7-Norbornadienylidene

MISSING IMAGE

Material Information

Title:
7-Norbornadienylidene
Physical Description:
xi, 99 leaves : ill. ; 28 cm.
Language:
English
Creator:
Brown, William T
Publication Date:

Subjects

Subjects / Keywords:
Norbornadienylidene   ( lcsh )
Genre:
bibliography   ( marcgt )
theses   ( marcgt )
non-fiction   ( marcgt )

Notes

Thesis:
Thesis--University of Florida.
Bibliography:
Includes bibliographical references (leaves 94-98).
Statement of Responsibility:
by William T. Brown.
General Note:
Typescript.
General Note:
Vita.

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 000065328
notis - AAH0539
oclc - 04338970
System ID:
AA00003492:00001

Full Text













7-NORBORNADIENYLIDENE


By

WILLIAM T. BROWN















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
1978


































tigitiqld ydiatotnet Archive
in 2010 with funding from
University of Florida, George A. Smathers Libraries with support from Lyrasis and the Sloan Foundation


http://www.archive.org/details/7norbornadienyli00brow














ACKNOWLEDGMENTS


The author would like to acknowledge his research

director, Dr. William M. Jones, for all the guidance and

moral support that he provided.















TABLE OF CONTENTS


Page

ACKNOWLEDGMENTS ..... .......... ........................... iii

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

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

ABSTRACT ................................................. ix

INTRODUCTION ........................................... 1

RESULTS AND DISCUSSION ................................. 19

EXPERIMENTAL ............................................. 61

General. ............................................ 61

7-t-Butoxynorbornadiene (67). ....................... 61

7-Chloronorbornadiene (50). ......................... 62

Silver Isocyanate (71). .............................. 62

7-Norbornadienylisocyanate (68) ..................... 63

Methyl N-7-Norbornadienylcarbamate (69) ............. 64

p-Chlorobenzoyl Nitrite (73). ....................... 65

Methyl N-Nitroso-N-7-norbornadienylcarbamate (64). .. 65

Bases Used in the Attempts to Generate 7-Norborna-
dienylidene (1) from Methyl N-Nitroso-N-7-norborna-
dienylcarbamate (64) and N-Nitroso-N-7-norborna-
dienylurea (65) .... ................ ................ 66

Decomposition of Methyl N-Nitroso-N-7-norborna-
dienylcarbamate (64) by Sodium Methoxide. ............ 68

Decomposition of Methyl N-Nitroso-N-7-norborna-
dienylcarbamate (64) by Lithium Ethoxide. .......... 69

7-Norbornadienylurea (70) .... ............... .... 70








Page
N-Nitroso-N-7-norbornadienylurea (65). .............. 71

Typical Decomposition of N-Nitroso-N-7-norborna-
dienylurea (65) under Basic Conditions. ............. 72

Decomposition of N-Nitroso-N-7-norbornadienylurea
(65) in Heptane at 900 ............................... 72

Decomposition of N-Nitroso-N-7-norbornadienylurea
(65) in Heptane in the Presence of Dimethylfumarate
at 900 ........................................... ... 73

Decomposition of N-Nitroso-N-7-norbornadienylurea
(65) in Heptane in the Presence of Lithium Carbonate
at 90 ................. .......................... 74

Decomposition of N-Nitroso-N-7-norbornadienylurea
(65) in Triglyme at 2000.. .......................... 75

Experimental Method for Thermolysis of N-Nitroso-
N-7-norbornadienylurea (65) in a Hot Tube. .......... 75

Typical Thermolysis of N-Nitroso-N-7-norbornadienyl-
urea (65) in a Hot Tube at 300. .................... 78

Typical Thermolysis of N-Nitroso-N-7-norbornadienyl-
urea (65) in a Hot Tube at 4000 .................... 78

Thermolysis of N-Nitroso-N-7-norbornadienylurea
(65) in a Hot Tube at 4000 and Trapping with Tetra-
cyanoethylene (TCNE) ....... ......................... 80

Sodium Salt of Benzaldehyde Tosylhydrazone. ......... 81

Thermolysis of the Sodium Salt of Benzaldehyde
Tosylhydrazone (113) in a Hot Tube at 4000.'......... 81

Heptafulvalene (14). ................................. 81

Thermolysis of Heptafulvalene (14) in a Hot Tube at
4000 .............. ......... .......................... 81

Isolation and Identification of syn- and anti-Penta-
cyclo[10.2.0.04,12.05,11.08,1]tetadeca-2,6,9,13-
tetraene (18) and (19) ............ ........ .......... 82

Thermolysis of syn- and anti-Pentacyclo
[10.2.0.04,12.0 ,5 08, 1-]tetradeca-2,6,9,13-tetraene
(18) and (19) in a Hot Tube at 3500. ............... 82









Page

APPENDIX ................................................. 84

REFERENCES AND NOTES ...................................... 94

BIOGRAPHICAL SKETCH ................................... 99















LIST OF TABLES


Page

Table 1. Decomposition of N-Nitroso-N-7-norborna-
dienylurea (65) Under Basic Conditions. ................ 72

Table 2. Thermolysis of N-Nitroso-N-7-norbornadienyl-
urea (65) in a Hot Tube ................................ 80


vii















LIST OF FIGURES



Page

Figure 1 ............. ......... ......................... 36

Figure 2 ................................................. 37

Figure 3 ............. ........ .......................... 38

Figure 4 .............. .................. ............... 39

Figure 5 ................. .......... .................... 40

Figure 6 ........................... .................... 41

Figure 7 .......................................... 42

Figure 8 .......... .............. ...................... 76

Figure 9 ............................................... 83


viii














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


7-NORBORNADIENYLIDENE

By

William T. Brown

June 1978

Chairman: William M. Jones
Major Department: Chemistry

7-Norbornadienylidene (1) has been particularly

resistant to generation in the past. Not only would it

fill an important gap in the present knowledge of the C7H6

energy surface but it would also be the premier example of

a group of carbenes known as foiled methylenes. A systema-

tic study has been carried out using methyl N-nitroso-N-

7-norbornadienylcarbamate (64) and N-nitroso-N-7-norborna-

dienylurea (65) as potential precursors for (1). Both (64)

and (65) were prepared by treating the corresponding

unnitrosated material with p-chlorobenzoyl nitrite (PCBN)

(73) at -450 in ether. Treatment of (64) and (65) with

various bases in a variety of solvents and under a number

of conditions gave no evidence for any carbene products.

Pyrolysis of (65) in heptane at 900 and in triglyme at 2000

gave no products which could be linked to (1).








When the normal method of flash vacuum thermolysis was

used on (65), no useful results were obtained owing to the

relatively low volatility of (65). However, when (65) was

pyrolyzed by dropping it down a heated inclined tube which

was being continuously evacuated and the products condensed

in a liquid nitrogen trap,the following results were

obtained: from 200 to 3000 a mixture of syn- and anti-

pentacyclo[10.2.0.04,12 .05,11. 0811]tetradeca-2,6,9,13-

tetraene (18) and (19) respectively was obtained in yields

averaging 14%, from 250 to 3500 heptafulvalene (14) in yields

averaging 15%, from 200 to 4000 benzene (22) in yields of

about 3%, and toluene (23) at 200 and 2500 in a 4% yield.

In one experiment at 4000 the volatile products were allowed

to react with tetracyanoethylene (TCNE). Nuclear magnetic

resonance analysis of the products showed weak signals

corresponding to those published in the literature for

5,5,6,6-tetracyano-7-vinylidenebicyclo[2.2.1]hept-2-ene (114).

Thus the presence of fulveneallene (12) was strongly

suggested.

These results suggest an attractive mechanism. Pyrolysis

of (65) results in the generation of (1) which then either

fragments to give (22) and a carbon atom or rearranges to

bicyclo[3.2.0]hepta-l,3,6-triene (6). The bicyclic triene

(6) then, depending on the temperature, dimerizes to give

the mixture of isomeric dimers (18) and (19), rearranges to

cycloheptatrienylidene (3) which then dimerizes to give the

observed (14), or (6) rearranges to (12). Three alternative








mechanisms, one involving phenylcarbene (2) as the

progenitor of (14); another involving dimerization of (1)

to give binorbornadienylidene (26) and subsequent

rearrangement of (26) to (14), (18), and (19); and

another involving generation of bicyclo[2.2.1]hepta-1,2,5-

triene (123) and subsequent rearrangement to 4-bicyclo-

[3.2.0]hepta-2,6-dienylidene (124) followed by rearrangement

of (124) to (6); cannot be entirely discounted.














INTRODUCTION


7-Norbornadienylidene (1) has been in the past particular-

ly resistant to generation. Either the popular methods for

generation of carbenes simply did not work or the precursors

needed for the method were not synthetically available.3

This left a significant void and was quite discouraging

since (1) remained as one of the last species on the C7H6

energy surface to be generated which was chemically reasonable.








(1)


For some time now interest has been high in the various

isomers which comprise the C7H6 energy hypersurface.427

The isomers which have been generated or implicated to date

are shown below. A considerable amount of energy has been

expended by various research groups in determining the

relationships between these isomers. For example, genera-

tion of phenylcarbene (2) from various precursors in the

gas phase led at temperatures below 5800 to heptafulvalene













(3)
(3)


(4)


, H


H


(8) (9)


(10)


(11)


(12)
(12)


(13)


(14), a ring expansion product.4'5 At temperatures above

5800 fulveneallene (12) and ethynylcyclopentadiene (8), both

ring contraction,products were formed.56

As regards the ring expansion the equilibria, which may

exist between the four species shown below have received

considerable attention.12,14,18-20,22,23 There is little

doubt as to the intermediacy of bicyclo[4.1.0]hepta-2,4,6-

triene (5).22 However, the identity of the intermediate

which leads to (14) whether it be cycloheptatrienylidene (3)










>5800


(12)


(8)


I I


(14)



and/or cycloheptatetraene (4) is still being investigated.


(4)


In the ring contraction ethynylcyclopentadiene (8) has

been shown to be a secondary product of fulveneallene (12)

whenever fulveneallene (12) is exposed to higher temper-
20
atures. Several mechanisms have been presented by workers


H


(12)








through the years to explain the ring contraction.64,20'24
24
One of these, first proposed by Wiersum and Hageman,4 involved

rearrangement of (2) to methylenecyclohexadienylidene (13)

which then rearranged to (12). Recently, however, evidence based



H >_-


(2) (13) (12)



on 1C labeling studies has been presented by C. Wentrup and

coworkers which discounts this and other previously suggested

mechanisms and favors one in which ring expansion occurs

and then is followed by rearrangement of (3) to bicyclo-
21
[3.2.0]hept-1,3,6-triene (6) at high temperature. The

bicyclic triene (6) then gives (12) presumably via the free

radical (15).



