SYNTHESES OF POLYFUNCTIONAL- CYCLOPROPANES

Th‘smmhe Desréo‘cfph..o. * ~ -

MECHBGAN STATE- UNIVERSITY
yam Chin Kim
~ {1996151} *****

W518

   

I LIBRTI’4 Ix’ g’.

i

 
     

Michigan State
University”

This is to certify that the
thesis entitled

SYN‘I‘HESES OF POLYFUNCTIONAL CYCLOPROPANES

presented by

Yoon Chin Kim

has been accepted towards fulfillment
of the requirements for

M degree in W

'1 M.
4 ' ~ *- ! '“'
-1 ' ere-vxe-L‘k 4k “TI-“inc. k--"'\- ’t

Major professor

Date September 21, 1965

 

0-169

 

 

 

ABSTRACT
SYNTHESES OF POLYFUNCTIONAL CYCLOPROPANES

by Yoon Chin Kim

Non-catalyzed condensation of the systems RR'C=CAB (I, R, R‘ = H,
alkyl, aryl; A, B = CN, COOCgHS, CONHQ) and CHBrYZ (II, Y, Z = CN, COOC2H5,
CONHQ) produced the corresponding polyfunctional cyclopropanes (III,
CRR'CABCYZ).

Condensation of alkylidenemalononitrile (I, R, R' = alkyl; A = B =
CN) and bromomalononitrile (II, Y = Z = CN) produced 3,5-dialkyl-l,l,2,2-
tetracyanocyclopropanes (III, R, R’ = alkyl; A = B = Y = Z = ON). The
reaction appeared to be influenced by steric and electronic effects of the
alkylidenemalononitriles. 5,3-Dicyclopr0pyl-, 5-ethyl-5-n-butyl-, 5,5-
nonamethylene-, 5,5-undecamethylene—, and 5,j-tetradecamethylene-l,l,2,2-
tetracywtnocyclopropanes which previously failed to form by the original
Wideqvist reaction (from the corresponding ketones and bromomalononitrile
in the presence of the iodide ion) were produced by the present method.

A number of 3—aryl-l,l,2,2-tetracyanocyc10pr0panes (III, R = H; R' =
phenyl or substituted phenyl; A=B=Y=Z=CN) were prepared from arylidenemalo-
nitriles (I, R = H; R' = phenyl or substituted phenyl; A = B = CN)and
bromomalononitrile. Introduction of an electron-withdrawing group into the
benzene ring did not seem to particularly facilitate the reaction. However,
introduction of an electron-releasing group did slow down the reaction. The
cyclOpropyl hydrogen of a number of 5-aryl-l,l,2,2-tetracyanocyc10propanes
were shown, in the NMR spectra, to couple to the ortho hydrogens of the

l

Yoon Chin Kim

phenyl group (four bonds away) with coupling constants ranging between
0.6 c.p.s. and l.O c.p.s.
Some 5,3-dialkyl-2-carbethoxy~l,l,2—tricyanocyclopropanes (III, R,R‘ =

CN) and 5-aryl-2-carbethoxy—l,l,2-tricyano-

1
H1

u
on

H

alkyl; A = COOCefls; B -

ll
33
50

ll

cyclOprOpanes (III, R phenyl or substituted phenyl; A = COOCQHS;
B = Y = Z = CN) were prepared by the following two routes: (i) Ethyl
alkylidene- or arylidenecyanoacetates (I, R, R' = H; aLKfih phenyl or sub-
stituted phenyl) and bromomalononitrile and (ii) alkylidene- or arylidene-
malononitriles and ethyl bromocyanoacetate (II, Y = CN; Z = COOCQH5). Route
(i) appeared to be superior to route (ii), since it produced a larger number
of compounds quicker and in better yield than route (ii). tereoisomers
of these compounds were studied by NMR spectra. 5-Methyl—5§;pr0pyl, 5-phenyl-,
and 5-pfmethoxyphenyl-2-carbethoxy-l,l,2-tricyanocyc10pr0pane were shown
to be produced as single stereoisomers in which the ieprOpyl, phenyl, or
pfmethoxyphenyl group is trans to the carbethoxyl group.

A few derivatives of 2-carboxamido—l,l,2-tricyanocyclopr0panes (III,
R,R‘ = H, alkyl, phenyl, or substituted phenyl; A = CONHQ; B = Y = Z =
ON), on boiling in methanol or ethanol, readily cyclized to form the corre-
sponding derivatives of l,5-dicyano—2—imino-j—aza—h—ketobicyclo[5.l.O]hexanes.
The acid treatment of these bicyclo[5.l.O]hexanes produced the correSponding
l,2—dicyano-l,2—carboximidocycloprOpanes.

Dimers of isoprOpylidenemalononitrile, 2-butylidenemalononitrile, and
cycIOpentylidenemalononitrile were prepared by treating the corresponding

monomers with pyridine. They were also obtained as by-products while pre-

paring the monomers.

 

Yoon Chin Kim

2,3—Benzocyclohexylidenemalononitrile and 2,5-benzocyc10pentylidene-
malononitrile, on refluxing with bromomalononitrile, produced 2-bromo-l-
dicyanomethylenetetralin and an unidentified compound C12H6Br2N2, respect-
ively. 2,5-Benzocyclohexylidenemalononitrile was found to condense with
bromomalononitrile, in 80% aqueous ethanol and at room temperature, to
form spiro[2,2,5,5-tetracyanocyclopropane-l,l'-tetralin].

Reaction of bromomalononitrile with ethanol produced l,l-dicyano-2-
amino-2-ethoxyethylene. Reaction of bromomalonitrile with l—nitro-2-
methylpropene gave 5-(2‘-bromo-2'-propyl)-l,l,2,2-tetracyanocyc10propane
and an unidentified compound C7H8Nu03'

Reaction of ethyl sodiocyanoacetate with bromine in carbon tetrachloride
produced l,2,5-tricarbethoxy-l,2,5-tricyanocyclopropane and the reaction of
ethyl sodiocyanoacetate with bromine in carbon disulfide formed compound
ClEH10N2OHSB which was tentatively identified as 5,5-bis(cyanocarbethoxy-

methylene)-l,2,u-trithiacyClopentane.

SYNTHESES OF POLYFUNCTIONAL CYCLOPROPANES

By

Yoon Chin Kim

A THESIS

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

Department of Chemistry

1965

SYNTHESES OF POLYFUNCTIONAL CYCLOPROPANES

By

Yoon Chin Kim

A THESIS

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

Department of Chemistry

1965

To my Father and in
memory of my Mother

ii

ACKNOWLEDGMENT

The author wishes to express his sincere appreciation to Professor
Harold Hart for his invaluable inspiration, guidance, and understanding
during the course of this study.

Appreciation is extended to Michigan State University for a Graduate
Teaching Assistantship from Fall, 1962 through Winter, 1964; to the National
Science Foundation for financial aid during the summer, 1965; to the
National Institutes of Health for financial aid from Spring, l96A through
Summer, 1965.

Appreciation is also extended to colleagues for their helpful
suggestions and discussions.

iii

INTRODUCTION . . . . . . . . . . . . . . . . . . . .
HISTORICAL . . . . . . . . .
RESULTS AND DISCUSSION . . . . . .
A. Reaction of Alkylidenemalononitriles with Bromomalononitrile.
B. Reaction of B-Arylalkylidenemalononitriles with Bromomalono-
nitrile . . . . . . . . . . . . . . . . .
C. Reaction of Arylidenemalononitriles with Bromomalononitrile .
D. 5, 5- Dialkyl- -2- -carbethoxy— —l, l, 2- -tricyanocyc10pr0panes and
5- -aryl- -2- -carbethoxy- -l, L 2- -tricyanocyc10propanes .
E. 5, 5- Dialkyl- and 5- aryl- -2- carboxamido- L l, 2- -tricyanocyclo-
prOpanes. . . . . . . . . . . . . . . . . . .
1. Reaction of Cyclohexylidenecyanoacetamide with Bromo-
malononitrile . . . . .
2. Reaction of Cyclohexylidenemalononitrile with Bromo-
cyanoacetamide. . . . . . . . . . . . .
5. Reaction of Arylidenecyanoacetamides with Bromomalono-
nitrile . . . . . . . . . . .
F. Some Dimers of Alkylidenemalononitriles . . . .
G. Miscellaneous . . . . . . . . . . .
l. Sulfur- -containing Compound obtained from Ethyl Sodio-
cyanoacetate and Carbon Disulfide . .
2. Reaction of 2- Methyl— l-nitroprOpene with Bromomalono—
nitrile . . . . . . . . . . . . .
5. Reaction of IsOprOpylidenemalononitrile with Bromonitro-
methane . . . . . -
EXPERIMENTAL . . . . . . . . . . . .
A. General Procedures.
B. Starting Materials.

TABLE OF CONTENTS

I. Bromomalononitrile.

Alkylidenemalononitriles.

Ethyl Alkylidenecyanoacetates
Arylidenemalononitriles .
l2. B—Arylalkylidenemalononitriles.
15. Miscellaneous

2. Ethyl Bromocyanoacetate

5. Ethyl Bromonitroacetate

A. Ethyl Nitroacetate.

5. Bromocyanoacetamide

6. Dibromocyanoacetamide . . . . . .
7. Dibromoacetonitrile . . . . . . . .,.
8. Bromonitromethane

9.
10.
ll.

iv

Page

TABLE OF CONTENTS Continued

C. Preparation of 5,5-Dialkyl-l,l,2,2-tetracyanocyc10propanes.

l. 5,5-Dimethyl-l,l,2,2-tetracyanocyclOpr0pane . . . .
2. 5-Methyl-5—ethyl-l,l,2,2-tetracyanocyclopr0pane . . .
5- 5-Methyl-5-n-pr0pyl-l,l,2,2-tetracyanocyc10propane.
h. 5-Methyl-5-i-pr0pyl-l,1,2,2-tetracyanocyc10pr0pane.
5. 5-Methyl-5—n-pentyl-l,l,2,2-tetracyanocyc10pr0pane.
6. 5,5-Diethyl:l,l,2,2-tetracyanocyclopropane. .
7. 5-Ethyl-5-n-butyl-l,l,2,2-tetracyanocyclopr0pane.

8. 5,5-Dicyc16pr0pyl-l,l,2,2-tetracyanocyc10propane.

9. 5,5-Tetramethylene-l,l,2,2—tetracyanocyclopr0pane .
lO. 5,5-Pentamethylene-l,l,2,2-tetracyanocyclopropane .
ll. l,l,2,2-Tetracyano-h-methylspiro[2.5]octane .
l2. 5,5-Nonamethylene-l,l,2,2-tetracyanocyclopr0pane.
l5. 5,5-Undecamethylene—l,l,2,2-tetracyanocyc10pr0pane. .
1h. 5,5-Tetradecamethylene-l,l,2,2-tetracyanocyclopropane .

D. Preparation of 5-Alkyl-5-aryl-l,l,2,2-tetracyanocyclopr0panes 7O

1. 5-Methyl-5-phenyl-l,l,2,2-tetracyanocyc10pr0pane

2. 5- -Methyl- -5-m-chlor0phenyl-l, l, 2 ,2-tetracyanocyclo-

prOpane . . . . . . . . . .
5. 5- Methyl- 5 -p-methylphenyl- -l, l, 2, 2- -tetracyanocyclo-

prepane . . . . . . . . .
h. 5- Methyl- 5-p _-methoxyphenyl-l, l, 2 ,2-tetracyanocyclo-

prOpane . . . . . . . . . . . . . . .

5. 5- Methyl- 5- penaphthyl- -l, l, 2, 2- -tetracyanocyc10propane.

6. 5— Ethyl- -5- phenyl- -l, l, 2, 2- -tetracyanocyc10pr0pane . . .

7. Spiro[2,2,5,5-tetracyanocycl0propane-l,l'-tetralin]

8. 5-Methyl-5-d—thienyl-l,1,2,2-tetracyanocyc10propane .

E. Preparation of Some Dimers of Alkylidenemalononitriles. . .

l. Isopropylidenemalononitrile Dimer .

2. 2-Butylidenemalononitrile Dimer . . . . . . . .

5. Cyclopentylidenemalononitrile Dimer .

F. Preparation of 5-Aryl-l,l,2,2-tetracyanocyc10propanes

l. 5-Phenyl-l,l,2,2-tetracyanocyclopropane . . . .

2. 5-g—Chlorophenyl-l,l,2,2—tetracyanocyc10pr0pane . . .

5. 5-m-Chlor0phenyl-l,l,2,2-tetracyanocyclopr0pane . . .

h. 5-p-Chlor0phenyl-l,l,2,2-tetracyanocyc10propane

5. 5-o-Nitr0phenyl-l l 2 2-tetracyanocyclopr0pane.

6. 5-m-Nitrophenyl- -l,l,2,2- tetracyanocyCIOprOpane.

7. 5 -p_—Nitrophenyl—l,1,2,2-tetracyanocyclopropane.

8. 5- gp-Bromophenyl— l,l,2,2- tetracyanocyclopropane.

9. 5 -p- Cyanophenyl- l,l,2,2- tetracyanocyclopropane.
lO. 5- p- Methylphenyl- -l,l,2,2-tetracyanocyc10pr0pane .

 

TABLE OF CONTENTS Continued

Page
ll. 5-p-Methoxyphenyl-l,l,2,2-tetracyanocyclopr0pane . . . 78
I2. 5- -T2', A'-Dichlorophenyl)-L l, 2 ,2-tetracyanocyclo-
propane . . . - - . . 78
l5. 5- (2, 6' -Dichlorophenyl) -l, L 2, 2- tetracyanocyclo-
propane . . - . - 78
1A. 5 (3 ,A'-Methylenedioxy)-l, 1, 2, 2 tetracyanocyclo-
propane . . . . . . . . 78
15- 5- GX-Naphthyl) L l, 2, 2- -tetracyanocyclopr0pane . . . . . 78
16- 5-(p-Naphthyl) -l, L 2, 2- -tetracyanocycl0propane . . . . . 78
I7. 5-(drFuryl)-l,l,2,2-tetracyanocyc10propane. . . . . . . 78
G. Preparation of 5, 5-Dialkyl-2-carbethoxy-l, l ,2-tricyanocyclo—
propanes. . . . 79
l. 5, 5 -Dimethyl— —2- carbethoxy— l, l, 2- tricyanocyclopropane. . 79
2. 5 —Methyl- 5- ethyl- -2- ~carbethoxy- -l, L 2- tricyanocyclo-
prOpane . . . . 8O
5. 5— Methyl- -5- n- propyl- -2- carbethoxy- L L 2- tricyanocyclo-
prOpane . . . . 80
A. 5 -Methyl- 5- i-propyl- -2- carbethoxy- -l, l, 2- tricyanocyclo-
prOpane . . . . . . 8l
5. 5, 5- Diethyl- 2- carbethoxy- L L 2- tricyanocycloprOpane . . 82
6. 5, 5- Tetramethylene- 2- -carbethoxy- L l, 2- tricyanocyclo-
prOpane . . . . . . . 82
7. 5, 5- Pentamethylene— 2- carbethoxy- -l, l, 2- tricyanocyclo-
prOpane . . . . . . . . . . . . . . . 85
H. Preparation of 5—Aryl-2-carbethoxy-l, l, 2-tricyanocyclo-
prOpanes . . . . . . . . 8A
1. 5- Phenyl-—2. carbethoxy-l, 1, 2——tricyanocyclopropane . . . 8A
2. 5- -p- Methoxyphenyl- 2- carbethoxy- -l, L 2- tricyanocyclo-
prOpane . . . . . . . . . . . . . . . . . . . . . 85
I. Preparation of 2-Carboxamido-l,l,2-tricyanocyc10pr0panes . . 86
l. 5, 5-Pentamethylene—2-carboxamido—l, l ,2-tricyanocyclo-
propane . . . . . . 86
2. 5— —Phenyl- 2- carboxamido- l, L 2- -tricyanocyclopr0pane . . . 87
5. 5- p— Chlorophenyl- -2- carboxamido— l, l, 2- tricyanocyclo-
propane . . . . . . . . . . . . . . . . . , . . . . . 87
J. Preparation of 1 ,5—Dicyano-2-imino—5-aza—A—ketobicyclo[5.l.d
hexanes . . . . . 88
l. l ,5 -Dicyano- -2- imino- 5— aza- A- keto- 6 ,6-pentamethylene-
bicyclo[5. l. O]hexane. . . 88
2. l ,5- -Dicyano- -2- -imino— 5- aza- A- keto- 6- phenylb1cyclo[5 l<fl
hexane . . . 9
5. l, S— —Dicyano- -2- imino- -5- aza- A— keto— 6— —p- chlorophenylbi—
cyclo[5. l. O]hexane. . . . . . . . . . . . . . . . . . . 90

vi

TABLE OF CONTENTS Continued

K. Preparation of l,2—Dicyano-l,2-carboximidocyclOpropanes .

Page

1. 5, 5- Pentamethylene-l, 2—dicyano-l, 2—carboximidocyclo-
propane . . . . . . . .
2. 5- Phenyl- ~l, 2- -dicyano- -l, 2- -carboximidocyclopropane.
5. 5-p- Chlorophenyl- -l, 2- dicyano- -l, 2- —carboximidocyclo-
prapane . . . . . . . . . . . . . . . . . . . . . .
L. Miscellaneous . . . . .'. .
1. Preparation of Compound ClngONQOuSB.
2. Oxidation of Compound C12H10N2OAS
5. Attempted Reduction of Compound 012H10N2048.. .
A. Treatment of Compound ClngONQOuSB with Sodium Boro-
hydride . . . . . . .
5. Attempted Desulfurization of Compound él2HlOfi2OAS5‘
6. Reaction of l- Nitro- -2-methylpr0pene with Bromomalono—
nitrile . .
7. Preparation of l, 2 ,5- Tricarbethoxy- -l, 2 ,5 -tricyanocyclo-
propane . . . . . . . . . . . . . .
8. Reaction of Isopropylidenemalononitrile with Bromo—
nitromethane . .
9. Reaction of Cyclohexylidenemalononitrile with Bromocyano-
acetamide . . . . . . . . . . .
lO. Preparation of Compound C13 H BrN2 . . . . . . .
ll. Treatment of Compound 015 H 9BgN2 with Pyridine . . .
l2. Preparation of Compound C12H6Br 2N2. . . . . . . . .
15. Reaction of Bromomalononitrile with Ethanol . . . .
SUMMARY . . . . . .

LITERATURE CITED . . . . . .

vii

TABLE

0\

LIST OF TABLES

3,5—Dialkyl—l,l,2,2—tetracyanocy010propanes .

l,l,2,2-Tetracyanocyclopropanes from Large-membered Cyclic
KEtones . . . . . . . . . . . . .

j-Alkyl-B-aryl-l,l,2,2-tetracyanocyclOpr0panes. . . . . .
5-Aryl-l,l,2,2-tetracyanocyclopropanes.
Some Spectral Data of B-Aryl-l,l,2,2-tetracyanocyc10pr0panes.

5,5-Dialkyl— and 5—Aryl-2-carbethoxy-l,1,2-tricyanocyclo-
prepanes. . . . . . . . . . . . . . . . . . . . . . . . .

viii

Page

. ll

. 15

. i7
26

.29

#2

LIST OF FIGURES

NMR Spectrum of CyclOprOpyl Hydrogen of j-g—Chlorophenyl-
1,1, 2,2-tetracyanocyclopropane. . . . . . . . . . . . . .

NMR Spectrum of Cyclopropyl Hydrogen of j—ofBromophenyl—
l, L 2, 2— —tetracyanocyc10pr0pane. . . . . . . . . . . . .

NMR Spectrum of Cyclopropyl Hydrogen of 3-(2‘,H'-Dichloro-
phenyl)-l,l,2,2-tetracyanocyclOpr0pane. . . . . . . . . . .

NMR Spectrum of CyclOpropyl Hydrogen of Bag-NitrOphenyl-
1,1,2,2-tetracyanocyc10pr0pane. . . . . . . . . . . . . .

NMR Spectrum of CycloprOpyl Hydrogen of 5-Phenyl-l,l,2,2-
tetracyanocyclOprOpane. . . . . . . . . . . . . . .

NMR Spectrum of CyclOprOpyl Hydrogen of 5-p§Chlorophenyl-
l,1,2,2-tetracyanocyclOpr0pane . . . . . . . . . . . .

NMR Spectrum of CyclOprOpyl Hydrogen of 3-pritrOphenyl-
l,l,2,2-tetracyanocyclOpr0pane . . . . . . . . . . . . .

NMR Spectrum of Cyclopropyl Hydrogen of B-p-Cyanophenyl-
l,1,2,2-tetracyanocy010propane . . . . . . . . . . . . .

NMR Spectrum of Cyclopropyl Hydrogen of B—mfNitrOphenyl-
l,l,2,2-tetracyanocyclopropane . . . . . . . . . .

NMR Spectrum of CyclOpropyl Hydrogen of 5— (3' ,h'—Methylene—
dioxyphenyl)— L L 2 2- -tetracyanocyclopropane . .

NMR Spectrum of CyclOpropyl Hydrogen of 3-m— ChlorOphenyl—
l, L 2, 2- -tetracyanocyclopropane. . .

NMR Spectrum of CycloprOpyl Hydrogen of 5—(2',6‘-Dichloro-
phenyl)-l,l,2,2—tetracyanocyc10pr0pane (in DMSO). . . . .

NMR Spectrum of CyclOprOpyl Hydrogen of 5-(2',6'-Dichloro-
phenyl)-l,l,2,2-tetracyanocyclopr0pane (in Acetone) . . .

 

NMR Spectrum of ortho-Hydrogen of 5- (2' ,M' —Dichlor0phenyl)-
l,1,2,2-tetracyanocyclOprOpane (in DMSO) . .

ix

Page

.51

.31

~ 55

- 55

- 55

55

- 55

. 55

- 57

57

- 57

- 57

.38

INTRODUCTION

Wideqvist and Ramberg obtained a crystalline compound, identified
as 3,5-dimethyl-l,l,2,2-tetracyanocyclopropane, by treating acetone with
two moles of bromomalononitrile in the presence of aqueous potassium iodide
(I). This reaction, which now is known as Wideqvist Reaction (2), was
soon extended to several different carbonyl compounds (ketones and
aldehydes) to prepare the appropriately substituted l,l,2,2—tetracyano-
cyclOprOpanes (5).

Recently, the sc0pe of the Wideqvist reaction was extensively studied
and the following mechanism for the formation of the tetracyanocycloprOpanes

was proposed (h).

 

 

 

 

 

 

 

R R OH
\
>C=O + CHBr(CN)2 < > /C<
R‘ R‘ C13r(cN)2 1A
R OH R +0H
\ 2 _
/C< + CHBr(CN)2 <———-=-—> \/c< + CBr(CN)2 1B
R' CBr(CN)2 R' CBr(CN)2
R +0R R CN
\\ // 2
< > >C=C/ + H0 + IBr lC
. \\ \\ 2
R l\<':(CN)2 R' CN
BTU—I
R CN _ R (:(cN)2
\ =C< + CBT<CN>2 *— \C<- 1D
R' CN R‘ C|(CN)2
Br
CN
R\ /9(CN)2 , CH --CN _
/C\) > >< + Br IE
R' C(CN)2 —-——-CN”
l CH3
Br
CN

Bromomalononitrile, with pKa of approximately 5 (5), is probably acidic

enough to protonate the carbonyl oxygen for the initial condensation

(step lA-C). The ease with which alkylidenemalononitrile is formed

(step lA-C) may be a primary factor for determining the success of the

Wideqvist reaction. Once the alkylidenemalononitrile is formed, it may

react further with bromomalononitrile in an irreversible process to yield

the product, provided there are no unfavorable steric or electronic effects.
Step ID is especially significant because, if this reaction proceeds

through formation of an alkylidenemalononitrile, then it should be possible

to prepare l,l,2,2-tetracyanocyclopr0panes from equimolar amounts of

alkylidenemalononitriles and bromomalononitrile without using iodide ion.

If cyclOprOpanes could be formed directly from alkylidenemalononitriles

and bromomalonitrile, this would lend support to the foregoing mechanism.
Another significance of step lD lies in a possible generalization of

this reaction as shown in the following equation, where A, B, Y, and Z are

A

R A R CAB R B
>C=C/ + CHXYZ ———> \c/ ') ——* 2
i \ 1/ \

R B R CYZ Y

)5] Z

electron-withdrawing groups such as -CN, —COOR, -CONH2, -N02, -SOQR, etc.
and X is a leaving group such as -Cl, —Br, —I, —OTs, etc. This reaction
scheme would make it possible to replace one or more of the cyano groups
of l,1,2,2-tetracyanocyclopr0panes by some other electron-withdrawing
group(s), if prOperly substituted reactants were used.

