• TABLE OF CONTENTS
HIDE
 Title Page
 Dedication
 Acknowledgement
 Table of Contents
 List of Tables
 List of Figures
 List of spectra
 Abstract
 Thermal reorganization of...
 Thermal reorganization of...
 General conclusions of isotope...
 Base-catalyzed elimination of small...
 Experimental
 Appendix
 Bibliography
 Bibliographical sketch






Title: Thermal reorganization and secondary deuterium isotope effect studies of small ring compounds
CITATION PDF VIEWER THUMBNAILS PAGE IMAGE ZOOMABLE
Full Citation
STANDARD VIEW MARC VIEW
Permanent Link: http://ufdc.ufl.edu/UF00098183/00001
 Material Information
Title: Thermal reorganization and secondary deuterium isotope effect studies of small ring compounds
Physical Description: x, 107 leaves. : illus. ; 28 cm.
Language: English
Creator: Alonso, Jorge Humberto
Publication Date: 1973
Copyright Date: 1973
 Subjects
Subject: Deuterium   ( lcsh )
Rearrangements (Chemistry)   ( lcsh )
Dideuteriobiscyclopropylidene   ( lcsh )
Chemistry thesis Ph. D
Dissertations, Academic -- Chemistry -- UF
Genre: bibliography   ( marcgt )
non-fiction   ( marcgt )
 Notes
Thesis: Thesis -- University of Florida.
Bibliography: Bibliography: leaves 102-106.
General Note: Typescript.
General Note: Vita.
Statement of Responsibility: by Jorge H. Alonso.
 Record Information
Bibliographic ID: UF00098183
Volume ID: VID00001
Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: alephbibnum - 000582531
oclc - 14120389
notis - ADB0906

Downloads

This item has the following downloads:

PDF ( 4 MBs ) ( PDF )


Table of Contents
    Title Page
        Page i
    Dedication
        Page ii
    Acknowledgement
        Page iii
    Table of Contents
        Page iv
    List of Tables
        Page v
    List of Figures
        Page vi
    List of spectra
        Page vii
        Page viii
    Abstract
        Page ix
        Page x
    Thermal reorganization of dideuteriobiscyclopylidene
        Page 1
        Page 2
        Page 3
        Page 4
        Page 5
        Page 6
        Page 7
        Page 8
        Page 9
        Page 10
        Page 11
        Page 12
        Page 13
    Thermal reorganization of 1, 1-Divinylcyclopropane
        Page 14
        Page 15
        Page 16
        Page 17
        Page 18
        Page 19
        Page 20
        Page 21
        Page 22
        Page 23
    General conclusions of isotope effect results
        Page 24
        Page 25
        Page 26
        Page 27
        Page 28
        Page 29
        Page 30
        Page 31
    Base-catalyzed elimination of small ring compounds
        Page 32
        Page 33
        Page 34
        Page 35
        Page 36
        Page 37
        Page 38
        Page 39
        Page 40
        Page 41
        Page 42
        Page 43
        Page 44
    Experimental
        Page 45
        Page 46
        Page 47
        Page 48
        Page 49
        Page 50
        Page 51
        Page 52
        Page 53
        Page 54
        Page 55
        Page 56
        Page 57
        Page 58
        Page 59
        Page 60
        Page 61
        Page 62
        Page 63
        Page 64
        Page 65
        Page 66
        Page 67
        Page 68
        Page 69
        Page 70
        Page 71
        Page 72
        Page 73
        Page 74
        Page 75
        Page 76
        Page 77
        Page 78
        Page 79
        Page 80
        Page 81
        Page 82
        Page 83
        Page 84
        Page 85
        Page 86
        Page 87
        Page 88
    Appendix
        Page 89
        Page 90
        Page 91
        Page 92
        Page 93
        Page 94
        Page 95
        Page 96
        Page 97
        Page 98
        Page 99
        Page 100
        Page 101
    Bibliography
        Page 102
        Page 103
        Page 104
        Page 105
        Page 106
    Bibliographical sketch
        Page 107
        Page 108
        Page 109
        Page 110
Full Text












THERMAL REORGANIZATION
AND
SECONDARY DEUTERIUM ISOTOPE EFFECT STUDIES
OF
SMALL RING COMPOUNDS











by

JORGE H. ALONSO







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


UNIVERSITY OF FLORIDA
1973















DEDICATION


To my parents, family and wife.















ACKNOWLEDGMENTS


The author wishes to acknowledge the financial support

of the Ford Foundation-Universidad del Valle fellowship

during the first two years of his graduate studies, financial

assistance as provided by the Graduate School of the Univer-

sity of Florida in the form of a teaching assistantship for

one year, and the research assistantship financed by the

National Science Foundation for two years.

The author also wishes to express his sincere appreci-

ation to Dr. W. R. Dolbier, Jr., his research director, for

all the encouragement, guidance, suggestions and advice

throughout the period of this research program.

The author extends his appreciation to the members of

his supervisory committee. The author is also thankful to

the members of his research group and classmates who made

his stay at the University of Florida most enjoyable.

The author's wife, Rebecca Ann, deserves special thanks

for her helpful assistance in preparing and typing this manu-

script.
















TABLE OF CONTENTS



ACKNOWLEDGMENTS ............... ...................

LIST OF TABLES......................................

LIST OF FIGURES.....................................

LIST OF SPECTRA.....................................

ABSTRACT............................................

CHAPTER

I THERMAL REORGANIZATION OF DIDEUTERIOBISCYCLO-
PROPYLIDENE .................................

II THERMAL REORGANIZATION OF 1,1-DIVINYLCYCLO-
FROPANE..................................... .

III GENERAL CONCLUSIONS OF ISOTOPE EFFECT
RESULTS......................................

IV BASE-CATALYZED ELIMINATION OF SMALL RING
COMPOUNDS...................................

V EXPERIMENTAL................................

APPENDIX ............................................

BIBLIOGRAPHY.................................. ......

BIBLIOGRAPHICAL SKETCH ..............................


Page

iii

v

vi
vii

ix




1


14


24


32

45

89

102

107















LIST OF TABLES


Table Page

1 Chemical shift comparison for undeuterated
and deuterated compounds..................... 6

2 Intramolecular isotope effects for the
dideuteriobiscyclopropylidene thermal reor-
ganization .................................. 9

3 Kinetic data for the thermal reorganization
of 1,1-divinylcyclopropane.................. 20

4 Intramolecular isotope effect for the thermal
reorganization of 1,1-divinylcyclopropane-d2. 21

5 Isotope effects observed in cyclization
studies...................................... 27

6 Observed frequencies of a planar methyl
radical...................................... 30

7 Infrared absorption frequencies of methylene
cyclopropane dO and d ...................... 30

8 nmr comparison of allyl/vinyl hydrogen ratios
for deuterated and undeuterated cyclobutenes. 40















LIST OF FIGURES


Figure Page

1 Energy surface for compounds 1, 6 and 12...... 10

2 Gas chromatographic retention time comparison
for compounds 1, 4, 3 and 6................... 51

3 Arrhenius plot for the reaction 22-21 ....... 63
4 Concentration v8. time plot for the reaction
22--21 at 233.5 .............................. 63

5 Concentration vs. time plot for the reaction
22- 12 at 242 ................................ 64
6 Concentration vs. time plot for the reaction
22-23 at 257 .................................. 64

7 Concentration vs. time plot comparison for
compounds 22 and 22-d 4........................ 69














LIST OF SPECTRA


Spectrum Page
1 nmr of methylenecyclopropane (2)............ 90
2 nmr of dideuteriomethylenecyclopropane
(2-d2) ...................................... 90

3 nmr of dibromospiropentane (2)............... 91
4 nmr of dibromospiropentane-d2 (2-d2)........ 91

5 nmr of vinylidenecyclopropane (4)........... 92
6 nmr of dideuteriovinylidenecyclopropane
(4-d2) ...................................... 92
7 nmr of dideuteriobiscyclopropylidene (1-d2). 93
8 nmr of methylenespiropentane (6) in C6D6.... 93

9 nmr of mixture of dideuterated methylene-
spiropentanes (x, y, z)..................... 94
10 nmr of 1,1-divinylcyclopropane (22).......... 94
11 nmr of 1-vinylcyclopentene (22)............. 95
12 nmr of 1,1-divinylcyclopropane-d2 (22-d2)... 95

13 nmr of mixture of dideuterated 1-vinyl-
cyclopentenes (1, 1) ...................... 96
14 nmr of 1,1-divinylcyclopropane-d4 (22-d4)... 96

15 nmr of 3-vinylmethylenecyclobutane (48).... 97
16 nmr of 4-methylenecyclohexene (49).......... 97

17 nmr of 1-vinylcyclobutene (65)............... 98
18 nmr of 2-vinylbutadiene (66)................. 98








Spectrum Page

19 nmr of cyclopropylcarbinyl bromide-d2...... 99
20 nmr of cyclopropane-l,l-dicarbinol di-p-
toluenesulfonate (28) in DMrSO-d6 .......... 99
21 ir of 1,1-divinylcyclopropane (22)......... 100

22 ir of 1-vinylcyclopentene (23)............. 100

23 ir of 1,1-divinylcyclopropane-d2 (22-d2)... 100
24 ir of 3-vinylmethylenecyclopropane (48) .... 101

25 ir of 4-methylenecyclohexene (4) .......... 101
26 ir of 1-vinylcyclobutene (65).............. 101


viii
















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


THERMAL REORGANIZATION AND SECONDARY DEUTERIUM ISOTOPE
EFFECT STUDIES OF SMALL RING COMPOUNDS

by

Jorge H. Alonso

March, 1973

Chairman: W. R. Dolbier, Jr.
'ajor Departmenti Chemistry

Three topics were investigated:

1) The synthesis of dideuteriobiscyclopropylidene is

reported and its thermal reorganization investigated. The

intramolecular isotope effects for the ring opening step and

for the cyclization step were determined from nmr analyses

of the dideuterated methylenespiropentane products obtained.

The ring opening isotope effect kH/kD was found to be 1.243

0.035 and the cyclization isotope effect kH/kD was found to

be 1.136%0.016. The most satisfactory rationalization of

these results requires a diradical intermediate and transi-

tion states that do not resemble and are less crowded than

either product or starting material.








2) The synthesis of 1,1-divinylcyclopropanes do, d2

and d are reported. The activation parameters for the

thermal reorganization of the undeuterated 1,1-divinylcyclo-

propane were determined. The intramolecular isotope effect

kH/kD determined from the thermal reorganization of 1,1-

divinylcylopropane d2 was found to be 1.062 1 0.018 and was

estimated from nmr analysis of the dideuterated 1-vinyl-

cyclopentene products. The intermolecular isotope effect

kH/kD was estimated from the rate constant ratio of undeu-

terated and tetra deuterated 1,1-divinylcyclopropanes and

was found to be 1.08 i 0.07. A multistep process is favored

on the basis of the isotope effects observed.

3) A novel base-catalyzed ring expansion of some
cyclopropylcarbinyl derivatives to the cyclobutene deriva-

tives was also investigated.















CHAPTER I

THERMAL REORGANIZATION OF DIDEUTERIOBISCYCLOPROPYLIDENE


Introduction

Secondary deuterium isotope effect studies have been

increasingly used in recent years by physical organic

chemists in an effort to clarify reaction mechanisms.

Rationalization of the results obtained has been related to

rehybridization changes,l steric interactions,2 hyperconju-

gation and inductive effects, their use depending on which

concept better explained the experimental data.

Isotope effects derive from changes in force constants

concerning the vibrational frequencies at the deuterated

position in the process of converting reactant to transition

state. A normal isotope effect KH/kD), would be observed if

there were a decrease in vibrational frequency at the deu-

terated position going from reactant to transition state.1

An inverse isotope effect kH/kDc1 would be observed if there

were an increase in vibrational frequency in the same process.

Streitweiser and coworkers1 made an analysis of the o-deute-

rium isotope effect on an SN1 reaction based on statistical

mechanics and indicated that the isotope effect was due

predominantly to the change of a tetrahedral C-H bending









vibration to an out-of-plane deformation in the transition

state.

That secondary deuterium isotope effects may be entirely

steric in origin has also been suggested by other investiga-

tors2'5 who have ascribed small inverse isotope effects to

the effectively small size of the C-D bond relative to the

C-H bond. Therefore a reaction that passes through a tran-

sition state that is sterically more strained than the re-

actant should give an inverse isotope effect kH/kD
Bartell2 has strongly related steric and force constant

change terminologies by using a harmonic approximation.

Crawford and Cameron reported a normal isotope effect

for the cyclization or biradical destruction of trimethylene-

methane and indicated that such an effect may arise from the

ponderal effect of the deuterium, as the CD2 group will have

twice the moment of inertia of the CH2 group. This isotope

effect study remained relatively isolated and uncorrelated

until the recent isotope effect studies on allene (2+2)

cycloadditions by Dolbier and Dai.7 It was also pointed out7

in the study of secondary deuterium isotope effects in allene

cycloadditions that the transition state for product forma-

tion in a two-step process may actually have little or no

bond character and that any observed isotope effect for such

a process might be relief of nonbonded interactions and tor-

sional interactions which would be found in a planar allylic

radical system, but not in a nonplanar one. Such an isotope

effect would be considered a "steric" isotope effect whereby









the hydrogen would rotate more rapidly out of a sterically

congested situation. From these studies much insight has

been gained as to the nature and predictability of such

isotope effects and has increased their credibility as a

mechanistic tool in distinguishing stepwise from concerted

cycloadditions.

Little use of secondary deuterium isotope effects in

the study of thermal reorganizations has been previously

attempted.9 It is then of special interest to be able to

compare the nature of the isotope effects in thermal reor-

ganizations with those obtained from cycloadditions, espe-

cially in those cases where diradicals are thought to be

involved, in order to acquire more knowledge about the nature

of the intermediates as derived from different processes.

Very few molecules with the biscyclopropylidene struc-

ture have been investigated with respect to their thermal

reorganization. The conversion of biscyclopropylidenes to

methylenespiropentanes and the equilibration of methylene-

spiropentanes seem to bear a very close analogy to the mecha-

nisns of simple methylenecyclopropane rearrangements which

have been significantly elucidated in recent years and such

reactions still generate much interest.

There has been considerable literature on the possible

intermediacy of trimethylenemethane diradicals in the bicyclo-

propylidene rearrangement and most recently there has been

much discussion as to the importance of nonplanar trimethyl-

enemethane species in the methylenecyclopropane rearrange-
ment10c
ment.









Dewar has recently determined that the singlet parent

trimethylenemethane species should be most stable in a non-

planar conformation.l HUckel calculations back up the non-

planarity of substituted trimethylenemethane diradicals7 and

accordingly the thermal conversion of substituted biscyclo-

propylidenes can be satisfactorily rationalized only by in-

voking nonplanar diradicals.10c

The probable intermediacy of nonplanar diradicals

raises questions concerning the nature of transition states

for bond-breaking processes leading to such intermediates

and bond-forming processes which destroy them. We believe

that secondary deuterium isotope effects may provide insight

into these processes. Such isotope effects have never been

obtained for the methylenecyclopropane rearrangement.



