Cycloheptatetraene or cycloheptatrienylidene from dehydrobromination of bromocycloheptatrienes

MISSING IMAGE

Material Information

Title:
Cycloheptatetraene or cycloheptatrienylidene from dehydrobromination of bromocycloheptatrienes
Physical Description:
viii, 72 leaves : ill. ; 28 cm.
Language:
English
Creator:
Harris, James Wesley, 1949-
Publication Date:

Subjects

Subjects / Keywords:
Cycloheptatrienylidene   ( lcsh )
Polymers   ( lcsh )
Polymerization   ( lcsh )
Allene   ( lcsh )
Genre:
bibliography   ( marcgt )
theses   ( marcgt )
non-fiction   ( marcgt )

Notes

Thesis:
Thesis (Ph. D.)--University of Florida, 1983.
Bibliography:
Includes bibliographical references (leaves 70-71).
Statement of Responsibility:
by James Wesley Harris, Jr.
General Note:
Typescript.
General Note:
Vita.

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 000506239
notis - ACS6556
oclc - 12203466
System ID:
AA00003433:00001

Full Text











CYCLOHEPTATETRAENE OR CYCLOHEPTATRIENYLIDENE
FROM DEHYDROBROMINATION
OF BROMOCYCLOHEPTATRIENES





BY

JAMES WESLEY HARRIS, JR.


A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN
PARTIAL FULFILLMENT OF THE REQUIREMENTS
FOR THE DEGREE OF DOCTOR OF PHILOSOPHY





UNIVERSITY OF FLORIDA


1983



















to my wife, Kathy













ACKNOWLEDGMENTS


The author wishes to express his deepest appreciation

to Dr. William M. Jones for providing the origin of this

work. Throughout the years of this work, Dr. Jones

extended the warmest friendship, gave the author the

technical assistance and the encouragement needed when

the task seemed impossible. The author also thanks

Dr. Jones for his constant belief in the author and

the work.

Additionally, the author thanks Dr. J. R. Sabin

for the molecular orbital calculations and the theoretical

ultraviolet spectra. And the author extends his

appreciation to all the members of the Chemistry

Department for their help, friendship, and assistance

through the years.













TABLE OF CONTENTS


PAGE

ACKNOWLEDGEMENTS iii

ABSTRACT vii

INTRODUCTION 1

RESULTS AND DISCUSSION 14

CONCLUSIONS 39

EXPERIMENTAL 42

General 42
Preparation of Tropone 43
Preparation of 2,4,6-Cycloheptatriene
p-Toluenesulfonylhydrazone 44
Preparation of the Sodium Salt of
2,4,6-Cycloheptatriene p-Toluene-
sulfonylhydrazone 45
Preparation of 1-, 2-, and 3-Chloro-
cyclohepatrienes 45
Preparation of the Lithium Salt of
2,4,6-Cycloheptatriene p-Toluene-
sulfonylhydrazone 47
Preparation of the Tetramethylammonium
Salt of 2,4,6-Cycloheptatriene
p-Toluenesulfonylhydrazone 47
Preparation of the Tetra-n-butyl-
ammonium Salt of 2,4,6-Cyclohepta-
triene p-Toluenesulfonylhydrazone 47
Preparation of N-Diphenylaziridine-
phthalimide 48
Preparation of N-Aminodiphenylaziridine 49
Preparation of 1-, 2-, and 3-Bromocyclo-
heptatrienes 49
Preparation of 3-Phenylphthalide 51
Preparation of 1,3-Diphenylisobenzofuran 52
Preparation of 1-Potassium Menthclate 53







Preparation of Lithium Aluminum
Deuteride-Quinine (LAD3-Q) 53
Reduction of 7,7-Dibromocycloheptatriene
with Lithium Aluminum Deuteride-
Quinine (LAD3-Q) 54
Cold Finger Deposit of the Reaction
Product of 1-, 2-, and 3-Chloro-
cycloheptatrienes with Base 54
Attempted Reaction of N-Aminodiphenyl-
aziridine and 7,7-Dichlorocyclo-
heptatriene 55
Attempted Reaction of N-Aminodiphenyl-
aziridine and Tropone 55
Attempted Sublimation of the Lithium
Salt of 2,4,6-Cycloheptatriene
D-Toluenesulfonylhydrazone 56
Attempted Sublimation of the Tetramethyl-
ammonium Salt of 2,4,6-Cyclohepta-
triene p-Toluenesulfonylhydrazone 56
Attempted Sublimation of the Tetra-n-
butylammonium Salt of 2,4,6-Cyclo-
heptatriene p-Toluenesulfonyl-
hydrazone 56
Attempted Matrix Isolation of the Sodium
Salt of 2,4,6-Cycloheptatriene
p-Toluenesulfonylhydrazone in Solid
Adamantane 57
The Matrix Photolysis of the Sodium Salt
of 2,4,6-Cycloheptatriene p-Toluene-
sulfonylhydrazone in 2-Methyltetra-
hydrofuran 57
The EPR-Matrix Photolysis of Some Benz-
annelated Sodium Salts of 2,4,6-Cyclo-
heptatriene p-Toluenesulfonylhydrazone
in 2-Methyltetrahydrofuran 59
The EPR-Matrix Photolysis of the Sodium
Salt of 2,4,6-Cycloheptatriene
p-Toluenesulfonylhydrazone in
2-Methyltetrahydrofuran 59
Reaction of 1-, 2-, and 3-Bromocyclo-
heptatrienes with 1-Potassium
Mentholate in the Presence of
1,3-Diphenylisobenzofuran 60
Reaction of Optically Active Deuterated
1-, 2-, and 3-Bromocycloheptatrienes
with Potassium tert-Butoxide in the
Presence of 1,3-Diphenylisobenzo-
furan 61








Reaction of Optically Active Deuterated
1-, 2-, and 3-Bromocycloheptatrienes
with Potassium tert-Butoxide in the
Presence of 1,3-Diphenylisobenzo-
furan and Toluene 63
Reaction of Optically Active Deuterated
1-, 2-, and 3-Bromocycloheptatrienes
with Potassium tert-Butoxide in the
Presence of 1,3-Diphenylisobenzo-
furan and Styrene 64
Reaction of Optically Active Deuterated
1-, 2-, and 3-Bromocycloheptatrienes
with Potassium tert-Butoxide in the
Presence of Styrene and a Reduced
Amount of 1,3-Diphenylisobenzo-
furan 66
Reaction of Optically Active Deuterated
1-, 2-, and 3-Bromocycloheptatrienes
with Potassium tert-Butoxide in the
Presence of Styrene 67

REFERENCES 70

BIOGRAPHICAL SKETCH 72













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





CYCLOHEPTATETRAENE OF CYCLOHEPTATRIENYLIDENE
FROM DEHYDROBROMINATION
OF BROMOCYCLOHEPTATRIENES

By

JAMES WESLEY HARRIS, JR.

December, 1983





Chairman: William M. Jones

Major Department: Chemistry

Cycloheptatetraene has been generated by-potassium

menthoxide promoted dehydrobromination of a mixture

of bromocycloheptatrienes to give an intermediate which,

when trapped with 1,3-diphenylisobenzofuran, gave an

optically active adduct. The same result was also

obtained when a mixture of optically active 7-deutero-

bromocycloheptatrienes were dehydrobrominated with

potassium tert-butoxide. These results lead to the

unequivocal conclusion that the initially formed inter-

mediate is chiral and is therefore best represented

by an allene structure. An ultraviolet spectrum of what


vii







may be the same intermediate in a 2-methyltetrahydrofuran

matrix is also reported. Evidence for a carbene valence

isomer of the allene as the progenitor of spirocyclo-

propanes is also reported.


viii













INTRODUCTION






For a period of time, the reaction mechanism that

occurred during the thermolysis or photolysis of the

sodium salt of tropone tosylhydrazone (1) was believed

to involve the carbene (2) as the intermediate leading

to the products. In fact, the following reaction

scheme was proposed to summarize the chemistry of the

supposedly carbene intermediate.







I /
w/ \ r^ / -e ia i ~ _


Krl
K^^


7 -
-"a-^


i/







However, it was soon discovered that an abnormality

occurred when a diene was used as the trapping reagent.2

It appeared that a cycloaddition of the diene took

place to yield an unexpected product as shown in the

following reaction.







1 + or
hv

8


The suspicion then arose that the intermediate might

possibly be the allene (9), or more likely, the allene

(9) and carbene (2) may exist in equilibrium.3







1 or
hv


2 9


Later, Untch et al. discovered that when a mixture

of 1-, 2-, and 3-halocycloheptatrienes (10) was treated

with base in the presence of butadiene, the same cyclo-

addition product (8) resulted.











aX +

X=C1
10


/ \ Base


or Br


Additionally, when carbene type trapping reagents were
used, carbene trapped products were obtained.


10 + PhCH=CH2


Base
___ ^-


Or, in the absence of traps, the dimer was isolated.


10 Base


Since a reasonable mechanism for the dehydrohalogenation
reaction is a beta-elimination (or vinylogous beta-

elimination from the 2- and 3-isomers), the observation
that the dehydrohalogenation gave an intermediate that
showed the same properties as those of the intermediate
generated from the tosylhydrazone salt, re-emphasized
the possibility that the common intermediate from the
two different sources might be the allene (9).5










H H
10a Br





la


10b


H H


Br

10c


However, as mentioned above, it also seemed reasonable

to represent the intermediate as an interconverting

mixture of the two valence isomers since the two




5

starting compounds (1 and 10), the tosylhydrazone
and halocycloheptatriene, would be expected to give
the different intermediates 9 and 2.







00

2 9


Assuming that both 9 and 2 represent energy
minima, the objective of the first portion of this
work was to determine which intermediate is the ground

state. This had been attempted some years earlier
through theoretical calculations.6 The total energies
of both intermediates were calculated and it was

found that the allene (9) should be the more stable
of the two intermediates by approximately 14 kilo-
calories/mole. These calculations did not explore
the energy surface around the carbene to see if it
represents an energy minima.

Chemically, the viability of the allene intermedicity
had been approached by investigating the mechanism of
the dehydrohalogenation. In principle, the mechanism
of the dehydrohalogenation of, for example 10b, can be
envisioned as a vinylogous alpha-elimination to yield







the carbene directly,


H H


9 Vinylogous >
a Elimination


10b


or a vinylogous beta-elimination to yield the allene.



