 

a) GC yields.

31

anomalous triacetylmethane (11) was formed when the reaction
was conducted in CHBCN (Entries 3 and 5, Table 6). Thus,
all subsequent reactions were performed in CHBCN.

A survey of various bases (triethylamine, pyridine,
dimethyl aniline, diethyl aniline, lutidine, and collidine)
revealed collidine as the most efficient of the bases tested
in the acylation of acetylacetone with isobutyryl chloride
(Table 7). Not only did collidine give the highest yield
of product (77% GC, 60% isolated), analysis of the reaction
mixture revealed that essentially no triacetylmethane was
formed (21% GC). This result is especially surprising since
lutidine, which differs from collidine only by not having a
methyl group at the 4 position, gave only a 51% yield (GC)
of product (12) and a significant amount of triacetylmethane
(210% GC). It appears, therefore, that the use of collidine
as a base somehow suppresses the transacylation reactions
which lead to the formation of triacetylmethane. Collidine
was also shown to give almost identical yields of C-acylated
product as pyridine in the acylation of acetylacetone with
acetyl chloride (81% isolated yield of triacetylmethane) and
benzoyl chloride (78% isolated yield of diacetyl benzoyl-
methane).

As mentioned previously, Masamune49 and Kobuke50 have
shown that magnesium salts of malonate half-esters can be
acylated with acyl imidazoles (eq 21). We therefore studied
whether acyl imidazoles might also be used in our acylation

procedure. It should be noted that acyl imidazoles can be

32

Table 7. Survey of Bases in the Acylation of Acetylacetone
with Isobutyryl Chloride.

 

 

 

 

 

MgClz, base
CH2 (COCH3) 2 + (CH3)2CHCOC1 CH3CH' 1 hr > 12 + 11

a Yield (%)b

Entry Base 13 21

1 Et3N 55 5

2 pyr 46 10

3 PhNMe2 68 l

4 PhNEt2 49 5

5 lutidine 51 10

6 collidine 77 (60%)C 1

 

a) 2 Equivalents of base used in all cases.
b) GC yields.

c) Isolated yield.

33

prepared directly from the corresponding carboxylic acid
whereas some acid chlorides cannot be prepared directly from
the corresponding acid. Thus, the successful use of acyl
imidazoles would increase the versatility of our procedure.
With acyl imidazoles only one equivalent of added base
should be necessary, since the imidazole anion released in
the reaction could act as the second equivalent of base
needed to convert the highly acidic triacylmethane to its

conjugate base (eq 33). Unfortunately, reaction of acetyl-

 

acetone with isobutyryl imidazole gave none of the expected
diacetyl isobutyrylmethane (12, eq 32); only triacetyl-
methane was observed (GC). This result may be due to the
fact that acyl imidazoles are less reactive than the cor-
responding acid chlorides. As the reactivity of the added
acylating agent decreases, one would expect the triacyl-
methane (1?, eq 34) to become competitive as an acylating
agent, thus decreasing the yield of 1?, and, at the same
time, increasing the amount of triacetylmethane formed.

As was the case with both diethyl malonate and ethyl
acetoacetate, no C-acylated product was formed when the

acetylation of acetylacetone was conducted in the absence of

34

 

.. 8 MUN“
M RCOX 2V
/ Mn
28
’b
29
”b
added MgCl2 (Entry 9, Table 5). Unlike the previous cases,

however, a 37% yield of O-acylated acetylacetone was isola-
ted. This result demonstrates the tendency of magnesium
enolates to acylate at carbon rather than at oxygen since,
in the presence of added MgClz, no O-acylated product was
observed.

In conclusion, the procedure which we have deve10ped
offers an extremely mild method for the acylation of B-
dicarbonyl compounds. Although there have been previous
reports of the acylation of magnesium enolates of B-dicar-
bonyl compounds, it was necessary to preform and isolate
the magnesium enolate; usually be reaction of the B-dicar-
bonyl compound with magnesium alkoxides. Also, many of the
previous methods require two equivalents of the magnesium
enolate per equivalent of acylating agent; the second
equivalent acting as a base and reacting with the product
triacylmethane. In our acylation procedure, the magnesium
enolate is generated in situ and reacted, without prior
isolation, with one equivalent (based on the starting 8-
dicarbonyl compound) of acid chloride in a one-pot reaction.

This not only decreases the amount of time needed to prepare

35

the triacylmethane, but also decreases the cost of the re-
action since a 1:1, not a 2:1 ratio of B-dicarbonyl to acy-
lating agent is sufficient. In fact, triacetylmethane,
which is used extensively in industry, can be prepared using
our procedure for approximately $10/mole (cost based on
materials used, does not include man hours). This same com-
pound is listed by Aldrich at a price of $1,000/mole (avail-
able in 5g samples). Thus, our procedure offers not only an
exceptionally mild method (essentially neutral conditions)
for the acylation of B-dicarbonyl compounds, but is also

extremely economical.

Experimental

 

Materials

 

Reagent grade methylene chloride was dried over 4A
molecular sieves. Acetonitrile was distilled from calcium
hydride as were all the commercially available amines.
Copper (II) chloride was obtained as an anhydrous reagent
from Fisher Scientific Co. The remaining Lewis acids were
obtained as anhydrous reagents from Aldrich Chemical Co.
Diethyl malonate, ethyl acetoacetate, acetylacetone, and the
acid chlorides were also obtained from Aldrich Chemical Co.,
and were purified by simple distillation. Isobutyryl imida-
zole was prepared from isobutyryl chloride and imidazole by

the method described by Staab54.