Results

In the thermal reorganization of dideuteriobiscyclopro-

pylidene three products should be formed irreversibly and

without interconversion. The ratio z/x+y will provide a

measure of (kH/kD)intra for the rate-determining ring cleav-

age process while the ratio x/y will be a measure of

(kH/kD)intra for the product-forming cyclization. In this

CD2



2 D2 D2
l-d2 Y_ z








section we report the synthesis of dideuteriobiscyclopro-
pylidene (l-d2), its thermal reorganization and isotope
effect studies.
Synthesis of Dideuteriobiscyclopropylidene (1-d2)
The synthesis was accomplished following a scheme
analogous to the one reported for the preparation of biscyclo-
propylidene (1).10a The last step in the synthesis, which
offered some difficulties, produced also methylenespiropen-
tane-d2 (6-d2) and dispiro (2.1.2.0) heptane-d2 (J-d2) when

Br
D2 2 > C= C
22 2
2-d2 2-d2 4-d2 1-d2
12
using modified Simmons-Smith reaction conditions.12 Spectral
analysis (Table I) along with comparison of the gas chromato-
graphic retention times with the undeuterated compounds,
CD2

C= -- CC D2'>+
D2 D2
4-d2 I-d2 -d2 6-d2

> C =CH2 'o 2
4 1 5 6

obtained in a parallel reaction, confirmed the structures of
the deuterated compounds.
The preparation of dideuteriomethylenecyclopropane
(2-d2) was accomplished by a series of straightforward
reactions via the deuterated alcohol Z-d2 and deuterated
















Chemical shift comparison
compounds


Table I

for undeuterated and deuterated


triplet, 4H
quintuplet, 2H

broad singlet


triplet, 4H
quintuplet, 2H

broad singlet


sharp singlet

broad singlet


singlet, 4H
singlet, 2H

singlet


a = TMS was the internal standard for nmr's taken in CC14

b = chemical shift taking 67.24 for benzene


2-d2


4


1

1-d2



2


2-d2


CC14


CC14


CC14


C6D6b
C6D6



CC14


CC14


J1.03
15.37

61.03


J1.49
14.73

61.49


J1.25

d1.25


J1.24
J1.97
61.24








bromide 8-d2. The dehydrohalogenation proceeded in good
yield and 2-d2 was used for the dibromocarbene reaction

>--CD20H -- >-CD2Br ---* >==CD2

Z-d2 8-d 2-d

without further purification. A pilot small-scale reaction
of the dehydrohalogenation step indicated that dideuterio-
methylenecyclopropane (2-d2) is produced in 80-90% yield
along with dideuteriocyclobutene (9-d2) that is apparently
formed from abstraction by base of the hydrogen attached to
the t carbon in the cyclopropyl ring, as discussed in
Chapter IV.


CD2 Br -- [> CD2 + 2
iH H
S2-d 2-d2

A previous attempt to prepare dideuteriobiscyclopropyli-
dene (1-d2) via the deuterated reduction of compound 10 gene-
rated from allene and dibromocarbene was made. The prepara-
tion was discontinued due to the very low conversion and
difficulty in purifying 10, which was observed to isomerize
to the novel compound 11 during distillation or when injected
into a thirteen-foot preparative chromatographic Carbowax
column at 1600


Br2








Thermal Reorganization of Biscyclopropylidene (1)

LePerchec and Conia reported the reorganization of un-

deuterated biscyclopropylidene (1)10a at 2100 and observed

some dimerization. We thermolyzed 20-30 mg samples of

biscyclopropylidene (1) at 21420 under vacuum, in a 200 ml

sealed tube which also contained 0.5 ml of benzene or pentane

to minimize dimerization, and we observed no dimerization.

The product of the reaction, methylenespiropentane (6), was

characterized by its spectral analyses. It is worth noting

that the nmr of methylenespiropentane (6) 0c in CClU gives a

spectrum which does not show completely separated cyclopropyl

and allyl hydrogens, while in C6D6 the cyclopropyl and allyl

hydrogens are separated by 21.7 cps which facilitates their

integration and that of the products of the thermal reorgani-

zation of dideuteriobiscyclopropylidene (1-d2).

Thermal Reorganization of Dideuteriobiscyclopropylidene

Three independent samples of dideuteriobiscyclopropyli-

dene (1-d2) were pyrolyzed at 21420 for forty-nine hours.

The ratio of x, y and z products was determined by nmr analy-

sis from the integration of the cyclopropyl hydrogens A,

allylic hydrogens B and vinylic hydrogens C. By solving the

three algebraic equations a through c we found x, y and z.

A+ EB- 2C
a) 4x + 4y + 2z = A x =-

A + C 2B
b) 2x + 2z = B Y= 6


S2B + 2C A
z= 6


c) 2y + 2z = C








The ratios x/z and z/x+y were then easily obtained and

the isotope effect determined.

x A + 2B /kD ring cyclization
y A + C 2B kH D
z 2B + 2C- A kH/k ring cleavage
x+ y 2A-B-C D

Table II
Intramolecular isotope effects for the dideuteriobiscyclopro-
pylidene thermal reorganization

Run x/ z/x+y

1 1.148 1.214
2 1.117 1.232
3 1.142 1.282
Average 1.13610.016 1.24310.035

Heating the reaction run 2 for ten more hours at 2440

gave x/y = 1.028 and z/x+y = 1.230.

A control sample of undeuterated biscyclopropylidene (1)

gave a ratio B/C = 1.001t0.012 and A/B+C = 0.99710.013.

The error for the three independent runs is expressed as

standard deviation. The nmr integration of hydrogens A, B

and C were repeated at least ten times for each run (spectrum

9). The maximum error of the nmr integration expressed as

standard deviation is~1.3%.

Discussion

Biscyclopropylidene (1), methylenespiropentane (6) and
1,2-dimethylenecyclobutane (12) may in principle be intercon-

vertedl0c via the pair of diradicals 13 and 14. The thermo-

dynamic driving force is apparently strong in the direction

1- 6 and in the direction 6-.12, with a significant barrier

between 13 and 14 videe infra). Heating of 6 produced none








of ,1 indicating that the equilibrium 1--6 lies far to the
right. An energy surface figure that accommodates 1, 6 and
12 interconversion is found in the literature.10c


T< <








15






16


k 1


14 12


Figure 1. Energy surface for compounds 1, 6 and 12.

The transition state diradical 15, which is considerably
less crowded than 1, would be formed upon ring cleavage of
biscyclopropylidene 1, a very crowded molecule. An isotope
effect kH/kD>l should be expected for the ring cleavage based
on steric arguments, and accordingly, we found an isotope
effect kH/kD = 1.243. Once the cleavage has taken place, the
diradical 15 may align itself to form the planar diradical 16
or the more stable orthogonal diradical intermediate 13, which
is somewhat more crowded than 15 but still considerably less


~













i16 12 i2

r l
1 6
crowded than 1. Cyclization to obtain methylenespiropentane
would require a 900 rotation of a methylene group. In such
a process the transition state diradical 17, which is less
crowded than 13, would be obtained. The ring cyclization
process 12-.17 should afford an isotope effect kH/kD>1 as
would be expected from Crawford and Cameron's work6 on the
ring cyclization of trimethylenemethane, from Dolbier and
Dai's work, and from our steric arguments. We observed an

isotope effect kH/kD = 1.136 for the ring cyclization process.
Some investigators follow the general thinking that the
transition state has to resemble the starting materials or
the products. What makes our system particularly interesting
is that we are postulating a transition state that does not
resemble, and is less crowded than, either product or starting
material.
Our ring cyclization isotope effect k,/kD = 1.136 can be
compared with Crawford and Cameron's isotope effect kH/kD =
1.37 for the cyclization of trimethylenemethane (19), formed
from the symmetrical breaking of the carbon-nitrogen bonds of
4-methylene-l-pyrazoline-3,3-d2 (18). In Crawford and
Cameron's system6 there are two possible rotations for 19,











-CD-
/CH -N CD2


CD2 D2

18 19 2-d2

since in the planar trimethylenemethane two T orbitals must

rotate to form a (bond, while in our system only one

orbital must rotate to form a o bond. Our isotope effect

should then be approximately the square root of Crawford and

Cameron's isotope effect. This is essentially the case,

since isotope effects are multiplicative. Seltzer's obser-

vation of a normal intermolecular isotope effect13 kH/kD

= 1.27 for the homolytic cleavage of carbon-nitrogen bonds

in compounds 20 and 21 has some analogy to the intramolecular

ring cleavage process of our dideuteriobiscyclopropylidene

(l-d2).

CH CH CH CH

-C-N--N C<)-C-N=N-IC-N-

H H D D

20 21


Change of hybridization in Seltzer's system, sp3 sp ,

is more similar to Streitweiser's solvolytic isotope effect
2 2
studies than it is to our system, sp 2sp radical, but it

is apparent that our steric argument seems to apply to

Seltzer's system; the carbon-nitrogen bond breaks faster in

compound 26 so as to relieve crowdedness in the starting

material.









That the observed isotope effect in the thermal reor-

ganization of dideuteriobiscyclopropylidene (1-d2) is a

kinetic isotope effect was proved by heating the sample used

in run 2 for ten more hours at 2440. We observed that the

ratio z/x+y remained essentially unaltered, while the ratio

x/y decreased to 1.028, an indication that an equilibrium7'14

between x and y was predominant at a higher temperature which

reversed the reaction to the intermediate 13.

The magnitude of the secondary deuterium isotope effect

that we have observed in the thermal reorganization of

dideuteriobiscyclopropylidene (l-d2) seems to exclude the

possibility of a concerted mechanism. A concerted mechanism

for which bond breaking is farther ahead than bond formation

would be expected to give a ratio z/x+y 1 and a ratio x/y

closer to 1 than the one we observed experimentally. A

concerted mechanism for which bond formation in the transi-

tion state is appreciable would be expected to give a ratio

x/y 1 by favoring y with the hydrogens in the vinylic posi-

tion over the more crowded x with the hydrogens in the allylic

position.













CHAPTER II
THERMAL REORGANIZATION OF 1,1-DIVINYLCYCLOPROPANE


Since the product-forming cyclization process of an
intermediate diradical can give rise to a significant normal
secondary deuterium isotope effect, it was our belief that
the intermediacy of a diradical 24 in the conversion of
22--2 could be proved utilizing the dideuterated and tetra-
deuterated species 22-d2 and 22-d4.
The ratio xl/Y1 should provide a measure of (kH/kD)intra
for the product-forming cyclization. The ratio of the rate
constant for the thermolysis of 22 to the rate constant for
the thermolysis of 22-d4 should provide a measure of (kH/kD)
inter for the ring cleavage process.


- CD2
CD2


22-d4


<:_: CD2 -


22-d
2


D2 D2 D2
Xl 1
24 23-d


><7 0-\\








In this chapter we will discuss the synthesis and
thermal reorganization of the novel 1,1-divinylcyclopropanes,
intramolecular and intermolecular isotope effect studies, as

well as the activation parameters of the conversion 22-*23.


Results

Synthesis of Cyclopropane-1,1-diacetic Acid
Cyclopropane-l,l-diacetic acid (21), the starting
material for the synthesis of the compounds 22, 22-d2 and
22-d4, was prepared according to a scheme analogous to the
one reported by Chamboux and coworkers.15
BrHH2C _CH20H 2 HCH20Ts
BrH2C CH20H / CH2OTs

2 CH20H 28 CH2CN

CH20H I / 2CN
~(COCH2CH3 H>< C2Br
j 11 2 3 2
0
26 29

Cyclopropane-l,l-dicarbinol (27) was easily prepared by
treating 2,2-bisbromomethyl-l,3-propanediol (25) with zinc in
refluxine ethanol. Zinc dibromide, a by-product of this
reaction, was completed and separated by filtration by
bubbling ammonia in the alcoholic solution at 00. The LiAlH4
reduction of compound 26 also produced 22 but the yield was
lower. Tosylation of 27 produced the ditosylate 28. Treat-
ment of 28 with LiBr in refluxing acetone formed the dibromide

29. Cyclopropane-l,l-dicarbonitrile O was generated from the








reaction of the ditosylate 28 or the dibromide 29 with
potassium cyanide in dimethyl sulfoxide (DMSO) at 800. The
reaction with the dibromide 29 was cleaner and no decomposi-
tion was observed as in the case of ditosylate 28 which
became dark and whose reaction product was difficult to work
up. Hydrolysis of 30 smoothly produced cyclopropane-l,l-
diacetic acid (31).
Synthesis of 1,1-Divinylcyclopropane
Via sequential (1) reduction by LiAlH4, (2) tosylation,
and (3) elimination using potassium tert-butoxide in DMSO at
260 we successfully prepared the novel l,1-divinylcyclopro-
pane 22. The yield for the elimination step was 45%. The
structure of 22 was verified by its nmr, ir, mass spectrum
and analysis. The nmr (CC14) showed a sharp singlet at
6 0.79 (4H) as well as doublets at J 4.89 (Jbd = 9.5 Jbc
= 1.7 cps, 2H), J 4.91 (Jcd = 18, Jbc = 1.7 cps, 2H), and
I 5.78 (Jbd = 9.5 Jcd = 18 cps, 2H); the ir spectrum showed
strong peaks at 1639, 991 and 902 cm-1 and the mass spectrum
at m/e 94 (P), 79 (base), 77 and 39.
0 Hd Hb
>s CH2COH > H2CH 20H > C H2CH20TsHc
c o-- HH22CH 20 _CH2 H /CH 2CH2OTs

21 21 22

Synthesis of 1,l-Divinylcyclopropane-d4
A path similar to the one followed for the preparation
of 22 easily afforded the novel compound 22-d4, whose nmr
showed a sharp singlet at J 0.79 (4H) and a broad multiple








at J 5.74 (2H); the ir showed peaks at 2967, 2326, 2222, 1590,
-1
1010, 952-893 and 746-709 cm-1
0
--"0 CD2
CH2COH r. CH2CD2OH > CH2CD OTs L/CD 2
V CH2COH CH2CD20H CH2CD20Ts
0
21 2-d4 2-d 22-d4
Synthesis of 1,l-Divinylcyclopropane-d2
The synthesis of the novel ,l1-divinylcyclopropane-d2
(22-d2) was successfully achieved by the following series of
reactions
0
> /CH C"(0 ),CH CH20H __. CH2CHZOH
-CH2C / CH2 OCH2CH3 CH2CD2OH
0 0
24 22 28 22-d
22-d2
rC H2COH CH2CC1 p CH2CH20Ts
L' H2OCH2CH3 C L'- > CH COCH2CH3 ,/ -CH2CD2OTs
0 0



The key step in this synthesis was the selective
reduction of the acyl halide group in )6 to an alcohol group
in 17. By usingl6 NaBH4 in anhydrous dioxane, 3Z was
obtained in 42% yield.
The nmr of 1,1-divinylcyclopropane-d2 (22-d2) in CC14
showed a sharp singlet at J 0.79 (4H), doublets at J4.89 (1H)
and 4.91 (1H), doublet of doublets at J5.78 combined with
a broad multiple at 5.74 (2H): the ir showed peaks at
3058, 2967, 2326, 2222, 1639, 1590, 1427, 991, 952-893, 736
and 725 cm-.