This thesis describes mainly the syntheses of substituted polyfunctional

cyclOprOpanes according to the above scheme. The reactant pairs that have

been used for this study are as follows:

i.

ii.

iii.

iv.

A

A

B:Y:Z:
COOCEHS; B
B=Y=CN3

CN; x = Br
= Y = z = CN;
2 = coochS;

~ Br

- Br

HISTORICAL

Polyfunctional cyclOprOpanes are often prepared by the base
catalyzed condensation of an activated d,p-unsaturated system with
thalO esters, ketones, nitriles, etc. followed by an intramolecular
dehydrohalogenation. The present work involves a similar type of
reaction, except that base is not used as a catalyst. In this
connection, it would be worthwhile to discuss some of the previous
work.

Widman (6,7) obtained 5-acetyl-j,A-phenacylidenecoumarin (III, X
= COCHB, Y = C6H5) by reacting j-acetylcoumarin (I, X = COCHB) with
phenacyl bromide (II, Y = C6H5) in the presence of sodium ethoxide.

The reaction was extended to prepare several analogs by substituting

X _ ' COY
+ BI‘CH2COY > X 5

 

O O
0 O
I II III

X=COCH5, coochS, CN; Y=C6H5, p-CHBOCéHu, mgoch6Hu, dFNaphthyl

various groups for X and Y.

Fraisse and co-workers condensed diethyl bromomalonate with ethyl
acrylate and acrylonitrile using sodium ethoxide and obtained the corres-
ponding cyCIOprOpanes (8). They also carried out further condensations
of various substituted acrylic esters and abhalo esters or drhalo nitriles

(9, 10, ll, 12).

westoo obtained compound VI by reacting l-phenyl-5-methyl-A-
bromo-2—pyrazoline-5-one (V) with l-phenyl-5-methyl—A-isoprOpylidene-2—

pyrazoline-S-one (IV) in the presence of sodium hydroxide (l5).

 

Bromomalononitrile and bromocyanoacetamide were condensed to activated
<X,fi-unsaturated system with sodium hydroxide to form the apprOpriately

substituted cyclOpropanes (lb, 15).

 

 

 

CH5 H5 CH5
+ CHBI‘(CN)2 L—
X CH5 x CN
0 0 CN
x=o, C6H5N
_ CN
CH50(R)—c(CN)cooch5
+ OH (4‘ CH5 OOCBHS 6
CHBr(CN)X R X
CN

R=CH5, 02H5; X=CN
R=CH5 ; X=CONH2

Warner reacted.ogp-unsaturated aldehydes or ketones with diethyl
bromomalonate in the presence of sodium ethoxide and obtained the corres-

ponding cyCIOprOpane derivatives (16).

 

* Use of base was not indicated for the reaction of 5-methyl—h—isopro-
pylidene-2-isoxazoline-5-one (X = O) and bromomalononitrile.

O\

 

’002H5 H H
RCH=CHZ + CHBr(COOCeHS)2 ~>
R 000ch5
000ch5
R=H, CH5; Z=CHO, CHBCO

Mousseron and co-workers made an extensive study on the synthesis
of polyfunctional cyc10pr0panes by base condensation of an ORB-unsatu-
rated system with drhalo esters (I7, 18, 19). Two representative

reactions employed are shown below.

 

R!
H --“Y ~
RCH=CR1Y + CHXZCOOCeHS ____§EEE_,_ R:><::1
———Z
COOCQHS

.R=H, CH3,.COOCQHS; RV=H,'CH3;' X=Cl,“Br5‘ Y:COOCQHS,

CH0, CN; Z=H, cooc2HS

 

OOCpHS
CH2==JcRcooc22H5 *
Base H R
+ >
R'CHXCOOCQHS H R'
000ch5

RzHfi,CH53 R'=H, CH5, C2HS, ETC5H73 X=Cl, Br, OTs
MCCoy studied base condensation of an obhalo ester with an(x43-

unsaturated ester to form l,2-cyc10pr0panedicarboxylic acid diesters in

toluene (20). By saponifying the diesters to the corresponding diacids

followed by analysis of the stereoisomers, the author found that the
less stable cis isomer (cis with respect to two carboxyl groups) was

often dominant. For example,

 

H
CH =CHCOOC H
2 2 5 - H COOH
1. 00 H
+ 25> 10
2. "OH H COOH
CHECICOOCQHS
H

Yield u5% (gig, 26%; trans, 22%)

It was also found that dbsubstituents in either or both of the reacting
esters favored the cis isomer, whereas fi-substituents in the acrylic

ester favored the trans isomers (see Equation ll).

 

 

H
l. NaH R L---COOH
RCH=CHCOOR' + CHBCHCICOOR" _ - :7 H;><::: ll

COOH
R=H 3 R'=CH3 ; R"=C2H5. Yield 70% (SEE, 65%; 33325, lu%)
R=CH53 R'=C2H5; R”=CH5. Yield 65% (gig, ; figggg, 66%)
The NaH catalyzed condensation of methyl methacrylate or methacrylonitrile
with dbhalo esters, nitriles, or ketones showed the same trend of the cis
predominance, that is, the isomer in which the activating groups (ester,
nitrile, or carbonyl) are gi§_to each other was predominant or exclusive (2l).

This cis dominance Was explained as following (22, 25, 24). Consider

intermediate anions such as VII or VIII formed by Michael addition of methyl

 

CH5.\\d://COOCH5 CHBOOC\\\C://CH5
CH5 .CooCH5 CH5::]:E CCCH5
H q H H
Cl Cl ‘

VII VIII

dbchloroprOpionate to methyl methacrylate. In solvents of low dielectric
constant such as benzene and toluene, intermediate VII would dominate
probably because the anion would be better stahflized by interaction

with the carbonyl of the second carboxyl group. In this case, the cis

isomer would be dominant in the product which was the case. For example,

CH5
H COCH5
H OCCH5
CH5
IX

95% of gig-IX was obtained when the reaction was carried out in benzene
or toluene. On the other hand, in a medium of high dielectric constant
and good solvating prOperties, intermediate VIII would be dominant. This
would be so, because the anion VIII would be not only stabilized by means
of external solvation (with solvent) but also would be sterically favored.

In this case, the trans isomer would be dominant and this was exemplified

 

by the presence of 66% and 6A% of transtX in the reaction product, when
the reaction was carried out in dimethylformamide or in a 50 : SO mixture
of benzene and hexamethylphosphoramide, respectively.

In the syntheses of cyCIOprOpanes substituted on all three of the ring
carbons from properly substituted acrylic ester and aRhalo ester, the solvent
effect is usually overshadowed by the steric effect (25, 12). However, a
fairly high stereospecificity was observed in certain cases. It was also
found that an dgp—disubstituted acrylic ester resulted in decreased yields

and that a BJB-disubstituted acrylic ester did not yield any isolable

product (presumably due to the steric hindrance).

Since the present work is largely concerned with polycyanocyclo-
propanes, it would be pertinent to mention the recent preparation and
reactions of dicyanocarbene. Recently, dicyanocarbene was generated
and trapped by benzene (26) and 2,5-dimethyl-2-butene (27) to yield 7,7—

dicyanonorcaradiene (X) and l,l,-dicyano—2,2,5,5-tetramethylcyclopropane

 

(XI).
W5
CN NC :><::j———CH5
. CN NC -——-CH5
CH5

RESULTS AND DISCUSSION

A. Reaction of Alkylidenemalononitriles with Bromomalononitrile

 

For a typical preparation of a 5,5-dialkyl-l,l,2,2-tetracyanocyclo-
prOpane, an alkylidenemalononitrile (XII) was allowed to react with an
equimolar amount or an excess of bromomalononitrile in aqueous ethanol.

The reaction was normally carried out at room temperature and the
concentration of ethanol ranged from 50% to 95% depending on the solubility
of XII. The mechanism of this reaction, as generally outlined in the
introduction involves initial attack of the bromodicyanocarbanion on the
fi-carbon of XII giving another carbanion XIII, which, by intramolecular

displacement of bromide ion, closes the ring.

CN
R CN Q(CN)2 R CN
R>C::C// + CHBr(CN) -—————> R\\ C//) -———+— 12
. \ 2 /C \
R CN R cl:( CN ) 2 R ' CN
(Br CN
XII XIII - XIV

Table 1 lists the tetracyanocycloprOpanes which were prepared by this
procedure. It includes the yields, and the time elapsed before crystalline
product began to separate from the reaction mixture. Although many factors
may influence this time interval, these times and yields may give some
indication of the ease with which the reaction proceeds. Some comments
on the ways in which structural changes in the alkylidenemalononitriles

affect the reaction, as judged by these criteria, may be worthwhile.

lO

ll

Table l. 5,5-Dialkyl-l,1,2,2-tetracyanocyclOpr0panes

 

 

 

N
R OH
R. CN
CN
R R' Reaction timea Yield, %

CH3 CH3 ’ 2-5 mins. 86

CH3 CQHS 5-6 mins. 91
CH3 E‘C5H7 5-6 mins. 97.5
CH3 i-CBH7 60 mins. 97.u

CH3 t—Cuflg ' b 0
CH3 n-CSHll 60 mins. 97.5
C2H5 CQH5 2—3 hrs. 88.5
C2H5 E-Cqu 21L hrs . 91L. 5

i-Cqu .i-CAH9 b o

E-‘CSHll E-CSHll b O

(CH2)2CH (CH2)2CH 10 mins. 62
(CH2)u 50 mins. 52.6

(CH5)CH(CH2)QCH(CH5) b o
(CH2)5 2-5 mins. 97.5
(CH2)4CH(CH5) 16 hrs. 39.3
(CH2)9 15 mins. 34.8

(CH2)11 60 mins. 9H

(CH2)lu 6-7 days 36C

 

a. Time elasped before crystalline material began to separate in the
reaction mixture.

b. .Total reaction time not less than a month.

c. The yield based on the amount of the ketone.

12

When R of the alkylidenemalononitrile (XII) is methyl, the time
necessary to produce product increases rapidly as the R' is changed
from methyl to isopropyl, or tabutyl. Despite the fact that the
reaction time, when R' = iSOpropyl, is long, the yield is very high.

This was accomplished by adding bromomalononitrile periodically to

CH3\\ //CN
A/C==C\\
R' CN
R: = CH3 _i_—C§H7 t-Cqu C2HS n-CBH7 n-CSHll
t(min.) = 2-5 60 (N.R.) 5-6 5-6 60

the reaction mixture. As will be seen later, bromomalononitrile reacts
with the solvent (see page 22). It is therefore necessary to use an
additional amount of it especially for a reaction that requires a long
reaction time.

When neither of the alkyl groups in the alkylidenemalononitrile
(XII) is methyl, the time required for reaction is usually much longer,

or the reaction may not occur.

R\>C=:C<:CN
R' CN
R = CQH5 C2H5 E’CBHY E‘CHH7
R' = CQHS n-CuH9 i’chY i-CuH7
t(hrs.)= 2-3 2h (N.R.) (N.R.)
Curiously, when R = R' = cyclOprOpyl, the reaction is unusually rapid.

This is particularly strange, since this reaction fails entirely when
carried out as a Wideqvist reaction. That is, 5,5-dicycl0pr0pyl-l,1,2,2—

tetracyanocyclopropane could not be obtained directly from dicyclopropyl

l5

ketone, bromomalononitrile, and iodide ion (28). The reaction does not
take place when R = R' = i50pr0pyl. . This may be explained in terms
of steric and electronic effects. When R = R’ = is0pr0pyl, two isopropyl
groups certainly will exert a high steric effect against the approaching
bromodicyanocarbanion. However, the steric effect of the cyclOprOpyl is
likely to be less than that of the isopropyl, since two methyls are tied
back to each other. Besides, the electron-withdrawing inductive effect
of the cycloprOpyl group is greater than that of the is0pr0pyl group (29).
This tends to make the fi-carbon of dicyclOpropylmethylenemalononitrile
more positive relative to that of diisOpropylmethylenemalononitrile and
thus a better center for the attack of bromodicyanocarbanion. .
5-Heptanone failed to produce the cyclOprOpane product by reacting '
with bromomalononitrile in the presence of the iodide ion (28); But
when 5-heptanone is converted into 5-heptylidenemalononitrile (XII, R =

02H5, R' = EfChH9)2 then the reaction with bromomalonitrile gave the

product, 5-ethyl-5-nfbutyl-l,l,2,2-tetracyanoacycIOprOpane (XIV, R =

CgHg, R' E'CHH9)' The failure of heptanone and dicyclOprOpyl ketone to
produce the product may be due to the failure of undergoing the initial
condensation to form the corresponding alkylidenemalononitriles under the
reaction conditions.

In the cyclic alkylidenemalononitriles, their ring size affects the

reaction rate and yield. For example, cyclohexylidenemalononitrile gives

 

 

CN . , CN , CH5 ON
ON * ‘ ‘ 5 CN CN
CH3
T (min-) = 2-5 50 30 days
Yield (%) = 97.5 52.6 (N.R.)

11+.

product within a few minutes, whereas cycl0pentylidenemalononitrile
requires 50 minutes for the formation of the product. This may be
explained in terms of internal ring strain and non—bonding interactions
(50). CyclOprOpanization of the exocyclic double bond involves a change
from sp2 to something approaching Sp5 hybridization. In case of the
cyclohexane ring, this conversion is generally favored because the six—
membered ring is least strained when all six carbons are saturated.
However, the reverse is true with the cyCIOpentane ring, because of an
increased non—bonding interaction in the saturated five-membered ring.
Therefore, conversion of cyclOpentylidenemalononitrile into the corres-
ponding cyclOprOpane would not be too highly favored. This view seems
to gain a support from the observation that 2,5edimethylcyclOpentylidene-
malononitrile did not produce the product at all apparently because of
even greater interaction among the substituents in the five-membered
ring of the product molecule.

Some large-membered cyclic ketones which had previously failed in
the Wideqvist reaction (28) have now been found to produce l,l,2,2-
tetracyanocyCIOpropanes in good yield if they are first converted into
the corresponding cycloalkylidenemalononitriles. A few l,l,2,2-tetracyano-

cyclopropanes thus prepared are listed in Table 2. In a large-membered ring

compOund -such' as cyclopentadecylidenemalononitrile (XV), the methylene
chain should be as free as if they were in an acyclic compound like

6- hendecylidenemalononitrile This point makes it difficult to

explain why XVI fails to form the product, whereas XV yields the product.

15

Table 2. l,l,2,2-Tetracyanocyclopropanes from Large-membered Cyclic Ketones

 

 

CN
ON
(CH2)n CN
CN
n Reaction timea m.p., °C Yield, % NMRb,'r
9 15 mins. 2lu—2l6 3u.8 8.u3 (s, 10.00 H)d
8.03 (m, 7.9M H)
ll 60 mins. 197—200 94 8.53 (s, lu.00 H)e
8.09 (m, 8.02 H)
In 6—7 days lll-ll2 36C 8.64 (s, 20.00 H)(1

8.30 (m, 7.85 H)

 

a. Time elapsed before the crystalline product appeared in the reaction
mixture.

b. Inside the parentheses are described multiplicity of the peak and the

number of hydrogens involved in that particular peak.

Yield based on the amount of the ketone.

NMR spectra run in DMSO-d6 solutions.

e. NMR spectrum run in acetone—d6.

{3.0

It is interesting to note that, in the NMR Spectra of the three tetra-
cyanocyCIOprOpanes prepared from large-membered cyclic ketones (Table 2),
eight hydrogens appear in a separate multiplet between’T8.05 and‘T8.50.
Incidentally, the eight hydrogens of 5,5-tetramethylene-l,l,2,2—tetracyano—
cyCIOpropane appeared in a peak at‘T7.9l. This may be an indication that
the eight hydrogens of the above three compounds come from those methylenes
o_(, [5, og’, p’ to the spiro carbon. However, it is possible that, due to
the flexibility of the large ring, some methylenes other than those men-
tioned may be brought to the proximity of the cyano groups and are

responsible for these multiplets of hydrogens. Perhaps, an experiment of

deuterium substitution would give a more unequivocal answer to this problem.

l6

B. Reaction of’fi-Arylalkylidenemalononitriles with Bromomalononitrile

 

Table 5 lists a group of 5—alkyl-5-aryl-l,l,2,2-tetracyanocyclo-
pr0panes prepared from the corresponding fl-arylalkylidenemalononitriles.
fB-Phenyl and fi-mfchlorOphenylethylidenemalononitrile (XVII and XVIII,
reSpectively) reacted much faster than fikpfmethoxyphenylethylidene-
malononitrile (XIX). This may be explained by inductive and resonance

effects. The phenyl and the mechlorOphenyl groups in XVII and XVIII,

CH3
C=C/CN C11 :1 C=C/CN C=C/CN
CH/ \CN CH/ \CN CH3/ \CN
5 5
XVII XVIII XIX
t (hrs.) = 1.5 2.0 20
Yield (%) = 86.6 68.1 51.2

respectively, will undoubtedly exert an electron—withdrawing inductive
effect. This will maintain the fi-carbon of XVII and XVIII relatively
positive, thus facilitating the attack of bromodicyanocarbanion. However,
in the case of XIX, resonance through the p-methoxyphenyl group is likely
to keep the p-carbon of XIX electron rich, thus retarding attack by the
bromodicyanocarbanion.

With 2,5-benzocyclohexylidenemalononitrile (XX), the reaction seemed
to take either of two courses, depending on the solvent and temperature.
When 80% aqueous ethanol was used, at room temperature, the normal cyclo-

pr0panization reaction occurred.

l7

8 C CH. <4!!s> CN
0 N + CHBr(CN)2 A 0% 2H5 > 1
._ Room Temp. . ‘ CN 5
ON
ON ON

XX XXI

Table 5. 5-Alkyl-5-aryl~l,l,2,2-tetraCyanocycIOprOpanes

 

 

 

N
Ar CN
R , CH
CN
R Ar Reaction timea Yield, %
CH5 C6H5 90 mins. 86.6
CH5 melC6Hu 2 hrs. 68.1
CH5 prHBC6Hu 15 hrs. 81.2
CH3 p§CH50C6Hu 20 hrs. 51.2
CH5 s-Naphthyl 60 mins. 5L8
C2HS C6H5 2-3 daysb 17.8
CH5 dehienyl 50 mins. 22.1
2,5-Benzocyc10pentylidene 6.7 hrs. 0
2,5-Benzocyclohexylidene 6'7 hrs. 5A.2

 

' ' ' v r w

a. Time elapsed before the crystalline product separated in the reaction
mixture.

b. This value is uncertain (see Experimental).

However, when the reaction was carried out at reflux or in 95% ethanol
at room temperature, a pinkish bromine-containing compound, m.p. 155-158o;

was obtained. Elemental analysis gave an empirical formula C15H9BrN2.

18

The IR spectrum (in nujol) showed three absorption bands in the double
bond region at 1605 cm-l, 1567 cm-1, and 15H5 cm-l, which are very

similar in shape and relative intensity to that of the starting material
XX (1599 cm-1, 156A cm-l, and 1551 cm'l). Also, the CEEN absorption

band in the IR spectrum appeared at a much lower frequency (22A5 cm_l)
than that of the tetracyanocycloprOpane XXI indicating that the carbon-
nitrogen triple bond is conjugated with an unsaturated system. The

UV Spectrum (in ethanol) showed two peaks, at 235 mu. (e 6,830) and 322.5
nnt(€l6,7h0), which again is similar to that of XX (229 my,£;7,161;

233 my, €6,868; 311 mu, £16,860). The absorption band at 233 my. strongly
suggests the presence of >C=C(CN)2 in the molecule, as the UV absorption
of a number of alkylidenemalononitriles falls between 252 mp,(:ll,920

(for isOpropylidenemalononitrile) and 258 mp, £12,918 (for 6-hendecylidene-
malononitrile). Finally, the NMR spectrum (in DMSO-d6) showed two
multiplets at 77.55 and 769+, a triplet at 111.36 (‘J = 3.6 c.p.s.) and the
aromatic multiplets at'r2.h7 andrrl.75 in an area ratio of 1.97 : 2.01 °
0.92 : u.00.

Possible structures for this compound are shown below (XXII-XXV).
NC

06 "" or 01,

XXII XXIII XXIV XXV

CH

    

w

19

The structure XXV is eliminated because the aromatic multiplets in the
NMR integrated as four hydrogens and also because the compound, on
heating with pyridine for two minutes, readily loses the bromine, a
phenomenon not usually expected from an aryl bromide. The debrominated
compound, m.p. 158-159° dec.,has not yet been completely identified.

The structure XXIV agrees well with the analytical, IR, and UV data.
However, it cannot comply with the NMR triplet at’TH.56 and is eliminated
on this ground. The structure of the brominated compound must, therefore,
be either XXII or XXIII. Both structures retain the:p=C(CN)2 moiety

and, in both compounds, the hydrogen attached to the brominated carbon
should appear as a triplet in the NMR spectrum. Although no conclusive
evidence is available, structure XXII appears to be more likely for the
following reason. The NMR chemical shift (in CD015) ofthe methylene
hydrogens of fi-phenylprOpylidenemalononitrile (XXVI) occurs at’T7.05

(51) and that of the<X~methylene hydrogens of tetralin (XXVII) at‘r7.2u
(52)- Imagine two brominated compounds XXVIII and XXIX. If Shoolery's
effective shielding constant for the bromine (2.55 p.p.m.) is substracted
from the methylene shifts of XXVI and XXVII, then the eXpected chemical
shifts for the methine hydrogen of XXVIII and XXIX would be’Th.72 and
7h.91, respectively. Now, the methine shift of XXIII probably would not
be much different from that of XXIX. However, the methine shift of XXII
would be lower than that of XXVIII (<1‘A.72), because, in a rigid molecule
like XXII, the methine hydrogen is likely to be held out of benzene plane.
Thus the methine hydrogen of XXII is more likely to appear at as low field‘

as’rh.56 (observed).

20

NC CN
at 00
T 7.05
E, H. 7’7.2h
XXVI XXVII
NC CN
o.
71+.72
(EXpected) Br H Tu.91
II XXIX (Expected)
NC ON NC CN
I Br I
O. a O.
T" .
((151 #:2d) Br H T1191
C e .
XXII Xpe XXIII (Expected)

It is speculated that the bromination of XX may involve a free

radical mechanism as shown in the Equation 1A.

BrCH(CN)2 ——>- Br- + 'CH(CN)2 14A
xx + -CH(CN)2 ——> 100 + CH2(CN)2 _ l-’+B
XX' + BrCH( CN )2 ——~- XX-Br + 'CH( CN)2 INC

A similar type of bromination was observed in the reaction of 2,5-
benzocyclopentylidenemalononitrile (XXX) and bromomalononitrile. The desired

product ,(XXXI)'was not obtained. Instead, there were obtained off-white

21

o ’ 0c
r .4

N
CN
N

CN
N

XXX XXXI
crystals, m.p. 208-2llfl'which, on standing, became magenta. Analysis
gave an empirical formula ClgHSBr2N2, which is an impossible formula,
but could very well be considered to be 012H6Br2N2. The IR spectrum
(in nujol) showed a CEEN absorption band at as low frequency as XXX
(22h0 cm_l), indicating a conjugated cyano‘group.‘ Furthermore, two IR
absorption bands at the double bond region, 1599 cm"1 and 1568 cm'l,
were similar in shape and 111 relative intensity to those of XXX (1601
cm-:L and 1568 cm-l).

The UV spectrum (in ethanol) showed a peak at 256 my.(€7,l#0) and
two ill-defined peaks at 337 mu (€15,859) and 3A6 mp.(615,731). These
data are favorably compared to 25A my, (€8,170),506 my. (£17,L+80), and
333 mu(e19,722) of the starting material'm*. Again, the peak at 236
my. (e 7,1l+0) is indicative of the presence of )C=C(CN)2 moiety in the
molecule. Unfortunately, it was not possible to find a suitable solvent
to determine the NMR spectrum. Possible structures for this compound are

shown below.

NC CN NC CN

 

Br

1 cis and trans

 

* The UV speCtrum.of 2,5-benzocyc10penty1idenemalononitrile (XXX) showed
an additional two peaks at 206 m}L(€7,229) and 211 m+L((6,258).

22

The following substituted l,l—dicyanoethylenes did not give any

isolable products on reaction with bromomalononitrile.