H

(2) ** **
(2) (3)


(12)


(15)








The bicyclic triene (6) has been generated independently

by two groups of workers.2527 Hoffman elimination of (16)26

and dehydrohalogenation of (17)25 both led to (6). However,

the monomer (6) could in no case be isolated. Instead a

pair of isomeric dimers, syn- and anti-pentacyclo-

[10.2.0.04'12.05'11.08'11]tetradeca-2,6,9,13-tetraene (18)

and (19) respectively, was obtained. Presumably these were

a result of 2 + 2 cycloaddition across the 1,2 double bond

of (6).26


NMe 3I1
3


(16)


(18)


(17)


(19)


Pyrolysis of benzocyclopropene (11) at 5000 gives (12).17

The cyclopropyl ring of (11) breaks to give (13) which then

rearranges to (12). Other precursors to (13) also give (12)

upon gas phase thermolysis.11,15
















(13)


Bergman found that gas phase flow thermolysis of 1,2-

diethynylcyclopropane (9) at 4800 and 1 atmosphere gave

bicyclo[3.2.0]hepta-1,4,6-triene (10).16 When (10) was

generated at a low pressure or when (10) was heated at 5800


(20)


7000





500


(13)


(12)


(21)



(12) and (8) were produced. Wolf and Shevlin generated

quadricyclanylidene (7) and found that it extrudes its C-7


(11)


(12)












4800
l/ Atm


(10)


580


480
Low Pressure


"*9 H

(9)


H





(12) (8)

(12) (8)


carbon atom to give benzene (22) and toluene (23) as products.8


(22) (23)

(22) (23)


In the case of 7-norbornadienylidene (1) a number of

possibilities come to mind. For instance, (1) could

rearrange by what is formally a 1,2 shift to give (6). Moss

et al. have found that 7-norbornenylidene (24) rearranges to

bicycle [3.2.0 hepta-1,6-diene (25).28 Another possible


(1) (6)
















(25)


fate is that (1) could simply dimerize to give binorbor-

nadienylidene (26). Another possibility would be loss of


(26)


the carbene bridge to give benzene (22) and a carbon atom.

This would of course correspond to the behavior found for
8
quadricyclanylidene (7). While on the subject of (7) it

is conceivable that (1) might isomerize to (7). However,

this is unlikely because from simple thermochemical


(22)


0C


(24)


ve










(1)





(1)


(7)


considerations (7) is calculated to be 40 kcal/mole higher

energy than (1). Finally, (1) could undergo what might be

called a retro-Skatteb 122 rearrangement to give the unknown

carbene norcaradienylidene (27). However, this type of

rearrangement is without precedent and is only known to

take place in the opposite sense as in the case of the

rearrangement of vinylcyclopropylidenes to cyclopentenyli-

denes.22
denes.


(27)


In addition to the significance of (1) as a C7H6 isomer

it remained as the premier example of a group of carbenes

known as foiled methylenes or nonclassical carbenes.28

By definition "these are systems in which an artificial

energy minimum is created as a result of initial stabiliza-




. IV


tion due to the inception of a typical facile methylene

reaction, such as addition to a double bond, which is

foiled by the impossibility of attaining the final product

geometry."29 5458 This can be viewed in a general way as shown

below. In (28) the carbene cannot add to the double bond










(28) (29)








(30) (31) (32)





because of the impossible geometry that would result. How-

ever, the carbene center could interact with the double bond

and one can draw two resonance structures which' together

result in the nonclassical structure (32). In principle

then, such an interaction would have a stabilizing influence on

the molecule.

Interest in these types of carbenes has been particularly

high since Gleiter and Hoffman stated in 1968 that foiled

methylenes may be a way to obtain a carbene with a singlet

ground state.29 The structures for singlet and triplet
ground state. The structures for singlet and triplet







carbenes are shown below. Gleiter and Hoffman stated that


T0
H C H



Triplet
sp Hyberdized


Singlet
2
sp Hyberdized


in order to obtain a singlet ground state the degeneracy of
the triplet methylenes p and p orbitals would have to be
destroyed and that one way to accomplish this would be to
force one of the p orbitals to interact with a high lying
29
occupied level.29 In principle this would be the case with a
foiled methylene. In addition it is clear from structures
(30), (31), and (32) that the carbene center may have taken
on increased neucleophilic character. In this way (1) may
be similar to cyclopropenylidene (33) and cycloheptatrienyli-
dene (3).29


-O


(33)








A considerable amount of work has been done by various

research groups to date on foiled methylenes.28-43 The

particular ones that have been studied so far are shown below.



00






(24) (34) (35) (36) (37)




00





(38) (39) *, (40) (41) (42)




A detailed review of this chemistry will not be attempted here.

However, Moss and others have generated 7-norbornenylidene (24) and

studied its behavior in detail.28,40-42 Moss et al. have found

that a mixture of products was formed when they pyrolyzed the

lithium salt of 7-norbornenone tosylhydrazone (43) at 1900

with the major product being the bicyclic diene (25).2 The

chemistry of the other foiled methylenes has been similar;

that is, products have been formed which suggest interaction

between the carbene and the double bonds. Nevertheless,

these products can all be explained by a simple 1,2 shift

not involving any kind of nonclassical structure. To this

date no definitive evidence for either case has been obtained.









Li
N-N-Tos
//


(43)


(24)


(25)


Attempts to generate 7-norbornadienylidene (1) by using

the popular methods for the generation of carbenes failed.1

These attempts centered on the Bamford-Stevens reaction4 and

a-elimination of hydrohalides.45 In the case of the Bamford-

Stevens reaction which involves the thermolysis of alkali

metal salts of tosylhydrazones the problem centered around

preparation of the tosylhydrazone itself.1 Tosylhydrazones


R- M A
C=N-N-Tos --
R


R
C: + N2 + MTos
/
R


are normally prepared from the corresponding ketone and

tosylhydrazide. However, (44) is inherently unstable.46


R. H+
C=O + NH NH-Tos --
2


R H
C=N-N-Tos + H20
2
R










0 0






(44)
Fe (CO)

(45)






Preparation of the iron tricarbonyl complex of 7-norbornadiene

(45) may have opened the door; however, attempts to prepare

the tosylhydrazone led only to the ketal type material (46).

Jones and Ledford attempted to prepare the tosylhydrazone



HH HH
TOS-NN NN-TOS





Fe(CO)

(46)



from 7,7-dimethoxynorbornadiene (47). Although this led
2
to (48), it could not be dehydrated. In the case of

hydrohalide elimination, an attempt to generate (1) from










NNTOS


)CH3


+ TOS-NNH2


(47)


-H 20
- Ly


(48) (49)


7-chloronorbornadiene (50) in the presence of n-butyl

lithium led only to products derived from attack on the

double bond by the base.3


1


+ n-BuLi


(50)


-- Products


(51)


Clearly, another type of precursor was called for if

(1) was to be generated. For some time now a method used

with success to generate carbenes has been the basic de-
47
composition of N-nitroso amide derivatives. As shown this

involves production of the corresponding diazoalkane which

then may lose nitrogen to give the carbene. For example,

H Base-N2
RR2 CNCOR3 >Ba RlR2C=N2 >-N R R1R2C:
NO








Jones and Muck discovered that treatment of (52) with
48
lithium ethoxide in ether gave the diazoalkane (53). In

addition Jones and Warner found that treatment of (54) under


Ph
Ph I


(52)


-NCOEt
N
NO


LiOEt
Ether
Ether


(53)


carefully controlled conditions with sodium methoxide in

the presence of dimethylfumarate gave up to 60% of the

spiro adduct (55).49 Moreover, decomposition of (56) with
spiro adduct (55). Moreover, decomposition of (56) with


Ph 0
"NO O
i COCH3
NCOCH + -3-
II 3
0 COCH3
Ph 0


(54)


0
Ph COCH3


NaOMe
Ph K COACH3
(55)
(55)


base led to yields of up to 13% of (14).50 In addition to

the treatment of N-nitroso amide derivatives with bases to

produce diazoalkanes, N-nitroso ureas have been thermolyzed

and given diazoalkanes in a few instances. It was found

that thermolysis of (57) in heptane at 800 gave up to 95%












Base



(14)


of 1,1-diphenylallene (58).51 Furthermore, the thermolysis


Ph
NO 80
-NCNH -----
i 2 n-Heptane
0


(57)


(53)


Ph
Ph


(58)


of N-nitroso-N-methylurea (59) gave diazomethane (60).

However, in this case the carbene was not formed because

(60) was trapped by cyanuric acid (62), formed as a by-pro-
52-54
duct in the reaction, to give trimethyl isocyanurate (63).52

NO 0
I A i
CHBNCNH2 -> CH2N2 + HNCO 3HNCO --- HN NNH
0 ] [


(60)


(61)


0
HN NH
CH2N2+ O N O

H


(62)


O"N 'ZO
(61) H
(62)


0
Me, NN Me



Me
(63)


NO

I
-NCOCH
O


(56)


(59)


(60)









NO NO
NCOCH NCNH
If 3 2
O O



(64) (65)


Therefore, two target molecules were picked, methyl

N-nitroso-N-7-norbornadienylcarbamate (64) and N-nitroso-

N-7-norbornadienylurea (65), to be synthesized which would

be used to attempt to generate (1).













RESULTS AND DISCUSSION


The syntheses of the two potential precursors (64) and

(65) were accomplished in five steps. The basic scheme

without reagents and conditions is shown below.

Two of these steps, the conversion of 7-chloronorborna-

diene (50) into 7-norbornadienylisocyanate (68) and

especially conversion of methyl N-7-norbornadienylcarbamate

(69) or 7-norbornadienylurea (70) into either (64) or (65)

respectively, were rather problematic.


(66)


(67)


(50)


-NCOR
NO


R=-OCH3 (69)
R=-NH2 (70)


R=-OCH3 (64)
R=-NH2 (65)


(68)


$-




L. U


7-t-Butoxynorbornadiene (67) was either used as obtained

from Frinton Labs or it was prepared according to the method

of Story and Fahrenholtz. This was an Organic Syntheses

preparation and involved treatment of norbornadiene (66)

with t-butyl perbenzoate in the presence of cuprous bromide.55


0


0
+ PhCO0-


(66)


CuBr
Benzene


(67)


The chloride (50) was prepared by the

Saunders with minor modifications.56

ether (67) with dry hydrogen chloride

chloride gave a satisfactory yield of

(50).