H H


+ B Vinylogous
x X B Elimination

10b



To distinguish between these, compound 11, 1,2-benzo-

5-bromo-l,3,5-cycloheptatriene, and compound 12,

1,2-benzo-4-bromo-3-methyl-1,3,5-cycloheptatriene,




CH3 Br


Br Br


were synthesized and allowed to react with base.







Compound 11 could eliminate via the proposed vinylogous

alpha-elimination mechanism to give the carbene 13

without difficulty; however, to undergo the vinylogous

beta-elimination aromaticity must be broken. In contrast,

compound 12 could eliminate to the allene (14) by

the vinylogous beta-elimination mechanism without

breaking aromaticity, while vinylogous alpha-elimination
5
could not occur.









Br Eliminatiof 1 /

C0


7


6 E
Elimination


- < 30









H3 Br





H

12 15



In fact, compound 11 did not react with potassium

tert-butoxide to give allene type products, and it

was recovered unchanged. The use of a stronger base,

n-butyllithium, reduced off the bromine. Based on

this evidence, the presence of the benzene ring had

little if any retarding effect on the rate of

dehydrobromination of 12. However, the benzene ring

had a severe retarding effect on the rate of dehydro-

bromination of 11.

Therefore, it seems that a halocycloheptatriene

can readily dehydrohalogenate to a cycloheptatetraene

but cannot readily eliminate directly to the carbene.

If an extrapolation is carried out from the benz-

annelated systems back to the parent halocyclohepta-

triene (10), the formation of 4, the cyclopropane,

from 10, the halocycloheptatriene, would appear to

require two different intermediates; initially the

non-planar allene (9), which by a simple conformational







change, isomerized to give planar carbene (2),

which then added to the double bond.

The problem still existed, though, to experimentally

determine which intermediate represents the ground

state. An investigation of the infrared and ultraviolet

spectra of the intermediate was therefore undertaken.

Since the aforementioned theoretical calculations

and most of the chemical evidence pointed toward

cycloheptatetraene (9) as the ground state intermediate,

it was expected that the cumulative double bonds of the

allene type structure would appear in a region in the

infrared spectrum that could easily be detected.

After this investigation was undertaken, this expectation

was confirmed by West, Chapman, and Le Roux for a number

of azacycloheptatrienes. Additionally, it was anticipated

that the carbene (2) type structure would not have a

similar characteristic infrared spectrum and that a

viable difference in the infrared spectra would exist

between the two proposed intermediates.

Two approaches to the infrared detection of the

ground state intermediate were devised. The first method

involved the reaction of a suitable precursor to deliver

the intermediate onto a cold finger for infrared

spectroscopy. If this approach failed, the second

method was to sublime a reactant onto the cold finger,

decompose it in some manner to the intermediate, and

then detect the intermediate by infrared spectroscopy.







A distinction between the two intermediates, based
on their ultraviolet spectra would rest on molecular
orbital calculations that predicted theoretical ultra-
violet spectra of both the carbene structure and the
allene structure. These theoretical spectra are depicted
in figures 1 and 2 (absorption intensities were surmised
from the intensities generated by the CNDO-TDHF program
modified with Jaffe's parameters).









L /






200 300 400 500 600 700
nm
Figure 1
Theoretical Ultraviolet Spectrum
of Cycloheptatrienylidene (2)
























200 300 400 500 600 700

nm

Figure 2

Theoretical Ultraviolet Spectrum
of Cycloheptatetraene (9)

The reasonable method to obtain the experimental

ultraviolet spectrum was to dissolve the tosylhydrazone

salt (1) in a matrix and photolyze it. Additionally,

since the technique of matrix isolation was to be used

for the ultraviolet detection of the intermediate, and

the off chance that the triplet carbene is the ground

state, it was planned, at the same time, to do electron

paramagnetic resonance studies (EPR). Of course, if

either the allene (9) or the singlet carbene (2) was the

ground state intermediate then there should be no observ-

able EPR spectrum for the intermediate.

Finally, to study which intermediate is responsible

for the different products, the classical technique of

chirality was used. The carbene structure is presumably








planar and certainly achiral whereas the allene structure

is chiral even though it has a C2 axis of symmetry.



C Axis
12
*o H H


H H

2 H H
9

Therefore, only the allene, generated under chiral conditions

or from an appropriate active precursor, could react with

an inactive trap to form an optically active adduct.

Formation of this optically active adduct would give

unequivocal evidence for the allene as the reacting

intermediate.

As a conclusion to the work, if the proposed equili-

brium does exist,













then there should be at least two methods to probe for the

proof. The first thought involved studying the racemization








of the allene as the temperature of the reaction is

increased. If the proposed equilibrium is real, then as

the temperature is increased the trapped allene should

lose optical activity because the unimolecular race-

mization through the carbene should be more sensitive

to change in temperature than the bimolecular trapping

of the intermediate. Secondly, if the racemization

did occur in this manner, then selectively removing some

of the carbene from the equilibrium should cause an

increase in the optical activity.













RESULTS AND DISCUSSION


Since the infrared spectrum of the intermediate was

deemed the most logical choice for detecting the presence

of the allene type structure, the problem became design-

ing a method in which the intermediate could be obtained.

It was known that the dehydrohalogenation of the halo-

cycloheptatrienes yielded products or trapped products

characteristic of both types of intermediates. Therefore,

if there was some way to react one or more of the halo-

cycloheptatrienes with base and deposit the intermediate

on a cold finger for infrared detection, then the allene

could possibly be detected.

A mixture of the isomers of 1-, 2-, and 3-chlorocyclo-
9
heptatrienes was prepared by a modified method of Folisch.

This mixture was placed in a reservoir of a vacuum system

designed with a heatable zone that could be packed with

various solid bases, and a cold finger that would trap

the products from the reaction of the halocycloheptatrienes

with the base. It was anticipated that as the chlorocyclo-

heptatrienes were vaporized through the heated base, de-

hydrohalogenation would occur and the intermediate would be

deposited on the cold finger for infrared detection.








The first attempt to deposit the intermediate was a

reaction with basic alumina as the base. Various amounts

of this base and temperatures up to 3000C yielded only

chlorocycloheptatrienes being deposited on the cold

finger. The solution to the nonreactivity of the

starting material seemed to be a stronger base. Several

bases were tried on a liquid solution of the chloro-

cycloheptatrienes to determine which would cause dehydro-

halogenation. Sodium methanolate was chosen. However,

in the hot tube-vacuum apparatus, the only compound

that was deposited on the cold finger was heptafulvalene,

the carbene or allene dimer. Again, the various temper-

atures and amounts of the base yielded the same results.

A precursor was then sought that was volatile

and that would decompose to the intermediate from a

thermolytic reaction. This would permit application

of flash vacuum pyrolysis techniques that are known

to be useful for detecting reaction intermediates.

It was known that when 2,5-diphenylfuran-3-carbaldehyde

(16) reacts with l-amino-trans-2,3-diphenylaziridine (17),

the hydrazone (18) is generated, which on photolysis or

pyrolysis gives products derived from 3-furylcarbene (19).10












,Ph

CHO Ph CH=N-N



Ph
16 17 18


A or hv


CH
Carbene Products ----- Ph
Ph
19



Since this method of carbene generation could be

adaptable for thermally generating the sought after

intermediate for infrared detection, l-amino-trans-2,

3-diphenylaziridine was allowed to react with tropone

in an attempt to make the hydrazone. Once the hydrazone

had been obtained, it could either be sublimed into a

flash vacuum pyrolysis apparatus for depositing on a

cold finger for infrared detection, or the hydrazone

itself could be coated on the cold finger and photolyzed

to decomposition for the infrared detection of the

intermediate. However, when the reaction between tropone

and the aziridine was carried out to generate the hydrazone,








the only extractable product from the reaction mixture

was trans-stilbene. Even lowering the reaction tempera-

ture (to the point where no reaction occurred) yielded

no positive results. It was thought that maybe changing

the reactants to 7,7-dichlorocycloheptatriene and the

aziridine would help. If the hydrochloride salt were

obtainable, then it could be neutralized to yield the

hydrazone, possibly under less stringent reaction

conditions where the hydrazone would not decompose.

Again, the only extractable product from the reaction

mixture (even at temperatures down to -78C) was trans-

stilbene. Apparently, even if the hydrazone did form,

it was too unstable to isolate.

Even though we were becoming pessimistic about the

possibility of obtaining the infrared spectrum of the

intermediate, one last series of attempts was made.

The method devised was to sublime some type of salt

of tropone tosylhydrazone onto the cold finger and

photolyze it to generate the intermediate. At first,

the tetra-n-butyl ammonium salt was prepared from the

sodium salt. All attempts to sublime this salt yielded

only heptafulvalene being deposited on the cold finger.

The salt apparently decomposed from the heat instead

of subliming. The next thought was to reduce the size

of the groups attached to the nitrogen atom in the hope

that sublimation would occur. However, even the








tetra-methyl ammonium salt would not sublime and also

decomposed to yield heptafulvalene being sublimed onto

the cold finger.

The final attempt to obtain the infrared spectrum

of the intermediate involved changing the salt to lithium

tropone tosylhydrazone. The lithium salt should be the

most covalent salt and possibly the most likely to sublime

onto the cold finger without decomposing. Again, as the

heat was gently applied to the sublimation apparatus

the salt began to decompose and the brown-black film of

heptafulvalene crystals began to deposit on the cold

finger. At this time it was deemed impractical to

continue searching for methods to obtain the infrared

spectrum of the intermediate since it was believed that

other spectroscopic methods of detecting the intermediate

would be easier to conduct and yield nearly as much

information.

The next spectroscopic method that was applied to

determining the ground state intermediate was ultraviolet

spectroscopy. Initially there were two problems to

overcome before the method could be used. First, what

would one predict for the ultraviolet spectrum of the

different intermediates? Secondly, matrix isolation of

the sodium salt of tropone tosylhydrazone and then

photolysis to the intermediate was deemed the reasonable

route to the spectrum of the intermediate, but what type of








matrix would hold enough of the salt to yield the

needed information?

The solution to the first problem was addressed

by molecular orbital calculations using a CNDO-TDHF

program modified with Jaffe's parameters.8 These cal-

culations yielded the theoretical ultraviolet spectra

shown in Figures 1 and 2.