R\__/CN
{/C-—C\\
R' CN
XXXII
XXXIIa R = R' = C6H5
XXXIIb R = H; R' = CQHSO
XXXIIe R = CH}; R' = CQHSO
XXXIId R, R‘ = Ethylenedioxy
XXXIIe R = R' = ON

The failure of diphenylunawaylenemalononitrile (XXXIIa) might be caused
by the steric effect of the two phenyl groups. The failure of fi-alkoxyl
derivatives (XXXIIb-d) might be ascribed to the resonance effect of
alkoxyl group, which would keep the g-carbon rich in electrons. It is not
possible to present a plausible reason why tetracyanoethylene (XXXIIe)
failed to produce the product.

When an excess of bromomalononitrile was allowed to react with an
alkylidenemalononitrile, especially for a prolonged time, there was
usually obtained, in addition to the desired tetracyanocyclOprOpane, a
small amount of crystals, which did not melt sharply but sublimed at 270-
500°. The substance contained bromine and nitrogen. The IR Spectrum (both
in nujol and KBr pellet) did not show any significant absorption bands.

The NMR spectrum (in D20) showed a singlet at‘T5.15. This substance was
identified as ammonium bromide.

When excess bromomalononitrile was allowed to react with 5,5-dimethyl-
2-butylidenemalononitrile in aqueous ethanol (about 85%) for a prolonged

time (a week to a month), no cyclOprOpane was produced. Instead, there was

. 23

obtained a brown precipitate which, on treating with Norit A, became
white crystals. This compound was also obtained by similarly treating
bromomalononitrile with other alkylidenemalononitriles which failed to
yield tetracyanocyclOprOpanes; for example, 2,4-dimethyl-5-penty1idene—
malononitrile, 6-hendecylidenemalononitrile, tetracyanoethylene, etc.

The fact that different alkylidenemalononitriles produced the same com-
pound suggested that the formation of this compound did not involve the
alkylidenemalononitrile. Indeed, when an aqueous ethanol solution of
bromomalononitrile was refluxed for several hours, this compound, as well
as ammonium bromide, was obtained.' The analysis of the sample, m.p. 255-
257° gave an empirical formula C6H7N50. The IR Spectrum (in nujol) showed
the absorption bands at 3355 cm'l, 3230 cm-1 (both for NH), 2248 cm'l,

1 -l -1
, 1550 cm , 1503 cm (C=C and/or NH),

2205 cm'1 (both for CN), 1658 cm'
10A5 cm-l'(C-0-C), and many other minor bands. The UV Spectrum (in
ethanol) gave a single absorption band at 255 mu (618,656). Finally, the
NMR spectrum (in DMSO-ds) showed a triplet and a quartet (J = 7.0 c.p.s.
for both peaks) at‘T8.7l and‘r5.72, respectively, (strongly indicative of
an ethoxyl group) and a Singlet at Tl.H5 in an area ratio of 5.00 : 1.98
1.98.

The analytical and spectral data suggest that the possible structures
are XXXIII and XXXIIIa.

NC// \\002H5

NC-C(NH2)==C(002H5)-CN
XXXIII ‘ XXXIIIa
However, XXXIIIa was eliminated because the compound appears to be identical

with XXXIII previously reported in the literature (55), m.p. 225-226° (IR:

S'OSP’) 5'19“: 6'05”) 6J6“; W3 25“" mp)'

2A

A possible mechanism for the formation of XXXIII is pr0posed in

 

Scheme 1.
Scheme 1
Br CHBrCN Br CHBrCN
' ,~‘| ' H+ é
NC—(IV («31 If 7 NC— —n m
CN N;> NH V

I
CN
C2H50H
H?/

 

 

Br BrON Br CHBrCN
i 02H50H l
NC--l I >NC-—C-———-C-—002H5
XXXV CN NH2 CN NH2 XXXVI

B : 'N

Br CHBrCN
NC—élc —-0C2H ———-> \0 =0 / 5

l | 5 NC’/ \\NH

CN NH2 2

XXXVI XXXIII

Bromodicyanocarbanion attacks on the cyano carbon of bromomalono-
nitrile to give the imino compound XXXIV. XXXIV tautomerized to XXXV
which adds ethanol across the double bond to give XXXVI. 0r XXXVI can
be formed by direct addition of ethanol to the imino double bond of
XXXIV. The exact nature of the products from the final elimination step
is not known. Possible candidates for the nucleOphile are bromide ion,
or bromodicyanocarbanion. Bromoacetonitrile or its hydrolysis product
may also be formed. A few possibleprecedents for the last step of the
elimination have been encountered and are described on pages N7, 50,

and 60, respectively.

25

C. Reaction of Arylidenemalonpnitriles with Bromomalononitrile*

 

 

 

CN
Ar CN Ar L— CN
\\ __ // , 1 15
/C—C\ + CHRr(CN)2 r
H CN H CH
XXXVII; ON
XXXVIII

Table A lists the 5-ary1-l,1,2,2-tetracyanocyCIOpropanes (XXXVIII)
prepared according to Equation 15, together with some of the pertinent
data. In most cases, an arylidenemalononitrile (XXXVII) seems to react
with bromomalonitrile to form the product much faster than an alkylidene-
malononitrile having an equal number of carbons. This may be due to the
fact that the electron—withdrawing nature of the phenyl makes the fircarbon
of XXXVII positive. It may also be ascribed to the steric effect, i.e.,
the arylidenemalononitriles have only one substituent at the fi-carbon,
whereas the alkylidenemalononitriles described earlier bear two fi—sub-
stituents.

When Ar of XXXVII was a phenyl or a phenyl with an electron-withdrawing
substituent, the time required for the formation of product was just a few
minutes. However, there were a few exceptions. For example, mechloro-
benzylidenemalononitrile (XXXIX) and genitrobenzylidenemalononitrile (XL)
required 50 minutes and 2,6-dichlorobenzylidenemalononitrile (XLI) an hour
The longer time required by XL and XLI may be explained by the steric
effect exerted by a nitro group or by two chlorine atoms in the ortho

positions.

 

* Arbitrafifly included with arylidenemalononitriles are those benzylidene-
malononitriles, with or without substituents, that retain the fi-hydrogen.

 

 

$38 3.3 Sada m I
I mm.:m os.m o:.m~ m.om ommIsmm .nsae om Ammo moIa
86an iodv “£81 I
I em.mm sm.m mw.wo s.>m sow .nsas eon < smooozIa
Aom.omv flow.wav Aos.av Amm.mmv I
mm.om mm.wa mm.a se.mm m.sw mmmImsm .nsas sot < amooamIo
1863 39: Eng I
I Om.om ma.m mm.mm so mmmImmm .nsae sou < smoozmoIa
I I I I s.ms mamIonm .naaa sou < a smoozmorm
H.. m mm.H n.m I
I Amm.wmv AH>.HV Ams.mmv m.mm mHmIHHm .nsae on smoozono
I I I I m.mm ohmesm .nsaa roe < o smooaoIm
6
2 Roll: Sada 86.3 8648 I
om.sa oa.mm .oa.m mm.ao mm mmHIme .msae om smooaoIe
Amos: $58 SIS 18.81 I
Hm.sa om.mm Hm.m sw.ao mm memIosm .nsae sea < smooHoIo
I I I I S.Hm omm-smm .nsaa Sea a o memo
x . z m o a .oaoaw o. .A.ooev.m.2 scene soaeosom .a<
_ e manageda
26
20
zo
zo

 

mmammopaoaozoonwhomnpopIm«maHaHIHzH<Im

.s manna

27

n o omfiv
mampmunho 09H.

0

mHmImom .m.e «CHSPRHE amoeoowIHoqmflpo an 809% UCNHHHmpmhhooH but onopHHm 0903
.CHSpmHmmaow EOOH pm mowam Mano m QH .mafl m Mom pmox mm: CHSpRHE one can Hoamnpo mSOCSwm

.Aem.asv .m.m.o

pogo...“ one page

Rem %o .HE OH ca ©o>HOmmH© who: Amoaoaa m.w a.m av CHHHPHGOQOHNSCSOHQ can Amoaoas m.m a.m av HHSS%H§@

"Rwanda mQHSDHHow 03p qfl dohmmohm on Oman BHdoo dadomfioo mflnp Swap dadow coop mpg PH
.w .Hoqmgpo ana womflaampmhhooa was 095pr8 coapomoh one Op Stem: wo .HE OOH maflppm
.mhg :m Mom manpmhogaop Hoop Hm 990M was Hoqmflpo mo .HE 00m SH ©0>Hommflp

.aIs Amv .ase .paq ,
so dmsuom mm: pmnp onpHmHoon
0903 Amoaoaa Om.> a.w mo.av mafiaaflaoaoamsoaohn paw Ammaoaa mo.m a.m av 0HHapflaocoamamsopflazaoaomflm

.GOHpHmogaooop pdonpfla dopamz .£

909mm voaflwpno mm: podpoam

.omo . oomIomH

.H

.Hoqmgpo Soaw coapwmflaampmhhoon an posoaaom pnm>aom one wo Hm>oaoh

.mhwd m Hog Chapmaomaop aooa Pm pmox was mOHPSHOm ogB

.oaapxas ameIHossseo

an mo .HE ow GH ©o>H0mmH© who: Amoaoaa m a.m mm.ov CHHHPHcCGOHHSCSOHQ pamflmoaoaa w:.: «.mav maflppflnocoama
.ooe .osmImsm .a.s Asv .Hos .eaq

quopflahmnopoaoanoflpI:«mu .m
.soapanoaaoooo psonpas ocean: Ie

.ooo .Hmm .a.s Amv .Hss .eaq .o

.m .ooo .HsmIosm .a.s Any .Hs>.waq
.Qowoawn mopocop N was mosaw> popwHSCHmo

.0

 

 

 

 

oossapsoo .s sabre

one out momCSpGCHmm SH mohzmflm .p .oadpxfla coapomoh map QH opmammom op qu09 powwowm ohoyop dommdao mafia .w
I I I I m.ws momImom .nsas 0H s HaasmIm
I I I I mm osmIomm .nsas Sou < assesaszIm
Amm.omv Aao.nv Ada.oev
I mm.om moIn Hm.os mm mmmImsm .saa a campus assesaszIH
Aem.amv AAMINV Ama.sov m m m
I om.Hm cm.m wH.so mo mmmImmm .has am a who 0 moIs.m
Aos.smv Adm.aav AosIav Amm.smv m
no.4m sm.mH 6:.H mm.:m o.am ommIemm .as H moomHoIo.m
Aos.sdv Amm.dav Aos.av Amm.smv m o m
oo.:m ms.mH Om.a 6:.sm m.mm a wmmImmm name m m m o HoIs.m
ASm.mmv Amm.nv Ae~.sov m I
I 0:.mm om.m mo.eo ma oamImom .mas sot < emcee moIa
x z . .m o I. $.6Hoas oo.A.ooev.m.z_ omega soaeosom a<
p mflmhadc<

28

If, however, Ar of XXXVII is a phenyl with an electron-releasing
substituent, then product formation requires a longer time presumably
because of the resonance effect caused by the substituent. This obser-
vation seems to suggest that an electron-withdrawing substituent may not
always facilitate the reaction, but an electron-releasing substituent
does Slow the reaction down.

The NMR spectra of arylidenemalononitriles Showed the vinyl hydrogen
between/71.15 (penitrobenzylidenemalononitrile) and7‘l.90 (pfmethoxy-
benzylidentuualononitrile). The vinyl hydrogen of l-naphthalmalononitrile
appeared at unusually low field (7'0.60). The vinyl hydrogen of an N-
substituted 1-amino-2,2-dicyanoethylene (XLII), which bears some relation-
ship to an arylidenemalononitrile, was reported to range between'T2.20
(R = CH5, R' = H) andfr2.92 (R, R' = pentamethylene) (34).

RR'N\\ —_ //CN
/C—C\
H CN
XLII

When an arylidenemalononitrile (XXXVII) is converted to a 5-aryl-1,
1,2,2-tetracyanocyc10propane (XXXIII), the vinyl hydrogen becomes the
cyc10pr0pyl hydrogen, appearing between’Th.51 (XXXVIII, Ar = 9702NC6H4)

and7'5.27 (XXXVIII, Ar = p-CH 006Hu). A remarkably low chemical shift

5
for the cyclopropyl hydrogen of XXXVIII has been ascribed to the fact,
thatIin XXXVIII, the cyano groups are oriented in such a position that
the anisotropy of carbon-nitrogen triple bond deshields the cycloprOpyl
hydrogen (A).

In the NMR spectra of 5-aryl-l,l,2,2-tetracyanocyc10propanes XXXVIII,

the cyclopropyl hydrogen was shown to couple with the ortho hydrogens of

29

 

 

Table 5, Some NMR Spectral Data of 5—Aryl-1,l,2,2-tetracyanocyclqpr0paneSa
k CN
Ar' CN
H CN
CN

Ar. H, Cyclopropylb H, Aromatic
06H5 5.l5(t, 0.8) 2.56(m) 2.26(m)
o ClC6Hu 5.03(d, 1.0) 2.30(m) 1.92(m)
m ClC6HLL 5.07(t, 1.0)C 2.50(m) 2.31(m) 2.07(m)
p C106Hu 5.06(t, 0.9) 2.46(d) 2.l4(d) (A232)
o-02NC6Hu 4.5l(d, 0.9)d 2.02(m) 1.65(m)
m-02NC6Hu 4.85(q, 0.9) 2.22(s) 2.0 7(s) 1.94(s)

l 56(m) 1.0 4(m)
p-02NC6Hu 4.94(t, 0.8) 1.93(d) 1.67(d) (A2B2)
p-NCCgH1+ 4.77(t, 0.6) 1.95(d) 1.73(d) (A2B2)
o-13rC6HIL 5.l4(d, 0.9) 2 58(m) 2.2 7(m)
p-CHBCgHI,f 5.16 2.64(d) 2.27(d) (A2B2)
p-CHBCCng 5.27 2.95(d) 2.27(d) (A2B2)
2,4-C12C6H5 5.02(d, 1.0) 2.13(m)
2,6-01206H5 4.94 2.26(8)
5,h-CH20206Hu 5.23(t, 0.95) 3.17(s) 3.02(s) 2.75(m)
l—Naphthyl 4.56 2.11(m) 1.85(m)
2-Naphthyl 4.84 2.34(m) 2.00(m) 1.50(s)
2-Fury1 4.60 3.41(q) 5.07(m) 2.17(m)

 

fits

b.

V

Unless otherwisestated, the SpectraIwere taken in an acetone solution

and the chemical Shifts were expressed inJTunits. In the parentheses,

small letter designate the multiplicity of the peak and the figure the
Central peak of this triplet was
This peak appears to

coupling constant (J) in c.p.s
further split into a triplet, J =

d.

consist 0f 2 doublets each with J =
run in dimethyl sulfoxide solution.
¢4.52 for H, cyclopropyl andxrl. 98(d ) and.71..81( (d) in an

for H, aromatic. f.

CIA c.p.s. d.

019 c.p.s.

The NMR spectrum was
The values in acetone solution are

The NMR spectrum was run in acetone

Anglmmtern

6 solution.

50

the aromatic ring with a coupling constant between 0.6 c.p.s. (XXXVIII,
Ar = prCC6Hu) and 1.0 c.p.s. (XXXVIII, Ar = g-ClC6Hu). In Table 5 are
listed the data of chemical shift, multiplicity, and coupling constant of
the cyclOprOpyl hydrogens.

5-Aryl-l,l,2,2-tetracyanocyclopropanes in which one of two ortho
hydrogens was substituted gave a doublets for the cyclopropyl hydrogens.
This would be expected if the cycloprOpyl hydrogen is coupled to one

ortho hydrogen.

 

 

X = C1, Y = H (d, J e.1.0 c.p.s.) ‘ Fig. l
X = Br, Y = H (d, J = 0.9 c.p.s.) Fig. 2
X = C1, Y = Cl (d, J = 1.0 c.p.s.) Fig. 3
X = N02, Y = H (2. doublets

J = 1.0 c.p.s. for both) Fig. A

However, 5-o-nitrophenyl-l,1,2,2-tetracyanocyclopropane Showed two doublets
for the cycloprOpyl hydrogen. Three different conformations are shown

below, i.e. XLIIIa, XLIIIb, and XLIIIc.

   

XLIIIa

51

.CQHQOMQ

IoaohoocwhomhpopIm“maHaHIAHSQoLQOHOHQCHQI_:a_mNIm mo QCWOthm HSQOHQOHCSU ao afihpoomm mzz
.mammoamoaohoosmhosapmpIm«NaHaHIHhcogmonHmwmIm wo memosphm Ahmoamoaoho yo afihpoogm mzz

.mammoamoaohoocmhomhpopIm«NaH«HIHSQCQQOHOHQOIOIM wo Gowosdhm Hamoamoaozo wo Haywoomm mzz

.m .mam
.m .mam
.H .maa

 

 

mo.mu.

 

 

 

 

omonHoo O .Hflhu

m .mam

_

 

 

emogoo mooub

m .mam

_
SHIWIP

 

 

.n.a.o o.HIe

 

_
mo.m.r

 

 

 

H .mam

 

 

 

52

An equilibrium between any two conformers could result in two doublets.

However, XLIIIC seems very unlikely because of steric reason. XLIIIb

would be sterically most favored (55). XLIIIa, although sterically less

favored than XLIIIb, may be electronically more favored; that is, in

XLIIIa, the w-electrons of the phenyl ring could overlap with those of

the cyclopropane ring and possibly with the two cyano groups. Therefore,

5-2fnitr0phenyl-l,1,2,2-tetracyanocyclopr0pane may be in equilibrium between

the two conformers, XLIIIa and XIJIIb.C This may possibly give rise to

two doublets for the cycloprOpyl hydrogen. Or the cyclOprOpyl hydrogen

may be split by the ortho hydrogen and the para hydrogen. No unambiguous

conclusion could be drawn from the data obtained.
5—Phenyl-1,1,2,2-tetracyanocyclopropane and 5-aryl-l,1,2,2-tetracyano-

cyclopropanes having a substituent at the para position gave a triplet for

the cycloprOpyl hydrogens. This is expected because, with free rotation

along the phenyl-cyclOpropyl bond, there would be two equivalent ortho

hydrogens. 5-p-Methylphenyl—l,1,2,2-tetracyanocyclopropane and 5-pf

methoxyphenyl-1,l,2,2-tetracyanocyclopr0pane did not give a resolved triplet,

0

but the peaks were shaped like a.triplet.

 

X CN
;——CN
H I—F-CN
CN
X = H (t, J = 0.8 c.p.s.) Fig. 5
X = Cl (t, J = 0.9 c.p.s.) Fig. 6
X = N02(t,, J~= 0.8 c.p.s.)i Fig. 7
X = CN (t, J = 0.6 c.p.s.) Fig. 8

 

55

.0QHQOHmoaohooqmzomeopIm«maHaHIHSQCQQOHOHQOImIm mo nowoapzm HSQOHQOHCSO Lo afihpoomm mzz .m .mflm
.CQSQOHQOHCSCOQHSCSHPCPIN«maHaHIakaoflmIm wo memosdhm HSQOHQOHCSQ mo afihpoomm mzz .m .mHm
.ossQOHQoaohooqwhomHPmpIm«NaH«HIHSGC£QOHPHZLMIM wo ammonpzm HSQOHQOHCSO wo adhpommm mzz .: .me
a

 

_ _
oqms mfime Hfsb

   

CMonHoO moon.”

w II n
omomoo moonhu .l—<I_

 

 

 

.mdo 04"??le
omogoo OOHHHJ

 

 

 

 

 

 

 

o .maa m .maa s .mam

 

 

 

 

 

 

34

In 5-aryl-1,l,2,2-tetracyanocy010propanes with a meta substituent,
the two ortho hydrogens are no longer equivalent. Therefore there should
be two doublets for the cyclopropyl hydrogen. This was the case with
5—mfnitrophenyl-l,l,2,2-tetracyanocyclopropane, which showed two doublets.
5-(5‘,H'-Methy1enedioxyphenyl)-1,l,2,2-tetracyanocyc10pr0pane showed a
triplet. This may be so probably because the original two doublets might
have been so placed as to make a triplet. 5-mfchlor0phenyl-l,l,2,2-
tetracyclopropane showed a complicated spectrum (Fig. 11), that is, the

Y ~ CN
XOICI

H "“ CN

 

CN

X = N02,, Y : H (2 doublets' , I ;
J = 0.8 c.p.s. forwboth) - Fig. 9

X, Y = -OCH20- ' (t, J €30.95 c.p.s.) . . Fig. 10
spectrum seemed to be a triplet, J = 1.0 c.p.s.,with the central peak
further Split into a smaller triplet, J = OIH c.p.s. A satisfactory
explanation for the complexity of this Spectrum cannot be offered at this
time.

More convincing evidence for four bond coupling between the cyClOprOpyl
hydrogen (H4) and the aromatic ortho hydrogen (H1) of a 5—ary1-l,l,2,2-
tetracyanocyclOpropane was obtained from the observation that, in 5—(2',h'-
dichlorophenyl)-l,l,2,2-tetracyanocyclopropane (XLIV), indeed H1 and HM
were Shown to couple to each other, J = 1.0 c.p.s. (Fig. 1H). The other

coupling constants observed were J (H1, H2) = 8.55 c.p.s., J (H2, H3) =

55

.ogmoamoaohoofimhomhpofim«maHaHIHhsogmoIHPfi/TMTM m8 demoed? Hammoamoaoho .Ho Sgpoomm mzz .m .mfim
.Aomzm

SIS oHQOHmoaosnooSwhomapopIm«maHaHIHhconmocdsmoImIm .Ho cowofiphm .TEOIaoaosmo mo geoomm ~92 .w .wfim

.mammogmoaohoofimhowhpofim«maHaHIHhQCQQOHPfi/HIQIM Mo ammonphm Hummoamoaoho mo Epoomm g .P .mflm

 

# _ a
Swine >541. Hmib

   

 

omegco moon.”

 

 

 

N.

OWQQOO QOOHHV

P( 7

1

 

 

 

om-goo moon-ha J

m .mam m .mam s .mam

 

 

 

 

 

 

 

56

 

XLIV

2.05 c.p.s., and J(H H5) = 0.45 c.p.s. The Hl appeared at the lowest

1’
field. This may be explained as follows. Due to restricted rotation of
the bond between the cyc10propyl and phenyl groups, XLIV would be likely
to retain the conformation Shown above. Consequently, H1 is the aromatic
hydrogen brought closest to the cyano groups and is least shielded.‘ This
would be probably true of most 5-aryl-l,l,2,2-tetracyanocyclopr0panes
haVing an ortho substituent.

If only the ortho hydrogensanmeinvolved in coupling with the cycloprOpyl
hydrogen, 5-(2',6‘-dichlor0pheny1)-1,l,2,2—tetracyanocyc10pr0pane (XLV)
where both ortho hydrogens are replaced by chlorine should give an unSplit

Singlet for the cyclopropyl hydrogen. This is the case with a DMSO

solution of XLV (Fig. 12). However, an acetone solution of XLV gave,

 

1 CN
——-CN
Cl
L_.
H ON
ON
XLV

in addition to the main peak, a small shoulder (Fig. 13). What caused this

shoulder cannot be answered at this time.

57

mammoagoaohoothomapopIm«maHaHIAahaonmonoanoHQI_ma.vam wo ammonphm Hamoamoaomo wo

.Aosopoom may

anasooam mzz .ma .maa
.Aomzm aav
mammoamoaohoosshomhpopIm«maHaHIAHSQmsmoHoagoHQI_ma.mNIm mo Gomoachm Ahmoamoaoho mo afihpommm mzz .mH .me
.summonmoaohooqwzowapopImam«HaHIHSQCQQOHOHQOIEIM mo somOHohm HSQOHQOHCSD mo sapwoomm mzz .HH .me
.deQoym
IOHCSOOQSSCHMPCpImamaH«HIAHhco£QSROHdocoamnpsz.3..vam mo cowoaphm HSQOHQOHohU mo adhpommm mzz .OH .mHm

 

 

_
all.

 

 

 

ma .maa

 

 

ma .maa

 

_
and.

h
so.m_.

