Group Title: study of some salts of 2,2-diphenylcyclo-propyldiazonium hydroxide
Title: A Study of some salts of 2,2-diphenylcyclo-propyldiazonium hydroxide
CITATION PDF VIEWER THUMBNAILS PAGE IMAGE ZOOMABLE
Full Citation
STANDARD VIEW MARC VIEW
Permanent Link: http://ufdc.ufl.edu/UF00097948/00001
 Material Information
Title: A Study of some salts of 2,2-diphenylcyclo-propyldiazonium hydroxide
Alternate Title: Some salts of 2,2-diphenylcyclo-propyldiazonium hydroxide
Physical Description: iii, 74 l. : illus. ; 28 cm.
Language: English
Creator: Tandy, Thomas King, 1937-
Publication Date: 1964
Copyright Date: 1964
 Subjects
Subject: Salts   ( lcsh )
Organometallic compounds   ( lcsh )
Chemistry thesis Ph. D
Dissertations, Academic -- Chemistry -- UF
Genre: bibliography   ( marcgt )
non-fiction   ( marcgt )
 Notes
Thesis: Thesis - University of Florida.
Bibliography: Bibliography: l. 72-73.
Additional Physical Form: Also available on World Wide Web
General Note: Manuscript copy.
General Note: Vita.
 Record Information
Bibliographic ID: UF00097948
Volume ID: VID00001
Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: alephbibnum - 000423926
oclc - 11032454
notis - ACH2331

Downloads

This item has the following downloads:

PDF ( 2 MBs ) ( PDF )


Full Text











A STUDY OF SOME SALTS OF

2,2-DIPHENYLCYCLOPROPYLDIAZONIUM

HYDROXIDE
















By
THOMAS KING TANDY, JR.


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











UNIVERSITY OF FLORIDA


December, 1964













ACKNOWLEDGMENTS


The author wishes to express his appreciation to

the members of his supervisory committee for their many

helpful suggestions and constructive criticism.

He is especially indebted to Dr. W. M. Jones,

director of this research, for his technical assistance

and advice. Moreover, during the course of this work,

he has been an unending source of encouragement,

inspiration and friendship.

The author would also like to thank his wife, Rhoda,

for her moral support, devotion, and unselfish contribution

of time, energy, and assistance.

The U. S. Army (Durham) and the National Science

Foundation are also to be acknowledged for providing

financial assistance.













TABLE OF CONTENTS

Page

ACKNOWLEDGMENTS . . . . . . . . ... ii

Chapter

I. THE MECHANISM OF THE LITHIUM ETHOXIDE INDUCED
CONVERSION OF N-NITROSO-N-(2,2-DIPHENYLCYCLO-
PROPYL)UREA TO 2,2-DIPHENYLDIAZOCYCLOPROPANE 1

Introduction . . . . . . . . 1

Results and Discussion . . . . . . 4

II. THE PREPARATION AND PROPERTIES OF SOME SALTS
OF 2,2-DIPHENYLCYCLOPROPYLDIAZONIUM HYDROXIDE. 17

Introduction . . . . . . . . 17

Results and Discussion . . . . . . 18

III. EXPERIMENTAL . . . . . . . .. 32

IV. SUMMARY. . . . . . . . . ... 70

LIST OF REFERENCES. . . . . . . . .. 72

BIOGRAPHICAL SKETCH . . . . . . . ... 74


iii













CHAPTER I


THE MECHANISM OF THE LITHIUM ETHOXIDE INDUCED CONVERSION
OF N-NITROSO-N-(2,2-DIPHENYLCYCLOPROPYL)UREA TO 2,2-DI-
PHENYLDIAZOCYCLOPROPANE


Introduction


As early as 1902 direct evidence was presented for

the course of von Pechmann's1 hydroxide induced conversion

of nitrosocarbamates to diazoalkanes when Hantzsch and
2
Lehmann actually isolated the potassium salts of methyl

and benzyl diazonium hydroxide (I) and showed that

treatment with water gave the corresponding diazoalkanes.

N=o
Rconc. 00
RCH2-N-C-O-Et -conc RCH2-N=N-O K
[ KOH
O I



H20



R = H, C6H5 R-CH-N


This scheme was expanded (Scheme 1) in 1955 by

Gutsche and Johnson as a result of a thorough study of the

methoxide induced conversion of a series of N-nitroso-N-

benzylcarbamates to the corresponding phenyldiazomethanes

1








2

(which were trapped with cyclohexanone).

N=O
I 0 0
(1) ArCHI2-N-CO2Et + OCH3 ArCH2N=N-0 + CH3-0-CO2Et



(2) ArCH2-N=N-O + CH30H -- - ArCH2N=N-OH

Scheme 1

0 0
(3) ArCH2-N=N-OH + OCH3 -- ArCH-N=N-OH





II

products -U ArCtN2



The evidence provided for the above scheme can be

summarized as follows:

1. The nitrosocarbamate is stable to methanol but

decomposes when potassium carbonate is added.

2. The potassium salt of benzyl diazonium hydroxide

has been isolated from reaction of ethyl N-

nitroso-N-benzylcarbamate with concentrated

potassium hydroxide.

3. Potassium benzyldiazotate reacts readily with water

or ethanol to give phenyldiazomethane and benzyl

alcohol or benzyl ethyl ether.











Although Gutsche and Johnson did not present direct

evidence for attack of the methoxide on the carbonyl to

give methyl ethyl carbonate, the isolation by Bollinger,

Hayes and Siegel of 67 per cent methyl ethyl carbonate

from the base induced (one equivalent of K2CO in MeOH)

conversion of N-nitroso-N-cyclohexylcarbamate (II) to

diazocyclohexane presents an excellent precedent for the

first step in the Gutsche and Johnson scheme.

N=O

N-C-0C2H5
| + K2CO3 + CH30H CH30-C02C2H5
0
67%
II


The general ideas of this scheme for the formation

of diazoalkanes have also been applied to the reaction of

N-nitrosoamides with base. Although there has not been

reported as thorough a study on the latter aspects of the

reaction (equivalent to steps 2 and 3 in Scheme 1), there

is ample evidence that the reaction of nitrosoamides with

alkoxide in alcohol proceeds via alkoxide attack on the

carbonyl carbon.5

By analogy, it has been natural to assume that N-

nitrosoureas also react with base by way of attack of the

base on the carbonyl carbon.6'7 This reaction path has










been most explicitly presented (Scheme 2) and discussed by

Applequist and McGreer6 for the ethoxide induced conversion

of N-nitroso-N-cyclobutylurea to diazocyclobutane.


i N=N-0


N=O
I
. N-CO-NH


0
II
+ H2N-C-OR


ROH


Scheme 2


ROH + OH


GO
+ N2
+


G
RO
*^----


Results and Discussion


In this section, data are presented that have

bearing on the course of the reaction of N-nitroso-N-(2,2-

diphenylcyclopropyl)urea with lithium ethoxide and with

pyrrolidine.

Reaction of N-nitroso-N-(2,2-diphenylcyclopropyl)-
8
urea with lithium ethoxide.--During the investigation on

the lithium ethoxide (in ether or hexane) induced con-

version of N-nitroso-N-(2,2-diphenylcyclopropyl)urea (III)

to 2,2-diphenyldiazocyclopropane (IV),8 two pieces of


RO













N=O

I-C-NH2

H 0



III


+ LiOEt


information were obtained that exclude the Applequist and

McGreer6 suggestion as a possible path for this reaction.

In the first place, less than 1 per cent of ethyl

carbamate (V) is formed in the above reaction. This was

demonstrated by assaying the reaction mixture for ethyl



N=O


+ LiOEt


0

Et-O-C-NH2


carbamate and by independently demonstrating that it is

stable to the reaction conditions. Secondly, lithium 2,2-

diphenylcyclopropyldiazotate (VI)* was isolated by an

alternate synthesis and found to be totally unaffected by


See Chapter II.


-C-NH2

I


--17 `










reaction conditions that lead to efficient conversion of


G @
N=N-O Li

H

VI

the nitroseurea to the diazocyclopropane.* For example,

whereas conversion of the nitrosocyclopropylurea to the

diazocyclopropane in ether at 0 is complete in 40 minutes,

the diazotate is stable for days under these conditions

(lithium ethoxide and lithium hydroxide in ether). In

fact, it is even recovered quantitatively from a homo-

geneous solution of lithium ethoxide and lithium hydroxide

in a mixture of tetrahydrofuran and water after 16 hours

at 0 followed by 4 hours at room temperature.+

For the base induced formation of 2,2-diphenyl-

diazocyclopropane, these observations quite effectively


Lithium 2,2-diphenylcyclopropyldiazotate is in fact
prepared under conditions more severe than reaction con-
ditions; for example, the lithium diazotate precipitates
from an ether solution of the pyrrolidinium 2,2-diphenyl-
cyclopropyldiazotate at room temperature upon addition of
lithium ethoxide.
+The diazotate is converted to 1,1-diphenylallene, 8
the decomposition product of 2,2-diphenyldiazocyclopropane,
upon heating with water. This shows that, although the
diazotate is not an intermediate, in the conversion of the
nitrosourea to the diazocyclopropane, it is capable of
conversion to the diazocyclopropane. This is consistent2
with the conversion of other diazotates to diazoalkanes.











exclude the reaction path as presented in previous papers

(Schemes 1 and 2). However, they do not, a priori,

exclude the diazotate as an intermediate. Two other

reasonable modes for decomposition of the diazotate are

reaction with ethyl carbamate and attack of the diazotate

on the carbonyl of a molecule of the nitrosourea. In both

cases, a diazourethan would be formed that could decompose

spontaneously to the diazocyclopropane.' '10 Both of
N=O 0

R-N=N-O + R-N-C-NH2 (or H2N-C-OEt)



o

R-N=N-O-C-NH2 + R-N=N-O (or OEt )



Rapid N2



R = 2,2-diphenylcyclopropyl


these possibilities were excluded by exposing the diazo-

tate to ethyl carbamate and to the nitrosourea under


For many examples of the thermal conversion of
diazoesters, see E. H. White and R. J. Baumgarten, J. Org.
Chem. 29, 2070(1964); E. H. White and C. A. Aufdermarsh,
J. Am. Chem. Soc., 83, 1174(1961); E. H. White, ibid., 77,
6013(1955); A. StrelEwieser, Jr., and W. D. Schaieffr,
ibid., 79, 2893(1957); A. R. Huisgen and R. Reimlinger,
Ann., 575, 183(1956).










reaction conditions. In neither case did reaction occur.

All of these observations lead to a single con-

clusion: lithium 2,2-diphenylcyclopropyldiazotate is not

an intermediate in the lithium ethoxide induced conversion

of N-nitroso-N-(2,2-diphenylcyclopropyl)urea to 2,2-di-

phenyldiazocyclopropane. Furthermore, these data suggest

that the reaction does not even involve attack of the base

at the carbonyl carbon of the nitrosourea.

As a possible alternate point for attack of the

base, the nitroso group is an obvious and not unprecedented

candidate. For example, Bollinger, Hayes and Siegel '11

found that two of the products from the reaction of N-

nitroso-N-cyclohexylcarbamate with potassium carbonate in

methanol were methyl nitrite and N-cyclohexylcarbamate.

These products obviously arose from attack of the methoxide
12
ion on the nitroso nitrogen. Backer and de Boer2 found



N=O H

N-C-OEt N-C-OEt

0 + OCH3 --O


+ CH 3-ONO











that treatment of N-methyl-N-nitroso-p-toluenesulfonamide

with piperidine gave only nitrosopiperidine and N-methyl-p-

toluenesulfonamide. Again, attack of base on the nitroso

nitrogen was obviously involved.