Synthesis of 3-Vinylmethylenecyclobutane

The usefulness of the sequence followed in the prepara-

tion of 1,1-divinylcyclopropane (22) was tested in the pre-
paration of the novel 3-vinylmethylenecyclobutane (48) which

offered special interest in diradical rearrangement studies.
All of the steps followed for the preparation of 48 proceeded

with high efficiency with the exception of the elimination
step which gave 48 in N40% yield. The structure of 48 was
0
--CN -- H --CHH CH2OTs

40 41 42 42

_- = -CH2CN_ H2 c0H 2--CH220H

44 4 46

=_0. --2CH2OTs

47 48
verified by its spectral characteristics and by its smooth

thermal reorganization to 4-methylenecyclohexene (4)17 when
a small sample of 48 was heated at 2400 in a sealed tube for

nine hours.
H He

H H
Ha
48 42


Thermal Reorganization of 1,1-Divinylcyclopropane
Compound 22 could be converted quantitatively at tempera-
tures above 2300 to 1-vinylcyclopentene (2)18 whose nmr gave

a broad doublet at J 1.93 (2H), a broad multiple at f 2.30-








2.50 (4H), doublets at J 4.94 (Jdg = 9.5, Jde = 2 cps, 1H)

andJ4.96 (Jeg = 18, Jde = 2 cps, 1H), a broad singlet at

J 5.62 (1H), and a doublet of doublets at J6.48 (Jdg = 9.5,

Jeg = 18 cps, 1H); the ir spectrum showed strong peaks at
2924, 990 and 902 cm-1; the uv had Amax 233 m/ (e 22,400).

The thermolysis of 22 was carried out in sealed tubes both

in the gas phase and in dilute benzene or pentane solution

to avoid any dimerization or polymerization.

Activation Parameters
The rate of disappearance of 22 in benzene solution

using pentane as an internal standard was followed by gas

chromatography using a ten-foot Carbowax column. The recor-

der of the gas chromatograph was attached to an Infotronics

Digital Readout System model CRS-100 to facilitate and mini-

mize the error in determining the peak areas.
HHb Hg

r c 233-2570 H


22 23

The conversion 22-.2, as in similar vinylcyclopropane
19
rearrangements,19 was unimolecular and a good Arrhenius plot

was obtained in the temperature range studied. Least-squares

analyses were performed to obtain the best fit of the data.

The activation parameters were calculated from the

kinetic data: Ea = 39.6+1 kcal/mol; log A = 12.5 7;H = 38.6

1t kcal/mol and ASi = -2.8t4 cal/deg.








Table III
Kinetic data for thermal reorganization of 22

Temp (C) Temp (OK) 10-3/Temp (OK) 10-5 K(sec-)

233.5 506.5 1.9743 3.148
242.0 515.0 1.9417 6.023

257.0 530.0 1.8868 18.342

These results indicate that the addition of a second

vinyl group lowers the activation energy for the vinylcyclo-

propane-*cyclopentene rearrangement by 11-12 kcal/mol. The

fact that one observes an activation energy lowering almost

equal to that due to the presence of the first vinyl group

( 14 kcal/mol)19 could be construed as being consistent with

either a concerted process or the intermediacy of diradical
species in the reaction. Our values may be compared with the

values of Ea = 42.7 and log A = 14.76 for the geometrical

isomerization of the spiroheptadiene 50.20





20 51
Intramolecular Isotope Effect

Five independent samples of l,l-divinylcyclopropane-d2

(22-d2) were thermolyzed at 2850 for twenty hours.


CD2 C D
2
21 2-
22-d2 23-d2


~








The relative amounts of compounds xl and yl were

determined by nmr integration of the vinyl hydrogens de, g
and f in either deuterated benzene (C6D6) or CC14 for which

the nmr resolution was slightly better.

Hg

Q\Hd
Hf e

The ratios of vinyl hydrogens f/de and g/de are directly
related to the ratio xl/y1 and consequently to (kH/kD)intra
for the ring cyclization process in the diradical mechanism.

Table IV
Intramolecular isoto-e effect for the thermal reorganization
22-d2

Run g/de f/de

1 1.05810.016 1.060o0.022
2 1.05010.017 1.0560.021

3 1.0550.012 1.054t0.010
4 1.08210.017 1.0770.021

5 1.05810.017 1.06810.022
Average 1.0620.018
An nmr analysis control on the undeuterated 1-vinylcy-
clopentene (23) gave g/de = 0.5005t0.024 and f/de = 0.5005
0.012.
Intermolecular Isotope Effect

The ratio of the rate constants for 22 and 22-d4 is
directly related to (kH/kD)inter for the ring cleavage process








in the conversion 22--2. We simultaneously determined the

rate constants for 22 and 22-d4 at 2420 and found (kHAD)

= 1.0810.07.


CD2

22 22-d4


Discussion

While the identity of the two secondary isotope effects

could be interpreted as deriving from a common transition

state for the rate-determining and the product-forming steps,

and this may be equated with a concerted process, we do not

favor this explanation. Most important, the preponderant

weight of analogy insists that a rate-determining step

process in which a sp2 carbon is transformed into a sp3

carbon should be associated with an inverse kinetic secondary

deuterium isotope effect. The only possible exception to

this observation is the yet anomalous ketene-styrene (2+2)

study of Baldwin and Fleming.21 In light of all past analogies

we favor the interpretation of the results as an indication of

a multistep process. In this case there are two possible

pathways: rate-determining diradical formation or rate-

determining diradical destruction.

In the former situation there is some analogy that a

normal intermolecular isotope effect is to be expected for

the conversion sp2- radical,7,22 in which case our observed

intermolecular isotope effect could be derived from a






23

rate-determining formation of the diradical 24. Thus our

observed intramolecular isotope effect, while being relatively

small, can be understood as deriving from the product-

forming destruction.

In the latter situation, a pre-equilibrium formation of

diradical 24 could result in the rate-determining and product-

forming transition states being one and the same. Thus the

two isotope effects should be nearly identical and resembling

in value those ordinarily observed for diradical cycliza-

tions.7















CHAPTER III

GENERAL CONCLUSIONS OF ISOTOPE EFFECT RESULTS


The determination of the nature of the intermediates in

a stepwise reorganization process is very important in the

elucidation of reaction mechanisms. From our experimental

results in the biscyclopropylidene system it is apparent that

changes in geometry leading to a transition state can be

correlated with the isotope effects observed. Our results

are consistent with previous studies of the secondary deu-

terium isotope effect in allene cycloadditions and biradical

destruction that similarly have allylic intermediates.

Hybridization changes have been used sometimes as a tool

for predicting isotope effects. Thus concerted cycloaddition

reactions, where an increase in vibrational frequency at the

deuterated position going from the reactant to product which

result in kH/kD being <1 in both the intra- and intermolecu-

lar competitions, have been associated with a hybridization
2 3
change sp -,sp The rate-determining step of a two-step
2 2
mechanism involving no change in hybridization (sp --sp

radical) should give at best a small isotope effect. Pryor
22a
and coworker 2a have demonstrated that the conversion

sp2 -sp2 radical not only should, but does, give rise to a

small normal deuterium isotope effect, a fact that was also

observed in Dolbier and Dai's cycloaddition studies.7
24








Intramolecular discrimination in the two-step process

of a cycloaddition reaction takes place in a fast, relatively

low activation energy step where the transition state should

occur early along the reaction coordinate. Dolbier and Dai

pointed out that a simple combination of radicals could be
H H
s n H2 H2C

Step CD2 +


Fast 2
two CH2 CH2
step .-J _

CD2

52 53

thought to have negligible activation energy. Since their

studies indicate that there is an apparent, small but signifi-

cant, activation energy for combination of 52, this could

derive from the rotation of a planar configuration toward the

orthogonal geometry 52 that is necessary for bond formation.

It was then suggested that the isotope effect observed should
not be due to a change in hybridization but due to a relief
of nonbonded interactions and torsional interactions which

would be found in the planar allylic but not in the nonplanar

radical system.7

The normal isotope effect observed by Crawford and

Cameron for the destruction of the trimethylenemethane

diradical remained uncorrelated until Dolbier and Dai's

results on the ring closure of allylic diradicals.7









The correlation between changes of hybridization and

isotope effects observed is at fault in some cases when the

isotope effect studies are put together. In the ring closure

step of the trimethylenemethane diradical a change of hybridi-

zation sp2 radical- sp2 gave an appreciable normal isotope

effect of 1.37, which would be anomalous if compared to Pryor

and coworker's calculations which indicate that a change of
2 2
hybridization sp -sp radical should have a normal isotope

effect.22 The ring closure step in the two-step (2+2) cyclo-

addition reactions having a hybridization change sp2 radical

--sp has a normal isotope effect of 1.14-1.20. The isotope

effect observed for the cyclization step (sp2 radical-sp2 )

in our studies of the dideuteriobiscyclopropylidene thermal

rearrangement, which has a trimethylenemethane environment,

correlates well with Crawford and Cameron's results. Our

1,1-divinylcyclopropane rearrangement gave an isotope effect

of 1.062 for the cyclization step involving a hybridization

change sp2 radical-sp The ring cleavage process for our

biscyclopropylidene rearrangement involved a hybridization

change sp2 -sp2 radical and a normal isotope effect of 1.243

was observed.

A comparison of the isotope effects of the cyclization

process regardless of the hybridization changes shows an

interesting similarity. As previously indicated, a possible

source of isotope effects in the cyclization process which

is also applicable to the ring-opening process might be

relief of nonbonded interactions and torsional interactions

also known as steric effect.









Table V
Isotope effects observed in cyclization studies

kH/kD



1.370


-1.136


S1.140-1.200


1.062

In the cyclization of the parent trimethylenemethane

diradical, which is probably planar, two orbitals must

rotate to form a i bond giving rise to an appreciably large

isotope effect of 1.370 based on steric considerations. In

spite of the predicted inherent stability of the planar tri-

methylenemethane diradical it seems clear that any alkyl

substitution is sufficient to prohibit planarity of the

system. In the thermal rearrangement of dideuteriobiscyclo-

propylidene, the more stable orthogonal intermediate is
favored and therefore only one r orbital must rotate to form

a bond and give consequently a smaller isotope effect of

1.136. The transition states postulated in the rearrangement

of dideuteriobiscyclopropylidene may have little or no

bond character and are accordingly less crowded than either

the starting material or the product. This argument and the








results obtained are in close agreement with the experimen-

tal results for the (2+2) reactions and dimerization of

allenes in which allylic diradicals that do not resemble the

product are suggested. The isotope effect observed is

derived from the rotation of a methylene group from the pla-

nar configuration 52 toward the orthogonal geometry 2

which is necessary for f bond formation.

The intramolecular isotope effects in the (2+2) allene

cycloadditions7 presented values varying from 1.140 to 1.200

which seem to be characteristic of the allyl radical. Our

intramolecular isotope effect 1.062 for the conversion

22--23 could also be explained by assuming that the penta-

dienyl radical 24 does not have the same activation barrier

of rotation as allyl radical 54.' A lower activation barrier

in 24 would be expected since rotation is made easier the

longer the conjugation.







24 4


The real source of the secondary deuterium isotope

effects is a change in the vibrational frequency at the

deuterated position which appears to be closely related to a

steric effect. In going from a very crowded configuration

to a less crowded configuration a normal isotope effect

should be observed. Such an effect was observed in Dolbier









and Dai's studies and in our thermal reorganization studies.

Using the out-of-plane bending frequencies of methylenecyclo-

propane23 as a model for biscyclopropylidene and the out-of-
24
plane bending frequencies of methyl radical as a best

model for the orthogonal allyl radical transition state,

after the fashion of Streitweiser,1 the predicted direction

of the isotope effect for the ring cleavage process is the

same as that found in our experiments (Tables VI and VII).

Besides the hybridization effects and steric effects

discussed previously, hyperconjugation and inductive effects

have been hypothesized as possible origins of secondary

deuterium isotope effects.

Inductive effects have been rationalized on the inter-

pretation that since the carbon-deuterium bond is shorter

than the carbon-hydrogen bond, it will have a higher charge

density and the deuterium will act as an electron-donating

relative to hydrogen. Inductive effects are usually low and

should be of minor importance in the thermal reorganization

of biscyclopropylidene as has been the case for cycloaddition

reactions.

Hyperconjugation and secondary -deuterium isotope

effects have been associated since Shiner3 indicated that

reactions with rate-determining steps that involve carbonium

ion or partial carbonium ion formation are slowed down by

the substitution of a deuterium atom for a hydrogen position

in a hyperconjugative position. The possibility of having

the carbon-hydrogen hyperconjugation resonance structure 55













Observed frequencies of a

CH3

CD3


Table VI

planar methyl radical24
-l
730.3 cm-l

567.0 cm-1


Table VII

Infrared absorption frequencies of
and d 23

Vibration Form of Vibration

4 A1 CH2 deformation

15 B1 CH2 deformation

7 Al CH2 wagging
17 B1 CH2 wagging

21 B2 CH2 twisting

23 E2 CH3 rocking


ethylene cyclopropane dO


C4H6(cm-1)

1436.5

1410.0

1002.6
1125.3

1073.0

748.6


C4D6(cm-1)

1168.0

1122.0

804.2

1017.0

835.5

537.5








could be discarded, based on the fact that an inverse

isotope effect would be observed for the ring cleavage of

dideuteriobiscyclopropylidene (1-d2), by favoring 5Z over

56; or because the geometry for hyperconjugation is not

favorable.25




D H



D H D
56 5Z 2 1-d2


Although hyperconjugation has been associated with

f-secondary deuterium isotope effects,3 a combination of
hyperconjugative and inductive effects could be taken into

consideration as an explanation for the intramolecular

isotope effect observed in the cyclization step of the

thermal reorganization of 1,1-divinylcyclopropane. Hydrogen,

being more electronegative than deuterium, would tend to

conjugate according to the pentadienyl radical structure 58

rather than 59 and then give rise to a normal isotope effect.














CHAPTER IV

BASE-CATALYZED ELIMINATION OF SMALL RING COMPOUNDS


Ring expansion and ring opening products in the cyclo-

propylcarbinyl system as well as ring opening products in
the cyclopropyl system have been reported in the literature.26








>-- H--







Ring expansions via anionic intermediates are less
27
common. Slobodin and Shokhor27 studied the reaction of
l,l-bisbromomethylcyclopropane (2Q) with zinc in ethanol
solution and obtained methylenecyclobutane (60). Mechanisms
28
for this reaction28 have been suggested as a base-induced

rearrangement with zinc metal acting as a base, an inter-
mediate organozinc compound 61, or an electrophilically

induced rearrangement via 62.