 

 

 

.m.@.o m.ou.n b VT F

It.
.mOQOO :OOHHJ

Ha .mam

 

 

 

 

_
mm.m.e

F V
1

 

 

 

dds moons

 

 

oH .maa

 

 

38

 

 

Fig. 1H

 

 

 

 

 

 

 

 

 

 

 

 

 

 

|-—~l |-'-| J(H1, H5)=O.)—l-5 c.p.s.
I
. J(H H )—1 0
I l; 2+ — o c.p.s. h—'—"i'
‘ J
(H1, H2)=8.55 CopoSo
(a >1
I
71.81 11.93

I l

 

Fig. 11+. NMR Spectrum of ortho-Hydrogen of 5-(2',lI'-Dichlor0phenyl)-
1,1,2,2-tetracyanocyclopr0pane.

 

Long range coupling between an aromatic hydrogen and a hydrogen of
the sp2 carbon directly attached to that aromatic ring seems to require
the hydrogens participating in the coupling to be on a "zig-zag path"

(or trans-trans path) such as shown by the heavy lines in XLVI and XLVII.

 

For example, in salicylaldehyde where the hydroxyl group is hydrogen-
bonded to the carbonyl oxygen (XLVI), the aldehydic hydrogen couples only

with the meta hydrogen at the CB-position. However, in other ortho sub—

H\ fls N /H

XLVI XLVII

stituted benzaldehydes where no intramolecular hydrogen bonding is possible
and the carbonyl oxygen, because of steric reasons, is Sided with the
unsubstituted part of the benzene ring (XLVII), the aldehydic hydrogen
couples only with the meta hydrogen at the C5 position (56).

Recently, long range coupling of the system represented by formula
XLVIII, where Z is an Sp2 hybridized carbon, nitrogen, or oxygen, has been
extensively investigated (57). With some exceptions, the data generally

confirms the trans-trans coupling. To quote a few examples, salicylaldehyde

 

(XLVIII, z = C0, x = 2-0H) showed H as a doublet (in CH2Br2) with J (H1,

1

H = 0.6 c.p.s. and J (H1, H5) = 0 c.p.s. In o-chlorobenzaldehyde

5)
(XLVIII, z = C0, X = 2-Cl), R1 was a doublet (in CH2Br2 and DMSO) with

J = (H1, H5) = 0.7 c.p.s. and J (H1, H5) = 0 c.p.s.l A triplet (in acetone)

MO

 

XLVIII .

was observed for H1 of penitrobenzaldehyde (XLVIII, Z = 00, X = A-NOB),

_ J = 0.55 c.p.s. It is so because, due to free rotation of the bond
joining the formyl group and phenyl group, Hl can be trans-trans oriented
to both H5 and H5, thus split by these two equivalent hydrogens.

By analogy with the trans-trans coupling between the aromatic hydrogens

2

 

and the hydrogen of the Sp carbon mentioned above, the cycloprOpyl hydrogen

2 character

of 5-ary1-l,1,2,2~tetracyanocy010pr0panes, because of some Sp
of the cycloprOpyl ring carbon (58), might be expected to couple with the

meta hydrogens, which can truly be trans-trans oriented with respect to

 

the cyclopropyl hydrogen. However, no coupling between these two kinds
of hydrogens was observed. This could suggest that either these hydrogens
do not couple at all or they couple but with such a small magnitude that

it cannot easily be observed.

D. 5,5-Dialkyl-2-carbethoxy-l,1,2-tricyanocyclopr0panes and 5—aryl—
2-carbethoxy-l,1,2-tricyanocyclOpropanes

 

For preparation of 5,5-dialky1- and 5-aryl-2-carbethoxy-1,l,2-
tricyanocycloprOpanes (XLIX), the following two routes were employed. The

products thus prepared are listed in Table 6.

Ml

 

 

 

 

 

Route A
CN
R CN R CN
\/C=C< + CHBr(CN)2 ———+ '16
R' C00C2HS R CN
000ch5
Route B XLIX
CN
R\ /CN R CN
C==C + CHBr(CN)CO0C H -————e» 17
/’ \\ 2 5 ———CN
R' CN R'
000C2H5
XLIX

Generally compounds of this group are produced in poorer yield than the
corresponding tetracyanocyclOpropaneS. 0f the two routes employed,
route A produces the products quicker and in better yield than route B.
Besides, several compounds that failed to form by Route B were produced
by route A. The relative superiority of route A.may be ascribed to the
greater acidity of bromomalononitrile. Since pKa of bromomalononitrile
(~45) is less than that of ethyl bromocyanoacetate (“J6) (5), the con-
centration of anion present will be greater in the former case. Further—
more, the bromocyanocarboxycarbanion is probably bulkier than the
bromodicyanocarbanion, so that its attack on the fl-carbon of an ethyl
alkylidenemalononitriE:would be more sterically hindered.

For the products (XLIX), where R t R', two stereoisomers would be
expected. The NMR Spectra (in acetone-d6) of 5—methyl—5—ethyl-2-carbethoxy-
1,1,2-tricyanocyclOpr0pane (L) and 5-methyl-5-nfpr0pyl-2-carbethoxy-l,1,2-
tricyanocycloprOpane (LI) Showed the cyclOprOpyl methyl Signals at two
positions,T8.26 and 78.39 for L and18.25 and 78.38 for LI. The intensity

of the two peaks in L and LI were comparable to each other. Therefore,

A2

 

 

 

Table 6. 5,5-Dia1kyl— and 5-Aryl-2-carbethoxy-l,1,2-tricyanocyc10pr0panes
0N
R CN
N
R R' (Ar) Route Reaction timea Yield, at
CH5 CH5 A 1 hr. 73.5
B 5 hrs 28.6
CH5 C2H5 A 20 hrs b 63.7
B c 0
CH5 g-C5H7 A 30 daysd 37
B c 0
CH l-C H A 14 days8 6.46
5 _' 5 7 B c 0
CgH5 02H5 A 2 daysf 35.2
B c 0
(CH2)IL A 2 days 14.5
B c 0
(CH2)5 A 1 hr. 97.5
B 8-9 hrs g 25.8
H C6H5 A 5-6 hrs.g 76.7
, B 5 hrs.g 29
H ETCH5OC6H4 A Overnight 56.2
B 2 days 60.5

 

OQI-tmpIoo‘

Time elapsed before the crystalline product separated in the

mixture.
Followed by refluxing for an additional 5 hrs.

reaction

Allowed to stand at room temperature for not less than a month.
Crystals obtained upon removing the solvent.
Initially refluxed for 5 hrs.

Followed by refluxing for an additional 6 hrs.
Reaction mixture was refluxed.

45

CN

R ———C00C2H5
CH ' ———CN

 

CN

L R = C2H5 78.39, 17826
LI R = n-CBHY 78.38, 18.25
LII R = 31;th7 78.35, (78.25)

these two compounds probably contained about equal amounts of the two
possible stereoisomers. But the NMR Spectrum (in acetone-d6) of 5-methyl-
5-iSOpr0pyl-2-carbethoxy-l,1,2-tricyanocy010pr0pane (LII) showed a very
strong peak at‘T8.55 and only a negligibly small peak at 78.25. This
suggests that only one isomer of LII was produced.

The configuration of the predominating stereoisomefof 5-methyl-5-
isopropyl-2-carbethoXy-l,1,2-tricyanocyclopropane (LII) was determined
as follows. In the NMR Spectra (in acetone-d6), the two methyls of 5,5-
dimethyl-1,1,2,2-tetracyanocyclopropane appeared at 78.25 and those of
5,5-dimethyl-2-carbethoxy-l,l,2—tricyanocyclOpropane (LIII) at 78.52 and

78.23.

CN CN

78.23 ’5

CN
CH; CN C55 ON C00C2H5

78.32
LIII

44

This suggests that the signal at 78.25 in compound LIII must be due-to the
methyl which is cis to two cyano groups and that the signal at T8.52

due to the methyl cis to one cyano and one carbethoxyl group. Likewise,
in compounds L, LI, and LII, the signal at the higher field must be

caused by the methyl cis'to carbethoxyl group. Therefore, the strong
methyl signal at 78.55 of 5-methy1-5-isopropyl-2-carbethoxy-l,1,2—tri—
cyanocycloprOpane (LII) must be due to the isomer in which the methyl

is cis to the carbethoxyl group (LIIa).

CN N
CN CH, CN
(CH5)2C 7‘8.25 5
\ \
N
8 CH5 C C00C2H5 (CH5)2CH CN 00002H5
T '55 LIIa LIIb

The almost exclusive formation of LIIa starting from ethyl 5-methy1-
2-butylidenecyanoacetate (LIV) containing comparable amounts of both
isomers* suggests that the reaction very likely proceeds through an
intermediate carbanion which can undergo inversion before forming

the product (see Scheme 2).

 

* The NMR Signals for the allylic methyls of the two isomers (LIVa and
LIVb) of this compound appeared at’77.84 and 77.76. The relative
intensities of these peaks were 55:65, respectively. LIVa appears to
be more favored and thus occurs in greater population (65%).

H 1'6.72 ' II T5'99
_ I /4_
(CH5)2C\\ ,/CN (CH5)2C\\ ‘_ //C0002H5
T884 ‘/£=w\\ 7901 //C—d\
CH5 C00C2H5 Cg; CN
. “77.84
17 76 LIVa LIVb
/}£‘T6. .83
g) C/
18 84 2 \c— —/C
CH/ \\:N Footnote continued on
7.84 _5 page 46.

 

1+5

20

m

  

 

 

..momAmmoV

 

 

m

m magnum

mmoooo

 

Ioz

 

I..Imomfimmov

amomnozv

5-Phenyl-2-carbethoxy-1,1,2-tricyanocyclopropane (LVI) and its pfmethoxyl
derivative (LVIII) appeared to be obtained as only one stereoisomer,
presumably the isomer in which the phenyl or pfmethoxyphenyl is trans to

the carbethoxyl.

 

The cyclopropyl hydrogen of the esters, LVI (T 5.74) and LVIII (7-5.79)
were appreciably more shielded than those of their cyano equivalents, LVII
(1'5.l5) and LIX (7 5.27), suggesting that the cyclOprOpyl hydrogens of the

esters are likely to be cis to the carbethoxyl groups.

 

That the iSOprOpyl group of LIVa is cis to the cyano group is supported by
the NMR chemical shifts of both kind's'Bf isopropyl hydrogens, 78.84 and
16.72, reSpectively, compared to those data of LV (7'8.84 and.76.85, res-
pectively). Similarly, that the allylic methyl group of LIVb is cis to the
cyano group is also supported by its NMR chemical Shift (7-7.84) CSEpared
to the NMR chemical Shift (7'7.84) of LV. It is interesting to note that,
in these particular compounds, the groups cis to carbethoxyl group (allylic
methyl in LIVa and the methine hydrogen in'II'be) are less shielded than
those cis to the cyano group (the allylic methyl in LIVb and the methine
hydrogEH—in LIVa). A possible reason for this observation would be that, in
the preferred conformation of LIVa and LIVb, the allylic methyl of LIVa and
the methine hydrogen of LIVb may come close to the carbethoxyl ether oxygen.

47

The two isoprOpyl methyl groups of 5-methyl-5-isopr0pyl-2-carbethoxy-
1,1,2-tricyanocyc10propane (LII) appeared in two doublets at 78.95 and
78.87, respectively. Ignoring the minor stereoisomer (because of its
negligibly small pOpulation) different conformers of the major isomer

are shown in the Newman projections as below.

 

NC
H CN NC H5 CN N CH5 CN
NC COOC2H5 NC OOCgH5 NC COOCEHS
' H H
CH5 CH5 CH5 H H C 5
LIIC LIId LIIe

Conformer LIBzwould probably be most stable since it holds the least
number of large groups in close proximity. If LII stays most of the
time in conformation LIIc, certainly two isopropyl methyls are non-
equivalent with each other. Therefore, two doublets would result. This
magnetic non-equivalence between two iSOprOpyl methyl groups has been
encountered and explained in terms of preferred conformation (59, 40).

E. 5,5-Dialkyl- and 5-Aryl-2-carboxamido-l,l,2-tricyanocyclopr0panes
and their Reactions

 

1. Reaction of Cyclohexylidenecyanoacetamide with Bromomalononitrile

When cyclohexylidenecyanoacetamide reacted with excess of bromo-
malononitrile in about 80% aqueous ethanol for two days at room temperature,
7-10% of 5,5-pentamethylene-l,1,2,2-tetracyanocyc10pr0pane (LX, identified
by m.p., m.m.p., and IR spectrum) was obtained in addition to 60% of crude

5,5-pentamethylene-2-carboxamido-l,1,2-tricyanocyclopr0pane (LXI).

#8

Two possible routes to the formation of LX are given in Scheme 5.

In one route, cyclohexylidenecyanoacetamide first adds bromomalononitrile
to form a Michael addition product LXII. LXII, eliminating bromocyano-
acetamide, forms cyclohexylidenemalononitrile, which easily reacts with
bromomalononitrile to form LX. The alternative route involves hydrolysis
of cyclohexylidenecyanoacetamide to cyclohexanone, which reacts with
bromomalononitrile to give XL (Cyclohexanone was found to react with
bromomalononitrile to form LX).

The crude major product, 5,5-pentamethy1ene-2-carboxamido-1,1,2-
tricyanocyclopropane (LXI) melts over a broad range of temperature (1M0-
180°). On recrystallizing from boiling methanolf the product is converted
to imino amide LXIII (Scheme 5). The imino amide (LXIII) showed IR
absorption bands (in nujol) at 5580 cm-1 (NH), 2260 cm_:L (CN), 1757

cm.1 (00), 1652 cm-1

, 1567 cm.1 (both strong and broad, C=N and/or NH).

The NMR spectrum (DMSO-d6 at 100°) showed ring hydrogens at 18.58 and
78.10 and the N-hydrogens at‘rl.22 in an area ratio of 10 : 1.8. A more
reliable structural proof for LXIII was provided by conversion of LXIII

to LXIV by treating LXIII with hydrochloric acid. LXIV'was identical with
authentic sample whichVMK3synthesized from LXV by a known method (#1).
Conversion of LXI into LXIII is believed to involve the attack of the amide
nitrogen at the cyano carbon on the vicinal position. Ring closure by

means of the attack of an amide nitrogen on the cyano carbon at a vicinal

position has been reported in the literature (15).

 

* Other organic solvents at an elevated temperature result in the same
effect on LXI.

1+9

m . ‘ 20
..\ .

.o
H A V .. nu
BS . mooom .m
. mam .H
‘ 20
no 20
>qu

 

0
gas
.__m_>

 

 

.363 o
m mMVII
2%... V0
£anme
83 75
£2835 78/ \oz Us \
Z Hm\0 NEH/HO \
v c .. v.
«I Ham/WV .
78 6'8
/ 78
oozmm

 

m oaoflom

SO

2. Reaction of Cyclohexylidenemalononitrile with Bromocyanoacetamide.

When cyclohexylidenemalononitrile reacted with bromocyanoacetamide
in aqueous ethanol for several days, 5,5-pentamethy1ene-1,1,2,2-tetra-
cyanocyclopropane (LX) was obtained in 19.1% yield. Upon working up the
reaction residue, unchanged bromocyanoacetamide was recovered. Possible
mechanisms for the formation of LX are presented in Scheme h.

Scheme h

CH(CN)2
___——+ CHBr(CN)2

*1.

NC //C\b Br CN
NCCHBrCONH2 ONHg C>=<
‘ . CN

N
N “‘ CN
0 .N
N CN LX
H20 \\\
CN
\ 9“ <:>=g
CHBr
H(CN)2 I N

CONHg

 

‘W ___.. CH2(CN)2 >CHBr(CN)2

H»?

5. Reaction of Arylidenecyanoacetamides with Bromomalononitrile

The reaction of some arylidenecyanoacetamides and bromomalononitrile
is summarized in Scheme 5. Like 5,5-pentamethy1ene-2-carboxamido-l,1,2-
tricyanocyclopropane (LXI), 5~phenyl-2-carboxamido—1,l,2-tricyanocyclo-
propane (LXVIII) and its pfchloro derivative (LXIX), on heating, were

exclusively converted into the imino amides LXX and LXXI, respectively.

51

 

Scheme 5
CN
Ar CN Ar t——CN
>0=c< + CHBr(CN)2 .___.
H CONH2 H‘ ———CN
LXVII Ar=prlC6H5 LXVIII Ar=C6H52\\\
LXIX Ari-p-ClC6H5 CN NH
H
ON
ON 0+ LXX Ar=C6H5

H3
LXXI Ar: -ClC H
Aflt>x<__<f) E. 6 u ‘
LXXII Ar=C6H5 0

CN

 

LXXIII Ar=ijlcéHu

Treatment of LXX and LXXI with hydrochloric acid yielded the carboximido
compounds, LXXII and LXXIII, respectively. (5-p-Methoxyphenyl-2-carboxamido-
l,1,2-tricyanocyc10pr0pane produced a mixture of compounds which could not

be identified.)

F. Some Dimers of Alkylidenemalononitriles

 

It was observed that cyclopentylidenemalononitrile (LXXIV), on treating
with pyridine, gave light yellow crystals, m.p. l87-l90°. This crystalline
product had also been fOrmed (in the distillation flask) while distilling
LXXIV. -Analysis gave an empirical formula (08H8N2)n. Since a number of
alkylidenemalononitriles are known to dimerize (42, #5, AM), this compound was
suspected of being cyclopentylidenemalononitrile dimer C16H16Nu' The IR Spec-

trum (in nujol) showed the presence of NH2 (5Hh0 cm-l, 5570 cm_l, 5260 cm-1)

52

and ON (22h0 cm-l). The absorption bands at 1650 cm.1 and 1589 cm-1

could be ascribed to 0:0 and/or NH. The NMR Spectrum (in DMSO-d6) showed
three multiplets at'r8.25, 17.61, and'T6.9l, a doublet (J = 1.8 c.p.s.)

at ‘74.52 and a singlet at‘T2.72 in an area ratio of (three multiplets as 15)
15 : 0.97 : 1.89. The UV spectrum (in ethanol) showed absorption bands

at 508 mpc(e 12,226), 2A1 my. (e 8,681), 224 mp.(£ 6,659), and 215 mph
(shoulder).

A.mechanism similar to the one advanced for the formation of the
isOprOpylidenemalononitrile dimer (#2) can be applied to account for the
formation of cyclopentylidenemalononitrile dimer (SchemeéS). Structures
LXXV and LXXVII are eliminated because the NMR Spectrum showed a hydrogen
in the vinyl region (T‘h.52)*. The NMR signal atq’h.52 is a doublet,

Scheme 6

N ,4 LXXV

c “I"

NH2

CN LXXVII

 

1!

If the compound is a mixture of these three isomers, the percentage of
LXXV and LXXVII must be negligibly small.

*-

53

J = 1.8 c.p.s., although it is expected to be a triplet for LXXVI.
Examination of a model of LXXVI reveals that the two dihedral angles
between H1 and two H2's are not the same. One is almost 90° whereas

the other is approximately us—6o°. Only the H with a dihedral angle

2

rvh5-60° should split H Therefore only a doublet would be observed

1'

for H . The magnitude of J (dihedral angleruh5-60°) = 1.8 c.p.s.
1 H1’ H2

(for LXXVI) is favorably compared to that of J (dihedral angle

H1: H2
65°) = 2.1 c.p.s. observed for cyclOpentene LXXVIII (45). A precedent

in which a vinyl hydrogen coupled to only one of two neighbouring

 

LXXVIII

methylene hydrogens was encountered, i.e., compound LXXIX showed H1 at

75.u5 (in CClu) as a doublet, J = 5.5 c.p.s. (M6).

 

LXXIX

Isopropylidenemalononitrile dimer (LXXX), m.p. 17l-l74°, and 2-butyli-
denemalononitrile dimer (LXXXI), m.p. l67-l69°, were prepared by treating
the corresponding monomer with pyridine. They were also formed (in the

distillation flask) while distilling the corresponding monomers*.

 

* IsOprOpylidenemalononitrile dimer was also found to form while storing
the monomer at room temperature for a few months.

54

Data for these dimers agreed with those reported in the literature (h2,hh).

CH5 CH3
CH5 , CHBCH
CN
NC NC ON
ON ON
NH2 NH2
LXXX LXXXI

G. Miscellaneous

 

l. Sulfur- containing Compound obtained from Ethyl Sodiocyanoacetate
and Carbon Disulfide

 

In an attempt to prepare ethyl bromocyanoacetate according to a
reported method (47), dry ethyl sodiocyanoacetate suspended in carbon
disulfide was allowed to react with bromine. Instead of the desired
ethyl bromocyanoacetate, there was obtained a yellow compound, m.p. 252-
256° (from benzene)-. Analysis gave an empirical formula C12H10N20485.

-l

The IR spectrum (in nujol) showed the absorption bands at 2210 cm (CN),

1661 cm'1

(co), 1181 cm'1 (ester band?) The NMR spectrum (in DMso-d6 at
80°) showed only a triplet at‘T8.62 and a quartet at 75.60; The UV spectrum
(in acetonitrile) gave only one peak at 55h mpt(€ 27,600). Compound
C12H10N20u85 rapidly decolorized potassium permanganate solution but did

not decolorize bromine appreciably. Structure LXXXII had been assigned to

this compound by Wenzel who first prepared this compound by the identical

procedure. However, some of the observed data do not satisfy structure

NC CN

ii

LXXXII

 

02H500C cooc2H5

55

LXXXII. For example, the IR bands at 2210 cm—1 is strongly indicative
of a conjugated cyano group and is too low for the cyano group of
LXXXII. Besides, LXXXII would not decolorize potassium permanganate
so rapidly. Therefore another structure LXXXIII is pr0posed for this
compound (stereochemistry is not known). The IR spectrum of LXXXIV

-1 -1
showed bands at 2209 cm for CEEN and 1669 cm for COOCH5 (M9). These
values are close to 2210 cm-1 (CN) and 1661 cm-1 (C00C2H5) of LXXXIII

and seem to suggest the presence of -S)EC=C(CN)COOC moiety in the

2H5

molecule. It is reported that methyl sodiocyanoacetate, on treating with

NC ‘ CN
CHBS\ _ /CN
/C—C\.
HS COOCH
CQHSOOC S COOC2H5 5
LXXXIII LXXXIV

carbon disulfide followed by oxidation with ammonium peroxysulfate, pro-
duced compound LXXXV (49). Assignment of the structure was based on the

low frequency of CEEN absorption band in IR spectrum (data not given).

NC CN
S

CHBOOC s-——— COOCH5

LXXXV

A possible mechanism for the formation of LXXXIII is pr0posed in

Scheme 7.

56

Scheme 7
CN
NC /s NC SH
“:2 02H5000 ' CQHSOOC s-
0000 H s
2 5

[0]Bromine

C2HSOOH 00002H5 C2H5 00 S———— OOCQHS

LXXXIII
Oxidation of LXXXIII (Cl2H10N2OhS5) using potassium permanganate in
acetone, produced a new compound, Cl2H10N20h32"m'p' 177-179° (from an
acetone-ethanol mixture). The IR spectrum (in nujol) showed absorption
bands at 2250 cm-l (CN), 1696 cm.1 (CO), 1560 cm-:L (C=C), 1188 cm_1 (ester
band?), and the UV spectrum (in ethanol) showed three absorption bands at
215 mp.(e 16,755), 555 m+L(€ M8,5u5), and 5&8 nui(€ 52,8ou). Structure LXXXVI

is proposed for this compound (stereochemistry HOt known).

NC>4HiN
C2H5OOC 0002H5
LXXXVI

An attempted reduction of LXXXIII using lithium aluminum hydride

or sodium borohydride gave an unidentified residue. Also, desulfurization
of LXXXIII using Raney nickel was tried, but there was obtained a dark

liquid residue with an amine odor, which could not be identified.

57

2. Reaction of 2-Methyl-l-nitr0pr0pene with Bromomalononitrile

 

Reaction of 2-methyl-l-nitropr0pene (above-85% pure as determined
by the NMR and v.p.c.) and bromomalononitrile in aqueous ethanol produced

two compounds C10H7BrNu (white crystals, m.p. 201-202°) and C H8NAO

7 5
(yellow crystals, m.p. 251-252°). The white compound crystallized out
first and, several hours later, the yellow compound followed.

Compound 0 BrNu showed CEEN band at 2290 cm—:L in the IR (in nujol)

10H7
spectrum. The compound did not have absorption for the entire accessible
UV and Visible range (in ethanol). The NMR.spectrum (in acetone-d6) showed
two singlets at‘T7.78 and 75.68 in an area ratio of 6 : 1. Three possible
structures LXXXVII, LXXXVIII, and LXXXIX are pr0posed for this compound

and a possible mechanism for the formation of each of them is shown in
Scheme 8. Structure LXXXVIII was eliminated for the following three
reasons. (i) The compound was transparent in the UV and Visible, whereas
alkylidenemalononitriles absorb intensely between 250-250 my.(due to
::C::C(CN2). (ii) The NMR singlet at 5.68 is too high for the vinyl
hydrogen in a compound of type XC, which usually appears between‘Tl.00 and

75.00 depending on the nature of R group. (iii) The IR spectrum lacked

\\C==C//

// \\

R CN

H CN

XC
absorption in the double bond region. Structure LXXXVII was also eliminated
because the NMR singlet at‘T7.78 is not compatible with two geminal methyl

groups, which should appear as two separate peaks.