0+
+ CH3CC1


(67)


method of Story and

Treatment of the

in refluxing acetyl

the desired material


HC1
___>V


(50)


Preparation of the isocyanate (68), however, was somewhat

more problematic. Although the following procedure was

sound the yield reported was not obtained immediately.

Synthesis of (68) was accomplished by treatment of the








chloride (50) with silver isocyanate (71) using refluxing

liquid sulfur dioxide as the solvent. The reaction was

carried out in darkness to protect (71) which is light

sensitive. The isocyanate (68) was isolated in a 94% yield

as an orange oil. Such a yield was not obtained in the early



Cl N=C=0

Darkness
+ AgNCO SO2


(50) (71) (68)



runs and low yields plagued the synthesis. Normally the

isocyanate (68) was immediately used upon its preparation

to make the carbamate (69) and the urea (70) since in a few

cases polymerization of (68) had taken place. Because of

this procedure the reason for the low yield was unclear.

However, examination of the crude reaction mixture in one

of the urea runs via its nmr spectrum showed another

material besides the urea to be present in significant

quantity. That this material had originated earlier in the

reaction sequence was shown by examining the nmr spectrum

of the isocyanate (68) used to prepare the urea. Indeed a

material with identical spectral characteristics was a

major proportion of the isocyanate mixture. This material

was isolated by column chromatography and identified as

7-norbornadienylether (72). The spectral data were as

follows: NMR (CDC13) T 3.3-3.6 (m, 8H, olefinic),








5.5-5.7 (m, 2H, C-7), 6.3-6.5 (m, 4H, bridgehead); ir (neat)
-l
1100, 1010 cm-1 on Beckman IR 10; ms m/e 198 (M+), 108, 107,

91 (base), 79, 78, 77, 65.

The silver isocyanate (71) that was being used was

prepared in the laboratory by precipitating (71) from an

aqueous solution of potassium cyanate with an aqueous

solution of silver nitrate. However, a check of the

literature uncovered the fact that (71) prepared from the

above materials without proper pH control gives a product

contaminated with silver oxide and silver carbonate.5 When

the method of Birkenbach and Linhard was substituted, the



0





(72)


pH of the potassium cyanate solution was brought to 6.2 with

1.0 M nitric acid before mixing with the silver nitrate

solution, a product free of oxides and carbonates was
58
obtained. Use of this silver isocyanate (71) gave the

high yield of (68) reported.

The carbamate (69) was prepared by stirring (68) at

room temperature with methanol in pentane. This gave (69)

as a white solid in a 19% yield based on starting chloride

(50). The urea (70) was prepared by treating (68) with

ammonia in benzene at room temperature. Recrystallization













+ MeOH Pentane


(68) (69)


of crude (70) from hot tetrahydrofuran gave (70) as a beige

solid in a 26% yield basedonthe isocyanate (68).

Synthesis of the two target molecules (64) and (65)




H
NCO NCNH2

+ NH3 Benzene

(68) (70)






from either (69) or (70) respectively was accomplished after

the correct nitrosating reagent was found. Treatment of

(69) or (70) with p-chlorobenzoyl nitrite (PCBN) (73) in the

presence of freshly fused anhydrous sodium acetate at -450

using anhydrous ether as a solvent gave (64) or (65)

respectively. The N-nitrosocarbamate (64) was obtained as a

yellow oil in 79% yield. The N-nitrosourea (65) was

obtained as canary yellow needles in a 38% yield. It was

initially felt that the most troublesome step in the

synthesis would be this last step and that turned out to








be the case. Although the reaction worked quite well as


+ Cl \ CONO


(69)
(70)


Ether
NaAc
-450


(73)


NCOR
NO




R=-OCH3 (64)
R=-NH2 (65)


reported above, this was not the case until the correct

nitrosating reagent was found. The traditional reagents

for nitrosating amide derivatives have been sodium nitrite

and hydrochloric acid, nitrosyl choride, or dinitrogen

tetraoxide.59 However, it is well known that reagents such

as nitrosyl chloride and dinitrogen tetraoxide readily
59
attack double bonds as well as nitrogen.59 For example,

treatment of norbornene (74) with nitrosyl chloride gave

the adduct (75).60 Reaction of (74) with dinitrogen tetra-

oxide gave the three products (76), (77), and i(78).61


+ NOC1 --


(74)


(NO
Cl

(75)


R=-OCH3
R=-NH2


--Dimer











+ N204 ~NNOO2 +NO2
(74) (76) (77 2
(77) 2

N02
+ ONO
(78)
Nevertheless, Jones and coworkers found that (79) and (80)
could be successfully nitrosated with dinitrogen tetraoxide

under the correct conditions without disturbing the double
bonds.50,62



Ph
H H 11
| NCOCH3 p NCOCH3
P h

(79) (80)


Therefore, the initial attempts to nitrosate were
made using dinitrogen tetraoxide. These experiments were
made using either the carbamate (69) or, in order to con-
serve starting material, a model system consisting of
methyl N-cyclohexylcarbamate (81) and (66). A wide variety
of conditions were tried in which solvent, temperature,
mode of addition of reagents, method of stirring, and other
conditions were changed. In all cases only dark brown oils
were obtained. No evidence of nitrosation of the nitrogen
was found in either (69) or the model system. Clearly, a










0
H I
-NCOCH3


+ N204


-COCH3 + + N204


(81) (66)


NO
NCOCH

(82)
(82)


different approach was called for and p-chlorobenzoyl nitrite

(73) was chosen as an alternative nitrosating reagent. It

was chosen primarily because of its ease of preparation and

the fact that it had been successfully employed to nitrosate

(83) in the presence of (84) and (85).63 Indeed, when (73)


(Ph) 4 CH
H[ I 3
PhNCCH + 0 + CH=CCO CH3
N( 3 2( 3

(83) (84) (85)


Cl- CONO


(73)


was used to nitrosate (69) and (70) the successful results

reported were obtained. Any explanation offered here for the

effectiveness of (73) as a nitrosating reagent would be

rather presumptuous.


(69)


(64)


NO

PhNCCH
If 3
0


(86)








Generation of diazoalkanes from N-nitroso amide
47
derivatives under basic conditions is a well known process.4

Also in a few instances diazoalkanes can be generated from
51-54
them by thermolysis. W. M. Jones and coworkers

studied the mechanisms involved in both the base and thermal

reactions of N-nitroso amide derivatives in detail.48,52,64

For the base induced decomposition of N-nitroso carbamates

Jones found that there are two mechanisms which might come

into play depending upon the conditions under which the

decomposition is done.48 However, the important thing

concerning diazoalkane production is not which of these

mechanisms is being followed but whether a most serious side

reaction may occur. Briefly, the two mechanisms which are

thought to be involved are as follows. First, the reaction

may proceed by initial attack of the base on the carbonyl

carbon leading to the diazohydroxide (90) which may then lose

water to give the diazoalkane (91). The mechanism by which

N=0 N=0 B
HI BE H IT / -R30CB
R2CN-COR R R2C-N-COR >
0 O9
(87) (88)



H E H H -H20
R R2C-N=N-O R R2C-N=N-OH > R R2-C=N2

(89) (90) (91)


water is lost is uncertain; however, in any case the hydrogen

on the carbon attached to the diazo group must be lost if a








diazoalkane is to be produced. Second, the initial attack

of the base may be on the nitroso nitrogen. Loss of either

(94) or (95) leads to either (96) or (97) respectively.

The intermediates (96) or (97) then may lose either (98)

or (99) respectively to give the diazoalkane (91). Which

one of these mechanisms is operable depends on the alkyl

group, the group attached to the carbonyl carbon atom,
48
the solvent, and the nature of the base. More importantly,

the intermediate diazohydroxide (90) or the related inter-

mediates (96) or (97) may give rise to a serious side

reaction.


N=O
H I B9
R R2CN-C-OR -
1 2 1, 3
0

(87)


1 a b
B ', a) -1
N-O
H Iy b) -
R R2C-N-C-OR
1 3
O0


B
N-O
HI i
R R2CN-C-OR 3

(92)


3 (94)

R30CO2 (95)


(93)


H 0
R1R2C-N=N-OCOR3

-B


-R3OCO2H (98)

-BH (99)


(96)

(97)


R1R2C=N2

(91)








Namely, ionization of (90), (96), or (97) to give the

diazonium ion (100) may occur before elimination, by what-

ever mechanism, of the hydrogen on the carbon attached to

the diazo group.6

This results in noncarbene products being formed.


H H
RIR2C-N=N-OR3 --> R R2C-N2+OR3 -> Noncarbene Products
(100)

R3=H (90)

R3=CO2R (96)


OR3=B (97)

Unfortunately this is sometimes the case. For example,

Gutsche and Johnson found that when (101) was treated with

potassium carbonate in methanol in the presence of cyclo-

hexanone (102) the expected product (103) was not isolated.