36

Methods of Analysis
1H NMR data were obtained on a Varian T-60 spectrometer
at 60 MHz. Chemical shifts are reported in parts per million
on the delta scale relative to TMS internal standard. Mass
spectral data were acquired with a Finnigan Model 4000 elec-
tron impact GC/Mass spectrometer. Gas chromatographic anal-
ysis were performed with a Varian 920 chromatograph equipped

with a 6 ft. by 0.25 in. stainless steel column packed with

15% SE-30 on Chromosorb W.

General Acylation Procedure Used to Survey Various Lewis
Acids

 

A flame-dried 50 mL round bottom flask equipped with
septum inlet, magnetic stirrer, and mercury bubbler was
flushed with argon and charged with 10 mL of dry methylene
chloride, diethyl malonate (10 mmol, 1.52 mL), and pyridine
(20 mmol, 1.60 mL). The Lewis acid (10 mmol) was added and
the resulting heterogeneous mixture was stirred for 15 min-
utes. The flask was immersed in an ice bath and acetyl
chloride (10 mmol, 0.72 mL) was introduced into the flask
via the septum inlet. After stirring the reaction mixture
for 15 minutes at 0°C, the cooling bath was removed and the
mixture was stirred for 12 hours. After cooling to 0°C, the
reaction was quenched with 5 mL of 6 M HCl. The resulting
solution was washed three times with 5 mL of diethyl ether

and the combined ether extracts were dried (MgSO The

4"

ether solution was analyzed by GC (dodecane as an internal

standard) for unreacted diethyl malonate.

37

General Procedure for the Acylation of Diethyl Malonate

 

A flame-dried 100 mL round bottom flask equipped with
septum inlet, magnetic stirrer, and mercury bubbler was
flushed with argon and charged with 25 mmol (2.389) of an-
hydrous magnesium chloride. Dry acetonitrile (25 mL) was
added to the flask. To the resulting heterogeneous mixture
was added 25 mmol (3.80 mL) diethyl malonate. The reaction
flask was immersed in an ice bath and 50 mmol (6.97 mL) of
triethylamine was added via the septum inlet. After stir-
ring for 15 minutes at 0°C, 25 mmol of acid chloride was
added. The resulting mixture was stirred 1 hour at 0°C and
12 hours at room temperature. After cooling to 0°C, the re-
action was quenched with 15 mL of 6 M HCl. The resulting
solution was washed three times with 20 mL of diethyl ether.

The combined ether extracts were dried (MgSO filtered,

4)!
and the solvent removed under vacuum. The resulting residue

was purified by bulb to bulb distillation.

Diethyl acetylmalonate was prepared from diethyl manur-

 

ate and acetyl cthride.: Bulb to bulb distillation (90°C/
0.25 mm) gave 4.7976g of a clear liquid which was shown to

be a mixture of diethyl malonate and diethyl acetylmalonate.

1H NMR analysis revealed that the mixture contained approxi-

mately 21 mmol (85%) of diethyl acetylmalonate. 1H NMR

(CDC13), 61.1-l.5 (m, 6H), 2.2 + 2.3 (3, total 3H), 4.0-4.5

(m, 4H), 13.3 (5). MS: (m/e) 203 (MT + l), 187 (M? - CH3),

160 (M? - o=c=cn 115, 86, 69, 43.

2).

38

Diethyl benzoylmalonate was prepared from diethyl malo-

 

nate and benzoyl chloride in 89% yield (b.p. 140°C/0.25 mm).

1

H NMR (CDCl 51.1-1.3 (t, J=7Hz, 6H), 4.0-4.4 (q, J=7Hz,

)
3 I
4H), 5.3 (s), 7.3-7.9 (m, 5H), 13.1 (5). MS: (m/e) 264

(HT), 105 (PhCOT).

Diethyl isobutyrylmalonate was prepared from diethyl

malonate and isobutyryl chloride in 92% yield (b.p. 100°C/

1

0.4 mm). H NMR (CDCl 61.0-1.5 (m, 12H), 2.5-3.0 (m, 1H),

),
3
4.0-4.4 (m, 4H), 4.6 (s), 13.2 (5). MS: (m/e) 230 (MT),

187 (m? - CH(CH3)2), 159 (M? - (CH3)2CHCO), 159, 141, 87,
71, 43.

Diethyl n-butyrylmalonate was prepared from diethyl

malonate and n-butyryl chloride in 86% yield (b.p. 100°C/

0.4 mm). 1H NMR (CDCl ao.7—2.o (m, 11H), 2.1-2.7 (m,

).
3
2H), 4.0-4.5 (m, 4H), 13.3 (s). MS: (m/e) 230 (MT), 187

(M? - CH CH CH 159 (M? - CH

2 2 3" CH

CHZCO), 141, 87, 71, 43.

