Regiospecific carbon-carbon bond formation via ring opening of vinyl oxiranes with an organoaluminum reagent


Material Information

Regiospecific carbon-carbon bond formation via ring opening of vinyl oxiranes with an organoaluminum reagent
Physical Description:
v, 80 leaves : ill. ; 28 cm.
Cuevas, Mapi M., 1952-
Publication Date:


Subjects / Keywords:
Organoaluminum compounds   ( lcsh )
Epoxy compounds   ( lcsh )
Diethyl-carbo-tert-butoxy methylalane   ( lcsh )
Prostaglandins -- Synthesis   ( lcsh )
Chemistry thesis Ph.D
Dissertations, Academic -- Chemistry -- UF
bibliography   ( marcgt )
non-fiction   ( marcgt )


Thesis (Ph. D.)--University of Florida, 1988.
Includes bibliographical references.
Statement of Responsibility:
by Mapi M. Cuevas.
General Note:
General Note:

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 001128589
oclc - 20117416
notis - AFM5793
sobekcm - AA00004797_00001
System ID:

Full Text









The author wishes to express her gratitude to

Professor Merle Battiste for his guidance and advice.

Special thanks go to Merle's Perles for helping to

maintain her sanity or lack thereof. Separate thanks go

to Jim Rocca for all the help he has given with

spectroscopic problems, and to Dr. Awartani for his

chemical advice. My deepest gratitude goes to the

Graduate School for the three year fellowship. Finally,

thanks go to Luis, for helping me through the "Ph.D.



ACKNOWLEDGEMENTS ............................ ii

ABBREVIATIONS ....... ............. ....... iv

ABSTRACT.............. ....................... v

CHAPTER I INTRODUCTION .................. 1




CHAPTER V EXPERIMENTAL................... 51
General ......................... ......... 51
Reagents and Solvents ................... 52
Apparatus and Technique.................. 52

HOUSES .... ...... ....... ... .... 76

BIBLIOGRAPHY................................. 77

BIOGRAPHICAL SKETCH ........................ ..... 80


Bu tert-butyl

DEAD diethylazodicarboxylate

DIBAL diisobutyl aluminum hydride

DME dimethoxy ethane

DMSO dimethyl sulfoxide

Et ethyl

eq equivalent

LDA lithium diisopropyl amine

MCPBA meta-chloroperbenzoic acid

Me methyl

MTM methylthiomethyl

mm millimole

m mole

M Molar

Py pyridine

THF tetrahydrofuran

TMS trimethylsilyl

B.P. boiling point

PDC pyridinium dichromate

Abstract of Dissertation Presented to the Graduate
School of the University of Florida in Partial
Fulfillment of the Requirements for the Degree of
Doctor of Philosophy



Mapi M. Cuevas

December 1988

Chairman: Merle A. Battiste
Major Department: Chemistry

The reaction of diethyl-carbo-tert-butoxy

methylalane with both cyclic and acyclic vinyl epoxides

was studied as a potential route to unsaturated enones

and carbocyclic compounds. Two formal syntheses, one of

cis-jasmone and the other of a prostaglandin

intermediate, exemplify the new methodology of this

acetate anion equivalent in its reaction with cyclic

vinyl epoxides. An interesting one-step conversion of

trans hydroxy acids to cis lactones involving the use

of DEAD reagent and triphenylphosphine is also shown.



The past 20 years have brought about tremendous

advancements in the area of organometallic chemistry.

The chemistry of organoaluminum, particularly alkenyl

and alkynyl alanes, has proven to be singularly useful

in the synthesis of natural products.1 Organoaluminum

compounds exhibit chemical properties that are somewhat

different from organolithium, organomagnesium and even

their boron congeners. In 1955, referring to aluminum

alkyls, Ziegler commented that even though the analogy

to the grignard reagent is tempting, these aluminum

alkyls behave "peculiarly".2 Only one Al-C bond reacts

in a grignard fashion. Once this Al-C bond is added to

a substrate such as a carbonyl compound, the aluminum

bond in the product formed is attached to oxygen.

Immediately, the reactivity of the other Al-C bonds

decreases greatly and no further addition occurs. These

differences in reactivity and selectivity of

organoaluminum compounds are instrumental to their

synthetic utility.

Epoxides may be considered pseudocarbonyls in

their reactions with organometals. Their synthetic

potential is greater, since many times they are far

more accessible as starting materials than carbonyls.

The organometallic reagent may act as a nucleophile or

as a Lewis Acid in its reactions with epoxides. Three

competing mechanistic pathways are possible as can be

seen in Figure 1.1.

R -^OH
r R' 2

rearrangement R R.

X 3

R^"^^ +

OH 2


L' j + R


R 5

Figure 1.1

This scheme illustrates the reactions of

organometals with alkyl epoxides. In general,

dimethylmagnesium, methyllithium and cuprates give


R 1

Table 1.1


Alkylation of Epoxides

R3AI Conditions

by Trialkyl Alanes

Products Yield (%)


C14H30; 800







' l*OH







Et3AI Same as above
Ratio 1:1

Me3AI Et20;35

Me3AI C6H14; 350




predominantly nucleophilic ring opening at the least

hindered site to furnish 2', while methylgrignards give

predominantly halohydrins. In the case of

trialkylaluminum reagents ring opening at the

substituted carbon often predominates. Some products

from rearrangement are also observed, though not

exclusively. The ratio of each seems to vary with

solvent and reaction conditions (Table 1.1).

The opening of alicyclic epoxides by

organoaluminum reagents and further elaboration of the

intermediates formed could provide a useful route to

important natural products. For example, an expedient

route to lactones or alpha-substituted ketones can be

achieved through nucleophilic addition of an acetate

anion equivalent to oxiranes (Figure 1.2).

S(CH2) A0

(C CH2) H2) OH
6 7

(CH 2) 0

Figure 1.2

The classic method of epoxide openings with

malonic ester enolates is not often practical since it

involves harsh conditions refluxingg ethanol) and is

sensitive to steric effects.4 Most recently,

organometallics have been used to modulate the

reactivity and selectivity of enolate type anions.

Application of aluminum enolates to ring opening of

allylic epoxides in this laboratory evolved from

consideration of alkynyl alane research and

methodology. Fried has developed useful synthetic

methods in prostaglandin synthesis which allowed the

opening of alicyclic epoxides with alkynyl aluminum

reagents. These alanes, prepared from addition of

diethylaluminum chloride to lithium acetylides in

toluene, gave satisfactory yields of the

trans-2-alkynyl cycloalkanols (Table 1.2).5

In 1976, the first acetate anion equivalent using

an organoaluminum reagent was reported by Danishefsky.6

Ultimately interested in the preparation of

trans-lactones from epoxides, Danishefsky reacted

cyclohexene oxide with 2.5 equivalents of

diethylcarbo-tert-butoxymethyl alane 13 at -30 to -40C

to give the hydroxy ester 12 in 34% yield (Fig. 1.3).

Only the trans product was observed. This reagent was

prepared by the addition of Et2AlCl solution to lithio

tert-butylacetate, 11, (Rathke's salt).7 Reaction of

Table 1.2 Reactions of Epoxides with Alkynyl Alanes


- Et2AIC--CR -

Epoxide Alone Temp.(C) Time(hrs) Yield(%)




B 850


.,..C' CR



72 59

Rathke's salt alone with cyclohexene oxide in toluene

had afforded only 8% of 12. Subsequently the yield of

this alane reaction was improved to 68% by allowing the

reaction temperature to rise to ambient temperature and

prolonging the reaction at this temperature for 6 hrs.

Attempts to utilize this alanyl methodology with a

ring-A steroidal epoxide, however, failed and

Danishefsky abandoned this approach.

( o + LiCH2COO'Bu OCH3 O '' COOtBu
v "^ OH
10 11 12

10 + Et2AICH2COOtBu ---- 12
13 68%

Figure 1.3

Spurred by Danishefsky's initial success with the

Rathke alane, Dr. Melean Visnick decided to utilize it

in his synthesis of ()-anastrephin (Figure 1.4).8 Only

one regio-and stereoisomer was isolated from the alane



H3 CH3 CH3

Figure 1.4

When the reaction medium was toluene, the yield

was 24%, but a solvent change to THF increased the

yield to 87%. Previously, standard literature

procedures had been carried out in hydrocarbon solvents

or toluene. A literature search did not fully reveal

why toluene had been the solvent of choice. Most

reaction temperatures were 250C or higher and the

stability of alanes in such polar solvents as THF or

DME seemed to have been questioned at these

temperatures. In 1975 Crosby and Stephenson reported

that the products formed in the reaction of 3,4-epoxy

cyclopentene with diethylhex-l-ynylaluminum was solvent

dependent (Figure 1.5).9 They proposed the following

rationalization. In the absence of polar solvents the

oxophilicity of the aluminum causes a rearrangement of

epoxide 14 to the enone 18 which then reacts with the

alkynyl alane to give the cyclopentenol 19.

Table 1.3 Reaction of Oxiranes with Et2A1CH2COO Bu


s COOtBu
O + Et2AICH2COOIBu ---

Epoxide Solvent Temp(C) Time(hrs) Yield(%)







THF 300

2 No Rxn

Visnick's studies of the solvent effect in the reaction

of diethylcarbo- tert-butoxymethyl alane 13 with

epoxides are summarized in Table 1.3.1. In all cases

the nucleophilic attack occurred at the allylic carbon

and no rearrangements were reported.

Since the aluminum reagent is better solvated and

less aggregated in polar solvents such as THF it should

be more reactive and less sensitive to steric and

entropic effects than in such solvents as toluene or



-200 "
14 15 16 17
75% 7% 8%

14 _1
-23;0CH3 OH
18 19

Figure 1.5

The main interest of this research work is (a) to

explore the scope of diethylcarbo-tert-butoxymethyl

alane 13 in its reactions with o: P-unsaturated

epoxides, (b) to investigate the generality of the

regiospecific opening at the allylic position, and (c)

to illustrate its applications to the synthesis of

natural products.

As can be seen from Figure 1.6, cyclic unsaturated

epoxides could be converted into cis- or trans-lactones

or substituted cycloalkenones in essentially two steps.