N=0 H

ArSO2-N-CH3 + + ASO2-N-CH3

N N
I I
H N=0


In the case of the nitrosocyclopropylurea, it is

suggested that the reaction proceeds by initial attack of

the base on the nitroso nitrogen followed by ring closure

to give the l-oxa-2,3-diazocyclobutane anion (VII) rather

than elimination of the urea anion. This ring closed

anion (VII) could then collapse by either of two routes.

Ring opening between carbon and nitrogen followed by

elimination of ethoxide would give the diazocarbamate

(VIII) that could decompose rapidly to give the diazo-

cyclopropane.910 Ring opening in the same way but with


This base induced conversion is identical with the
heat induced conversion of nitrosoureas, carbamates, etc.,
to diazocarbamates, esters, etc. See Ref. 15, pp. 153-
159; see also Ref. 10.










elimination of the carbamate anion would give rise to the

diazoether (IX) which could conceivably collapse to the

diazocyclopropane.


N=0 EtO-N-0

IN-C-NH2 N-C-NH2 N=N-OEt
I Oft -02CNH2
1 o 0


VII IX


0 0
-OE t OEt




0

)N=N-0-C-NH2 -N2

> 2


VIII



Several experiments were run in an attempt either

to add credence to the suggested reaction scheme or help

distinguish between the two suggested modes of collapse.

Although neither venture has been entirely successful,

some isolated observations were made that should be

mentioned.











In the first place, it was obvious that attempts

should be made to synthesize the diazoether to determine

its stability under the conditions of the reaction. These

attempts included treatment of the lithium and silver

salts of the diazotate with both methyl and ethyl iodide.13

In each case, reaction with the iodide was extremely slow

and, although traces of 1,1-diphenylallene (the decompo-

sition product of the diazocyclopropane) were found, it

was not possible to conclude anything about the stability

of the ether under the typical reaction conditions.

In an attempt to trap an intermediate analogous to

the diazoether, the nitrosourea was treated with

pyrrolidine. Although the large part of this reaction

proceeds via carbonyl attack (see next section), when the

reaction was run in hydrocarbon solvents, small quantities

of the triazine (X) were isolated (maximum yield 19%).

This can be pictured as resulting from a reaction path

analogous to that which would give the diazoether (IX).

Furthermore, it was demonstrated that this product was

not a result of the


N=0

N-C-NH2 hexane N=N-N


197. X










reaction of 2,2-diphenyldiazocyclopropane with pyrrolidine

by decomposing the pyrrolidinium diazotate (see the next

section) in the presence of pyrrolidine. Although it is

not felt that this observation can be taken as direct

evidence for the suggested mechanism involving attack of

the ethoxide ion on the nitroso nitrogen, it is certainly

consistent with this suggestion.

In an attempt to gain some insight into the source

of the apparent anomaly in our mechanism, a preliminary

extension of these investigations was carried out. N-

nitroso-N-(2,2-diphenylcyclopropyl)benzamide (XI) was

synthesized and treated with lithium ethoxide in ether.

It was found that, under these conditions, a high yield of

1,1-diphenylallene was formed but that only 2.9 per cent of

ethyl benzoate (from carbonyl attack) was produced and 68

per cent lithium benzoate (from nitroso attack) was formed.

Again, ethyl benzoate was independently shown to be stable

to the reaction conditions.


N-0

N-C-Q 0G
i1 + LiOEt ---- QCO2Et + OC02 Li
0
o Q 2.9% 68%

+ ,C=C=CH2











It was also found that the lithium ethoxide induced

conversion of the nitrosocyclopropylurea to 2,2-diphenyl-

diazocyclopropane was completed in about one minute when

anhydrous ethanol was the solvent. Furthermore, potassium

2,2-diphenylcyclopropyldiazotate was quantitatively

recovered after stirring six hours under these conditions.

These two results make one wonder if this mechanism

is unique to the 2,2-diphenylcyclopropyl-lithium ethoxide

system.

Finally, it was found that the nitrosourea reacts

with lithium hydroxide in ether to give almost 80 per cent

of the calculated nitrogen and a mixture of lithium 2,2-

diphenylcyclopropyldiazotate, lithium carbamate and 1,1-

diphenylallene. Because carbonyl carbon attack initially

would give the lithium diazotate which is stable to lithium

hydroxide under the reaction conditions, the allene

probably arose from attack of the hydroxide on the nitroso

nitrogen to give the diazocyclopropane which then decomposed

under the reaction conditions.


N=O HO-N-0

N-C-NH2 + LiOH ----2 N2
II 0N2




c c-N2

C=C=CH2










Reaction of N-nitroso-N-(2,2-diphenylcyclopropyl)-

urea with pyrrolidine.--It is interesting that, despite the

fact that lithium ethoxide (and hydroxide) attack on the

nitrosourea apparently proceeds almost exclusively at the

nitroso group, attack of pyrrolidine proceeds almost ex-

clusively at the carbonyl carbon. For example, treatment

of the nitrosourea with pyrrolidine in ether at 0 (in

the presence of magnesium sulfate) gave rise to a quanti-

tative yield of N-pyrrolidylcarboxamide (XII). When this

reaction was run in hydrocarbon solvents, there was formed

slightly less of the N-pyrrolidylcarboxamide. This is

believed to have resulted from formation of the triazine X.

Considering the mild reaction conditions, this

product most certainly arose from attack of the pyrrolidine

on the carbonyl carbon. Furthermore, when ether was the

solvent, there was also isolated about 40 per cent of the

pyrrolidinium diazotate (XIII). However, in addition to

the N-pyrrolidylcarboxamide and the diazotate, there was

invariably noted about 60 per cent nitrogen evolution in

ether and 80-90 per cent in pentane. In a hydrocarbon

solvent, however, pyrrolidinium 2,2-diphenylcyclopropyl-

diazotate was not a product. Examination of the reaction

mixtures showed that the rest of the starting nitrosourea

could be accounted for by 1,1-diphenylallene. As evidence

that the allene resulted from 2,2-diphenyldiazocyclopropane










8
and 2,2-diphenylcyclopropylidene in the normal fashion,

it was shown that the decomposition with pyrrolidine in

trans-2-butene gave the expected 1,2-dimethyl-4,4-diphenyl

spiropentane.


N=0 0

N-C-NH2 etherNH
I1 + 0
1 000

H 100% XII


N=N-0 H2N
+ + N2 + C=C=-CH2


40% XIII

In view of the quantitative formation of N-pyrrolidyl-

carboxamide (XII), the 1,1-diphenylallene and nitrogen

must have arisen from attack of the pyrrolidine on the

carbonyl carbon. However, it was independently demon-

strated that pyrrolidinium 2,2-diphenylcyclopropyl-

diazotate does not decompose to the diazocyclopropane (and,

hence, the allene and nitrogen) in ether or pentane at 0.


Although the pyrrolidinium diazotate is stable to
reaction conditions, it does decompose to 2,2-diphenyl-
diazocyclopropane under more drastic conditions. In
refluxing hexane, it gives quantitative nitrogen and 1,1-
diphenylallene. In the presence of diethyl fumarate
(ether, room temperature, overnight) it gives the expected
pyrazoline and in refluxing cyclohexene (one hour) it gives
23 per cent of the expected spiropentane and the rest 1,1-
diphenylallene. See Chapter II.










These results suggest that the formation of the allene and

nitrogen is induced by the relatively strong acid (XIV)

which would result from the collapse of the intermediate

formed from attack of pyrrolidine on the carbonyl carbon

of the nitrosourea.


0=N 0

R-N-I-NH2

N-H








R-N=N-0


0

+ N-C-NH2


XIV


,iii


Consistent with this suggestion was our finding that

pyrroliiinium 2,2-diphenylcyclopropyldiazotate (XIII) is

rapidly decomposed by p-toluenesulfonic acid in ether at

0 to give 1,1-diphenylallene (contaminated with what is

probably some ring-opened tosylate) and nitrogen.


N=O

R-N-C-NH2
II
2J
o


N
H












CHAPTER II


THE PREPARATION AND PROPERTIES OF SOME SALTS OF
2,2-DIPHENYLCYCLOPROPYLDIAZONIUM HYDROXIDE


Introduction


During the investigations on the base induced con-

version of N-nitroso-N-(2,2-diphenylcyclopropyl)urea to

2,2-diphenyldiazocyclopropane discussed in the preceding


N=O

0,-


G@
--N2


p


chapter, several salts of 2,2-diphenylcyclopropyldiazonium

hydroxide (XV) were isolated.


SLi Na K, Ag


H WH


+ LiOEt --"










The purpose of this chapter is to report the method

of synthesis of this interesting and relatively rare type

of molecule, and to report some of the chemical properties

of these salts that do not relate directly to the course of

the conversion of N-nitroso-N-(2,2-diphenylcyclopropyl)urea

to 2,2-diphenyldiazocyclopropane.


Results and Discussion


Preparation of pyrrolidinium 2,2-diphenylcyclo-

oropyldiazotate.--Pyrrolidinium 2,2-diphenylcyclopropyl-

diazotate (XIII) was isolated in 40 per cent yield from

the reaction of N-nitroso-N-(2,2-diphenylcyclopropyl)urea

(III) with an excess of pyrrolidine in ether at 0. The

diazotate formation was accompanied by ca. 60 per cent

nitrogen evolution (apparently due to the formation of

2,2-diphenyldiazocyclopropane, which spontaneously

decomposes to 1,1-diphenylallene and nitrogen) and

quantitative precipitation of N-pyrrolidinylcarboxamide

(XII).


To the best of our knowledge, the only other
aliphatic diazotates that have been reported are some
salts of methyl and benzyl diazonium hydroxide.2,14,15
For a general discussion on both aliphatic and aromatic
diazotates, see Ref. 13, pp. 44-67.











N=O
o o I :? Go
N-C-NH2 N2
ii + ON



III IV

C=C-CH2
0
0 + 60% N

=N-O H2N + 100% N-C-NH2

40% XIII XII

Pyrrolidinium 2,2-diphenylcyclopropyldiazotate (XIII)

is a white solid that melts sharply with decomposition at

850. It is unstable to moisture, decomposing rather

rapidly in aqueous solution and slowly when exposed as a

solid to the atmosphere. Decomposition either thermally

or by moisture is accompanied by the liberation of

pyrrolidine. Reproducible analyses could not be obtained.

The infrared spectrum was instructive in that it showed a

group of rather broad peaks between 3.5 and 4.5 microns

that can be assigned to the NH2+ stretching vibrations.

It also showed typical mono-substituted phenyl absorptions.

Perhaps the strongest evidence for the suggested

diazotate structure is the fact that XIII can be converted










by heat to 2,2-diphenyldiazocyclopropane (IV). This was

demonstrated in the following way. In hot hexane, the

diazotate was thermally unstable and gave the typical

decomposition product of 2,2-diphenyldiazocyclopropane,

1,1-diphenylallene. Thermal decomposition in the presence

of an olefin acceptor gave the spiropentane (XVI) and 1,1-

diphenylallene. Finally, decomposition in the presence

of diethyl fumarate gave the pyrazoline (XVII) and 1,1-

diphenylallene. From these observations, there can be




hexane C=C=CI"
/




C02E CO2Et

XIII DEF + C=C=CH




xvi i





XVI

This is despite the fact that it has been clearly
shown in the previous chapter that this is not the pre-
cursor of 2,2-diphenyldiazocyclopropane in the pyrrolidine
induced conversion of the nitrosourea to this diazo
compound.











virtually no doubt about the conversion of this salt to

2,2-diphenyldiazocyclopropane. Thus, in summary, the

primary points favoring the suggested diazotate structure

are the following. The salt is formed in a reaction that

gives a quantitative yield of N-pyrrolidinylcarboxamide.