Instead (104) was isolated as the product.66







0 0 \/ (103)
NO K O CH3
CH -/ CH2NCO2Et + -C
CH3OH
(101) (102)



CH3 CH20CH2


(104)








As regards diazoalkane generation from N-nitroso ureas

the same side reaction, ionization, may occur whether the

decomposition is induced with a base or thermally. For

N-nitroso ureas the initial site of attack by the base is

thought to be on the nitroso nitrogen to give (106). A

proton transfer gives (107) which can lose isocyanic acid

to give either (90) or (97) depending on whether route a or

b is taken. If (90) or (97) loses (108) or (94) respectively

the diazoalkane (91) is the product.48'64

N=O N-0
HI B8 HI
R R2CNCNH2 R1R2CNCNH2
0 0
(105) (106)



a E8 (94)
B b a) -B
N-OH b) -OH (108)
H 1 ~e- -HNCO
R1R2C-N'-CNH
0
(107)


-OH8 (108)
H _
H or -B (94)
R1R2C-N=N-OH (90) ) R R2C=N2
-B (97) (91)


In the case of thermal decomposition of N-nitroso ureas the

reaction is thought to involve loss of isocyanic acid to

give the diazohydroxide (90).52 For simplicity this

reaction is shown as being concerted. The diazohydroxide

may then proceed as discussed previously. At any rate









N=0/O H
H I IK H A H -HO2
RR2CN-C-N -HNCO > RR2C-N=N-OH > RRC=N
O
0

(105) (90) (91)

whether base or thermally induced, ionization may take place

before the hydrogen on the carbon attached to the diazo group

is lost. In addition to ionization there is another problem

associated with the thermal reaction. Namely, the isocyanic

acid or its trimer (62) produced as aby-product may trap the

diazoalkane once it is produced.52 A reaction of this type

was found when Huisgen and Reimlinger decomposed (59) in

benzene and isolated (63) as the product.53



0

NO CH 311CH3
CH NCNH
-3 n 2 Benzene
O N

(59) CH3

(63) i


Unfortunately, these problems and possibly others

plagued all attempts to generate (1) from (64) or (65) in

solution. When (64) was allowed to react with a mixture

of sodium methoxide and methanol using ether as a solvent

at either -500 or at reflux,only complex mixtures of

unidentifiable products were produced. One run was made in

the presence of dimethylfumarate in an attempt to trap (1)








if it were present; however, nothing except the product of

Michael addition of sodium methoxide to dimethylfumarate

was found. In no run was there any evidence found for a

carbene product. When (64) was treated with a 50% excess


NO
NCOCH
11 3
O -50 or
+ NaOC/CHOH Reflux
+ NaOCH /CH OH r >- No Carbene Products
3 3 Ether


(64)


of lithium ethoxide at room temperature in ether, methyl

7-norbornadienylcarbonate (109) was isolated as the sole

product in a 26% yield. This was quite discouraging since


NO

NCOCH
11 3
0


+ LiOEt
50% Excess


(64)


OCOCH3
0


RT
Ether


(109)


a product such as (109) is evidence of the ionization side

reaction discussed previously. A mechanism which may

explain (109) is shown below. When (65) was treated with

a variety of bases under a variety of conditions and in

some runs in the presence of carbene traps (see Table 1),

again only complex mixtures of unidentifiable products were












EtO-N-O
-NCOCH
I f) 3


EtO


(64)


-EtO9


(110)


(111)


-N2


(112)


(109)


obtained. In no run was there any evidence for a carbene

product. At this point attempts to generate (1) via

reaction of (64) or (65) with base were abandoned.


ICONH2


+ Base --No Carbene Products


(65)


Pyrolysis of (65) in n-heptane at 90 under an argon

atmosphere gave a mixture of products with the nmr spectrum

shown in Figure 1. The presence of 7-norbornadienylisocyanate

(68) as one of the products in this mixture is suggested by

the multiplets which appear at T3.1-3.3 and 6.2-6.5 since















90
n-Heptane-
(65)





they correlate with the spectrum of authentic (68). As

outlined earlier the origin of an isocyanate such as

(68) could possibly arise from entrapment of the diazo-

alkane or conceivably even the carbene by isocyanic acid.52

Unfortunately, when a reaction was carried out under the

same conditions in the presence of excess lithium

carbonate no differences in the products were observed.

Possibly more interesting than the possible presence of (68)

are the other nmr signals in Figure 1. It is believed that

these signals T3.4-3.9 and 6.9-7.1 with an integral ratio of

5 to 2 respectively probably belong to a single material.

(In another run under similar conditions an almost identical

nmr spectrum (see Figure 2) was obtained of the crude

products. However, the integral ratio of the signals at

T3.4-3.9 and 6.9-7.1 was still 5 to 2 whereas the ratio

between these signals and the signals at T3.1-3.3 and 6.2-

6.5 was different.)

Presumably, a rearrangement has taken place since the

resonance pattern from T3.4-3.9 no longer resembles the








the pattern so characteristic of the olefinic protons of

a norbornadienyl moiety. Quite unfortunately all attempts

to isolate this material failed miserably. Even low

temperature column chromatography failed. Nevertheless,

some insight may be gained into the nature of this

material by examining the results of a series of trapping

experiments that were run.

In one experiment methanol was added to the crude

product mixture and the progress followed by nmr spectroscopy.

The nmr spectrum that was obtained is shown in Figure 3.

Several observations can be made. The rearranged material

reacted rapidly with the methanol. In fact it appears to

have reacted faster with the methanol than the material

which is presumably (68). In addition when an attempt to

isolate the adduct was made via preparative thin layer

chromatography only a small amount of the methyl carbamate

(69) was obtained.

In another experiment the reaction was carried out in

the presence of an excess of dimethylfumarate.. The nmr

spectrum of the product mixture is shown in Figure 4. The

major observation which may be made here is that the fumarate

has formed an adduct with something. An attractive explan-

ation would be that it has trapped some material intermediate

to the rearranged material. Isolation of this material was

foiled by its inadvertent exposure to ethanol. The results

of this exposure are shown in Figures 5 and 6. Apparently

the new adduct has reacted with the ethanol. An attempt to




















































a)
- r
44











;i


~.U t~


S -4
r:I
- >-










* 9


) 4




:I
r34


---


-7



























































tt~

a,
k



Fr(
H
























































Ln

S ,



r-I






























































a,
k
~
~I
-r(
cLI























































4
4
: 0
Nr








isolate some meaningful material from this mixture via

column chromatography failed as shown in Figure 7. At this

point attempts to identify the rearranged material were

stopped mainly for pragmatic reasons and since it was most

likely that it was not a carbene product especially in

light of the integral ratio of 5 to 2.

A final attempt to generate (1) in solution was made

when (65) was thermolyzed in triglyme at 2000. No evidence

was obtained for any carbene product. In fact, the only

product isolated was a trivial one identified as a contamin-

ant in the triglyme. Therefore, in conclusion, no evidence

was found for any carbene type products in any of the

experiments that were undertaken in solution although it

may be interesting if only for curiosity's sake, to learn

the identity of the rearranged material in the heptane

experiments.






NCONH
12
NO
2000
--Tr--ly> No Carbene Products
Triglyme

(65)



With the absence of any evidence for carbene products

in any of the solution experiments, gas phase pyrolysis

experiments of the N-nitroso urea (65) were initiated.

Although a static pyrolysis system would normally have been








preferred it was not attempted because of the anticipated

thermal instability of the expected products. When an

attempt was made to thermolyze (65) using the normal method

of flash vacuum pyrolysis (the substrate is allowed to slowly

sublime through a heated, evacuated tube), no useful results

were obtained. Apparently (65) was not volatile enough to

sublime, at least without heating relatively strongly,

through the pyrolysis tube. With the normal technique of

flash vacuum pyrolysis not readily workable another type

of flow system was chosen. A system used with success in

the past has been the hot tube technique. This technique

was used by W. M. Jones and coworkers when they pyrolyzed

the sodium salt of benzaldehyde tosylhydrazone (113) to

give phenylcarbene (2).

Briefly, the technique consists of dropping the

substrate of interest down an inclined tube which is being

continuously evacuated and is heated to the desired

temperature. In addition the apparatus is continuously

swept with a stream of inert gas to help remove products

from the hot zone. The products are then condensed in a

liquid nitrogen trap. For a detailed description of the

procedure used in this work see the Experimental Section.

When (65) was pyrolyzed by the hot tube method at 1200

only a complex mixture of unidentifiable products was

obtained. No evidence for the presence of any carbene

products was found. No further attempts at this temperature

were made. However, when (65) was pyrolyzed at higher







temperatures the results shown in Table 2 were obtained.

At 2000 a mixture of the isomeric dimers (18) and (19) was

obtained in yields of 13% and up. In addition benzene (22)

was also formed in yields as high as 19%. Although no

quantitative yield was obtained for toluene (23) it was also

present in a significant amount (see Note b, Table 2). When

the temperature was raised to 2500 (18) and (19) were again

isolated in yields exceeding 14%. In addition heptafulvalene

(14) was found in one run in a 5% yield. Again benzene and

toluene were products in yields of 18 and 4% respectively.

At 3000 a trend began to appear. The dimers (18) and (19)

were found in yields from 9 to 14% whereas heptafulvalene (14)

was found in yields from 6 to 15%. The yield of benzene was

typically 3 to 4% but toluene was not observed in any of the

experiments at this temperature. At 3500 the trend continued.

The dimers (18) and (19) were no longer present in the product

mixture. The yield of heptafulvalene (14) had increased to

above 18%. Again benzene was a product but no toluene. No

evidence was found for either heptafulvalene (14) or the dimers

(18) and (19). That (14) was not simply decomposing at 4000

was shown when authentic (14) was pyrolyzed at 4000 and re-

covered in a 32% yield. In a typical experiment at 4000 the

only products identified were benzene in yields of 3% and

better and some aromatic hydrocarbons in an undetermined

amount. A few of these hydrocarbons have been identified by

gc-ms as biphenyl, bibenzyl, and two materials with molecular

formulae of C14H12. That similar aromatic hydrocarbons were

also formed at the lower temperatures was suggested by nmr







resonances in their analyses that were similar to resonances

found in 4000 nmr analyses. In addition, in one experiment

not shown in Table 2 the volatile constituent of the product

mixture was allowed to react with tetracyanoethylene (TCNE),

a known fulveneallene (12) trap. When the products from

this reaction were analyzed by nmr some weak signals

corresponding to those reported in the literature6 for

5,5,6,6-tetracyano-7-vinylidenebicyclo[2.2.1]hept-2-ene (114)

were found. However, the yield was too low to be quantified.

Nevertheless, the presence of (12) as a product at 4000 was

strongly suggested. To summarize quickly, the isomeric






NC CN
//



+ CN
NC CN & CN
CN
(12) (114)



dimers (18) and (19) were obtained from 200 to 3000; hepta-

fulvalene (14) from 250 to 3500; benzene from 200 and up;

toluene at 200 and 2500; and at 4000 benzene, a mixture of

aromatic hydrocarbons, and most likely fulveneallene (12).

In addition similar aromatic hydrocarbons may have been

present at the lower temperatures.












200-400 0 +
Hot Tube

(65)
(18) (19)




+ Q + \ + + Ar-H
--- 2 -

(14) (22) (23) (12)



These results suggest an attractive mechanism. Pyrolysis

of the N-nitroso urea (65) above 2000 may lead to 7-norborna-

dienylidene (1). This carbene may then fragment to give

benzene and a carbon atom in a manner similar to that which

Wolf and Shevlin found when they generated quadricyclanyli-

dene (7). This is supported by the isolation of (22) and

(23). Rearrangement of (1) to (7) is not likely since from

simple thermochemical considerations (7) is calculated to be

40 kcal/mole higher energy than (1). Instead of fragmenting

(1) may also rearrange to the bicyclic triene (6) by a

reaction that is formally a 1,2 carbene shift. This rearrange-

ment of course parallels the reaction which Moss and coworkers

found when they generated 7-norbornenylidene (24) to give
28
(25) as a major product.28 As to whether the mechanism for

rearrangement isa 1,2 shift or involves a nonclassical inter-

mediate is,however, open to question.