The applicability of the oxirane opening by alane 13

will be exemplified in the formal synthesis of

cis-jasmone as well as a known prostaglandin





(CH2)n/ \-


(CH2)n 0


I cootBu

(CH2)n O

Figure 1.6

0 O ----s



The initial goal of this research was to

investigate the generality of the alane reaction with

various cyclic unsaturated epoxides. Visnick had

ascertained that a minimum of 2.3 equivalents seemed to

be necessary in order for the reaction to proceed with

good yields.8 The working assumption was that one

equivalent of the aluminum species was required to

coordinate with the epoxide oxygen while possibly

another delivered the acetate anion. This simplistic

assumption invoked the existence of a monomeric


(CH2), ---- (CH.2),


Figure 2.1

Based on the regioselectivity shown, the first

mechanism for the reaction was presumed to be as shown

in Figure 2.1. As shown, one equivalent of the alane

complexes with the epoxide weakening the carbon-oxygen

bond and rendering the allylic site partially positive.

A second equivalent of the reagent probably delivers

the acetate anion equivalent.

This type of mechanism suggests that the reaction

could be run with only one equivalent of Rathke's salt

and an excess of diethylaluminum chloride. Recent

stoichiometric studies done by a member of Dr.

Battiste's research group seem to support the need for

an excess amount of the Rathke alane. The yields of the

reaction were poor to nonexistent when one equivalent

of Rathke's salt and a slight excess of Et2AlCl were

used.11 The formation of chlorohydrin seems to

predominate in cases where Et2AlCl is in excess.

Several attempts have been made to try and

elucidate the structure of the Rathke alane. NMR

studies have proven inconclusive and seem to suggest

the possibility of several species in solution. The NMR

samples were prepared as in the general procedure (see

experimental) except that after the solvents were

removed under vacuum, THF-d8 was added. NMR samples of

the dimethylcarbo-tert-butoxymethylalane, 25, were also

prepared in hopes that conversion from methyl to ethyl

groups would permit a clearer view of the metalated

alkyl region of the spectrum. Both the proton and

carbon NMR showed more peaks than would be expected for

a simple monomeric species.12


Figure 2.2

Rathke prepared lithio-t-butyl acetate by treating

t-butyl acetate with LDA. In his characterization of

this salt in benzene, Rathke reports two partially

resolved doublets at 3.14 and 3.44 ppm and an infrared

band at 1620 cm-1. No band was observed between 1675
-1 12
and 2000 cm This leads to the conclusion that

Rathke's salt is a true enolate with the structure as

shown in Figure 2.3.

/O -Li
O -- Bu

Figure 2.3

One of the questions this research wished to

address was whether the aluminum metal was on carbon as

in structure 26 or on oxygen as in 27. In three

separate NMR experiments, there was no evidence of

vinyl protons for the alanes made from either Me 2ACl

or Et A1C1.

0 CH,
II //
26 27
R =Me or Et

Figure 2.4

The literature offers conflicting structures for

aluminum enolates, no doubt enhanced by the scarcity of

aluminum enolates known as compared to those of other

metals. Japanese workers have published a number of

papers in which they use several aluminum enolates in

aldol type reactions (Figure 2.5).13 They draw the

enolate formed as shown in 28 without giving direct

literature precedents for it. In 1974, Jeffrey, Meister

and Mole reported the isolation and characterization of

the aluminum enolates formed by reaction of mesityl

oxide with Me3Al in the presence of nickel acetyl

acetonate using ether or cyclopentane as a solvent.14

EtO Br
O t


Figure 2.5

The Z- and E-enolates shown in Figure 2.6 were

unusually stable. Cryometric studies showed that the
structure of the Z-enolate was dimeric while the

structure of the E-enolate was composed of dimers and

trimers. Spectral studies proved that a definite

vinyloxy structure existed- as shown in 27. The dimers

were held together by Al-O-Al bridges rather than by

Al-O-C-C-Al bridges. The lack of vinyl protons,

however, in our spectra still worried us.


+ AIMe3 2
Et2 or CsH12

C =C
H/ c=/ N
H 30 Me

tBu /Me
C =C
31 1

Figure 2.6


2 33

Figure 2.7

The Reformatsky reagent is analogous to the alane

enolate and the identity of its structure has also been

the object of discussion. The unsettling question of

whether the zinc intermediate possesses either

structure 32 or 33 seems to have been finally answered

in two papers.

In 1982, Orzinni, Pelizonni and Ricca conducted

spectroscopic studies of the Reformatsky reagent

prepared from t-butyl bromo-acetate.5 No significant

change was seen for the C-1 carbon, either in the

proton or carbon NMR, (see Table 2.1) as would have

been expected if a change in hybridization from an sp

to an sp2 carbon had occurred to give structure 33.

They concluded that there was no evidence to support an

oxygen metallated species.

Table 2.1 Spectroscopic Data for Reformatsky



Solvent C1

DMSO 22.2




1 2






1H (S)


1.84 (s.3H)

1.93 (s.3H)

1.85 (s,3H)

23.1 170.2

21.3 168.8




- CH2-







Py = pyridine

1 2

In 1983, Dutch researchers published X-ray data

that gave the first molecular stucture for the

Reformatsky reagent 35.6 X-ray diffraction analysis in

THF showed a dimer in which each zinc was surrounded by

two oxygen, one bromine and one carbon atom as in

Figure 2.8.


Br I

I I Br
tO ,Zn

Figure 2.8

The dimer, an eight-membered, non-planar ring

showed normal single bond distances for the Zn-C and

Zn-O bonds. The researchers conclude that it is

incorrect to describe the reagent as either C- or O-

metallated. They proposed two possible mechanisms for

the reaction of the Reformatsky reagent with

electrophiles, such as a ketone (Figure 2.9).

Mechanism 38 is a six centered one which, according to

studies, shows less steric hindrance and is favored by

the investigators.


Z3. /CB"
Zn '- c
0 .0
u C .Za' Zn-Br
But0/ cI t

,, OBu ,
// .*CH2---


ButO-c-- CH
O-Zn-. Br

Figure 2.9

In as much as the Rathke alane 13, behaves as a

Reformatsky reagent, we propose a similar structure to

that in 36. The NMR data we have collected, though

inconclusive, does suggest-a non-monomeric species

which has no vinyloxy protons. With these in mind, we

would like to propose the following structure for

reagent 13 (Figure 2.10).

Et CCH2 /OtBu

+ Al
tBuOZ El
IBuO"^ ''CH2

Figure 2.10

In fact, Fried has proposed a similar intermediate

for the reaction of alkynylalanes with epoxides (Figure

2.11) 17

Based on the requirement for at least two

equivalents of alane, we can also speculate on a

mechanism for its reaction with epoxides (Figure 2.12).

Evidence to support both the structure and mechanism

will be discussed further in the following chapters.

+ Me2AIC=C-R'--

Me -C--CC

Figure 2.11



C H2 C

V-7 0I


/ \ __ --c.
/ \2o-.i

Figure 2.12



In general, reactions of vinyl epoxides with

organocopper, organolithium and organomagnesium

reagents follow a predominantly SN2' process. Some of

these results are summarized in Table 3.1.18 When the

metal is lithium or magnesium, a variety of

nucleophilic attacks are seen, depending on reaction

conditions. Cuprates and alanates give exclusively S 2'


In 1987, Naruta and Maruyama reported that highly

regioselective 1,2 addition products to vinyl epoxides

had not been developed.19 They published results

describing successful additions of allylstannanes to

vinyl epoxides using BF3 OET2*

As can be seen by their mechanism (Figure 3.1),

the BF3 complexes with the epoxide and induces ring

opening before attack of the stannane reagent. In fact,

when a substituent was placed at the olefinic terminus

that could stabilize a positive charge, such as a

phenyl group, the 1,4 adduct is formed in good yield.

Table 3.1 Reactions of Metaloalkyl Reagents with Vinyl


+ RM


CH3Li (LiBr)




R HO R+ + R
a b R d
% Products
a b c d

7 38 55 0

1 36 44 19

0 6 94 0

R'= Me MeAIMe3




R'= H

0 0 100 0

The reaction occurs at the site with the more

stabilized cationic character.

R=H HO-'* R

0 BF3-OEt2 0 3 ] SnBu3 4Z

45 46

R=0 HO-


Figure 3.1

In our estimation, the addition of acetate alanes

to vinyl epoxides would then be a new complimentary

approach to those already available. The first step in

our research involved studying the reaction of the

acetate alane with a variety of cyclic vinyl epoxides

as shown in Table 3.2. Epoxides 49, 51-53, and 55 were

prepared from the enones. Alkaline epoxidation was

followed by a Wittig or Peterson olefination reaction

as shown in Table 3.3. Epoxides 50 and 54 were prepared

by peracid epoxidation of the corresponding diene

(Figure 3.2).

In all cases except one the reaction with the

alane gave the corresponding trans-hydroxy ester in

good to moderate yields. Compound 49 was difficult to

work with due to its volatility and the low yields

shown may be attributed to this. Yields for compound 51

were, as might be expected, poor due to the

neutralization of 1 equivalent of alane by the hydroxyl

proton and formation of t-butyl acetate in the reaction

medium which subsequently reacts with another

equivalent of the alane. This was avoided by protecting

the alcohol functionality.




Figure 3.2

Methylene cycloheptenyl oxide, 55, to our

surprise, failed to react cleanly. Its reaction with

the alane was sluggish. If reaction times were extended

over three hours, the substrate decomposed into a

myriad of products. A catalytic amount of BF3 OEt2 was

added in order to induce ring opening, without success.

The best yield of the hydroxy ester achieved with the

Rathke Alane was 10%. These results seem to indicate

that some stringent electronic requirements exist in

Table 3.2 Reactions of Vinyl Epoxides with Rathke




59 OH

: E


Reactions in THF at -60 to -40OC








Yield (%)








Table 3.3 Synthesis of Vinyl Epoxides




H202 /OH-








94 J3PMe/BuLi

86 COOEt

56 H

" 67 H

a *


the transition state. We know that the reaction does

not occur through a free carbonium ion intermediate

since only the trans-adduct is seen in all cases.

Instead, partial weakening of the C-O bond occurs

before the nucleophile attacks. Nucleophilic addition

occurs at the position most capable of stabilizing the

incipient positive character on the epoxide after

complexation. Coplanarity of the p orbitals of the

double bond with the incipient cationic site as shown

in Figure 3.3 is necessary.


