WI

‘
1

‘

l
i

”H

t

 

_{
1833 III
CID—LP

33mm}; T0 WON EEEZE N639 ‘3? Ws‘RHCS

as;
”fig
. WE‘RE I, 5-3EACETYLENES

’E‘hesis ior the Degree of v1.3
Mi CREME STAT ”REVERSE. W
REEBERT ':.. RRAUSE
13387

.5fl‘ \

LIBRARY "2

Michigan 5 are H
Universit .7

  
 
  

3.3.1213);

ABSTRACT

APPROACHES TO NON-BENZENOID AROMATICS
FROM 1.5-DIACETYLENES

by Robert E. Krause

Theoretical calculations have suggested three examples
of non-benzenoid aromatic compounds which lend themselves
to preparation via 1,5-diacetylenic compounds.

The first of these examples studied was that of the

1,4-dehydrobenzene.

1,4-dehydrobenzene

Because of the similarities between 1,4—dehydrobenzenes and
cyclobutadienes, methods which have been successful in the
preparation of cyclobutadiene derivatives were applied.
After initially failing to produce the necessary
intermediates to a 1,4-dehydrobenzene from meso-1,6-diphenyl-
3 ,4-dimethyl-1, 5-‘hexadiyne—3 ,4-diol (v11); g—bis (phenyl-
ethynyl)benzene (XVI) was synthesized, and attempts were
made to form a metal complex of a 1,4-dehydrobenzene from
this compound. Regrettably, only polymeric products could
be isolated from reactions with this intermediate, and the
synthesis of a stable 31,4-dehydrcbenzeh'e by these routescould‘not

be successfully completed.

Robert E. Krause

 

(v11) (XVI)

A second non-benzenoid aromatic species of interest
was gyphenylene-bis(phenylcyclopropenone) (IV) which was
considered interesting because of the close proximity of

the two aromatic cyclopropenone groups to each other.

0..

(Iv)

Also, isolation of this compound would facilitate the form—

ation of a cyclobutene dication derivative (VI) by photo-

lytic rearrangement to obtain a third non-benzenoid aromatic
compound.

3 Robert E. Krause

 

(VI)

Attempts to form the gfphenylene-bis(phenylcyclopropenone)
(IV) by reaction with dichlorocarbene on 27bis(phenyl—
ethynyl)benzene (XVI) failed, however, to insert two di-
chlorocarbene units, and only the mono insertion product
(gfphenylethynylphenyl)phenylcyclopropenone (XXI) was

isolable.

(XXI)

APPROACHES TO NONvBENZENOID AROMATICS

FROM 1,5-DIACETYLENES

BY

Robert E..Krause

A THESIS

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

MASTER OF SCIENCE

Department of Chemistry

1967

ACKNOWLEDGMENT

The author wishes to express his gratitude and ap-
preciation to Dr. Eugene LeGoff for his able guidance
and helpful suggestions throughout the course of this

investigation.

ii

TABLE OF CONTENTS

INTRODUCTION
-RESULTS AND DISCUSSION . . . . . . . .
'EXPERIMENTAL . . . . . . . . . . . . . . . . . . .

General Procedures and Apparatus

A.

M.

SPECTRA

1, 6-diphenyl- -3, .4-dimethyl-1, 5-hexadiyne-
3, 4-diol . . . . . . . .

Dibenzoate of 2, 3— —diphenylethynyl- -2, 3—
butanediol . . . . . . . . .

1,3—dioxa-2-ethoxy-4,5-dimethyl-4,5-~
diphenylethynylcyclopentane, Attempted .

2, 5-dioxa-3, 4-dimethyl- -3, 4- -diphenylethynyl-
cyclopentanone . . . . . . . .

1,3-dioxa-2-phenyl-4,5—dimethyl-4,5-
diphenylethynylcyclopentane . . . . . . . .

Reaction of 1, 6-diphenyl-3, 4- -dimethyl-1, 5-
hexadiyne-3, 4-diol with .triphenylphosphine
dibromide . . . . . . . . .
27bis(pheny1ethynyl)benzene
27bis(2—phenylethyl)benzene

Bis(benzonitrile)palladium II chloride

Bis(acetonitrile)palladium II chloride . .

<Reaction of o-bis(phenylethynyl)benzene

with dichlorocarbene; Method 1. . . . . .

Reaction of 97bis(phenylethynyl)benzene
with dichlorocarbene; Method II . . . .

Infrared Spectrum of 27bis(phenylethynyl)benzene

Ultraviolet Spectrum of .o-bis(phenylethynyl)-
benzene . . . . . . . . . . .

iii

Page

23
23

23

24

25

26

26

27

28

29

29

30

31

32
34
35

36

TABLE OF CONTENTS (Cont.)
Page
Infrared Spectrum of 27bis(2-phenylethyl)benzene 37

Ultraviolet Spectrum of 97bis(2-phenylethyl)—
benzene . . . . . . . . . . . . . . . . . . . . . 38

Infrared Spectrum of (gfphenylethynylphenyl)-
phenylcyclopropenone . . . . . . . . . . . . . . 39

Ultraviolet Spectrum of (gfphenylethynylphenyl)—
phenylcyclopropenone . . . . . . . . . . . . . . 4O

LITERATURE CITED . . . . . . . . . . . . . . . . . . . 41

iv

SCHEME

II.

III.

REACTION SCHEMES

Page
-Unsuccessful preparation of 2,3-dimethyl-5,6-
diphenyl-l,4—dehydrobenzene from meso-1,6-
diphenyl-3,4-dimethyl-1,5-hexadiyne-3,4—diol. 9
Synthesis of 27bis(phenylethynyl)benzene, and
unsuccessful conversion to 2,3-diphenyl-1,4-
dehydronaphthalene . . . . . . . . . . . . . 14
AConversion of 27bis(phenylethynyl)benzene to
cyclopropenones . . . . . . . . . . . . . . . 20

INTRODUCTION

A useful precursor for the formation of non—benzenoid
aromatic systems is the-acetylenic group. Perhaps the best
known uses of this functional group for this purpose are in
the formation of cyclobutadiene-metal complexes and cyclo—
propenone aromatic systems. It is also possible, however,
that other non—benzenoid aromatic systems can be made via
an acetylenic intermediate.