3 2

Diethyl pivaloylmalonate was prepared from diethyl
malonate and pivaloyl chloride in 90% yield (b.p. 100°C/

1

0.4 mm). H NMR (CDCl 51.1-1.4 (m, 15H), 4.0-4.4 (m, 4H),

3),
4.9 (s, 1H). MS: (m/e) 245 (M? + 1), 159 (m? - (CH3)3CCO),
85, 57, 41.
1H NMR Study of the Diethyl Malonate/Magnesium
Chloride System

A flame-dried 50 mL round bottom flask equipped with

septum inlet, magnetic stirrer, and mercury bubbler was

39

flushed with argon and charged with 10 mL of CD3CN and 5
mmol (0.76 mL) of diethyl malonate. A 1H NMR of the result-
ing solution was taken.

1H NMR (CD3CN), 51.0-1.4 (m, 6H), 3.3 (s, 29), 3.9-4.3

0n, 4H).

The NMR sample was returned to the flask and 5 mmol
(0.66 mL) of triethylamine was added. A 1H NMR was taken of
the resulting solution.

1H NMR (CD3CN), 60.8-1.4 (m, 15H), 2.2-2.7 (m,6H), 3.3

(s, 2H), 3.9-4.3 (m, 4H).

The NMR sample was returned to the flask and 5 mmol
(0.489) of anhydrous magnesium chloride was added. The re-
sulting heterogeneous mixture was stirred for 15 minutes at
room temperature. As the mixture was stirred it became more
viscous. An aliquot of the mixture was removed, filtered,
and a 1H NMR was taken of the filtrate.

1H NMR (CD3CN), 51.0-1.4 (m, 15H), 2.8-3.3 (m, 6H), 3.8-

4.2 (m, 5H).

Isolation of the Magnesium Enolate of Diethyl Malonate

A 0.5 M diethyl ether solution (20 mL; 10 mmol) of di-
ethyl malonate was prepared under an atmosphere of argon.
To this solution were added 10 mmol (1.39 mL) of triethyl-
amine and 10 mmol (0.959) of anhydrous magnesium chloride.
The resulting heterogeneous mixture was stirred for 1.5
hours. The mixture was filtered under an atmosphere of

argon and the solvent was removed from the filtrate under

40

reduced pressure leaving 0.869 of a white solid.

1H NMR (CDC13), 50.9-1.4 (m, 6H), 3.7—4.3 (m, 5H).

General Procedure for the Acylation of Ethyl Acetoacetate
A flame-dried 100 mL round bottom flask equipped with
septum inlet, magnetic stirrer, and mercury bubbler was
flushed with argon and charged with 25 mmol (2.389) of dry
magnesium chloride. Dry solvent (25 mL, CHBCN or CH2C12,
see Table 4) was added to the flask. To the resulting
heterogeneous mixture was added 25 mmol (3.19 mL) of ethyl
acetoacetate. The reaction flask was immersed in an ice
bath and 50 mmol of base (pyridine or triethylamine, see
Table 4) was added through the septum inlet. After stir-
ring for 15 minutes at 0°C, 25 mmol of acid chloride was
added. The resulting mixture was stirred for 15 minutes
at 0°C and 1 hour at room temperature. After cooling to
0°C, the reaction was quenched with 15 mL of 6 M HCl. The
resulting solution was washed three times with 20 mL of
diethyl ether. The combined ether extracts were dried
(M9504), filtered, and the solvent removed under reduced
pressure. The resulting residue was purified by bulb to

bulb distillation.

Ethyl-3-oxo-2-acety1butanoate was prepared from ethyl

 

acetoacetate and acetyl chloride in 91% yield (b.p. 45°C/

1

0.2 mm). H NMR (CDC13), 61.3 (t, J=7Hz, 3H), 2.4 (S, 6H),

4.3 (q. J=7Hz, 2H), 17.5 (s, 1H). MS: (m/e) 172 (M1), 157

(MT - CH 129 (M? - CH3CO), 98, 85, 43.

3)!

41

Ethyl-3-oxo-2-benzoylbutanoate was prepared from ethyl

acetoacetate and benzoyl chloride in 81% yield (b.p. 140°C/

1

0.25 mm). H NMR (CDC13), 60.7-1.4 (m, 3H). 2.0-2.4 (5,

total 3H), 3.7-4.3 (m, 2H), 5.3 (s), 7.2-7.9 (m, 5H), 12.9
(s), 16.3 (bs). ms: (m/e) 234 (MT), 233 (MT - H), 219

(M? - CH3), 187, 105 (Phcof), 77, 43.

Ethyl-3-oxo-2-acetyl-4-methyl pentanoate was prepared

from ethyl acetoacetate and isobutyryl chloride in 77% yield

1

(b.p. 55°C/0.2 mm). H NMR (CDCl 61.0-1.5 (m, 9H), 2.3

3).
(s, 3H), 2.9-3.4 (m, 1H), 4.0-4.5 (m, 2H), 17.3 (5, 1H).

MS: (m/e) 200 (M1), 185 (M? - CH 155, 71.

3)!

Ethyl-3-oxo-2-acety1hexanoate was prepared from ethyl

acetoacetate and n-butyryl chloride in 78% yield (b.p. 54°C/

1

0.2 mm). H NMR (CDCl 50.8-2.0 (m, 8H), 2.3 (s, 3H),

3).
2.3-2.8 (m, 2H), 4.1-4.5 (m, 2H), 17.4 (s, 1H). MS: (m/e)
201 (M? + 1), 185 (MT - CH3), 157 (MT - CH3CO), 139, 129,

111, 71, 43.