Figure 3.3

The 3-methylene-l,2-oxidocycloheptane,55, being

conformationally more flexible than the cyclopentyl or

cyclohexyl substrates, does not seem to favor the

appropriate configuration. At approximately the same

time that these experiments were carried out, a paper

by Mas, Malacria and Gore appeared involving the

reaction of lithium trialkynylborates with acyclic
vinyl epoxides.20 They stressed the fact that

regioselective reagents that induce ring opening at

either site oc or ( to the vinyl group in acyclic vinyl

epoxides are few (Figure 3.4). The alkynyl borates they

reacted with vinyl epoxides gave SN opening only,

making these reagents a complementary approach to the

more abundant SN' processes available for vinyl


R1 R1 R2
R3 3 R3

R R4
1 2 34
R= Bu; R= Et;R R4= H 46% 16%
R= 0; R= Et; R, R4=H 36% 20%

Figure 3.4

With this in mind, the alane opening of the cyclic

epoxides seemed encouraging and it would be interesting

to find out if the acyclic cases, which were more

flexible, less conformationally biased than the cyclic

vinyl epoxides molecules discussed previously would be

as clean. The synthesis of various acyclic vinyl

epoxides was undertaken. The reaction of the epoxides

with the alane was performed in the usual way, with 3

equivalents of the alane in THF. Reaction times varied

slightly. The results are summarized in Table 3.4. As

can be seen, in general, the alane gave predominantly

Table 3.4 Reactions of Acyclic Vinyl Epoxides with

Rathke Alane




6 E




69 7:1 0
69 70



1 OH




*Isolated yield





1,2 addition at the site to the double bond, though

this was not exclusive as in the cyclic cases. It was

to be expected that the lack of rigidity of these

molecules would lower the regioselectivity of addition.

No conjugate addition, however, was observed. To our

surprise 74 did not react at all. It was even recovered

unchanged after three hours of reaction.

The possible explanation for this lack of

reactivity is the following. A steric interaction

between the alkyl group and the vinyl hydrogen as shown

in Figure 3.5 would disfavor alignment of the p

orbitals of the double bond with the C-O bond of the

epoxide. This would not allow for partial C-O bond

breaking. The alane does not appear to be a

sufficiently strong Lewis acid to open the epoxide by

complexation alone. Two other oxiranes similar to 74,

with substituents cis to the vinyl were synthesized in

order to corroborate this assumption.


CHf H CH 3 H

Figure 3.5

2,2-Cyclohexyl-l-vinyl epoxide, 75, was prepared

using dimethyl allyl sulfonium ylide as shown in Figure

3.6. 1,l-Dimethyl-2-vinyl epoxide, 76, was synthesized

from the reaction of 2-methyl-2,4-pentadiene with MCPBA

(Figure 3.6). When combined with Rathke alane, both of

these substrates failed to react, even at extended

reaction times or when the reaction mixture was allowed

to come to ambient temperatures. It was possible to

confirm this sterically induced misalignment through

NMR studies.




Figure 3.6

The best substrate to use seemed to be

1,1-dimethyl- 2-vinyl epoxide, since the chemical shift

of the two methyl groups should provide information as

to the preferred orientation of the vinyl group. It

would be expected that in order to avoid bumping

between the methyl and the vinyl hydrogen as shown in

Figure 3.5, the preferred conformation would place the

p orbitals of the double bond almost orthogonal to the

C-O bond. This in turn would shield the vicinal methyl


As can be seen in Figure 3.7, the two methyl

groups in the epoxide are distinctly separate when

compared to a small quantity of the starting diene. The

methyl assignments are based on analogy to the chemical

shifts of methyls in similar compounds as well as NOE

studies done on the dimethyl vinyl epoxide.

The methyl group syn to the vinyl group shows a

shielding of 0.08 ppm when compared to the anti methyl

indicating a preferred conformation such as that shown

in Figure 3.5. The vinyl hydrogen is staggered between

the two methyl groups while the p orbitals of the

double bond are aligned with the methyl group and

almost orthogonal to the conformation required for

reaction with the alane. Sauleau and coworkers report a

similar effect in the reaction of sodium phenoxide with

substituted oxiranes (Figure 3.8).21 Depending on the

R', R",and R"' substituents the amounts of conjugate

versus direct addition of phenoxide to the oxirane

vary. In the case where R', R"= CH and R, R'" = H,

direct addition at carbon 1 accounted for 33% of the

products. Twenty two percent was conjugate addition,

but of this only the E isomer was seen. When the system

.J- -d

is loaded with R' R" and R"' = CH3 only direct addition

was observed due to the unfavorable steric interactions

that would be required for conjugate addition.

R R"
.-O-C-C-C= CH2
R O /R' OH R"
2 ff'-ONa A
R R" --'-O-CH2--C=C--R

R R" Rt
/ \- CH2-C=C-C-R

Relative 96 Yields
R=R"=H R'=R"=CH3 33 22 II

R=H;R'=R'-R'=CH3 100 -

Figure 3.8

General and mild methods providing regioselective

1,2 addition products with vinyl oxiranes are scarce.

The Rathke alane provides such a route in exclusion of

the 1,4 addition so frequently seen with other

organometallic reagents. Ring opening of cyclic

epoxides is regiospecific and the trans-hydroxy esters

produced are important intermediates in natural product

synthesis. In acyclic cases, the regioselectivity of

the aluminum enolate is far superior to that observed

with other organometallic compounds.



Following the study of the vinyloxiranes with the

acetate alane, its synthetic applicability needed to be

explored. As was shown in Chapter I, the hydroxy esters

offer a potential route to lactones as well as 2,3-

disubstituted enones. Many of the important natural

products shown in Figure 4.1 could be conveniently

prepared from the synthetic sequence shown in Figure


cis -jasmone




Figure 4.1



The jasmone family, as exemplified in cis-jasmone

and dihydrojasmone, are important compounds in

perfumery. The prostaglandins, in their immense variety

have proven to be of great interest in pharmaceuticals

for their vasodepressant and muscle relaxant


The 2,3-disubstituted alkanone or alkenone pattern

seen in all of these compounds may be achieved by

conversion of the hydroxy esters produced from the

Rathke alane reaction with the appropriate epoxide. In

this research a formal synthesis of cis-jasmone and a

prostaglandin PG analog previously synthesized by Corey

were undertaken to illustrate the synthetic

applicability of the alane reactions. Corey's

intermediate has been subsequently converted to

11-deoxy-PGE2 and 11-deoxy-PGF222

Conversion of the hydroxy esters prepared from the

alane and vinyl epoxides to enones involves oxidation

and subsequent isomerization of the double bond.

Various oxidative processes were tried, the most

convenient one being a variation of Czernecki's PDC

oxidation.23 The PDC oxidations were carried out in dry

methylene chloride with molecular sieves and a

catalytic amount of dry acetic acid. The sieves were

ground and activated by heating in the reaction vessel

under vacuum for several minutes. After cooling, the

other materials were added. For the oxidation procedure
to work well the reaction should be done under an inert
atmosphere and all the reagents freshly distilled or
dried. Table 4.1 shows the results of the PDC
Table 4.1 PDC Oxidations of Hydroxy Esters






78 E





Exocyclic double bonds isomerized to give the
conjugated enone under the reaction conditions. In the
case were the double bond was inside the ring, the
isomerization did not occur but could be readily
achieved by reaction with p-toluene sulfonic acid. In

the case of 79, the ester could be hydrolyzed to the

carboxylic acid in K2CO3 in methanol. This constitutes

a formal synthesis of cis-jasmone. The subsequent steps

in Figure 4.2 were reported in the literature by



81 82

Figure 4.2

The use of 3-methylene cyclopentene oxide 49 in

the synthesis of another prostanoid intermediate was

halted due to the difficulty in isolating this

compound. The yields after Wittig reaction on

2,3-oxidocyclopentanone were not encouraging. Its

subsequent reaction with the alane also suffered due to

the extreme volatility of the compound. It was almost

impossible to isolate completely from solvent. Even

when it was stored in the refrigerator in a paraffin

wrapped vial it evaporated in a few days. In order to

avoid this difficulty in handling, a separate synthesis

of Corey's prostaglandin intermediate 85 was undertaken

as can be seen in the retrosynthetic scheme in Figure

4.3. The cis-lactone structure could be constructed

from the trans-hydroxy ester by isomerizing the

hydroxyl carbon and lactonizing. This could be

accomplished in a single step by a novel reaction

utilizing DEAD reagent and triphenylphosphine. The

hydroxy ester in turn could be derived from the

reaction of Rathke alane and the vinyl epoxide, 52. The

protected allylic alcohol in 52 evolves from Peterson

olefination of 2,3-oxido-cyclopentanone.

11-deoxy- PGE2
11 -deoxy-PGF2,c






83 5


Figure 4.3



Peterson olefination of 2,3-oxido-cyclopentanone

gave an 88% mixture of the Z- and E-epoxy esters in a

ratio of 7:4 DIBAL reduction of the esters afforded

the Z- and E-allylic alcohols 51. Subsequent reaction

with an excess of Rathke alane gave the trans-hydroxy

ester 58 in poor yield. As mentioned previously the

presence of a hydroxyl group neutralizes 1 equivalent

of the alane which allows the formation of t-butyl

acetate and ensuing side reactions. When the Z- and

E-epoxy alcohols were protected as the MTM ethers, 52,

the yield was then improved to 94%.

The Z- and E-trans-hydroxy esters 59 were hydrolyzed

under mild conditions to the trans-hydroxy acids

(Figure 4.4). Lactonization and epimerization of the

hydroxyl carbon could be accomplished by utilizing

triphenylphosphine and the DEAD reagent. This method

had not been described in the literature previously

even though the use of the DEAD reagent in conversion
of alcohols to esters was well known.2

The accepted mechanism of conversion of an alcohol

to an ester is shown in Figure 4.5.