Originally it was thought that adamantane might

serve as the solid matrix to hold the sodium salt because

the use of most solution type glass matrices were in-

compatible in one way or another with the salt. An

adamantane matrix had been used at room temperature to

observe the optical absorption spectra of the hydro-

carbon free radicals cyclohexenyl, cyclopentenyl, cyclo-

hexadienyl, benzyl, cycloheptanyl and cyclooctanyl in

the ultraviolet region.1 The method involved pressing

disks of adamantane (10 mm. o. d., approximately 1 mm.

thick) containing the parent hydrocarbon and then placing

them in the sample and reference beams of the spectro-

photometer. No absorption was observed unless the disk

had been previously irradiated.1 Therefore, it was

hoped that a method patented after this one would yield

the ultraviolet spectrum of the intermediate.

When a cyclohexane solution of the salt and ad-

amantane was evaporated to dryness and pressed into a

pellet with an infrared press-pellet maker, no detectable








ultraviolet spectrum of the salt could be detected.

Therefore, the pellet matrix was redissolved in cyclo-

hexane and the ultraviolet spectrum of the solution

indicated the presence of the salt. This indicated the

possibility that the salt had not been trapped in the

crystal-cage lattice of the adamantane, and a matrix

had not been formed. In order to increase the solubility

of the salt in the adamantane, 15-Crown-5, the crown

ether specifically designed to work with sodium was

added to the cyclohexane solution of the salt and
12
adamantane.12 The solvent was evaporated, the crystalline

powder pressed into a pellet, and the ultraviolet

spectrum taken. Again, no observable ultraviolet

spectrum of the salt was obtained, but when the pellet-

matrix was dissolved in cyclohexane a representable

spectrum of the salt was obtained. Even increasing

the concentration of the salt far beyond that required

for any solution ultraviolet spectrum yielded no positive

results. The adamantane matrix idea was then abandoned

in favor of a glass matrix.

As indicated above, the use of the sodium salt

posed a problem in itself as far as using any liquid

solutions frozen to a glass matrix. Many of these solvents

are alcohols which would react with the salt, or they

are not polar enough to cause solution of the salt
13
with the solvent. It was found that a saturated








solution of 2-methyltetrahydrofuran and the salt gives

an ultraviolet spectrum of the salt, and this solution

freezes to clear glass.13

The next difficulty to overcome was designing the

equipment to obtain the matrix ultraviolet spectrum.

The method required an apparatus for freezing the sol-

ution of the salt and 2-methyltetrahydrofuran to a glass,

obtaining the ultraviolet spectrum of the salt and

obtaining the ultraviolet spectrum of the intermediate

after photolysis of the matrix to compare with the

theoretical spectra. Therefore, a quartz windowed dewar

was constructed which would hold liquid nitrogen and a

cell for the solution.









The dewar was designed so that it would fit into the

compartment of a Cary spectrophotometer, and Super Cel

quartz windows of the dewar could be aligned with the

optics of the spectrophotometer. The cell for the

solution was constructed of quartz and of the same

dimensions as a standard ultraviolet cell except that

it had a long neck so that the cell could be sealed

when placed in the dewar of liquid nitrogen.

















The apparatus and method (detailed below) initially

were checked for reproducibility by photolyzing a matrix

solution of phenyldiazomethane in 2-methyltetrahydrofuran

and also the tosyl hydrazone sodium salt of benzaldehyde

in 2-methyltetrahydrofuran. Both compounds, upon pho-

tolysis in their matrices, exhibited the literature

ultraviolet spectrum of phenyl carbene and the apparatus
14,15
was deemed usable. 15

A saturated solution of the sodium salt of tropone

tosyl hydrazone in 2-methyltetrahydrofuran was prepared

and this solution was filtered and pipetted into the

special cell. The matrix solution was freeze-thawed

three times to remove any oxygen and the cell was

sealed and placed into the dewar of liquid nitrogen.

The dewar was aligned in the spectrophotometer, a sample

of the 2-methyltetrahydrofuran was placed in the reference

compartment and the ultraviolet spectrum of the salt

in the glass was obtained (see Figure 3). The dewar

was removed from the sample compartment, and the glass

matrix was photolyzed with either a 450 or 550W Hanovia

mercury arc lamp through the quartz windows. Photolysis








was conducted in 15 minute intervals until there was no

further change in the ultraviolet spectrum. After each

15 minute photolysis period the dewar was replaced in

the spectrophotometer and the ultraviolet spectrum was

taken. Gradually, the ultraviolet spectrum of the salt

began to disappear and the appearance of a shoulder at

390 nm. appeared (see Figure 3). As can be seen by

comparison to the theoretical spectra of the carbene and

allene, the experimental ultraviolet spectrum looked

somewhat similar to that of the allene. However, there

did not seem to be enough similarities to definitely

conclude that the spectra were the same. Finally, after

the ultraviolet spectrum of the intermediate showed no

further change upon photolysis, the glass-matrix was

allowed to melt by slow warming to room temperature. It

was observed that upon melting there was vigorous gas

evolution (sometimes to the point of causing the cell

to crack) and the solution darkened in color. After

melting, the solution was refrozen to a glass and a

final ultraviolet spectrum was taken. This final spectrum

(see Figure 3) indicated the presence of heptafulvalene,

the dimer of the carbene, and some starting material.

Additionally, if the solution was spotted on a silica

gel thin layer chromatography plate along with a spot

of known heptafulvalene, and the plate developed in

pentane, only heptafulvalene eluted from the matrix


















































































< w O !:IQ r0 C 0 w


I
zgz


(I)
ci)



rd


L








solution. And as a final check for this dimer, if

known heptafulvalene was salted into the matrix solution

and another ultraviolet spectrum was taken, the ultra-

violet spectrum of heptafulvalene was greatly enhanced.

Based on the ultraviolet data, there was no long

wavelength absorption as predicted for the carbene.

The similarity of the theoretical allene spectrum to the

experimental spectrum of the intermediate seemed to show

that the allene is the intermediate observed.

Essentially the same technique of matrix isolation

of the sodium salt of tropone tosylhydrazone was also

applied to obtaining the electron paramagnetic resonance

(EPR) spectra of a number of cycloheptatrienyliden-

cycloheptatetraene intermediates. Saturated solutions

of the following sodium salts and the sodium salt of

tropone tosylhydrazone were made in 2-methyltetrahydro-

furan.




Na Na
0 E
N-N-Ts N-N-Ts









20 21










Na
Na Na
N-N-Ts N
N-N-Ts





22
-- 1



Each one was placed in a standard EPR tube. The tube

was then immersed in the EPR dewar containing liquid

nitrogen and frozen to a glass. All were photolyzed with

a 550W Hanovia mercury arc lamp and tested for an EPR

spectrum. None showed EPR spectra. All samples liberated

gas upon warming the glass to melt. This gas evolution

was interpreted as the release of nitrogen from the

decomposition of the tosylhydrazone salt when the inter-

mediate was generated during the photolysis. The absence of

an EPR spectrum for these intermediates is consistent with

either the allene or singlet carbene intermediate.14

The problems of trying to obtain an infrared spectrum

of the intermediate, the lack of a definitive ultraviolet

spectrum, along with the absence of an EPR spectrum and

the seemingly small amount of chemical evidence pointing

toward the allene as a reactive intermediate, demanded

a new approach to probe the structure. As indicated in

the introduction, the carbene is probably planar and








certainly achiral, whereas the allene, even though it

has a C2 axis of symmetry, is chiral.


C Axis
0O








Therefore, the powerful technique of chirality could

unequivocally identify the reacting intermediate.

If the allene was the reacting species, as most of the

evidence indicated, then an optically active product

could be formed from the intermediate and an inactive

trap, if the intermediate was generated under chiral

conditions.

As noted in Table 1, generation of the intermediate

from 1-, 2-, and 3-bromocycloheptatrienes in the presence

of 1,3-diphenylisobenzofuran with 1-potassium menthoxide

gave the allene adduct (allene and 1,3-diphenylisobenzo-

furan, 24) with a specific rotation. However, there

was a real reason for concern about this data.








Table 1

Specific Rotation Data for the Intermediate Generated
from the Reaction of the Bromocycloheptatrienes and
1-Potassium Menthoxide

Temperature Reaction Time Amount of Adduct Specific
(oC) Rotation

-40 2 days 100 mg. 1.92

-40 2 days 98 mg. 2.200

-40 2 days 58 mg. 2.170

0 1 day 58 mg. 3.400

25 2 hours 114 mg. 3.200

45 1 hour 116 mg. 2.600

65 1 hour 121 mg. 0.740

NOTE: Rotations were determined at 589 nm. (Sodium D
line) and dissolving all of the adduct in 5.0 ml.
of methylene chloride in a 1.0 decimeter polarimeter
cell. All rotations are accurate to 0.150.



Once the active base had reacted with the bromides or was

destroyed on the workup of the reaction with water, the

reaction mixture contained 1-menthol, an optically active

by-product. Although excess 1,3-diphenylisobenzofuran

could be removed by reaction with maleic anhydride to

give an adduct that did not elute with pentane from a

column of grade 2 basic alumina with heptafulvalene, the

other by-product eluted from the column as the first

fraction followed by the allene adduct. The allene adduct

could not be obtained pure according to nmr; there were

invariably absorptions in the upfield portion of the








spectrum corresponding to menthol. In addition the isolated

compound had the odor of menthol. Further, and presumably

at least in part due to this impurity, the specific rotation

data were by no means consistent nor were they reproducible

from run to run. All these results seemed to indicate

contamination of the allene by 1-menthol. This type of

inconsistency was intolerable so various exotic chromato-

graphic techniques were tried.

The first technique used was preparative thick layer

chromatography plates made with silica gel. The allene

adduct isolated from the column above was streaked onto

the plate and the plate was developed in pentane. Numerous

fractions containing the adduct were isolated. Unfortun-

ately, all the fractions either contained the same menthol-

type nmr absorption or some unknown nmr absorptions.

Additionally, the specific rotations varied from fraction

to fraction. Therefore, this technique was abandoned.

The second technique of dry column chromatography

involved packing a column with grade 2 basic alumina,

and then adding this coated alumina to the top of the

column. Pentane was used as the elutant. However, as

with the thick layer method, the various fractions

isolated containing the allene adduct exhibited some

type of menthol nmr absorption and each fraction varied

in specific rotation. Again, this technique was given up

for a final method.








The third technique was low pressure column chroma-

tography. A system was designed with a two liter solvent

tank which could be pressurized up to 15 psi with nitrogen.

The three foot column, one inch in diameter, was packed with

140 mesh, or greater, silica gel powder and pentane was

used as the elutant. The pre-isolated allene adduct

was absorbed onto silica gel and added to the dry column.