32











SCH2Br CH2Br CH H2Br
S2HBr I 0CH2ZnBr CH2 +

2e 60 61 62

The ring opening of the cyclopropylcarbinyl system via
anionic intermediates has been studied more accurately.29.30
Roberts and coworker found29 that the products derived from
the Grignard reagent of cyclopropylcarbinyl bromide 63 had
the allylcarbinyl structure and determined the reversibility
of the rearrangement using 1C or deuterium-labeled allyl-
carbinyl bromide.

CH2=CH-CH2CD2MgBr = -CH2MgBr CH2=CH-CD2CH2MgBr
D2


As an approach to making 1,1-divinylcyclopropane (22)
via the ditosylate 64, we decided to react the ditosylate 28
as a control in the reaction with potassium tert-butoxide in
DMSO. We obtained 1-vinylcyclobutene (65) as the major vola-
tile product. This result led us to work on a different
approach with the eventual preparation of 22, which we
discussed in the previous chapter.

- CCH2OTs CH20Ts -m.
>







We became interested in the ring expansion 28-*6~ mainly

for three reasons,

1. 65 was a new compound.
2. 65 contained an extra carbon.

3. To the best of our knowledge the ring expansion was
of a type previously unknown.

In this chapter we will consider the preparation and
thermal reorganization of 65, mechanistic possibilities that
account for the extra carbon and the ring expansion, and

studies using simpler cyclopropylcarbinyl derivatives in an

attempt to elucidate the mechanism for the ring expansion.


Results and Discussion
Synthesis of 1-Vinylcyclobutene

From the reaction of ditosylate 28 with potassium tert-
butoxide in DXSO, 1-vinylcyclobutene 65 was produced in a

10-15% yield as the major volatile component. The nmr
spectrum of 65 showed a broad multiple at S 2.35-2.70 (4H),

doublets at J 5.05 (Jdf = 10 cps, 1H) and 6 5.08 (Jef = 17.5
cps, 1H), a broad singlet at J 5.86 (1H) and doublet of
doublets at 9 6.27 (Jcf = 10 cps, Jef = 17.5 cps, 1H). The

ir spectrum (neat) showed peaks at 3195, 2899, 2817, 1770,

985, 847 and 769 cm-1; the uv had Xmax 233 mT (E 15800); the
mass spectrum presented a peak m/e = 80 (F). The structure

was confirmed by the reaction of 65 with TCNE to give an adduct
whose nmr was similar to that of the adduct of 23 and TCNE.








Thermal Reorganization of 1-Vinylcyclobutene
Thermolysis of 1-vinylcyclobutene (65) at 2100 for three
hours in a small sealed tube produced 2-vinylbutadiene (66)
which has been prepared,31 although by a totally different
method. The nmr of 66 in CC14 showed a broad multiple at





6566

1 5.10-5.35 combined with a broad singlet at f 5.54 (6H),
doublet of doublets at J6.2-6.8 (2H). The ir spectrum
(CC14) showed peaks at 3077, 2985, 1587, 1418, 1379, 999,
923, 913 and 892 cm-1; the uv had Amax 209 mp (e 13400), and
1233 my (e 9700).
An analysis of the products for the pyrolysis of 1,2-
diacetoxymethylcyclobutane (6L), at temperatures between 450
and 500 leads us to think that previous investigators may
not have observed the intermediacy of 65 due to the high

0
CH2OCCH _
E C 200C3 0
0 + CH2=CH-CH20CH + 66
CH20 CH3 CH2OCCH3
6 CH
6Z 68 H 3
HI

H CH2OCCH CH2
0








temperatures used in the pyrolysis of 67. It seems possible

that 66 was produced from 68 via 62 and 70, although a less

likely eight-membered ring transition state in 68 could also
produce 66 directly.
Mechanistic Analysis for the Conversion 28-65

The base-catalyzed elimination of the ditosylate 28 to
give 65 could proceed by initial t-hydrogen abstraction from

the cyclopropyl ring to form 71 which would then be attacked
by dimsyl anion32 to produce 73. Base-catalyzed elimination

of 73 would generate 65. A variation would be initial attack
of dimsyl anion on 28 to form 72 which would either ring
expand to i7 or suffer-basic elimination to give 74. Base-
catalyzed elimination of 74 would also produce 65.




CH20Ts CCH2Ts H2CH2SCH
CH20Ts H2 D
28 b CH2s2C5 H20Ts

CH CH CH

71 0 74

The key point of the various possibilities is that the
base-catalyzed ring expansion via 28, 72 or 74 seems to
proceed by the novel /-hydrogen abstraction on the cyclopropyl
ring.
The displacement by dimsyl anion of tosylates and bro-
mides, formed from high molecular weight alcohols, has been
reported in only two instances.33 Tosylates and bromides are








known to directly eliminate to produce olefins in the pre-

sence of potassium tert-butoxide in DMSO,3 tosylates

favoring elimination and bromides favoring substitution.

Similarly the base-catalyzed elimination of the mixed
0
CH2 CH 0
CnH2n+1CH2OTs C Cn H2n+CH2SCH ----- C H2n+1CH=CH2
H22n s -2 n 2n+l 2 3n 2n+lD ]S0
DMSO
n=12,15

sulfoxide 7 is known to produce isobutene (Z6).35 Alkenes

are also formed when the mixed sulfoxides are just heated in

DMSO.

H 0 t-BuOK
CH3-C-CH2SCH3 CH3-C=CH2
CH3 DMSO CH3

21 76

The displacement by dimsyl anion of tosylates, prepared

from lower molecular weight alcohols, was further established

when we prepared the mixed sulfoxides 77 in 85% and 78 in

71% yields. The mixed sulfoxide 79 was similarly prepared

0 0 0 CH 0
I n H \3 2
CH (CH2)CH2SCH CH2CH2CH2SCH CH2CH2SCH CH C-CH2SCH
CH3

2 78 22 80

although in a low yield, the reaction being more complex due

to the extreme instability of the starting material, benzyl

tosylate.6 We also attempted to prepare 80, and although

the reaction proceeded as in previous cases, the apparent

high solubility of 80 in water precluded its isolation.








Applying similar conditions for the preparation of
olefin via the mixed sulfoxide 81 on the ditosylate 28
resulted in only traces of 22.


X CH2 OTs
> CH20Ts

28


0
DMSO IH2CH2SCH3 t-BuOK 22
2 2 22
NaH H2CH2SCH DMSO
0
81


Treating 28 with a mixture of sodium hydride in DMSO
and potassium tert-butoxide in DMSO produced only J65 no 22
was observed.
From these results it seems that the tosylate substi-
tution by dimsyl anion is taking place to a lesser degree
than is the ring opening under the reaction conditions for
the conversion 28-65.


D+ +>-
S ~-1:1 2



S-~1-2:9-8 2


-- CH20Ts
82

>-_CH2Br
8


>--CH20PNB
82


CH2X


>--CH20H

7








The low yield of 1-vinylcyclobutene (6) and the novelty
of the rearrangement led us to work on the simple parent
cyclopropylcarbinyl system. Cyclopropylcarbinyl tosylate
(82), cyclopropylcarbinyl bromide (8) and cyclopropylcarbinyl
E-nitrobenzoate (83) were prepared37 and reacted with potas-
sium tert-butoxide in DMSO. The tosylate 82 and the bromide
8 produced cyclobutene (Q) and methylenecyclopropane (2) but
the P-nitrobenzoate gave only cyclopropylcarbinol (2), a
hydrolysis product obtained from the work-up of the nonvola-
tile portion of the reaction.
Basically two mechanisms could be operating under the
reaction conditions for the formation of cyclobutene (2)s
a) a carbenoid mechanism.
H H

> C---x -X C:

b) a base-catalyzed t-hydrogen abstraction in the
cyclopropyl ring.

r CH2X -
H H. B

To distinguish between the two possibilities, we pre-
pared the dideuterated tosylate 82-d2 and the dideuterated
bromide 8-d2. The position of the deuterium in the cyclo-
butene produced by base-catalyzed elimination should clearly
indicate which mechanism is apparently favored.
An allyl/vinyl ratio of 2/2 was observed when both
82-d2 and 8-d2 were treated with potassium tert-butoxide in








DMSO, thus favoring the base-catalyzed Y-hydrogen abstraction
mechanism.
Table VIII
Comparison of allyl/vinyl hydrogen ratios for deuterated and
undeuterated cyclobutenes
nmr
allyl/vinyl







Hj- 4/2
HH





Another possibility which could affect our proposed
mechanism would be a carbonium ring expansion of the tosy-
lates 28, 82 and 82-d2 and of the bromides 8 and 8-d2 in
DMSO previous to attack by the base. Based on our reaction
conditions, we do not favor this explanation. If carbonium
ring expansion were to take place, then cyclobutenes x2 and

Y2 would be obtained from the base-catalyzed elimination of
the intermediate 85. A primary deuterium isotope effect

>--CD2OTs X

82- D2
S-CD2Br 5 2 Y2
8-d2
kH/kD=8 should be observable8 for the elimination step;
since the ratio x2/y2 is directly related to kH/kD we can









easily calculate the theoretical ratio of allyl/vinyl

hydrogens. By solving the algebraic equations a) and b) we

obtain the proportions of x2 and x2, replacing the values

obtained for x2 and y2 in the algebraic equations c) and d).
A ratio 1.12 that should be observable by nmr analysis is

obtained for the allyl/vinyl hydrogen ratio. This ratio is

relatively close to the one observed and proposed in the

anionic ring expansion.

a) x2 + y2 = 1
b) x2x2 = 0.889, Y2 = 0.111
b) x2 y2 =

c) x2 + Y2 = allyl hydrogens
allyl/vinyl = 1.12
d) x2 + Y2 = vinyl hydrogens

Snyder and Soto39 ruled out the existing possibility of
extensive carbonium ion or carbene formation in the reactions
of primary alkylbenzenesulfonates with sodium methoxide in

DMSO and DMF, reaction conditions that are similar to ours.

A third possibility, extensive formation of alkoxydimethyl-

sulfonium cation,40 (CH3)2SO+R, was not ruled out. Although
previous studies on the solvolysis of alkylbenzenesulfonates
indicated the absence of extensive (CH3)2SO R formation under
41
the reaction conditions,41 preferential reaction of alkoxide
with a small equilibrium concentration of this cation was not
precluded. The possibility of (CH3)2SO+R formation which has
been suggested as being responsible for the relative effi-
ciency0 of olefin formation from many alkyl arenesulfonates
in DKSO as a solvent does not modify the key point of our

proposed mechanism for the conversion 28-6J, the base-catalyzed.

i-hydrogen abstraction in the cyclopropyl ring.








We tested the extent of rearrangement of ditosylate 28
in DMSO, prior to the attack by base, by taking an nmr
spectrum of 28 in DYSO-d6 at room temperature after one hour,
reaction time for the conversion 28-65. We did not observe
any peaks in the spectrum of 28 that could be credited to a
cyclobutyl derivative or the intermediacy of (CH3)2SO R. We
should mention nevertheless that some rearrangement or decom-
position is observed in the nmr spectrum of 28 after ten
hours at room temperature; the decomposition is almost com-
plete after thirty-nine hours at room temperature.
We similarly took nmr spectra of the bromide 8-d2, con-
taining traces of DYF, in DMSO-d6 and observed no decomposi-
tion or rearrangement at all after twelve hours at room
temperature some slight decomposition was nevertheless
noticed after one month.
We encountered some limitations to the general use of
the ring expansion to cyclobutene derivatives. The dibromide
29 did not produce 6 and only the substitution product 86,
which had also been obtained as a nonvolatile by-product for
the base-catalyzed reaction of 29, was observed.

CH2Br CH20C(H3 )3
CH2Br > 29 86
CH2OTs 86 + l
[> CH2OTs








The reaction of 1-methylcyclopropylcarbinyl bromide (87)
with potassium tert-butoxide afforded mainly the ether 88;
methylenecyclobutane (60) and 1-methylcyclobutene (89) were
also present in a 2/1 ratio (20% yield). Competition of
primary and secondary hydrogen abstraction by base would
give 60 and 8 respectively. We do not favor the possible
base-catalyzed isomerization42 60 89 since as soon as 60
and 89 are produced they are pumped out of the reaction
mixture.

CH3 CH3
HL-CH2Br --- -CH2OC(CH33+ ~f + Ll

81 88 60

We also prepared the bromide 20 which was subjected to
the same basic conditions as 87. The major products of this
reaction corresponded to a 2/1 mixture (70% yield) of vinyl-
cyclopropane (84) and ethylidenecyclopropane (91); traces of
the corresponding tert-butyl ether 92 were also detected in
the volatile fractions collected in the gas traps.

SCHCH CH
>_-CH-3r --+ H>-- > H + [--CH-OC(CH )2

90 84 91 22

Different methods were followed for the preparation of
the bromides 29, 8, 81 and 90. The dibromide V2 was best
prepared from the reaction of the corresponding ditosylate 28
and LiBr in refluxing acetone. The bromides 8, 8 and 90 were








best prepared from the reaction of their corresponding

alcohols in DMF with triphenyl phosphine and bromine. A

previous attempt to prepare 87 from its corresponding alcohol

93 and PBr3 in ether resulted in the formation of 94.43

3PBr

CH H2C

ether CH 3
Br
22 2



Conclusion
Our experimental data support the base-catalyzed ring

expansion mechanism, via /-hydrogen abstraction, of the

cyclopropylcarbinyl derivatives to the cyclobutene deriva-

tives. Some limitations were encountered in the general

applicability of the reaction by a competing substitution

process and by competing f and /-hydrogen abstractions due

to the presence of additional methyl groups in the cyclopro-

pylcarbinyl system.
















CHAPTER V

EXPERIMENTAL


Melting points were determined on a Thomas-Hoover capil-

lary melting point apparatus. All melting and boiling points

are uncorrected. Elemental analyses were performed by Atlan-

tic Microlab, Inc., Atlanta, Georgia. Preparative glpc was

performed with a Model A-90-P3 Varian Aerograph gas chromato-

graph equipped with a Varian Model G2010 ten-inch strip chart


recorder. Columns

as follows

DMS-20s


Carbowax-lO


Carbowax-13:


Carbowax-15:


DC-703:


used in glpc are characterized and coded


1/4" x 20', 5% dimethyl sulfolane in
60/80 Chromosorb P.

3/8" x 10', 10% Carbowax 1500 on 60/80
Chromosorb P.

3/8" x 13', 10f Carbowax 1500 on 60/80
Chromosorb P.

3/8" x 15', 10? Carbowax 1500 on 60/80
Chromosorb P.

1/4" x 5', 10% Dow Corning Silicone Oil
703 (phenyl methyl) on 60/80 Chromosorb P.