64 %


O0 86%


51 R=H
52 R=MTM


""'' COOtBu KO H
95 %
59 OH

Figure 4.4




E-N=N-E P03 -- E--N-N-E -- E-N-N--E RO
03P+ + P03

E-N-N-E+-ROP03 > EN-NE R'COO- + R-O-PO3
SR'COOR + 03pP + EN--NE

Figure 4.5

A similar process could be used to form the

cis-lactone 88. After formation of the triphenyl

phosphine adduct 89, shown in figure 4.6. The

carboxylate anion acts as an intramolecular nucleophile

with backside displacement of triphenylphosphine oxide.



89 88

Figure 4.6

Since previous examples of this reaction were not

known, the reaction was attempted with the

cyclohexenehydroxy acid, 90, previously prepared by

reaction of the 1,3-cyclo-hexadiene epoxide with the

Rathke alane and hydrolysis to the hydroxy acid. The

trans-hydroxy acid was then treated with triphenyl

phosphine and DEAD reagent in THF to furnish 78% of the

cis-lactone, 91.


90 91

Figure 4.7

Encouraged with this reaction the lactonization

process was attempted with the-prostaglandin

intermediate 87. The reaction was clean and afforded

the lactone in 90% yield. The Z- and E-lactones were

separable by flash chromatography at this point.

Separation of isomers, however, was not essential to

the success of the synthesis as can be seen in the

subsequent steps.

Confirmation of the cis orientation for the Z- and

E-MTM-lactones was determined by spectral studies. The

respective proton spectra may be compared in Figure

4.8. The most pronounced difference between isomers is

shown in the D and D' protons which exhibit a complex

multiple for the E-isomer due to coupling with H and L

protons. The Z-isomer shows only a coupling to the

vinylogous proton. The vinyl proton, in turn, shows a

)-CH H

B O 0

b I I

5.5 5.0 q.5 4.0 3.5 3.0 2.5 2.0 1.5 PPM


CH2-O-CH2- S-CH3


B O 0


d f

5.5 5.0 4.5 4.0 3.5 3.0 2.5 2.0 1.5 PPm

Figure 4.8

broad triplet for the Z-isomer while proton A for the E

isomer exhibits more fine structure due to coupling to

D, D' as well as F and G. NOE studies done by Mr. Jim

Rocca confirm the structures shown. The Z-MTM-lactone

showed a 4.1% NOE between proton E and D, D', as well

as a 5.4% NOE between E and B. No NOE was observed

between protons E and A. This is consistent with the

Z-cis lactone. The E isomer on the other hand showed a

4.8% NOE between E and B as well as a 4.2% NOE between

protons E and A.

Following lactonization, the MTM protecting group

was removed using AgNO3. PDC oxidation to the aldehyde,

utilizing the PDC oxidation with molecular sieves

described previously, afforded the Z- and

E-aldehydes 93 (Figure 4.9).

It was conceived that the trans orientation of the

side group in Corey's lactone 85, could best be

achieved by preparing the OC, ( -unsaturated vinyl

ether, 94, which upon acid hydrolysis should give the,

unsaturated aldehyde, trans to the lactone ring as the

thermodynamically preferred product. Further

elaboration of the side chain should provide access to

the 11-deoxy-prostaglandins of the PG series.

92 OH

92 0O


93 0


95 0





Figure 4.9


0 r,"






Melting points were recorded using a Thomas-Hoover

capillary melting point apparatus. Analyses were

performed by Atlantic Microlab, Inc. of Atlanta,



Infra-red spectra were recorded on a Perkin-Elmer

283B Spectrophotometer. Routine mass spectra were

obtained on an Associated Electronics Industries model

MS-30 mass spectrometer at 70 eV. High resolution mass

determinations were handled on the same instrument

further equipped with a Nova Systems 4 computer. Proton

and/or Carbon NMR were recorded on a Varian Model

EM-360, a JEOL Model FX-100, Varian XL-200, or 300.

Chemical shifts were recorded relative to

tetramethylsilane, unless otherwise noted, at 0.00

Carbon chemical shifts were relative to the

deuterochloroform resonance at 77.00 ppm, unless

otherwise noted.

Reagents and Solvents

Many of the reagents used in this work were

obtained from several chemical supply houses. The alkyl

lithium solutions were titrated using 2,5-dimethoxy-

benzylalcohol. Some of the liquid reagents were

purified by distillation. Solvents such as

tetrahydrofuran (THF) and diethyl ether were distilled

from sodium-benzophenone prior to use in an apparatus
described by Brown.27 Dimethylsulfoxide (DMSO),

diisopropyl amine and methylene chloride were distilled

from calcium hydride and stored over activated sieves

or used immediately.

Apparatus and Technique

All air sensitive reactions were run in glassware

that had been flame-dried under vacuum. The glassware

was filled with an inert atmosphere of nitrogen or

argon by successive evacuation and backflushing through
a Firestone valve. Liquid reagents were added to the

reaction vessel through standard syringe techniques.

Flash chromatography or distillation was used for the

isolation of pure materials.29

General procedure for epoxidation of cycloalkenones

The enone (1.0 eq.) was added to a mixture of 30%

H202 (3.0 eq.) and methylene chloride at 15 OC.

Dropwise addition of 6 N NaOH (0.5 eq.) followed

maintaining the reaction temperature between 15 200

C. The mixture was then allowed to come to ambient

temperature. Upon completion the reaction was quenched

with water and extracted with methylene chloride.

Extracts were washed with brine and dried over Na2SO4.

After removal of solvent, Kugelrohr distillation of

crude afforded desired keto epoxides in 90 100%


2,3-oxidocyclohexane-l-one : B.P. 66-70 oC, 10 mm Hg;
-1 1
IR (neat) 1760 (s), 865 (s), 795 (s) cm ; H NMR (60

MHZ, CDC13) 3.65 (bd, 1H), 3.20 (dd, 1H), 2.80 1.50

(m, 6H) ; C NMR (CDC13) 205.5, 55.5, 54.8, 36.0,

22.5, 16.7. mass spectrum (70 eV) m/e 112 (32), 83

(18), 55 (100), 28 (24); 94% yield.

2,3-oxidocyclopentane-l-one : IR (neat) 1745 (s), 1174

(s), 840 (s), 730 (s) cm-; H NMR (60 MHZ, CDC13)

3.93 (d, 1H), 3.30 (d, 1H), 2.41 1.90 (m, 4H); 13C

NMR (CDC13) 209.2, 57.2, 54.0, 29.7, 22.4 ppm; High

resolution mass spectrum calculated mass of 98.03678,

found 98.03711; 100% yield.

2,3-oxidocycloheptane-l-one : IR (neat) 1700 (s), 930

(m), 835 (s); 1H NMR (60 MHZ, CDC13) 3.43 (d, 2H),

2.92-1.38 (m, 8H); 13C NMR (CDCl3) 205, 59, 55, 40, 27,

24, 23 ppm; mass spectrum (70 eV) m/e 126 (11), 97

(27); 83 (19); 70 (58); 55 (85); 41 (100); 39 (43); 27

(49); 28 (24). B.P. 450C at 15 mm Hg. Yield:

quantitative. In the case of cycloheptenone, methanol

was used as solvent.

General preparation of methylene epoxides

A three necked flask with 1.0 to 1.3 equivalents

of methyltriphenylphosphonium bromide and stirrer was

dried overnight in a vacuum oven at 40 -50 oC. The

flask was then fitted with vacuum and thermometer

adapters as well as septa. The flask was flushed with

argon by means of a firestone valve and charged with

dry THF. After cooling to -78 C, 1.0-1.3 equivalents of

n-butyllithium were added slowly. The colored solution

was stirred for 1 hr at -780C. 1.0 equivalent of the

keto epoxide dissolved in a small amount of THF was

added to the ylide. The reaction mixture was allowed to

come to room temperature slowly and quenched with

saturated NH4CL solution, extracted with ether or

pentane, washed with brine and dried over Na2SO4'

Purification of crude was achieved by Kugelrohr


3-methylene-1,2-oxidocyclohexane: IR (neat) 3000-2900

(s), 900 (m), 730 (m); 1H NMR (60 MHZ, CDC13) 5.37 (s,

1H), 5.26 (s, 1H), 3.43 (bs, 2H), 2.4-1.2 (m, 6H); 13C

NMP (CDC13) 143, 116, 55, 54, 29, 24, 20 ppm; 56%


3-methylene-l,2-oxidocyclopentane: B.P. 48 oC at 10 mm

Hg; H NMR (60 MHZ, CDC13) 5.16 (bs,lH), 4.85 (bs, 1H)

3.68- 3.40 (m, 2H), 2.23- 1.5 (m, 4H); 49% yield. This

compound was extraordinarily volatile and was not fully

separated from solvent before reaction with the alane.

3-methylene-l,2-oxidocycloheptane: B.P. 300C at 10 mm

Hg; 1H NMR (60 MHZ, CDC13) 5.00 (s, 1H), 4.92 (s,lH),

3.37 (d,lH), 3.10 (t, 1H), 2.4-1.2 (m, 8H); C NMR

(CDC13) 147, 115, 60, 57, 34, 29, 28, 24; 67% yield.

General preparation of methylene hydroxy esters

A 250 ml three necked flask was fitted with a

magnetic stirring bar, thermometer and septum inlet. It

was flame dried, flushed with argon and charged with 3

equivalents diisopropyl amine and hexane. This was then

cooled to -780C and 3 equivalents n-butyllithium, 2.5 M

in hexane were added. The solution was stirred from 15

minutes to an hour after which 3 equivalents of t-butyl

acetate were added dropwise, keeping the temperature

between -75 to -700C. After stirring for 30 minutes,

the solution was allowed to come up to 0C at which

time the volatiles were removed in vacuo. This was

followed by addition of THF and subsequent cooling to

-780C. The solution was then charged with 3 equivalents

of diethylaluminum chloride, 1.8 M in toluene. After a

few minutes, 1 equivalent of the methylene epoxide was

added in 5 ml THF. It was allowed to react at

approximately -500C until reaction was complete.