The system was closed and pressurized with nitrogen

gas to start the elution. As with the other two methods,

several fractions were isolated containing the allene

adduct. These fractions also showed the same menthol-

type nmr contamination and variation in the specific

rotation.

The idea of using chirality to seek the answer as

to which intermediate is involved in the chemistry was

not abandoned, just the method of using 1-potassium

menthoxide as the asymmetric inducing reagent. Therefore,

a method of generating the intermediate from an optically

active precursor was pursued. If the intermediate

generated in this manner could be trapped, and if this

trapped product was optically active, then the allene

could be proposed as the ground state intermediate.

The method used was the reduction of 7,7-dibromo-
16
cycloheptatriene with lithium aluminum deuteride-quinine.1

Due to the bulkiness of this reducing agent, it was

believed that steric hinderance would allow more of one





31


optical isomer than the other. This was found since the

specific rotation of the deuterated bromides was -0.164.


+ LAD3-Q


The reaction of the deuterated bromides with base

(potassium tert-butoxide) was run in the presence of

1,3-diphenylisobenzofuran.


Ph

Br + 0

h


Base
>








Since cis-elimination of DBr from the bromides would

give one enantiomer while elimination of HBr would give

the other, the primary isotope effect should lead to an

excess of one enantiomer over the other. The crude

material was chromatographed on alumina with pentane

and in this series of reactions, two isomers (endo and

exo) of the allene adduct (24) were formed as indicated

by nmr. Additionally, these two isomers were separated

by an activated AgNO3-silica gel column eluting with 7%

ether in pentane. The isolated endo isomer gave a

specific rotation of 0.9820. This result is taken as

unequivocal evidence for a chiral intermediate as the

precursor to the allene adduct (24).

Further, as can be noted from the specific rotation

data in Table 2, when the reaction is run at elevated

temperatures the specific rotation of the allene adduct

decreases. This result suggests that the intermediate

racemizes at higher temperatures. To racemize, the

allene must convert to an achiral intermediate which is

most certainly the carbene. Therefore, this implies that

the allene is converting to the carbene as indicated

in the proposed equilibrium.


/








Table 2

Specific Rotation Data for the Intermediate Generated
from the Reaction of the Deuterated Bromocycloheptatrienes
and Potassium tert-Butoxide

Temp. Reaction Time Amount of Endo/Exo Specific
(C) Adducts Ratios Rotation

25 2 hours 95 mg. 2.9/1 -1.88

25 2 hours 112 mg. 2.8/1 -1.840

53 1 hour 145 mg. 2.2/1 -1.38

53 1 hour 100 mg. 2.5/1 -1.500

100 30 minutes 100 mg. 1.1/1 -0.950

100 30 minutes 100 mg. 1.1/1 -0.910

140 30 minutes 75 mg. 0.8/1 -0.600

NOTE: Rotations were determined at 589 nm. (Sodium D
line) and by dissolving all of the adduct in 5.0 ml.
of methylene chloride in a 1.0 decimeter polarimeter
cell. All rotations are accurate to 0.150.



To continue the investigation into the equilibrium,

another series of reactions was conducted in which the

deuterated bromides were reacted with base in the presence

of two trapping reagents, the 1,3-diphenylisobenzofuran

and styrene. Styrene is known to be a trapping reagent

of cycloheptatienylidene,17 and if the carbene is being

formed as proposed in the equilibrium, it is anticipated

that the styrene would trap the carbene. It can also be

envisioned that chirality could be an unequivocal test

for this precursor. Since isomerization of the allene

to the carbene is a racemizing process, removing the








carbene with styrene should affect the optical purity

of the allene adduct. In doing so, it was hoped there

would be an observed increase in the optical activity of

the allene adducts. Unfortunately, such a test was not

feasible due to a leveling effect from heptafulvalene

formation (from either the carbene or the allene) as can

be seen from the following scheme and assumptive inter-

pretation.


9 2


1,3-Diphenyl- Styrene
isobenzofuran (23)


Ph




Ph








For a typical reaction of the deuterated bromides

with base in the presence of 1,3-diphenylisobenzofuran

at 0C, the assumption can be made that there are 1000

molecules of allene (9) formed. Then, based on typical

reaction yields, 800 molecules are converted to hepta-

fulvalene (3), 100 to the allene (9) and the other 100

to the carbene (2). When the reaction is run in the

presence of only 1,3-diphenylisobenzofuran (23), the

allene (9), is trapped, but the other 100 molecules

of carbene (2), are free to return to the allene side

of the equilibrium. When they do, 80 of them lead to

heptafulvalene (3), 10 will be racemized and trapped

by 1,3-diphenylisobenzofuran (23), and 10 will return

to carbene (2). Again, the remaining 10 carbene molecules

exist in the equilibrium and 8 will lead to heptafulvalene

(3), 1 will be racemized and trapped by the 1,3-diphenyl-

isobenzofuran (23), and 1 will return to a carbene (2).

This is summarized in Table 3 below.










Summary of Data
Leveling Effect



Passes Through
Equilibrium

1st Pass

2nd Pass

3rd Pass


Table 3

for Assumptive Interpretation of the
from Heptafulvalene Formation

Molecules Formed Molecules Formed
at 0C at 1000C
Hept. All.* Car. Hept. All.* Car.
(3) (9) (2) (3) (9) (2)

800 100 100 800 20 180

80 10 10 144 4 32

8 1 1 36 1 5

*Total of 111 mole- *Total of 25 mole-
cules of allene os cules of allene of
which 100 are opti- which 20 are opti-
cally active and 11 cally active and 5
are racemic. are racemic.


Thus, even if styrene trapped all carbene molecules formed,

the increase in specific activity would only be about

10%. Therefore, even if the principle is sound, the

leveling effect of heptafulvalene (3) formation would

make the actual experimental change in activity upon

addition of reasonable amounts of styrene too small to

comfortably interpret.

However, chirality can still be used to answer the

question about the intermediate responsible for the

formation of the spirononatriene (4). If the styrene is
2 8
trapping the allene (9) in an allowed (2 s + T s) reaction,

the mode of attack on the double bond should be as

pictured below.









P H
H H

+ PhCH=CH2 -


25
9



Furthermore, based solely on steric grounds, the phenyl

ring should have a favored orientation (most likely

the one pictured). If the phenyl ring does favor one

orientation, then the optically active allene (9) should

give an excess of one enantiomeric transition state,

and therefore the trapped spirononatriene (4) should

be optically active. However, this activity would

result from asymmetric induction and most likely would

be small. A rather large quantity of the spirononatriene

(4) was prepared -(at -450C to accentuate the assymetric

induction) from the deuterated bromides and styrene

in the presence of potassium tert-butoxide and its

rotation taken. No rotation was detected. This is

taken as negative evidence, but it is the result that

would be expected if the carbene is the intermediate being

trapped. Additionally, it has been reported that the

intermediate generated from either the sodium tosyl-

hydrazone salt of tropone or the mixture of 1-, 2-, and

3-bromocycloheptatrienes is trapped by nucleophilic







reagents to give 26. This is the type of product

expected of a carbene, not a strained cyclic allene

which should give 27.18,19


X OCH3
\/ 3


CHO3D CH3OD
+ isomers I---- or 10


H3CO
3 1












CONCLUSIONS


The technique of chirality proved that the dehydro-

bromination of the 1-, 2-, and 3-bromocyclohepatrienes gave

an intermediate that is best represented by the allene

structure (9), and that this intermediate is responsible

for the allene adduct, 24. Both approaches, the generation

of the intermediate from 1-, 2-, and 3-bromocyclohepatrienes

in the presence of 1,3-diphenylisobenzofuran with 1-

potassium menthoxide and the generation of the intermediate

from an optically active precursor in the presence of

1,3-diphenylisobenzofuran, gave the optically active

adduct, 24. These results are unequivocal evidence for
20
a chiral intermediate as the progenitor of adduct 24.2

The effect of change in temperature on the specific

rotation of the adduct 24, generated from 1-, 2-, and 3-

bromocycloheptatrienes in the presence of 1,3-diphenyl-

isobenzofuran and 1-potassium menthoxide, is recorded in

Table 1. However, as described, the endo adduct of 24

could not be separated from small and varying amounts of

the exo isomer of 24. Also, the effect of change in

temperature on the specific rotation of the adduct 24

generated from optically active deuterated 1-, 2-, and







3-bromocycloheptatrienes in the presence of 1,3-diphenyl-

isobenzofuran (23) and potassium tert-butoxide is recorded

in Table 2. The only valid conclusion from these results

is that up to 65C at least some of the activity persists.

This requires that if the allene (9) is competitively

racemizing, at least some is bled off before complete
20
racemization occurs.

Since the results of the ultraviolet spectrum from

the matrix photolysis of the sodium salt of tropone

tosylhydrazone were inconclusive, chirality was also

used to assign the structure of the precursor to the

spirononatriene (4). If styrene does trap cyclohepta-

tetraene (9) in the allowed (T2s + T s) reaction, then

reacting optically active allene (9) with styrene would

have produced the optically active spirononatriene (4).

Since no activity was observed, the achiral carbene

structure (2) was assigned to the progenitor of the

spirononatriene ('). Additionally, other evidence has

been reported that the intermediate generated from the

sodium salt of tropone tosylhydrazone or the mixture

of 1-, 2-, and 3-bromocycloheptatrienes is the carbene.18

Therefore, the results have proved that the dehydro-

bromomination of the 1-, 2-, and 3-bromocycloheptatrienes

gave an intermediate that is best represented by the allene

structure (9) and this intermediate is responsible for

the allene adduct, 24. In addition, the evidence also




41


favored a carbene (2) as the progenitor of the spiro-

nonatriene (4).












EXPERIMENTAL


General

Melting points were determined on a Thomas Hoover

Capillary Melting Point Apparatus and are uncorrected.

Boiling points were recorded as the temperature at which

the sample distilled, are uncorrected, and were recorded

at atmospheric pressure unless reported otherwise.

Nuclear magnetic resonance spectra (60 MHz) were

recorded on either a Varian Associates A-60A or JEOL PMX

60 Spectrometer and are reported in parts per million (6)

downfield from tetramethylsilane, the internal standard.

Exo/endo isomer ratios were calculated from nmr integration

obtained on a JEOL FMX 100 Spectrometer. Infrared data

were recorded on a Perkin-Elmer Model 137 Sodium Chloride

Spectrophotometer. Mass spectra were obtained on an AEI

MS 30 Spectrometer. Ultraviolet and visible spectra were

recorded on either a Cary 14, 15, or 17 double-beam

spectrophotometer. Electron spin resonance spectra were

obtained on a Varian Associates E-3 EPR Spectrometer.