Infrared spectra were recorded on a Perkin-Elmer Model

137. Mass spectra were determined on a Hitachi Model RMU-6E

spectrometer. Ultraviolet spectra were recorded on a Cary 15

Recording Spectrophotometer. Nuclear magnetic resonance (nmr)

spectra were determined on a Varian A-60 A spectrometer with








compounds dissolved in either CC14, acetone-d6, benzene-d6

or otherwise specified, with tetramethylsilane (tms) as an

internal or external reference. Chemical shifts are given
in units of J.
Cyclopropylcarbinol-d2 (Z-d2)
The standard reduction procedure, 37c using LiAlD4, on

cyclopropanecarboxylic acid or cyclopropanecarbonyl chloride

produced Z-d2 in 70% and 50% yield, respectively.

bp: 97-980/250 mm
ir (NaC1 plates): 3333, 2985, 2198, 2083, 1361, 1176,
1099, 971, 909 and 833 cm-1
nmr (CC14): 0.28-0.81 multiple 4H
0.90-1.48 multiple 1H
4.18 singlet 1H
Cyclopropylcarbinyl Bromide-d2 (8-d2)
Two methods were followed for its preparation: a) reac-
tion of the alcohol Z-d2 with PBr3 in ether37c in n70% yield,

and b) reaction of 2-d2 with triphenyl phosphine and bromine

in DMF, conditions that induce substitution without rearrange-
ment,4 in n60% yield.
bps 62-640/70 mm
ir (NaC1 plates): 3077, 3012, 2179, 1689, 1429, 1205,
1073, 966 and 833 cm-1
nmri Jo 0.25-0.92 multiple 4H
1.00-1.50 multiple 1H
Dideuteriomethylenecyclopropane (2-d2)
Small-scale reactions were run by mixing 0.01 mol of

compound 8-d2 with 25 ml DMSO immediately before the reaction.








The solution was slowly added over a period of fifteen
minutes to a three-necked flask containing a magnetic
stirrer and 0.03 mol (excess) of potassium tert-butoxide.

The reaction flask, maintained at room temperature, was
connected to two liquid nitrogen traps which in turn were

attached to a vacuum line (120 mm). Vacuum was applied
(20 mm) for an additional forty-five minutes after the addi-

tion was completed. Gas chromatographic analysis, using a
twenty-foot 5T dimethylsulfolane column in Chromosorb P at

300, indicated that dideuteriomethylenecyclopropane (2-d2)
was obtained in N85% yield, along with dideuteriocyclobutene

(0-d2). The retention times for 2-d2 and 2-d2 were compared
with those of undeuterated methylenecyclopropane and cyclo-
butene obtained in a similar reaction using undeuterated
cyclopropylcarbinyl bromide.

Compound 2-d2:

ir (gas): 2985, 2941, 2778, 2309, 2212, 2037, 1404,


nmr (CC14)
Compound 9-d2s


1109-952, 889, 803-797, 780 and 741-667 cm-1
6 1.03 broad singlet


ir (gas)i 3021, 2899, 2841, 2222, 2165, 1377, 1290,
1143, 862-847, 840 and 725 cm-1
nmr (CC14): 1 2.54 singlet 2H
5.97 singlet 2H
gem-Dibromospiropentane-d2 (3-d2)
Crude dideuteriomethylenecyclopropane (2-d2) obtained
from the reaction of 17.2 g (0.126 mol) of cyclopropylcarbinyl
bromide-d2 (8-d2) in 150 ml DMSO and 44 g (0.39 mol) of








potassium tert-butoxide, using a system similar to the one

described for the small-scale preparation of 2-d2, was added

without further purification to 800 ml of pentane kept at
-750 in a 1000 ml three-necked flask containing 22.4 g

(0.2 mol) of potassium tert-butoxide and equipped with a dry

ice condenser, a mechanical stirrer and a pressure-equalizing

dropping funnel charged with 50.5 g (0.2 mol) of bromoform.

The bromoform was slowly added to the reaction mixture over

a period of two hours; the temperature of the reaction flask

was maintained at -750 for an additional two hours and then

gradually increased to room temperature by removing the

cooling system (dry ice-acetone). The mixture, dark brown,

was poured over 300 ml of ice water, extracted twice with

pentane, and the solvent fractions dried over sodium sulfate.

The solvent was rapidly removed. Distillation of the residue

gave 10 g of compound 2-d2 bp 73-760/25 mm (354 yield based

on cyclopropylcarbinyl bromide-d2). A parallel reaction

using undeuterated methylenecyclopropane (2) produced undeu-

terated gem-dibromospiropentane10a bp 73-760/25 mm in 43%

yield (based on methylenecyclopropane).

Compound 2-d2t

bpt 73-760/25 mm

ir (NaC1 plates): 2976, 2198, 1488, 1418, 1122, 1020,
1009, 966, 881, 858 and 729 cm-1

nmr (CC14): 6 1.24 singlet 4H








gem-Dibromospiropentanel 0a

bp: 73-760/25 mm (lit.10a 63-660/15 mm)

ir (NaCl plates): 2985, 1493, 1427, 1391, 1086, 1053,
1036, 1018, 904, 897, 851 and 689 cm-1
nmr (CC14) J 1.24 singlet 4H
1.97 singlet 2H
Vinylidenecyclopropane-d2 (4-d2)

A 100 ml three-necked flask was equipped with a magnetic

stirrer, a dry nitrogen inlet, a dry ice condenser protected

from moisture by an additional dry ice trap and a CaC12 tube,

and a rubber serum cap. In the flask were placed 6.0 g

(0.0264 mol) of gem-dibromospiropentane-d2 ()-d2) and 5 ml of

anhydrous ether. Through the serum cap was slowly injected

14 ml of 4.8% methyllithium (CH3Li) in ether (0.0305 mol) for

twenty minutes. Upon initial addition of the methyllithium,

cloudiness due to the immediate formation of lithium bromide

was observed. The reaction was followed by gas chromatography

using a fifteen-foot SE-30 column at 1500. All of compound

2-d2 reacted with methyllithium to give almost exclusively
compound 4-d2 along with CH3Br, a by-product of the reaction.

The reaction mixture was stirred for two additional hours

then cooled to -100 and 5 ml of water was injected. The

layers were quickly separated; the water layer was extracted

twice with 10 ml of anhydrous ether. The ether layers were

combined, dried over Na2SO4 and slowly distilled to remove
the bromomethane and most of the ether. The concentrated

residue ,10 ml was flash distilled and collected in a small

flask cooled with dry ice. An analytical sample of








vinylidenecyclopropane-d2 (4-d2) was obtained by gas chroma-
tography using a fifteen-foot SE-30 column at 1100. The

amount of highly volatile 4-d2 recovered was estimated at

1.1 g (60% yield); its retention time was comparable to that
of an undeuterated sample of vinylidenecyclopropane (4)10a

prepared in a parallel reaction.

Compound 4-d2i

ir (gas): 2976, 2865, 2020, 1031-1015, 985, 971
and 826 cm-1

nmr (CC14): J1.49 broad singlet
mass spectra: 69 (P), 67, 41, 40 and 39

Compound 4i10a

ir (gas)s 2976, 2865, 2020, 1449, 1075-1010,
948 and 939 cm-1

nmr (CC14) J 1.49 triplet 4H (J = 3.9 cps)
4.73 quintuplet 2H
mass spectra (m/e): 66 (P), 65, 40 and 39

Dideuteriobiscyclopropylidene (1-d2)
A modified Simmons-Smith reaction was followed for this
step. A 100 ml three-necked flask was equipped with a mag-

netic stirrer, a dry nitrogen inlet, a sprial reflux condenser

protected from moisture by attaching to its top an additional

dry ice trap and a CaC12 tube. The spiral condenser was kept
at -780 by continually circulating isopropanol cooled with

dry ice. The flask was charged with 7.5 g (0.114 mol) of
powdered zinc, 1.23 g (0.114 mol) of cuprous chloride and 10
ml of anhydrous ether. The mixture was refluxed for thirty
minutes and then cooled to room temperature before slowly









adding v10 ml of ether solution containing 4-d2 and 2.34 ml

(0.0288 mol) of diiodomethane. Over a period of three days

and two nights the mixture was refluxed, for a total of

thirty-four daytime hours, and maintained overnight at room

temperature for a total of twenty hours. The contents of

the reaction mixture were poured into 50 ml of ice water

containing 5% HC1 and left to achieve complete hydrolysis

for forty-five minutes. The solution was quickly filtered!

the organic layer was separated, washed with 50 ml of 5%

sodium thiosulfate and 50 ml of water and then dried over

Na2SO4. The ether solution was flash distilled and the vola-

tile products collected in a flask cooled in dry ice. Gas

chromatography using a five-foot DC-703 column at 900 indi-

cated that compound 4-d2 was still partially unreacted and

that new products were formed which were identified as dideu-

teriomethylenespiropentane (6-d2), dispiro (2.1.2.0) heptane

-d2 (J-d2) and dideuteriobiscyclopropylidene (1-d2), in an
approximate ratio of 1:2:2, by the similarity of retention

times to those of authentic undeuterated samples prepared in

a parallel reaction. Small traces of at least two unidenti-

fied products were also found. The retention times were, in

order: CH3Er(ether<4t6

4 \ 1


CH3Br ether benzene
Figure 2. Gas chromatographic retention time comparison for
compounds 1, 4, 5 and 6.








of pure dideuteriobiscyclopropylidene (1-d2) (11.3% yield)

was collected by gas chromatography.

Compound l-d2

ir (gas): 3077, 2985, 2326, 2222-1613, 1266-
1250, 1087-1047, 980, 968, 870-855,
787 and 781-766 cm-1
nmr (C6D6): J 1.25 broad singlet
mass spectra (m/e): 82 (P), 81, 80, 67, 66, 65, 54 and
52
Compound 3-d2

nmr (CC14): 0.57 multiple 3H symmetrical
0.78 multiple 3H
1.14 singlet
mass spectra (m/e)i 96 (P)
Thermolysis of Dideuteriobiscyclopropylidene (1-d2)

Three independent samples were prepared, two using C6D6

as solvent and one using pentane. A general method for the

thermolysis consisted of transferring via vacuum line 0.025 g

of 1-d2 to a 200 ml glass pyrolysis tube. Deuterated benzene
or pentane, 0.3 ml, was similarly transferred via vacuum line
to the glass pyrolysis tube. The pyrolysis tube was degassed,

sealed under vacuum, wrapped with glass wool and heated in a
furnace tube at 2140 for forty-nine hours. After pyrolysis
the narrow part of the pyrolysis tube was chilled at -1900
to condense the sample. This part was quickly cut while cold.
An nmr, when C6D6 was the solvent, indicated no dimer formed

during the thermolysis. The sample was dried by passing it

through MgSO4 via vacuum line. Traces of dispiro (2.1.2.0)









heptane-d2 (5-d2) present in the original sample were

removed after the pyrolysis by gas chromatography using a

five-foot DC-703 column at 900. A pure sample containing

deuterated methylenespiropentanes x, y and z was obtained

and analyzed by nmr (see Table II).

2,2-Dibromomethylenecyclopropane (10)

A 1000 ml three-necked flask was equipped with a mecha-

nical stirrer, a dry ice condenser and a pressure-equalizing

dropping funnel and protected from moisture by attaching an

additional dry ice trap to its top and a CaCl2 tube. The

flask, charged with 800 ml of dried pentane and 22.4 g

(0.187 mol) of potassium tert-butoxide, was cooled to -750

with a dry ice-acetone bath. Slowly 65 g (1.63 mol) of

allene was condensed in the pentane, and 38 g (0.15 mol) of

bromoform contained in the funnel was added to the pentane

solution over a period of four hours. The reaction mixture

turned light yellow upon initial addition of the bromoform

and deep brown at the end of the addition. The reaction

flask was allowed to reach room temperature by removing the

dry ice-acetone bath. The contents of the dark reaction mix-

ture were poured into 600 ml of ice water: the layers quickly

separated. The water layer was extracted with 300 ml of pen-

tane. The combined organic layers were dried over MgS04 and

the solvent quickly evaporated using a rotor water vacuum

evaporator. Approximately 15 g of dark residue remained. An

nmr of this residue showed peaks corresponding to 10 and CHBr3

in a ratio of 1/5.2, along with some other peaks. Flash









distillation of the crude product produced 11 g showing nmr

peaks corresponding to 10 and CHBr3 (in a ratio of 1/7.5).

The yield of 10 was calculated as N~5 from the initial

amount of bromoform used for the reaction. An nmr of frac-

tions from an attempted slow distillation of 10 indicated

the presence of a new peak. When samples containing 11

were injected into a thirteen-foot Carbowax column at 1600

the proportion of 10 decreased and that of the new peak

increased (ratio 1/4.5). The collected new peak showed

spectral characteristics for dibromomethylenecyclopropane

(11). The nmr of both 10 and 11 were similar to the dichloro
45
analogs. 5

Compound 10:

nmr (CC14): 4 2.27 triplet 2H (J = 2.4 cps)
5.63 multiple 1H
6.04 multiple 1H (symmetrical)

Compound 10 + ll:

ir (NaC1 plates): 2941, 1692, 1634, 1460-1391, 1198,
1149, 1042, 1022, 911, 749, 722 and
697 cm-1
Compound 11:

ir (NaC1 plates): 2941, 1692, 1634, 1449, 1372, 1096,
1042, 1022, 911, 735, 722 and 697 cm-1

nmr (CC14): 1.40 singlet

2,2-Eisbromomethyl-l,3-propanediol (25)

This compound was prepared by the improved method of
46
M. Saucier and coworkers.

mp: 109-1100









Cyclopropane-l,l-dicarbinol (27)

This compound was prepared by the method of B. Chamboux
47
and coworkers, and by lithium aluminum hydride reduction

of diethyl cyclopropane-l,1-dicarboxylate 26.48

bp: 1210/12 mm

ir (NaC1 plates): 3322, 2857, 1429, 1212, 1053-990,
917 and 870 cm-1
0
II
nmr (CD3CCD3): d 0.40 singlet 4H
3.52 doublet 4H
4.06 triplet 2H

Cyclopropane-l,l-dicarbinol Di-p-toluenesulfonate (28)

The method of B. Chamboux and coworkers49 was used for

the preparation of this compound.

mp 1140

ir (KBr): 2985, 1942, 1818, 1786, 1667, 1600,
1361, 1326, 1183, 1098, 1047, 1010,
940, 855-838 and 759 cm-1
0
nmr (CD3CCD3): 6 0.63 singlet 4H
2.45 singlet 6H
3.93 singlet 1H
7.44 doublet 2H (J = 8 cps)(mme_
7.77 doublet 2H (J = 8 cps) tca1)

Cyclopropane-1,1-dicarbinyl Dibromide (29)

Dry acetone was distilled from potassium permanganate

and anhydrous potassium carbonate; 42.7 g (0.104 mol) of

cyclopropane-l,l-dicarbinol di-p-toluenesulfonate and 50.5 g

(0.58 mol) of lithium bromide were refluxed in 300 ml of this
acetone, protected from moisture. A magnetic stirrer preven-

ted serious bumping. After the reaction was completed the









the acetone was removed on a rotary evaporator and the

residual oil was taken up in ether. The ether material was

washed well with water and brine and dried over anhydrous

potassium carbonate. Distillation afforded 20 g (84% yield)

of a colorless liquid that was best stored over anhydrous

potassium carbonate.

bp: 91-930/19 mm (lit.50 82-870/20 mm)

ir (NaC1 plates): 3049, 2976, 1429, 1325, 1227, 1054,
1022, 970, 961, 943, 883 and 830 cm-1

nmr (CC14): J 0.9 singlet 4H
3.45 singlet 4H
Cyclopropane-l,l-diacetonitrile (20)
Two methods were followed to prepare this compound.