This indicates quantitative attack of the pyrrolidine on

the carbonyl with displacement of the diazotate ion.

The infrared spectrum and the liberation of pyrrolidine

upon aqueous decomposition of the salt indicate the

presence of the pyrrolidinium cation. Conversion to 2,2-

diphenyldiazocyclopropane proves conclusively that the

2,2-diphenylcyclopropyl ring is intact in the salt and,

furthermore, is highly suggestive of the diazotate

structure. Finally, the pyrrolidinium salt is readily

converted to alkali metal salts for which there is addi-

tional evidence for the diazotate structure.

Preparation of alkali metal salts of 2,2-diphenyl-

cyclopropyldiazonium hydroxide.--When a sample of the

pyrrolidinium diazotate was added to a small amount of 5

per cent aqueous sodium, potassium, or lithium hydroxide,

the salt completely dissolved. However, after standing

for a few minutes, white crystals began to form. Similar


For qualifications to this statement as well as a
discussion of the formation of the diazocyclopropane in
this synthesis, see the preceding chapter.










to the pyrrolidinium salt, these white solids are stable

in aqueous base but decompose slowly when dissolved in tap

water. The decomposition is accompanied by the evolution

of a gas. Acidification of a freshly prepared aqueous

solution of any of these salts gave a precipitate (probably

the diazonium hydroxide) which very rapidly decomposed with

the evolution of a gas (presumably nitrogen). Moreover,

when dry carbon dioxide was passed through an ether slurry

of the lithium salt, 1,1-diphenylallene, the decomposition

product of 2,2-diphenyldiazocyclopropane, was formed. All

of these properties are consistent with those reported for

sodium methyldiazotate.2'14,15 Difficulty in the

manipulation and drying of the salts prepared in the above

manner prompted efforts to alternatively synthesize the

alkali metal derivatives.

It was found that the potassium (or lithium) salt

could be rather easily prepared by reduced pressure

evaporation of a solution of the pyrrolidinium diazotate

in methanol containing one equivalent of potassium (or

lithium) hydroxide. When the resulting thick brown oil

was leached with anhydrous ether, the potassium (or

lithium) diazotate crystallized as a pale yellow solid,

normally in yields of about 75 per cent. The infrared

spectra (KBr) showed that these salts are probably

alcoholates. This is consistent with the observations of











other investigators who have found that diazotates commonly

crystallize as alcoholates. The lithium salt was also

conveniently prepared by the addition of one equivalent of

lithium ethoxide to an ether solution of the pyrrolidinium

diazotate. Upon stirring overnight, the lithium salt

precipitated in nearly quantitative yield.

Reaction of lithium 2,2-diphenylcyclopropyldiazotate

with acylating agents.--In an attempt to obtain additional

evidence for the diazotate structures, the lithium salt was

treated with benzoyl chloride, acetyl chloride and acetic

anhydride. By analogy with the acylation of potassium

methyldiazotate and aromatic diazotates, ,15 the formation

of a mixture of the corresponding nitrosoamides (XVIII)

and the products that would result from the decomposition
of the diazoester (XIX) (presumably the diazocyclo-

propane*1' ) were anticipated.


For a general discussion on both aliphatic and
aromatic diazotates, see Ref. 13, pp. 44-67, 154.
+See footnote on p. 7.








24



N=O

SN- N-C-R
N-0 + R-l-X
0


XVIII
0

a, R C6H5 X = C + N=N--C-R

b, R = CH3 X = C1
0
c, R = CH3 X = O-C-CH3

To our initial surprise, infrared spectra of the product

oils from all of these reactions showed no absorption

maxima between 5.75 and 5.85 microns or around 6.65 microns,

thus suggesting that virtually none of the nitrosoamides

were formed. Alternate synthesis of both nitrosoamides by

the following scheme and demonstration of their stability

to reaction conditions




COOH () CO-N3 S N=C=O
(1) SOC1 benzene
(2) NaN3
0 0












CH3CO2H | HONO
o o


N=C=0
XVIIIb
H N=O

-N-C-
(2) <
(2) NaOAc

XVIIIa

eliminated the possibility that they had been formed and

then decomposed. The absence of these products was further

confirmed by chromatography of the crude reaction mixtures

on silica gel using conditions which were independently

shown to be effective for isolating the nitrosoamides.

The search for O-acylation products proved to be

more fruitful. Acylation on the diazotate oxygen would

give rise to a diazoester (XIX). In line with the rather

extensive studies that have been reported on this type of

compound, the diazoester would be expected to decompose

rapidly to give 2,2-diphenyldiazocyclopropane and the

corresponding carboxylic acid.


0

N=N-O-C-R+ R
N2 + RCO2H


NO










The formation of 2,2-diphenyldiazocyclopropane was sug-

gested with the detection of 1,1-diphenylallene in all of

the acylation reactions and was confirmed by effecting the

acylation of the diazotate in the presence of diethyl

fumarate. Under these conditions, the pyrazoline (XVII)

was isolated.

0
11 0 0
=N-0 M C-C1 11 Ii
+ + EtO-C-CH=CH-C-OEt ---L XVII





Additional evidence for the formation of 2,2-

diphenyldiazocyclopropane from the acylation of the

diazotate was obtained from the reaction of lithium or

potassium 2,2-diphenylcyclopropyldiazotate with excess

benzoyl chloride in ether at 00. Under these conditions,

a yellow-orange color developed almost immediately and

persisted throughout the reaction. Evaporation of the

reaction mixture to dryness gave an unstable orange oil.

The infrared spectrum of this oil showed, in addition to

the absorptions that are typical of l,l-diphenylallene

and benzoyl chloride, two peaks at 4.75 and 6.2 microns,

absorptions that suggested a diazoketone structure.16

Unfortunately, this particular component of the mixture










was too unstable to isolate in pure form. For example, all

attempts to chromatograph the crude reaction mixture led to

nitrogen evolution and loss of the unknown. We also

attempted to characterize this material by Wolff rearrange-

ment,17 pyrazoline formation and tosylate formation by
18
the method of Crowther and Holt. All of these attempts

were unsuccessful. However, some insight into the

structure of this material was gained by approaching the

problem from an entirely different direction. As a working

hypothesis, it was assumed that the diazoketone resulted

from reaction of 2,2-diphenyldiazocyclopropane with

benzoyl chloride. To test this, 2,2-diphenyldiazocyclo-

propane was generated (from reaction of the nitrosourea

(III) with lithium ethoxide) in the presence of benzoyl

chloride and, indeed, under these conditions, the orange

oil was again observed. However, in addition, there was

isolated a yellow solid which showed proper analysis,

infrared and n.m.r. for the hydroxy diazoketone (XX).


0
NhO11 6 0
-- + LiOEt + C-C -CH *

A OH N2










Furthermore, when the same reaction was run in the presence

of magnesium sulfate (to assure anhydrous conditions) the

unknown orange oil was formed at the expense of hydroxy

diazoketone (XX). These observations are readily

rationalized by the following reaction scheme in which the

orange oil and the hydroxy diazoketone have the common

precursor XXI.


O N2






N
H20 2
XX /C -- CH2- C--0
4) oi
0

XXI



Orange Oil---


There is excellent analogy for this suggested scheme in
19
the well-known Arndt-Eistert 9 reaction in which diazo7

methane reacts with an acid chloride to give a diazoketone.

The difference between the Arndt-Eistert and our reaction

rests in the manner in which the intermediate diazonium










ketone collapses. In the former case, it loses a proton

and in the latter which does not have a proton to lose,

the ring opens to give the delocalized diphenylmethyl

type of carbonium ion.


0 0 0
II II _ II
11 11 e -H 1
R-C-Cl + CH2N2 R-C-CH2-N2 R-C-CHN2



Furthermore, the similarity between the spectra (I.R. and

U.V.) of the two compounds suggests that the orange oil is

the product of reaction of the carbonium ion with some

available nucleophile (e.g., C1-, benzoic acid) and/or,

possibly loss of a proton to give the olefin. Nonetheless,

whatever its complete structure, there can be little doubt

but that it also contains an a-diazoketone group which

arose from reaction of 2,2-diphenyldiazocyclopropane with

benzoyl chloride.

All of these acylation results not only present

additional evidence for the suggested diazotate structures

for the salts but also point up an interesting anomalous

behavior of these salts: whereas the methyl and benzyl

diazotates undergo acylation almost exclusively on the

nitrogen,15 the 2,2-diphenylcyclopropyldiazotate undergoes

acylation exclusively on the oxygen. Actually, the reason

for this is probably quite simple since the nitrogen of








50

the diazotate in the 2,2-diphenylcyclopropyl system is

bonded to a rather bulky secondary carbon that could easily

retard attack of the acylating group on the nitrogen.

Reaction of lithium 2,2-diphenylcyclopropyldiazotate

with benzoic acid.--The reaction of the lithium diazotate

with benzoic acid in refluxing benzene resulted in the

formation of the ring-opened ester, 1,l-diphenylpropen-3-yl

benzoate (XXII), and 1,1-diphenylallene. The benzoate



G 0 0
N=N-0 Li A II
+ CO2H -- C=CH-CH2-0-C -4


o\ XXII

+ C=C=CH2



probably arises from reaction of 2,2-diphenyldiazocyclo-

propane with benzoic acid. Consistent with this suggestion

is the fact that both the thermal decomposition of N-

nitroso-(2,2-diphenylcyclopropyl)urea in the presence of
*
benzoic acid and the thermal decomposition of N-nitroso-

N-(2,2-diphenylcyclopropyl)benzamide (XVIII-a), both of


Unpublished results of D. L. Muck, Department of
Chemistry, University of Florida, Gainesville, Florida.










which probably proceed through 2,2-diphenyldiazocyclo-

propane, give the same mixture of products.


N=O
[ CO2H

X I' A
o o
0

C=CH-CH -0- C-


N=O +

CN-C-0 C CH
o


XVIIIa












CHAPTER III


EXPERIMENTAL


The melting points were taken in a Thomas Hoover

Uni-melt apparatus and are uncorrected. The infrared

spectra were recorded with a Perkin-Elmer Infracord

spectrophotometer and the ultraviolet spectra were recorded

with a Cary 14. The elemental analyses were carried out by

Galbraith Laboratories, Inc., Knoxville, Tennessee. G.l.c.

analyses were performed with an Aerograph Hy-Fi model

600-B or Aerograph Model A-350-B converted to an autoprep.

The nuclear magnetic resonance spectra were determined in

deuterated chloroform solution employing a Varian 4300-2

high resolution spectrometer operating at 56.4 Mc.

Materials.--The solvents used in this work were all

Fisher Certified Reagents and were used without further

purification as were the ethyl carbamate (Matheson), ethyl

benzoate (Eastman White Label), diethyl fumarate (Eastman

White Label), benzoyl chloride (Allied), acetyl chloride

(Allied), glacial acetic acid (Allied), acetic anhydride

(Baker), phenylmagnesium bromide (Peninsular Chem Research),

and vinylmagnesium bromide (PCR). The pyrrolidine










(Eastman Practical Grade) and cyclohexene (Eastman White

Label) were purified by fractional distillation before use.

Lithium ethoxide.--Lithium ethoxide was prepared by
20
the method of Brown, Dickerhoof and Bafus. The product

resulting from this preparation was invariably contaminated

with material that evolves hydrogen with sodium hydride.

The contaminant is probably ethanol and, possibly, some

lithium hydroxide (or water).

Quantitative determination of ethyl carbamate

resulting from the reaction of N-nitroso-N-(2,2-diphenyl-

cyclopropyl)urea with lithium ethoxide in ethyl ether.--To

0.60 g. (0.00214 mole) of N-nitroso-N-(2,2-diphenylcyclo-

propyl)urea8 and 30 ml. of anhydrous ether stirred at 00
was added 0.310 g. (1.5 equivalents) of lithium ethoxide.