All chemicals were reagent grade and were used as

supplied unless otherwise stated. Tetrahydrofuran and

dioxane were dried by distillation from lithium aluminum







hydride. Diglyme was dried by distillation from calcium

hydride and stored over calcium hydride until used. The

2-methyltetrahydrofuran was dried by distillation from

lithium aluminum hydride, fractionally distilled by

spinning band distillation, and stored over sodium-

potassium alloy until used.

Column chromatography was conducted using Baker 60-

200 mesh silica gel or Fisher basic alumina. Low pressure

column chromatography was conducted using Fisher silica

powder (140 mesh maximum).

Matrix irradiations were carried out using either

a 450-W or 550-W Hanovia high pressure mercury arc lamp,

centered in a water cooled quartz immersion well.

Photolysis of the matrix was conducted through a quartz

windowed dewar and 1-cm quartz cells.

Preparation of Tropone

In a two liter round bottom Morton flask equipped

with a mechanical stirrer and a reflux condenser was

stirred 13.5 g. (0.10 mole) of cycloheptatriene, 53.0

g. (0.48 mole) of selenium dioxide, 33 ml. of water, and

350 ml. of dioxane. The mixture was heated to 90'C in

an oil bath and stirred for sixteen hours. The mixture

was allowed to cool to room temperature, the residual

selenium was filtered, and the filtrate was poured into

750 ml. of water. The aqueous solution was extracted

with methylene chloride (3 X 300 ml.). The extracts







were washed with 250 ml. of 10% sodium bicarbonate

solution and then dried (MgS04). The methylene chloride

was rotary evaporated and the residual brown oil was

distilled under reduced pressure. A pale yellow

liquid (34.0 g., 63% yield) was obtained with a boiling

point of 88-90C/3.0 mm. (lit. 91-920C/4.0 mm.).21

Preparation of 2,4,6-Cycloheptatriene p-Toluenesulfonyl
hydrazone

In a 250 ml. three neck round bottom flask equipped

with a mechanical stirrer, a condenser, a capped addition

funnel, and a nitrogen inlet tube was stirred 5.0 g.

(0.47 mole) of tropone in 75 ml. of dry methylene

chloride. The flask was immersed in an ice bath and over

a 15 minute period 6.3 g. (0.49 mole) of oxayl chloride

in 25 ml. of dry methylene chloride were added dropwise.

The resulting slurry was stirred for an additional 30

minutes at room temperature and then the solvent was

removed by aspiration.

The remaining muddy brown crystals were dissolved in

50 ml. of absolute ethanol. This alcoholic solution was

poured rapidly into a 150 ml. solution of absolute

ethanol and 8.8 g. (0.47 mole) of p-toluenesulfonyl-

hydrazine. This dark red solution was kept under a nitrogen

atmosphere and warmed over a steam bath for 30 minutes.

The mixture was allowed to cool to room temperature and

then placed in an ice bath. The yellow-brown solid which

formed was filtered and the crystals were washed with







ether (13.1 g., 89% yield, melting point 170-1C dec.).

All 13.1 g. of the hydrochloride salt were added

to a mixture of 750 ml. of 10% sodium bicarbonate and

750 ml. of methylene chloride. This resulting mixture

was stirred vigorously for 30 minutes, separated, and

the aqueous layer extracted with methylene chloride

(2 X 250 ml.). The extracts were dried (MgSO ) and

the methylene chloride was rotary evaporated leaving red-

brown crystals (11.9 g., 94% yield, melting point 142-

40C, lit. 144-5C).1

Preparation of the Sodium Salt of 2,4,6-Cycloheptatriene
p-Toluenesulfonylhydrazone

The 2,4,6-cycloheptatriene p-toluenesulfonylhydrazone

(10.0 g., 0.036 mole) was weighed, placed in a 250 ml.

flask in a dry box, and dissolved in 100 ml. of dry

tetrahydrofuran. As this solution was stirred with a

magnetic stirrer, 0.88 g. (0.036 mole) of sodium hydride

powder were added in small amounts. When the gas

evolution ceased after the last of the sodium hydride

powder had been added, the resulting slurry was stirred

for another 15 minutes. The brown precipitate (10.6 g.,

100% yield) was filtered and stored in the dry box until

used.1

Preparation of 1-, 2-, and 3-Chlorocycloheptatrienes

In a 250 ml. three neck round bottom flask equipped

with a mechanical stirrer, a condenser, a capped addition







funnel, and a nitrogen inlet tube was stirred 5.0 g.

(0.47 mole) of tropone in 75 ml. of dry methylene

chloride. The flask was immersed in an ice bath and

over a 15 minute period 6.3 g. (0.49 mole) of oxayl

chloride in 25 ml. of dry methylene chloride was added

dropwise. The resulting slurry was stirred for an

additional 30 minutes at room temperature and then the

solvent was removed by aspiration.

In the same flask, the yellow residue was dissolved

in 100 ml. of dry tetrahydrofuran and the addition

funnel was replaced with a solid addition tube containing

3.8 g. (0.10 mole) of lithium aluminum hydride. The

flask was immersed again in the ice bath and the lithium

aluminum hydride was added in small portions over a

20 minute period. The resulting suspension was stirred

for an additional hour and allowed to come to room

temperature. The excess hydride was destroyed with

water and the precipitate filtered. The filtrate was

extracted with 250 ml. of methylene chloride and the

methylene chloride extract was dried (MgS04). The

solvent was rotary evaporated leaving a light brown oil

which was purified by column chromatography on silica

gel using pentane as the elutant. The yellow oil

resulting after rotary evaporation of the pentane (3.5 g.,

60% yield) is a mixture of 1-, 2-, and 3-chlorocyclohepta-

trienes and has the nmr spectrum reported in the literature.







Preparation of the Lithium Salt of 2,4,6-Cycloheptatriene
p-Toluenesulfonylhydrazone

The 2,4,6-cycloheptatriene p-toluenesulfonyl-

hydrazone (0.30 g., 0.001 mole) was dissolved in 30 ml.

of dry tetrahydrofuran in a dry box. This solution was

placed in a 100 ml. round bottom flask equipped with

a magnetic stirrer. To this was added dropwise via

syringe 0.5 ml. (0.0011 mole) of 1.8 molar phenyl

lithium solution. This mixture was stirred for 15

minutes and then transferred to the atmosphere where

the tetrahydrofuran was rotary evaporated. The residue

was not isolated and used immediately.

Preparation of the Tetramethylammonium Salt of 2,4,6-
Cycloheptatriene p-Toluenesulfonylhydrazone

The sodium salt of 2,4,6-cycloheptatriene p-toluene-

sulfonylhydrazone (0.070 g., 0.0002 mole) and tetra-

methylammonium bromide (0.035 g., 0.0002 mole) were

dissolved in 50 ml. of water in a 100 ml. round bottom

flask. The water was rotary evaporated and the residue

was extracted with 50 ml. of tetrahydrofuran. The

insoluble residue from the extraction was filtered and

the tetrahydrofuran filtrate was rotary evaporated

leaving an orange precipitate (0..060 g., 87% yield).

Preparation of the Tetra-n-butylammonium Salt of 2,4,6-
Cycloheptatriene p-Toluenesulfonylhydrazone

The sodium salt of 2,4,6-cycloheptatriene D-toluene-

sulfonylhydrazone (0.3 g., 0.001 mole) and tetra-n-

butylammonium bromide (0.32 g., 0.001 mole) were







dissolved in 50 ml. of water in a 250 ml. flask. To this

mixture 50 ml. of methylene chloride were added and the

mixture was stirred for 15 minutes. The reaction mixture

was transferred to a separatory funnel and the

methylene chloride layer was separated. The methylene

chloride layer was dried (MgSO4) and rotary evaporated

to dryness leaving a purple oil. The oil was stirred

twice with two 50 ml. portions of pentane. A purple

precipitate (0.52 g., 100% yield) resulted; it was

filtered and was stored in a dry box until used.

Preparation of N-Diphenylaziridinephthalimide

In a 300 ml. three neck round bottom flask equipped

with a mechanical stirrer, a solid addition tube, and a

stopper was added 6.5 g. (0.04 mole) of N-amino-

phthalimide and 36.0 g. (0.20 mole) of trans-stilbene

dissolved in 100 ml. of dry methylene chloride. To this

solution 20.0 g. (0.04 mole) of lead tetraacetate were

added in small portions with the aid of the solid addition

tube over a 10 minute period with vigorous stirring. The

reaction mixture was stirred an additional hour, filtered

through celite, and the filtrate rotary evaporated. The

residual material was column chromatographed on silica

gel with methylene chloride. The excess trans-stilbene

eluted first, followed by two unidentified compounds,

and then the adduct. The adduct was recrystallized by

dissolving it in a minimum amount of hot chloroform and







then adding enough pentane to cloud the solution.

Beautiful pale yellow crystals were obtained (7.5 g.,

55% yield, melting point 176-9C, lit. 175C).22

Preparation of N-Aminodiphenylaziridine

In a 250 ml. three neck round bottom flask equipped

with a mechanical stirrer, a capped addition funnel,

and a stopper, 2.0 g. (0.006 mole) of N-diphenyl-

aziridinephthalimide were added dissolved in 60 ml. of

ethanol. In the addition funnel 60 ml. (1.2 mole) of 64%

hydrazine were placed. The flask was placed in a dry

ice/isopropanol bath at -45C. After coming to

temperature the dropwise addition of the hydrazine was

started and continued for one hour with stirring. After

final addition the reaction mixture was allowed to warm

to room temperature and was added to 100 ml. of ice

water. The resulting solution was extracted with 250 ml.

of ether. The ether layer was separated, dried (K2C03),

and rotary evaporated to dryness leaving 1.0 g. of

crystals (81% yield, melting point 90-20C, lit. 93-4C).22

Preparation of 1-, 2-, and 3-Bromocycloheptatrienes

In a 250 ml. three neck round bottom flask equipped

with a mechanical stirrer, a condenser, a capped addition

funnel, and a nitrogen inlet tube was stirred 10.6 g.

(0.10 mole) of tropone dissolved in 50 ml. of dry

methylene chloride. The flask was immersed in an ice

bath and over a 15 minute period 25 g. (0.12 mole) of







oxayl bromide in 25 ml. of dry methylene chloride were

added dropwise. The resulting slurry was stirred for

an additional 30 minutes at room temperature and then

the solvent was removed by aspiration.