In 1952, Roberts, Streitwieser, and Regan suggested
the existence of a stable 1,4-dehydrobenzene by their cal-
culations attributing a resonance energy of 1.666 or about
28 kcal. to the parent bicyclo[2.2.0]hexa-1,3,5—triene,

more commonly referred to as 1,4-dehydrobenzene or butalene

(1) -

7:7

 

 

 

 

1,4—dehydrobenzene

Further calculations were done by Dewar and Gleicher
who calculated a resonance energy of 0.984 e.v. or 22.7
kcal. by Hfickel molecular orbital calculatiOns;_ O.191'e.v.
or 4.40 kcal. by POple, Pariser, and Parr methods; and
0.391 e.v. or 9.02 kcal. by a modified split prorbital

approach (2).

2

Computer caICulations done by Lee also duplicate the
results obtained by Roberts, Streitwieser and Regan (3).

One possible observation of the formation of stable
1,4-dehydrobenzene was reported by Berry, Clardy, and Schafer
in 1965 (4). Analysis of spectroscopic and mass spectral
data gathered from the flash initiated decomposition of
benzenediazonium-4-carboxylate indicated the formation of
a species of mass 76, attributed to 1,4-dehydrobenzene,
having a lifetime of two minutes. However, no conclusive
evidence was presented to rule out other isomers of this
compound.

AThis empirical evidence, together with the theoretical
calculations, lends encouragement to a possible successful
synthesis of a stable 1,4-dehydrobenzene.

Because of structural similarities between 1,4-dehydro-
benzenes and cyclobutadienes, stabilizing factors for the
cyclobutadiene system may be applied to the 1,4-dehydro—
benzene system as well.

Roberts, Streitwieser, and Regan have completed
Hfickel molecular orbital calculations for a number of the
cyclobutadiene systems (1). While their calculations have
shown cyclobutadiene to exist in a triplet ground state
with no delocalization energy, vinyl cyclobutadiene was
shown to have a delocalization energy of 0.605 and 1,2-
divinyl cyclobutadiene to have a delocalization energy of
1.213. -Most significant was phenyl cyclobutadiene which was

shown to have a delocalization energy of 2.535.

3

The applicability of this data to the 1,4-dehydro-
benzene system is indicated by the fact that the delocali-
zation energy of 1.665 for the parent 1,4—dehydrobenzene
is increased to a delocalization energy of 2.475 when a
vinyl group is added across the 2,3 positions of the 1,4-
dehydrobenzene nucleus. It would thus seem that the best
chances for synthesizing.a stable 1,4-dehydrobenzene would
exist when the 1,4-dehydrobenzene being synthesized was
highly substituted, preferably with phenyl groups. (Such
substitution would also serve to prevent complicating re-
arrangements due to hydrogen migrations.

In attempting a synthesis of such a compound, it was
considered advantageous, because of structural similarities
with the cyclobutadiene system, to use reactions and tech-
niques which have led to the successful preparation of metal
complexes of cyclobutadiene derivatives. ~Decomposition of
this complex, then, should be expected to produce the free
1,4-dehydrobenzene derivative. Although this last step has
failed for cyclObutadienes, it was felt that the increased
delocalization energy of the 1,4-dehydrobenzene system, plus
the existence in a singlet ground state as against a triplet
ground state for cyclobutadienes, should greatly increase
the chances for its isolation.

Utilization of this approach necessitates the isola-
tion of an intermediate with either of the following groups,
where -X indicates a good leaving group which will allow

for the in situ formation of intermediate I.

   

(I) (II)

The initial attempt involved converting meso—1,6—di-
phenyl-3,4-dimethyl-1,5-hexadiyne-3,4—diol (VII) toaa
suitable intermediate from which a compound of the follow-

ing type could be synthesized.

O
/

“W

O M CH3

A second route involved forming the intermediate

 

 

 

 

27bis(phenylethynyl)benzene (XVI), which could be converted

to the following metal complex.
\
\ /

l .
\ a
“O

 

 

 

5
This intermediate was synthesized as a new compound by the
method of Stevens and Castro (5), but it has since been re-
ported in the literature (6).

A second non-benzenoid aromatic Species derivable from
an acetylene is cyclopr0penone. A number of workers have
synthesized diphenylcyclopropenone and have shown that it
exists in the form of species III (7). This is consistent
with theory, as such a species, according to the Hfickel
4n + 2 rule, would exhibit aromatic character and thus
gain stability.

By analogy with diphenylcyclopropenone (III), it should
be possible to convert 27bis(phenylethynyl)benzene (XVI)

.

to species IV.

   

(III) (IV)

Formation of such a species would be interesting for two
reasons. First, it would be interesting to note the prop—
erties of such a molecule containing two cyclopropenone
groups in close proximity to each other. Secondly, iSola-

tion of this intermediate would allow for the formation of

6
an interesting derivative of the following non-benzenoid
aromatic species (V), which should also exhibit aromatic

character as predicted by the Hfickel 4n + 2 rule.

(V)
Rearrangement of the dicyclopropenone compound (IV),
either photolytically, or with heat, might be expected to

lead to formation of the species VI, which should exhibit

most interesting properties.

0—

5°30

RESULTS AND DISCUSS ION

The initial goal of this investigation involved the
synthesis of a stable 1,4-dehydrobenzene. The method pro-
posed to accomplish this synthesis was to convert meso-1,6—
diphenyl-3,4-dimethyl-1,5—hexadiyne-3,4-diol (VII) to its
orthoformate derivative (VIII)(Scheme I). Pyrolysis of‘
this intermediate was expected to result in the loss of
carbon dioxide and ethanol with the formation of the olefin
(IX)(8) which could be reacted with a palladium chloride
derivative to form the palladium chloride complex of 2,3-
dimethyl-4,5-diphenyl-1,4—dehydrobenzene (XI). Careful.de—
composition of the complex with a phosphine would then
produce the expected 1,3—dimethyl-4,5-diphenyl-1,4-dehydro-
benzene (XII) (9).

To accomplish this, 1,6—diphenyl-3,4-dimethyl-1,5-hexa-
diyne-3,4-diol (VII) was synthesized by the method of Wilson
and Hyslop (10). The isomers were separated by fractional
crystallization and the low melting isomer was shown to be
the meso form by its behavior in the nmr.