Ethyl-3-oxo-2-acetyl-4,4-dimethylpentanoate was prepared

 

from ethyl acetoacetate and pivaloyl chloride in 75% yield

(b.p. 65°C/0.25 mm). 1

H NMR (CDC13), 61.1-1.5 (m, 12H),
2.3 (3, 3H), 4.0-4.4 (m, 2H), 5.0 (s, 1H). MS: (m/e) 214

(MT), 199 (M? - CH 173, 155, 131, 85.

3)!

General Procedure for the Acylation of Acetylacetone

 

Acetylacetone was acylated in a manner identical to that

described for ethyl acetoacetate substituting 25 mmol

42

(2.57 mL) of acetylacetone for 25 mmol of ethyl acetoacetate.

Triacetylmethane was prepared from acetylacetone and

acetyl chloride in 83% yield (b.p. 55°C/0.5 mm). 1H NMR

 

(CDCl 52.1 (s, 6H), 2.3 (8, 3H), 16.7 (s, 1H). MS:

3).
(m/e) 142 (MT), 127 (M? - CH3), 100 (MT - CH2=C=O), 85,

67, 43.

Benzoyl diacetylmethane was prepared from acetylacetone

and benzoyl chloride in 98% yield (b.p. 125°C/0.25 mm). In

 

NMR (CDCl 62.0 (s, 6H), 7.4-8.0 (m, 5H), 16.6 (s, 1H).

3).

MS: (m/e) 204 (MT), 189 (MT - CH 161 (M? - CH3CO), 147,

3)!
127, 105, 85, 77, 43.

Pivaloyl diacetylmethane was prepared from acetylacetone

 

and pivaloyl chloride. Bulb to bulb distillation (50°C/
0.05 mm) yielded 1.69729 of a clear liquid. 1H NMR analysis

showed the liquid to be a 3:1 mixture of pivaloyl diacetyl-

1

methane : triacetylmethane giving a 30% H NMR yield of

pivaloyl diacetylmethane.

1

H NMR (CDCl 51.1-1.3 (m, 9H), 2.2-2.4 (3, total 6H),

).
3
5.6 (s) + 15.4 (bs, total 1H). MS: (m/e) 184 (M1), 142

(m? - CH =C=O), 127 (H? - C(CH 101, 85, 57, 43.

2 3’3"

Isobutyryl diacetylmethane was prepared from acetyl-
acetone and isobutyryl chloride following the procedure de-
scribed above and using the reagents listed in Entry 6 of
Table 7. Bulb to bulb distillation (b.p. 40°C/0.1 mm)

yielded 2.679 of a clear liquid. 1H NMR analysis revealed

43

the liquid to be a 12:1 mixture of isobutyryl diacetyl-
methane and triacetylmethane giving a 60% 1H NMR yield of
isobutyryl diacetylmethane. 1H NMR 01.0-l.3 (m, 6H), 2.1—
2.3 (m, 6H), 2.4-3.2 (m, 1H), 5.0 (s) + 16.5 (s) + 16.7 (5,
total 1H). MS: (m/e) 171 (M? + 1), 152, 127 (MT - CH3CO),

85, 71, 43.

Diisobutyryl acetylmethane was formed as a by-product

 

in the reaction of acetylchloride with isobutyryl chloride
and was identified by its GC/MS: (m/e) 198 (MT), 155 (M? -

CH3CO), 137, 128, 113, 85, 71, 43.

Acetylacetone enol acetate was prepared from acetylace-

 

tone and acetyl chloride in the absence of magnesium chlo-
ride in 37% yield (b.p. 38°C/0.25 mm). 1H NMR 62.0-2.4 (m,
11H), 5.8 (s) + 6.1 (5, total IR). 143 (M? + 1), 127 (M? -

CH3), 101, 85, 43.

GC Study of the Acylation of Acetylacetone with Isobutyryl
Chloride

 

Acetylacetone was reacted with isobutyryl chloride ac-
cording to the procedure described for the acylation of ethyl
acetoacetate. Following acid work-up, the combined ether
extracts were dried (M9804) and analyzed for isobutyryl di-

acetylmethane by GC (tetradecane as an internal standard).

Acylation of Acetylacetone with Isobutyryl Imidazole

 

Acetylacetone was reacted with isobutyryl imidazole in

CH2C12 with triethylamine according to the procedure for the

44

acylation of ethyl acetoacetate. GC analysis revealed tri-

acetylmethane as the only product.

Chapter III

AN INTRODUCTORY STUDY OF THE CARBOMETHOXYLATION
OF KETONES USING WEAK BASES

Introduction

After our success in acylating B-dicarbonyl compounds
with acyl halides, using triethylamine and M9Cl2 to promote
enolate formation, we decided to extend the procedure to the
acylation of ketones. Previous work in this laboratory55
has shown that the acetylation of cyclic ketones with acyl
imidazoles, using triethylamine and M9C12 to promote enolate

formation, gave C-acylated products in moderate yields (eq

35). However, acyclic ketones failed to react under these

 

O (J O
2 Et3N A (35)
+ CH3COIm 2'
(CH2)" 2 M9C12 cu-I2n
CH CN
n=2,3'4 3 48-67%

conditions.