Quenching was best achieved by cannula transfer of

solution to an erlenmeyer flask containing ice and 1.2

M HC1. The mixture was vigorously stirred by means of a

mechanical stirrer. This was followed by extraction

with ether. The extracts were dried over MgSO4 and

evaporated to give the crude product. Flash

chromatography on silica gel eluting with ethyl

acetate/hexane (or pentane) afforded the desired

hydroxy esters.


pentanol: 1H NMR (60 MHz, CDC13 ) 4.93 4.92 (q, 2H),

3.70 (s, 1H), 2.70 1.70 (m, 8H), 1.40 (s, 9 H) ppm;
3C NMR (CDC13 ) 173.4, 151.9, 106.6, 81.1, 78.1,

48.0, 38.5, 32.3, 29.4, 28.0 ppm; mass spectrum (70 eV)

m/e 139 (28.6), 138 (84.7), 110 (14.7), 93 (63.3), 57

(100), 41 (33.7).

Reaction of 3,4-epoxy-cyclopent-l-ene30 with Rathke


Procedure as described in general procedure.

Isolated yield 50%, after flash chromatography with 40%

ethyl acetate/ pentane.

2-(methylenecarbo-t-butoxy)-3-cyclopentene-l-ol (57):
-1 1
IR (CC14) 3400, 1710, 1150 cm ; HNMR (NT-300, CDC13)

5.68 (m, 1H), 5.51 (m, 1H), 4.18 (m, 1H), 3.06 (bs,

1H), 2.87 (m, 1H), 2.78-2.15 (m, 4H), 1.45 (s, 9H);

3CNMR (CDC13) 173.1, 131.4, 129.0, 80.9, 78.3, 50.7,

40.8, 39.3, 28.0 ppm; Mass spectrum (70 eV) m/e 142

(7%), 124 (50%), 107 (16%), 96 (19%), 83 (23%), 79

(23%), 57 (100%), 41 (41%).

Reaction of 3,4-epoxy-cyclohex-l-ene31 with Rathke


As in general procedure for alane. Isolated yield

after flash chromatography 84%. IR (neat) 3430, 1725,

1140 cm ; HNMR (NT-300,CDC13) 5.65 (m, 1H), 5.43 (d,

1H), 3.55 (m, 1H), 3.19 (bs, 1H), 2.58 (d, 1H), 2.53

(bs, 1H), 2.23 (sextet, 1H), 2.12 (m, 2H), 1.92 (m,

1H), 1.70-1.57 (m, 1H), 1.45 (s, 9H) ppm; 1CNMR

(CDCl3) 172.9, 128.0, 127.3, 80.7, 71.6, 41.1, 39.9,

30.4, 28.0, 24.2 ppm; Mass spectrum (70 eV) m/e 138

(16%), 112 (5%), 79 (13%), 74 (13%), 59 (27%), 57(31%),

45 (18%), 31 (45%), 28 (100%).

Peterson Olefination of 2,3-epoxy cyclopentanone

To 1.50 ml diisopropyl amine (10.7 mmol) in THF,

under argon, 6.40 ml of n-butyl lithium, 1.6 M in

hexane (10.7 mmol) were added slowly at -150C. The

solution was cooled to -780C, followed by addition of

ethyltrimethylsilylacetate, 1.90 ml (10.4 mmol). The

reaction was stirred for 30 min. at -780C. 1.00 g of

2,3-epoxy cyclopentanone in THF was added (10.2 mmol).

Allowed to react for 2 hrs. at -780C, then allowed to

come to room temperature. The orange solution was

quenched with 25 ml of saturated ammonium chloride and

extracted with ether. Volatiles were stripped under

vacuo. Kugelrohr distillation afforded 1.37 g of the

Z- and E-isomers 86Z and 86E (80% yield) in a ratio of

7:4 (Z:E),b.p. 72C at 0.5mm Hg. IR (neat) 2960 (m),
-1 1
1715 (s), 1653 (m), 1222 (s), 1135 (s) cm ; H NMR

(100 MHZ, CDC13) Z-isomer 5.95 (bs, 1H), 4.82 (d, 1H),

4.24 (q, 2H), 3.82 (m, 1H), 2.56-1.83 (m, 4H), 1.30 (t,

3H); E-isomer 5.96 (bs, 1H), 4.21 (q, 2H), 3.69 (d,

1H), 3.02 (bs,lH), 2.56-1.83 (m, 4H), 1.29 (t, 3H); 13C

NMR (CDC13) Z-isomer 165.7, 159.9, 117.9, 59.9, 59.6,

54.2, 27.5, 25.4, 14.1; E-isomer 165.7, 160.4, 116.3,

60.1, 59.8, 54.2, 26.5, 24.9, 14.1; mass spectrum (70

eV) m/e 168 (0.81), 140 (23), 123 (35), 112 (100), 97

(26), 67 (45), 55 (40), 41 (50), 39 (77), 29 (46), 27


Preparation of epoxy allylic alcohols (51Z and 51E)

A three-necked flask fitted with vacuum adapter,

stirring bar and thermometer was flame dried under

vacuum. After cooling to room temperature the flask was

charged with 5.05 g (30.1 mm) of the ester epoxides 86E

and 862 and 100 ml of dry THF. The solution was cooled

to -780C. 60.2 ml of a 1.0 M DIBAL solution was added

slowly. The reaction mixture was stirred at -780C for 1

hour then at -500C for 2 hours after which it was

quenched with 25 ml of methanol. The gelatinous product

was filtered through Celite us-ing hot methanol.

Evaporation of solvent under vacuum afforded 3.79 g of

E and Z epoxy allylic alcohols 51Z and 51E,

(quantitative yield). 1H NMR (300 MHZ, CDC13) Z-isomer

5.76 (m, 1H), 4.32 (d, 2H), 4.20 (m, 1H), 3.92 (d, 1H),

3.76 (dd, 1H), 2.40- 1.60 (m, 4H); E-isomer 5.9 (m,lH),

4.15 (dxAB, 2H), 4.10 (m, 1H), 3.64 (dd, 1H), 3.73 (dd,

1H), 2.40-1.60 (m, 4H); 13C NMR (CDC 3) Z-isomer

142.0, 126.2, 59.9, 58.7, 54.3, 26.5, 22.1; E-isomer

141.7, 125.5, 59.8, 58.5, 54.3, 25.9, 22.1; High

resolution mass spectrum calculated for C7 H1002

126.0680, found: 126.0683.

Protection of the epoxy allylic alcohols as the MTM

ethers (52Z and 52E)

NaH (2.56 g, 60% oil dispersion) were added to a

flame dried flask under argon and washed with hexane,

3x10 ml. The flask was then charged with 100 ml of THF

and cooled to approximately 0C before adding 3.50 g of

the epoxy allylic alcohols (27.8 mm) in 5 ml THF; 2.38

ml of methylthiomethyl chloride were added at

approximately 50C and allowed to react for 6 hrs. The

mixture was quenched at room temperature with 25 ml

saturated NaHCO3, extracted with ether and washed with

brine and more saturated sodium bicarbonate. The

organic extracts were dried over Na2SO4. The solvent

was evaporated under vacuum and the crude product

purified by passing through a sintered glass funnel

filled with silica gel using gradient elution

increasing from 10% ethyl acetate/pentane to 50%. 4.17

g (81% yield) of MTM-ethers 52Z and 52E were recovered.

H NMR (300 MHZ, CDC13) Z-isomer 5.68 (t, 1H), 4.67 (s,

2H), 4.24 (dd, 1H), 4.14 (dd, 1H), 3.8 (d, 1H), 3.68

(m, 1H), 2.36-1.50 (m, 4H), 2.10 (s, 3H); E-isomer

5.86 (t, 1H), 4.64 (s, 2H), 4.12 (dd, 1H), 4.07 (dd,

1H), 3.72 (m, 1H), 3.63 (d, 1H), 2.36-1.50 (m, 4H),

2.15 (s, 3H); 13C NMR (CDC13) Z-isomer 144.1, 122.4,

73.9, 63.8, 59.5, 54.1, 26.5, 25.9, 13.6; E-isomer

143.4, 121.8, 73.9, 63.9, 59.2, 58.4, 25.5, 22.1, 13.6;

Mass spectrum (70 eV) m/e 186 (0.06%), 138 (15%), 125,

109 (54%), 108 (26%), 81 (100%), 79 (70%), 67 (20%), 61

(100%), 53 (41%), 41 (51%).

Preparation of Z- and E-hydroxy esters (59)

Procedure as in general preparation of hydroxy

esters. To 0.79 g of LDA prepared in hexane, 0.80 g of

t-butylacetate was added. After formation of Rathke's

salt, 3.67 ml Of 1.8M solution of Et2AlCl in toluene

were charged into the flask followed by 0.412 g of

oxirane 52. The final product was a pale yellow oil,

0.6237 g of the hydroxy ester 59 (94% yield) after

flash chromatography using 10% acetone/CH2Cl2 1H NMR

(300 MHZ, CDC13) Z-isomer 5.52 (t, 1H), 4.63 (m, 2H),

4.07 (m, 2H), 2.89 (m, 1H), 2.69-1.98 (m, 6H), 2.21 (s,

3H), 1.70-1.63 (m, 2H), 1.47 (s, 9H); E-isomer 5.35 (m,

1H), 4.62 (m, 2H), 3.97-3.86 (m, 2H), 2.80 (m, 2H),

2.69-1.98 (m, 6H), 2.15 (s, 3H), 1.64 (m, 2H), 1.43 (s,

9H); 13C NMR (CDC13) Z-isomer 172.6, 148.8, 118.7,

81.3, 78.8, 74.6, 64.7, 46.0, 39.6, 32.5, 30.9, 28.0,

14.0; E-isomer 173.4, 147.5, 117.8, 81.3, 77.9, 74.2,

64.5, 48.2, 38.8, 32.4, 28.0, 26.0, 14.0 ppm.

Reaction of Rathke Alane with 3,4-epoxy-3-methyl-


A dry three necked flask fitted with vacuum and

thermometer adapters as well as a stirring bar was

charged with 5.02 ml of diisopropyl amine (0.036m)

under argon. The flask was cooled to -78 oC after the

addition of 20 ml of hexane. To this solution 14.4 ml

of a 2.5 M solution of n-butyllithium in hexane (0.036

m) were added and the reaction mixture allowed to stir

for 20 min. at -780C; 4.8 ml of t-butyl acetate were

added dropwise and again allowed to stir for

approximately 20 min after which the solution was

brought to 00C and the solvents removed under vacuum.