Method lt The method of Chamboux and coworkers was followed.49

Method 2: To a 2000 ml three-necked flask equipped with a

pressure-equalizing dropping funnel, a reflux condenser and

a mechanical stirrer were added 400 ml of DMSO dried over NaH

and 220 g (3.4 mol) of potassium cyanide. The system was

heated to 700. Slowly 83.6 g (0.366 mol) of cyclopropane-l,1-

dicarbinyl dibromide dissolved in 200 ml of dried DMSO was

added. During the addition the temperature was maintained

at around 80 Once the addition was completed the system

was heated to 1000 and kept at that temperature for twenty-

one hours. The cooled solution was filtered and the precipi-

tate (potassium bromide) was washed with ether. The filtrate

was treated with water and extracted with ether. The com-

bined ether extracts were dried with sodium sulfate. Frac-

tional distillation gave 32 g (73% yield) of a colorless

liquid.






57

bpi 100-1190/0.6-0.8 mm (lit.49 1400/11 mm)
ir (NaC1 plates): 2950, 2262, 1420 and 1031 cm-1
nm4 (CC14): J 0.77 singlet 4H
2.53 singlet 4H
Cyclopropane-l,l-diacetic Acid (31)
This compound was prepared by the method of B. Chamboux
and coworkers.15
bp: 870
ir (KBr): 3333, 2778, 2632, 1724-1695, 1408, 1316-
1176, 1060, 1029, 971-905, 889, 791 and
708 cm-1
nmr (CD3CCD 3): 0.53 singlet 4H
2.42 singlet 4H
5.9-6.9 broad singlet 1H
Cyclopropane-l1,-diethanol (32)
Cyclopropane-l,l-diacetic acid was reduced by the stan-
dard method using lithium aluminum hydride37c in 80% yield.
mp: 68-700
ir (KBr): 3390, 2907, 1481, 1447, 1427, 1366, 1058,
S 1028, 1015, 960, 909 and 865 cm-1
nmr (CD3CCD): J 0.28 singlet 4H
1.51 triplet 4H (J = 7 cps)
3.65 triplet 4H (J = 7 cps)
3.3-3.8 broad singlet 2H
Cyclopropane-l,l-diethanol Di-2-toluenesulfonate (22)

In a 250 ml three-necked flask equipped with a mechanical
stirrer and an ice-salt bath to keep the temperature of the
flask between 0 and -100 were placed 9 g (0.069 mol) of cyclo-
propane-l,l-diethanol and 30.2 g of dry pyridine: 36.6 g









(0.38 mol) of p-toluenesulfonyl chloride was added in small

portions. The reaction mixture was stirred at 00 for one-

half hour and then for another hour letting the system slowly

reach room temperature. The slurry formed was poured over


200 ml of

filtered.

late 29 g

mpt

ir (K


nmr (


ice water the precipitate formed was quickly

Vacuum was applied overnight and the crude tosy-

mp 54-570 was used for the next reaction.

54-570

:Br): 2985, 2924, 1942, 1802, 1667, 1600,
1471, 1429, 1350, 1198, 1176, 1110,
1058, 1021, 998, 935, 885, 855, 820
and 775 cm-1
0
CDCDCCD): 0.26 singlet 4H
1.55 triplet 4H (J = 7 cps)
2.42 singlet 6H
4.1 triplet 4H (J = 7 cps)
7.42 doublet 4H (J = 8 cps) summerr.
7.78 doublet 4H (J = 8 cps) cal)


1,l-Divinylcyclopropane (22)

A 500 ml dried three-necked flask was equipped with a

micro-distillation head, a pressure-equalizing dropping fun-

nel, a bent tube sealed on one end and a magnetic stirrer.

To the flask were added 2.6 g (0.02 mol) of cyclopropane-

1,1-diethanol di-p-toluenesulfonate and 30 ml of dried DMSO.

The sealed tube was charged with 5.6 g (0.05 mol) of potas-

sium tert-butoxide and the pressure-equalizing funnel charged

with 50 ml of DSSO. The reaction flask was kept at room

temperature during the reaction. Vacuum was applied (50 mm)

while the potassium tert-butoxide was added in small portions









during fifteen minutes. The reaction mixture changed color,

from green to brown. The remaining DMSO was added and 50 mm

of vacuum was applied for fifteen more minutes. Full vacuum

was then applied for one hour. The volatile compounds col-

lected in the gas traps were degassed and transferred to a

small tube that was sealed and kept in the refrigerator

until separation was accomplished using a fifteen-foot Carbo-

wax column.

ir (NaC1 plates): 3058, 2967, 1639, 1427, 991, 951,
917, 902 and 863 cm-1
nmr (CC14) S 0.79 singlet 4H
4.89 doublets 2H
(Jbd = 9.5 Jbc = 1.7 cps)
4.91 doublets 2H
(Jcd = 18, Jbc = 1.7 cps)
5.78 doublets 2H
(Jbd = 9.5. Jcd = 18 cps)

mass spectral 94 (P), 79 (base), 77 and 39

elemental anal.
for C7 H10 calc. Hs 10.71 C: 89.29
found H: 10.88 CI 89.12
Thermolysis of 1,1-Divinylcyclopropane (22)

Compound 21, 1-vinylcyclopentene,18 was smoothly

obtained when 1,1-divinylcyclopropane 22 was thermolyzed.

The general method used is described below: 0.030 g of com-

pound 22 was transferred to a 900 ml pyrolysis tube via vacu-

um line. Pentane or benzene, 0.3 ml, was similarly trans-

ferred via vacuum line to the pyrolysis tube. The sample was

degassed and sealed under full vacuum. The pyrolysis tube









was covered with glass wool and heated in a tube furnace at

2850 for twenty hours. The temperature of the furnace was

lowered and when it reached 1000 the tube was removed. The

narrow part of the pyrolysis tube was chilled at -1900 to

condense the sample and quickly cut while cold. The pyroly-

sis product was separated by gas chromatography using a ten-

foot Carbowax column at 800. To the collected sample 0.3 ml

of CCl4 was added to take an nmr. If traces of water were

present after the nmr was taken, they were eliminated by

passing the solution through anhydrous MgS04 via vacuum line.

Yields were M90%.

Compound 231

ir (NaCl plates)s 2924, 2825, 1639, 1587, 990, 902
and 818 cm"l


nmr (CCl4),











uv



mass spectra (m/e),

elemental anal.
for C7H10:


6 1.93 broad doublet 2H
2.30-2.50 broad multiple 4H
4.94 doublets 1H
(Jdg 9.5. Jde = 2 cps)
4.96 doublets 1H
(Jeg = 18, de = 2 cps)
5.62 broad singlet 1H
6.48 doublet of doublets 1H
(Jdg = 9.5, Jg = 18 cps)

Amax 233 my (e 22,400),A240 my
(( 16,650),A 227 mu (6 21,111) and
2202 my (e 12,200)

94 (P)


calc. H: 10.71 Ci 89.29
found Hi 10.88 C: 89.14









Reaction of 1-Vinylcyclopentene (f2) with TCNE

Small portions of tetracyanoethylene (TCNE) were slowly

added to an nmr tube containing 0.03 g of compound 23 in

C6D6. The nmr peaks corresponding to 23 disappeared and

those corresponding to the Diels-Adler adduct appeared. The

solvent C6D6 was quickly evaporated and an ir of the crude

product was taken.

ir (NaC1 plates): 2778, 2174, 1399, 1235 and 850 cm-1

nmr (C6D6), ) 0.9-2.9 multiplets 9H
4.5-4.7 multiple 1H

Rate Constant for Compound 22

A standard sample was prepared by condensing via vacuum

line in a 10 ml tube 3 ml of benzene (to be used as solvent),

0.03 g of 1,1-divinylcyclopropane and 0.03 g of n-heptane

(to be used as internal standard). The homogenized content

of the 10 ml tube was divided into six portions, each placed

in a 3 ml tube and sealed at atmospheric pressure. The

tubes were heated in a well-insulated, well-stirred bath

containing GE-SF-1093 (100) silicone fluid. This oil did not

decompose even at the highest temperature used in our study.

The bath was heated by one 500-W bar heater to a temperature

100 below that desired. A smaller 100-W coiled-wire heater

enclosed in glass tubing was used in conjunction with a

Hallikainen platinum resistance thermometer (model 1146) and

a Hallikainen Resistotrol temperature regulator, off-on type,

to bring the temperature to the desired value and to maintain

it there. Temperatures were monitored using a calibrated









iron-constantan thermocouple in conjunction with a Honeywell

rodel 2702 potentiometer. Each tube was withdrawn from the

bath at an appropriate time, cooled in dry ice, opened, and

the contents analyzed by gas chromatography using a ten-foot

Carbowax column at 800. The gas chromatograph was attached

to an Infotronics Digital Readout System Model CRS-100 to

facilitate and minimize the error in the determination of

the peak areas. Each tube's gas chromatographic result

constituted a single point in determining a rate constant.

Rate constants at three different temperatures were obtained

(see Table III).

Temp(C) Temp(oK) 103/Temp(OK) k(10-5)sec-1

233.5 506.5 1.9743 3.148
242.0 515.0 1.9417 6.023
257.0 530.0 1.8868 18.342

Cyclopropane-l,l-diethanol-d4 (32-d4)

The standard method of reduction,5 using LiAlD4, pro-

duced 32-d4 in A80% yield.
0


nmr (CD3CCD 3), 0.28 singlet 4H
1.51 broad singlet 4H
3.4 broad singlet 2H

Cyclopropane-l,l-diethanol-d4 Di-D-toluenesulfonate (32-d4)

The method used for the preparation of compound 13 was

followed for the preparation of 23-d4. The crude ditosylate

was used in the elimination step without further purifica-

tion.











1.3 \
0
1.2

1.1

1.0

0.9

o 0.8

0.7

0.6

0.5

0.4
1.90 1.95 2.00

T(10 ()

Figure 3. Arrhenius plot for the reaction 22--23.






2








2 4 6 8 10 12 14 16

t(sec 103)
Figure 4. Concentration vs. time plot for the reaction 22-*23
at 233.50.




























Figu


pZI


o 2






1
5 lo 15 20
t(sec 103)
ire 5. Concentration vs. time plot for the react
at 2420.





3




0


0


1 2 3 4 5 6 7
t(sec 103)
re 6. Concentration vs. time plot for the react


ion 2-232


at 2570.


Figu


;ion 22--, 2









l,l-Divinylcyclopropane-d4 (22-d4)

The same procedure used in the preparation of 22 from

22 was followed to prepare 22-d4 from 33-d4.
ir (gas): 2967, 2326, 2222, 1590, 1010, 952-893 and
746-709 cm-1
nmr (CC14): J 0.79 singlet 4H
5.74 broad multiple 2H
Cyclopropane-l,l-diacetic Anhydride (34)

This compound was prepared by refluxing 19 g (0.12 mol)
of cyclopropane-l,l-diacetic acid (31)15 with 30 g (0.38 mol)

of acetyl chloride for two hours. After the reaction was

completed the excess of acetyl chloride was mostly distilled

off. Then full vacuum was applied to eliminate last traces
of acetyl chloride. The paste that remained in the reaction

pot was utilized in the next step without further purifica-
tion.
Cyclopropane-l-acetic Acid, 1-Ethyl Acetate (35)

To compound 34 was added 10 ml of absolute ethanol and
the mixture was refluxed for one and one-half hours. Frac-
tional distillation gave 17.6 g of compound 25 (73.5% yield)

based on compound 31.

bpi 131-1330/ 0.4 mm

ir (NaC1 plates): 3175, 3049, 2950, 2632, 1980, 1739-
1709, 1408, 1370, 1307, 1253-1143,
1036. 966 and 935 cm-1








nmr (CC14): J 0.54 singlet 4H
1.26 triplet 3H (J = 7 cps)
2.36 singlet 2H
2.46 singlet 2H
4.02 quartet 2H (J = 7 cps)
11.56 singlet 1H
Cyclopropane-l-acetylchloride, 1-Ethyl Acetate (26)

To a 250 ml round-bottomed flask with an attached con-
denser was added 17.6 g (0.095 mol) of compound 47. The
flask was ice water cooled while adding slowly 27 g (f = 1.65)
(0.226 mol) of thionyl chloride. The mixture was refluxed for
two hours, then cooled and finally fractionally distilled. A
total of 17.4 g (90%o yield) of compound 36 was obtained.
bpi 950/0.45 mm
ir (NaC1 plates): 3058, 2959, 2933, 1812, 1739, 1399,
1374, 1314, 1250, 1205-0043, 1099,
1064, 1036, 1005, 962, 943, 909, 866,
784, 746 and 697 cm-1
nmr (CC14): d 0.58 singlet 4H
1.26 triplet 3H (J = 7 cps)
2.32 singlet 2H
3.03 singlet 2H
4.11 quartet 2H (J = 7 cps)
Cyclopropane-l-ethanol, 1-Ethyl Acetate (37)

To a well-stirred solution of 16.2 g (0.43 mol) of NaBH4
in 70 ml of anhydrous dioxane was added a solution of 17.4 g

(0.85 mol) of compound 36 in 25 ml of anhydrous dioxane over
a period of forty-five minutes. After refluxing the mixture
for five hours it was cooled and slowly and carefully poured
over 600 ml of ice water and extracted with six 200 ml









portions of ether. After extraction, the aqueous layer was

acidified with HC1 to pH 4 and again extracted with four

more 200 ml portions of ether. The ether extracts were

combined and dried over sodium sulfate. After removal of

the ether the reaction product was fractionally distilled,

giving 6.2 g (424 yield) of the expected product.


bpi

ir (NaCl plates):



nmr (CC14):


95-970/ 0.2 mm or 106-1080/1.75 mm

3448, 2941, 2899, 1727, 1460-1366,
1307, 1250, 1176-1143, 1093, 1053,
1036, 981 and 952 cm1

6 0.42 singlet 4H
1.25 triplet 3H (J = 7 c]
1.63 triplet 2H (J = 6 c]
2.23 singlet 2H
2.72 broad singlet 1H
3.62 triplet 2H (J = 6 c]
4.11 quartet 2H (J = 7 c]


ps)
ps)



ps)
ps)


Cyclopropane-l,l-diethanol-d2 (38)

This compound was prepared by the standard method of

reduction,51 treating compound 3 with LiAlD4 in U80o yield.

ir (KBr): 3448, 3279, 2899, 2841, 2183, 2092, 1473,
1445, 1422, 1348, 1130, 1099, 1055, 1032,
1010, 960, 881 and 810 cm-1


0
H


nmr (CD3CCD3), 6 0.28 singlet 4H
1.51 triplet 2H (J = 7 cps)
1.51 broad singlet 2H
3.33 broad singlet 2H
3.65 triplet 2H (J = 7 cps)

Cyclopropane-l,l-diethanol-d2 Di-p-toluenesulfonate (29)

The preparation of this compound was similar to that of









23. The crude ditosylate was used without further purifi-

cation in the preparation of 22-d2.