A possible explanation for the low yield of C-acylated
products from reactions involving acetyl imidazole is a self-
condensation similar to the Claisen ester condensation (eq
36). Staab has observed such a reaction in an attempted
synthesis of E-butyl acetate from acetyl imidazole and

45

46

o 0 o
g M9Cl u H
2 CH -Im j} CH CCH C-Im (36)

base 3 2

 

E-butanol in the presence of sodium E-butoxide56. Clearly,

use of an acylating agent which lacked alpha protons would
eliminate this side reaction.

We chose to study the reaction of ketones with methyl
chloroformate in the presence of triethylamine and MgCl2 (eq

37). Methyl chloroformate does not have protons alpha to the

9g M
+ ClCO CH I> (37
1 2 3 MoCl n n1 0cm3 )

 

R ‘ 2

carbonyl function and thus cannot undergo a condensation re-
action of the type shown in equation 36.
Carboalkoxylations of ketones to form B-keto esters
have been accomplished by reaction of ketones with diethyl
carbonate (10, X=0Et) or ethyl chloroformate (10, X=Cl) in
the presence of such bases as sodium ethoxide or sodium hy-
dride (eq 38)S7. Diethyl oxalate (1}, eq 39) and ethyl di-
ethoxyphosphinyl formate (12, eq 40) have also been employed
as acylating reagents in the carboethoxylation of ketones.
All of the carboalkoxylation reactions referred to above

require the use of a strong base to form the ketone enolate.

47

/

O 1) base 0 O (38)

2

 

' O C) 0
P, 1) 11 OEt A 30025: 0025189)
. + (EtOC)2 ‘a + 7‘ . -—->
%} 2) H30
0 O 0 o O
u Most) co (5: (40)
. + EtOfiP (OEt)2 £139 &3 2 ...—9 2

(£3 0

Also, to insure a good yield of product, many of the pro-
cedures require the use of excess ketone enolate. We thought
that these problems could be avoided if the procedure which
we developed for the acylation of B-dicarbonyl compounds
could be extended to the carboalkoxylation of ketones (eq
37). The development of such a mild carboalkoxylation pro-
cedure would provide a useful alternative to the conven-

tional procedures.

48

Results and Discussion
Cyclohexanone was reacted with methyl chloroformate in
the presence of MgCl2 and triethylamine in CH3CN solution

(eq 41). It was necessary to use two equivalents of tri-

o —O 7

.‘
M9C1 021110 002141. (41)
' \
. + ClCOZMe 2 Et3N z . "3 1;?) 313

CH CN

3 L. 39 __

N

i

 

 

ethylamine, as the second equivalent is consumed by the
acidic B-keto ester product 13. The resulting white sus-
pension was allowed to stir overnight before quenching with
aqueous HCl. Unfortunately, analysis of the reaction mixture
(GC) revealed no B-keto ester (13) was formed.

Previous work in our laboratory had revealed that in the
acetylation of cyclohexanone under similar conditions, acetyl
imidazole gave a much higher yield of 2-acetyl cyclohexanone
(67%) than acetyl chloride (28%, eq 35, n=4). Thus, we de-
cided to attempt the carbomethoxylation using carbomethoxy
imidazole as the acylating agent (eq 42). Only one equiva-
lent of triethylamine was used, as the imidazole anion which
is released can act as the second equivalent of base and re-
act with the B-keto ester 39. Unfortunately, this procedure
provided only a 9% isolated yield of 13.

A possible explanation for the low yield of product ob-
served here may be the occurrence of a side reaction between

M9Cl2 and the imidazole produced in the reaction (eq 43).

49

O o (42)
N

O
I
1 Et 3N \ OzMe _
+ MeOC-Im 7 + Im >
MgCl2

C) e _
‘2” .m

 

MgCl2 + H-Im + Et3N —> Et3NoHCl + ClMg-Im (43)

14

The chloromagnesium imidazole (14) produced in this reaction
would be a less active Lewis acid than MgCl2 and therefore
should not be as effective as MgCl2 in promoting enolate for-
mation of cyclohexanone. A possible solution to this prob-
lem would be to add a second equivalent of MgCl2 to the re-
action mixture. Also, because amines weakly chelate to
M9C1228, a second equivalent of triethylamine would probably
be needed. When these conditions were applied to this re-
action, only a 38% yield of 39 was obtained (eq 44).

We next reacted cyclohexanone with carbomethoxy imidazole
in the presence of a 2:1 mixture of sodium iodide and M9C12,
with triethylamine as the base (eq 45). It was assumed that

this mixture would generate magnesium iodide, which, being a

stronger Lewis acid than M9C12, would better promote enolate

l
2 Et N 0M0

. + ImCOZMe 3 > . 2 (44)
2 MgCl

13, 38%

 

O

Et3N
\
+ ImCOZMe + MgCl2 + 2 NaI CH3CN /. 19 (45)

 

formation from cyclohexanone. The result of this change was
an increase in the yield of 13 from 9% (Entry 2, Table 8) to
47% (Entry 6, Table 8). Further adjustments of the reaction
conditions increased the yield of 19 to 64%. The optimum re-
action conditions were found to be a 2:2:1 ratio of MgClZ:
triethylamine:sodium iodide and these conditions were em-
ployed for all our further studies.