In the same flask, the orange-yellow residue was

dissolved in 100 ml. of dry tetrahydrofuran and the

addition funnel was replaced with a solid addition tube

containing 4.0 g. (0.11 mole) of lithium aluminum hydride.

The flask was immersed again in the ice bath and the

lithium aluminum hydride was added in small portions

over a 20 minute period. The resulting suspension was

stirred for an additional hour and allowed to come to

room temperature. The excess hydride was destroyed

with water and the precipitate filtered. The filtrate

was extracted with 250 ml. of methylene chloride and the

methylene chloride extract was dried (MgSO4). The

solvent was rotary evaporated leaving a light brown

oil which was purified by column chromatography on

silica gel using pentane as the elutant. The yellow oil

resulting after rotary evaporation of the pentane

(13.5 g., 79% yield) is a mixture of 1-, 2-, and 3-

bromocycloheptatrienes and has the nmr spectrum reported

in the literature.







Preparation of 3-Phenylphthalide

In a two liter round bottom flask equipped with a

reflux condenser, 200 g. of zinc dust, 100 g. (0.44 mole)

of ortho-benzoylbenzoic acid, 200 ml. of water, and

800 ml. of glacial acetic acid were refluxed vigorously

for two hours. The resultant yellow solution was

decanted from the residue. The residue was then washed

with 200 ml. of hot glacial acetic acid. The solutions

were combined and allowed to stand overnight.

White needles were separated by suction filtration.

The needles were washed with 500 ml. of water and a

second crop of crystals was suction filtered from the

mother liquor. Again the second crop of crystals was

washed with 500 ml. of water. The combined crystals

were added slowly to one liter of 10% sodium bicarbonate

solution with stirring. After all the crystals had

been added, the solution was checked for basicity (solid

sodium bicarbonate was added to maintain the basic

condition). Stirring was continued for an additional

hour to allow the flocculent material to digest. The

solution was filtered and the'precipitate was recrystallized

from approximately one liter of ethanol. Hot filtration

was necessary to remove a small amount of insoluble

impurity. After recrystallizing, 51 g. (55% yield) of

material was isolated (melting point 114-5C, lit.

114-50C).23







Preparation of 1,3-Diphenylisobenzofuran

In a 500 ml. addition funnel capped at the bottom

with a 50 ml. round bottom flask and equipped with a

reflux condenser, phenyl magnesium bromide was generated

under nitrogen from 40 g. (0.25 mole) of bromobenzene

and 6 g. (0.25 mole) of magnesium turnings.

The funnel was transferred to a one liter three

neck round bottom Morton flask equipped with a condenser,

a nitrogen inlet tube, and a mechanical stirrer. The

Grignard reagent was dripped into 42 g. (0.20 mole)

of 3-phenylphthalide in 250 ml. of dry tetrahydrofuran

over a 45 minute period. After an additional hour of

stirring, 300 ml. of methylene chloride were added, and

the resulting solution was washed two times with 500

ml. of one normal hydrochloric acid. Next the organic

phase was washed two times with 500 ml. of 10%

potassium carbonate. The water washings were combined

and extracted with 250 ml. of methylene chloride. The

methylene chloride phases were combined, dried (MgSO4),

and rotary evaporated to dryness. The residue was

recrystallized from benzene. The crystals were washed

with 100 ml. of ethanol. The filtrate was rotary

evaporated to one-half volume and a second crop of crystals

was isolated. Both crops were combined to yield 38 g.

(87% yield) of yellow needles (melting point 128-30C,

lit. 130-1C).24







Preparation of 1-Potassium Mentholate

A 250 ml. round bottom three neck flask was equipped

with a condenser, an argon inlet tube, and a mechanical

stirrer. Introduced into the flask was 10.0 g. (0.064

mole) of 1-menthol dissolved in 70 ml. of dry benzene.

The flask was placed in an oil bath at 1000C and 2.5 g.

(0.64 mole) of potassium metal were added piecewise.

After all of the potassium metal had been added, the

reaction was refluxed overnight.

The next day the reaction was allowed to cool to

room temperature and the flask transferred to a dry

box. The white precipitate (12.4 g., 100% yield) was

filtered and washed with dry ether. The precipitate

was stored in a dry box until used.

Preparation of Lithium Aluminum Deuteride-Quinine (LAD3-Q)

To a 1000 ml. three neck round bottom Morton flask

equipped with a mechanical stirrer, a condenser, a

nitrogen inlet tube, and a solid addition tube was

added 17.0 g. (0.5 mole) of dry quinine in 250 ml. of

dry ether. The lithium aluminum deuteride (2.0 g.,

0.45 mole) was placed in the solid addition tube and

added slowly to the quinine solution which had been

immersed in an ice bath. After the addition had been

completed the ice bath was removed and the solution was

refluxedd for two hours. The LAD3-Q was not isolated and
ws ud 16,25
was used immediately.







Reduction of 7,7-Dibromocycloheptatriene with Lithium
Aluminum Deuteride-Quinine (LAD3-Q)

The reaction was conducted in the same Morton

flask containing the prepared LAD3-Q. The solid

dibromide (prepared from 2.1 g. (0.02 mole) of tropone

and 6.5 g. (0.03 mole) of oxayl bromide) was added

slowly to the prepared LAD3-Q, which was well stirred

and cooled by an ice bath. After all of the dibromide

had been added the reaction was allowed to stir for

sixteen hours at 0C. Then 20 ml. of water was added

slowly, followed by 100 ml. of 20% sulfuric acid. The

ether layer was separated, washed with another 100 ml.

of 10% sodium carbonate, and finally a washing with 100 ml.

of water was conducted. The ether layer was dried (MgS04)

and rotary evaporated leaving 3.1 g. (91% yield) of the

deuterated bromide (typical mass spectrum indicated

99+% monodeuteration and the specific rotation of

1.0 g. of the material was -0.1640).16,25

Cold Finger Deposit of the Reaction Product of 1-, 2-,
and 3-Chlorocycloheptatrienes with Base

The chlorocycloheptatrienes (0.5 g.) were frozen

in a resevoir with liquid nitrogen cooling. A vacuum

was applied to the system and the nitrogen was removed

from the resevoir. The chlorocycloheptatrienes were

vaporized through a 6 inch plug of basic alumina heated

to 100C. The only material deposited on the cold finger

was unreacted chlorocycloheptatrienes. The temperature








was increased to 3000C, and the chlorocycloheptatrienes

failed to react.

The 6 inch plug of basic alumina was replaced with

a one inch plug of sodium metholate. Again, the

chlorocycloheptatrienes were vaporized through this plug

at 1000C. A brown crystalline deposit was noted on

the cold finger. This deposit was confirmed as hepta-

fulvalene by nmr.

Attempted Reaction of N-Aminodiphenylaziridine and
7,7-Dichlorocycloheptatriene

In 25 ml. of dry methylene chloride, 200 mg. (0.001

mole) of freshly prepared 7,7-dichlorocycloheptatriene

and 263 mg. (0.001 mole) of N-aminodiphenylaziridine

were stirred overnight. After rotary evaporation of

the solvent, the nmr spectrum indicated the only

extractable product of the reaction was trans-stilbene.

The reaction was repeated several times at various

low temperatures (0, -20, -400, -780C) with similar

negative results. Also, pyridine was added to some

of the attempts to possibly remove any hydrochloric acid

formed. Again, only decomposition to trans-stilbene

occurred.

Attempted Reaction of N-Aminodiphenylaziridine and Tropone

In 25 ml. of dry tetrahydrofuran, 200 mg. (0.002

mole) of tropone and 400 mg. (0.002 mole) of N-amino-

diphenylaziridine were stirred at reflux for 20 hours.







Aliquots were taken at 0.5, 2, 5, 12, and 20 hours for

nmr spectrums. After 20 hours the only extractable

product was trans-stilbene.

Attempted Sublimation of the Lithium Salt of 2,4,6-
Cycloheptatriene p-Toluenesulfonylhydrazone

The residue from the preparation of the lithium

salt of 2,4,6-cycloheptatriene p-toluenesulfonyl-

hydrazone was placed in a sublimation apparatus. The

apparatus was slowly heated to 100C under vacuum. At

that temperature the salt began to decompose and a

brown film began to form on the cold finger. The

sublimation was ceased and the brown film was identified

as heptafulvalene by nmr.

Attempted Sublimation of the Tetramethylammonium Salt
of 2,4,6-Cycloheptatriene p-Toluenesulfonylhydrazone

The tetramethylammonium salt of 2,4,6-cyclohepta-

triene p-toluenesulfonylhydrazone (0.060 g.) was placed

in a sublimation apparatus. The apparatus was slowly

heated to 1400C under vacuum. At that temperature the

salt began to decompose and a brown film began to

deposit on the cold finger. The sublimation was ceased

and the brown film was identified as heptafulvalene
1
by nmr.

Attempted Sublimation of the Tetra-n-butylammonium Salt
of 2,4,6-Cycloheptatriene p-Toluenesulfonylhydrazone

The tetra-n-butylammonium salt of 2,4,6-cyclohepta-

triene p-toluenesulfonylhydrazone (0.25 g.) was placed

in a sublimation apparatus. The apparatus was slowly








heated to 95'C under vacuum. At that temperature the

salt began to decompose and a brown film began to

deposit on the cold finger. The sublimation was ceased

and the brown film was identified as heptafulvalene by
1
nmr.

Attempted Matrix Isolation of the Sodium Salt of 2,4,6-
Cycloheptatriene p-Toluenesulfonylhydrazone in Solid
Adamantane

In 25 ml. of cyclohexane, 3 mg. of the sodium salt

of 2,4,6-cycloheptatriene p-toluenesulfonylhydrazone

and 100 mg. of adamantane were dissolved. The solvent

was rotary evaporated and the residual powder was

pressed into a pellet with an ir pellet maker. However,

the ultraviolet spectrum of the pellet-matrix showed no

indication of the salt even though when redissolved in

cyclohexane the liquid solution gave the reported

literature ultraviolet spectrum.1

Other attempts were made to increase the solubility

of the salt in the adamantane with the Crown ether,

15-Crown-5. Again, the pellet matrix showed no indication

of the salt in the ultraviolet spectrum. Further attempts

yielded the same negative results.