In a solvent of low polarity, such as deuterochloroform,
the racemic isomer would be expected to assume a configura-
tion where minimal interactions between methyl hydrogens
will occur. This may result in a very slight deshielding
in the nmr, since this places the methyl hydrogens in the
vicinity of the acetylene groups, but it places both methyl

7

8
groups in an equally unhindered environment. The high
melting isomer behaves in this manner and exhibits a single
methyl peak at 18.19 in the nmr.

In the same solvent, the meso isomer can only orient
one of its methyl groups in the same unhindered orientation
as in the racemic isomer. The other methyl group must then
assume a position where it interacts with a hydroxyl group
rather than with an acetylene group. This might be ex—
pected to result in less deshielding in the nmr, and the
meso isomer should therefore exhibit methyl peaks of slightly
different chemical shifts in the nmr. The low melting iso-
mer behaves in this manner and exhibits two methyl peaks
at 18.17 and 18-25 in the nmr.

Further confirmation of the meso structure comes from
the behavior of the low melting (meso) isomer in a polar
solvent. In dimethylsulfoxide, intramolecular hydrogen
bonding takes place and the methyl groups are drawn into
a slightly hindered but unshielded or deshielded position.
In this solvent, the methyl groups exhibit a singlet at
18.29 in the nmr and the hydroxyl groups exhibit a doublet
at T4.33(J= 6 cps). Introduction of a trace of strong
acid increases proton exchange and moves this hydroxyl peak
to 73.8 as a broad singlet. This behavior is consistent
with what should be expected, as the intramolecular hydrogen
bonding draws the;methyl groups out of their slightly dif-

ferent environments and into a single common environment.

 

 

 

 

 

 

19

Unfortunately, reaction of meSo-1,6—diphenyl—3,4-di-
methyl-1,5-hexadiyne-3,4-diol (VII) with ethyl orthoformate
to produce the orthoformate derivative (VIII) did not appear
to form the expected derivatiVe. While a high yield of a'
liquid product was obtained, its nmr spectrum was in no way
iconsistent with the expected product.

In an attempt to synthesize the olefin 1,6-diphenyl-
3,4-dimethyl-3-hexene-1,5—diyne (IX) directly, two other
reactions were tried.

First, numerous sets of conditions were tried by which
a modified Chugaev type reaction might be used to convert
1,6-diphenyl-3,4—dimethyl-1,5-hexadiyne-3,4—diol (VII) to
1,6—diphenyl-3,4-dimethyl-3-hexene-1,5-diyne (IX). How-
ever, in all cases only decomposition products and carbonyl
containing rearrangement products were obtained.

*An attempt was also made to form 1,6—diphenyl-3,4-di-
methyl-3-hexene—1,5-diyne (IX) from 1,6-diphenyl-3,4—di-
methyl-1,5-hexadiyne-3,4-diol (VII) by reduction with stan-
nous chloride in methanolic HCl (11). In this case, analy-
sis indicates that some 1,6-diphenyl-3,4—dimethyl-3—hexene-
1,5—diyne (IX) may have been formed. Unfortunately, there
appears to be a competing reaction taking place which makes
conditions extremely critical. After a number of attempts,
the best reaction that could be obtained yielded only a
moderate yield of crude red oil product along with some
carbonyl containing by—product and much contaminated start-

ing materials. Extending the reaction time to consume all

a 11.
the starting material only resulted in less olefin fraction
and more carbonyl containing fraction.

In Pettit's work on cyclobutadiene (12), the second
double bond of the cyclobutadiene was formed by the abstrac-
tion by diiron enneacarbonyl of two cis vicinal chlorines
to form the cyclobutadiene which was then immediately
complexed by more diiron enneacarbonyl.

With the prospect that this same type of double bond
formation might be possible in the formation of the 1,4-
dehydrobenzene, attempts were made to replace the hydroxyl
groups on the diol with other groups which could be easily
displaced, while not causing decomposition of the reagents
in the reaction system.

The dihalo derivatives were considered to be the most
desirable, and thus were given the most attention. At-
tempts were made to obtain the dichloroderivative (X) of
1,6-diphenyl-3,4-dimethyl-1,5—hexadiyne-3,4-diol (VII) by
treatment with thionyl chloride, but after numerous sets of
reaction conditions had been tried, no dichloride was isol—
able. The reaction either produced black intractable
masses of decomposition products, or formed impure non-
chlorine containing products which appeared from their infra—
red spectra to be sulfite esters, a reasonable reaction pro—
duct in view of the 1,2 arrangement of the hydroxyl groups.

Replacement of the thionyl chloride by phosphorous tri—
chloride again gave only non-chlorine containing products
which again gave an infrared spectrum consistent with ester

formation, in this case a phosphite ester.

12
A near sucéess was observed in the reaction of 1,6-di—
phenyl-3,4-dimethy171,5-diyne-3,4-diol:dVII) with triphenyl-
phOSphine dibromide in dimethylformamide solution (13).
~However, even running the reaction without external heating
gave only an undesired product which gave analyses and nmr
spectra consistent with 1,6-diphenyl-3—methylidene—4-methyl—

4-bromo-hexa-1,5—diyne (XIII).

 

(XIII)

It is believed that the desired dibromo compound may
actually have been formed, but because of the sterically
hindered conditions resulting from the 1,2-arrangement of
the bromines,‘a more stable product was formed by loss of
a molecule of hydrogen bromide to give the isolated product.

An attempt to make the dichloroderivative (X) with
triphenylphosphine and carbon tetrachloride under reflux
gave only starting material containing a trace of some
unidentifiable product. Apparently these reaction condi-
tions were too mild.

Two other derivatives were attempted, but neither was

13
entirely successful. «Reaction of 1,6—diphenyl-3,4-dimethyl-
1,5-hexadiyne-3,4-diol (VII) with ethyl chloroformate gave a
product with only phenyl and methyl peaks in the nmr. No
complete structure proof was done, but the spectral evidence
is consistent with a carbonate product. Reaction of 1,6—
diphenyl-3,4-dimethyl-1,5-hexadiyne-3,4-diol (VII) with
benzaldehyde to form an acetal was also attempted, but in
spite of a moderate yield of product, as identified by nmr,
the experimental problems of heavy foaming and poor chroma-
tographic separation resulted in this reaction being set
aside from further work.

Because of a lack of success with the diol system, it
was decided to try another system which contained a built-
in 3,4 double bond. Modeled on the work of Castro and
Stevens in synthesizing tolanes (5), it was decided that
97bis(phenylethynyl)benzene (XVI) would be a useful inter-
mediate since it contains three benzene rings to stabilize
the 1,4-dehydrobenzene product.