Extension of this procedure to the carbomethoxylation
of acetophenone gave a 64% isolated yield of methyl benzoyl-
acetate. Carbomethoxylation of diethyl ketone (5 mmol
scale), however, gave a mixture of three B-dicarbonyl com-
pounds with a combined weight of 0.409 (256%, eq 46). The
compounds were identified by GC/mass spectrometry and by 1H
NMR analysis of GC isolated samples. A possible explanation

for the occurrence of products 16 and 17 is offered in

51

Table 8. Carbomethoxylation of Cyclohexanone.

O o

Et3N Oils
+ MgCl2 + NaI + ImCOZMe >
CH CN

 

 

# of equivalents

 

 

 

Entry MgCl2 NaI Et3N Yield
1b 1 - 2 0
2 1 - 1 9
3 2 — 1 26
4 2 - 2 38
5 1 2 1 47
6C 1 2 1 41
7 2 2 1 36
8 2 2 2 64
9 2 4 2 59

10 2 1 2 64
11 2 0.2 2 48
12b 1 2 2 0

 

a) Based on 5.0 mmol of cyclohexanone.

b) Methyl chloroformate used in place of carbomethoxy
imidazole.

c) Refluxed overnight.

52

O 2 Et N, 2 M9Cl

 

3 2 \
W + ImCOZMe 1 NaI, CHBCN , (46)
o o o o
Mm + + Me 0M0
15 16 3]

Figure 4. Diethyl ketone may be his acylated to give the
triacylmethane 18. Triacylmethanes, especially those which
are unable to enolize, have been known to act as acylating
a9ents34. Thus, the triacylmethane 1§ may acylate diethyl
ketone to give 16 and 17 after acidification. Triethylamine
may also react with 18 in a manner similar to that of equa-
tion 30 to give ketene 19 and the B-dicarbonyl 13. This may
explain why 1] qualitatively appears to be the major pro-
duct (analyzed by GC).

The transacylation reactions shown in Figure 4 are not
expected to occur in the carbomethoxylation of cyclic ketones.
Should bis carbomethoxylation of the cyclic ketone occur, the
resulting triacylmethane 49 would probably react with remain-
ing ketone enolate 4} to give two equivalents of the expected
product 1? (eq 47). Alternatively, triacylmethane 10 may be
hydrolyzed during the acidic work-up (eq 48).

In conclusion, this procedure has the potential to be a

53

Figure 4. Carbomethoxylation of Diethyl Ketone.

 

(K) ImCOZMe [fin/00¢“ ImCOzMe > RjAcozugh

365 .18

0’

O 00
fi+18 -——> M ”‘Mm
16 17

 

n) In
H co Me) CH
Et3N:) W 2 2 3>C=C=O + 7

38 H
'b

\J/
8‘”

54

o
W002M85+ A 2N02Me ( 4 7 )
c112)n (012),, 5 (0142),,
44,0 4,3 4%2

O O
NCOZMQMZ HC 1 9 $02.“ HO/u\OMe (4 8 )
CH5)"

mild and convenient method of carbomethoxylating cyclic
ketones. Additional study is required, however, to increase
the generality and effectiveness of this procedure, espe-
cially in the carbomethoxylation of non-cyclic ketones.
Perhaps the use of a less reactive carbomethoxylating agent
will eliminate the transacylation reactions in the carbometh-

oxylation of diethyl ketone.

Experimental

 

Materials

 

Acetonitrile and triethylamine were distilled from cal-
cium hydride. Methyl chloroformate was obtained from Al—
drich Chemical Co. and distilled prior to use. Carbomethoxy
imidazole was prepared from methyl chloroformate and imida-

zole by the method described by Moodie57. Sodium iodide,

 

55

purchased from J.T. Baker Chemical Co., was dried by heating
under vacuum prior to its use. Magnesium chloride, acquired
as the anhydrous reagent from Aldrich Chemical Co., was

stored in a glove bag under argon. All ketones used in this
investigation were commercially available and were distilled

from calcium hydride prior to use.

Methods of Analysis
1

 

H NMR data were obtained on a Varian T-60 spectrometer
at 60 MHz. Chemical shifts are reported in parts per million
on the delta scale relative to TMS internal standard. Mass
spectral data were acquired with a Finnigan Model 4000 elec-
tron impact GC/Mass spectrometer. Gas chromatographic analy-
sis were performed with a Varian 920 chromatograph equipped
with a 6 ft. by 0.25 in. stainless steel column packed with

15% SE 30 On Chromosorb W.

General Procedure for the Carbomethoxylation of Cyclohexanone

 

A flame-dried 50 mL round bottom flask equipped with a
septum inlet, magnetic stirrer, and mercury bubbler was

flushed with argon and charged with 5 mL of CH CN and 0.52

3
mL (5.0 mmol) of cyclohexanone. The appropriate amounts of
magnesium chloride, sodium iodide, and triethylamine (see
Table 8) were added and the resulting heterogeneous mixture
was stirred for 15 minutes. Carbomethoxy imidazole (5.0

mmol, 0.489) was added and the resulting white mixture was

stirred overnight at room temperature. The reaction mixture

56

was quenched with 4 mL of 6 M HCl and the resulting solution
was washed three times with 5 mL of diethyl ether. The com-
bined ether extracts were dried (M9804), filtered, and the

solvent removed under reduced pressure. The resulting resi-

due was purified by bulb to bulb distillation.