After Rathke's salt was allow to dry at room

temperature under vacuum, 35 ml of THF were added and

the solution cooled to -780C. After cooling 20.0 ml of

a 1.8 M solution of Et2AlCl in toluene (0.036 m) were

added followed after 5 min with 1.0 g of

3,4-epoxy-l-butene (0.0119 m). After 2 hours at -600C,

the reaction was quenched by transferring the solution

via cannula to an erlenmeyer flask containing ice and

30 ml 1.2 M HC1. The mixture was stirred vigorously

with a magnetic stirring bar to avoid formation of

gels. Extracted with ether (3 x 30 ml), washed the

organic extracts with brine and dried over Na2SO4'

After filtration and removal of solvents, flash

chromatography using 20% ethyl acetate/hexane furnished

two regioisomers in the ratio of 7:1. Yield 63%. The

major isomer was 3-(hydroxymethyl)-3-methyl-t-butyl-

4-pentenoate (69) and the minor isomer was 4-hydroxy-


3-(hydroxymethyl)-3-methyl-t-butyl-4-pentenoate (69):

IR (neat) both isomers 3440 (s), 1720 (s), 1170 (s)
-1 1
cm ; HNMR (200MHz, CDC13) 5.85 (dd, 1H), 5.15-5.03

(m,2H), 3.48 (d,2H), 2.33 (d,3H), 1.46 (s,9H), 1.11

(s,3H); CNMR (CDC13) 171.7, 143.0, 113.6, 80.8, 69.5,

43.1, 41.5, 28.0, 21.2 ppm; Mass spectrum (70eV) m/e

145 (7%), 127 (27%), 114 (69%), 96 (14%), 85 (25%), 71

(43%), 57 (100%), 43 (25%), 41 (31%).


HNMR (200 MHz, CDC13) 5.79 (dd, 1H), 5.18 (dd, 1H, J=

1.2 Hz and 17 Hz), 5.01 (dd, 1H, J= 1.2 Hz and 11 Hz),

2.24 (m, 2H), 1.75 (m, 3H), 1.38 (s, 9H), 1.22 (s, 3H);
1CNMR (CDC13) 173.8, 144.3, 112.3, 80.4, 72.5, 36.5,

30.4, 28.3, 28.0 ppm; Mass spectrum (70 eV) m/e 182

(.07%), 144 (10%), 126 (71%), 111 (26%), 85 (35%),

71(26%), 57 (100%), 41 (30%).

Reaction of Rathke alane with butadiene monoxide

A flame dried three necked flask fitted with

magnetic stirring bar, vacuum and thermometer adapter

under argon was charged with 1.80 ml diisopropyl amine

(12.8 mm) and 15 ml of hexane. The solution was cooled

to -780C and then 8.60 ml of 1.5 M solution of

n-butyllithium in hexane were added and stirred for 30

min 1.72 ml of t-butyl acetate (12.8 mm) were added

dropwise at -78 C. The reaction mixture was stirred for

another 30 min before allowing the solution to come to

0C and the solvents evaporated under vacuum. After the

white Rathke salt was dry, 20 ml of THF were charged

into the flask and cooled to -780C At this

temperature 7.20 ml of a 1.8 M solution of Et2AlCl

(12.96 mm)in toluene were carefully added followed by

0.326 g of epoxide (0.428 mm). The solution was allowed

to react at -50C for 1 hr. It was quenched with 15 ml

of 3% HC1 and extracted with ether. The organic

extracts were washed with brine and dried over MgSO4.

Two regioisomers were formed in a ratio of 3:1. They

were purified by flash chromatography using 30% ethyl


Yield: 22% The major one being 3-(methylenehydroxy)-

t-butyl-4-pentenoate (66) and the minor product was

4-hydroxy-t-butyl-5-hexeneoate (67).

3-(hydroxymethyl)-t-butyl-4-pentenoate (66):

IR (neat) both isomers 3415 (bs); 1723 (s); 1150 (s)

cm ; HNMR (300 MHZ, CDC3 ) 5.78-5.66 (m, 1H), 5.16

(t, 2H), 3.57 (d,2H), 2.72 (m, 1H), 2.46-2.24(dAB

pattern, 3H), 1.45 (s, 9H); CNMR (CDC1 ) 171.9,

137.8, 116.9, 80.7, 65.2, 42.8, 37.4, 28.0 ppm; Mass

spectrum (70eV) m/e 113(12%), 100(30%), 71(22%),

57(100%), 54(16%), 43(16%), 41(38%), 29(25%).

4-hydroxy-t-butyl-5-hexeneoate (67):

IR (neat) both isomers 3415 (bs); 1723 (s); 1150 (s)
-1 1
cm1 ; HNMR (60 MHZ, CDC1 ) 5.9 (M,1H), 5.3 (M, 2H),

4.25 (bs,lH), 2.6-1.0 (m. 5H), 1.5 (s,9H) ppm; 13CNMR

(CDC13) 173.3, 140.5, 114.8, 80.4, 72.1, 31.8, 31.5,

28.0 ppm; Mass spectrum (70 eV) m/e 130 (3%), 112(37%),

71 (30%), 57(100%), 41(34%), 28(10%).

Reaction of styrene oxide with Rathke alane

Procedure as in general procedure for alane

reaction. Yield 64%. Two products isolated by flash

chromatography using 15% ethyl- acetate/pentane.The

ratio of regioisomers was 4:1. The major product was

4-hydroxy-3-phenyl-t-butyl-butanoate (76a) and the

minor 4-hydroxy-4-phenyl-t-butyl-butanoate (76b).

4-hydroxy-3-phenyl-t-butyl-butanoate (72): White solid,

melting point 44-480C. IR (neat) both isomers 3010 (s),

1712 (s), 1368 (s), 1212 (s) cm ; HNMR (XL-200,CDC13)

7.28 (m, 5H), 3.78 (d, 2H), 3.30 (m, 1H), 2.65 (dAB,

2H), 2.63 (bs,lH), 1.35 (s, 9H); 1CNMR (CDC13) 171.9,

141.1, 128.5, 127.8, 126.9, 80.6, 66.9, 44.7, 38.6,

27.9; Analysis: calculated 71.16% C, 8.53% H Found

71.10% C, 8.57% H.

4-hydroxy-4-phenyl-t-butyl-butanoate (73):

IR (neat) both isomers 3010 (s), 1712 (s), 1368 (s),
-1212 (s) cm; R (300 MHz,CDC) 7.34 (m, 5H), 4.73
1212 (s) cm ; HNMR (300 MHz,CDC13) 7.34 (m, 5H), 4.73

(t, 1H), 2.58 (bs, 1H), 2.33 (t, 2H), 2.02 (q, 2H),

1.44 (s, 9H) ppm; 13CNMR (CDC13) 173.3, 144.2, 128.4,

127.5, 125.7, 80.5, 73.6, 34.0, 31.9, 28.1 ppm; Mass

spectrum (70 eV) m/e 180 (15%), 162 (21%), 161 (40%),

117 (40%), 107 (67%), 105 (39%), 91 (28%), 79 (34%), 77

(38%), 57 (100%), 41 (66%), 28 (50%).

Preparation of 3,7-dimethyl-2,3-oxo-6-octen-l-al
Molecular sieves (2.76 g, 4A) were ground and

activated by flame drying under vaccum in the reaction

vessel. After cooling to room temperature and flushing

with argon, 24.37 g of PDC (0.065 m) were added under

an argon atmosphere; 30 ml of CH2Cl2 were charged into

the flask, cooled with an ice bath, and 5.23 g of

3,7-dimethyl-2,3-oxo-6-octen-1-o133 (0.0231 m) were

added in 20 ml of CH2C12. The mixture was allowed to

react overnight. The crude mixture was filtered through

Celite followed by filtering through a plug of SIO2

with 10% MgSO4 using CH2C12. Kugelrohr distillation of

the crude afforded 1.23 g of pure aldehyde (32% yield).
-1 1
IR (CC14) 3020-2840 (s), 1710 (s) cm ; H NMR (300

MHz,C6D6) 9.20 (s, 1H), 4.97 (bs, 1H), 2.90 (s,lH),

2.25-1.0 (m, 4H), 1.61 (s, 3H), 2.4 (s, 3H), 1.45 (s,

3H); 13C NMR (C6D6) 198.6, 132.2, 123.3, 63.4, 38.4,

25.7, 23.7, 17.6, 17.0 ppm.

Wittig reaction of 3,7-dimethyl-2,3-oxo-6-octen-l-al

A 150 three necked flask was dried in a vacuum

oven overnight with 3.10 g of methyltriphenylphospho-

nium bromide (8.70 mm) at 560C. The flask was then

fitted with vacuum and thermometer adapters as well as

a stirrer. It was evacuated and flushed with argon

several times. 50 ml of dry THF were charged into the

flask and the mixture chilled to -500C. At this

temperature, 3.56 ml of n-butyllithium 2.5 M in hexane

were added (8.70 mm). The ylide color was bright yellow

after approximately 30 min, 1.49 g of the aldehyde were

added and the solution allowed- to come to room

temperature. After 3 hrs, it was quenched with 50 ml of

water, extracted with ether and dried over MgSO4 after

washing with brine. The solvents were evaporated under

vacuum at which point a large amount of

triphenylphosphine oxide crystallized. The liquid

remaining was decanted and the salts washed with

pentane. The organic liquids were combined, and the

solvents evaporated. Kugelrohr distillation afforded

0.823 g of the vinyl epoxide 74 (74% yield). IR (CC14)
-1 1
3059 (s), 1634 (s), 1971 (s) cm ; H (300 MHz, C6D6

5.56-5.49 (m, 1H), 5.18 (d, 1H); 4.98 (m, 2H), 3.09 (d,

1H), 2.07-1.99 (dd, 2H), 1.55-1.24 (m, 2H), 1.52 (s,

3H), 1.38 (s, 3H) 1.02 (s, 3H); 13C NMR 134.6, 131.6,

124.3, 119.0, 63.1, 69.1, 33.8, 25.7, 24.1, 17.6, 16.7;

High resolution mass spectrum calculated for C11HI80,

166.1362, found: 166.1357.