The 97bis(phenylethynyl)benzene (XVI) was synthesized
by tha.methodrof Castro and Stevens using Qfdiiodobenzene
(XV) which was synthesized by the method of Friedman (14)
(Scheme II). The yield of product was 75% of purified
crystalline material which analyzed properly. The struc—
ture was confirmed by a four mole hydrogen uptake per mole
of hydrocarbon during atmospheric pressure hydrogenation
over Adam's catalyst. This is consistent with the presence

of two acetylene groups. The nmr spectra of both the

14

SCHEME II

(XIV)

F + - (XV)

 

 

 

H2 N2 I
i-amONO
”OH > ("2‘ I
c .J I

O O

 

 

 

 

 

 

 

'15
acetylenic and the saturated compounds were also consistent
with their assigned structures and thus served as additional
proof of structure.

‘Rearrangement of the 97bis(phenylethynyl)benzene (XVI)
to form the 1,4-dehydrobenzene product was attempted using
two different palladium reagents. Stirring a benzene solu-
tion of the acetylene with the benzonitrile complex of
palladium chloride formed no product (15). Changing to the
use of an acetonitrile solution with the acetonitrile com—
plex of palladium chloride, because of the weaker bonding
in this complex, was more productive. However, again no
useful product was obtained. Instead, along with a large
recovery of nitrile—palladium complex, only a quantity of
brown powder was obtained. Fractional crystallization gave
a number of fractions with various melting points. However,
all fractions gave essentially the same infrared spectrum,
and it was concluded that they were polymers of various
molecular weights.

Since molybdenum hexacarbonyl, molybdenum carbonyls
complexed with glycol ethers, and cyclopentadienyl cobalt
dicarbonyl have all been-reported to form tetraphenylcyclo-
butadiene from tolane (16), reactions were tried between
27bis(phenylethynyl)benzene (XVI) and each of these complex—
ing agents. Again, as in the palladium cases, the only
products, other than starting acetylene, were polymers of

various molecular weights.

16

Diiron enneacarbonyl is also well known for its ability
to react with certain precursors to form stabilized cyclo-
butadienes (12). While the use of this compound has been
restricted to cases where it was first necessary to abstract
two chlorines, thus forming the cyclobutadiene, before com-
plexing with the resultant cyclobutadiene, there was no
reason to believe that it should not also function with
acetylenes in the manner of other metal carbonyls such as
molybdenum. However, even after stirring a reaction mixture
of diiron enneacarbonyl and gébis(phenylethynyl)benzene
(XVI) in benzene at 40°C. for 72 hours, only contaminated
iron carbonyls could be cryStallized from the reaction mix-
ture.

Refluxing 27bis(phenylethynyl)benzene (XVI) in benzene
with cobalt octacarbonyl for two hours was slightly more
successful, but still yielded only a mixture of starting
material and polymer.

It has been reported in the literature (17) that tolane
reacts with lithium in ether to form the 1,4-dilithio—
1,2,3,4-tetraphenyl-1,3-butadiene which slowly rearranges
to the 1,4-dilithio-1,2,3-triphenylnaphthalene. Reaction
of the dilithiobutadiene with cupric bromide apparently re—
sulted in the formation of tetraphenylbutadiene diradical,
but this species immediately dimerized to form the cyclo-
octatetraene instead of forming the cyclobutadiene.

However, when this lithium reaction was applied to

27bis(phenylethynyl)benzene (XVI), the reaction was found to

17

proceed extremely slowly. After 90 hours of contact time,
hydrolysis of the products yielded much unreacted starting
material, and only polymer could be isolated as a product.

Only one fact can be concluded from the consistent form—
ation of polymer from these many attempts to form the 1,4—
dehydrobenzene. It appears that the benzene ring, of which
the 3,4 double bond is a part, is preventing the system from
reacting properly. This fact is made clear when the geometry

of this system is examined.

 

First of all, the fact that the 3,4 double bond is part of

a benzene ring means that the two cis vicinal bonds will
extend at a 600 angle to each other. -With this geometry,
the two acetylene bonds are placed at such a large distance
from each other that they cannot easily achieve the pror-
bital overlap necessary for bond isomerization and formation
of the cyclobutadiene nucleus. This problem is further com—
plicated by the presence of the 3,4 double bond in a rigid

benzene ring, which prevents the bond deformation which is

18
necessary to bring the two acetylene bonds into close prox—
imity with each other.

The result of this is that rather than obtaining intra-
molecular reactions, intermolecular reactions take place to
give polymers which are probably highly cross—linked, since
the monomer is bifunctional.

In 1966, Mfiller, Sauerbier, and Heiss attempted to
prepare 1,3-diphenyl-1,4-dehydronaphthalene (XII) by di-
rect photolysis of 27bis(phenylethynyl)benzene (XVI) in
methylcyclohexane solution (18). They obtained, instead, a
fused azulene derivative. Hoping that sensitization might
aid the photochemical reaction, the reaction was repeated
using a Hanovia high pressure mercury lamp to photolyze an
acetone solution of 27bis(phenylethynyl)benzene (XVI). In-
stead of obtaining the desired dehydronaphthalene, however,
infrared analysis and melting points of the resultant pro-
ducts indicated that a free radical polymerization, probably
initiated by an acetoxy radical, had taken place.

Since it had become quite apparent that the acetylene
groups in 97bis(phenylethynyl)benzene (XVI) are too rigidly
oriented at too great a distance from each other to readily
cyclize to a cyclobutadiene system, it was decided to at-
tempt to form other non-benzenoid aromatic systems from
27bis(phenylethynyl)benzene (XVI) in which this condition
would not act as so great a hindrance.

The synthesis of diphenylcyclopropenone (III) from

tolane has been reported in the literature (7), and the

19
similarity between tolane and 27bis(phenylethynyl)benzene
(XVI) made it desirable to synthesize the dicyclopropenone
gfphenylene-bis(phenylcyclopropenone) (IV) from 27bis-

(phenylethynyl)benzene (XVI), and to examine its properties.

 

Also, it was hypothesized that such a dicyclopropenone
should rearrange under irradiation to form the substituted
cyclobutene dication, which fits the theoretical conditions
for aromaticity.