4




(1)


O(2

(22)


Apparently the fate of the bicyclic triene (6) is
dependent on the temperature. Below 3000 dimerization


1,2


*0


(115)


(6)


may occur to give the isolated isomers (18) and (19). In
addition if the temperature is between 250 and 3500
rearrangement of (6) to cycloheptatrienylidene (3) may
occur. Dimerization of (3) then accounts for the observed

heptafulvalene (14). The trapping experiment with TCNE at
4000 suggests that at this temperature (6) may be rearrang-
ing to fulveneallene (12). Rearrangement of (6) to (3) and


:C
0L




-1 w


(12) gains support from the recent work of Wentrup and

coworkers whose 1C labeling experiments with phenyl-


S


(15)


(12)


(14)


(18) (19)

carbene (2) and fulveneallene (12) are most easily

explained by a mechanism involving interconversion of

(6), (3), and (12).21 To summarize, an attractive scheme

which would explain all of these results is as follows:

the carbene (1) either fragments to give (22) and a

carbon atom or rearranges to (6) which depending on the

temperature dimerizes, rearranges to (3), or rearranges to (12).









** **




(2) (3)











(12) (15) (6)



However, there are a number of alternative mechanisms

which come to mind that could at least explain portions

of the results. These include the following: One,

fragmentation of (1) may give benzene and a carbon atom

which could combine to give (27). Rearrangement of (27) to

(3) and subsequent dimerization of (3) would explain the









(1) (22) (27)








(3) (14)








heptafulvalene (14). In fact, there are examples in the

literature where carbon atoms generated from an arc or in

an accelerator have reacted with benzene to give products

which suggested a mechanism involving addition of the

carbon atom to (22) to give (27) and subsequent rearrange-
22
ment to (3). However, the fact that benzene (22) and (14)

do not always appear simultaneously as products, as one

might expect if this mechanism were important in the chemistry

of the N-nitroso urea (65), casts doubt upon the validity

of this mechanism in this case. Two, (1) may rearrange to

norcaradienylidene (27) by direct rearrangement. This
22
would be a retro-Skattebzl rearrangement.2 However, as

mentioned before there is no literature precedent for a

reaction such as this. Three, it is possible that (65)

could rearrange to (116) which, in turn, could give hepta-

fulvalene (14). That (65) does not rearrange to (116) is

shown by Jones and Manganiello who find that (116) does

not give any (14) when pyrolyzed under the same conditions

as used in this work.67 Four, phenylcarbene (2) could account




NO

nf 2 NO
0 1
\ NCNH
\n 2
---


(116)


(65)








for the formation of (14) from 250 and 350. However, the

fact that (14) is no longer a product at 4000 while authentic

(2) generated at 4000 gives a 33% yield of (14) casts some

doubt on the importance of this possibility although it

does not entirely rule it out.

Five, electrocyclic ring opening of (18) to (14) is in

principle a plausible alternative mechanism for explaining

the increasing yield of (14) with temperature. This

rearrangement is particularly interesting because according

to the Woodward Hoffman rules the reaction is allowed for the

syn isomer (18) and not allowed for the anti isomer (19).68

In theory then the observation that (18) is being selectively

depleted as the temperature is raised would be evidence for

this rearrangement. Unfortunately, it was not possible to

obtain conclusive evidence for or against depletion of (18)

in the experiments undertaken in this work. Nevertheless,

it is possible to exclude this mechanism. When an authentic

mixture of (18) and (19) was pyrolyzed at the temperature

which should have yielded the highest yield ofi (14), 3500,

no (14) could be found by nmr analysis.


(18)


(14)



















































(ra



NJ

ko >1






+ 4-
En

U)







to
(N
















EL,




/:



/:

/
/


0o
+
(u 0

oN

+
+ .Q
-I-)


0


'I







Six, in none of the pyrolysis experiments is there any

evidence for binorbornadienylidene (26).69 On the other hand

it is conceivable that dimerization of (1) to (26) may occur

followed by rearrangement of (26) to (14) or rearrangement

to (18) and (19). Of course rearrangement of (26) to (14)

would have to be favored at higher temperatures in order

for this mechanism to be consistent with the experimental

results obtained. Although norbornadienyl compounds are









(18) (19)





(26)





(14)
70
known to rearrange to cycloheptatrienyl compounds on

heating, not much is contained in the literature on the

thermal stability of methylene norbornadienes. Although in

one instance Martin and Forster report that dimethyl

methylenenorbornadiene is stable at 1000 for 24 hours.71

Seven, an interesting possibility although admittedly

rather wild, comes to light when the recent results of R.








Hoffman217273 and those of Chan and Massuda74 are viewed

along side the results in this work. R. Hoffman found that

when 7-acetoxynorbornadiene (117) was subjected to flash

vacuum pyrolysis at 4500 and 7000 acetic acid and hepta-

fulvalene (14) or fulveneallene (12) respectively were
21
obtained as products.21 An attractive explanation for

this would have been generation of (1) which then could have

led to (14) and (12). However, when (117d) was pyrolyzed at

4500 (14d) was isolated still containing most of the

7000

O0
OCCH
(12)



(117)

450

(14)



0
OCCH /I
3O -CH CO H



(117) (1)



D D
DJ OCCH
OI 3
0


(117d)


(14d)








72
deuterium.72 Therefore, (1) could not have been involved.

Hoffman and coworkers have proposed a free radical mechanism

to explain the products obtained in the case of 7-phenyl-

7-acetoxynorbornadiene (118).73 Chan and Massuda found



O
II
Ph OCCH 3
Ph OCCH
II 3




(118) (119)




strong evidence when (120) was treated with fluoride ion

for the formation of the bridgehead olefin (121) and what

is apparently a retro 1,2 carbene shift to give the carbene
74
(122).74 They were able to trap both of these intermediates

with appropriate traps. Therefore, it is conceivable that

decomposition of (117) could proceed by elimination across

SiMe3
Br (Br






(120) (121)


(122)
(122)








the 1,7 bond of (117) to give the very strained anti-

Bredt compound bicyclo[2.2.l]hepta-l,2,5-triene (123). This

could then rearrange to 4-bicyclo[3.2.0]hepta-2,6-dienylidene

(124) via a retro 1,2 carbene shift followed by a 1,2 shift

to give (6). Of course a deuterium on the 7 position would

still be contained in the products. In addition it is also

possible that some intermediate in the case of (65) could

undergo the same process. As to whether this mechanism is of

importance remains to be seen. However, it should be noted




H O-C-CH H
1 3 f*3



H

(117) (123) (124) (6)




that an interesting dilemma surfaces if (117) is giving rise

to (123) assuming that temperatures are directly comparable

between the hot tube experiments and Hoffman's experiments.

That is, how could (6) give rise to heptafulvalene (14) at

4500 in the case of (117) and give no (14) at 4000 in the

case of the N-nitroso urea (65)?

In conclusion, (1) is strongly suggested as an inter-

mediate in the hot tube pyrolysis experiments. The mechanism

that is most consistent with the current literature and

experimental evidence is one involving rearrangement of (1) to

























01









CC
I *0
*







N
z
O U=0




VV




(6) followed by, depending on the temperature, dimerization

to give (18) and (19) or rearrangement to give (3) or (12).

Nevertheless, mechanisms involving (2), (26) or (123)

cannot be entirely ruled out.














EXPERIMENTAL


General. All melting points were obtained on a

Thomas Hoover melting point apparatus and are uncorrected.

All 60 MHz nuclear magnetic resonance data were obtained

on a Varian A-60A spectrometer and are reported in units

of T from tetramethylsilane which was used as an internal

standard. The 100 MHz nuclear magnetic resonance data

were obtained on a Varian XL 100 spectrometer and are

reported in the same manner. Infrared data were obtained

on either a Beckman IR 10 or Perkin-Elmer 137 spectrophoto-

meter. Ultraviolet spectra were obtained on a Cary 15

spectrophotometer. Combustion analyses were performed by

Atlantic Microlab Inc., Atlanta, Georgia. Mass spectra

were obtained on an AEI MS 30 spectrometer.

Analytical thin layer chromatographies were performed

on either Eastman silica gel or alumina chromatograms

impregnated with florescent indicator. Preparative thin

layer chromatographies were carried out on 8 in. x 8 in.

silica gel plates coated to a thickness of 1.0 mm with Merk

PF 254 preparative thin layer chromatography silica gel.

Unless otherwise noted solvents were reagent grade and

were used as obtained.

7-t-Butoxynorbornadiene (67). This material was

either used as obtained from Frinton Laboratories Inc.,




DL


Vineland, New Jersey, or was prepared according to the

method of P. R. Story and S. R. Fahrenholtz.55

7-Chloronorbornadiene (50). 7-Chloronorbornadiene (50)

was prepared by the method of P. R. Story and M. Saunders56

with the following modifications. The reaction solution

was cooled with an ice water bath while the initial hydrogen

chloride was bubbled in. In addition, instead of distilling

the product on a spinning band column a 30 cm Vigreux

column was used to give a product which was essentially pure

except for acetic anhydride. The amount of acetic anhydride

was reduced to an acceptable level by chromatography with

pentane on a silica gel column.

The spectral data were as follows: NMR (CDC13) T3.2-3.5

(m, 4H, olefinic), 5.7-5.9 (m, 1H, C-7), 6.3-6.5 (m, 2H,

bridgehead); ir (neat) 3130, 3110, 3080, 3000, 1650, 1550,

1370, 1310, 1200, 950, 880, 810, 730, 650 cm-1 on Beckman IR

10.

Silver Isocyanate (71). This procedure is essentially

that of Birkenbach and Linhard58 with some refinements.

The entire preparation was carried out in the dark and the

product was protected from light at all times. To a

solution of 34.0 g (0.20 mol) of silver nitrate in 1000 ml

of deionized water in an amber bottle was added a solution

of 17.9 g (0.22 mol) of potassium cyanate in 1000 ml of

deionized water. The potassium cyanate solution had been

previously brought to a pH of 6.2 with 1.0 M nitric acid.