1,1-Divinylcyclopropane-d2 (22-d2)

The same procedure used in the preparation of 22 from

31 was followed in the preparation of 22-d2 from 32.

ir (gas): 3058, 2967, 2326, 1639, 1590, 1427, 991,
952-893, 763 and 725 cm-1
nmr (CC14) J 0.79 singlet 1H
4.89 doublets 1H
4.91 doublets 1H
5.74 broad multiple 1H
5.78 doublet of doublets
Thermolysis of l,1-Divinylcyclopropane-d2 (22-d2), Intramo-

cular Isotope Effect

Five independent samples were prepared. The experimen-

tal procedure was similar to the one followed for the thermo-

lysis of 22. Traces of water present in collected samples

containing xl and kl were eliminated by treating the samples

with powdered MgSO4 via vacuum line. Nmr analysis afforded

the ratio of vinyl hydrogens f/de and g/de in the mixture of

xl and Yl (see Table IV, p. 21).
ir (NaC1 plates): 2924, 2825, 2174, 2088, 1639, 1587,
1546, 990, 943, 902, 820-763 and
722 cm-1

Thermolysis of 22 and 22-d4, Intermolecular Isotope Effect

The intermolecular isotope effect kH/kD is directly

related to the ratio of the rate constants of compounds 22

and 22-d4. The preparation of the samples and kinetic pro-

cedure were similar to those followed for the determination






69

of rate constants for compound 22. For the intermolecular

isotope effect compounds 22 and 22-d4 were run simultaneously

at 2420. An intermolecular isotope effect kH/kD = 1.080.07

was obtained. The ratio of the rate constants of 22 and

22-d4 for each point of the kinetic plot was determined.

Point kH/kD Point kH/kD

1 0.9753 4 1.1011
2 1.0844 5 1.1557
3 1.0762 Average: 1.080.07


Figure 7.


* 22
S22-d4


20

t(sec 103)


Concentration vs. time plot comparison for
compounds 22 and 22-d4.








3-Methylenecyclobutanecarbonitrile (40)

This compound was prepared by the method of Caserio and

coworkers51 from allene and acrylonitrile in a 1.5 liter

stainless steel Parr bomb. Toluene was used as a solvent

and traces of hydroquinone were added to prevent dimeriza-

tion of allene.

bps 74-760/31 mm

ir (NaC1 plates): 2967, 2924, 2874, 2237, 1686, 1664,


1408 and 889 cm-1

nmr (CC14) J 3.08 multiple
4.87 multiple

3-Kethylenecyclobutanecarboxylic Acid (41)

The method of Caserio and coworkers51

paring this compound.


was used in pre-


bps 106-1080/8 mm (lit.51 99-1010/9 mm)

ir (NaCl plates)i 2941, 2667, 3236, 1709, 1418, 1333,
1290, 1212, 939 and 881 cm-1

nmr (neat) J 3.00 multiple 5H
4.80 multiple 2H
12.10 singlet 1H

3-Methylenecyclobutanecarbinol (42)
This compound was prepared following the standard meth
of reduction,51 with LiAlH4, in 80% yield.
bpi 101-1030/78 mm

ir (NaC1 plates): 3279, 2899, 1669, 1408, 1044, 1020,
971 and 877 cm-1


0
nmr (CD 3CD 3)


6 2.52 broad multiple 5H
3.50-3.60 broad doublet 3H (2+1)
4.69 multiple 2H


od









3-Methylenecyclobutylcarbinyl Tosylate (43)

A 1000 ml three-necked flask was equipped with a mecha-

nical stirrer, an ice-salt bath to maintain the temperature

between 0 and -100, and was charged with 33 g (0.338 mol) of

3-methylenecyclobutanecarbinol (42) and 67.5 g of dry pyri-

dine. A positive pressure of nitrogen was maintained at all

times and the flask was protected from moisture. In small

portions, 81 g (0.424 mol) of p-toluenesulfonyl chloride was

added with vigorous stirring over a period of forty-five

minutes. The reaction mixture was stirred at NO for one-

half hour and then allowed to reach room temperature. The

contents of the reaction flask were poured over 300 ml of

ice water with vigorous stirring. A precipitate formed and

was quickly filtered. The dried, crude 80.5 g of tosylate

( 95% yield) was used for the next reaction.
mp: 42-44o
0
nmr (CD3CCD): 3 2.1-2.9 broad multiple 5H
2.45 singlet 3H
4.10 broad multiple 2H
4.71 multiple 2H
7.45 doublet 2H (J = 8 cps)
7.81 doublet 2H (J = 8 cps)

3-Kethylenecyclobutaneacetonitrile (44)

To a 500 ml three-necked flask equipped with a pressure-

equalizing dropping funnel, a reflux condenser, a thermometer

and a mechanical stirrer, were added 51.5 g (0.79 mol) of

potassium cyanide and 100 ml of DMSO. The mixture was gently

heated to 70 The dropping funnel was charged with a








suspension of 40 g (0.16 mol) in 50 ml of DMSO and the sus-

pension was slowly added to the reaction flask so that the
temperature was below 800 during the addition. After the

addition was completed the reaction was heated over a period
of fifteen hours at 900. The system was cooled down and

filtered; water was added to the filtrate, and a small organic
layer formed on top. The lower aqueous layer was extracted

with ether. All the fractions containing 3-methylenecyclo-

butaneacetonitrile were mixed and dried over MgS04. Distil-

lation afforded 15.5 g of the product (91.3% yield).

bpi 102-1040/54 mm
ir (NaC1 plates): 2941, 2899, 2247, 1678, 1416, 1062
and 883 cm-1
nmr (CC14) 2.48 doublet 2H
2.2-3.1 broad multiple 5H
4.79 multiple 2H
3-Methylenecyclobutaneacetic Acid (4_)
To a solution of 82 g (1.26 mol) of 86% potassium hydrox-
ide pellets in 450 ml of 50% aqueous ethanol was added 28.7 g
(0.268 mol) of 3-methylenecyclobutaneacetonitrile (44). The
moisture was heated on a steam bath; after about three hours
the ammonia evolution had ceased and the solvent was evapo-
rated under water aspirator vacuum. The residual solid was
dissolved in 100 ml of water and evaporated again to eliminate

completely traces of ethanol. About 100 ml of water was again
added and concentrated hydrochloric acid was slowly added

until the solution became acid to Congo red. Two layers were
formed; the lower aqueous layer was extracted with four








portions of 100 ml of ether. The top layer was mixed with
the ether extracts and dried over anhydrous magnesium sulfate.
Fractional distillation afforded 27.8 g of the product
(M82% yield).

bps 1320/23 mm

ir (NaCl plates): 3058, 2899, 2632, 1709, 1408, 1307,
1212, 943 and 881 cm-1
nmr (CC14)s: 2.54 doublet 2H
2.2-3.2 broad multiple 5H
4.72 multiple 2H
11.64 singlet 1H
3-Methylenecyclobutane Ethanol (46)
This compound was prepared following the standard method
of reduction,51 with LiAIH4, in 91% yield.

bp: 95-960/23 mm
ir (NaC1 plates), 3333, 3049, 2915, 1678, 1410, 1060
and 877 cm-1
0
nmr (CD3CCD 3): 1.73 broad quartet 2H
2.1-2.9 broad multiple 5H
3.53 broad triplet 3H (2+1)
4.70 multiple 2H
nmr (CDCL3), J 1.73 broad quartet 2H
2.1-3.0 broad multiple 6H (5+1)
3.65 broad multiple 2H
4.73 multiple 2H
Tosylate of 3-Yethylenecyclobutane Ethanol (47)
The same method followed for the preparation of 3-methy-
lenecyclobutanecarbinyl tosylate (43) was used in preparing
this compound. Tosylate 47 was obtained in N98% yield as an









oily compound which without further purification was used in

the next step.

ir (NaCl plates): 3030, 2933, 1678, 1600, 1493, 1353,
1190, 1178, 1099, 990, 948, 877,
-1
820 and 775 cm1
0
nmr (CD3CCD3)h J 1.78 broad quartet 2H
2.1-2.9 broad multiple 5H
2.40 singlet 3H
4.01 triplet 2H
4.67 multiple 2H
7.42 doublet 2H
7.80 doublet 2H

3-Vinylmethylenecyclobutane (48)

A 300 ml three-necked flask was equipped with a micro-

distillation head connected to two gas traps kept in liquid

nitrogen and a pressure-equalizing dropping funnel. To the

flask were added 5.6 g (0.05 mol) of potassium tert-butoxide

and 30 ml of DMSO and the dropping funnel charged with 6.7 g

(0.025 mol) of tosylate 47. The contents of the funnel were

slowly added to the flask which was kept at room temperature

during the addition with the aid of an ice water bath.

During the addition, which took approximately twenty minutes,

the reaction mixture initially became greenish, then dark

blue and finally dark brown. During this time 25 mm of pres-

sure was applied to pump the volatile products out of the

reaction mixture. After the addition was over full vacuum

was applied for forty-five minutes. The volatile compounds

collected in the gas traps were degassed and transferred to

a small tube that was sealed and kept in the refrigerator









until separation was accomplished using a ten-foot Carbowax

column. A less volatile fraction remained in the gas traps.

The most volatile fraction N3 g consisted of traces of

dimethyl sulfide (CH SCH3), an unidentified product in 4%

yield, 3-vinylmethylenecyclobutane 48 in 40% yield and some

tert-butyl alcohol formed during the reaction. An nmr of

the less volatile fraction (~1.5 g) showed peaks character-

istic of tert-butyl alcohol, DMSO and the tert-butyl ether

derivative that resulted from direct displacement reaction

of the tosylate 48 by tert-butoxide.

Compound 481

ir (neat), 3040, 2268, 1681, 1639, 1410, 995, 913 and


nmr (CC14):


877 cm-1

6 2.40-3.00
4.68
4.90


broad multiple
multiple
doublets
(Jbc = 9 cps)


4.92 doublets 1H
(Jac = 18 cps)

5.62-6.25 broad multiple 1H

Thermolysis of 3-Vinylmethylenecyclobutane (48)

Approximately 0.08 g of compound 48 was transferred to

a 10 ml glass tube which was then sealed at room temperature.

The tube was wrapped with glass wool and placed in a tube

furnace at 2400 for nine hours (probably more than sufficient).

Spectral characteristics of the product indicated that com-

pound 48 rearranged quantitatively to 4-methylenecyclohexene

(40).1








Compound 49t

ir (neat): 3021, 2899, 2841, 1661, 1449-1418 and
889 cm-1

nmr (CC14), 2.21 broad singlet 4H
2.71 broad singlet 2H
4.68 broad singlet 2H
5.61 broad singlet 2H
1-Vinylcyclobutene (65)

Compound 65 was obtained when 10 g (0.0244 mol) of di-

tosylate 28 in 20 ml of dried DMSO was slowly added (over

a period of fifteen minutes) to a 250 ml reaction flask

equipped with a mechanical stirrer and dry ice and liquid

nitrogen traps connected to a vacuum line (50 mm). Before

the addition the flask was charged with 20 ml of dried DMSO

and 10.9 g (0.0973 mol) of potassium tert-butoxide (1.32

excess if 1 mol of 28 requires 3 mol of potassium tert-

butoxide to generate 1 mol of 65). The reaction mixture

changed from greenish to brown and the flask was maintained

at room temperature using a cooling water bath. After the

addition was completed full vacuum was applied for a period

of forty-five minutes. The volatile products collected in

the traps were identified as traces of dimethyl sulfide

(CH3SCH3), 1-vinylcyclobutane (65) and tert-butyl alcohol.

Separation was accomplished by gas chromatography using a

ten-foot Carbowax column at 1000. The yield of 1-vinylcyclo-

butene (65) varied between 10 and 15% in three different runs.

The reaction mixture remaining in the reaction flask of

one of the runs was extracted with 200 ml of ethyl ether and









200 ml of benzene. The ether extracts were acidified and

washed twice with 50 ml of water. The organic layer was

separated and the ether quickly removed by distillation.

There remained a residue of 3.3 g which contained ethyl

ether, cyclopropane-l,l-dimethyl ditert-butyl ether (86) and

tert-butyl alcohol. The benzene extracts did not contain

any 86. The amount of 86 recovered from the work-up of the

reaction mixture was estimated at 1.27 g (30% yield).

The reaction mixture remaining in another run was addi-

tionally heated at 1500 for twenty hours. The only major

volatile product corresponded to dimethyl sulfide. No more

compound 65 formed.

A variation on the reaction conditions was attempted by

adding 0.01 mol of ditosylate 28 in 30 ml of DMSO to a mix-

ture of 0.023 mol of potassium tert-butoxide and 0.04 mol of

dimsyl anion at room temperature. The dimsyl anion was pre-

pared by heating 0.04 mol of sodium hydride and 15 ml of

dried DMSO at 75-800 for one hour under a nitrogen atmos-

phere. Compound 65 was observed but the yield was not im-

proved.

Another variation was attempted by mixing the ditosylate

with dimsyl anion before addition of potassium tert-butoxide.

No volatile compounds were collected until the reaction mix-

ture was heated at 550 for thirteen hours. No 65 was ob-

served and only traces of 1,1-divinylcyclopropane 22 were

detected along with tert-butyl alcohol, dimethyl sulfide and

traces of other unidentified compounds.