The initial attempt at forming the dicyclopropenone
gfphenylene-bis(phenylcyclopropenone) (IV) was by the method
of Seyferth and Damrauer using dichlorocarbene generated
from phenyl(trichloromethyl)mercury (Scheme III) (19). This
method of carbene generation had the advantage of requiring
no harsh reagents, and it produces easily removed by-products.
A cyclopropenone product was isolated, as indicated by the
characteristic infrared peaks of 1840 and 1620 cm-1, although
the presence of an acetylenic peak at 2200 cm.1 indicated
that only the monocyclopropenone had been formed (7). The

elemental analysis, however, was inconsistent with a simple

20

SCHEME.III

 

 

¢HgCC13 CHElEOt bu
K ..
Heat

 

 

 

 

21
monocyclopropenone. A mass spectrum of the product showed
mercury peaks as well as chlorine and chloromethyl peaks.
With this data in mind, the elemental analysis confirmed a
product composed of two molecules of (gephenylethynylphenyl)-
phenylcyclopropenone coordinated to one molecule of phenyl-
(trichloromethyl)mercury (XXII).

In an effort to increase the rate of decomposition of
the phenyl(trichloromethyl)mercury to dichlorocarbene, and
thus, increase the concentration of dichlorocarbene inthe
reaction mixture at any time, the Seyferth method using
phenyl(trichloromethyl)mercury with sodium iodide to gener-
ate dichlorocarbene was tried (20). However, only phenyl—
mercuriciodide and starting acetylene were isolated. Ap-
parently, the problem is confined to the lack of reactivity
of the acetylene and not to the slowness of formation of
the carbene.

As a possible means of overcoming this, the phenyl-
(trichloromethyl)mercury was added incrementally over 40
hours. However, only the same complexed (gfphenylethynyl-
phenyl)phenylcyclopropenone (XXII) product was isolable,
along with some starting material and other by—products.

A reaction of 27bis(phenylethynyl)benzene (XVI) with
dichlorocarbene, generated by reaction of chloroform with
potassium tertiary bUtDXide (21) gave only (prhenylethynyl-
phenyl)phenylcyclopropenone (XXI), as indicated by its
infrared spectrum which contained the characteristic peaks

of 2200, 1840, and 1620 cm-1; and by the elemental analysis.

22
Even when an excess of chloroform and potassium tertiary
butoxide were used, only the monocyclopropenone could be
isolated.

It is apparent that 95bis(phenylethynyl)benzene (XVI)
does not add two moles of dichlorocarbene readily. However,
exactly what type of association, or other factor, prevents
addition of the second mole of dichlorocarbene is unknown.

Moreover, no insight into the nature of this phenom-
enon could be gathered from the photochemical reactions of
this compound or its hydrochloride. Irradiation of (27
phenylethynylphenyl)phenylcyclopropenone (XXI) with a Hanovia
high pressure mercury lamp succeeded only in splitting out
carbon monoxide to reform 27bis(phenylethynyl)benzene (XVI).
Photolysis of the hydrochloride in methanolic HCl formed a
number of uncharacterizable products, none of which gave
infrared spectra consistent with the product which might
be expected from an intramolecular rearrangement to a benzo-

dehydrotropyliumwsalt.

EXPERIMENTAL
General Procedures and Apparatus

Melting points were determined on a Thomas Hoover
Capillary melting point apparatus and are uncorrected.

Infrared spectra were obtained on a Perkin—Elmer
237B Grating Infrared Specrophotometer. The sample
holders were sodium chloride cells or plates.

Ultraviolet spectra were obtained on a Unicam SP-800
Ultraviolet Spectrophotometer. The sample holders were
one centimeter quartz cells.

Nuclear magnetic resonance spectra were obtained on
a Varian A-60 high resolution spectrometer using tetra—
methylsilane as internal reference.

Microanalyses were performed by Spang Microanalytical

Laboratory, Ann Arbor, Michigan.

A. 1,6-diphenyl—3,4-dimethyl-1,5-hexadiyne—3,4-diol (v11)

 

To 24.4g.(1.0 mole) magnesium turnings in a flame dried
apparatus was added 15 ml. of a solution of 108.Gg.(1.0
mole) ethyl bromide in 250 ml. anhydrous ether. Once the
exothermic reaction had begun, 50 ml. ether was added and
the remaining bromide solution was added over one hour.
Following one further hour of refluxing on a steam bath,
102.1Zg(1.0 mole) phenylacetylene was added over 15 minutes.
After adding 100 ml. ether to obtain a solution, this solution

23

24
was refluxed for four hours. After cooling to 0°C. and
adding 300 ml of ether, a solution of 43.0g.(0.5 mole)
diacetyl in 200 ml. ether was added over 40 minutes to the
vigorously stirred solution. The mixture was stored at 0°C.
for three days, after which 200g. ice, followed by a mix-
ture of 759. coficentrated sulfuric acid and 200g. ice,
was added and the layers were separated. The aqueous
layer was extracted with ether (3 x 100 m1.). The ether
extracts were combined, dried over sodium sulfate, and
evaporated to leave an oily yellow-brown solid. This
solid was washed with carbon tetrachloride (4 x 100 ml.)
and the remaining solid was fractionally crystallized from
chloroform-petroleum ether (30 to 60°) to give 4.19g.
(3.36%) high melting isomer, m.p. 125°C., nmr (CDCl3):
T2.48(multiplet,10H), T6.81(singlet,2H), and 18.19(singlet,
6H); and 55.1g. (37.95%) low melting isomer, m.p. 115-7°c.,
v§§§13z 3550(broad), 3000, 2950, 2180, 1600, 1490, 1445.
1378, 1340, 1110, 1065, 945, 925, and 880 cm-1; nmr (cpc13):
T2.48(multiplet,10H), 16.79(singlet,2H), 18.17(singlet,3H).

and 18.25(singlet,3H).

B. ‘Dibenzoate of 2,3-diphenylethynyl-2,3-butanediol

 

To 1.45g.(0.005 mole) 2,3-diphenylethynyl-2,3-butane-
diol (m.p. 117°C.), partially dissolved in 0.79g.(0.010
mole) pyridine, was added 1.4lg.(0.010 mole) benzoyl chlor-
ide. (After heating the mixture on a steam bath for one

hour, it was poured into 25 ml. ice water. The resultant

25

oil was extracted into 5 ml. chloroform. The chloroform
was evaporated and the resultant oil was partially redis—
solved in ether, leaving behind 0.9599 pale gold crystals
(m.p. 70°C.). The ether solution was evaporated and the
resultant oil was redissolved in methanol. The addition
of water precipitated 0.38g. tan solid. Recrystallization
from methanol gave 0.30g. (12.0%), m.p. 163-5°C., nmr

(CCl4): 12.4(mu1tiplet,20H), and T7.65(singlet,6H).