After vigorous mixing the white precipitate was collected








on a medium frit and was washed with water, alcohol, and

ether. Drying the precipitate in a vacuum desiccator over

phosphorus pentoxide for 3 days gave 26.0 g (87%) of (71)

as a snow white solid.

The spectral data were as follows: IR (nujol mull
-i
and fluorolube mull) 2180 cm on PE 137. No absorptions

attributable to silver oxide or silver carbonate (1300-
-i
1500 cm ) were present.

7-Norbornadienylisocyanate (68). Over a period of

several minutes 15.3 g (0.12 mol) of a solution of 7-chloro-

norbornadiene (68) in approximately 60 ml of dry liquid

sulfur dioxide was added to a mixture of 19.5 g (0.13 mol) of

silver isocyanate and 150 ml of dry liquid sulfur dioxide.

After the addition was complete the reaction mixture was

allowed to reflux while stirring with a magnetic spin bar

for 1.5 hours. During this time the system was kept under

an inert atmosphere of argon and in complete darkness. After

this time the reaction mixture was allowed to warm to room

temperature by letting the sulfur dioxide evaporate. During

the evaporation of the sulfur dioxide 200 ml of anhydrous

ether was added. After this the reaction mixture was filtered

through a medium frit containing a layer of filter aid.

Removal of the solvent in vacuo (rotary evaporator) from the

resulting ether solution gave 15.1 g (94%) of the isocyanate

(68) as an orange oil. No attempt was made to purify this

compound and it was used immediately as obtained.

The spectral data were as follows: NMR (CDC13) T3.1-3.3

(m, 4H, olefinic), 6.2-6.5 (m, 3H, bridgehead and C-7); ir

(neat) 2200 cm-1 on PE 137.








Methyl N-7-Norbornadienylcarbamate (69). A solution

of 7-norbornadienylisocyanate (68) in 70 ml of pentane just

prepared from 3.8 g (0.03 mol) of the chloride (50) was

added to 50 ml of anhydrous methanol containing a few drops

of acetic acid. This solution was then stirred with a

magnetic spin bar under an inert atmosphere of nitrogen.

After maintaining the reaction under the above conditions

for 20 hours the resulting brown mixture was concentrated

to a volume of approximately 5 ml in vacuo (rotary evapora-

tor). The concentrate was dissolved in a minimum amount of

ether and chromatographed on a short silica gel column using

20% v/v ether-pentane as an elutant. Collection of the first

yellow band gave 1.3 g of a yellow solid on removal of

solvent via rotary evaporator. Trituration of this material

with pentane and concentration of the resulting clear,

colorless pentane solution via rotary evaporator afforded

930 mg [19% based on the chloride (50)] of (69) as a white

solid, mp 63.5-65.5 An analytical sample was prepared for

CHN analysis by sublimation at room temperature and a pressure

of 5 microns.

The spectral data were as follows: NMR (CDCl3) T3.2-

3.4 (m, 4H, olefinic), 4.2-5.0 (broad d, 1H, N-H), 5.7-6.1

(broad d, 1H, C-7), 6.4 (s, 3H, methyl), 6.4-6.6 (m, 2H,

bridgehead); ir (nujol mull) 3440, 3405, 2920, 1740, 1375, 1350

1310, 1210, 1070, 1050, 780, 740 cm-1 on Beckman IR 10;

ms m/e 165 (M+), 164, 150 (base), 134, 133, 132, 106, 91,

79, 78, 77.








Anal. Calcd for C9 11NO2: C, 65.44; H, 6.71;

N, 8.48. Found: C, 65.19; H, 6.74; N, 8.42.

p-Chlorobenzoyl Nitrite (73). This material was

prepared using the method of Baigrie, Cadogan, Mitchell,

Robertson, and Sharp.63 The material was carefully weighed

and dissolved in anhydrous ether in a volumetric flask

immediately after preparation and stored at Dry Ice

temperature. In this way the actual amount of solution

needed for each reaction was easily calculated. The solution

was allowed to warm to room temperature just before use.

Best results were obtained if the material was made within

a few days of its anticipated use in about the amount needed

instead of trying to make large quantities of the material

and store it over a long period of time since the material

slowly decomposes.

Methyl N-Nitroso-N-7-norbornadienylcarbamate (64). A

volume, in this case 42.0 ml, of p-chlorobenzoyl nitrite

solution containing 1.0 equivalents of p-chlorobenzoyl

nitrite (73) was added to a solution of 1.07 g (0.0065 mol)

of methyl N-7-norbornadienylcarbamate (69) in 100 ml of

anhydrous ether containing 5.2 g of freshly fused anhydrous

sodium acetate. The p-chlorobenzoyl nitrite solution was

at room temperature. The mixture was stirred with a

magnetic spin bar between -40 and -550 under an argon atmos-

phere for 4.5 hours. The yellow reaction mixture was then

poured into 200 ml of ice cold aqueous 5% sodium bicarbonate

solution and 300 ml of ice cold ether. After mixing and







separation of layers the ether phase was washed twice with

200 ml portions of ice cold aqueous 5% sodium bicarbonate

solution and then twice with 200 ml portions of ice cold

water. Drying of the ether phase over anhydrous magnesium

sulfate, filtration, and removal of the solvent in vacuo

(rotary evaporator) gave an orange yellow oil. The orange

yellow oil was then chromatographed on a specially prepared

1.5 x 35 cm silica gel column using wet 3% v/v ether-pentane.

The column was prepared by making it up using 3% v/v ether-

pentane which had been shaken in a separatory funnel with

water. Ether-pentane prepared in this same manner was used

for elution. The yellow band was collected. The yellow

solution was dried over anhydrous magnesium sulfate, filtered,

and the solvent evaporated in vacuo (rotary evaporator) to

give 930 mg (79%) of (64) as a yellow orange oil.

The spectral data were as follows: NMR (CDC13) T3.1-3.6

(m, 4H, olefinic), 6.0 (s, 3H, CH3), 5.9-6.1 (m, 3H, bridge-

head and C-7).

Bases Used in the Attempts to Generate 7-Norbornadienyli-

dene (1) from Methyl N-Nitroso-N-7-norbornadienylcarbamate

(64) and N-Nitroso-N-7-norbornadienylurea (65).

(A) Sodium Methoxide. This was obtained as a 25%

solution in methanol from Aldrich (54.0 g sodium methoxide/

216 g solution).

(B) Lithium Ethoxide. To 100 ml of absolute ethanol

under an atmosphere of argon was added 3.5 g (0.5 mol) of

freshly cut lithium ribbon. After the lithium dissolved the

mixture was refluxed for 13.5 hours. Upon cooling the mixture








solidified into a single solid mass. The excess ethanol

was taken off in vacuo at room temperature overnight.

This afforded 28.4 g of a white solid. To titrate 0.1527 g

of this solid to a phenolphthalein endpoint 2.52 milli-

equivalents of 0.1000 N hydrochloric acid were required.

Calculations showed that 1.56 milliequivalents would be

needed if the solid were lithium ethoxide-ethanol and 2.96

if it were lithium ethoxide.

(C) Lithium Methoxide. To 150 ml anhydrous methanol

was added slowly 5.1 g (0.75 mol) of lithium ribbon. The

reaction system was a 200 ml three neck round bottom flask

fitted with an efficient reflux condenser and drying tube

plus an argon inlet. After most of the lithium had

reacted an additional 15 ml of methanol were added and the

system was brought to reflux for 23 hours. After cooling

to room temperature the excess methanol was distilled off

in vacuo to give 26.6 g (93%) of the product as a white

solid.

The spectral data were as follows: IR (nujol mull)

1150, 1050 cm-1; (fluorolube mull) 2900, 2800, 2750, 2575,

1450 cm- on PE 137.

(D) Lithium Ethoxide-Ethanol. To 50 ml of absolute

ethanol under nitrogen was added lithium metal until the

lithium metal reacted slowly. More ethanol, approximately

25 ml, was then added and the mixture brought to reflux

until all of the lithium had dissolved. This gave a white

gelatinous precipitate which upon cooling with an ice water








bath crystallized in a form which could be filtered.

Filtration, washing with a small portion of ethanol, and

air drying gave a white solid.

The spectral data were as follows: IR (potassium

bromide pellet) 3580, 3000, 1580, 1520, 1450, 1000, 860,
-i
680, 640 cm on Beckman IR 10.

Decomposition of Methyl N-Nitroso-N-7-norbornadienyl-

carbamate (64) by Sodium Methoxide. In a typical reaction

0.295 g of a 25% sodium methoxide in methanol solution

which had been diluted with 0.3 ml of anhydrous methanol was

added dropwise over a six minute period to a solution of

102 mg (0.53 mmol) of methyl N-nitroso-N-7-norbornadienyl-

carbamate (64) in 100 ml of anhydrous ether. The reaction

solution was held at reflux while stirring with a high speed

mechanical stirrer in a Mortan flask under an argon

atmosphere. After the addition was complete the reaction

solution was allowed to stir under the above conditions for

24 minutes. The resulting brown orange solution was poured

into a mixture of 300 ml of ice cold ether and 200 ml of ice

cold water. After mixing and separation of phases the ether

phase was washed twice with ice cold water. The ether phase

was dryed over anhydrous magnesium sulfate, filtered, and the

solvent evaporated in vacuo (rotary evaporator) to give

42.6 mg of a red brown oil. Analysis of the crude product

via nmr, tic, and glpc and attempts to isolate products by

column chromatography gave no evidence for the presence of

carbene products. Carrying the reaction out in the presence








of dimethylfumarate gave the Michael adduct of sodium

methoxide and dimethylfumarate as the only isolatable

product.

Decomposition of Methyl N-Nitroso-N-7-norbornadienyl-

carbamate (64) by Lithium Ethoxide. By way of a solid

addition tube was added 1.5 equivalents, as determined by

titration, of lithium ethoxide in one lot to a solution of

101 mg (0.52) mmol) of methyl-N-nitroso-N-7-norbornadienyl-

carbamate (64) in 100 ml of anhydrous ether. The mixture

was stirred with a high speed mechanical stirrer in a Mortan

flask under an argon atmosphere. The reaction was held at

00 for 30 minutes after the addition and when no visual

change took place it was warmed to room temperature for 4.25

hours. At the end of this time the resulting orangish brown

mixture was poured into a mixture of ether and water. After

mixing and separation of phases the ether phase was washed

with two additional portions of water. Most of the color

washed into the aqueous phases. The organic phase was dryed

over anhydrous magnesium sulfate, filtered, and the solvent

removed in vacuo (rotary evaporator) to give 64.1 mg of a

orangish brown oil. Column chromatography of this oil on

silica gel with 3% v/v ether-hexane afforded 22.2 mg (26%) of

methyl 7-norbornadienylcarbonate (109) as a yellow oil.