57

2. Reaction of 2-Methyl-l-nitr0propene with Bromomalononitrile

 

Reaction of 2-methyl-1-nitr0pr0pene (above-85% pure as determined
by the NMR and v.p.c.) and bromomalononitrile in aqueous ethanol produced

two compounds ClOH7BrN,+ (white crystals, m.p. 201—2o2°) and C HBNuo

7 5
(yellow crystals, m.p. 251-252°). The white compound crystallized out
first and, several hours later, the yellow compound followed.

Compound C Ber showed CEEN band at 2290 cm'1 in the IR (in nujol)

10H7
spectrum} The compound did not have absorption for the entire accessible
UV and Visible range (in ethanol). The NMR spectrum (in acetone-d6) showed
two singlets at‘T7.78 and 75.68 in an area ratio of 6 : 1. Three possible
structures LXXXVII, LXXXVIII, and LXXXIX are pr0posed for this compound

and a possible mechanism for the formation of each of them is shown in
Scheme 8. Structure LXXXVIII was eliminated for the following three
reasons. (i) The compound was transparent in the UV and Visible, whereas
alkylidenemalononitriles absorb intensely between 250-250 my.(due to
::C=:C(CN2). (ii) The NMR singlet at 5.68 is too high for the vinyl
hydrogen in a compound of type XC, which usually appears between‘Tl.00 and
15.00 depending on the nature of R group. (iii) The IR spectrum lacked
R\\C::c//CN
H// \\CN
XC

absorption in the double bond region. Structure LXXXVII was also eliminated
because the NMR singlet at'T7.78 is not compatible with two geminal methyl

groups, which should appear as two separate peaks.

58

mm

mm

mm

<®

m1

NHVOOQH

m

6mAm26v

zo 2m\

6
22662226 2 m26
mA26V6 _)¢

§Qm8

 

2A26V6n26-226mflm26v

NUOOQH

 

2
26 62
m mm D
26 (y 26 226-2_ , .+||||||.
. 2 mi. 622662226 A V t - 22- \\\w 2
26 A 266 62 . m
A a 62
26

HHHBOOS
m 02

,(L

 

 

 

HH>xqu 2 26
26 m. f\66\
2 26 .1 2W6/ 2
226 mA26vam6.|.§ 22266226-,zlmwu
26Vs26 6m2moa 262

m ofionom

ml iHWJH:

£26626.|.26/.

 

 

26
m a
wn266226.|||.w.||26 mA26Vsm6-\\\\t 2
m26 62 m62

 

266
2
m VnuuA
Azovpmmo moz
2
mA26Vsm26
mo
2
mA26Vsm26 V A
mo2

59

Structure LXXXIX agrees well with all the spectral data obtained.
When propane, with TCH5 = 9.10 (50), is converted into 2-bromopropane,

with 1’CH = 8.29 (51), the magnitude of the deshielding effect caused

3
by bromine on the vicinal hydrogens is 0.81 p.p.m. Now, 5-iSOpr0pyl-
1,1,2,2-tetracyanocyclopr0pane Shows the two iSOpropyl methyl groups at
7'8.65 and the cyclopropyl hydrogen at T6.50 (in acetone-d6) (28). If
this vicinal bromine effect is applied, then the expected chemical shifts
for two methyls and the cyclOprOpyl hydrogen in LXXXIX would be T(8.65 -
0.81) or 17.82 and 1(6.50 - 0.81) or‘r5.69, respectively. These eXpected
values are close to the observed values of‘T7.78 and.75.68,respectively.
The transparency in UV and the high CEEN frequency in IR agree with

structure LXXXIX. Thus it is very likely that compound ClOH BrNu is

7
5-(2'-bromois0pr0py1)-l,l,2,2-tetracyanocyclopr0pane (LXXXIX). 0f the
two possible mechanisms proposed for the formation of LXXXIX (see Scheme
80 and 8D), 8D seems to be less likely because the carbonium ion may not
be stable.

The yellow compound C7H8Nu0 showed IR absorption bands (in nujol)

5
at 5550 cm-l, 52h5 cm—l, 5175 cm-l, 2255 cm-1, 1659 cm-l, 1585 cm-1, 1556
cm-1. Two peaks appeared in the NMR spectrum (in DMSO-d6) at‘T8.h9 (s)
and.71.02 (broad), reSpectively, in an area ratio of 6 : 1.82. These data
suggest an amide for this compound. However, a plausible structure has

not yet been deduced for this product.

5. Reaction of Is0pr0pylidenemalononitrile with Bromonitromethane

 

When is0pr0pylidenemalononitrile was allowed to react with bromonitro-
methane, a small amount of 5,5-dimethyl-l,l,2,2-tetracyanocyclopr0pane (XCI)
was obtained instead of the desired product, 5,5-dimethy1—2-nitro-l,l-dicyano—

cyclOpropane.

60

Two possible mechanisms for the formation of XCI are shown in
Scheme 9.

 

Scheme 9
\\\C CH(CN)2 < )
=' CHBr CN
/// 2
\VHHFLBr CN
- N0
CHQBrNOQ 2 >=<
ON
ON
N

 

 

XCI
Hr N
CH > <
\\ CH(CN)2 I 2 cm
/ ,3 CH2(CN)2 ( >2
0'.

EXPERIMENTAL

A. General ProcedUres

 

Unless otherwise Specified, all IR spectra were obtained on a
Unicam S P' 200 Infrared SpectrOphotometer and the absorption bands
were expressed in cm.1 (u); all UV spectra were obtained on a Beckman.
DB SpectrOphotometer and the absorption bands were expressed in mpL
(E); all NMR spectra were obtained on a Varian A260 Spectrometer using
tetramethylsilane as an internal standard and the signal positions
were expressed in T units.
All melting points were taken in a sealed capillary tube and were
uncorrected. _All microanalyses were performed by the Spang Microanalytical

Laboratory, Ann Arbor, Michigan.

B. Starting Materials

 

All compounds described in this section were identified by IR, UV, and
NMR Spectra (as far as possible) and by comparing the physical data such
as b.p., m.p., refractive index, etc. with those reported in the literature
(in case the compounds were known).

1. Bromomalononitrile

 

Prepared from malononitrile and bromine (28).

2. Ethyl Bromocyanoacetate

 

Prepared from ethyl cyanoacetate and bromine (52).

5. Ethyl Bromonitroacetate

 

Prepared from ethyl nitroacetate and bromine in the same manner

61

62

described for the preparation of ethyl bromocyanoacetate, bu 79-80°.
NMR (neat): 1-8.6u (t), CHSCH20, 1’5.62 (q), CH50H20, 1-5.u0 (s),
OQNCEBI.

4. Ethyl Nitroacetate

 

Prepared from ethyl acetoacetate, acetic anhydride, and fuming

nitric acid (55), b6 91°, ugh 1.u220. NMR (neat): 1‘8.70 (t), CHSCH2O;

1’ 5.70 (q), CHBCHZ‘; 1-u.72 (s), 02NCH2.

5. Bromocyanoacetamide

 

Prepared from dibromocyanoacetamide and cyanoacetamide (5h), m.p.
llh-ll7°. NMR (in acetone) showed two singlets at T h.58 and 1‘2.6h in
an area ratio of 0.9h : 2.

6. Dibromocyanoacetamide

 

Prepared from cyanoacetamide and bromine (55), m.p. l21-l2h°.

7. Dibromoacetonitrile

 

Prepared from ethyl cyanoacetate, bromine, and magnesium oxide

2M

D 1.5594. NMR (neat): 7'u.15 (s).

(56)) bQO 70-730) n

8. Bromonitromethane

 

Prepared from nitromethane, bromine and barium oxide (57), b. lh5-
150°. NMR (neat): 1 u.l7 (s).

9. Alkylidenemalononitriles

 

Isopropylidenemalononitrile and 5,5-dimethyl-2-butylidenemalononitrile
were prepared according to the procedure described by Frout (58). All
other alkylidenemalononitriles were prepared in the manner described by

COpe and Hancock for the preparation of ethyl 5~pentylidenecyanoacetate (59).

63

10. Ethyl Alkylidenecyanoacetates

 

Ethyl isoprOpylidenecyanoacetate was prepared according to the
procedure described by Frout (58). All other ethyl alkylidenecyano-
acetates were prepared in the manner described by COpe and Hancock for
the preparation of ethyl 5-pentylidenecyanoacetate (59).

ll. Arylidenemalononitriles

 

p—Methylbenzylidenemalononitrile was prepared in the manner described
by COpe and Hancock for the preparation of ethyl 5-pentylidenecyanoacetate
(59). All other benzylidenemalononitriles were prepared according to
the procedure of Corson and Stoughton (60).

12. fl-Arylalkylidenemalononitriles

 

Prepared in the manner described by COpe and Hancock for the prep-
aration of ethyl 5-pentylidenecyanoacetate (59).

15. Miscellaneous

 

Ethyl benzylidenecyanoacetate (61), ethyl pfmethoxybenzylidenecyano-
acetate (60), benzylidenecyanoacetamide (62), pfmethoxybenzylidenecyano-
acetamide (65), and p-chlorobenzylidenecyanoacetamide (62) were prepared
according to the procedure: reported previously. 2,5-Benzocyc10pentylidene-
cyanoacetamide and 2,5—benzocyclohexylidenecyanoacetamide were prepared in

the manner described by COpe and Hancock for the preparation of ethyl

5-penty1idenecyanoacetate (59).

C. Preparation of 5,5-Dia1kyl-l,1,2,2—tetracyanocyc10pr0panes

 

1. 5,5-Dimethy1-l,l,2,2-tetracyanocyclopr0pane
Method I. In a 50-ml. Erlenmeyer flask were placed 0.5 g. (h.72 mmoles)

of isOpropylidenemalononitrile and 5 ml. of 50% aqueous ethanol. To this

6h

solution was added 1 g. (6.90 mmoles) of bromomalononitrile. A precipitate
formed within 2 mins. After one-half hr., the crystals were collected by
means of filtration and recrystallized from an ethanol-acetone mixture,
m.p. 206—208° (lit. val. (1) 209.5-210°), 0.69 g. (86%). The IR and

NMR spectra agreed with those obtained previously (28).

This reaction was found to proceed also in 50% aqueous acetone (71%)
and in 95% ethanol (68.2%) and more slowly in water (72%).

Method II. In a 50-ml. Erlenmeyer flask was dissolved 0.5 g. (5.h5
mmoles) of bromomalononitrile in 10 ml. of 50% aqueous acetone. The
solution was kept at room temperature for 2h hrs. A white precipitate
that had formed was worked up as described in Method I, m.p. 20h—206°,
0.25 s- (78.5%).

2. 5-Methy1-5-ethy1-1,l,2,2-tetracyanocyc10pr0pane

 

One gram (8.5 mmoles) of 2-butylidenemalononitrile and 1.h5 g.
(10.0 mmoles) of bromomalononitrile were dissolved in 15 m1. of 80% aqueous
ethanol. Crystals formed within a few mins. After a few hrs., the
crystals were filtered and recrystallized from an ethanol-acetone mixture,
m.p. 20u-206° (lit. val. (5), 202—202.5°), 1.59 g. (91%). IR (in nujol):
2265 (u.u2),CN; 982 (10.18), cyclopropane ring (?). NMR (in acetone-d6):
’r 8.72 (t), CH20H

5
5. 5-Methy1—5en-propyl-l,1,2,2-tetracyanocyclOpropane

5 1'8.25 (5), ring methyl; 1'7.92 (q), CH20H5.

 

Eight-tenths gram (5.95 mmoles) of 2-pentylidenemalononitrile and 1.5
g. (8.97 mmoles) of bromomalononitrile were dissolved in 20 m1. of 75%
aqueous ethanol. The precipitate which had formed in a few mins. was
filtered and recrystallized from ethanol, m.p. 16h-166° (lit. val. (A),

167.5-168°), 1.15 g. (97.5%). IR (in nujol): 2280 (u.59), CN; 985 (10.17),

65

cyclOprOpane ring (?). NMR (in acetone-d6); 1'8.9h (t), CHSCHQCHg;
’T 8.25 (s), ring methyl; 1‘8.07 (m), CH5(CH2)2.

A. 5-Methyl—5-isopr0py1-1,l,2,2-tetracyanocyclopr0pane

Five-tenths gram (5.75 mmoles) of 5emethyl-2—buty1idenemalononitrile
and 0.87 g. (6.00 mmoles) of bromomalononitrile were mixed in 12 m1. of
80% aqueous ethanol. A precipitate started forming in 1 hr. After a few
hrs., the first cr0p was collected and after 2h hrs., the second crop was
collected. To the filtrate obtained after removal of the second crop was
added an additional 0.87 g. (6.00 mmoles) of bromomalononitrile. There-
after crystals were collected every 2h hrs. for 5 days. Total yield,
after recrystallization from ethanol, amounted to 0.72 g. (97.h%), m.p.
201-202.5° (lit. val. (u), 187—188°). IR (in nujol): 2260 (4.42), CN;
985 (10.17), cyclOpropane ring (?). NMR (in DMSO-d6): 1‘8.82 (d),
(CH5)2CH; 7'8.h9 (5), ring methyl; 1‘8.52 (septet), (CH5)2CH,

5. 5-Methy1-5-n-pentyl-l,1,2,2-tetracyanocyc10pr0pane

2-Hepty1idenemalononitrile (1.25 g., 7.71 mmoles) and bromomalononitrile
(l.h5 g., 10.0 mmoles) were dissolved in 10 ml. of 90% aqueous ethanol. A
precipitate formed in 1 hr., which was filtered after a few hrs. Recrystal-
lization from ethanol yielded 1.70 g. (97.5%) of white crystals, m.p. 105—
105°. IR (in nujol): 2280 (u.58), CN; 982 (10.18), cycloprOpane ring (?)
NMR (in DMso-dg): 1-9.08 (t), CH5(CH2)BCH2; 7'8.75 (m), CH5(CH2)BCH2;
1' 8.u1 (s), ring methyl; 1 8.27 (m), CH5(CH2)ECH2.

Anal. Calcd. for ClBHluNuz C, 69.00; H, 6.28; N, 2u.76

Found: C, 69.14; H, 6.573 N, 24.78

6. 5-5-Diethy1-l,1,2,2-tetracyanocyc10pr0pane

 

Method I. 5~Pentylidenemalononitrile (1.55 g., 10.1 mmoles) and

66

bromomalononitrile (2.00 g., 15.8 mmoles) were dissolved in 15 ml. of
ethanol. A precipitate formed in a few hrs. and was filtered after
standing for an additional several hrs. Recrystallization from ethanol
gave 1.75 g. (88.5%) of white crystals, m.p. l65-l65° (lit. val. (4),
167-168°). IR (in nujol): 2276 (4.59), CN; 980 (10.20), cyclopropane
ring (?). NMR (in acetone-d6): 1 8.76 (t), CHBCHQ; 1—7.96 (q), CHBCHQ.

Method II. In a 50-ml. Erlenmeyer flask were placed 0.86 g. (10.0

 

mmoles) of 5—pentanone, 1.00 g. (6.9 mmoles) of bromomalononitrile, and
10 ml. of 50% aqueous ethanol. The reaction mixture was shaken until a
complete solution was obtained and set at room temperature for a month
(actually the time required for reaction may not have been this long).
Then the contents in the flask were transferred to a beaker, when the
product crystallized out. Recrystallization from ethanol yielded 0.52
g. (46.8%) of white crystals melting at 165—166°.

7- 5-Ethyl-5-n—butyl-1,1,2,2-tetracyanocyclopr0pane

 

One gram (6.17 mmoles) of 5-heptylidenemalononitrile and 1.5 g.
(10.5 mmoles) of bromomalononitrile were dissolved in 15 m1. of 90%
ethanol and set aside for 24 hrs. The precipitate which had formed was
filtered and, upon recrystallization from ethanol, yielded 1.52 g. (94.5%)
of white crystals, m.p. 119-12l°. IR (in nujol): 2275 (4.40), CN; 998
(10.02), cyclopropane ring (?). NMR spectrum (in acetone-d6) showed an
irresolvable complex multiplet from T'9.10 to T 8.00.

Anal. Calcd. for ClBHluNu: C, 69.00; H, 6.24; N, 24.76

Found: (5 68.97; IL 6.27; DL 24.85

 

8. 5,5-Dicyc10pr0pyl-l,l,2,2-tetracyanocyclopropane

DicyclOpropylmethylenemalononitrile (0.79 g., 5.0 mmoles) and bromo-

67

malononitrile (1.00 g., 6.9 mmoles) were dissolved in 20 m1. of 60%
aqueous ethanol. A precipitate formed in 10 mins. The first cr0p
was collected after a few hrs. and the second crOp after standing over-
night. Recrystallization from ethanol yielded 0.69 g. (62%) of fluffy
crystals, m.p. 185—187°. IR (in nujol): 2262 (4.42), CN; 1040 (9.62),
cyclOprOpane ring (?). NMR spectrum (in acetone-d6) showed two multiplets
at T 9.09 and 78.76.

Anal. Calcd. for ClBHlONu: C, 70.25; H, 4.54; N, 25.21

Found: C, 70.58; H, 4.47; N, 25.52

9. 5,5-Tetramethylene-l,1,2,2-tetracyanocyclopropane

 

One gram (7.57 mmoles) of cyc10pentylidenemalononitrile and 1.5 g.
(10.5 mmoles) of bromomalononitrile were dissolved in 14 ml. of 85% aqueous
ethanol. The precipitate which had formed in 50 mins. was filtered after
standing for a few hrs., and, upon recrystallization from acetone, yielded
0.78 g. (52.6%) of greyish white crystals, m.p. 240-245° dec. (lit. val.
(4), 259-240°). IR (in nujol): 2270 (4.41), CN; 968 (10.55), cyclopropane
ring (?). The NMR spectrum (in DMSO-d6 at about 110°) showed a broad peak
at 1 7.91 with reference to nitromethane (1 5.67) as an internal standard.

10. 5,5-Pentamethylene-l,1,2,2-tetracyanocyclopr0pane

 

Method I. One gram (6.84 mmoles) of cyclohexylidenemalononitrile and
1.7 g. (11.07 mmoles) of bromomalononitrile were dissolved in 15 ml. of
80% aqueous ethanol. A precipitate formed within 2 mins., which, after a
few hrs., was filtered and recrystallized from an ethanol-acetone mixture,
m.p. 177-179° (lit. val. (4), 180-181°), 1.4 g. (97.5%). IR (in nujol):
2260 (4.42), CN; 985 (10.17), cyclopropane ring (2). NMR Spectrum (in

acetone-d6) showed two multiplets at T 8.22 and 7 7.97.

r—- -'-=— -——--—---— . .

68

Method II. Oneehalf gram (5.10 mmoles) of cyclohexanone and 2.22 g.
(15.5 mmoles) of bromomalononitrile were dissolved in 20 m1. of ethanol
in a 50-ml. Erlenmeyer flask. The reaction mixture was kept overnight.
The white precipitate that had formed was filtered and recrystallized
from ethanol, m.p. 176—178°, 0.77 g. (71.9%).

11. l,1,2,2-Tetracyano—4—methylspiro[2.5]octane

One gram (6.25 mmoles) of 2-methylcyclohexylidenemalononitrile and
1.5 g. (10.5 mmoles) of bromomalononitrile were dissolved in 50 m1. of
ethanol. A precipitate formed in 16 hrs. The first crop was filtered
after 24 hrs. and the second crOp after standing for an additional 24 hrs.
Recrystallization from ethanol yielded 0.55 g. (59.5%) of white crystals,
m.p. 165-166°. IR (in nujol): 2275 (4.40), CN; 984 (10.16), cyclopropane
ring (?). NMR (in acetone-d6): 1‘8.45 (d, J=6.5 c.p.s.), ring methyl;
1‘ 8.22 (m), 05’ C69 and C7 methylene hydrogens; 1*7.95 (m), C8 methylene
hydrogens; 1 7.56 (m), methine hydrogen.

Anal. Calcd. for ClnggNu: C, 69.62; H, 5.59; N, 24.98

Found: C, 69.75; H, 5.29; N, 25.10

12. 5,5—Nonamethylene-1,1,2,2-tetracyanocyclopr0pane

Crude cyclodecylidenemalononitrile (0.98 g., 4.85 mmoles) and bromoma—
lononitrile (2.22 g., 15.5 mmoles) were dissolved in 12 ml. of ethanol. A
precipitate formed in 15 mins., which, after standing for a few hrs., was
filtered and recrystallized from an ethanol-acetone mixture, m.p. 214-216°,
0.45 g. (54.8%). IR (in nujol): 2280 (4.59), CN; 978 (10.25), cyclopropane
ring (?). The NMR spectrum (in DMSO-d6) showed two peaks at 1 8.45 and

’T 8.05 in an area ratio of 10 : 7.94.

69

Anal. Calcd. for ClnggNu: C, 72.15; H, 6.81; N, 21.04
Found: C, 72.14; H, 6.81; N, 21.11

15. 5,5-Undecamethylene-1,l,2,2-tetracyanocyclopr0pane

 

One-half gram (2.17 mmoles) of crude cyclododecylidenemalononitrile
and 2 g. (15.8 mmoles) of bromomalononitrile were dissolved in 10 ml. of
ethanol. A precipitate was formed in 1 hr., which, after standing for a
few hours, was worked up as described in section 12, m.p. l97-200°, 0.6
g. (94%). IR (in nujol): 2265 (4.42), CN; 971 (10.50), cyclOprOpane
ring (?). NMR spectrum (in acetone-d6) showed a large broad singlet at
1 8.55 and a multiplet at 1 8.09 in an area ratio of 14 : 8.02.

Anal. Calcd. for C18H22N4: C, 75.44; H, 7.553 N, 19.05

Found: C, 75.59; H, 7.56; N, 19.12

14. 5,5-Tetradecamethylene-l,1,2,2—tetracyanocyclopropane

 

Crude cyc1opentadecyclidenemalononitrile (obtained from 0.65 g. (2.89
mmoles) of cyclopentadecanone) and 2 g. (15.8 mmoles) of bromomalononitrile
were dissolved in 15 ml. of ethanol in a 50-ml. Erlenmeyer flask. The
solution was kept (at room temperature) for 6 days. Then 5 ml. of water
was added to the flask. An oily residue formed at the bottom which gradually
solidified upon scratching the flask wall. Filtration of the solid
followed by recrystallization from ethanol yielded 0.55 g. (56% based on
the amount of the ketone) of white crystals melting at lll-ll2°. IR (in
nujol): 2280 (4.59), CN; 975 (10.28), cyclopropane ring (?). The NMR
spectrum (in DMSO-dg) showed a large peak at T 8.64 and a smaller broad
peak at T 8.50 in an area ratio of 20 : 7.85.

Anal. Calcd. for CElH28N4: C, 74.96; H, 8.59; N, 16.65

Found: C, 74.86; H, 8.58; N, 16.61

70

D. Preparation of 5-Alky1-5-aryl-l,l,2,2-tetracyanocyc10propanes

 

1. 5-Methy1-5-pheny1-1,1,2,2{tetracyanocycloprOpane_

 

One gram (5.96 mmoles) of paphenylethylidenemalononitrile and 1.45 g.
(10.0 mmoles) of bromomalononitrile were dissolved in 40 m1. of ethanol.
Crystals formed in 1.5 hr., which, after standing overnight, were filtered.
To the filtrate was added an additional 1.45 g. of bromomalononitrile, and
thereafter crystals were collected every 24 hrs. for 5 days. The last
filtrate was added to 50 m1. of water with stirring and the resulting
crystals were filtered. The combined cr0ps, upon recrystallization from
an ethanol-acetone mixture, gave 1.2 g. (86.6%) of crystals, m.p. 249-252°
dec. (lit. val. (5), 225°). IR (in nujol): 2270 (4.41), CN; 988 (10.12),
cycloprOpane ring (?); 764 (15.09) and 700 (14.29), monosubstituted
benzene ring. NMR spectrum (in acetone-d6) showed a singlet for the ring
methyl at T 8.15 and a multiplet for the aromatic hydrogens, centered at
T 2.50.