C. 1,3-dioxa-2-ethoxy-4,5-dimethyl—4,5-diphenylethyn l-
gyclopentane, Attempted

 

One average sized crystal of paratoluenesulfonic acid
was added to a mixture of 1.45g.(0.005 mole) 2,3-diphenyl-
ethynyl-2,3—butanediol (m.p. 117°C.) and 3.70g.(0.025 mole)
ethyl orthoformate. The clear solution was allowed to
stand at room temperature for 180 hours, and the volatile
products were then aspirated off. The remaining colorless
oil was chromatographed on silicic acid with an eluent of
chloroform. A yield of 1.5g. (85.9%) colorless oil was ob—
tained. vgzit: 2980, 2930, 2250, 1600, 1485, 1440, 1375,
1325, 1270, 1100(broad), 985, 760, and 690 cm-1; nmr (cc14):
12.5(multiplet,10H), T3.88(doublet, J=2cps.,1H), T6.17(pen-
tuplet, J-7.5cps.,2H), 18.09, 18.15, 18.29, 18.34Qcomplex

multiplet,6H), and T8.72(triplet, J=7.5cps.,3H).

26

D. 2,5-dioxa-3,4-dimethyl—3,4-diphenylethynylcyclopentanone

To a solution of 1.45g.(0.005 mole) 2,3-diphenylethynyl-
1,3-butanediol (m.p. 117°C.) and 1.11g.(0.011 mole) triethyl—
amine in 10 ml. anhydrous ether, stirred at 0°C., was
slowly added 1.09g.(0.01 mole) ethyl chloroformate. The re-
sultant mixture was stirred at 0°C. for five minutes, and
at room temperature for 24 hours. Filtration removed 0.24g.
white solid (m.p. 250-4°C.), and the filtrate was evaporated
to dryness. »The remaining solid was redissolved in 10 ml.
ether and a small amount of remaining solid was filtered
off. -Another evaporation produced 1.409. residue, m.p.
101-6°C.. Chromatography on silicic acid with a chloro-

form eluent gave 0.35g. (22.0%) white solid, m.p. 90°C.,

VCHC13:
max

1115, 1075, 1055, and 1000 cm’l; nmr (cc14): 12.42(multiplet,

3000, 2235, 1810, 1485, 1445, 1375, 1250(broad),

10H), T7.99(singlet,3H), and 78.16(singlet,3H).

E. 1,3-dioxa-2-phenyl-4,5-dimethyl-4,5-diphenylethynyl-

cyclogentane

 

Two crystals of paratoluenesulfonic acid were added to
a solution of 2.90g.(0.01 mole) 2,3—diphenylethynyl—2,3-
butanediol (m.p. 117°C.) and 1.06g.(0.01 mole) freshly
distilled benzaldehyde in 50 ml. dry benzene contained in
the flask of a Soxhlet extractor. Calcium hydride was
placed in the extractor thimble and the reaction mixture was

refluxed through the extractor for three hours. The

27
resultant green solutiOn was treated with solid sodium car-
bonate to neutralize the acid, and the solvent was evaporated
to leave 3.90g. green oil This oil was chromatographed
on silicic acid with chloroform eluent to give 1.70g. orange
semisolid oil. Crystallization from chloroform-petroleum
ether (30-60°) gave 0.65g. (17.2%) light tan crystals, m.p.
111-20c., v§§§l3z 3000, 2245, 1600, 1485, 1445, 1375, 1365,
1100(broad), and 975 cm-1; nmr (CHCla): 72.40(multiplet,
15H), 13.53(singlet),1H), and 77.95 and 18.00(two singlets,

6H).

F. Reaction of 1,6-diphenyl-3,4-dimethyl-1,5-hexadiyne-
3,4-diol with triphenylphosphine dibromide

To a nitrogen protected refluxing solution of 3.36g.
(0.0128 mole) triphenylphosphine in 25 ml. carbon tetra-
chloride was added 2.05g. (0.0128 mole) bromine in 8.5 ml.
carbon tetrachloride over 15 minutes. The resultant cream
suspension was refluxed one hour, cooled and filtered, and
the collected solid was added unweighted to 15 ml. nitrogen
purged dimethylformamide. While stirring the resultant
orange suspension, a solution of 1.85209.(0.0064 mole) 1,6-
diphenyl-3,4-dimethyl-1,5-hexadiyne-3,4-diol (m.p. 117°C.)
in 10 ml. dimethylformamide was added dropwise over 15
minutes. This solution was stirred at room temperature for
48 hours. The solvent was evaporated at reduced pressure
and the remaining tan mush was purified by elution from a

silicic acid chromatography column with a 10:2 carbon

28

tetrachloride:benzene eluent to yield 1.159. yellow solid.
the solid was recrystallized from ethanol to produce tan
crystals of 4-bromo-1,6-dipheny1-3-methylidene-4—methyl-;
1,5-hexadiyne, m.p. 122-4°C., Analysis: C20H15Br requires
0, 71.055; H, 4.56%; Found: 0, 71,765: H, 4.46%; viii:
3050, 2200, 1305, 1205, 905, 750, and 680 cm-1; nmr (cc14):
12.69(multiplet,10H), 15.61(singlet,2H), and.T7.77(sing-

let,3H).

G. 27bis(phenylethynyl)benzene (XVI)

 

A mixture of 23.59.(0.143 mole) cuprous phenylacetylide,
235529.(0.072 mole) gfdiiodobenzene, and 350 ml. pyridine
were refluxed in a nitrogen atmosphere for 18 hours, and
the resulting mixture was poured into 1500 ml. water. After
extracting with ether (3 x 500 ml.), the combined ether ex-
tracts were washed successively with 10% hydrochlorid acid.
(3 x 250 ml.), 5% sodium bicarbonate (3 x 250 ml.), and
water (3 x 250 ml.). The Ether solution was dried over
magnesium sulfate and was evaporated to leave 20.559. dark
oil. This oil was crystallized from methanol to give 15.59.
brown crystals. The crystals were redissolved in 100 ml.
methylene chloride and this solution was decolorized by
filtering through a thin pad of Norit. The solution was

then evaporated and the product was recrystallized from

VCHC13:

methanol to give 15.09. (75.4%) white crystals. max

3050, 3000, 2200, 1600, 1485, and 1435 cm‘l; xfizgfl (log a):

221.5(4.39), 302(4.30), 312(4.32), and 331(3.95) mu.;

29
Analysis: C22H14 requires C, 94.93%; H, 5.07%; Found:

c, 94.92%; H, 5.30%.