The spectral data were as follows: NMR (CDCl3) T3.3-

3.6 (m, 4H, olefinic), 5.5 (m, 1H, C-7), 6.3 (s, 3H, CH3),
-I
6.3-6.5 (m, 2H, bridgehead); ir (CHC13) 1730, 1250, 1050 cm

on Beckman IR 10; ms m/e 166 (M+), 122, 121, 107, 91 (base),

79, 78, 77, 59, 44.







7-Norbornadienylurea (70). Into 100 ml of a saturated

benzene ammonia solution at room temperature was added 15.1 g

(0.114 mol) of 7-norbornadienylisocyanate (68) dissolved in

a small amount of benzene. An orangish precipitate formed

almost immediately upon addition of the isocyanate (68) and

the flask became rather warm. The reaction mixture was

stirred with a magnetic spin bar while bubbling in ammonia for

an additional 1.5 hours. The resulting precipitate was

collected on a medium frit to give 10.3 g of a dark beige

solid. A dark hard lump of material which seemed to form

quickly just before the orangish precipitate began to form was

kept separate from the product. Recrystallization of the

beige solid from hot tetrahydrofuran afforded 4.4 g (26%

based on (68)) of (70) as a beige solid. An analytical

sample was prepared by recrystallizing from tetrahydrofuran

three times to give a white solid, mp 175-184o d.

The spectral data were as follows: NMR (DMSO-d6)

T3.1-3.4 (m, 4H, olefinic), 3.5-3.9 (broad d, 1H, N-H),

4.3-4.8 (broad s, 2H, NH2), 5.8-6.2 (broad d, 1H, C-7),

6.4-6.7 (m, 2H, bridgehead); ir (potassium bromide pellet)

3450, 3310, 3210, 3100, 3000, 1660, 1600, 1570, 1550, 1540,

1395, 1330, 1315, 1240, 1210, 1170, 740, 650, 630 cm-1 on

Beckman IR 10; ms m/e 150 (M+, weak), 149, 133, 132, 106

(base), 91, 79, 65.

Anal. Calcd for C8H1ON20: C, 63.68; H, 6.71; N, 18.65.

Found: C, 63.91; H, 6.73; N, 18.60.








N-Nitroso-N-7-norbornadienylurea (65). To a mixture

of 2.1 g (0.014 mol) of 7-norbornadienylurea (70) and

11.5 g of freshly fused anhydrous sodium acetate in 100 ml

of anhydrous ether stirred magnetically under argon at -450,

was added 1.0 equivalents of p-chlorobenzoyl nitrite

solution at room temperature. The reaction was allowed to

run under the above conditions for 4.5 hours. After this

time the entire contents were poured into a mixture of 300 ml

of ice cold ether and 200 ml of ice cold aqueous 10% sodium

bicarbonate solution. After mixing and separation of layers

the ether phase was washed twice more with 200 ml portions

of ice cold aqueous 10% sodium bicarbonate solution and

finally once with 200 ml of ice cold water. The ether phase

was dried over anhydrous magnesium sulfate, filtered, and

the solvent removed in vacuo (rotary evaporator) to afford

1.45 g of a yellow brown solid. Chromatography of the crude

product on a silica gel column with 50% v/v ether-pentane

gave on collection of the bright yellow band 944 mg (38%)

of (65) as bright canary yellow needles. A small amount of

material was rechromatographed to provide a sample with mp

99-101 d.

The spectral data were as follows: NMR (CDCl3) T3.1-3.5

(m, 2H, olefinic), 3.8 (very broad s, 2H, NH2), 5.8-6.0

(m, 2H, bridgehead), 6.0-6.2 (m, 1H, C-7); ir (potassium

bromide pellet) 3440, 3310, 3250, 3170, 3080, 3010, 2970,

1710, 1610, 1590, 1500, 1400, 1340, 1310, 1240, 1200, 980,
-1
700 cm on Beckman IR 10; ms m/e 179 (M+, very weak), 149,

132, 119, 91, 79, 77, 43 (base).








Typical Decomposition of N-Nitroso-N-7-norbornadienyl-

urea (65) under Basic Conditions. In a typical reaction

18.9 mg of lithium methoxide was added as a solid in one batch

to a stirred solution of 48.8 mg (0.27 mmol) of N-nitroso-

N-7-norbornadienylurea (65) in 50 ml of anhydrous ether at

0 The mixture was stirred at 0 for 5.5 hours. During

this period the bright yellow color of the solution became

dull although no bubbles or transient colors were observed.

The reaction was worked up by diluting with 150 ml of ether

and washing with three 150 ml portions of water. Drying

over anhydrous magnesium sulfate and removal of solvent in

vacuo (rotary evaporator) gave 28.6 mg of a brown oil. No

carbene products could be detected by normal analysis methods.

Other conditions attempted are recorded in Table 1. Under

none of these conditions have any carbene products been

detected.


Table 1. Decomposition of N-Nitroso-N-7-norbornadienylurea
(65) Under Basic Conditions.


Base Solvent T (C ) Trap

Lithium Ethoxide Ether RT None
Lithium Methoxide Ether 0 Dimethylfumarate
Lithium Methoxide Ether 0 None
Lithium Methoxide Pentane 0 None
Lithium Methoxide Isobutylene -7 Isobutylene
Sodium Methoxide Isobutylene -7 Isobutylene
Lithium Ethoxide- Isobutylene -7 Isobutylene
Ethanol


Decomposition of N-Nitroso-N-7-norbornadienylurea (65)

in Heptane at 900. In a typical reaction 97.7 mg (0.55 mmol)

of N-nitroso-N-7-norbornadienylurea (65) was added as a solid







in one batch to 10 ml of heptane which was stirred at 90

under an argon atmosphere. Gas evolution occurred

immediately and was quite rapid. The gas evolution was

complete within a few minutes and no transient colors were

observed. Some brownish insoluble material formed and the

appearance of the solution changed from bright yellow to

dull yellow. The reaction solution was held at 900 for

4 minutes after the addition and was quenched by cooling

with an ice water bath. Evaporation of the heptane to near

dryness in vacuo (rotary evaporator) gave an undetermined

amount of an orange oil containing some brown solid with the

nmr spectrum shown in Figure 1. In another run a similar

material was obtained with the nmr spectrum shown in

Figure 2.

In another run methanol was added to the crude product

mixture and the progress followed by nmr to give the spectrum

shown in Figure 3.

Decomposition of N-Nitroso-N-7-norbornadienylurea (65)

in Heptane in the Presence of Dimethylfumarate at 90. As

a solid in one batch 41.1 mg (0.23 mmol) of N-nitroso-N-7-

norbornadienylurea (65) was added to a solution of 400 mg

(12 fold excess) of dimethylfumarate in 10 ml of heptane.

The solution was stirred with a magnetic spin bar at 900

under an argon atmosphere. Gas evolution was immediate and

complete within 2 minutes. No transient colors were observed.

After 2.5 minutes at 900 the reaction was cooled with an ice

water bath. The excess dimethylfumarate which crystallized








upon cooling was filtered off and the solvent removed in

vacuo (rotary evaporator) from the filtrate to give an

undetermined amount of an orange oil and white solid. The

nmr spectrum of this material is shown in Figure 4.

The crude product mixture was then unintentionally

exposed to ethanol which resulted in the appearance of new

signals in its nmr spectrum as shown in Figure 5. Most of

the remaining dimethylfumarate was removed by sublimation.

When an nmr of this material was obtained it resulted in

the following spectrum shown in Figure 6. Chromatrography

of this material on a silica gel column with 75% v/v ether-

hexane gave 13.2 mg of an orange oil with the nmr spectrum

shown in Figure 7.

Decomposition of N-Nitroso-N-7-norbornadienylurea (65)

in Heptane in the Presence of Lithium Carbonate at 90 To

a stirred suspension of 52 mg (0.7 mmol, 5 fold excess) of

lithium carbonate in 10 ml of heptane at 90 under argon was

added as a solid in one batch 25.2 mg (0.14 mmol) N-nitroso-

N-7-norbornadienylurea (65). Gas evolution was immediate

and quite rapid but no transient colors were observed. The

reaction mixture was held at 900 for an additional four

minutes and then quenched by cooling with an ice water bath.

The lithium carbonate was filtered off on a medium frit and

the solvent removed in vacuo (rotary evaporator) to give an

undetermined amount of an orangish oil. Spectral analysis

by nmr showed the products to be the same as in experiments

where no lithium carbonate had been used.








Decomposition of N-Nitroso-N-7-norbornadienylurea (65)

in Triglyme at 2000. To 15 ml of triglyme which was stirred

with a magnetic spin bar under an argon atmosphere at 2000

was added dropwise over a 15 minute period a solution of

50.8 mg (0.28 mmol) of N-nitroso-N-7-norbornadienylurea (65)

in 5 ml of triglyme. The color of the reaction solution

changed from colorless to yellow to a bright brown as the

reaction progressed. The reaction was cooled with an ice

water bath after heating at 2000 for five minutes. After

diluting with 400 ml of pentane the solution was washed

with ten 50 ml portions of water to remove the triglyme.

Drying of the pentane solution over anhydrous magnesium

sulfate and removal of the solvent in vacuo (rotary

evaporator) afforded an undetermined amount of an orange

brown oil. Spectral analysis by nmr and gc-ms showed only

a trivial product tentatively identified as dimethoxybenzene.

This material was trivial since a material with the same

spectral properties was independently found to be a

contaminant in the triglyme.

Experimental Method for Thermolysis of N-Nitroso-N-7-

norbornadienylurea (65) in a Hot Tube. The experimental

system as shown in Figure 8 consisted of an open tube

inclined at approximately 450. At the head of the tube,

which was at an angle of 450 to the inclined portion, was a

solid addition tube containing the substrate and downstream

from it a nitrogen inlet. The inclined portion of the tube

was heated by a heating tape and the temperature monitored










Substrate

I


Nitrogen








Hot Zone

















Trap
Pump -


Figure 8







by an external thermocouple. Products were trapped by a

liquid nitrogen trap at the end of this tube. A plug of

glass wool was placed in the head of the trap and cooled

with liquid nitrogen to help break up any aerosoling.