2. 5-Methyl-5-g:chlor0phenyl-l,l,2,2-tetracyanocyclopr0pane

 

One gram (4.95 mmoles) of flamechlorOphenylethylidenemalononitrile and
1.5 g. (10.5 mmoles) of bromomalononitrile were dissblved in 50 m1. of
ethanol. A precipitate formed in 2 hrs., which, after standing for an
additional several hrs., was filtered and recrystallized from ethanol, m.p.
206—208°, 0.9 g. (68.1%). 1R (in nujol): 2266 (4.41), CN; 988 (10.12),
cycloprOpane ring (?); 794 (12.59) and 905 (11.05), 1,5-disubstituted benzene.
NMR Spectrum (in acetone-d6) showed a singlet for the cyclopropyl methyl
at T 7.95 and three multiplets for the aromatic hydrogens at 7 2.42, T 2.05,

mfl‘TL75

71

Anal. Calcd. for CluH7ClNu: C, 65.05; H, 2.65; Cl, 15.50; N, 21.01
Found: C, 62.26; H, 2.81; 01, 15.57; N, 20.77

5. 5-Methyl-5-p-methylpheny1-l,1,2,2-tetracyanocyc10propane

 

One gram (5.49 mmoles) of fifip?methylpheny1ethylidenemalononitrile and
1.45 g. (10.0 mmoles) of bromomalononitrile were dissolved in 50 ml. of

ethanol. A precipitate formed in 15 hrs., which, after standing for an
additional 5 hrs., was filtered. To the filtrate was added 0.9 g. of
bromomalononitrile and the second crOp was collected after standing
overnight. The combined product, upon recrystallization from ethanol,
gave 1.1 g. (81.2%) of white crystals, m.p. 222—224°. IR (in nujol):
2262 (4.42), CN; 985 (10.15), cyclOpropane ring (?); 817 (12.24), 1,4-
disubstituted benzene. NMR (in acetone-d6): 7'7.98 (S), cyclopropyl
methyl; 1 7.62 (s), phenyl methyl; 1 2.65 (d) and 7 2.18 (d) in an
A2B2 pattern, aromatic hydrogens.

Anal. Calcd. for ClSH10N4‘ C, 75.16; H, 4.09; N, 22.75

Found: C, 75.18; H, 4.05 N, 22.62

4. 5-Methyl-5-p-methoxypheny1-1,l,2,2-tetracyanocyc10pr0pane

 

One gram (5.04 mmoles) of P-pfmethoxyphenylethylidenemalononitrile

and 2.18 g. (15.0 mmoles) were dissolved in 50 ml. of ethanol. Crystals
formed in 20 hrs., which, after standing for an additional 24 hrs., were
filtered. To the filtrate was added an additional 2 g. of bromomalononitrile
and, after 24 hrs., the second crOp was collectedl' The combined product,

on recrystallization from ethanol, gave 0.46 g. (51.2% based on the amount

of the reacted P-pfmethoxyphenylethylidenemalononitrile) of white crystals,

9

m.p. 215—217°. IR (in nujol): 2500 (4.55), CN; 990 (10.10), cyclopropane

ring (?); 849 (11.78), 1,4-disubstituted benzene. NMR (in acetone-d6):

 

 

 

72

7.98 (s), cycloprOpyl methyl; 6.15 (s), methoxyl methyl; 2.95 (d) and
2.14 (d) in an AQBQ pattern, aromatic hydrogens.
Anal. Calcd. for ClSHlONAO: C, 68.69; H, 5.84; N, 21.56
Found: C, 68.52; H, 5.78; N, 21.58

Removal of solvent (using a rotary evaporator) from the final
filtrate resulted in a solid residue, which, on recrystallization from
ethanol, gave 0.52 g. (1.61 mmoles) of unchanged pipamethoxyphenylethyli-
denemalononitrile.

5. 5-Methy1-5-[3-naphthy1-l,1,2,2—tetracyanocyc10propane

One gram (4.58 moles) of p-2 -naphthy1ethylidenemalononitrile and
2.18 g. (15.0 mmoles)of bromomalononitrile were dissolved in 50 m1. of
ethanol. The precipitate that had formed in 1 hr., was filtered after
standing for 24 hrs., and upon recrystallization from an ethanol-acetone
mixture, yielded 0.55 g. (54.8% based on the amount of the reacted B-EB-
naphthylethylidenemalononitrile) of white crystals, m.p. 255-260°. IR (in
nujol): 2280 (4.58), CN; 985 (10.15), cycloprOpane ring (?); 870 (11.49),
816 (12.25), and 749 (15.55), p-naphthyl ring. NMR (in acetone-d6): T 7.84
(s), cycloprOpyl methyl; 7 2.57 (m), 7 2.02 (m), and 1 1.45 (s), aromatic
hydrogens.

Anal. Calcd. for Cl8HlON4: C, 76.58; H, 5.57; N, 19.85

Found: C, 75.05; H, 5.54; N, 19.50

Removal of solvent from the filtrate followed by recrystallization
(from ethanol) of the resulting residue gave 0.25 g. (1.15 mmole) of
recovered (3~2anaphthylethylidenemalononitrile.

6. 5-Ethyl-5-phenyl-l,l,2,2-tetracyanocyclopropane

 

‘fl-PhenylprOpylidenemalononitrile (0.91 g., 5.0 mmoles) and bromomalono-

nitrile (1.20 g., 8.0 mmoles) were dissolved in 20 ml. of ethanol and allowed

75

to stand for a few days. The reaction mixture was then poured into a
50-ml. beaker and 0001ed in an ice bath. Crystals that had formed

were filtered and recrystallized from ethanol, m.p. 225-227°. 0.22 g.
(17.8%). IR (in nujol); 2275 (4 40), CN; 990 (10.10), cyclopropane
ring (?); 774 (12.94) and 702 (14.25), the monosubstituted benzene.

NMR (in acetone-d6): T 8.89 (t), CH5CH2; 1 7.68 (q), CHBCHQ; 1 2.21 (m),
aromatic hydrogens. Area ratio: 5.08 : 2.18 : 5.00.

Anal. Calcd. for C Nu: C, 75.16; H, 4.09; N, 22.75

lSHlO
Found: C, 75.08; H, 4.11; N, 22.68

7. Spiro[2,2,5,5-tetracyanocyclopropane-1,l‘—tetralrm

 

Six-tenths (5.1 mmoles) of 2,5-benzocyclohexylidenemalononitrile
was placed in a 50-ml. Erlenmeyer flask containing 50 m1. of ethanol and
5 ml. of water. The flask was gently warmed on a hot plate until a com-
plete solution was obtained and then the solution was allowed to cool to
room temperature. One and one-half gram (10.5 mmoles) of bromomalononitrile
was dissolved in this solution. Crystals that had formed in 6-7 hrs. were
filtered after 12 hrs. To the filtrate was added 2 m1. of water and the
second crOp was collected after standing for an additional several hrs.
The combined product was recrystallized from ethanol, m.p. 167—170., 0.29
g. (54.2% based on the amount of the reacted 2,5-benzocyclohexylidenemalono-
nitrile). IR (in nujol); 2265 (4.41), CN; 968 (10.55), cyclopropane ring
(?); 745 (15.46), benzene ring.

Anal. Calcd. for C16H10N4: C, 74.40; H, 5.90; N, 21.69

Found: C, 74.51; H, 5.98; N, 21.55

74

The final filtrate, after removal of solvent followed by recrystalli-
zation of the resulting solid, gave 0.2 g. (1.5 mmoles) of unchanged 2,5-
benzocyclohexylidenemalononitrile.

It seemed very important to use a solvent made up of ethanol and
water according to the above pr0portion and to run the reaction at room
temperature. An elevated reaction temperature or use of solvent richer
in ethanol (90% aqueous ethanol, for example) was found to cause formation
of brominated compound described on page 98.

8. 5-Methy1-5-ok-thieny1-l,l,2,2-tetracyanocyclopropane

One gram (5.7 mmoles) of fl—EZ—thieny1ethylidenemalononitrile and lg.
(6.9 mmoles) of bromomalononitrile were dissolved in 50 m1. of ethanol and
gently warmed on a steam bath for severalnfins. Precipitate that had formed
in 50 mins. was filtered and recrystallized from ethanol, m.p. 207-210°
J dec., 0.5 g. (22.1%). IR (in nujol): 2275 (4.40), CN; 985 (10.15),
cyclOprOpane ring (?); 717 (15.98), thiOphene ring. NMR (in DMSO-d6):

7 8.16 (s), cyclOprOpyl methyl; 1 2.87 (m) and 7 2.25 (m), aromatic
hydrogens.

Anal. Calcd. for Cl2H6NuS: C, 60.49; H, 2.54; N, 25.52; S, 15.46

Found: C, 60.47; H, 2.76; N, 25.52; S, 15.51

E. Preparation of Some Dimers of Alkylidenemalononitriles

 

1. IsOpropylidenemalononitrile Dimer

 

Method I. When isoprOpylidenemalononitrile was allowed to stand at
room temperature for a long time (6-10 months), a precipitate formed at the

bottom of the container. Recrystallization of this precipitate from ethanol

75

gave pale yellow crystals of isOprOpylidenemalononitrile dimer, m.p.
171-174° (lit. val. (64) 168-170°). IR (in nujol): 5440 (2.91), 5540
(2.99), 5250 (5.10), NH; 2215 (4.51), CN; 1465 (6.08), 1614 (6.20),
1578 (6.54), C=C and/or C=N. UV (in ethanol): 218.5 (11,540), 240.0
(5,687) (sh ), 510-0 (5,770). NMR (in DMSO-d6)3 7'8.74 (d), gem.
dimethyls; T 8.15 (d., J=1.7 c.p.s.), allylic methyl; 1 4.88 (q., J=1.6
c.p.s.); r 2.57 (s), NHQ: Area ratio was 6 : 2.62 : 0.75 : 1.69 in the
order described.

The condensation reaction of acetone and malononitrile gives a
small amount of isopropylidenemalononitrile dimer as well as the monomer.
Usually the dimer is obtained by recrystallizing the solid residue remaining
in the bottom of the distillation flask after the monomer has been distilled
out.

Method II. One gram (9.45 mmoles) of is0pr0pylidenemalononitrile

 

was dissolved in 7 ml. of ethanol contained in a 50-ml. Erlenmeyer flask.
To this solution was added 5 drOps of piperidine. The mixture was kept at
room temperature for 1.5 hr. Then the contents in the flask were poured
into 20 m1. of water contained in a 50-ml. beaker. A precipitate formed
immediately. After one-half hr., crystals were filtered and recrystallized
from ethanol, m.p. 168-171°, 0.75 g. (75%). The m.m.p. with the compound
-prepared by Method I did not depress.

2. 2—Butylidenemalononitrile Dimer

 

Method I. In a 100-ml. round-bottomed flask fitted with a Barrett
distilling receiver and a water condenser, were placed 14.4 g. (0.2 mole)
of 2-butanone, 15.2 g. (0.2 mole) of malononitrile, 5 g. of ammonium acetate,

6 ml. of glacial acetic acid, and 50 m1. of benzene. The mixture was

76

refluxed for 4 hrs. Then reflux was discontinued and the reaction
mixture was allowed to cool to room temperature. An additional 20 m1.
of benzene was added to the flask and the entire reaction mixture was
washed with three 50-ml. portions of water. The organic layer was dried
over 20 g. of anhydrous sodium sulfate for several hrs. The drying agent
was removed by filtration and the solvent was stripped with the aid of a
water aspirator. Further fractionation gave 9.00 g. (57.5%) of pale
yellow liquid of 2-butylidenemalononitrile, b4 82° (lit. val. (65) b8
102°), nb9 1.4707.

After the 2-butylidenemalononitrile was fractionated, there was

formed a gummy residue in the distillation flask. This was purified by
twice recryStallizing it from ethanol. There was obtained 2.70 g. (11.2%)
of 2-butylidenemalononitrile dimer, m.p. 159-164° (lit. val. (44) 167—
169°). IR (in nujol): 5450 (2.98), 5560 (2.98), 5260 (5.07), NH; 2220
(4.50), CN; 1652 (6.05), 1595 (6.28), C=C and/or C=N. UV (in ethanol):
501 (15,020), 245 (9,661), 217 (6,248), 215 (sh). NMR (in DMSO-d6):r9.07

(t), CH CH 1‘8.45 (q), CHECH ' ’T8.80 (3), ring methyl; 7 8.27 (d),

33 5’
allylic methyl; 1‘7.66 (s), ring methylene hydrogens; T 4.44 (q., J=7.0

2

c.p.s.), olefinic hydrogen; 7-2.95 (s), NH2. Area ratio: 'r9.07—8.27 :
1 7.66 : 1 4.44 : T 2.95 = 11 : 1.95 : 0.76 : 1.75.
Anal. Calcd. for Clqu6Nh: C, 69.97; H, 6.71; N, 25 52
Found: C, 70.01; H, 6.74; N, 25.25

Method II. In a 50-ml. Erlenmeyer flask were placed 2 g. (16.6

 

mmoles) of 2-butylidenemalononitrile and 10 ml. of ethanol. While stirring
the solution using a magnetic stirrer, 10 drOps of piperidine was added to

the reaction mixture. After 1.5 hr., the stirring was discontinued and the

77

contents in the flask was treated in the same manner described in
Method II of preparation of isOpropylidenemalononitrile dimer.
Recrystallization from ethanol yielded 1.55 g. (67.5%) of light yellow
crystals melting at l58-162°. The m.m.p. with the compound prepared by
Method I did not depress.

5. Cyclopentylidenemalononitrile Dimer

Method I,Cyclopentanonec(20 gs, 0.250 mole), 14.8 g. (0.224 mole) of
malononitrile, 5.0 g. of ammonium acetate, 6.0 ml. of glacial acetic acid,
and 50 ml. of benzene were treated in the manner described in Method I
for preparation of 2-butylidenemalononitri1e dimer. While cyclopentylidene-
malononitrile was distilled, there began to form a precipitate in the
flask*. Distillation was discontinued and the contents of the flask were
transferred to a 100-ml. beaker. The beaker was kept at room temperature
until crystallization of the reaction mixture appeared to be completed
(about an hr.). Filtration followed by recrystallization from ethanol
yielded 11.72 g. (59.2%) of lemon yellow crystals, m.p. 187-190°. IR (in
nujol): 5440 (2791), 5570 (2.97), 5260 (5.07), NH; 2240 (4.46), CN; 1650
(6.06), 1589 (6.29), C=C and/or C=N. UV (in ethanol): 508 (12,226), 241
(8,681), 224 (6,659), 215 (sh.). NMR (in DMSO- -d6): ‘T 2. 72 (s),m 47-4. 52

(d, J=1.8 c.p.s.), olefinic hydrogen; 7‘8.25 (m), 1 7.61 (m), 1‘6.97 (m),

 

* No cyclOpentylidenemalononitrile monomer was obtained in this particular
experiment. However, in one of previous experiments, the monomer was
obtained in a yield of 22.1%, b4 115-114°, n35 1.4995. IR (in CClu):

2250 (4.44),, CN; 1618 (6.18), C=C. UV (in ethanol). 258 (10, 885). The
NMR spectrum (neat) showed two multiplets at 7 8.11 (C - and Cu-hydrogens)
and 7'7. 24 (C2— and C -hydrogens) in an area ratio of 88 . The
residue obtained in t is latter case was a tar which could not be
identified.

78

rest of hydrogens in the molecule. Area ratio: 7 2.72 : 7 4.52
7‘6.97 — 8.25 = 1.89 : 0.97 : 15.

Anal. Calcd. for C16H16N4: C, 72.70; H, 6.10; N, 21.20

Found: C, 72.82; H, 6.05; N, 21.50

Method II. Four and six-tenths grams (0.057 mole) of cyclopentanone,
5.50 g. (0.050 mole) of malononitrile, 0.5 g. of ammonium acetate, 1.0 ml.
of glacial acetic acid, and 10 ml. of benzene were treated in the manner
described in Method I to obtain a dry benzene solution of the cyclo-
pentylidenemalononitrile monomer (25 m1.). The benzene solution was placed
in a 250-ml. Erlenmeyer flask and 0.6 ml. of piperidine and 50 m1. of
water were added. The solution was stirred using magnetic stirrer for 1
hr. A precipitate formed. After a few hrs., the precipitate was filtered
and recrystallized from ethanol, m.p. l87-189.5°, 4.06 g. (58.5%). The

m.m.p. with the compound prepared by Method I did not depress.

F. Preparation of 5-Aryl-l,1,2,2-tetracyanocyclopropanes

 

An ethanolic solution of arylidenemalononitrile was mixed with an
ethanolic solution of equimolar or excess (twice or less) amount of bromo-
malononitrile. A precipitate formed in sometime between a few mins. and
a few hrs. depending on the aryl group. The reaction mixture was allowed
to stand for several hrs. after the first precipitate appeared. Then the
precipitate was filtered and purified by recrystallizing from ethanol or
more often from an ethanol-acetone mixture. Compounds thus prepared are
presented in Table 4 and some of their NMR spectral data appear in Table 5

(page 26 and 29, respectively).

79

A typical preparation follows: To 1 g. (6.5 mmoles) of benzylidene-
malononitrile dissolved in 10 m1. of ethanol contained in a 50-ml.
Erlenmeyer flask was added 1.16 g. (8.0 mmoles) of bromomalononitrile
dissolved in 10 ml. of ethanol. A.precipitate formed within a few mins.
After 5 hrs, the precipitate was filtered. Recrystallization from an
ethanol-acetone mixture yielded 1.55 g. (91.7%) of crystals, m.p. 227-250°

dec.

G. Preparation of 5,5—Dialky1-2-carbethoxy-l,1,2-tricyanocyclopropanes

 

l. 5,5-Dimethy1-2-carbethoxy-l,1,2-tricyanocyc10propane

 

Method I. One-half gram (5.26 mmoles) of ethyl is0propylidenecyano-
acetate and 1.0 g. (6.90 mmoles) of bromomalononitrile were dissolved in
12 ml. of 85% aqueous ethanol contained in a 50-ml. Erlenmeyer flask. The
solution was kept at room temperature for 1 hr., when the precipitate
started forming. The reaction mixture was allowed to stand for an additional
few hrs. Then the precipitate was filtered. Recrystallization from ethanol
gave 0.52 g. (75.5%) of white crystals, m.p. 141-145° (lit. val. (14) 155°).
IR (in nujol): 2270 (4 41), CN; 1747 (5.72), 00; 1280 (7.81), 1248 (8.01),
C-O-C; 980 (10.20), cyclopropane ring (?). The NMR Spectrum (in acetone-
d6) showed COOCEHS at 1’8.62 (t) and 1'5.62 (q) and two ring methyls at

T 8.51 (s, cis to COOC2H5) and 7 8.21 (s, trans to 00002H ).

 

When the reaction mixture was warmed on a steam bath, the product
formed in 10 mins.

Method II. Two grams (18.9 mmoles) of isopropylidenemalononitrile and
5.84 g. (20 mmoles) of ethyl bromocyanoacetate were dissolved in 25 m1. of

50% aqueous ethanol. The mixture was'allowed to stand at room temperature.

8O

Crystals formed after 5 hrs. The product was filtered after 24 hrs.
Recrystallization from an. ethanol-acetone mixture yielded 1.17 g.
(28.6%) of white crystals, m.p. l57-l59°. The m.m.p. with the compound
prepared by Method I did not depress.

2-.5-Methyl-5~ethyl-2-carbethoxy-l,l,2-tricyanocyclopropane

 

One gram (5.98 mmoles) of ethyl 2-butylidenecyanoacetate and 1.01 g.
(7.00 mmoles) of bromomalononitrile were dissolved in 12 m1. of 80%
aqueous ethanol contained in a 50-ml. Erlenmeyer flask and kept at room
temperature for about 20 hrs. The contents of the flask were transferred
to a 50-ml. round-bottomed flask. Two millilitres of water was added and
the mixture was refluxed for a few hrs. The contents were then poured
into a 50-ml. beaker and cooled in an ice bath until a precipitate formed.
Filtration followed by recrystallization from ethanol yielded 0.88 g.
(65.7%) of white crystals, m.p. 87-89° (lit. val. (14) 89°). IR (in nujol):
2260 (4.42), CN; 1745 (5.75), CO; 1275 (7.84), C-O-C; 978 (10.25), cyclo-
propane ring (?). NMR (in acetone-d6): 1 8.80 (m, 4 triplets), CHQCHB and
CCCCHQCH5; 1 8.59 (s) and T 8.26 (s), ring methyl gig. and 4%, respect- _,
ively,to COOCH2CH3; 1'7.94 (m), CH2CH5; 1'5.64 (q, 2 quartets when expanded),
COOCHQCHB. An attempt to prepare this compound from 2-butylidenemalono-
nitrile and ethyl bromocyanoacetate was unsuccessful.

5. 5-Methy1-5-p-propy1-2-carbethoxy-l,1,2-tricyanqu910pr9pane

One gram (5.52 mmoles) of ethyl 2-pentylidenecyanoacetate and 1.02 g.
(7.04 mmoles)of bromomalononitrile were mixed in 50% aqueous ethanol in
50-ml. Erlenmeyer flask and warmed on a steam bath for several mins. The

dark brown reaction mixture was kept at room temperature for about a month.

81

ter this time, the solvent was removed with aid of a rotary evaporator.
To the concentrated residue thus obtained was added 10 m1. of water and
the mixture was cooled in an ice bath for half an hr. A precipitate was
obtained and recrystallized from ethanol, m.p. 65-65°, 0.5 g. (57%). IR
(in nujol): 2260 (4.42), CN; 1747 (5.72), 00; 1269 (7.88), C-O—C; 979

(10.21) cyclopropane ring (?). NMR (in acetone-d6): 1‘8.98 (t), CH20H20H5;

'T8.58 (s) and 1‘8.25 (5), ring methyl gi§ and trans, respectively, to

C00C2H 1'8.64 (t) and q~5.62 (q), COOCQH. The methylene hydrogens in

5’ 5'
the nfprOpyl group appeared in a multiplet above T‘8.05.
Anal. Calcd. for ClBHlSNBOE: C, 65.66; H, 6.16; N, 17.15
Found: C, 65.65; H, 6.05; N, 17.51
An attempt to prepare this compound from 2-pentylidenemalononitrile

and ethyl bromocyanoacetate was not successful.

4. 5-Methy1—5-is0pr0py1-2-carbethoxy-1,1,2-tricyanocyclopr0pane

 

Ethyl 5-methy1—2-butylidenecyanoacetate (2.18 g., 12 mmoles) and bromo-
malononitrile (2.18 g., 15 mmoles) were dissolved in 20 m1. of aqueous
ethanol (about 80%) contained in a 100-ml. round-bottomed flask. The mix-
ture was refluxed for 5 hrs. and then set aside at room temperature for 2
weeks. Work-up of the reaction mixture yielded 0.19 g. (6.46%) of crystals,
m.p. 151—154° (from ethanol). IR (in nujol): 2275 (4.40), CN; 1745 (5.75),
00; 1256 (7.96), C-O-C; 1017 (9.85), cyclopropane ring (?). NMR (in acetone-
d6%'1899(d)wfl.7879(d% flfiqghy’r865(t)mflw1562(qb
00002H5; 1 8.55 (s), ring methyl; 7-7.76 (m), CH(CH5)2.

Anal. Calcd. for C H N 0 - C, 65.66, H, 6.16; N, 17.15

15 15 5 2°
Found: C, 65.14; H, 6.01; N, 17.51

82

An attempt to prepare this compound from 5-methyl-2-butylidenemalono—
nitrile and ethyl bromocyanoacetate was not successful.

5. 5,5-Diethy1-2-carbethoxy-1,1,2-tricyanocyclopr0pane

Two grams (11 mmoles) of ethyl 5-pentylidenecyanoacetate and 2.2 g.
(15 mmoles) of bromomalononitrile were dissolved in 50 m1. of ethanol in a
lOO-ml. round—bottomed flask. The mixture was set aside at room temperature
for 2 days and then refluxed for several hrs. Then the contents were
cooled to room temperature. The reaction mixture was poured into a 100—ml.
beaker and further cooled in an ice bath. Crystals that formed were
filtered and recrystallized from ethanol, m.p. 95-95°, 0.95 g. (55.2%).
IR (in nujol): 2255 (4.45), CN; 1741 (5.74), CO; 1272 (7.86), C-o-C; 975
(10.26), cycloprOpane ring (?). NMR spectrum (in acetone-d6) showed three;
triplets at T 8.95, 1'8.80 and T 8.65 and three quartets at T 8.00, T 7.97
and 15.62 with an area ratio of triplets vs. quartets being 6.00 : 4.22.