Reaction of Rathke alane with vinyl epoxide 74

Compound 74 was reacted with 3 equivalents of

alane as described in the general procedure for alane

reactions After three hours and allowing reaction

temperatures to rise from -60 to -100C. No product

arising from reaction with the epoxide was seen. In

order to corroborate lack of reactivity 3 more

equivalents of the alane reagent were quickly in a

separate reaction vessel and cannula transferred to the

first. It was allowed to react overnight at room

temperature. After the usual workup, the vinyl oxirane

was recovered unchanged.

Preparation of l,l-cyclohexyl-2-vinyl-oxirane (75)

A three necked flask fitted with vacuum adapter,

stirring bar and thermometer was flame dried under

vacuum and 19.6 g (0.11 moles) of allyldimethyl

sulfonium bromide over an atmosphere of argon. The

sulfonium salt was prepared by stirring a mixture of

dimethylsulfide and allylbromide overnight. THF, 100

ml, was charged into the flask and cooled to 00C; 44 ml

of 2.5 M n-butyl lithium (0.11 moles) were added and

stirred for 30 min. The color of the ylide was deep

red. When 1.8 g (0.018 moles) of cyclohexanone were

injected into the flask, the color turned from red to

faint yellow after addition. The reaction mixture was

allowed to come to room temperature and react for 2

hrs. It was quenched with 75 ml of ice water; extracted

with ether and dried over Na2SO4. Yield was 57% by G.C.

Short path distillation, 25 mm Hg at 65-700C, afforded

the oxirane contaminated with cyclohexanone. Washing

with 10% NaHSO3 and passing through a small plug of

SiO2 using 20% ether/pentane gave the pure oxirane. IR

(neat) 3100- 2860 (s), 1640 (m), 1447 (s) cm 1; H NMR

(300 MHz, CDCL3) 5.82-5.71 (m, 1H), 5.47 (dd, 1H, J=

17, 2Hz), 5.31 (dd, 1H, J= 11, 2 Hz), 3.18 (d, 1H, J= 5

Hz), 1.77- 1.68 (m, 2H) 1.60-1.45 (m,8H); 13C NMR

(CDC13) 133.2, 119.8, 64.7, 64.3, 35.4, 29.3, 25.5,

25.0, 24.7 ppm.

Synthesis of l,l-dimethyl-2-vinyl-oxirane (76)

To a flame dried flask fitted with vacuum adapter

and septa, 5.16 g of Na2CO3 ( 48.7 mm) were added.

MCPBA (7.89 g, 45.7 mm) were added under an atmosphere

of argon as well as 14 ml of ether (dry). The flask was

cooled to 0C and 1.4 ml (12 mm) of 2-methyl-2,4-

dimethyl-pentadiene were charged into the flask. The

mixture was allowed to react for 45 min. The reaction

mixture was filtered to remove solids and most of the

ether distilled through a vigreaux column. Vacuum

transfer of the remaining residue afforded 9 mg (0.092

mm) of epoxide 94% pure which would be used directly in

reaction with the Rathke alane.

H NMR (300 MHz, CDCL3) 5.64 (m, 1H), 5.38 (dd, 1H, J=

17, 2 Hz), 5.27 (dd, 1H, J= 11 Hz,2 Hz), 3.14 (d, 1H,

J= 5 Hz), 1.28 (s, 3H), 1.20 (s, 3H); 1C NMR (CDC13)

125.9, 120.0, 64.4, 60.2, 30.9, 24.6 ppm.

Reaction of 1,l-dimethyl-2-vinyl-oxirane with Rathke


A 0.3 mm solution of the Rathke alane was prepared

in 8 ml of THF following the general procedure for the

alane formation. An ether solution of the oxirane 76

along with a C16 standard were added at -600C. The

reaction was monitored for 50 minutes no observable

product arising from reaction with the epoxide 76 was

seen. In order to test the formation of the alane 0.01

ml of methylene cyclohexane oxide was added to react

with the alane. Conversion of 53 to the hydroxy ester

60 was evidenced by G.C. monitoring within 20 min.

Oxidation of hydroxy ester (60) to enone (80)

Hydroxy ester 60 (0.632 g) was oxidized using
1.789 g of ground 3A molecular sieves, 3.764 g of PDC

and a catalytic amount of dry acetic acid in methylene

chloride. The procedure has been previously described

on page 66. After 3 hrs, the slurry was filtered

through a sintered glass funnel filled with a mixture

of SiO2 and 10% MgSO4, eluting with ether. This gave

0.516 g of enone 80 (82% yield). HNMR (300 MHz, CDC13)

3.28 (s, 2H), 2.40 (m, 4H), 1.98 (m, 2H), 1.94 (s, 3H),

1.43 (s, 9H); 13C 199.8, 170.6, 157.9, 129.8, 80.4,

37.1, 32.7, 31.8, 27.9, 22.0, 21.5.

PDC oxidation of methylene hydroxy ester (56)

Compound 56 (58.8 mg) was oxidized with 0.506 g

PDC, 0.722 g molecular sieves, a catalytic amount of

acetic acid and 10 ml of methylene chloride as

previously described. It was allowed to react

overnight. Filtration through SiO2 and 10% MgSO4 gave

the crude enone 79 which was purified by using 30%

ethyl acetate/pentane (63% yield). 1HNMR(60 MHz, CDC13)

3.16 (s, 2H), 2.80-2.20 (m, 4H), 2.09 (s, 3H), 1.47 (S,

9H); CNMR (CDC13) 208.0, 172.8, 169.5, 149.8, 80.9,

34.1, 31.8, 29.6, 28.0, 17.5 ppm.

Oxidation of cyclopentene hydroxy ester (57)

The alcohol, 0.149 g (0.753 mm), was oxidized with
the following: 0.73 g of ground molecular sieves, 3A,

1.500 g of PDC and a catalytic amount of acetic acid in

methylene chloride. After reacting for 1 hr the slurry

was filtered through SiO2 and 10% MgSO4, to give 0.106

g of 2-(methylenecarbo-t-butoxy)-3-cyclopenten-l-one

(77) (84% yield). 1HNMR (100 MHz, CDC13) 6.11 (m, 2H),

3.24-3.10 (m, 1H), 2.93 (s, 2H), 2,61-2.30 (m, 2H),

1.43 (s, 9H); 1CNMR (CDCl3) 217.0, 170.5, 132.2,

128.4, 80.9, 48.5, 42.4, 36.0, 28.0 ppm.

Oxidation of cyclohexene hydroxy ester (61).
Finely ground molecular sieves (0.786 g, 3A) were

activated by heating vigorously for 5 min. in a 25 ml

flask under vacuum After allowing to come to room

temperature 1.18 g of PDC (3.14 mm) were added along

with a stirring bar and and 12 ml of freshly distilled

methylene chloride. The mixture was placed under argon,

and cooled with ice water. At this time the hydroxy

ester, 0.224 g (1.06 mm), was charged into the flask

via syringe followed by a catalytic amount of dry

acetic acid. The slurry turned dark brown upon addition

of the alcohol. After 1 hour, the mixture was filtered

through 10% MgSO4/SiO2 and the solids washed with

ether. The yield of 2-(methylenecarbo-t-butoxy)-3-

cyclohexen-l-one (78) was 97% (0.2163 g). IR (neat)
-1 1
1720, 1148 cm ; HNMR (100 MHz, CDC13) 6.0-5.5 (m,

2H), 3.4 (bs, 1H), 2.8-2.2 (m, 6H), 1.4 (s, 9H); 13CNMR

(CDC13) 209.7, 171.2, 128.4, 127.5, 80.7, 45.1, 38.1,

36.3, 28.1, 26.4 ppm.

Hydrolysis of t-butyl ester (61) to the acid (90)

To a 100 ml flask were added the following, 2.0 g

of hydroxy ester 61 (0.013 m), 40 ml of methanol and

1.79 g of K2CO3. The mixture was stirred for 2 days

after which it was acidified with 6M HC1 to a pH of 2,

extracted with ether (4x35 ml) and dried over MgSO4.

Evaporation of solvent gave 1.4821 g of 90 in 73%

yield. IR (neat) 3680-2800 (s), 1710 (s), 910 (m) cm-1
HNMR (200 MHz, CDC13) 6.25 (bs, 1H), 5.64 (m, 1H),

5.42 (d, 1H), 3.60 (m, 1H), 2.66-1.10 (m, 7H); CNMR

(CDC13) 177.7, 127.7, 127.5, 71.8, 40.6, 38.4, 30.2,

24.2 ppm; Mass spectrum (70 eV) m/e 138 (3), 110 (2),

84 (12), 74 (47), 59 (80), 45 (51), 31 (100), 28 (69).

Lactonization of cyclohexene hydroxy acid (90)

A 25 ml three necked flask was fitted with

thermometer, stirring bar, and vacuum adapter. It was

flame dried under vacuum and flushed with argon.

Triphenyl phosphine (0.5128 g) were added under a

positive pressure of argon followed by 20 ml of dry

THF. The hydroxy acid (0.210g) dissolved in THF was

charged into the flask and the reaction mixture cooled

to 10C. At this point, 0.29 ml of DEAD reagent was

added and a catalytic amount of acetic acid. After

reaction was complete, the solvent was evaporated in

vacuo and the crude cis lactone was purified by flash

chromatography using 40% ethyl acetate/pentane to give

0.140 g of lactone 91 in 78% yield. IR (neat) 3011,
1762, 1273 cm 1 R (CDC XL-200) 5.90 (m, H),
1762, 1273 cm ; HNIIR (CDC13, XL-200) 5.90 (m, 1H),

5.48 (d, 1H), 4.76 (m, 1H), 3.02 (m, 1H), 2,76 (AB

quartet, 1H), 2.30 (dd, 1H), 2.20- 1.10 (m, 4H); 1CNMR

(CDC13) 176.8, 128.7, 125.6, 78.1, 35.8, 34.4, 24.6,

19.1 ppm; High resolution mass spectrum calculated:

138.0680, found: 138.0685.