After 1 1/2 hours, 36.5 ml. (68%) of nitrogen had been

collected and no more was being evolved. The reaction

mixture was filtered and the filtrate evaporated under

reduced pressure to a small volume. After adding 5.25 mg.

of menthol (internal standard), the ether solution was

analyzed by g.l.c. (5 ft. x 1/8 in. column charged with

Carbowax 4000 on firebrick) using a disc chart integrator.

Although 1.6 mg. (corresponding to 0.86% yield) of ethyl

carbamate would have been easily detected, none was observed.

An infrared spectrum on the oil obtained by the evaporation










of the remaining ether showed this product to be almost

pure 1,1-diphenylallene.8 An infrared spectrum (KBr) on

the solid filtered from the reaction was consistent with a

mixture of lithium ethoxide, lithium carbamate (or,

possibly, the cyanate21) and lithium 2,2-diphenylcyclo-

propyldiazotate.

Quantitative determination of ethyl carbamate re-

sulting from the reaction of N-nitroso-N-(2,2-diphenyl-

cyclopropyl)urea with lithium ethoxide in petroleum ether.--

To 0.60 g. (0.00214 mole) of N-nitroso-N-(2,2-diphenyl-

cyclopropyl)urea and 30 ml. of petroleum ether (A.R.)

stirred at 00 was added 0.207 g. (0.00214 mole) of lithium

ethoxide. After 2 hours, 46.7 ml. (87%) of nitrogen had

been collected and no more was being evolved. The reaction

mixture was filtered and the filtrate evaporated under

reduced pressure to a small volume. After adding 3.23 mg.

of menthol (internal standard), the solution was analyzed

by g.l.c. (5 ft. x 1/8 in. column charged with Carbowax

4000 on firebrick) using a disc chart integrator. Although

1.78 mg. (corresponding to 0.94% yield) of ethyl carbamate

would have been easily detected, none was observed. An

infrared spectrum (film) of the oil obtained by evaporation

of the solvent was superimposable with that of 1,1-

diphenylallene.










Stability of ethyl carbamate to reaction condi-

tions.--To an anhydrous ether solution of 0.1 g. (0.001124

mole) of ethyl carbamate and 0.08 ml. of diglyme (internal

standard) stirred at 00 were added 0.109 g. (0.001124 mole)

of lithium ethoxide and 27 mg. (0.001124 mole) of lithium

hydroxide. The resulting mixture was stirred with cooling

for 5 hours. G.l.c. analyses (5 ft. x 1/8 in. column

charged with Carbowax 4000 on firebrick) of the ether

solution before the addition of base and again upon com-

pletion of the experiment showed that the ethyl carbamate

was 99 per cent unreacted.

Stability of lithium 2,2-diphenylcyclopropyl-

diazotate in aqueous tetrahydrofuran.--To 15 ml. of tetra-

hydrofuran and 5 ml. of water at 0 was added 99 mg. of

lithium 2,2-diphenylcyclopropyldiazotate (contaminated

with a small amount of lithium ethoxide and lithium

hydroxide) prepared by reacting pyrrolidinium 2,2-diphenyl-

cyclopropyldiazotate with lithium ethoxide. The solution

was stirred at this temperature for about 16 hours and then

allowed to warm slowly to room temperature and stir for an

additional 4 hours. By recooling the system to 00, it was

ascertained that no gas had been evolved. The solvents

were removed in vacuo and the residual solid was powdered

under ether. Filtration and infrared analysis (KBr) of










the white solid (99 mg.) demonstrated quantitative recovery

of the starting lithium diazotate.

Decomposition of lithium 2,2-diphenylcyclopropyl-

diazotate in refluxing aqueous tetrahydrofuran.--To a

refluxing solution of 5 ml. of tetrahydrofuran and 5 ml.

of water was added dropwise a tetrahydrofuran solution of

lithium 2,2-diphenylcyclopropyldiazotate. After heating

under reflux overnight, the reaction mixture was cooled

and poured into ether. The organic layer was washed once

with 5 per cent HC1 and twice with water, dried over

anhydrous MgSO4, and evaporated to an oil. G.l.c. analysis

(5 ft. x 1/8 in. column charged with Carbowax 20M on 60-80

mesh Gas Chrom Z) of the oil showed 92.1 per cent of a

material with a retention time identical to that of a

sample of 1,1-diphenylallene prepared by the thermal

decomposition of N-nitroso-N-(2,2-diphenylcyclopropyl)urea.10

There were no peaks observed with retention times corres-

ponding to those of l,l-diphenylpropen-3-ol* or 3,3-diphenyl-

propen-3-ol. An infrared spectrum (film) of the product

oil was consistent with the g.l.c. results by being super-

imposable with that of 1,1-diphenylallene.


This compound was supplied by D. L. Muck, who pre-
pared it by a reported method.








37

Preparation of 5,3-diphenylpropen-3-ol.--To a flask

flushed with Argon and cooled in a water bath was intro-

duced 22 ml. (0.0559 mole) of a 2.54 molar tetrahydrofuran

solution of vinylmagnesium bromide. Then 5.0 g. (0.0274

mole) of benzophenone dissolved in 50 ml. of anhydrous

tetrahydrofuran was added to the Grignard reagent dropwise.

The reaction mixture was stirred until the red color

disappeared and then heated under reflux for one hour.

After cooling the solution to room temperature, 1.6 g.

(0.03 mole) of ammonium chloride was added.
After the initial vigorous bubbling ceased, the

mixture was heated under reflux for several minutes and

then poured into water with stirring. This aqueous mixture

was extracted with ether and the ether extracts were washed

successively with water, 5 per cent NaHCO3 and water again

before being dried over anhydrous MgSO4. The oil obtained

upon evaporation of the ether was dissolved in pentane

and stirred. A white solid, m.p. 154-1680, of undetermined

nature precipitated and was filtered. The oil obtained

upon evaporation of the filtrate (5.5 g.) was purified by

molecular distillation to give 3,3-diphenylpropen-3-ol,

b.p. 900/0.1-0.3 mm. Characteristic infrared absorption

maxima (neat): 2.82 P, 5.2 u, 6.25 p, 6.7 p, 6.9 u, 7.1 p,

7.5 u, 8.5 u, 9.15 p, 9.45 p, 10.2 p, 10.8 p, 11.1 p,
13.0 p, 13.2 p, 14.3 p.










Anal. Calcd. for C15H140: C, 85.68; H, 6.71. Found:

C, 85.45; H, 6.54; Mol. Wt. (Rast) 209.

Stability of N-nitroso-N-(2,2-diphenylcyclopropyl)urea

to lithium 2,2-diphenylcyclopropyldiazotate.--To 0.300 g.

of N-nitroso-N-(2,2-diphenylcyclopropyl)urea dissolved in

20 ml. of anhydrous ethyl ether at 00 was added a catalytic

amount (about 50 mg.) of lithium 2,2-diphenylcyclopropyl-

diazotate. After stirring with cooling for 6 1/2 hours,

the reaction mixture was filtered. When the filtrate was

evaporated to a small volume, a yellow solid, m.p. 92-70

(dec.), formed. An infrared spectrum (KBr) of this solid

was superimposable with that of a known sample of starting

nitrosourea.

Stability of lithium 2,2-diphenylcyclopropyl-

diazotate in the presence of ethyl carbamate.--To a slurry

of 100 mg. (0.453 mmoles) of lithium 2,2-diphenylcyclo-

propyldiazotate in 25 ml. of anhydrous ether cooled to 00

was added 40 mg. (0.453 mmoles) of ethyl carbamate. After

stirring for 7 hours, the reaction mixture was filtered to

give 91 mg. (91% recovery) of the lithium diazotate. The

filtrate was evaporated to a white solid whose infrared

spectrum (KBr) was superimposable with that of ethyl

carbamate. No 1,1-diphenylallene was detected.










Attempted preparation of 2,2-diphenyldiazocyclo-

propane alkyl ethers.--To a slurry of 257 mg. (0.00106 mole)

of lithium 2,2-diphenylcyclopropyldiazotate in 25 ml. of

anhydrous ether was added 86 pl. (2 equivalents) of ethyl

iodide. This mixture was allowed to stir at room temperature

for 24 hours. Filtration yielded an 85 per cent recovery

of starting diazotate. A small sample was dissolved in

H20, acidified with HNO3 and tested with AgNO The

resulting clear solution indicated an absence of KI.

The remaining lithium diazotate (0.222 g.; 0.914

mmole) was dissolved in tetrahydrofuran and cooled to 00.

After 0.114 ml. (2 equivalents) of methyl iodide was

added, the reaction was stirred with cooling for 8 hours.

By this time, no gas evolution had occurred and so the

reaction mixture was allowed to warm to room temperature

and stir for 2 days. The solvent was then evaporated under

reduced pressure and ether was added. A small amount of

the lithium diazotate was recovered. However, an infrared

spectrum (film) on the ether soluble oil was superimposable

with that of 1,1-diphenylallene.

To a solution of 0.105 g. (0.433 mmoles) of lithium

2,2-diphenylcyclopropyldiazotate in 3 ml. of anhydrous

ethanol was added 0.074 g. (1 equivalent) of AgNO After

stirring for 30 minutes, the dark brown silver diazotate

was filtered and dried.










To an ether slurry of this salt stirred at 00 was

added 3 equivalents of ethyl iodide. After 9 hours, only

1.4 ml. of gas had been collected (14.5%) and no more was

being evolved. The reaction mixture was filtered and the

filtrate evaporated under reduced pressure. An infrared

spectrum on the small amount of remaining oil was super-

imposable with that of 1,1-diphenylallene. The recovered

silver diazotate was added to an ether solution of 3

equivalents of methyl iodide and the resulting mixture was

stirred overnight at room temperature. After removal of

the unreacted silver diazotate by filtration, the ether

solution was evaporated to an oil. An infrared spectrum

(film) of this oil was superimposable with that of 1,1-

diphenylallene.

Preparation of l-(2,2-diphenylcyclopropyl)-3,3-

tetramethylenetriazine.--To 0.5 g. (0.0018 mole) of N-

nitroso-N-(2,2-diphenylcyclopropyl)urea in 20 ml. of

petroleum ether (A.R.) stirred at 00 was added 0.4 ml.

(0.0048 mole) of pyrrolidine. Nitrogen was evolved over

a 2-2 1/2 hour period and N-pyrrolidinylcarboxamide slowly

formed. After gas evolution ceased (usually 80-93%) the

reaction mixture was filtered. Reduced pressure evapora-

tion of the solvent gave a heavy yellow oil which was

dissolved in a minimum amount of hot pentane. The solution










was then cooled to 0 and this temperature was maintained

overnight. The yellow crystals which had formed were

filtered and washed once with pentane to yield 30 mg. to

98 mg. (6-19%) of triazine, m.p. 77-800. Successive

recrystallizations from hexane gave a pure white solid,

m.p. 80-810.

Significant infrared absorptions (KBr): 3.35 p,

6.25 p, 6.7 p, 6.95 P, 7.55 p, 7.75 u, 9.85 p, 9.9 p,

12.95 p, 13.3 u, 14.2 u.

Anal. Calcd. for C19H21N3: C, 78.50; H, 7.27;

N, 14.45. Found: C, 78.38; H, 7.23; N, 14.24.