2-Carbomethoxycyclohexanone was prepared by the procedure
described above (10 mmol MgClz, 10 mmol triethylamine, 5.0
mmol NaI) in 64% yield (b.p. 110°C/1.4 mm). In NMR (CDC13),
01.3-2.5 (m, 8H), 3.2-3.5 (m) + 12.0 (5, total 1H), 3.7 (s,

3H).

Carbomethoxylation of Acetophenone
A flame-dried 50 mL round bottom flask equipped with a
septum inlet, magnetic stirrer, and mercury bubbler was

flushed with argon and charged with 5 mL of CH CN and 0.59

3
mL (5.0 mmol) of acetophenone. Magnesium chloride (0.959,

10 mmol), sodium iodide (0.759, 5.0 mmol), and triethylamine
(1.40 mL, 10 mmol) were added to the flask and the resulting
heterogeneous mixture was stirred for 15 minutes. Carbometh-
oxy imidazole (0.489, 5.0 mmol) was added and the resulting
mixture was stirred overnight at room temperature. The
reaction mixture was quenched with 4 mL of 6 M HCl and the
resulting solution was washed three times with 5 mL of di-
ethyl ether. The combined ether extracts were dried (M9804),
filtered, and the solvent removed under reduced pressure.

The resulting residue was purified by bulb to bulb distilla-

tion giving a 64% yield of methyl benzoylacetate (b.p.

 

57

l75°C/1.4 mm). 1

H NMR (CDCl3), 03.70 (s) and 3.77 (5, total
3H), 4.00 (s) and 5.67 (5, total 2H), 7.30-8.00 (m, 5H).

ms: (m/e) 178 (HT), 147 (HT — OCH3), 105 (phcot), 77, 51.

Carbomethoxylation of Diethyl Ketone

 

Diethyl ketone was carbomethoxylated according to the
procedure described above, substituting diethyl ketone (0.53
mL, 5.0 mmol) for acetOphenone. After acidic work-up, the
solution was washed three times with 5 mL of diethyl ether.
The combined ether extracts were dried (M9804), filtered,
and the solvent removed under reduced pressure giving 0.409
of a slightly yellow liquid. Analysis (GC) revealed three
products which were purified by GC preparation and identified

by their mass spectrum and 1H NMR.

Methyl-3-oxo-2-methylpentanoate (15): 1H NMR (CDC13),

01.07 (t, J=7Hz, 3H), 1.33 (d, J=7Hz, 3H), 2.53 (q, J=7Hz,
2H), 3.50 (q, J=7Hz, 1H), 3.71 (s, 3H). MS: (m/e) 144 (M1),

115 (Ht - CH CH3), 113 (Mt - OCH3), 88, 57.

2

4-Methyl-3,5-heptanedione (16): 1H NMR (CDCl 01.03

3)!
(t, J=7Hz, 6H), 1.30 (d, J=7Hz, 3H), 2.47 (q, J=7Hz, 4H),

 

3.70 (q. J=7Hz, 1H). MS: (m/e) 142 (MT), 113 (M? - CH CH

2 3"

86, 57.

1

Dimethyl methylmalonate (33): H NMR (CDC13), 01.38

(d, J=7Hz, 3H), 3.60 (q, J=7Hz, 1H), 3.71 (s, 6H). MS: (m/e)

146 (MT), 115 (M+ - OCH3), 87 (M? - CH 0C0), 72, 59.

3

REFERENCES

12.

13.

References

A. Wissner, J. Org. Chem., 44, 4617 (1979).

 

Y-N. Kuo, J.A. Yahner, and C. Ainsworth, J. Amer. Chem.
Soc., 13, 6321, (1971).

 

L. Crombie, P. Hemesley, and G. Pattenden, Tetrahedron
Lett., 3021 (1968).

 

R.B. Herbert, F.B. Jackson, and I.T. Nicholson, Chem.

Commun., 450 (1976).

For a recent review on the mechanism of decarboxylation
of B-keto acids see: R.M. Pollack in R.D. Gandor and
R.I. Schowen, Eds., "Transition States of Biochemical
Processes," Plenum Press, N.Y., 1978, p. 467.

M.W. Logue, R.M. Pollack, and V.P. Vitullo, J. Amer.
Chem. Soc., 17, 6868 (1975).

 

E. Stenhagen, Ark. Kemi, a, 381 (1951).

 

M.A. Mitz, A.E. Axelrod, and K. Hofmann, J. Amer. Chem.

 

Soc., 23, 1231 (1950).

M.W. Logue, J. Org. Chem., 19, 3455 (1974).

 

M. Stiles, J. Amer. Chem. Soc., 11, 2598 (1959).

 

G. Bottaccio and G.P. Chiusoli, Chem. Commun., 618
(1966).

 

G. Bottaccio, G.P. Chiusoli, and M.G. Felicioli, Gazz.
Chim. Ital., 199, 105 (1973).

 

A.P. Kraocho, D.S. Kashdan, and E.G.E. Jahugen, Jr.,
J. Org. Chem., 12, 1189 (1977).

 

58

 

14.

15.

16.

17.

18.

19.

20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

30.

31.

32.

59

F.L. Breush and H. Kreskin, Rev. Faculte Sci. Univ.
Istanbul, 119, 24 (1946).