H. 27bis(2-phenylethyl)benzene (XVII)

 

Into a 50 ml. round bottom flask assembled in an atomos-
pheric pressure micro hydrogenation apparatus was placed
40mg. Adam's catalyst covered by minimal methanol. 'After
prereducing the catalyst, 0.5576g.(0.002003 mole) 27bis-
(phenylethynyl)benzene dissolved in 35 ml. methanol was
added and the hydrogenation was allowed to proceed until
no further hydrogen uptake was noted. Hydrogen uptake was
188.8 ml. (105% of theoretical at S.T.P.). After removal
of the catalyst by filtration, the methanol was evaporated
to leave an oil. Vacuum distillation gave 0.49909. (89.5%)

neat

of pale yellow oil, b.p. 144-50/0.175 mm., Vmax . 3055,

3025, 2925, 2855, 1600, 1495, 1450, 750, and 700 cm-1;

1::3H:(1og s): 242.5(2.60)s, 248(2.76)s, 253.5(2.87)s,
259(2.93), 261(2.93), 264.5(2.92), 268(2.86), and 271.5(2.69)s
mu.; nmr (CC14): 12.89 and 12.97(two singlets,14H), and

T7.19(singlet,8H).
J. Bis(benzonitrile)palladium II chloride

To 30 ml. freshly distilled benzonitrile was added
2.10309.(0.0118 mole) palladium chloride, and this mixture
was stirred at 120°C. for one hour. The resultant red-
brown suspension was filtered hot, and the filtrate was

then cooled to precipitate an orange-yellow solid which

30

was filtered off. -The filtrate was recombined with the
previously isolated residue, and this mixture was again
heated to 120°C. for one hour. Again, the hot solution
was filtered from a trace of undissolved material and was
then cooled to precipitate more product. ‘The filtrate was
then diluted with petroleum ether (30-60°), and the solution
was chilled to yield a further crop of product. The com-
bined product fractions were dried over phosphorous pent-
oxide in a vacuum oven at room temperature for 18 hours

. C

to yield 3.90509. (86.1%), vm§§13;3450(broad), 2990, 2280,

2225, 1600, 1485, 1450, and 1200(broad) cm‘1
K. Bis(acetonitrile)palladium II chloride

To 30 ml. freshly distilled acetonitrile was added
2.11409. (0.0119 mole) palladium chloride, and this mixture
was refluxed for one hour. The mother liquor was then de-
canted from the undissolved solid. »This liquor was cooled
and the resultant precipitate was filtered off. The fil—
trate was then returned to the reaction flask, and the re—
action mixture was refluxed for another hour. -Again the
mother liquor was decanted, cooled and filtered to yield
more product. ‘This process was repeated until no more
solid remained undissolved. rThe isolated yellow-tan solid
was then dried over phosphorous pentoxide in a vacuum oven
at toom temperature overnight to yield 2.81419. (91.0%)

1

complex. vfizfiachlorobutadlene: 2960, 2900, and 2300 cm- 7

Nujol. -1
vmax . 1150, 1020, and 725 cm .

31

L. Reaction of 27bis(phenylethynyl)benzene with dichloro-

 

carbene;
Method I.

A solution of 2.789.(0.010 mole) 27bis(phenylethynyl)
benzene and 7.929.(0.020 mole) phenyl(trichloromethyl)mercury
in 50 m. anhydrous nitrogen purged benzene was refluxed
under nitrogen for 48 hours. After cooling and filtering
off 4.909. phenyl mercuric chloride, 30 ml. 95% ethanol
was added, and the mixture was refluxed for 30 minutes.

The resultant solution was evaporated to dryness under re-
duced pressure to leave 5.809. red oil. This oil was
chromatographed on silicic acid using an eluent of 10:12
benzene:methylenechloride to remove unreacted starting_
material fractions and by-products. ;An eluent of chloro—
form removed the product fraction as 3.399. golden crystals.
Crystallization from methylene chloride-n, hexane gave

1.10 g. (21.8% as 2(C23H140) to 1(C7H5HgC13) complex),

m.p. 146-7°C., vggil3: 3000, 2200, 1840, 1620, 1490, 1470,
1440, 1330, and 1235 cm-1; 13:2H(log e); 221(4.43), 234(4.36),
260(4.37), 273(4.47), 279.5(4.42), 302(4.24), 331(3.93),

and 350(3.66) mu.; nmr (CC14): 12.6(broad multiplet);
Analysis: .C53H33HgC1302 requires C, 63.10%; H, 3.30%; Cl,
10.5%; Found: C, 63.55%; H, 3.58%; Cl, 9.05%.

32

M. *Reaction of 27bis(phenylethynyl)benzene with dichloro-

 

carbene;
Method II.

A solution of 1.19g.(0.01 mole) chloroform in 10 ml.
benzene was added dropwise during one hour to a rapidly
stirred ice cold suspension of 3.369.(0.03 mole) potassium
tertiary butoxide and 1,39g.(0.005 mole) 27bis(phenyl—
ethynyl)benzene in 25 ml. benzene protected by a nitrogen
atomodphere. When the addition was complete, the ice bath
was removed and the mixture was stirred at room temperature
for 30 minutes. The reaction mixture was then poured into
50 ml. water and the layers were separated. The aqueous
layer was extracted with benzene (3 x 10 ml.) and the com-
bined benzene extracts were washed with water (2 x 10 ml.).
After drying the benzene solution over magnesium sulfate,
the solvent was evaporated to leave 1.709. red oil. After
dissolving this oil in 20 ml. anhydrous ether, hydrogen
chloride gas was bubbled in to saturate the solution, and
the solution was chilled to 0°C. overnight. Filtration
isolated 0.359. off-white solid which was partially dis-
solved in a mixture of 15 ml. water and 10 ml. ethanol.
The mixture was made basic with sodium carbonate, and the
mixture was stirred vigorously for 30 minutes to convert
the salt form to the free cyclopropenone. The resulting
mixture was extracted with ether (2 x 10 ml.), the ether
extracts were dried over magnesium sulfate, and the ether

was evaporated to leave 0.309. off-white crystals.