The Matrix Photolysis of the Sodium Salt of 2,4,6-
Cycloheptatriene p-Toluenesulfonylhydrazone in
2-Methyltetrahydrofuran

A saturated solution of the sodium salt of 2,4,6-

cycloheptatriene p-toluenesulfonylhydrazone was made

in 2-methyltetrahydrofuran by stirring an excess amount

of the salt in the solvent for 30 minutes. The solution








was filtered into a 1 cm. matrix ultraviolet cell and

degassed by the freeze-thaw method three times. The cell

was sealed and placed into a quartz windowed dewar con-

taining liquid nitrogen. Once the solution had frozen

into a clear glass, the ultraviolet spectrum was taken

and absorptions were noted at 250 nm. and 320 nm. The

matrix was photolyzed for 15 minute intervals up to one

hour, and after each photolysis period the ultraviolet

spectrum was retaken. There appeared to a gradual dis-

appearance of the ultraviolet spectrum of the starting

material and the appearance of a developing spectrum of

an intermediate (shoulder at 390 nm.).

At the end of the photolysis, the matrix was allowed

to melt (gas evolution and darkening of the solution

was noted upon melting), the solution was refrozen,

and the final spectrum was taken. The final ultraviolet

spectrum indicated the presence of starting material

and heptafulvalene (determined by comparison to literature

spectra and an authentic sample). Additionally, when

the matrix solution was spotted on a silica gel thin

layer chromatography plate and developed in pentane,

the only compound that eluted, corresponded to hepta-

fulvalene.








The EPR-Matrix Photolysis of Some Benzannelated Sodium
Salts of 2,4,6-Cycloheptatriene p-Toluenesulfonyl-
hydrazone in 2-Methyltetrahydrofuran

The sodium salts of 4,5-benzo-2-methyl-2,4,6-

cycloheptatriene p-toluenesulfonylhydrazone, 2,3,4,5-

dibenzo-2,4,6-cycloheptatriene p-toluenesulfonyl-

hydrazone, and 2,3,6,7-dibenzo-2,4,6-cycloheptatriene

p-toluenesulfonylhydrazone had been previously

prepared in our laboratory. All of the salts were

stirred with pentane and filtered before use. A
3
3 X 10- molar solution of each was made in dry

2-methyltetrahydrofuran. The matrix solution was

placed in a standard EPR tube and quartz dewar filled

with liquid nitrogen. When the 2-methyltetrahydro-

furan had frozen in a clear matrix, the sample was

photolyzed for 30 minutes. The EPR spectrum was

taken and no apparent absorption was noted. The sample

was photolyzed for another 30 minutes and the EPR

spectrum was retaken. Again, no apparent absorption

was indicated in the EPR spectrum. All samples showed

gas evolution upon warming the matrix to melt.

The EPR-Matrix Photolysis of the Sodium Salt of 2,4,6-
Cycloheptatriene p-Toluenesulfonylhydrazone in
2-Methyltetrahydrofuran

The sodium salt of 2,4,6-cycloheptatriene

p-toluenesulfonylhydrazone (0.010 g., 3 X 10-5 mole)

was dissolved in 10 ml. of dry 2-methyltetrahydrofuran
-3
to make a 3 X 103 molar solution. The solution was

placed in a standard EPR tube and quartz dewar filled








with liquid nitrogen. Once the 2-methyltetrahydrofuran

had frozen in a clear matrix, the sample was photolyzed

for 30 minutes. The EPR spectrum was taken and no

apparent absorption was noted. The sample was

photolyzed for another 30 minutes and the EPR spectrum

was retaken. Again, no apparent absorption was

indicated in the EPR spectrum. The sample did show

gas evolution and darkening of the solution upon

warming the matrix to melt. A sample of the solution

was spotted on a silica gel thin layer chromatography

plate and the plate developed in pentane. The only

spot that eluted corresponded to heptafulvalene.

Reaction of 1-, 2-, and 3-Bromocycloheptatrienes with 1-
Potassium Mentholate in the Presence of 1,3-Diphenyl-
isobenzofuran

General procedure. To a three neck 50 ml. round

bottom flask equipped with a magnetic stirrer, a condenser,

a nitrogen inlet, and an addition funnel, 1.0 g.

(0.003 mole) of 1-potassium mentholate and 1.0 g.

(0.004 mole) of 1,3-diphenylisobenzofuran were added.

The bromocycloheptatrienes (0.5 g., 0.003 mole) were

weighed and placed in the addition funnel along with

2 ml. of dry tetrahydrofuran. The flask was then

brought to temperature (see Table 1, page 28) and the

dropwise addition of the bromide was started. The mixture

was allowed to react for the indicated time (see Table 1,

page 28) and then poured into 75 ml. of water to which had








been added solid maleic anhydride (to react with the excess

1,3-diphenylisobenzofuran). The aqueous solution was

extracted three times with 100 ml. portions of ether.

The ether extracts were dried (MgSO4) and rotary

evaporated to dryness. The resulting solid was

chromatographed on grade 2 basic alumina with pentane.

Elution of the allene adduct (24) formed between

cycloheptatetraene (9) and 1,3-diphenylisobenzofuran (23)

was followed by using a handheld ultraviolet light.

H nmr 5 ppm. 3.68 (m, 1 H), 4.75 (q, 1 H),

5.85 (m, 1 H), 6.13 (m, 1 H), 6.20 (m, 2 H), 7.0-

7.8 (m, Ph, 14 H).

Note: Due to the large discrepancy of the specific

rotations between -400C and 0C (see Table 1, page 28),

the latter two samples of the -40C runs and the 0C

and 25'C runs were subjected to low pressure column

chromatography on a 100g., one inch, silica column

using hexane as the elutant. However, the procedure

failed to produce any significant or conclusive results.

Reaction of Optically Active Deuterated 1-, 2-, and 3-
Bromocycloheptatrienes with Potassium tert-Butoxide
in the Presence of 1,3-Diphenylisobenzofuran

General procedure. The 1,3-diphenylisobenzofuran

(1.0 g., 0.004 mole) and the potassium tert-butoxide

(0.5g., 0.004 mole) were weighed and placed in a

three neck, 50 ml., round bottom flask equipped with

a magnetic stirrer, a condenser, a nitrogen inlet tube,








and a capped addition funnel. Solution was effected

with the addition of 10 ml. of dry tetrahydrofuran.

The deuterated bromides were weighed (0.5 g., 0.003 mole),

dissolved in 2 ml. of dry tetrahydrofuran, and placed in

the addition funnel. The flask was then brought to

temperature (see Table 2, page 33) and the dropwise

addition of the bromides was started. After all of the

bromide solution had been added, the reaction was allowed

to stir for the indicated time (see Table 2, page 33).

Then the mixture was transferred to a separatory funnel

with 100 ml. of ether, 50 ml. of water, and solid maleic

anhydride (to react with the excess 1,3-diphenyliso-

benzofuran). The aqueous layer was then separated and

the ether layer was washed twice more with 50 ml. portions

of water. The ether layer was dried (MgS04) and rotary

evaporated to dryness. The crude material was

chromatographed on grade 2 basic alumina with pentane.

The elution of the isomers of the allene adduct (24)

(endo and exo) formed between cycloheptatetraene (9)

and 1,3-diphenylisobenzofuran (23) was followed by

handeld ultraviolet light.
1H nmr 6 ppm. Endo isomer: 3.68 (m, 1 H), 4.75

(q, 1 H), 5.85 (m, 1 H), 6.13 (m, 1 H), 6.20 (m, 2 H),

7.0-7.8 (m, Ph, 14 H). Exo isomer: 2.60 (m, 1 H),

5.15 (q, 1 H), 6.05 (m, 1 H), 6.32 (m, 1 H), 6.43 (m,

2 H), 7.0-7.8 (m, Ph, 14 H).








Reaction of Optically Active Deuterated 1-, 2-, and 3-
Bromocycloheptatrienes with Potassium tert-Butoxide
in the Presence of 1,3-Diphenylisobenzofuran and
Toluene

The 1,3-diphenylisobenzofuran (0.9 g., 0.003 mole)

and the potassium tert-butoxide (0.5 g., 0.004 mole)

were weighed and placed in a three neck, 50 ml. round

bottom flask along with 5 ml. of dry tetrahydrofuran

and 3 ml. of toluene. The flask was equipped with a

magnetic stirrer, a condenser, a nitrogen inlet tube,

a capped addition funnel, and a rubber septum. The

deuterated bromides were weighed (0.694 g., 0.004 mole),

dissolved in 5 ml. of dry tetrahydrofuran, and placed in

the addition funnel. The mixture was cooled to 0C,

and the bromides were added dropwise over a period of

five minutes. The reaction mixture was allowed to stir

for one hour at 0C after the addition of the bromides.

Maleic anhydride was added until all the excess

furan was reacted (checked by thin layer chromatography).

The reaction mixture was placed in a separatory funnel

with 100 ml. of ether and 50 ml. of water. The water

was separated and the ether layer was then extracted

with 50 ml. of aqueous silver nitrate. A black precipitate

(probably silver) was noted in the aqueous layer.

The ether layer turned yellow. The ether layer was

dried (MgSO ) and evaporated to dryness. The crude

material was chromatographed on a short silica gel








column with 90/10 pentane/ether. This chromatography

yielded 0.184 g. of the exo and endo allene-

diphenylisobenzofuran adducts (24).

These adducts were chromatographed on a silver

nitrate-silica gel column. The silver nitrate-silica

gel column was prepared by dissolving 12.5 g. of silver

nitrate in 150 ml. of water and adding 50 g. of silica

gel. The water was evaporated and the silver nitrate-

silica gel was activated by placing it in an oven at

1300C overnight. The chromatography yielded 0.112 g.

of pure endo product by eluting with 90/10 pentane/ether

(checked by nmr).

Observed rotation. 0.112 g. in 1.1 ml. of methylene

chloride gave 0.100 0.003.

Specific rotation. +0.9820.


Rotations were determined at 589 nm. (Sodium D

line) and in a 1.0 decimeter polarimeter cell.

Reaction of Optically Active Deuterated 1-, 2-, and 3-


Bromocycloheptatrienes with Potassium tert-Butoxide
in the Presence of 1,3-Diphenylisobenzofuran and Styrene


The 1,3-diphenylisobenzofuran (0.9 g., 0.003 mole)

and the potassium tert-butoxide (0.5 g., 0.004 mole)

were weighed and placed in a three neck; 50 ml. round

bottom flask along with 5 ml. of dry tetrahydrofuran

and 3 ml. of freshly distilled styrene. The flask

was equipped with a magnetic stirrer, a condenser, a

nitrogen inlet tube, a capped addition funnel, and a

rubber septum. The deuterated bromides were weighed








(0.694 g., 0.004 mole), dissolved in 5 ml. of dry

tetrahydrofuran, and placed in the addition funnel.