Compound 65:

ir (NaC1 plates):


nmr (CC14):











uv:


mass spectra (YM/e):
Compound 86:


3195, 2899,
769 cm-1

J 2.35-2.70
5.05

5.08

5.86
6.27



Amax 233 my


2817, 985, 847 and


broad multiple L
doublets ]
(Jdf = 10 cps)
doublet I
(Jef = 17.5 cps)
broad singlet 1
doublet of doublets 3
(Jdf = 10 cps,
Jef = 17.5 cps)

(e 15800) and A216 my


(e 8500)

80 (P), 79 and 77


bps 850/18 mm or 630/1 mm

ir (NaCl plates): 2941, 1471, 1387, 1361, 1233, 1199,
1075, 1020, 952 and 893-870 cm-1
nmr (CC14)i 0.32 singlet 4H
1.12 singlet 18H
3.14 singlet 4H
Thermolysis of 1-Vinylcyclobutene 65

Compound 66, 2-vinylbutadiene, was produced essentially
quantitatively when 0.1 g of 65 was placed in a 900 ml pyro-
lysis tube, sealed under vacuum, and heated at 2100 for three

hours. Purification was accomplished using a fifteen-foot
Carbowax column at 1150
ir (CC14): 3077, 2985, 1587, 1418, 1379, 999,
923, 913 and 892 cm-1









nmr (CC14): J 5.10-5.35 broad multiple 6H
5.54 broad singlet
6.20-6.80 doublet of doublets 2H

uvi Amax 208 mp (e 13400) and A233
(E 9700)
mass spectra (m/e): 80 (P), 79 and 77

Reaction of 1-Vinylcyclobutene 65 with TCNE

To an nmr tube containing 0.03 g of compound 65 in a
mixture of CC14 and C6D6 were added small portions of tetra-
cyanoethylene (TCNE). The peaks corresponding to 6 disap-

peared and those corresponding to the Diels-Alder adduct
appeared (similar to the reaction of 23 with TCNE).

nmr (C6D6 + CC14): 61.3-2.5 multiple 7H
2.5-3.2 multiple
4.5-4.7 multiple 1H

?ixed Sulfoxide 77

The mixed sulfoxide 77 was prepared by reacting n-hexyl

tosylate with dimsyl anion. A 200 ml three-necked flask was

equipped with a magnetic stirrer for continuous stirring
throughout the reaction, a dried nitrogen inlet and outlet to

maintain positive pressure of nitrogen, and a rubber serum
cap. The flask was charged with 13 g (0.32 mol) of sodium

hydride 56% oil dispersion; the sodium hydride dispersion
was washed three times with 50 ml portions of dried pentane
to eliminate the oil. Approximately 50 ml of dried DMSO was

injected through the serum cap and the mixture heated to

75-800 for one and one-half hours until hydrogen evolution
ceased, an indication that the dimsyl anion formation was









completed. The reaction flask was then cooled down to room

temperature and 23.4 g (0.091 mol) of crude n-hexyl tosylate

in 40 ml of DMSO was slowly injected over a period of twenty

minutes to the reaction flask containing the freshly prepared

dimsyl anion. The reaction mixture was cooled down to room

temperature during the tosylate addition with the aid of an

ice water cooling bath. A change of color from light red to

purple was observed during the addition. Stirring was con-

tinued for twenty minutes after the addition was completed.

Then 150 ml of water was slowly added, maintaining the reac-

tion flask at room temperature, and the purple solution

became yellow-brown. The reaction mixture was extracted

twice with 250 ml of ethyl ether, the organic layer separated,

dried over Na2SO4 and the ether distilled off; 13.3 g of

crude liquid (85% yield) which showed the spectral charac-

teristics of the sulfoxide 77 remained.

The crude, oily n-hexyl tosylate used in the above reac-

tion was prepared using a procedure similar to the one fol-

lowed for the preparation of tosylate 47 in 92% yield.

Compound 77:

ir (NaC1 plates): 2899, 2841, 1653, 1449 and 1031 cm-1

nmr (CC14): 6 0.73-1.70 multiple 13H
2.49 singlet 3H
2.66 triplet 2H

n-Hexyl Tosylate:

ir (NaC1 plates): 2976, 2924, 1637, 1471, 1370, 1196,
1183, 1099, 935 and 820 cm-1









nmr (CC14): J 2.83-1.70 multiple 11H
2.44 broad singlet 3H
3.98 triplet 2H
7.32 doublet 2H
7.75 doublet 2H
Nixed Sulfoxide 78

This sulfoxide was prepared in 71% yield following a

method similar to the one described for the preparation of

the mixed sulfoxide 72 by reacting benzylcarbinyl tosylate

with dimsyl anion. The nmr and ir spectra of the mixed sul-

foxide 78 were identical to the spectra of an authentic

sample of 8 prepared by the reaction of styrene and dimsyl

anion.52

Mixed Sulfoxide 72

A method similar to the one followed in the preparation

of 7 was used in trying to obtain 79. Benzyl tosylate6

was reacted with dimsyl anion and the reaction product worked

up. The crude product contained trans-stilbene53 in vl0%

yield and mixed sulfoxide 72 in v30% yield. The low yield of

the reaction was probably due to the unstable benzyl tosylate

that partially polymerized after its preparation.6

Cyclopropylcarbinyl Tosylate 8254

In a 200 ml round-bottomed flask was placed 0.06 mol of

cyclopropylcarbinol in 45 ml of anhydrous ether. The tempera-

ture of the flask was lowered to -6 using an ice-salt bath

and 0.06 mol of p-toluenesulfonyl chloride was added to the

ether solution followed by 1.0 g of powdered potassium hydrox-

ide over a period of one hour, maintaining the temperature at









-6o. Then a stopper was put on the mouth of the flask and

the flask was kept in a freezer (-18) for five hours. Ice

and water were added to the reaction mixture; the organic

layer separated and the water layer was extracted with cold

ethyl ether. The ether layers were combined, maintained at
'00 and dried with MgSO4. An oily residue remained after

quick evaporation of the ether. The crude tosylate obtained

was contaminated with p-toluenesulfonyl chloride; the yield

of tosylate was estimated at '60%. Attempts to prepare

solid samples of tosylate 82 were fruitless.5

Reaction of 82 with Potassium tert-Butoxide in DMSO

To the crude tosylate 82 (N0.036 mol) in 30 ml of dried

DMSO was added 0.11 mol of powdered potassium tert-butoxide,

using a system similar to the one used for the preparation

of 1,1-divinylcyclopropane (22). The volative products col-

lected in the gas traps cooled at -1900 with liquid nitrogen

were identified as methylenecyclopropane and cyclobutene (in

an approximate ratio of 1/1), along with traces of isobutene,

which had been observed in the preparation of methylenecyclo-

propane from 2-methylpropenyl chloride and sodium amide.55

Cyclopropylcarbinyl Tosylate-d2 (82-d2)

This compound was prepared using a procedure similar to'

the one followed for the preparation of 82, but using cyclo-

propylcarbinol-d2 (Z-d2).

nmr (CD3Cl): J 0.05-0.80 multiple 4H
0.80-1.50 multiple 1H
2.43 singlet 3H
3.90 doublet 2H
7.10-8.00 doublets 4H









Reaction of Tosylate 82-d2 with Potassium tert-Butoxide

A procedure similar to the one used for the reaction of

82 with potassium tert-butoxide was followed. An nmr of the

cyclobutene formed showed an allyl to vinyl ratio of approxi-

mately 2.3/2 (Table VI).

Cyclopropylcarbinyl p-Nitrobenzoate (81)

This compound56 was prepared in 80% yield from the reac-

tion of p-nitrobenzoyl chloride and cyclopropylcarbinol in

pyridine.

nmr (CD3CCD3): J 0.35-0.75 multiple 4H
0.95-1.60 multiple 1H
4.25 doublet 2H
8.38 singlet 4H

Reaction of 83 with Potassium tert-Butoxide

A system similar to the one used for the reaction of 82

with potassium tert-butoxide was followed. To 2.5 g (0.0117

mol) of 81 in 7 ml of ISO was slowly added 6 g (0.0535 mol)

of powdered potassium tert-butoxide. The reaction flask was

kept at room temperature and vacuum was applied (20 mm) during

the addition. The milky reaction mixture changed colors from

purple to brown during the addition. No volatile compounds

were collected in the gas traps during the addition nor after

full vacuum was applied for an additional forty-five minutes.

The mixture remaining in the reaction flask was treated with

water and extracted with three 200 ml portions of ether. The

ether extracts were combined, dried over MgSO4 and the ether

quickly rotor water pump evaporated. About 1 g of residue

remained which was identified as a mixture of tert-butyl








alcohol and cyclopropylcarbinol: both are hydrolysis products

formed during the work-up of the mixture remaining in the

reaction flask.

Reaction of Cyclopropane-l,l-dicarbinyl Dibromide 29 with
Potassium tert-Butoxide

We used the same system and procedure as the one used

for the preparation of 1-vinylcyclobutene (65) from the reac-

tion of cyclopropane-l,l-dicarbinyl ditosylate (28) in DMSO

with potassium tert-butoxide. The volatile products collected

in the gas traps did not indicate the presence of any 1-vinyl-

cyclobutene, the major volatile product being dimethyl sul-

fide and traces of at least three more compounds. The con-

tents of the reaction flask were treated with ice water and

extracted with ether; the ether layers were combined, dried

over MgSO4 and the ether evaporated. The residue remaining

after this treatment consisted only of tert-butyl alcohol and

cyclopropane-l,l-dicarbinyl dimethyl ether (86) (bp 850/18 mm

or 630/1 mm) which was similarly obtained in the reaction of

28 with potassium tert-butoxide. The yield of recovered 86

was estimated at i40o%.

Yethyl, 1-Wethylcyclopropanecarboxylate

This compound was prepared from the reaction of methyl

methacrylate with diazomethane.57

ir (NaC1 plates): 2950, 1730, 1468, 1437, 1342, 1527,
1198, 1163, 1047, 1031, 957, 877, 763
and 758 cm-1









nmr (neat):, 0.63 multiple 2H (symmetrical)
1.12 multiple 2H
1.28 singlet 3H
3.57 singlet 3H
1-Yethylcyclopropylcarbinol57

This compound was prepared in N80% yield from the LiAlH4
reduction of methyl, 1-methylcyclopropanecarboxylate.57

bp: 126-1270/739 mm

ir (NaCl plates), 3333, 3030-2857, 1449, 1429, 1387,
1070-1010, 919, 862 and 724 cm-1
nmr (CC14), 0.25 multiple 2H(symmetrical)
0.37 multiple 2H
1.13 singlet 3H
2.25 broad singlet 1H
3.21 singlet 2H
1-Yethylcyclopropylcarbinyl Bromide (87)58

A 200 ml three-necked flask was equipped with a magnetic

stirrer, a pressure-equalizing dropping funnel and a nitrogen
inlet and outlet to maintain a positive nitrogen pressure

during the reaction. The reaction flask was placed in an ice
water bath and then charged with 1.6 g (0.0187 mol) of 1-

methylcyclopropylcarbinol, 20 ml of DMF, and 5.25 g (0.02 mol)

of triphenyl phosphine. Bromine, contained in the dropping

funnel, was added drop by drop until an orange color persisted
after one minute of continuous stirring. The reaction was

then immediately connected to a vacuum line and all the vola-
tile products were collected in a 200 ml flask cooled in dry

ice. To the volatile products were added 50 ml of pentane
(bp 20-400) and 10 ml of ice water. Two layers formed; the








organic layer was separated and the water layer washed with

50 ml of pentane. The pentane extracts were combined, dried
over Na2SO4 and then the pentane was evaporated. The dark

residue that remained was flash distilled to give 1 g (46%

yield) of a slightly orange-tinted liquid which was identi-
fied as the bromide 87.

bp: 650/130 mm

nmr (CC14): J 0.58 singlet 4H
1.12 singlet 3H
3.23 singlet 2H
Reaction of 87 with Potassium tert-Butoxide
To 4.15 g (0.037 mol) of potassium tert-butoxide con-
tained in a reaction flask was added 0.8 g (0.054 mol) of

bromide 87 in 20 ml of dried DSSO, a set-up similar to pre-

vious reactions of bromides with potassium tert-butoxide,
although the order of addition was inverted. The major vola-

tile products were identified59 as methylenecyclobutane (60)
and 1-methylcyclobutene (89) along with a small amount of

1-methylcyclopropylcarbinyl methyl ether (88). Work-up of

the contents of the reaction flask produced more mixed ether
88. Approximately 0.07 g (v20% yield) of a mixture of 60 and

82 (in a ratio of 2/1) were obtained. The yield of 88 re-
covered was estimated at 0.38 g (50% yield). An analytical
sample of 88 was collected by gas chromatography using a ten-
foot Carbowax column at 1350.

Compound 88:

ir (NaC1 plates): 2950, 2874, 1456, 1389, 1361, 1212,
1078, 1027, 885 and 865 cm-1









nmr (CCl4): o 0.19 multiple 2H (symmetrical)
0.31 multiple 2H
1.04 singlet 3H
1.12 singlet 9H
3.05 singlet 2H
Cyclopropylcarbinyl Methyl Bromide (90)

The preparation of 90 (containing traces of DMF) was

accomplished in -50% yield by reacting cyclopropylcarbinyl

methyl alcohol60 in DMF with triphenyl phosphine and bromine

following the reaction procedure used for the preparation of

8?.
bps 700/128 mm

nmr (CCl4) 60.40 multiple 2H (symmetrical)
0.75 multiple 2H
0.90-1.40 multiple 1H
1.75 doublet 3H
3.37 quartet 1H
Reaction of 90 with Potassium tert-Butoxide
The same set-up and reaction conditions as the ones used

for the reaction of 87 with potassium tert-butoxide were used.
The major volatile product corresponded to a 2/1 ratio of a

mixture (v70% yield) of compounds 8461 and 962 Traces of

other compounds collected in the gas traps corresponded to
tert-butyl alcohol, dimethyl sulfide and mixed ether 92. No
attempts were made to recover 92 from the nonvolatile reac-
tion mixture remaining in the reaction flask after the vola-

tile products were pumped out of the reaction mixture. An

nmr of the volatile products did not show peaks that could
be credited to 3-methylcyclobutene.








Yethylcyclobutyl Bromide (24)

Unlike the reaction of 1-methylcyclopropylcarbinol with

triphenyl phosphine and bromine that produced the bromide 87,

the reaction of 1-methylcyclopropylcarbinol with phosphorous

tribromide produced the rearranged bromide 94. In a flask

protected from atmospheric moisture were placed 1.72 g (0.02

mol) of 1-methylcyclopropylcarbinol and 10 ml of anhydrous

ether. The solution was cooled in a dry ice-acetone bath and
0.66 ml (0.007 mol) of phosphorous tribromide was slowly added

while stirring. The mixture was allowed to reach room tem-

perature and 2 ml of water was added. The ether layer sepa-

rated and was washed with sodium bicarbonate and dried over

drierite. The ether was removed and the dark residue was

flash distilled to give 2 g of a colorless liquid containing

the bromide 94 (which had been prepared from the reaction

of methylenecyclobutane and hydrobromic acid), along with

traces of the bromide 87.

bp: 710/90 mm (lit.43b 111-1120)

ir (NaC1 plates): 2950, 2865, 1439, 1429, 1372, 1242,
1125, 926-889, 787 and 690 cm-1
nmr (CC14): J1.80-3.00 multiple 6H
1.95 singlet 3H















APPENDIX

nmr and ir Spectra


(Solvents and absorption positions are recorded in the Experi-
mental Section.)






90

























*e 1 nI U ler----

Spectrum 1. nmr of methylenecyclopropane (2)55

i^" --- f v o -- -- r - Tc f


Spectrum 2. nmr of dideuteriomethylenecyclo-
propane (2-d2)




University of Florida Home Page
© 2004 - 2010 University of Florida George A. Smathers Libraries.
All rights reserved.

Acceptable Use, Copyright, and Disclaimer Statement
Last updated October 10, 2010 - - mvs