Pressure and nitrogen flow were regulated by evacuating the

system to approximately 20 microns and then adjusting the

nitrogen flow so as to achieve the final desired pressure.

The substrate was added to the hot tube slowly over a 10 to

15 minute period. Nonvolatile and volatile products were

worked up separately by allowing the volatiles to distill into

the bottom of the trap and then dissolving them in chloro-

form-d for direct spectral analysis by nmr. Nonvolatiles

which remained at the top of the trap and in the cool portions

of the tube before the trap were collected by washing them

out with methylene chloride. Removal of the methylene

chloride in vacuo (rotary evaporator) afforded the non-

volatile products which were analyzed by their nmr and uv

spectra.

Yields of products were obtained by addition of a

known amount of cyclooctane to the nmr sample and integration

of peaks. In several runs yields of heptafulvalene (14) were

also obtained from its ultraviolet spectra. The yield of

benzene (22) in one instance was determined by gas chromato-

graphy. Identification of (14) was done by comparison of its

nmr spectrum with that of an authentic sample and by tlc by

comparing its Rf value with that of an authentic sample.

Toluene (23) and (22) were identified by comparison of their







nmr spectra with those of authentic samples. In addition

(22) was also identified in one instance by gc-ms. The

isomeric dimers (18) and (19) were identified by comparison

of their very characteristic nmr spectra with those published

in the literature26 and from their mass spectra.

Typical Thermolysis of N-Nitroso-N-7-norbornadienylurea

(65) in a Hot Tube at 3000. In a typical experiment 47.1 mg

(0.263 mmol) of N-nitroso-N-7-norbornadienylurea (65) was

exposed to the hot tube with the temperature set at 3000 and

a pressure of 3 torr. The products were worked up and

analyzed in the manner described in the experimental method.

A brown material was found in the cool part of the tube before

the trap and in the top of the trap. A yellow liquid which

partially solidified on warming to room temperature to give

a white solid and a yellow liquid was found in the bottom of

the trap. The products had a rather strong odor. The

analysis showed the presence of the two isomeric dimers (18)

and (19) in an 11% yield and heptafulvalene (14) in a 9%

yield as nonvolatiles. The volatile portion consisted of

benzene (22) in a 3% yield. In addition some materials)

with a complex nmr resonance from T2.4 to 3.0 was/were a

part of the nonvolatiles.

Typical Thermolysis of N-Nitroso-N-7-norbornadienylurea

(65) in a Hot Tube at 4000. In a typical experiment 43.4 mg

(0.242 mmol) of N-nitroso-N-7-norbornadienylurea (65) was

exposed to the hot tube with the temperature set at 4000 and

a pressure of 3 torr. The products were worked up and







analyzed in the manner described in the experimental

method. A brown oil was found in the cool part of the tube

before the trap and in the top of the trap. A yellow liquid

which partially solidified on warming to room temperature

to give a white solid and a yellow liquid was found in the

bottom of the trap. The products had a rather strong odor.

The volatile portion consisted of benzene (22) in a 13% yield.

The nonvolatile portion consisted of an undetermined amount

of materials which have been identified as aromatic hydro-

carbons.

The spectral data were as follows: NMR (CDCl3)

T2.4-3.0 complex pattern; gc-ms (5' x 1/4" 3% SP 2100 column,

2000) m/e 154 (base), 153, 152, 77, 76; 182, 91 (base); 180,

179 (base), 178, 165, 89, 76; 180 (base), 179, 178, 165, 89,

88. The mass spectral data given here are only a represent-

ative sample of the gc-ms data obtained. Tentative assign-

ments for the portion of the data given here are biphenyl,

bibenzyl, and two materials with molecular formulae of

C14H12 respectively. The results from the other thermolysis

experiments are shown in Table 2.








Table 2. Thermolysis of N-Nitroso-N-7-norbornadienylurea
(65) in a Hot Tube.

T(oC) Run (18) + (19) (14) (22) (23)

200 A 61 0 19a b
B 13 0 0.5 0
250 A 34 0 18 4
B 14 5 2 0
300 A 14 15-20 c c
B 13 7 4 0
C 9 3d 4 0
D 11 9e 3 0
E 12 6f trace 0
350 A 0 38 11 0
B 0 18 1 0
400 A 0 0 13 0
B 0 0 3 0
C 0 0 3 0

aThe yield of (22) in this instance was determined by
hc and its identity was determined from its mass spectrum.
The presence of (23) as a product at 2000 was established
from nmr spectra in earlier runs where no yields were
obtained. In this run the volatiles were not analyzed by
nmr and the amount of (23) was too low to be detected by gc.
cThe volatile products were not worked up separately in this
run. dThe yield was 3% by UV analysis. eThe yield was 6%
by UV analysis. fThe yield was 4% by UV analysis.

Thermolysis of N-Nitroso-N-7-norbornadienylurea (65)

in a Hot Tube at 4000 and Trapping with Tetracyanoethylene

(TCNE). In this experiment 70.0 mg (0.391 mmol) of

N-nitroso-N-7-norbornadienylurea (65) were pyrolyzed in the

hot tube at 4000 and 3 torr. The volatile products were

allowed to distill at -35 and 20 microns into a flask

containing a dilute solution of tetracyanoethylene (TCNE) in

tetrahydrofuran. The contents of the flask were then brought

to atmospheric pressure and allowed to react at -300 for 30

minutes. When the solvent was removed in vacuo (rotary

evaporator) a small amount of a brown solid was obtained.








Analysis by nmr showed weak signals corresponding to those

published for 5,5,6,6-tetracyano-7-vinylidenebicyclo[2.2.1]-

hept-2-ene (114).6

The spectral data were as follows: NMR (acetone-d6)

weak signals corresponding to the literature values for

(114). The literature spectral data were as follows:

NMR (acetone-d6) T3.1 (2H, t), 4.6 (2H, s), 5.15 (2H, t).

Sodium Salt of Benzaldehyde Tosylhydrazone (113). This

material was prepared by the method of R. C. Joines,

A. B. Turner, and W. M. Jones.

Thermolysis of the Sodium Salt of Benzaldehyde Tosyl-

hydrazone (113) in a Hot Tube at 4000. Exposure of 54 mg

(0.18 mmol) of the sodium salt of benzaldehyde tosylhydrazone

(113) to the hot tube at a temperature of 4000 and a pressure

of 3 torr resulted in the formation of a brown solid on the

cool parts of the pyrolysis tube and in the upper parts of

the trap. Workup, as described in the experimental method,

gave heptafulvalene (14) in a 33% yield.

Heptafulvalene (14). Authentic heptafulvalene (14) was

prepared in the manner used by W. M. Jones and C. L. Ennis.50

Thermolysis of Heptafulvalene (14) in a Hot Tube at

4000. Addition of 39.7 mg (0.22 mmol) of heptafulvalene (14)

to the hot tube at a temperature of 4000 and a pressure of 3

torr resulted in a dark solid being deposited on the cooler

parts of the pyrolysis tube and the upper parts of the trap.

Workup, as described in the experimental method, gave hepta-

fulvalene (14) in a 32% recovery.







Isolation and Identification of syn- and anti-Pentacyclo-

[10.2.0.0'412.05,11 0811]tetradeca-2,6,9,13-tetraene (18) and

(19). Samples of (18) and (19) obtained from a series of runs

were combined. Chromatography of these materials on a short

silica gel column with pentane gave a material that was pre-

dominately (18) and (19) except for some unidentified aromatic

materials and silicone grease. A substantial amount of the

silicone grease was removed by trituration with acetone.

The spectral data were as follows: NMR (100 MHz, CDC13)

see Figure 9; ms m/e 180 (M+), 179, 178 (base), 165. The
26 75
literature spectral data were as follows: NMR7 dimer A,

T3.7-3.9 (4H, AB, J = 1.3 Hz, cyclobutenes), 4.2 (4H, s, cyclo-

pentenes), 6.45 (2H, m, methines), 7.35 (2H, m, methines);

dimer B, T3.8 (4H, s, cyclobutenes), 4.2-4.4 (4H, AB, J=2.9Hz,

cyclopentenes), 6.4 (2H, m, methines), 6.7 (2H, m, methines);

ms m/e 180.

Thermolysis of syn- and anti-Pentacyclo[10.2.0.04'12

05'11.08'11]tetradeca-2,6,9,13-tetraene (18) and (19) in a

Hot Tube at 3500. The mixture of dimers (18) and (19)

isolated in the proceeding section was coated in its

entirety on glass beads with a mesh size of 60-80 and

pyrolyzed in the hot tube at 350 In a previous run the

N-nitroso urea (65) coated on the same type beads under the

conditions used here gave a 13% yield of heptafulvalene (14).

Workup as described in the experimental method followed by

nmr analysis did not reveal any (14) as a product.





83























C,
















s-I

aT-















APPENDIX


(1)







(2)






(3)


(4)


*


/H



^^--H


(10)


(11)


(12)








(13)


(21)


(14)




(15)




(16)




(17)





(18)




(19)


(22)





(23)




(24)





(25)


(26)


K>





I-


(20)








(27) ll3

(34)

fJ/f

(28) (35)




(29) (36)





(30) (37)






(31) (38)





(32) (39)




(33) >: (40)


S*









(41)


(47)


(42) (>


(48)


Li
NN-Tos


(43)








(44)


(45)


Fe(CO)3


HH HH
TOS-NN NN-TOS


(46)


Fe(CO)


(49)







(50)








(51)


(52)


CH3 OCH3







HO HH
NN-TOS





H
NN-TOS









^c1


Ph
Ph -NO

NCO2Et











(60) CH2N2







(61) HNCO


(55)


Ph CO2CH




Ph "C02CH3


(56) jNCO2CH





SPh
Ph
NO

(57) NCNH
1 2
0




Ph

(58) h
Ph



NO
(59) CH NCNH
3 2
0


(62) O O

HN0
O


Me

(63) N

N NN
Me T NMe




NO
-NCOCH


(64) c


(65)


Ph
Ph I
-h i


(53)


(54)









(73) Cl / CONO


(66)







(67)







(68)







(69)







(70)






(71) AgNCO


(74)


(75)


H
NCO2CH3


H
-NCONH2


NO2


(77)






(78)


(72)


(76)


NO2


-NO2

- ONO




0
H I
-NCOCH3


(79)