The mixture was cooled to 0C, and the bromides were

added dropwise over a period of five minutes. The

reaction mixture was allowed to stir for one hour

at 0C after the addition of the bromides.

Maleic anhydride was added until all the excess

furan was reacted (checked by thin layer chromatography).

The reaction mixture was placed in a separatory funnel

with 100 ml. of ether and 50 ml. of water. The water

was separated and the ether layer was washed twice

with water (50 ml. portions). The ether layer was

then extracted with 50 ml. of aqueous silver nitrate.

The ether layer was dried (MgSO ) and evaporated. The

excess styrene was removed by vacuum at room temperature.

The crude material was chromatographed on a short

silica gel column with 90/10 pentane/ether. The

spiro adduct (4) of styrene and cycloheptatrienylidene

(2) eluted first followed by the exo and endo allene-

diphenylisobenzofuran adducts (24).

These allene adducts were chromatographed on a

silver nitrate-silica gel column as described previously.

The chromatography yielded 0.119 g. of the pure endo

product by elution with 90/10 pentane/ether.

Observed rotation. 0.119 g. in 1.1 ml. of methylene

chloride gave 0.102 0.0020








Specific rotation. +0.94.

Rotations were determined at 589 nm. (Sodium D

line) and in a 1.0 decimeter polarimeter cell.

Reaction of Optically Active Deuterated 1-, 2-, and 3-
Bromocycloheptatrienes with Potassium tert-Butoxide
in the Presence of Styrene and a Reduced Amount of
1,3-Diphenylisobenzofuran

The 1,3-dihenylisobenzofuran (0.15 g., 0.0005 mole)

and the potassium tert-butoxide (0.5 g., 0.004 mole)

were weighed and placed in a three neck, 50 ml. round

bottom flask along with 5 ml. of dry tetrahydrofuran

and 3 ml. of freshly distilled styrene. The flask was

equipped with a magnetic stirrer, a condenser, a

nitrogen inlet tube, a capped addition funnel, and

a rubber septum. The deuterated bromides were weighed

(0.694 g., 0.004 mole), dissolved in 5 ml. of dry

tetrahydrofuran, and placed in the addition funnel.

The mixture was cooled to 0C, and the bromides were

added dropwise over a period of five minutes. The

reaction mixture was allowed to stir for one hour

at 0C after the addition of the bromides.

Maleic anhydride was added until all the excess

furan was reacted (checked by thin layer chromatography).

The reaction mixture was placed in a separatory

funnel with 100 ml. of ether and 50 ml. of water. The

water was separated and the ether layer was washed

twice with water (50 ml. portions). The ether layer








was then extracted with 50 ml. of aqueous silver nitrate.

The ether layer was dried (MgSO ) and evaporated. The

excess styrene was removed by vacuum at room temperature.

The crude material was chromatographed on a short

silica gel column with 90/10 pentane/ether. The spiro

adduct (4, 0.35 g.) of styrene and cycloheptatrienylidene

(2) eluted first followed by the exo and endo allene-

diphenylisobenzofuran adducts (24).

These allene adducts were chromatographed on a

silver nitrate-silica gel column as described previously.

The chromatography yielded 0.042 g. of the pure endo

product by elution with 90/10 pentane/ether.

Observed rotation. 0.042 g. in 1.2 ml. of methylene

chloride gave 0.046 0.002.

Specific rotation. +1.310.

Rotations were determined at 589 nm. (Sodium D

line) and in a 1.0 decimeter polarimeter cell.

Reaction of Optically Active Deuterated 1-, 2-, and 3-
Bromocycloheptatrienes with Potassium tert-Butoxide
in the Presence of Styrene

The potassium tert-butoxide (1.8 g., 0.016 mole)

was weighed and placed in a three neck, 50 ml. round

bottom flask along with 5 ml. of dry tetrahydrofuran.

The flask was equipped with a magnetic stirrer, a

condenser, a nitrogen inlet tube, a capped addition

funnel, and a rubber septum. The mixture was cooled

to -300C, and 15 ml. of freshly distilled styrene were








added via syringe through the septum. The deuterated

bromides were weighed (2.031 g., 0.012 mole), dissolved

in 5 ml. of dry tetrahydrofuran, and placed in the

addition funnel. The bromides were added dropwise

over a period of one hour. The reaction mixture was

allowed to stir for 5.5 hours at -300C after the addition

of the bromides. The temperature was raised to 0C,

and the reaction was stirred for an additional 2 hours.

The reaction mixture was transferred to a

separatory funnel, and 100 ml. of ether were added.

The organic layer was washed with 50 ml. of dilute

nitric acid, 50 ml. of aqueous silver nitrate, and

finally with 50 ml. of water. The ether layer was

dried (MgSO ) and evaporated. The excess styrene

was removed by vacuum at room temperature. The crude

material was chromatographed on a silica gel column

with 90/10 pentane/ether. After chromatography,

1.1 g. of a yellow oil were obtained. This oil was

mainly 9-phenylspiro(2.6)nona-2,4,6-triene (4,

confirmed by nmr); however, it was too dark in color

to obtain a good rotation. The oil was dissolved in

ether and washed again with aqueous silver nitrate

to remove all traces of heptafulvalene (3). After

drying the ether (MgSO4) and evaporating it, the oil

was rechromatographed on a silica gel column with





69


90/10 pentane/ether. This chromatography yielded 0.550

g. of the pure 9-phenylspiro(2.6)nona-2,4,6-triene

(4) compound.

Observed rotation. 0.550 g. in 5.4 ml. of methylene

chloride gave 0.002 0.0010

Specific rotation. +0.002 0.002'.

Rotations were determined at 589 nm. (Sodium D

line) and in a 1.0 decimeter polarimeter cell.













REFERENCES





1. Jones, W. M.; Ennis, C. L. J. Am. Chem. Soc. 1969,
91, 6391.

2. Waali, E. E.; Jones, W. M. J. Am. Chem. Soc. 1973,
95, 8114.

3. Waali, E. E.; Lewis, J. M.; Lee, D. E.; Allen, E. W.,
III; Chappell, A. K. J. Org. Chem. 1977, 42, 3460.

4. Untch, K.; Matthews, G.; Hylton, T. First International
Symposium on the Chemistry of Nonbenzenoid Aromatic
Compounds 1980.

5. Mayor, C.; Jones, W. M. Tetrahedron Lett. 1977, 3855.

6. Tyner, R. L.; Jones, W. M.; Ohrn, Y.; Sabin, J. R.
J. Am. Chem. Soc. 1974, 96, 3765.

7. West, P. R.; Chapman, 0. L.; LeRoux, J. P. J. Am.
Chem. Soc. 1982, 104, 1779.

8. Molecular Orbital Calculations and Theoretical Ultra-
violet Spectra provided by J. R. Sabin, Department of
Chemistry, University of Florida.

9. Folisch, B. Chem. Ber. 1968, 101, 2717; idid., 1971,
104, 2324.

10. Gilchrist, T. L.; Pearson, D. P. J. Chem. Soc. Perkin
I 1976, 11, 1257.

11. Jordan, J. E.; Pratt, D. W.; Wood, D. E. J. Am. Chem.
Soc. 1974, 96, 5588.

12. Pedersen, C. J. J. Am. Chem. Soc. 1967, 89, 7017;
idid., 1970, 92, 386, 391. Moss, R. A.; Pickiewicz,
J. J. Am. Chem. Soc. 1974, 96, 5632.

13. Scott, D. R.; Allison, J. B. J. Phys. Chem. 1962,
66, 561.





71


14. Trozzolo, A. M.; Gibbons, W. A. J. Am. Chem. Soc.
1967, 89, 239.

15. Trozzolo, A. M.; Murrary, R. W.; Wasserman, E. J.
Am. Chem. Soc. 1962, 84, 4990.

16. Wynberg, H.; Marsman, B. J. J. Org. Chem. 1980,
45, 158.

17. Reference 4. Christensen, L. W.; Waali, E. E.;
Jones, W. M. J. Am. Chem. Soc. 1972, 94, 2118.

18. Kirmse, W.; Loosen, K.; Sluma, H. D. J. Am. Chem.
Soc. 1981, 103, 5935.

19. Mohanakrishnan, P.; Toyal, S. R.; Vaidyanathaswamy,
R.; Devaprabhakara, D. Tetrahedron Lett. 1972, 2871.
Bottini, A. T.; Anderson, B. ibid., 1973, 3321.

20. Harris, J. W.; Jones, W. M. J. Am. Chem. Soc. 1982,
104, 7329.

21. Radlick, P. J. Org. Chem. 1964, 29, 960.

22. Wiley, J. Organic Synthesis 1976, 55, 114.

23. Hauser, C. R.; Tetenbaum, M. T.; Hoffenberg, D. S.
J. Org. Chem. 1958, 23, 861.

24. Newman, M. S. J. Org. Chem. 1961, 26, 2630.

25. Cervinka, 0. Collect. Czech. Chem. Commun. 1965,
30, 1684, 2403.













BIOGRAPHICAL SKETCH


James Wesley Harris, Jr. was born in Richmond,

Virginia on June 28, 1949. He graduated from Henrico

High School in June, 1967, and entered the University

of Richmond in September, 1967. In June, 1971, he

received a Bachelor of Science Degree, graduated with

honors in chemistry, was inducted into Phi Beta Kappa,

and joined the American Chemical Society. He continued

his education at the University of Richmond and while

teaching undergraduate chemistry earned a Master of

Science Degree in May, 1973. He taught at the University

of Richmond for two years before entering the University

of Florida in September, 1975, to earn his Doctor of

Philosophy Degree. At the University of Florida, he

received the DuPont Award for teaching excellence. In

September, 1980, he was employed by Merck Chemical

Company in Elkton, Virginia. While employed at Merck

Chemical Company, he completed his degree requirements

and on September 4, 1982, he married the former Kathy

Kay Lohkamp. Currently, he is employed at Merck Chemical

Company as a Senior Chemist.













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.




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.


Merle A. Battiste
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.




Martin T. Vala
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.




James E. Keesling
Professor of Mathematics







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 for
partial fulfillment of the requirements of the degree of Doctor
of Philosophy.

December, 1983


Dean for Graduate Studies
and Research






















































UNIVERSITY OF FLORIDA


3 1262 08553 5283