Thermal decomposition of pyrrolidinium 2,2-diphenyl-

cyclopropyldiazotate in the presence of pyrrolidine.--A

mixture of 182 mg. of pyrrolidinium 2,2-diphenylcyclo-

propyldiazotate and 0.4 ml. of pyrrolidine in 30 ml. of

hexane was heated under reflux for 1 1/2 hours. The

reaction mixture was then filtered to remove a small amount

of white solid (probably N-pyrrolidinylcarboxamide present

as starting diazotate impurity) and the filtrate was

evaporated to an oil. An infrared spectrum (film) of this

oil showed it to be essentially pure 1,1-diphenylallene.

No absorptions consistent with 1-(2,2-diphenylcyclopropyl)-

3,3-tetramethylenetriazine were observed. Neither did the
product oil evolve a gas when treated with acetic acid.

The latter is a fairly sensitive qualitative test for the

triazine.










Reaction of N-nitroso-N-(2,2-diphenylcyclopropyl)urea

with lithium hydroxide.--To a solution of 0.300 g. (0.00107

mole) of N-nitroso-N-(2,2-diphenylcyclopropyl)urea in 30 ml.

of anhydrous ether stirred at 0 was added 39 mg. (1.5

equivalents) of lithium hydroxide. Nitrogen slowly evolved

for 5 hours (21.2 ml.; 78.5%) after which the reaction
mixture was filtered. An infrared spectrum (KBr) of the

white solid (129 mg.) was consistent with a mixture of
lithium 2,2-diphenylcyclopropyldiazotate and lithium

carbamate. The filtrate was evaporated to an oil (183 mg.)

whose infrared spectrum (film) was consistent with 1,1-

diphenylallene contaminated with N-(2,2-diphenylcyclo-

propyl)urea.

Reaction of N-nitroso-N-(2,2-diphenylcyclo-

propyl)benzamide with lithium ethoxide.--To 0.861 g.
(0.00252 mole) of N-nitroso-N-(2,2-diphenylcyclo-

propyl)benzamide dissolved in 30 ml. of anhydrous ether
was added 0.366 g. (1.5 equivalents) of lithium ethoxide

alcoholate. After stirring at room temperature for 2 days,
the reaction mixture was filtered and the filtrate was
evaporated to a small volume. Upon adding 55 mg. of

naphthalene (internal standard), the ether solution was

analyzed by g.l.c. (5 ft. x 1/8 in. column charged with

Carbowax 4000 on firebrick) which showed a 2.9 per cent

yield (average of two reactions) of ethyl benzoate. An










infrared spectrum (film) of the oil obtained upon evapora-

tion of the ether was consistent with 1,1-diphenylallene

contaminated with a small amount of unreacted nitroso-

benzamide.

The solid which was filtered from the reaction

mixture was dissolved in water and the aqueous solution

was washed twice with ether. Upon acidification with 5

per cent HC1, the solution was extracted with ether. The

ether extracts were then washed with water, dried over

anhydrous MgS04, and evaporated to a white solid. An

infrared spectrum (KBr) of this solid (209 mg.) was

superimposable with that of benzoic acid and represents a

68 per cent yield of lithium benzoate from the reaction.

In a subsequent reaction, no lithium 2,2-diphenylcyclo-

propyldiazotate was detected in the solids filtered from

the reaction mixture by physical (I.R.) or chemical (HOAc,

BzC1) means.

Stability of ethyl benzoate to reaction conditions.--

To an anhydrous ether solution of ethyl benzoate and

naphthalene (internal standard) were added lithium ethoxide

alcoholate (1.5 equivalents) and LiOH (1.5 equivalents).

The resulting mixture was stirred at room temperature for

48 hours. G.l.c. analyses (5 ft. x 1/8 in. column charged

with Carbowax 4000 on firebrick) of the ether solution










before the addition of base and again upon completion of

the experiment showed that the ethyl benzoate was 99.1 per

cent unreacted (average of two experiments).

Reaction of N-nitroso-N-(2,2-diphenylcyclopropyl)urea

with pyrrolidine in ethyl ether.--To 0.600 g. (0.00214 mole)

of N-nitroso-N-(2,2-diphenylcyclopropyl)urea dissolved in

25 ml. of anhydrous ether stirred at 0 was added 0.4 ml.

(0.34 g.; 0.0048 mole) of pyrrolidine (and usually 0.60 g.
of anhydrous MgSO4). Nitrogen evolved for 13 minutes

(usually about 28 ml.; 58%) after which the reaction

mixture was filtered to give a quantitative yield of N-

pyrrolidinylcarboxamide, m.p. 216-2250. Successive

recrystallizations from ethanol gave a pure white solid,

m.p. 220-221 reportedd2 m.p. 2180). Significant infra-
red absorptions (KBr): 2.95 u, 3.1 y, 6.1 ,, 6.3 p,

6.75 p, 6.9 ), 7.14 p and 12.9 p.

Anal. Calcd. for C5H10N20: C, 52.61; H, 8.83; N,

24.54. Found: C, 52.39; H, 8.65; N, 24.37.

When the above ether filtrate was evaporated under

reduced pressure while allowing the flask to get very

cold, pyrrolidinium 2,2-diphenylcyclopropyldiazotate usually

crystallized to a white solid. Sometimes, however, it was

necessary to add pentane very slowly to effect crystalli-

zation. This solid was filtered, washed with a small

volume of cold, anhydrous ether and dried to give an average










yield of 265 mg. (40%), m.p. 79-830 (dec.). Recrystalli-

zation from anhydrous ether gave the pyrrolidinium

diazotate, m.p. 850 (dec.).

Significant infrared absorption bands (nujol):

4.05 p (broad), 6.05 p, 6.15 p, 6.25 p, 6.65 u, 8.8 p,

9.5 p, 9.9 p, 10.8 p (very broad), 11.6 p (very broad),

15.15 p, and 14.35 p. An elemental analysis on this

compound was precluded by its instability.

The ether solution from which the pyrrolidinium

diazotate had crystallized was evaporated under reduced

pressure to a yellow oil. An infrared spectrum (film)

of this oil was superimposable with that of 1,1-diphenyl-

allene.

Reaction of N-nitroso-N-(2,2-diphenylcyclopropyl)urea

and pyrrolidine in trans-2-butene.--Into a flask containing

1.0 g. (0.00356 mole) of N-nitroso-N-(2,2-diphenylcyclo-

propyl)urea and equipped with a dry ice-acetone condenser

was added 15 ml. of trans-2-butene (Matheson C.P.). After

the addition of 0.8 ml. (0.681 g.; 0.0096 mole) of

pyrrolidine, the mixture was magnetically stirred at room

temperature. After an hour, 60 ml. (67%) of N2 had been

collected and no more was being evolved. The butene was

then allowed to evaporate after the addition of 25 ml. of

petroleum ether. The reaction mixture was filtered and the

filtrate was dried over anhydrous Na2SO4 and evaporated to










an oil. This oil was heated on a steam bath for 20 minutes

to polymerize the 1,1-diphenylallene and then chromato-

graphed through basic alumina (Fisher; 80-200 mesh). After

eluting with 250 ml. of petroleum ether, the solvent was

evaporated to give 200 mg. of a light yellow oil. This oil

was then separated by means of preparative g.l.c. (9 ft.

x 1/2 in. column charged with 25% G.E. SF-96(50) silicone

fluid on 30-60 mesh Chromosorb P) and the desired spiro-

pentane was collected as a colorless oil. The addition of

a small amount of cold ethanol effected crystallization to

give a white powder, m.p. 51-52 reported8 m.p. 51 1/2-

520). No melting point depression was observed on

admixture with a known sample of 2,2-diphenyl-4,5-dimethyl-

spiropentane.8

Stability of pyrrolidinium 2,2-diphenylcyclopropyl-

diazotate to reaction conditions.--An ether solution of

240 mg. of pyrrolidinium 2,2-diphenylcyclopropyldiazotate

was magnetically stirred at 00 for 7 hours. The solution

was then evaporated to a small volume and pentane was added

to crystallize the starting material. Filtration gave

160 mg. (67% recovery) of diazotate, m.p. 70 (dec.).

It was also found that a 96 per cent recovery of the

pyrrolidinium diazotate was possible after stirring a

mixture of 189 mg. of diazotate and 0.4 ml. of pyrrolidine

in 30 ml. of petroleum ether at 00 for 4 hours.










Decomposition of pyrrolidinium 2,2-diphenylcyclo-

propyldiazotate by p-toluenesulfonic acid.--To 174 mg.

(0.565 mmoles) of pyrrolidinium 2,2-diphenylcyclopropyl-
diazotate, which had been leached with anhydrous ether to

remove all traces of 1,1-diphenylallene, in ether at 00

was added an ether solution of 107 mg. (0.563 mmoles) of

p-toluenesulfonic acid during 10 minutes. After stirring
for one hour, the reaction mixture was washed twice each

with water, 5 per cent NaHCO3 and water again before being

dried over anhydrous MgSO4. An infrared spectrum (neat)

of the oil obtained upon evaporation of the solvent showed

1,1-diphenylallene to be a major product of the reaction.

Thermal decomposition of pyrrolidinium 2,2-diphenyl-

cyclopropyldiazotate in hexane.--A 50 mg. sample of

pyrrolidinium 2,2-diphenylcyclopropyldiazotate was heated

in hexane to reflux. Reduced pressure evaporation of the

solvent gave a brown-yellow oil whose infrared spectrum

(film) was superimposable with that of 1,1-diphenylallene.

Thermal decomposition of pyrrolidinium 2,2-diphenyl-

cyclopropyldiazotate in the presence of diethyl fumarate.--

A solution of 240 mg. of pyrrolidinium 2,2-diphenylcyclo-

propyldiazotate and 0.546 ml. (0.574 g.; 4 equivalents) of

diethyl fumarate in 30 ml. of anhydrous ether was stirred

overnight at room temperature. Evaporation of the solvent










to a small volume and filtration yielded 145 mg. (47.6%)

of white solid, m.p. 140-1435, whose infrared spectrum

(KBr) was superimposable with that of a known sample of

pyrazoline (XVII).8

Thermal decomposition of pyrrolidinium 2,2-diphenyl-

cyclopropyldiazotate in cyclohexene.--A mixture of 100 mg.

(0.324 mmoles) of pyrrolidinium 2,2-diphenylcyclopropyl-

diazotate in 10 ml. of cyclohexene was heated under reflux

for one hour. The resulting solution was cooled, dried

over anhydrous MgSO4, filtered, and evaporated to an oil.

After 5.42 mg. of methyl a-methylcinnamate (internal

standard) was added, the product oil was analyzed by g.l.c.

(10 ft. x 1/8 in. column charged with 5% Apiezon L on 60-

80 mesh Gas Chrom Z) which showed a 23.3 per cent yield of

2,2-diphenyl-4,5-tetramethylenespiropentane. The adduct

was identified by comparison of its retention time with

that of a known sample.8

Preparation of sodium 2,2-diphenylcyclopropyl-

diazotate.--A small portion of freshly prepared pyrrolidinium

2,2-diphenylcyclopropyldiazotate was added to a few drops

of 5 per cent NaOH. After the sample completely dissolved,
crystals of sodium 2,2-diphenylcyclopropyldiazotate pre-

cipitated out of solution upon standing. This precipitate

was filtered and dried to give a white solid. Character-

istic infrared absorption maxima (nujol and KBr): 6.1 p,










6.25 p, 6.7 p, 8.5 p, 8.8 y, 9.1 p, 9.5 p, 9.75 p, 11.5 u,

15.0 p, 13.5 p, and 14.5 (broad) p.
When this sodium diazotate was dissolved in water

and acidified with 5 per cent HC1, a white precipitate

formed which rapidly evolved a gas (presumably nitrogen).

The lithium and potassium diazotates were prepared

in the same manner.