 

 

R. Locquin, Bull. Soc. Chim. France, 31, 595 (1904).

 

G.C. Swain, R.F. Bader, R.M. Estreve, and R.N. Griffen,
J. Amer. Chem. Soc., 83, 1951 (1961).

 

H. Sakurai, A. Shirahata, and A. Hosomi, Tetrahedron
Lett., 21, 1967 (1980).

 

N. Matsumura, Y. Sakaguchi, T. Ohba, and H. Inoue,
Chem. Commun., 326, (1980).

 

N. Matsumura, T. Ohba, and H. Inoue, Bull. Chem. Soc.
Jpn., 55, 3949 (1982).

 

J.F.W.K. Barnick, J.L. van der Baan, and F. Bickelhaupt,
Synthesis, 10, 787 (1979).

 

M.W. Rathke and J. Deitch, Tetrahedron Lett., 2953
(1971).

 

R.L. Augustine, "Carbon-Carbon Bond Formation," Marcel
Decker Inc., Vol. 1, 1979, p. 255.

J.F. Klebe, H. Finkbeiner, and D.M. White, J. Amer.
Chem. Soc., 88, 3390 (1966).

 

o-CH3C6H4COC1 was prepared and donated by Mr. Joel R.
Christensen of Michigan State University.

T.S. Straub and M.L. Bender, J. Amer. Chem. Soc., 94,
8881 (1972).

 

 

E. Wahlberg, Ber. Dtsch. Chem. Ges., 43, 2071 (1911).

Y. Oikawa, K. Sugano, and O. Yonemitso, J. Org. Chem.,
43, 2087 (1978).

 

For a general discussion see: R.P. Houghton, "Metal
Complexes in Organic Chemistry," Cambridge University
Press, Cambridge, 1979.

G. Schulz and G. Mathias, C.A., 79 P3482? x, 1970
(Ger. Offen. 1, 803, 510).

E.C. Taylor and I.J. Turchi, Org. Prep and Procd. Int.,
19, 221 (1978).

 

M. Viscontini and K. Adank, Helv. Chem. Acta, 35, 1342
(1952).

 

P. Pollet and S. Gelin, Synthesis, 142 (1978).

 

33.

34.

35.

36.

37.

38.

39.

40.

41.

42.

43.

44.

45.

46.

47.

48.

49.

60

B. Trost and R.A. Kunz, J. Org. Chem., 39, 2648 (1974).

 

U. Schmidt and M. Schwochau, Monatsh. Chem., 98, 1492
(1967).

 

U. Schmidt and M. Schwochau, Tetrahedron Lett., 10,
875 (1967).

 

a. H. Rohner, C.A., 84, P122916 h, 1976 (Ger. Offen.
2, 433, 987).

b. M. Chochemore and M. Gay, C.A., 85, P125135 j,
1976 (Ger. Offen. 2, 600, 516).

c. K. Wilhelm, N. Wolfram, E. Konrad, and A. Albert,
C.A., 62, P11805 b, 1965 (U.S. 3, 193, 525).

H. Kato, A. Takahashi, H. Tamiya, M. Kono, and M. Shimo,
Shakuhin Eiseigaku Zasshi, Z, 60 (1966).

 

D.S. Tarbell and J.A. Price, J. Org. Chem., 22, 245
(1957) .

 

M. Viscontini and N. Merckling, Helv. Chem. Acta, 35,
2280 (1952).

 

 

S. Forsen and M. Nilsson, Acta Chem. Scand., 13, 1383
(1959).

V.A. Zagorexskii and I.S. Akhrem, Zhur. Obshchei Khim.,
29, 619 (1959).

 

M. Viscontini and K. Adank, Helv. Chem. Acta, 35, 1342
(1952).

 

 

E.C. Taylor and A. McKillox, Tetrahedron, 23, 897
(1967).

D.C. Nonhebel and J. Smith, J. Chem. Soc. (C), 1919
(1967).

 

H.D. Murdoch and D.C. Nonhebel, J. Chem. Soc., 2153
(1962).

 

D.C. Nonhebel, J. Chem. Soc., 4628 (1962).

 

H.D. Murdoch and D.C. Nonhebel, J. Chem. Soc. (C),
2298 (1968).

 

R.L. Augustine, "Carbon—Carbon Bond Formation"; Marcel
Decker Inc., New York, N.Y., 1979, Vol. 1, pp. 250-263.

D.W. Brooks, L. Lu, and S. Matsumune, Agnew. Chem. Int.
Ed- Engl-. 1‘8, 72 (1979).

 

50.

51.

52.

53.
54.

55.

56.

57.

61

Y. Kobuke and J. Yoshida, Tetrahedron Lett., 367
(1968).

 

H.O. House, "Modern Synthetic Reactions," W.A. Benjamin,
Inc., 2nd Ed., 1972.

M.E.H. Howden and M. Tyler, Tetrahedron Lett., Zé,
1979 (1975).

 

D.C. Nonhebel, J. Chem. Soc. (C), 1716 (1967).

 

H.A. Staab, Chem. Ber., 90, 1321 (1957).

 

R.E. Tirpak, Ph.D. Dissertation, Michigan State Univer-
sity, 1982.

H.A. Staab, Agnew. Chem. Int. Ed. Engl., 1, 351 (1962).

 

R.B. Moodie and R. Towill, J. Chem. Soc. Perkin II,
184 (1972.

 

9
2
1
3

"
III
S”
H
II).