Anal. Calcd. for 015H15N502: C, 65.66, H, 6.16; N, 17 15

Found: C, 65.55; H, 6.09; N, 17.04

An attempt to prepare this compound from 5-pentylidenemalononitrile

and ethyl bromocyanoacetate was not successful

/

o. 5,5-Tetramethy1ene-2jcarbethoxy-l,1,2-tricyanocyclopropane

 

One and eight-tenths gram '(10 mmoles) of ethyl cyclopentylidenecyano—
acetate and 1.9 g. (15 mmoles) of bromomalononitrile were dissolved in 15
m1. of ethanol contained in a 50-ml. Erlenmeyer flask and kept at room
temperature. After a few hrs., 2 ml. of water was added to the flask, when
there was formed a brown oily layer at the bottom of the flask. After 2

days, the oil solidified. Recrystallization of the solid mass from ethanol

85

yielded 0.55 g. (14.5%) of needles (pale gray), m.p. 156-159°. IR
(in nujol): 2260 (4.42), CN; 1745 (5.74), 00; 1282 (7.80), C-O-C; 970
(10.54), cycloprOpane ring (?). NMR spectrum (in DMSO-d6) showed the
COOCQHS at 1'8.66 (t) and 7 5.72 (q) and the ring hydrogens at T 8.02
(m, consisted of 5 main peaks).

Anal. Calcd. for 013H15N§02z C, 64.19; H, 5.59; N, 17.28

Found C, 64.16; H, 5.51; N, 17.55
An attempt to prepare this compound from cyclopentylidenemalononitrile

and ethyl bromocyanoacetate was not successful.

7. 5,5-Pentamethy1ene-2—carbethoxy-1,1,2-tricyanocyclopr0pane

 

Method 1. One-half gram (2.59 mmoles) of ethyl cyclohexylidenecyano-
acetate, 0.75 g. (5.00 mmoles) of bromomalononitrile, and 10 m1. of 80%
aqueous ethanol were placed in a 50-ml. Erlenmeyer flask. The flask was
gently shaken until the solution was completed and kept at room temperature.
One hr. later, a precipitate appeared. After 4 hrs., the precipitate was
filtered and recrystallized from ethanol, m.p. 129-151°, 0.65 g. (97.5%).
IR (in nujol): 2275 (4 40), CN; 1745 (5.75), 00; 1274 (7.85), C-0-C; 981
(10.19), cycloprOpane ring (?). NMR spectrum (in acetone—d6) showed ethyl
hydrogens at'T 8.64 (t) and 1 5.61 (q) and the ring hydrogens at T 8.55 (m)
and 1 7.96 (m).

Anal. Calcd. for CluH C, 65.55; H, 5.88; N, 16.55

15N502:
Found: C, 65.50; H, 5.80; N, 16.58
Method II. In a 50-ml. round—bottomed flask were placed 1 g. (6.85
mmoles) of cyc1ohexylidenemalononitrile, 1.54 g. (8.0 mmoles) of ethyl

bromocyanoacetate, and 15 m1. of 50% aqueous ethanol and the mixture was

84

refluxed overnight. Reflux was discontinued and the reaction mixture

was kept at room temperature for a few hrs. During this time a precipitate
formed in the flask. Filtration followed by recrystallization from ethanol
gave 0.42 g. (25.8%) of white crystals, m.p. 15l-155°. The m.m.p. with

the compound prepared by Method I did not depress.

H. Preparation of 5-Ary1-2-carbethoxy-l,1,2-tricyanocyclopr0panes

1. 5-Phenyl-2-carbethoxy-l,1,2-tricyanocyclopr0pane

Method I. One gram (4.98 mmoles) of ethyl benzylidenecyanoacetate
and 0.74 g. (5.1 mmoles) of bromomalononitrile were dissolved in 20 m1. of
50% aqueous ethanol. The mixture was refluxed for several hrs., during
which time a precipitate formed in the flask. Filtration followed by
recrystallization from ethanol yielded 1.0 g. (76.7%) of white crystals,
m.p. 124.5—126°. IR (in nujol): 2260 (4.45), CN; 1754 (5.77), CO; 1288
(7.81), 1255 (8.11), C-O-C; 740 (15.51), 700 (14.28), monosubstituted
benzene; 1015 (9.85), cyclopropane ring (?). NMR (in acetone): 7 8.61 (t)
and 1 5.61 (q), COOCQHS; 7-5.74 (s), H, cycloprOpyl; 1 2.52 (m), H,
aromatic.

Anal. Calcd. for ClSHllNEOQ: C, 67.92; H, 4.18; N, 15.84

Found: C, 67.82; H, 4.11; N, 16.00

Method II. One gram (6.5 mmoles) of benzylidenemalononitrile and 1.25
g. (6.8 mmoles) of ethyltuomocyanoacetate were placed in a 50-ml. round-
bottomed flask containing 27 m1. of 60% aqueous ethanol. The contents in

the flask were refluxed for 5 hrs. An oily layer which formed at the

bottom of the container solidified on being kept in a refrigerator for a

85

few hrs. The solid mass was filtered and recrystallized from ethanol,
0.5 g. (29%), m.p. 125-125°. The m.m.p. with the compound prepared by
Method I did not depress.
2. 5-p—Methoxyphenyl-2—carbethoxy-l,1,2—tricyanocyc10pr0pane
Method I. In a l25-ml. Erlenmeyer flask were placed 1.16 g.
(5 mmoles) of ethyl anisylidenecyanoacetate, 1.5 g. (10.5 mmoles) of
bromomalononitrile, and 10 m1. of ethanol. The mixture was shaken
until a complete solution was obtained, then kept overnight at room
temperature. The precipitate that had formed was filtered and treated
with Norit A. Recrystallization from ethanol gave 0.55 g. (56.2%) of
crystals, m.p. 95-97°. IR (in nujol): 2275 (4.40), CN; 1745 (5.74),
00; 1267 (7.89), 1180 (8.47), C-O-C; 990 (10.10), cycloprOpane ring (?);
820 (12.19), 1,4-disubstituted benzene. NMR (in acetone): 1'8.65 (t)
and T 5.58 (q), COOC2H53'T6.19 (s), CH50;1'5.79 (S), H, cycloprOpyl;
7 5.00 (d) and 1-2.40 (d) (A2B2), H, aromatic.

Anal. Calcd. for Cl6H C, 65.08; H, 4.44; N, 14.25

15N503:
Found: C, 64.74; H, 4.55; N, 14.29

Method II. In a 50-ml. Erlenmeyer flask were placed 0.92 g. (5
mmoles) of anisylidenemalononitrile, 2.00 g. (10 mmoles) of ethyl bromo-
cyanoacetate, and 50 m1. of ethanol. The mixture was shaken until a
complete solution was obtained and set at room temperature for 2 days,
after which time crystals started forming. After a week, the crystals
were filtered and recrystallized from ethanol, m.p. 105-106°, 0.89 g.

(60.5%). The m.m.p. with the compound prepared by Method I did not

depress.

l

86

1. Preparation of 2-carboxamido-l,l,2-tricyanocyclopr0panes

 

 

1. 5,5-Pentamethylene-2-carboxamido-1,1,2—tricyanocyclopropane

In a 50-ml. Erlenmeyer flask was placed 1.6 g. (10 mmoles) of
cyclohexylidenecyanoacetamide, 4.55 g. (50 mmoles) of bromomalononitrile,
and 25 m1. of 80% aqueous ethanol. The flask was shaken until solution
was completed and kept at room temperature for 2 days. Then the contents
in the flask warztransferred into a lOO-ml. beaker and 20 m1. of water
was added. The beaker was placed in an ice bath for a few hrs. (5-4
hrs.). The precipitate which had formed during this time period was
filtered and washed with water. Recrystallization from ethanol gave
0.15 g. (7.15% based on 10 mmoles of cyclohexylidenecyanoacetamide) of
5,5-pentamethylene-l,1,2,2-tetracyanocyclopr0pane, m.p. 176-178°.

The filtrate was further kept in an ice bath for several hrs.
(6—7 hrs.) occasionally scratching the wall of the beaker. The pre-
cipitate that formed was filtered and recrystallized by dissolving in
a minimum amount of ethanol at room temperature and then filtering
followed by cooling in an ice bath. The crystals thus obtained were
crude, beginning to melt at 140° and finally decomposing at 180°, 1.40
g. (61.5%). IR (in nujol): 5450 (2 91), 5560 (2.98), 5200 (5.15), NH;
1700 (5.88), 00: 1615 (6.19), amide II band; 970 (10.51), cyc10pr0pane
ring (?). NMR Spectrum (in acetone-d6) showed the cyclohexane ring
hydrogens in 2 multiplets at 7 8.27 and 1 8.07 and the amide hydrogens
in 2 broad peaks at 7 2.58 and 7 2.02 in an area ratio of 10 : 1.75.
This compound could not be recrystallized in the ordinary manner without

undergoing a chemical change.

87

2. 5-Pheny1-2-carboxamido-l,1,2-tricyanocyclopropane

 

One and seven-tenths gram (10 mmoles) of benzylidenecyanoacetamide
was dissolved in 50 m1. of ethanol contained in a l25-ml. Erlenmeyer
flask. To this solution was added 5 g. (20 mmoles) of bromomalononitrile
and the flask was shaken until the solution was completed. The solution
was kept at room temperature. Within one-half hr., a precipitate
appeared. After a few (5-4) hrs., the first crOp of 1.75 g. was col-
lected. After overnight the second crop of 0.55 g. was collected, 2.1
g. (89%). Crude crystals decomposed sharply at 185° (m.p. for this
compound was taken by heating the apparatus at a rate of 20-50°/min.).

TB (in nujol): 5500 (2.86), 5500 (5.05), 5200 (5.15), NH; 2270 (4.41),
CN; 1710 (5.85), 1687 (5.95), 00; 1605 (6.25), amide 1: band; 740 (15.51),
700 (14.29), monosubstituted benzene. NMR (in DMSO-dg): 1 5.65 (s), H,

cyclopropyl; T 2.57 (m), H, aromatic; 7 1.47, CONH

_2. flfiscmmmmd

could not be recrystallized in the ordinary manner without undergoing a
chemical change.

5. 5-prhlor0phenyl-2-carboxamido-l,1,2-tricyanocyclopr0pane

 

One and nine-tenths gram (9.18 mmoles) of pfchlorobenzylidenecyano-
acetamide was dissolved in 250 m1. of warm (40-45°) ethanol contained in
a 500-ml. Erlenmeyer flask. To this solution was added 5 g. (20 mmoles)
of bromomalononitrile and the mixture was Shaken until solution was nearly
complete. The resultant solution was filtered to remove undissolved

impurities and allowed to stand at room temperature for 5 hrs. Then

ethanol was evaporated using a rotary evaporator below 40° until the

residual volume decreased to about 70 ml. The residue was transferred

88

into a lOO-ml. beaker and cooled in an ice bath. Crystals that formed
were filtered and washed thoroughly with water. The filtrate was
further evaporated and treated in the same manner to obtain the second
crop. White crystalline product that had not been recrystallized
(this compound could not be recrystallized in the ordinary manner
without undergoing a chemical change) sharply decomposed at 202° (m.p.
measured at a heating rate of 20-50°/min.), 2.2 g. (88.5%). IR (in
nujol): 5455 (2.91), 5365 (2.97), 5295 (5.05). 5225 (5.10). NH; 2280
(4.58), CN: 1704 (5.87), 1684 (5.94), C0; 1605 (6.24), amide 11 band;
812 (12.51), 1,4-disubstituted benzene. NMR (in DMSO-d6): 1‘5.42 (s),
H, cyclopropyl; 1 2.50 (d) and 7-2.55 (d) (A2B2), H, aromatic; 7 1.48
(s, broad), CONH2.

Anal. Calcd. for C ClNMO: C, 57.68; H, 2.61

15H?
Found: C, 57.71; H, 2.77

J. Preparation of l,5-Dicyano-2-imino-5-aza-4-ketobicyclo[5.1.0]hexanes

 

1. l,5—Dicyano-2-imino-5-aza-4-keto-6,6-pentamethylenebicyclo[5.l.0]
hexane

 

One-half gram (2.19 mmoles) of 5,5-pentamethylene—2-carboxamido-1,1,2-
tricyanocyclopropane was dissolved in a few (5-4) ml. of boiling methanol.
When the solution was completed, 1 ml. of water was added and the resultant
aqueous solution was further heated on a hot plate for a min. The solution
was filtered and then Cooled in an ice bath. Crystals thus obtained were
recrystallized from ethanol, m.p. 225-228° dec., 0.47 g. (94%). IR (in
nujol): 5580 (2.96), NH; 2260 (4.42), CN; 1757 (5.76), 00; 1652 (6.05),

1567 (6.58), (both strong and broad), C=N and/or NH; 1007 (9.95), cyclo-

89

pr0pane ring (?). NMR spectrum taken in a dimethyl sulfoxide-d6 solution
at a probe temperature of about 100° showed the cyclohexane ring hydrogens
in a large peak at 7 8.58 and a small adjoining peak at 7'8.10. The imino
and amido hydrogens appeared in a weak broad peak at 7 1.22 (the chemical
shift of this peak was found to increase, the longer the sample was

kept in the probe). Area ratio: ring hydrogens : N-hydrogens, 10 : 1.8.
A half drop of ordinary dimethyl sulfoxide (7‘7.58) was used as an
internal reference.

Anal. Calcd. for C NuO: C, 65.14; H, 5.50; N, 24.55

12H12
Found: C, 65.25; H, 5.25; N, 24.51

2. 1,5-Dicyano-2-imino—5-aza-4-keto-6-pheny1bicyclo[5.1.0]hexane

 

In a lOO-ml. beaker were placed 1.92 g. (8.12 mmoles) of 5-pheny1—
2-carboxamido-1,l,2-tricyanocyclopropane and 50 ml. of ethanol. The
mixture in which a large part of solid remained undissolved was heated
on a hot plate under constant stirring for about 20 mins. The amount of
crystals in the container increased during the process of heating. The
reaction mixture was allowed to cool to room temperature and crystals
were filtered. For recrystallization, the crystals were dissolved in a
minimum amount of warm dimethyl sulfoxide and filtered. To the filtrate
water was slowly added until the solution was turbid. 0n cooling in an
ice bath, 1.85 g. (96.5%) of powdery crystals were produced, m.p. 257° dec.
(A heating rate of 20—50°/min. was employed). IR (in nujol): 5250 (5.08),
NH; 2280 (4.58), CN; 1717 (5.82), 00; 1680 (5.95), 1555 (6.44) (both
broad and strong), C=N and/or NH; 771 (12.97), 697 (14.55), monosubstituted

benzene. NMR Spectrum (in DMSO-d6) showed cycloprOpyl hydrogen at 7 5.89

90

(s), aromatic hydrogens at T 2.49 (s), and N-hydrogens at 7'0.41 (S,
broad) in an area ratio of 1.06 : 5 : 1.86.
Anal. Calcd. for ClBHBNAO: C, 66.10; H, 5.41; N, 25.72
Found: C, 65.86; H, 5.58; N, 25.68

5. 1,5-Dicyano-2-imino-5-aza-4-keto-6-p-chlor0phenylbicyclo
[5.1.0]hexane I

 

One and one-tenth gram (4.06 mmoles) of 5-pfchlor0pheny1-2-carboxamido-
1,1,2-tricyanocyclopr0pane was mixed with 40 ml. of ethanol contained in a
100-ml. beaker. The mixture was vigorously heated on a hot plate under
constant stirring. At the beginning, there was obtained a clear solution,
which, a moment later, became turbid with forming precipitate. After
20 mins. of heating, the reaction mixture was allowed to cool to room
temperature and then was further cooled in an ice bath. Crystals thus
obtained were recrystallized from dimethyl sulfoxide in the same manner
described for the 6-pheny1 analog in the preceding section, m.p. 250°
dec. (by heating at a rate of 20-50°/min.), 1.05 g. (95.5%). IR (in
nujol): 5540 (2.99), NH; 2282 (4.58), CN; 1724 (5.80), CO; 1690 (5.92),
1684 (5.94), 1560 (6.41), C=N and or NH; 852 (12.02), 1,4-disubstituted
benzene. NMR (in DMSO—d6): 7'5.86 (s), H, cyc10pr0pyl; T 2.57 (s), H,
aromatic; 1 0.58 (s, broad), N-hydrogens. Area ratio: 0.95 : 4 : 2.05
in the order described.

Anal. Calcd. for C ClNMO: C, 57.68; H, 2.61; C1, 15.10; N, 20.70

15H7
Found: C, 57.51; 'H, 2.74; Cl, 15.18; N, 20.77

91

K. Preparation of l,2-Dicyano-l,2-carboximidocyclOpropanes

1. 5,5-Pentamethylene-l,2-dicyano—l,2-carboximidocycloprOpane

 

One and eight-tenths gram (7.88 mmoles) of 1,5-dicyano-2-imino-
5-aza-4-keto-6,6-pentamethylenebicyclo[5.1.0]hexane was dissolved in
2 m1. of ethanol in a 5—ml. beaker. To this solution was added 5 dr0ps
of concentrated hydrochloric acid diluted with 10 drOps of water.
The mixture was gently heated on a hot plate for several mins. until
boiling. Then the solution was quickly filtered and the filtrate was
cooled in an ice bath. The crystals were filtered and recrystallized
from ethanol, m.p. 255—256°, 0.15 g. (85.1%). The m.m.p. with the
compound prepared by the known method (41) did not depress. IR (in
nujol): 5540 (2-99>, NH; 2280 (4.58), CN; 1790 (5.58), 1755 (5.70).
00. NMR spectrum (in acetone-d6) showed cyclohexane ring hydrogens in
2 adjoining peaks of a comparable intensity at T'8.26 and 7 8.05 and
the imido hydrogen at 7'-0.81. Area ratio: 10 : l for ring hydrogens
vs. N-hydrogen.

2. 5—Phenyl-1,2-dicyano-1,2-carboximidocycloprOpane

 

Three-tenths gram (1.27 mmoles) of 1,5-dicyano-2-imino-5-aza—4-
keto-6-pheny1bicyclo[5.1.0]hexane was mixed with 50 m1. of ethanol con-
tained in a lOO-ml. beaker. A mixture of 2 m1. of concentrated hydro-
chloric acid and 2 ml. of water was added to the beaker. The contents
of the beaker were boiled on a hot plate for 20 mins., cooled to room
temperature and then in an ice bath. The precipitate thus obtained was
filtered and recrystallized from an ethanol-acetone mixture, m.p. 268-
271° dec, 0.27 g. (89.6%). IR (in nujol): 5202 (5.12), 5102 (5.55),

NH; 2280 (4.58). CN; 1799 (5.56). 1722 (5.81). 00; 772 <12-95). 696

92

(14.55), monosubstituted benzene. NMR spectrum (in DMSO) showed the
cycloprOpyl hydrogen at T 5.44 (s) and aromatic hydrogens at T 2.48 (8).
However, there did not appear the peak corresponding to the imido hydrogen.
Anal. Calcd. for C15H7N502: C, 65.82; H, 2.97; N, 17.72
Found: C, 65.76; H, 2.91; N, 17.80

5. 5—prhlor0pheny1-1,2-dicyano-l,2-carboximidocyclopr0pane

 

Eight-tenths gram (2.95 mmoles) of 1,5-dicyano-2-imino-5-aza—4-
keto-6-pfchlor0phenylbicyclo[5.1.0]hexane was mixed with 10 ml. of ethanol
in a 20-ml. beaker. One millilitre of concentrated hydrochloric acid
diluted with 1 m1. of water was added to the beaker. The reaction mix-
ture was treated in the same manner described for the preparation of
5-pheny1 analog (in the preceding section). Recrystallization of the
precipitate from an ethanol-acetone mixture yielded 0.72 g. (90%) of
crystals which decomposed with effervescence at 284-290°. (A heating
rate of 20~50°/min. was employed). IR (in nujol): 5270 (5.06), NH;

2277 (4.59), CN; 1800 (5.56), 1725 (5.80), CO; 828 (12.08), 1,4-disub-
stituted benzene. The NMR (in DMSO-d6) showed the cycloprOpyl hydrogen
at T'5.42 (s) and aromatic hydrogens at 7 2.45 (s), but failed to Show
the N-hydrogen as was the case with the 5-phenyl analog in the preceding
section.

Anal. Calcd. for C15H6N502:

Found: C, 57.69; H, 2.05; 01, 15.08; N, 15.46

C, 57.48; H, 2.25; C1, 15.05; N, 15.47

93

L. Miscellaneous

 

1. Preparation of Compound Cl2HlON2O4S

3

In a 2-1. three-necked round-bottomed flask fitted with a mechanical
stirrer, a 250-ml. drOpping funnel, and a water condenser, were placed
550 m1. of carbon disulfide and 156 g. (1 mole) of dry ethyl sodiocyano-
acetate. While the contents were stirred, 44 ml. (about 0.82 mole) of
bromine diluted with 100 m1. of carbon disulfide was added over a 5-6
hr. time period. A yellow precipitate formed. The mixture was stirred
for an additional few hrs. after complete addition of bromine. Then the
precipitate was filtered* and washed with water to remove sodium bromide.
Recrystallization from benzene (also recrystallizable from THF or ace-
tonitrile) gave 44 g. (25.6%) of yellow crystals, m.p. 252-256° (lit.
val. (48), 250°). IR (in nujol): 2210 (4.25), CN; 1661 (6.02), CO;
1508 (7.64); 1181 (8.47), C-O-C. UV (in acetonitrile): 554 (27,600).
NMR (in DMSO-d6 at 80°): T-8.62 (t) and 1*5.60 (q) in an area ratio of
5 : 2 (A dr0p of unlabelled DMSO was used as an internal standard).

Anal. Calcd. for C N N 048 C, 42.09; H, 2.94; N, 8.18; S, 28.09

12 10 2 5‘

Found: C, 42.02; H, 5.40; N, 8.12; 3, 27.94
2. 0x1dat10n of Compound Cl2H10N2O4S5

In a 500-ml. three-necked round-bottomed flask equipped with a
mechanical stirrer and an air condenser, were placed 5.44 g. (0.01 mole)

of C H N 048 and 100 m1. of acetone. 'While stirring the contents, 15 g.

12 10 2 5

of powdered potassium permanganate was added at such a rate that the

 

* The filtrate was worked up to possibly isolate ethyl bromocyanoacetate.
However, no product was obtained.

94

reaction temperature did not exceed about 50°. Stirring was discontinued
at the point where an aliquot, on adding to water, did not precipitate
the starting material. The reaction mixture was filtered followed by
removal of acetone. A light yellow residue was obtained which, upon
recrystallization from an acetone—ethanol mixture gave a small amount
(about 0.5 g.) of light yellow crystals, m.p. 177—179°. IR (in nujol):
2250 (4.48), CN; 1696 (5.90), CO; 1560 (6.41), C=C (?). UV (in ethanol):
348 (52,804). 555 (48, 545), 215 (16,755).

Anal. Calcd. for Cl2HlON 2OMS2: C, 46.44; H, 5.25; N, 9.05; S, 20.66

Found: C, 46.57; H, 5.64; N, 8.80; S, 20.70

5. Attempted Reduction of Compound C H 048

12 10N 2 5

In a 500-ml. three-necked round-bottomed flask fitted with a
mechanical stirrer and an air condenser, were placed 5 g. (8.76 mmoles)

of C H 048

12 10N2 and 150 m1. Of dry tetrahydrofuran. While stirring the

5
contents, 5.5 g.(87.2 mmoles) of lithium aluminum hydride was added little
by little. After 50-60 mins., the reaction was discontinued and the
contents of the flask were poured to 150 m1. of water. The mixture was
then acidified with dilute hydrochloric acid (hydrogen sulfide evolved

at this time) followed by extraction with ether. Work-up of the ether
layer gave a gummy residue which could not be identified*.

4. Treatment of Compound 0 H 048 with Sodium Borohydride

12 10 N2 5

Three grams (8.76 mmoles) of C H 048 was dissolved in 70 m1. of

12 10N 2 5
tetrahydrofuran contained in a 500-ml. Erlenmeyer flask. To the flask

 

* It appears that an undentified reduction residue may cause severe
irritation and itching on the skin followed by swelling. It is
recommended to wear gloves while carrying out the reduction of this
compound using lithium aluminum hydride.