33
Recrystallization from methylene chloride-2, hexane yielded
0.2959. (19.3%) white crystals of phenyl(2—phenylethynyl-
phenyl)gyclopropenone, m.p. 107-8°C., VCHC13: 3000, 2200,

max
1840, 1620, 1485, 1470, 1435, 1330, and 1230 cm'l; xfing
(log a): 221(4.37), 232.5(4.31), 260(4.40), 273(4.51),
279.5(4.45), 301(4.30), 330(4.04), and 347(3.94) mu;
Analysis: C23H14O requires C, 90.17%; H, 4.60%; Found:

c, 89.62%; H, 4.83%.

SPECTRA

‘34

35

come

 

.TGTNGTQAHMCMguoamsonmvaQhW mo Enuuummm UmnmumcH

H EU mocmsqmnm

coca ooem ooom

 

 

 

 

 

 

 

 

 

 

.ON

r¢¢

vow

 

:ow

 

 

F 1

 

 

 

OOH

uorssrmsuexm aueo zed

36

0mm

 

.mamusmnnahcmnumahcmnmvmanta mo Eduuowmm umHOH>MHuHD

mGOHUHEHHHflE,£umc0Hm>m3.

oLom Mme. !. cow.

 

 

o

1w.o

I I

«n

(o
eoueqzosqv»

I
N
H

 

 

37

 

 

.wcmuconfiHanuoamcmnmlmvmflntm mo Ednuommm.omnmnmcH

H EU mocwsumum

OOON ooom ooov

- p b p p p - rb

 

 

 

 

 

16m

.bw

be

 

 

 

.om

 

 

 

 

- . 63

 

uorssrmsuexm quao Isa

38

on”

 

.mswncmnAngumamcmnmlmvmflflhm mo Esuuommm DTHOH>mHuHD

mGOHoHEHHHaE numcoaw>mz

com
-

p)

0mm

(-

CON

 

 

 

 

 

.NA

6..“

 

 

eoueqxosqv

39

 

 

1i;

.mzosmmoumoaomoamcwnmAHmconmamcmgumamcmgmlov mo Eduuommm.ooumumcH,

«I80 mocmswmum
coon coca ooom . . 000%

o

D P D (P

 

row

Pow

 

 

 

e .8

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

., ooa

uorssrmsuexm auea Isa

 

.mcosmmonmoaowwA>cm£mAHmcmnmamcmnumamcmsmuwd mo Eduuowmm u0H0H>MHuHD

TGOHUHEHHHHE apmcwam>mz
0mm oom mrm .. oou

- . . p .

 

40

 

°§
O

)

 

o.N

 

aoueqxosqv

10.

11.

12.

13.

14.

LITERATURE CITED

J. D. Roberts, A. Steitwieser, and C. M. Regan,
J. Am. Chem. Soc., 18, 4579 (1952).

M. J. S. Dewar, and G. J. Gleicher, J. Am. Chem. Soc.,
81, 685 (1965).

H. S. Lee, Chemistry (Taipei), 22 (1963).

R. S. Berr , J. Clardy, and M. E. Schafer, Tet. Letters,
1003 (1965 .

R. D. Stevens, and C. E. Castro, J. Org. Chem., 88,
3313 (1963).

E. Mfiller, M. Sauerbier, and J. Heiss, Tet. Letters,
2473 (1966).

H. W. Whitlock Jr., and D. E. Sandvick, J. Am. Chem.
s_oc_.. £51, 4525 (1966).
C. F. Castro, E. J. Gaugham, and D. C. Owsley, J. Org.
Chem., 8;, 4071 (1966).

D. Se ferth, and R. Damrauer, J. Org.-Chem., 88, 1660
1966 .

R. Breslow, T. Eicher, A. Krebs, R. A. Peterson, and

J. Possner, J. Am.-Chem. Soc., 81, 1320 (1965).

D. N. Kursanov, M. E. Vol‘pin, and Yu.-D. Koreshkov,

J. General Chem. USSR., 88, 2855 (1960).

H. W. Post, The Chemistry of the Aliphatic Orthoestegg,
Reinhold Publishing Co., New York, N.Y., 1943, pp. 117.

P. M. Maitlis, and F. G. Stone, Proc. Chem. Soc., 330
(1962).

F. J. Wilson, and W. M. Hyslop, J. Chem. Soc., 123,
2612 (1923).

<W. Ried, H. J. Schmidt, and A. Urschel,-Chem. Ber.,

91,. 2472 (1958) .

G.-F. Emerson, L. Watts, and R.-Pettit, J. Am. Chem.
Soc., 81, 131 (1965).

G. A. Wiley, B. M. Rein, and R. L. Herschkowitz,-Tet.
Letters, 2509 (1964).

L. Friedman, and F. M. Logullo, Angew. Chem., 11, 217
(1965).

41

15.

16.

17.

18.

19.

20.

21.

‘Soc., 88, 4406 (1961).

42'

P. M. Maitlis, D. Pollock, M. L. Games, and W; J.
Pryde, Can. J. Chem., 88, 470 (1965).

W; Huebel, and R.-Merenyi, J. Organometallic Chem.,
8, 213 (1964).
J. L. Boston, D. W. A.-Sharp, and G. Wilkinson, 8;

‘Chem. Soc., 3488 (1962).

L. I. Smith and H. H. Hoehn, J. Am. Chem. Soc., 88,
1184 (1941). '

AF. C. Leavitt, T. A.-Manuel, G. Johnson, L.-U;

Matternas, and.D. S. Lehman, J. Am. Chem. Soc., 88,
5099 (1960).
E. H. Braye, W. Hflbel, and I. Caplier, J. Am. Chem.

E.-Mflller, M. Sauerbier, and J. Heiss, Tet. Letters,
2473 (1966).

D.-Seyferth, and R. Damrauer, J. Org. Chem., 88, 1660
((1966)..' ‘V 1 . .
D. Seyferth, M. E. Gordon, J. Mui, and J. M. Builitch,
J. Am. Chem. Soc., 88, 959 (1967).

D.-N.-Kursanov, M. E. Vol'pin, and Yu. D. Koreshkov,
J. General Chem. USSR., 88, 2855 (1960).