Decomposition of lithium 2,2-diphenylcyclopropyl-

diazotate with carbon dioxide.--Dry carbon dioxide (Linde)

was bubbled through an ether slurry of lithium 2,2-

diphenylcyclopropyldiazotate for 15 minutes. The ether

solution was then filtered and evaporated to an oil. An

infrared spectrum of the product oil was superimposable

with that of pure 1,1-diphenylallene, the decomposition

product of 2,2-diphenyldiazocyclopropane.

Preparation of potassium 2,2-diphenylcyclopropyl-

diazotate.--To freshly prepared pyrrolidinium 2,2-

diphenylcyclopropyldiazotate was added an equimolar amount

of KOH dissolved in anhydrous methanol and cooled to 00.

This solution was stirred magnetically as the solvent was

evaporated in vacuo to a very thick brown oil. Upon adding

a small amount of anhydrous ether, the potassium diazotate

crystallized to a pale yellow solid (normally 75% yield).

When the product was heated in an oil bath, it slowly











darkened up to its melting point (2080), but did not

decompose as evidenced by loss of gas from the cooled melt

on acidification with acetic acid. A strong and broad

infrared absorption maximum in the OH region indicated

that the potassium salt was probably an alcoholate. The

lithium diazotate was also prepared in this manner.

Preparation of lithium 2,2-diphenylcyclopropyl-

diazotate.--To an ether solution of pyrrolidinium 2,2-

diphenylcyclopropyldiazotate was added one equivalent of

lithium ethoxide. After stirring overnight at room

temperature, the white solid was filtered giving the

lithium diazotate (contaminated with lithium ethoxide) in

good yield.

Reaction of lithium 2,2-diphenylcyclopropyldiazotate

with benzoyl chloride.--To an ether suspension of lithium

2,2-diphenylcyclopropyldiazotate stirred magnetically at O0

was added 1.5 equivalents of benzoyl chloride. After

several hours the yellow reaction mixture was washed with

5 per cent NaHCO3 and water and dried over anhydrous MgSO4.

An infrared spectrum (neat) of the orange oil obtained upon

evaporation of the solvent showed strong absorption bands

at 4.75 p, 6.2 p and 6.55 p in addition to those consistent

with a mixture of benzoyl chloride and 1,1-diphenylallene.

There was a distinct absence of absorption bands at 5.85 p











and 6.65 p. This eliminates N-nitroso-N-(2,2-diphenyl-

cyclopropyl)benzamide as a possible product.*

The orange product oil containing the above mixture

was dissolved in acetonitrile and stirred at room temperature

overnight. The precipitate which had formed was filtered,

washed with ether, and recrystallized from benzene-pentane

to give white crystals, m.p. 197-1980. An infrared

spectrum of this solid was superimposable with that of the

C44H5202 material obtained by the decomposition of the a-
diazoketone (orange oil) prepared by decomposing N-nitroso-

N-(2,2-diphenylcyclopropyl)urea with lithium ethoxide in

the presence of benzoyl chloride.

Preparation of N-(2,2-diphenylcyclopropyl)acetamide.--

An ether solution of 2,2-diphenylcyclopropanecarboxylic

acid azide was prepared in the reported manner8 from 10 g.

(0.042 mole) of 2,2-diphenylcyclopropanecarboxylic acid.

To a refluxing mixture of 100 ml. of anhydrous benzene and

100 ml. of glacial acetic acid was added the anhydrous


Despite the fact that the infrared spectrum did not
show a detectable amount of the nitrosobenzamide, a sub-
sequent reaction mixture was carefully chromatographed on
silica gel (Fisher) using the conditions known to isolate
this compound. None was obtained.
Completely analogous results were obtained when
acetyl chloride or acetic anhydride was used as the
acylating reagent.










ether solution of acid azide at such a rate that the ether

was distilled away. When the temperature of the refluxing

vapor reached 750, the water to the condenser was turned

on and the reaction mixture was heated under reflux over-

night. After gas evolution stopped, about 15 hours, the

reaction mixture was cooled, neutralized with a saturated

sodium carbonate solution, and diluted with ether. A white

solid which precipitated at this point was removed by

filtration and identified as the symmetrically disubstituted

urea.

The organic layer was separated, washed twice with

water, dried over anhydrous MgSO4, and evaporated under

reduced pressure to a small volume. After the oil was

dissolved in ether, a white solid precipitated which was

filtered, washed with ether and dried to give 5.5 g.

(0.0219 moles; 52.2% from acid) of the desired amide, m.p.

125-1270. Recrystallization from benzene-ether yielded

3.5 g. of product, m.p. 125 1/2-1260; infrared maxima
(nujol), 3.02 p, 6.05 p, 6.52 y, 6.67 p, 7.24 p, 9.8 p,

10.63 p, 13.27 p, 14.17 p and 14.35 p.

Anal. Calcd. for C17H17NO: C, 81.24; H, 6.82; N,

5.57. Found: C, 81.17; H, 6.80; N, 5.44.

Preparation of N-nitroso-N-(2,2-diphenylcyclo-

propyl)acetamide.--Method B of White23 was employed in










this nitrosation. A mixture of 0.8 g. (3.2 mmoles) of N-

(2,2-diphenylcyclopropyl)acetamide, 3.2 ml. of glacial

acetic acid and 16 ml. of acetic anhydride was stirred at

00 as 4.8 g. (0.07 mole) of solid sodium nitrite were

added over a five hour period. After stirring an addi-

tional five hours at that temperature, the solution was

warmed to 150, poured into a mixture of ice and water,

and extracted with ether. These ether extracts were

washed successively with water, 5 per cent NaHCO3 and

water again before being dried over anhydrous MgS04. The

yellow oil obtained upon solvent evaporation was re-

crystallized twice from a pentane-ether mixture at dry

ice-acetone temperature. The viscous yellow oil was

unstable enough at room temperature to preclude elemental

analyses. However, it was stable in solution at 00 for

long periods of time. It was found that elution

chromatography on silica gel using 90 per cent pentane-

ether as eluent also purified the nitrosoamide. Significant

infrared absorption maxima (neat): 3.22 p, 3.36 p, 5.73 u,

6.25 p, 6.6 p, 6.9 p, 7.3 p, 7.5 p, 8.67 p, 9.2 p, 10.65 p,

11.05 p, 13.0 u and 14.36 p.

Preparation of N-(2,2-diphenylcyclopropyl)benzamide.--

A benzene solution of 2,2-diphenylcyclopropylisocyanate,

made in the usual manner from 2,2-diphenylcyclopropane-

carboxylic acid (10.0 g.; 0.042 mole), was cooled to 00










and maintained in an atmosphere of dry argon as a slight

excess of phenylmagnesium bromide in ether was slowly added

with stirring. After stirring an hour, the reaction

mixture was allowed to warm to room temperature and stir

for an additional hour. The solution was then cooled to

00 and an aqueous solution of 2.61 g. (0.0488 mole) of

ammonium chloride was added. The organic layer was

separated, washed successively with 5 per cent HC1, 5 per

cent NaHCO3 and water, dried over anhydrous magnesium
sulfate, and evaporated to a solid residue. The solid

was leached with pentane and filtered to give 9.53 g.

(72.5% from acid) of product amide, m.p. 156-158.
Successive recrystallizations from hot benzene or

benzene-pentane gave analytically pure product, m.p. 157-

1580; infrared maxima (KBr): 2.95 p, 3.25 p, 6.1 p,
6.25 6, 6.35 p, 6.56 v, 6.72 p, 6.93 p, 7.68 p, 7.75 u,

9.33 u, 9.75 u, 10.12 u, 10.85 p, 11.33 p, 11.55 p,
12.47 1, 15.14 u, 13.37 p, 13.9 p, 14.17 p, and 14.5 p.

Anal. Calcd. for C22H19NO: C, 84.31; H, 6.11; N,

4.47. Found: C, 84.27; H, 6.08; N, 4.31.

Preparation of N-nitroso-N-(2,2-diphenylcyclo-

propyl)benzamide.--A solution of 0.5 g. (0.0016 mole) of

N-(2,2-diphenylcyclopropyl)benzamide in 40 ml. of ether
was stirred magnetically and cooled to -200. To this was

added 0.394 g. (0.0048 mole) of sodium acetate and 0.0024











moles of N204 (Matheson C.P.) dissolved in ether. The

mixture was allowed to stir until all the green color had

disappeared and additional N204 was then added. When the

green color had again vanished, the yellow reaction

mixture was washed successively with water, 5 per cent

NaHCO3 and water again, dried over anhydrous MgSO4 and

evaporated to an oil. This oil was chromatographed on

silica gel (Fisher) and the nitrosobenzamide separated as

a bright yellow band with 95 per cent pentane-ether as

eluent. The nitrosoamide crystallized from pentane at dry

ice temperature, but was a liquid at room temperature. The

product was unstable enough at ordinary temperature to

preclude elemental analyses.

Characteristic infrared absorptions (neat): 3.25 P,

5.85 u, 6.65 p, 6.7 p, 6.9 u, 7.5 p, 7.85 u, 9.35 p,

10.5 p, 11.35 p, 11.5 p, 12.7 p (broad), 15.3 p, 14.4 p

(broad).

Reaction of potassium 2,2-diphenylcyclopropyldiazo-

tate with benzoyl chloride in the presence of diethyl

fumarate.--To 0.163 g. (0.591 mmoles) of potassium 2,2-

diphenylcyclopropyldiazotate and 0.585 ml. (0.406 g.;

2.36 mmoles) of diethyl fumarate (Eastman Organic Chemi-

cals) at 00 was added an ether solution of 68 pl. (0.591

mmoles) of benzoyl chloride over a two hour period.











Although a white solid had precipitated by this time, the

reaction mixture was stirred for an additional two hours.

After enough ether had been added to dissolve the white

solid, the solution was washed with water, 5 per cent

NaHCO3 and water, dried over anhydrous MgS04, and

evaporated to a small volume. This mixture was filtered

to give 55 mg. (23%) of white solid, m.p. 1350. Re-

crystallization of the solid from hot methanol gave white

crystals, m.p. 146-1470 (reported 147-1490), whose infra-

red spectrum matched perfectly with that of a known sample

of spiro(3,4-dicarboethoxy-2-pyrazoline-5, 1,-2',2,-

diphenylcyclopropane), (XVII).

Reaction of N-nitroso-N-(2,2-diphenylcyclopropyl)-

urea with lithium ethoxide in the presence of benzoyl

chloride.--To an ether solution of 0.60 g. (0.00214 mole)

of N-nitroso-N-2,2-diphenylcyclopropyl)urea and 0.5 ml.

(0.61 g.; 0.00434 mole) of benzoyl chloride cooled to 0

and stirred magnetically was added 0.40 g. (0.00408 mole)

of lithium ethoxide. When nitrogen evolution ceased

(usually 12-15 ml. evolved within an hour), the reaction

mixture was filtered and the filtrate evaporated to an oil.

When this oil was dissolved in a small amount of ether

and pentane was very slowly added with stirring, a yellow

solid (usually 40-80 mg., m.p. ca. 1400 dec.) precipitated










from solution. Repeated recrystallizations from ether

gave pure yellow solid (XX), m.p. 135-135 1/20 (dec.);

N max(iso-PrOH) 302 mp (e = 7280).

Infrared absorption maxima (KBr): 2.85 p, 4.75 p,

6.2 p, 6.35 p, 6.7 p, 6.9 p, 7.35 p, 7.76 p, 8.45 p,

9.45 p, 9.85 p, 12.7 p, 12.85 P, 13.32 p, 13.72 p and

14.3 p.

Anal. Calcd. for C22H18N202: C, 77.17; H, 5.30;

N, 8.18. Found: C, 77.12; H, 5.41; N, 8.13.

n.m.r. resonances: 2.34, 2.47 and 6.06 tau.