Preparation of MTM-hydroxy acid (87)

To a 50 ml one neck flask were added the

following: 1.03 g of hydroxy ester 87 (3.41 mm), 15 ml

of absolute ethanol, 1 ml of water, and 0.660 g of KOH

(11.8 mm). The mixture was stirred and heated at 43-47

C under argon for 13 hrs. The ethanol mixture was

concentrated in the rotovap. The slurry was acidified

with 1.2 M HC1 to a pH of 2, extracted with ether (4 X

20 ml) and dried over MgSO4. This afforded 0.6842 g of

a brown, very viscous oil (82% yield) which was

lactonized without further purification. IR (neat)

3700-2800 (s), 1715 (s), 1430 (w), 1245 (w), 1050 (s)

Preparation of cis-MTM-lactone (88)

A 25 ml three necked flask was fitted with

thermometer, stirring bar and vacuum adapter. It was

flame dried and flushed with argon. Triphenylphosphine,

0.197 g (0.730 mm), was added under an argon

atmosphere; 15 ml of dry THF were charged into the

flask followed by 110 mg of the hydroxy acid 87

(0.447mm) in THF. The mixture was cooled to 100C.

Dropwise addition of the DEAD reagent, 0.11 ml (0.699

mm), followed and allowed to react for 1 hr. The

solvent was evaporated and flash chromatography using

35% ethyl acetate/pentane gave the E and Z MTM lactones

(90% yield by G.C.). IR (CC14) 3300-2810 (m),1775 (s),
-1 1
1165 (m) cm ; H NMR (300 MHZ, CDCL3) Z-isomer 5.58

(t, 1H), 5.09 (t, 1H), 4.63 (AB quartet, 2H), 4.04 (d,

2H), 3.51 (m, 1H), 3.02-2.92 (dd, 1H, J= 18, 10 Hz),

2.65-2.48 (m, 1H), 2.43-2.40 (dd, 1H, J= 18, 2.5 Hz),

2.44-2.35 (m, 1H), 2.22-2.18 Cm, 1H), 2.16 (s, 3H),

1.85-1.72 (m, 1H); E-isomer 5.57- 5.50 (m, 1H), 5.01

(t, 1H), 4.63 (s, 2H), 4.17-4.01 (m, 2H), 3.34 (m, 1H),

2.98-2.89 (dd, 1H, J=18,9 Hz), 2.63-2.55 (m, 1H),

2.54-2.47 (dd, 1H, J=18, 1.6 Hz), 2.43 (m, 1H), 2.31-

2.22 (m, 1H), 2.15 (s, 3H), 1.94-1.81 (m, 1H); 1C NMR

(CDC 3) Z-isomer 176.7, 148.3, 121.1, 85.4, 74.6, 65.0,

44.3, 37.3, 31.4, 26.6, 14.0; E-isomer 176.6, 148.5,

119.6, 85.7, 74.6, 64.3. 40.25, 36.3, 31.5, 30.8, 14.0;

High resolution mass spectrum calculated: 228.0820:

found 228.0853.



2-cyclopentenone- Aldrich

DIBAl- Aldrich

n-butyllithium- Aldrich

diethyl aluminum chloride- Aldrich

PDC- Aldrich

MCPBA- Aldrich

triphenylphosphine- Aldrich

MTM chloride- Aldrich

Ethyltrimethylsilylacetate- Aldrich

t-butyl acetate- Aldrich

sodium hydride- Alfa

Aldrich Chemical Co.
P.O. Box 355,
Milwaukee, Wisconsin 53201

Alfa Products- Thiokol/Ventron Division
P.O. Box 299
152 Andover Street
Danvers, Massachusetts 01923


1. G. Zweifel and J.A. Miller, "Synthesis Using
Alkyne-Derived Alkenyl- and Alkynylaluminum
Compounds", Organic Reactions, Vol. 32, John Wiley
and Sons, New York, 1984, Chapter 2.

2. K. Ziegler, Experentia Suppl. II, 278, (1955).

3. (a) A. J. Lundeen and A.C. Oehlschlager, J.
Organometal. Chem., 25, 337 (1970). (b) J. L. Namy,
E. Henry-Basch and P. Freon, C.R. Acad. Sc. Ser. C,
269, 1222, (1969). (c) J. L. Namy, E. Henry-Basch
and P. Freon, Bull. Soc. Chim., 6, 2249 (1970).

4. M. S. Newman and C. A. VanderWerf, J. Am. Chem.
Soc., 67, 233 (1945).

5. J. Fried, C. H. Lin and S.H. Ford, Tetrahedron
Lett., 1379, (1969). J. Fried and J.C. Sih, ibid.,
3899 (1973). J. Fried M. Mehra, C. Lin and W.
Kao, Ann. N.Y. Acad. Sci., 180, 38, (1971).

6. S. Danishevsky and R.K. Singh, J. Org. Chem., 41,
1669, (1976).

7. M. W. Rathke and D. F. Sullivan, J. Am. Chem. Soc.,
95, 3050 (1972).

8. M. Visnick, Ph. D. dissertation, University of
Florida, 1982.

9. G. A. Crosby and R. A. Stephenson, J.C.S. Chem.
Comm., 287 (1975).

10. M.Visnick, Ph. D. dissertation, University of
Florida, page 15, 1982.

11. C. Campbell and M. Battiste unpublished results.

12. Spectroscopic studies done by Jim Rocca, Curt
Campbell and Mapi Cuevas.

13. K. Oshima N. Tsuboniwa, S. Matsubara, Y, Morizawa
and Hitosi Nozaki, Bull. Chem. Soc. Jpn., 57, 3242,
(1984). K. Oshima, N. Tsuboniwa, S. Matsubara, Y,
Morizawa and Hitosi Nozaki, Tetrahedron Letters,
25, 2569 (1984).

14. A. Jeffrey, A. Meister and T. Mole, J. Organometal.
Chem., 74, 365 (1974).

15. F. Orsini, F. Pelizzoni and G. Ricca, Tetrahedron
Letters, 23, 3945 (1982).

16. J. Dekker, J. Boersma and G. J. M. Van der Kerk, J.
Chem Soc. Chem. Commun., 553 (1983).

17. J. Fried et al., J. Am. Chem. Soc., 94, 4342

18. (a) C.R. Johnson, R.W. Herr, J. Am. Chem. Soc., 92,
4979 (1970); (b) J.M. Mas. M. Malacria, and J.
Gore, J. Am. Chem. Soc. Commun., 1161 (1985).

19. Y. Naruta and K. Maruyama, Chemistry Letters, 963

20. J. M. Mas, M. Malacria, and J. Gore, J.
Chem Soc. Chem. Commun., 1161 (1985).

21. M. David, J. Sauleau and A. Sauleau, Can. J. Chem,
63, 2449 (1985).

22. E. J. Corey, Tet. Lett., 4753 (1971)

23. S. Czernecki, C. Georgoulis, C.L. Stevens and K.
Vijayakumaran, Tet. Lett., 26, 1699 (1985).

24. A. J. Birch,K. S. Keogh and V. R. Mamdapur,
Australian Journal of Chemistry, 26, 2671 (1973).

25. 0. Mitsunobu, Synthesis, 16, 1 (1981).

26. R. M. Coates, J. Am. Chem. Soc., 97, 1619 (1975).

27. H. C. Brown, Organic Synthesis via Boranes, John
Wiley and Sons, New York, 1975.

28. U.S. Patent 4, 31, 129.

29. W.C. Still, J. Org. Chem., 43, 2923 (1978).


30. J. K. Crandall, D. B. Banks, R. A. Colyer, R. J.
Watkins and J. P. Arrington, J. Org. Chem., 33, 423

31. J. K. Crandall, D. B. Banks, R. A. Colyer, R. J.
Watkins and J. P. Arrington, J. Org. Chem., 33, 423

32. E.J. Reist, Irene G. Junga and B.R. Parker, J. Org.
Chem., 25, 1673 (1960).

33. K.B. Sharpless And T. Katsuki, J. Am. Chem. Soc.,
102, 5974 (1980).


Mapi Cuevas was born on March 18, 1952, in Rio

Piedras, Puerto Rico. She received a B.S. in chemistry,

magna cum laude, from the University of Puerto Rico in

May 1973 after three years of undergraduate studies.

From August 1973 to June 1975 she taught science and

mathematics at Baldwin High School, Guaynabo, Puerto

Rico and was named Science Chairperson from 1974-1975.

In the summer of 1975, after moving to Gainesville, she

worked for three years at Golden Hills Academy. A

career move to North Fort Myers to teach at the local

high school followed. Ms. Cuevas was science instructor

at North Fort Myers High school until 1983 when she was

admitted to graduate studies at the University of

Florida. From 1983 to 1986 she was awarded a GPOP

Fellowship by the Graduate School. From 1987 to 1988

Mapi co-authored a physical science textbook for

Harcourt Brace Jovanovich and was curriculum

coordinator of the undergraduate organic laboratory at

UF for the 1986-87 school year. Most recently she is an

adjunct instructor of chemistry at Santa Fe Community


I certify that I have read this study and that in
my opinion it conforms to acceptable standards of
scholarly presentation and is fully adequate, in scope
and quality, as a dissertation for the degree of Doctor
of Philosophy.

Merle A. Battiste, Chairman
Professor of Chemistry

I certify that I have read this study and that in
my opinion it conforms to acceptable standards of
scholarly presentation and is fully adequate, in scope
and quality, as a dissertation for the degree of Doctor
of Philosophy.

William R. Dolbier, Jr /
Professor of Chemistry'-

I certify that I have read this study and that in
my opinion it conforms to acceptable standards of
scholarly presentation and is fully adequate, in scope
and quality, as a dissertation for the degree of Doctor
of Philosophy.

William M. Jones,
Professor of Chemistry

I certify that I have read this study and that in
my opinion it conforms to acceptable standards of
scholarly presentation and is fully adequate, in scope
and quality, as a dissertation for the degree of Doctor
of Philosophy.

J in q. Dorsey
As oc ate Professor of C emi try

I certify that I have read this study and that in
my opinion it conforms to acceptable standards of
scholarly presentation and is fully adequate, in scope
and quality, as a dissertation for the degree of Doctor
of Philosophy.

Mih'a l Kilberg'
Professor of Biochemistr and
Molecular Biology

This dissertation was submitted to the Graduate Faculty
of the Department of Chemistry in the College of
Liberal Arts and Sciences and to the Graduate School,
and was accepted as partial fulfillment of the
requirements for the degree of Doctor of Philosophy.

December 1988

Dean, Graduate School

111 ll II II 1111111111111111 I llll ll l ll II
3 1262 08556 7732