95

was added 0.7 g. (18.5 mmoles) of sodium borohydride little by little.
Heat was evolved and the solution bubbled with generation of hydrogen
sulfide. After bubbling stopped completely, the reaction mixture was
sloWbracidified with dilute hydrochloric acid. Then the mixture was
extracted with ether. Work-up of the ethereal solution gave a hygro-
scopic residue which could not be identified.

5. Attempted Desulfurization of Compound 0 H N 048

12 10 2 5

In a 500-ml. pressure bottle, were placed 150 m1. of tetrahydrofuran,

5 g. (8.76 mmoles) of C H N ous

12 10 2 and 6 teaspoonfuls of Raney nickel.

5).

The mixture was placed in a Paar Hydrogenator and Shaken for 2 days. The
contents were filtered and the filtrate was concentrated on a rotary
evaporator until the residual volume was about 50 ml. The residue was
poured into 200 m1. of water containing 10% of sodium chloride. The
mixture was extracted with three 50-ml. portions of ether. Work-up of
the ethereal layer gave about 2 m1. of dark brown liquid having an amine
odor. IR (a smear between NaCl plates): 5500 (2.86), 5400 (2.94), 2950
(5.59), 2220 (4.51), 1720 (5.81), 1675 (5.97), 1640 (6.10), 1469 (6.81),
1580 (7.20), etc. No further attempt was made to identify this residue.

6. Reaction of l-Nitro-2-methylpropene with Bromomalononitrile

 

Five and four-tenths grams (50 mmoles) of 1-nitro-2-methy1propene
(66) (purity better than 85% checked by NMR and v.p.c.*) and 7.0 g. (48
mmoles) of bromomalononitrile were dissolved in 70 m1. of 50% aqueous
ethanol contained in a l25—ml. Erlenmeyer flask. The reaction flask was

kept at room temperature for 22 hrs. White crystals that had formed were

 

* The other component (15%) was 2-nitromethy1pr0pene.

96

filtered. An additional crop was obtained after an additional 20 hrs.
The crude product amounted to 1.25 g. The crystals can be purified by
recrystallizing from ethanol or better by subliming at 140-145°/4 mm.,
m.p. 201-202°. IR (in nujol): 2290 (4.57), CN; 1250 (8.15), 1128 (8.86),
1111 (9.00), 740 (15.51). UV (in ethanol): transparent for whole range.
NMR (in acetone-d6): T'7.78 (s) and.‘r5.68 (s) in an area ratio of

6 : 0.9. This compound was tentatively identified as 5-(2'-bromo-2L
propyl)—l,1,2,2-tetracyanocyclopr0pane (see page 22).

Anal. Calcd. for C BrNfi: C, 45.65; H, 2.68; Br, 50.57; N, 21.50

10H7
found: C, 45.70; H, 2.64; Br, 50.52; N, 21.25

The filtrate obtained after removal of the white crystals was
allowed to stand further at room temperature for 15-20 hrs. A yellow
precipitate formed. It was filtered and purified either by recrystallizing
from ethanol or by subliming at around 150-155°/4 mm., m.p. 251-252°
dec. (Yield was not calculated.) IR (in nujol): 5550 (2.98), 5245 (5.08),
5175 (5.15), NH2; 2255 (4.43). CN; 1659 (6.05). 00; 1585 (6.51). 1556
(6.51). NMR (in DMSO-d6): 1‘8.49 (s) and 1~l.02 (broad) in an area ratio
of 6 : 1.82.
Anal. Found: C, 42.82; H, 4.05; N, 28 61 (C H8Nu0 )

7 5

7. Preparation of 1,2,5-Tricarbethoxy-l,2,5-tricyanocyclopr0pane

 

In a 2-1. three-necked round-bottomed flask equipped with a mechanical
stirrer, an air condenser, and a dropping funnel, were placed 87 g. (0.644
mole) of ethyl sodiocyanoacetate and 550 m1. of carbon tetrachloride.

While the suspension was stirred, 96 g. (52 ml., 0.60 mole) of bromine was

added for 2-5 hrs. The bromine was decolorized almost immediately. The

97

mixture was stirred for an additional 5—6 hrs., then filtered and washed
with water followed by dilute carbonate solution. (The sodium carbonate
layer became dark brown). The organic layer was dried over anhydrous
35 1.5261); b2
60-65° (1 g. n35 1.4555). These two liquid products were not identified

magnesium sulfate and fractionated: b2 45-47° (14 g. n

but did not seem to be ethyl bromocyanoacetate, the desired product of
this reaction.

The high-boiling liquid left over in the distillation flask
solidified on standing at room temperature for several hrs. The solid
mass was recrystallized from ethanol (1.5 g.). These crystals, m.p.
119-120°, were identified as 1,2,5-tricarbethoxy-1,2,5-tricyanocyclo-
pr0pane, m.p. 122-125° (67). IR (in nujol) lacked CN but Showed CO at
1750 (5.71), C—O-C at 1272 (7.86), and cyc10pr0pane ring (?) at 1025
(9.78). UV (in ethanol): 270 (5,214). NMR (in acetone-d6): 1‘8.61 (t)
and T 5.58 (q) (Each peak of the triplet and quartet was Slightly split
indicating that the three carbethoxyl groups are not all equivalent to
one another). Mass Spectrum (m/e): 555 (parent peak, p), 506 (p -

C H or p - HCN), 505 (p - CgHu), 288 (p - CQHSO).

2 5

8. Reaction of Is0propylilenemalononitrile with Bromonitromethane

 

One gram (9.45 mmoles) of iSOpropylidenemalononitrile and 1.67 g.
(12.00 mmoles) of bromonitromethane were dissolved in 10 ml. of aqueous
ethanol contained in a 50-ml. Erlenmeyer flask. The flask was set aside
at room temperature for a month. The contents in the flask were dark.
The solvent was removed and an oily residue was obtained. An attempt to
crystallize this residue from aqueous ethanol gave a small amount of 5,5-

dimethyl-l,1,2,2-tetracyanocyclopropane (identified by m.p., m.m.p., and IR).

98

9. Reaction of Cyclohexylidenemalononitrile with Bromocyanoacetamide

 

Nine-tenths gram (6.15 mmoles) of cyclohexylidenemalononitrile and
2.2 g. (15.5 mmoles) of bromocyanoacetamide was dissolved in 10 m1. of
ethanol. No precipitate was formed on standing for 15 hrs. at room
temperature. The reaction mixture (two layers) was poured into 50 ml.
of water and was allowed to stand for an additional several days at room
temperature. The precipitate that had formed was identified (by m.p.,
m.m.p., and IR) as 5,5-dimethy1-1,1,2,2-tetracyanocyc10propane, m.p. 177—
179° (from ethanol), 0.2 g. (19.1%). Work—up of the filtrate gave
bromocyanoacetamide.

10. Preparation of Compound Cl5H9BrN2

 

In a lOO-ml. round—bottomed flask were placed 0.6 g. (5.1 mmoles) of
2,5-benzocyclohexylidenemalononitrile and 55 m1. of 85% aqueous ethanol and
the solution was completed by warming the flask in an oil bath. Then 1.5
g. (10.5 mmoles) of bromomalononitrile was added to the flask and the
contents were refluxed for 4 hrs. The solvent was removed on a rotary
evaporator. A dark residue, on treating with Norit A followed by recrystal-
lization from ethanol, produced 0.54 g. (40.2%) of pinkish crystals, m.p.
155-158°. A Beilstein test suggested the presence of bromine. IR (in
nujol): 2245 (4.45), CN; 1605 (6.24), 1567 (6.58), 1545 (6.48), C=C;

755 (15.61), ortho disubstituted benzene. ‘UV (in ethanol): 522.5 (16,740),
255 (6, 850). NMR (in DMSO-d6): 1-7.55 (m), 1-6.94 (m), 1 4.56 (t, J =
5.6 c.p.s.), T 2.47 (m), 7 1.75 (m), Area ratio: T 7.55 :1 6.94 : 7'4.56

1'2.47, 1.75 = 1.97 ; 2.01 : 0.92 : 4.00.

99

Anal. Calcd. for Cl5H9BrN2: C, 57.16; H, 5.52; Br, 29.26; N, 10.26
Found: 0, 57.55; H, 5.45; Br, 29.18; N, 10.20

11. Treatment of Compound C H BrN

13 9 2 with Pyridine

Three-tenths gram of compound Cl5H9BrN2 wanissolved in 2 ml. of
pyridine. The solution was boiled on a hot plate for 2 mins. The dark
reaction mixture was poured to 10 m1. of cold water. A caky precipitate
that had formed was filtered and treated with Norit A followed by recrys-
tallization from ethanol. There were obtained deep magenta crystals, m.p.
158—159° dec. The compound was negative to the Beilstein test for
halogen. IR (in nujol): 2290 (4.57), 2245 (4.45), CN; 1605 (6.25),

1565 (6.40), C=C; 778 (12.85), 756 (15.59), aromatic. UV (in ethanol):
505 (1,522), 509 (5,701), 252 (11,504), 207 (29,216).

12. Preparatlon of Compound Cl2HI6Br2N2

 

In a 250-ml. round-bottomed flask were placed 1 g. (5.55 mmoles) of
2,5-benzocyclopentylidenemalononitrile, 100 ml. of ethanol, and 10 ml. of
water. The solution was completed by heating the mixture in an oil bath.
Three grams (20.7 mmoles) of bromomalononitrile was added. The contents
of the flask were refluxed for 5-6 hrs., during which time a precipitate
formed in the flask. Filtration followed by recrystallization from acetone
gave light magenta crystals, m.p. 208.211°. (Yield not calcualted) IR
(in nujol): 2240 (4.46), CN; 1599 (6 25), 1568 (6.58), C=C; 790 (12.66),
aromatic. UV (in ethanol): 546 (15,751), 557 (15,859), 256 (7,140). An
attempt to measure the NMR spectrum.(in DMSO-d6) was not successful due

to insufficient solubility.

100

Anal. Calcd. for Cl2H6Br2N2: C, 42.64; H, 1.79; Br, 47.28; N, 8.29
Found: C, 42.72: H, 1.52; Br, 47.65; N, 8.11
Work-up of the reaction filtrate and the acetone filtrate(obtained
after recrystallization of the bromine-compound) gave some recovered
starting material, 2,5—benzocyclopentylidenemalononitrile.

15. Reaction of Bromomalononitrile with Ethanol

 

Five grams (54.5 mmoles) of bromomalononitrile was dissolved in 15 ml.
of aqueous ethanol and the solution was refluxed for 8 hrs. The solvent
was removed on a rotary evaporator. To the black residue thus obtained
was added 50 ml. of water and the mixture was cooled in an ice bath.

The solid part was filtered, treated with Norit A, and recrystallized
from ethanol. There was obtained 0.2 g. (8.4%) of l,l-dicyano-2-amino
2-ethoxyethylene, m.p. 251—255° (255-257° on further recrystallization).
(lit val. (52), 225-226°) IR (in nujol): 5555 (2.98), 5250 (5.10), NH;
2248 (4.45), 2205 (4.54), CN; 1658 (6.05), 1550 (6.45), 1505 (6.66),

C=C and/or NH. UV (in nujol): 255 (18,656). NMR (in DMSO-d6): 1 8.71
(t, J = 7.0 c.p.s.), CH50H20; 1‘5.72 (q, J = 7.0 c.p.s.), CH50H20; 7 1.45
(s) N112.

Anal. Calcd. for 06H C, 52.55; H, 5.15; N, 50.64

7N50:
Found: C, 52.71; H, 5.07; N, 50.75
This compound was obtained as a precipitate while bromomalononitrile
was allowed to react with 5,5-dimethyl-2-butylidenemalononitrile, 2,4-
dimethyl-5-pentylidenemalononitrile, 6-hendecylidenemalononitrile, or

tetracyanoethylene. The reaction was carried out in ethanol and at room

temperature for a time period of 2—4 weeks.

101

The aqueous filtrate obtained after removal of the l,l-dicyano-2-
amino-2-ethoxyethylene. was concenteratéd on a rotary evaporator. The
solid residue obtained was dissolved in a minimum amount (2 ml.) of
water followed by filtration. To the clear filtrate was added acetone
until precipitate formed. The precipitate was filtered and further
recrystallized in the same manner. The compound thus obtained did not
melt sharply but sublimed at 270-500°. The compound contained bromine
and nitrogen. The IR spectrum (in nujol and KBr pellet) did not Show
significant absorption bands. The NMR (in D20) spectrum showed a

rather broad singlet at 7 5.15. The compound was ammonium bromide.

SUMMARY

1. A number of 5,5-dialky1-l,l,2,2-tetracyanocyclopropanes were
prepared by reacting the corresponding alkylidenemalononitriles with
bromomalononitrile. The reaction appeared to be influenced by steric
and eleCtronic effects of the alkylidenemalononitriles.

2. Some of 5,5-dialkyl-l,l,2,2-tetracyanocyclopropanes that
failed to form by the original Wideqvist reaction (reaction of the
carbonyl compounds with bromomalononitrile in the presence of iodide
ion) were obtained by this method.. The compounds thus prepared are
5-ethyl-5-nfbuty1-, 5,5-dicyclopropyl-, 5,5-nonamethylene-, 5,5-
undecamethylene-, and 5,5-tetradecamethylene-l,1,2,2-tetracyanocyclo-
pr0pane.

5. A number of 5-aryl-1,1,2,2-tetracyanocyc10pr0paneS and 5—alkyl-
5-ary1-l,l,2,2-tetracyanocy010propanes were prepared by reacting the
corresponding benzylidenemalononitriles or 5-alkylbenzylidenemalonos
nitriles with bromomalononitrile. Introduction of an electron-with-
drawing group in the benzene ring did not seem to particularly facilitate
the reaction but the introduction of an electron-releasing group did slow
down the reaction.

4. The cyclopropyl hydrogen of a number of 5-aryl-l,l,2,2-tetracyano-
cycloprOpanes was shown to couple to the 25222 hydrogens of the aromatic
group. However, exceptions were encounteredin 5gg-nitr0phenyl- and 5-mf
chlor0phenyl-l,l,2,2-tetracyanocy010pr0pane which showed a complex splitting
for the cyclopropyl hydrogens.

102

10.5

5. Reaction of 2,5-benzocyclohexylidenemalononitrile with
bromomalononitrile produced two compounds depending on the reaction
conditions, i.e., at room temperature and in 80% aqueous ethanol, spiro-
[2,2,5,5-tetracyanocyclopropane-1,1'-tetralin] was produced, whereas,
at reflux or in 95% ethanol, there was obtained a compound ClegBrN2
which was identified as 2-bromo-l-dicyanomethy1enetetralin. Reaction
of 2,5-benzocyclopentylidenemalononitrile with bromomalononitrile
produced a compound 012H6Br2N2 which has not yet been identified.

6. A number of 5,5-dialkyl- and 5-ary1-2-carbethoxy-l,1,2-tricyano-
cyclopropanes were prepared by reaéting the corresponding ethyl alkylidene-
or benzylidenecyanoacetates with bromomalononitrile. Alternatively,
these compounds were prepared by reacting alkylidene- or benzylidene-
malononitriles with ethyl bromocyanoacetate. The first route appeared
to be superior to the second route, because it produced the compounds
quicker and in better yield and furthermore, some compounds which failed
to form by the second route were produced by the first route.

7. 5-Methyl-5-ifpr0pyl-, 53phenyl—, and 5—pfmethoxyphenyl-
2_carbethoxyal,1,2-tricyanoCyclopr0paner were shown, by NMR spectra, to
be produced as Single stereoisomers in which the ieprOpyl, phenyl, and
pfmethoxyphenyl groups are EEEE§.t° the carbethoxyl group.

8. 5-Methy1-5-iepropyl-2-carbethoxy-l,1,2-tricyanocyc10propane
showed the ifprOpyl methyl groups in two separate doublets in NMR spectrum.
This was explained in terms of magnetic non-equivalence of the two methyl
groups due to a preferred conformation.

9. Cyclohexylidene-, benzylidene-, and p-chlorobenzylidenecyano-

acetamides, on reacting with bromomalononitrile, produced 5,5-pentamethy1ene-

10'4

5-phenyl, and 5-pfchlor0pheny1-2-carboxamido-l,l,2-tricyanocyclopropane,
respectively. These carboxamides, on treating with boiling methanol or
ethanol, were converted to 6,6-pentamethy1ene-, 6-phenyl-, and 6-pf
chlorophenyl-l,5-dicyano-2—imino~5~aza-4—ketobicyclo[5.1.01hexane,
respectively. The acid treatment of these bicyclo[5.l.0]hexanes pro-
duced 5,5-pentamethylene-, 5-phenyl, and 5-pfchlorophenyl-l,2-dicyano-
l,2-carboximidocyclOprOpane, respectively. 7

10. Dimers of isopropylidenemalononitrile, 2-butylidenemalono-
nitrile, and cyclopentylidenemalononitrile were prepared by treating the
corresponding monomers with pyridine. These dimers were also formed
in the reaction residue while distilling the monomers.

11. Reaction of bromomalononitrile with ethanol produced l,l-
dicyan>2~amino-2-ethoxyethylene.

12. Reaction of l-nitro-2-methylpr0pene with bromomalononitrile
produced two compounds ClOH7BrNN and C7H8N505. Compound ClOH7BrN4 was
.identified as 5-(2'-bromo-2'-pr0pyl)-1,l,2,2-tetracyanocyclopropane.

15. Reaction of ethyl sodiocyanoacetate with bromine in carbon

disulfide produced compound C12H10N20u8 which was tentatively identified

5
as 5,5-bis(cyanocarbethoxymethylene)-l,2,4-trithiacyclopentane.
14. Reaction of ethyl sodiocyanoacetate with bromine in the carbon

tetrachloride produced l,2,5-tricyano-l,2,5-tricarbethoxycyclopropane.

LITERATURE CITED

1. L. Ramberg and S. Wideqvist, Arkiv. Kemi, Mineral. Geol., 12A, No. 22,
12 pp. (1957).

2. H. Hart and F. Freeman, J. Am. Chem. Soc., 85, 1161 (1965).

5. S. Wideqvist, Arkiv. Kemi, Mineral. Geol., 29;, No. 4, 8 pp. (1945).
4. H. Hart and F. Freeman, J. Org. Chem., 88, 1220 (1965).

5. R.c1 fearson and R. L. Dillon, J.‘Am. Chem. Soc., 75, 2459 (1955).
6. 0. Widman, Ber., 58, 555 (1918).

.7. 0. Widman, Ber., 51, 907 (1918).

8. R. Fraisse and R. Jacquier, Bull.Soc. Chim. France, 986 (1957).

9. R. Fraisse, Bull.Soc. Chim. France, 1102 (1959).
10. R. Fraisse and M. Guitard, Bull. Soc. Chim. France, 418 (1960),
11. R. Fraisse and M. Guitard, Bull. Soc. Chim. France, 788 (1960).
12. R. Fraisse and M. Guitard, Bull. Soc. Chim. France, 200 (1961).
15. G. Westoo, Acta Chem. Scand¢,11, 1290 (1957).
14. G. Westoo, Acta Chem. Scand., 12, 685 (1959).
15. G. Westoo, Acta Chem. Scanda,15, 689 (1959).
16. D. T. Warner, J. Org. Chem., g8, 1556 (1959).
17. M. Mousseron and R. Fraisse, Compt. Rend. (Acad. Sci.), 858, 887 (1959).

18. M. Mousseron, R. Fraisse, R. Jacquier, and G. Bonavent, Compt. Rend.

(Acad. Sci.), £88, 1465 (1959).

19. M. MOusseron, R. Fraisse, R. Jacquier, and G. Bonavent, Compt. Rend.

(Acad. Sci.), 248, 2840 (1959).
20. L. L. McCoy, J. Am. Chem. Soc., 89, 6568 (1958).

21. L. L. McCoy, J. Org. Chem., g5, 2078 (1960).

105

55-
54.
55-
56.

37.
58.

59-

40.

41.

42.
45.
44.

l06

L. L. MCCoy, J. Am. Chem. Soc., 88, 6416 (1960).

L. L. McCoy, J. Am. Chem. Soc., 88, 2246 (1962).

L. L. McCoy, J. Org. Chem., 32: 240 (1964).

L. L. MCCoy and G. W. Nachtigall, J. Org. Chem., 87, 4512 (1962).
E. Ciganek, J. Am. Chem. Sco., 81, 652 (1965).

J. S. Swenson and D. J. Renaud, J. Am. Chem. Soc., 81, 1594 (1965).
F. Freeman, Ph.D. Thesis, Michigan State University, 1962.

H. Hart and J. M. Sandri, J. Am. Chem. Soc., 8;, 520 (1959).

H. C. Brown, J. Chem. Soc., 1248 (1956).

H. Hart and Y. C. Kim, Unpublished Results.

N. S. Bhacca, D. P. Hollis, L. F. Johnson, and E. A. Pier, "NMR
Spectra Catalog,” Vol. 2, Varian Associates, Palo Alto, California,
1965, Spectrum No. 557.

W. J. Middleton and V. A. Engelhardt, J. Am. Chem. Soc., 88, 2788 (1958).
s. Trofimenko, J. Org. Chem., g8, 2755 (1965).

G. L. Gloss and R. A. Mess, J. Am. Chem. Soc., 88, 4042 (1964).
C. N. Banwell and N. Sheppard, DiscussiOns Faraday Soc., 58, 122 (1962).
G. J. Karabatsos and F. M. Vane, J. Am. Chem. Soc., 82, 5886 (1965)-

D. J. Patel, M. E. H. Howden, and J. D. Roberts, J. Am. Chem. Soc.,
532. 5218 (1963).

T. H. Siddall, III and C. A. Prohaska, J. Am. Chem. Soc., 84, 2502
(1962). "—

S. Goodwin, J. N. Shoolery, and L. F. Johnson, J. Am. Chem. Soc., 88,
5065 (1959).

S. F. Birch, W. H. Gough, and G. A. R. Kon, J. Chem. Soc., 119, 1515
(1921).

J. K. Williams, J. Org. Chem., g8, 1054 (1965).
M. R. S. Weir and J. B. Hyne, Can. J. Chem., 8;, 2905 (1965).

M. R. S. Weir and J. B. Hyne, Can. J. Chem., 83, 1440 (1964).

 

 

45.

51.

52.

55.
54.

55.
56.

57.
58.

59.

66.
67.

107

G. V. Smith and H. Kriloff, J. Am. Chem. Soc., 85, 2016 (1965).
R. Weiss, Ph. D. Thesis, Michigan State University, 1964.

J. F. Thorpe and W. J. Young, J. Chem. Soc., 11, 956 (1900).

G. Wenzel, Ber., 58, 1045 (1901).

R. Gompper and_W. Topfel, Ber., 92, 2861 (1962),

L. M. Jackman, ”Application of Nuclear Magnetic Resonance SpectrOSCOpy
in Organic Chemistry,° Pergamon Press, New York, 1959, p. 52.

N. S. Bhacca, D. P. Hollis, L. F. JohnsOn, E. A. Pier, "NMR Spectra
Catalog," Vol. 2, Varian Associates, Palo Alto, California, 1965,
Spectrum No. 591.

G. Errera and F. Perciabosco, Ber., 22, 2976 (1900),

F. Arndt and J. D. Rose, J. Chem. Soc., 1 (1955),

S. Wideqvist, Acta Chem. Scand. 7, 696 (1955),

B. C. Hesse, Am. Chem. Jo, gé. 725 (1896).

D. G. I. Felton, J. Chem. Soc., 515 (1955),

J. Tscherniac, Ber., 28, 2588 (1897).

F. S. Frout, J. Org. Chem., 18, 928 (1955).

A. C. Cope and E. M Hancock, "Organic Syntheses,” 0011. Vol. III,
John Wiley and Sons, Inc. , New York, 1955, p. 599.

B. B. Corson and R. W. Stoughton, J. Am. Chem. Soc., 88, 2825 (1928),
E. P. Kohler and M. Reimer, Am. Chem. J., 55, 555 (1905).

J. Zabicky, J. Chem. Soc., 685 (1961).

G. P. Schiemenz, Ber., 98, 485 (1962),

A. C. Cope and K. E. Hoyle, J. Am. Chem. Soc., 88, 755 (1941).

J. D. Kendall, H. G Suggate, and J. HoiMayo, Brit. 704, 770, Mar. 5,
1954; C. ,48, 9850i (1954)

H. Shechter and J. W. Shepherd, J. Am. Chem. Soc., 18, 5617 (1954).

A. P. Krapcho and P. S. Huyffer, J. Org. Chem., 88, 2904_(l965).