The pentane-ether solution from which the yellow

a-diazoketone (XX) had precipitated was evaporated to an

oil. An infrared spectrum (neat) of the oil showed strong

absorption bands with maxima at 4.75 p and 6.2 p and the

absence of a 2.85 u band. The ultraviolet spectrum showed

a maximum at 303 my (iso-PrOH), G = 5670.* Repeated

attempts to crystallize this second diazoketone were un-

successful. Attempts to purify the oil by elution

chromatography (neutral alumina or silica gel) resulted in

decomposition to a solid, m.p. 198-201. Successive re-

crystallizations of this decomposition product from


Since the infrared spectrum of this oil showed a
considerable amount of benzoyl chloride, this extinction
coefficient is only a minimum value.










benzene-pentane gave a pure white solid, m.p. 199 1/2-2000

(very dark melt).

Infrared absorption maxima (KBr): 3.21 P, 5.98 p,

6.25 p, 6.35 p, 6.7 p, 6.9 p, 7.4 p, 8.12 p, 8.47 p,

9.77 u, 10.2 p, 10.85 p, 11.4 p, 11.66 p, 13.0 p (broad),

13.72 p and 14.4 p (broad).

Anal. Calcd. for C44H3202: C, 89.16; H, 5.44;

M.W. 593. Found: C, 88.78; H, 5.82; M.W. 590 (Rast).

When the above reaction was run taking no pre-

cautions to exclude moisture, 5.47 per cent of diazo-

ketone (XX) was isolated. Treatment of the remaining

reaction product dissolved in acetonitrile with p-toluene-

sulfonic acid gave 45.2 per cent nitrogen evolution. When

another run was made under strictly anhydrous conditions,

no diazoketone (XX) was isolable. Treatment of the re-

action product dissolved in acetonitrile with p-toluene-

sulfonic acid gave 56.6 per cent nitrogen evolution.

Reaction of lithium 2,2-diphenylcyclopropyldiazotate

with benzoic acid.--To a refluxing benzene solution of

benzoic acid (3 mole excess) was added dropwise a slurry

of lithium 2,2-diphenylcyclopropyldiazotate in chloroform.

After the addition was complete, the reaction mixture was

heated under reflux for an additional two hours. The

lithium benzoate precipitate was then filtered and the











filtrate washed twice each with water, 5 per cent NaHCO3

and water again before being dried over anhydrous MgSO4.

Evaporation of the solvents under reduced pressure gave a

yellow oil. When this oil was dissolved in pentane, white

needles very slowly formed. These were filtered and washed

with pentane, m.p. 87-890; no depression on admixture with

the material obtained from the reaction of 1,1-diphenyl-

propen-3-ol with benzoyl chloride was observed. The

infrared spectra of the two materials were also identical.

G.l.c. analysis (5 ft. x 1/8 in. column charged

with S.E. -30 silicone fluid on 60-80 mesh Chromosorb W)

of the reaction product oil showed peaks with retention

times identical to those of 1,1-diphenylallene and 1,1-

diphenylpropen-3-yl benzoate in the ratio 63:79. There

was no peak observed with a retention time corresponding

to that of 3,5-diphenylpropen-3-yl benzoate.* An infrared

spectrum of the product oil was consistent with the g.l.c.

results in that all absorption peaks could be explained by

a simple mixture of the allene and 1,l-diphenylpropen-3-yl

benzoate.


This was prepared by reacting 3,3-diphenylpropen-
3-ol and benzoyl chloride.










Thermal decomposition of N-nitroso-N-(2,2-diphenyl-

cyclopropyl)benzamide.--Approximately 100 mg. of nitroso-

amide was heated at 1000 for 30 minutes to effect decompo-

sition. An infrared spectrum of the crude oil showed the

product to be almost entirely 1,l-diphenylpropen-3-yl

benzoate with small amounts of 1,1-diphenylallene and

benzoic acid also present. The result of g.l.c. analysis

(5 ft. x 1/8 in. column charged with PDEAS fluid on 60-80

mesh Gas Chrom. Z) was consistent with the above mixture.

When the product oil was dissolved in ether and the solvent

allowed to slowly evaporate, the benzoate crystallized,

m.p. 87-890. No melting point depression was observed on

admixture with the material obtained from the reaction of

benzoyl chloride and 1,l-diphenylpropen-3-ol and their

infrared spectra were superimposable.

Reaction of N-nitroso-N-(2,2-diphenylcyclopropyl)-

urea with lithium ethoxide in ethanol.--A solution of

0.60 g. (0.00214 mole) of N-nitroso-N-(2,2-diphenylcyclo-

propyl)urea in 10 ml. of anhydrous ethanol was stirred for

5 hours at 260. No gas evolution occurred under these

conditions. Then, 0.206 g. (0.00214 mole) of lithium

ethoxide alcoholate was added. Gas evolution was spon-

taneous and ceased within one minute after the addition of

base. After the solvent was evaporated in vacuo, ether

was added to the gummy residue. An infrared spectrum (KBr)










of the white solid obtained upon filtering the resulting

mixture was superimposable with that of a known sample of

lithium carbamate. The filtrate was evaporated to a

yellow oil. Infrared analysis (film) showed this product

to be 1,1-diphenylallene.

A subsequent experiment demonstrated the stability

of potassium 2,2-diphenylcyclopropyldiazotate to the above

reaction conditions. The potassium diazotate (181 mg.)

was quantitatively recovered after stirring 6 hours at

room temperature in 10 ml. of anhydrous ethanol.

Reaction of N-nitroso-N-(2,2-diphenylcyclopropyl)-

urea and potassium carbonate in ethanol.--To 0.600 g.

(0.00214 mole) of N-nitroso-N-(2,2-diphenylcyclopropyl)-

urea dissolved in 10 ml. of absolute ethanol stirred at

240 was added 29 mg. (0.00021 mole) of K2CO3. Nitrogen

evolution began almost immediately and continued for 6

hours to give 91.7 per cent (average of 2 runs) of the

quantitative amount. The reaction mixture was filtered

and the solvent evaporated in vacuo to a thick tan oil.

After the addition of 14 mg. of menthol (internal standard),

the reaction product was analyzed by g.l.c. (5 ft. x 1/8 in.

column charged with Carbowax 4000 on firebrick). No ethyl

carbamate was observed although an amount corresponding

to a 5 per cent yield would have been easily detected.







62

Infrared analysis of the oil showed an absence of the

absorption bands characteristic of 1,1-diphenylallene.

When the oil was dissolved in ether and the solvent allowed

to slowly evaporate, a solid, m.p. 139-1410, precipitated.

Successive recrystallizations from benzene gave pure white

crystals, m.p. 143-1440. Characteristic infrared absorption

bands (KBr): 2.93 p, 3.05 p, 3.17 p, 3.31 p, 5.79 p,

5.93 p, 6.41 p, 6.54 p, 6.69 u, 6.93 P, 7.32 p, 8.02 p,
9.8 p (broad), 12.9 p, 13.06 p and 14.25 p (broad).

Anal. Found: C, 69.67; H, 6.47; N, 8.81.

Reaction of N-nitroso-N-(2,2-diphenylcyclopropyl)-

benzamide and potassium carbonate in ethanol.--To 0.607 g.

(0.001775 mole) of N-nitroso-N-(2,2-diphenylcyclopropyl)-
benzamide dissolved in 10 ml. of anhydrous ethanol stirred

at room temperature was added 25 mg. (0.0001775 mole) of

K2CO3. After stirring this mixture for several days, the
solvent was evaporated in vacuo to a brown oil. Infrared

analysis of the product showed an absence of the absorption

bands characteristic of 1,1-diphenylallene. However, there

were observed absorption maxima analogous to those obtained

from the ethanolic K2CO3 induced decomposition of N-

nitroso-N-(2,2-diphenylcyclopropyl)urea described above.

Significant infrared absorption maxima (neat):

3.24 p, 3.35 p, 3.45 p, 5.83 p, 5.93 p, 6.28 p, 6.71 u,
6.91 p, 7.3 p, 7.62 p, 7.89 y, 8.53 y, 9.1 u (broad),

9.36 9, 9.77 i, 13.0 p, 13.14 and 14.3 u (broad).










Decomposition of unknown diazoketone by D-toluene-

sulfonic acid.--The crude diazoketone, prepared by the

reported procedure from nitrosourea III, benzoyl chloride

and lithium ethoxide, was obtained as an oil after the

diazoketone (XX) had been removed. The oil was dissolved

in 10 ml. of acetonitrile and cooled to 0 by means of an

ice bath. To this solution stirred magnetically was added

a previously cooled solution of 0.5 g. of p-toluenesulfonic

acid (monohydrate) dissolved in 10 ml. of acetonitrile.

Nitrogen evolution was spontaneous and the solution turned

very dark. After five minutes, the solvent was evaporated

under reduced pressure. The residual oil was dissolved in

25 ml. of ether and this solution was stirred for 15

minutes with 15 ml. of water. The organic layer was then

separated and dried over anhydrous MgSO4. The oil obtained

upon evaporation of the solvent was chromatographed through

neutral alumina (Woelm). A bright orange band eluted with

20 per cent ether-pentane. Upon evaporation of the

solvent, 92 mg. of orange solid crystallized, m.p. 234-

256. Successive recrystallizations from benzene-pentane

gave pure solid, m.p. 235 1/2-236.

Infrared absorption maxima (KBr): 3.22 p, 6.06 p,

6.26 u, 6.7 y, 6.92 9, 7.92 p, 8.42 9, 9.32 p, 9.5 y,

9.76 p, 10.55 u, 13.0 p (broad), 13.25 p, 13.5 y and

14.4 u (broad).










Anal. Found: C, 90.87; H, 5.17; S, 0.10.

Elution with pure ether gave a yellow solution from

which 27 mg. of solid crystallized upon evaporation, m.p.

196-1980. An infrared spectrum (KBr) of this solid was

superimposable with that of the C44H5202 dimeric material

obtained upon elution chromatography or thermal decompo-

sition of the unknown diazoketone.

Preparation of 3,3-diphenylpropen-3-yl benzoate.--

To 5.8 g. (0.0274 mole) of 5,5-diphenylpropen-3-ol

dissolved in 22 ml. of pyridine was added 3.85 g. (3.16

ml.; 0.0274 mole) of benzoyl chloride. This solution was

heated on a steam bath overnight and then poured into

water. Ether was added and the mixture was allowed to

stir for several hours. The organic layer was separated,

washed twice each with water, 5 per cent HC1, 5 per cent

NaHCO and water, and dried over anhydrous MgSO4. Attempts

to purify the benzoate by column chromatography (Woelm

silica gel and neutral alumina) of the oil obtained upon

solvent evaporation were unsuccessful. Thermal instability

of the product precluded purification by distillation or

preparative g.l.c.

Significant infrared absorption maxima (neat):

3.2 u, 5.8 u, 6.25 u, 6.7 p, 6.9 u, 7.85 9, 9.0 p, 13.0 p

(broad) and 14.2 u (broad).










Preparation of N,N-dimethyl-N'-(2,2-diphenylcyclo-

propyl)urea.--To 50 g. (0.21 mole) of 2,2-diphenylcyclo-

propanecarboxylic acid was added 50 ml. (0.695 mole) of

thionyl chloride and the mixture was allowed to stand at

room temperature overnight. After the reaction mixture

was heated under reflux for a short time, the excess

thionyl chloride was evaporated under reduced pressure

from a hot water bath and the final traces were removed

in vacuo.

The remaining heavy oil of acid chloride was added

to about 250 ml. of cold anhydrous acetone. While this

acetone solution was being mechanically stirred, 17.0 g.

(0.262 mole) of sodium azide dissolved in a minimum amount

of water was added rapidly. After stirring at 0 for

three hours, the reaction mixture was added to water and

extracted with ether. The combined ether extracts were

washed twice with water, 5 per cent NaHCOC solution, and

water again. The bicarbonate washings were acidified to

yield 3.7 g. of starting acid, m.p. 167-170 1/20. The

ether solution of acid azide was then dried over anhydrous

magnesium sulfate.

To approximately 500 ml. of refluxing anhydrous

benzene was added the dry ether solution of acid azide

through the top of the reflux condenser at such a rate

that the ether was vaporized as fast as it was added.










The water in the reflux condenser jacket was drained during

this operation to maintain a constant liquid level in the

flask. The benzene solution was then heated under reflux

until no more nitrogen was evolved.

This benzene solution of isocyanate was cooled to

0 and anhydrous dimethylamine was bubbled through the

solution for 45 minutes. The resulting solution was

stirred for several hours and then evaporated under re-

duced pressure to a heavy oil. This oil was dissolved

in ether with stirring and the solution was seeded with

crystals of the urea obtained by cooling a concentrated

ether solution of the above oil in a dry-ice bath. The

solution was cooled to 0 and the resulting precipitate of

N,N-dimethyl-N,-(2,2-diphenylcyclopropyl)urea, after being

filtered, triturated with ether, and dried, weighed

50.4 g. (92.5% from unrecovered acid), m.p. 111-1130.

Successive recrystallizations from chloroform-pentane

mixtures gave white needle-like crystals, m.p. 112-1135;

infrared maxima (melt), N-H 2.98 p, C=O 6.1 p.

Anal. Calcd. for C18H20N20: C, 77.11; H, 7.19;

N, 9.99. Found: C, 77.10; H, 7.10; N, 10.00.

Preparation of N-N-dimethyl-N,-(2,2-diphenylcyclo-

propyl)-N,-nitrosourea.--Five grams (0.0179 mole) of N,N-

dimethyl-N'-(2,2-diphenylcyclopropyl)urea dissolved in








67

18 ml. of a 70 per cent by volume mixture of acetic acid in

acetic anhydride was cooled to 0. To this solution,

stirred mechanically, was added dropwise 2.46 g. (2 equiva-

lents) of NaNO2 dissolved in 6 ml. of water over a one

hour period. The solution was allowed to stir at 0 for

an additional thirty minutes and then 11 ml. of water were

slowly added dropwise. The yellow nitrosourea precipitated

and was filtered, washed three times with water and dried

in a desiccator. Yield 4.1 g. (74.5%); m.p. 88.5-90.5.

Successive recrystallizations from ether-pentane or

chloroform-pentane mixtures yielded a pale yellow powdery

solid, m.p. 87-87 1/20 dec.; infrared maximum (nujol

mull), C=O 5.9 p.

Anal. Calcd. for C18H19N 02: C, 69.87; H, 6.19;

N, 15.58. Found: C, 69.69; H, 5.98; N, 13.59.

Preparation of N-(diphenylmethyl)urea.--A solution

of 20 g. (0.0945 mole) of diphenylacetic acid dissolved in

thionyl chloride was allowed to stand overnight at room

temperature after which all traces of the solvent were

removed by distillation under reduced pressure. The

residual acid chloride was dissolved in anhydrous acetone

and cooled to 00. To this solution stirred mechanically

was added an aqueous solution of 8 g. (50% excess) of

sodium azide. After stirring at 0 for 2 hours, the

reaction mixture was poured into water and extracted with










ether. The ether solution was washed once each with 5 per

cent NaHCO3 and water and dried over anhydrous MgSO4. At

this point, the acid azide, which had begun to decompose

slowly during the aqueous washes, rapidly evolved a gas.

An infrared spectrum of an aliquot of the dried ether

solution was consistent with a mixture of the acid azide

and isocyanate. The ether solution of this mixture was

added dropwise to 250 ml. of dry refluxing benzene at such

a rate that the ether vaporized as fast as it was added.

After the addition was complete, the benzene solution was

refluxed overnight. The benzene solution was then cooled

in an ice bath and anhydrous ammonia was passed through

the cold solution for one hour. After evaporation of the

solution to about 100 ml., the white solid was filtered to

give 15.35 g. (72%), of N-(diphenylmethyl)urea, m.p. 144-

145 1/20. Recrystallization from a mixture of benzene,

ethanol and ether gave an analytical sample, m.p. 144 1/2-

145 1/20.

Significant infrared absorptions (KBr): 2.81 p,

2.95 p, 3.05 p, 3.25 p, 6.05 p, 6.27 p, 6.45 p, 6.71 y,

6.91 p, 7.31 u, 7.61 p, 8.45 u, 8.86 p, 9.31 p, 9.72 u,

11.58 1, 13.2 1, 13.5 u, 14.3 1, 14.5 p.

Anal. Calcd. for C14H14N20: C, 74.31; H, 6.24;

H, 12.38. Found: C, 74.50; H, 6.26; N, 12.59.










Attempted preparation of N-nitroso-N-(diphenyl-

methyl)urea.--Initial attempts to nitrosate N-(diphenyl-

methyl)urea employing method B of White25 (NaN02, acetic

anhydride, acetic acid and the urea) were unsuccessful.

After stirring three hours at 00, the starting urea was

recovered. Subsequent attempts by White's method E23

(N204, sodium acetate and urea in chloroform) resulted in

the consumption of starting urea, but the absence of

isolable nitrosourea. An infrared spectrum of the product

oil showed a large amount of what appeared to be the

isocyanate.













CHAPTER IV


SUMMARY


In the course of the investigations on the lithium

ethoxide induced conversion of N-nitroso-N-(2,2-diphenyl-

cyclopropyl)urea to 2,2-diphenyldiazocyclopropane, two

pieces of information were obtained which exclude the

accepted mechanistic path for this reaction. Ethyl

carbamate, the compound which would result from an initial

carbonyl carbon attack by the base, was found not to be a

reaction product. Moreover, lithium 2,2-diphenylcyclo-

propyldiazotate, the normally presumed intermediate, was

synthesized and found to be totally unaffected by reaction

conditions that led to efficient conversion of the

nitrosourea to the diazocyclopropane. An alternate

mechanism involving attack of the base on the nitroso

nitrogen has been offered. The pyrrolidine induced

decomposition of the nitrosourea proceeded by the tra-

ditional mechanism.

In addition to the lithium diazotate, several other

salts of 2,2-diphenylcyclopropyldiazonium hydroxide were

isolated. The methods of preparation of this unusual and

interesting type of compound were discussed. The reaction

70










of the alkali metal diazotates with acylating agents was

studied and found to give 0-acylation products exclusively.

This anomaly was explained as being due to the nitrogen of

the diazotate being bonded to a rather bulky secondary

carbon that could easily retard attack of the acylating

group on the nitrogen. The reaction of lithium 2,2-

diphenylcyclopropyldiazotate with benzoic acid was also

studied and found to give the ring-opened ester, 1,1-

diphenylpropen-3-yl benzoate.











LIST OF REFERENCES


1. H. von Pechmann, Ber., 27, 1888(1894).

2. A. Hantzsch and M. Lehmann, ibid., 35, 897(1902).

3. C. D. Gutsche and H. E. Johnson, J. Am. Chem. Soc.,
77, 109(1955).
4. F. W. Bollinger, F. N. Hayes and S. Siegel, ibid.,
72, 5592(1950).
5. R. Huisgen and J. Reinertshofer, Ann., 575, 174(1952);
R. Huisgen, ibid., 573, 173(1951); C. D. Gutsche and
I. Y. C. Tao, J. Org. Chem., 28, 883(1963). Also see
C. D. Gutsche, Organic Reactions, Vol. VIII, John
Wiley and Sons, Inc., New York, N.Y., 1954, PP. 389-
390.
6. D. E. Applequist and D. E. McGreer, J. Am. Chem. Soc.,
82, 1965(1960).

7. J. Tempe, H. Heslot and J. Morel, Compt. Rend., 258,
5470(1964).
8. Cf. W. M. Jones, M. H. Grasley and W. S. Brey, Jr.,
J. Am. Chem. Soc., 85, 2754(1963) and references
cited therein.

9. E. Mueller, W. Hoppe, H. Hagenmaier, H. Haiss,
R. Huber, W. Rundel and H. Suhr, Ber., 96, 1712(1963).

10. W. M. Jones, M. H. Grasley and D. G. Baarda, J. Am.
Chem. Soc., 86, 912(1964).

11. F. W. Bollinger, F. N. Hayes and S. Siegel, ibid., 75,
1729(1953).
12. H. J. Backer and T. J. de Boer, Proc. Koninkl.
Nederland Akad. Wetenschap, 4B, 191(1951). Also see
K. Heyns and A. Heins, Ann., 604, 133(1957).










13. Cf. H. Zollinger, Diazo and Azo Chemistry Aliphatic
and Aromatic Compounds, Interscience Publishers, Inc.,
New York, N.Y., 1961, p. 150.

14. J. Thiele, Ann., 376, 239(1910).

15. E. Mueller, W. Rundel, H. Haiss and H. Hagenmaier,
Z. Naturforch, 15b, 751(1960); also see ref. 9.

16. P. Yates, B. Shapiro, N. Yoda and J. Fugger, J. Am.
Chem. Soc., 79, 5756(1957).

17. W. E. Bachmann and W. S. Struve, Organic Reactions,
Vol. I, John Wiley and Sons, Inc., New York, N.Y.,
1942, pp. 39-41.

18. A. L. Crowther and G. Holt, J. Chem. Soc., 1963, 2818.

19. H. Staudinger and C. Machling, Ber., 49, 1973(1916);
see also ref. 17.

20. T. L. Brown, D. W. Dickerhoof and D. A. Bafus, J. Am.
Chem. Soc., 84, 1371(1962).

21. C. D. Gutsche, Organic Reactions, Vol. VIII, John
Wiley and Sons, Inc., New York, N.Y., 1954, pp. 389-
390; see also ref. 7.
22. V. W. Reppe, et al., Ann., 596, 150(1955).

23. E. H. White, J. Am. Chem. Soc., 77, 6009(1955).

Supplementary References

A. R. Huisgen and R. Reimlinger, Ann., 599, 183(1956).

A. Streitwieser, Jr. and W. D. Schaeffer, J. Am.
Chem. Soc., 79, 2893(1957).

E. H. White, ibid., 77, 6013(1955).

E. H. White and C. A. Aufdermarsh, ibid., 83,
1174(1961).

E. H. White and R. J. Baumgarten, J. Org. Chem., 29,
2070(1964).












BIOGRAPHICAL SKETCH


Thomas King Tandy, Jr., was born May 22, 1937, at

Welch, West Virginia. In June, 1955, he was graduated

from Big Creek High School, War, West Virginia. The

following September he entered West Virginia University

where he received the degree of Bachelor of Science in

May, 1960, and the degree of Master of Science in August,

1961. He then continued his education by entering the

Graduate School of the University of Florida in September,

1961. During graduate study he held both graduate and

research assistantships and the position of interim

instructor in the Department of Chemistry.

Thomas King Tandy, Jr. is married to the former

Rhoda Lee Hauck and is the father of two children. He is

a member of the American Chemical Society and Phi Lambda

Upsilon.











This dissertation was prepared under the direction

of the chairman of the candidate's supervisory committee

and has been approved by all members of that committee.

It was submitted to the Dean of the College of Arts and

Sciences and to the Graduate Council, and was approved as

partial fulfillment of the requirements for the degree of

Doctor of Philosophy.

December 19, 1964



Dean, Colle rts and Sciences



Dean, Graduate School


Supervisory Committee:


Chairman







VJL) ^-'~Ac7-Rd




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

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