”1‘

ill

1H

1

  

\‘ ‘
’
‘

'lH‘l

 

W

I

 

W

335 WHII
(I) —\ —\ I

  

.pHoTocHEMICAL asawmzmon 2N THE
TETRACYCDO [4.3.013 2'4.03'7]N0NA-8-ENE SYSTEM

Thesis for the Degree of Ph. D.

MICHXGA‘Q STATE UNIVERSITY

Raymond Joseph Barreras
1966

LIBRARY

This is to certifg that the

thesis entitled
PHOTOCHEMICAL ISOMERIZATION

2,4

IN THE TETRAcycwE4.3.o.o .03'7JN0NA-8-ENE

SYSTEM
presented by

Raymond Joseph Barreraa

has been accepted towards fulfillment
of the requirements for

Ph. D. degree in Chemistry

/

Major professor

Date______//'/;/’6 6

0-169

 

 

 

 

ABSTRACT

?HOTOCHEMICAL ISOMERIZATION

2,4

SYSTEM

by Raymond Joseph Barreras

The first example of photochemical cyclo-
addition of a cyclopropane ring to an olefin was

reported by Prinzbach (I -——9 II) (1).

CH301C \ COLCH}
15’

II

 

This investigatit- has disclosed a second
example of intramoleculgr cycloaddition of a cyclopropane
system to an olefin. 1:3 cycloaddition occurs readily
even though the conversion of 8,9—dicarbomethoxy-
2’4.03’7]nona-8-ene (III) to

V 6 8
3,7-dicarbomethoxypentacyclo[3.3.1.02’4.O ’ .03’7i-

tetracyclo[4.3.0.0

nonane (IV) results in an increase in strain.

Raymond Joseph Barreras

\

1
co CH
C0195 z. 3

The isomerization of III to IV was effected by

irradiating a dilute pentane solution (100 mm. III in

C)

300 ml. solvent) for five hours. A 2 0 watt medium

a.

pressure immerSion lanp (Hanovia) equipped with a

#6:)

Vycor filter was used to ef cct a 35% conversion to IV.
The structure of the pentacyclic photoproduct
was deduced from mass spectral, infrared, and nuclear
marnetic resonance data, and the relative therna
stability of IV. I-{eltin at 53-40, IV exhibited
infrared absorption at 3055, 2950, 2860, 1740, 1435,

h ,A a r- ..1 a -
13 o, 1105, and 1070 m. . The mass stectrum OI iV

(h

showed a parent peak at n/e 23% ard the n.m.r. spectrum

sonance at 17: 8.11, 7.40, 7.27, and 6.39 (area

D"

U)

I‘

(I)

,.
(.4:

ratios 1:2:1:3 respectively).

H... fix? . ~,‘ 3"! ~ "‘

-ne photoisomer, -4, is stasis to heatinC to
::(‘:O "‘ (1 v- ') C " k. '3‘. " r "' \F"f\ " for‘"" c. q‘mq
JUU M \A-eCO M ()9CU “U V 5“.st v U v A; V; “Udrb . fi‘c‘!
+"‘ flr‘pwr ‘1 " +Afl ~ - I w - / -fi~ ~+‘—-‘2 ni- A‘AO
UbV‘MUV-C—-c CieSbc- , -.....‘..’ *u UAAG ...... .h.f DUQKJ—~\a LXU CVU
n sq A -‘ I - f :O O

Raymond Joseph Barreras

only recoverable material is unreacted III. No trace
of the photoproduct is found.

An intermolecular reaction between an olefin and
a cyclopropane ring did not occur when mixtures of
maleic anhydride/norcarane or cyclohexene/1,1,2,2-

tetracarbomethoxycyclopropane were irradiated.

H'“ ~ ~-\\\T T5
Rh... “A C A.

1. H. Prinzbach, W. E? ‘xtch, an; G. Von Veh,
Ang. Chemie (Int -: 31.1, 4, 436(1965)

PHDTOCHEMICAL ISOMERIZATION

291+.0397

IN THE TETRACYCLO[4.3.0.0 ]NONA—8-ENE

SYSTEM

BY

Raymond Joseph Barreras

A THESIS

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

DOCTOR OF PHILOSOPHY

Department of Chemistry

1966

ACKNOWLEDGMENTS

The author wishes to express his appreciation
to Professor William H. Reusch for his interest,
inspiration and encouragement throughout the course
of this investigation.

Grateful acknowledgment is extended to the
National Institutes of Health for personal financial
assistance from September, 1963, to September, 1965.
Appreciation is also extended to the Dow Chemical
Company for a summer fellowship in 1966.

Special thanks are also extended to my wife
and four children whose patience and quiet endurance
has made it all possible. Finally to my parents
for giving me every opportunity to advance, a warm
De Colores.

W’é‘} Vim-$154.3 firth-lb“

ii

TABLE OF CONTENTS
Page

INTRODUCTION 0 O O O O O O O O O O O O C C 1
RESULTS AND DISCUSSION . . . . . . . . . . 4
mmniENTfl O O O O O O O O O O O O O O O 19

9-Dicarbzmethoxytetracyclo—
[1}. 3. O. 02 ’ .30 ’ 7]n0na~8~ene, v o o o 20

3 7-Dicarb methoxygentacyclo-
[3 3.1.02 0 ]nonane, VI . . 20

Tetracyclo[4. 3 O. 02 4 .03 7]nona-
8-6116-8, 9-6-100 361d, VIIa-o o o o o o 25

MethOd II o 0 o o o o o o o o o 25

8, 9- -Dicargomethoxytetracyclo-

[4.3.0 7]nonane,x..... 27
9-
30

8, Dih d ox thyltetracyclo-
[4: o 0.0§’£o Gym?]nona-8-ene O o o o o 28

MethOd II o o o o o o o o o o o 32

Attempted synthesig: zf g, 3-dimethyl-
tetracyclo[ 3.0 O ]nona-
8-6ne, VIIG. o o o o o o o 32

Attempted synthesis zfo @: phenyl-
tetracyclo[ 3 0 O ]nona-
8-8116, IXb o o o o o o o o 34

Attempted synthesia 3f 9,9-diphenyl-
tetracyclo[4. 3.0.0 ’ ]nona-
8-6116, 1X8 0 o o o o o o o o o o o o 34

Pyrolysis of 8,9-dicir $3m$thoxy-

tetracyclo[4.3.0.02 :i ]nona-
8-6116. 0 o o o o o o o o o o 35

iii

 

 

 

TABLE OF CONTENTS - Continued

Page
Pyrolysis of 3 7-d c rb m th x -
pentacyclo[3.3:1.Oisifigvgflg’¥ -
nonane, VI 0 O O O O O O O O O O O O 36
Deuterium Exchange of 8,9—dicagb9-
methoxytetracyclo[h.3.0.0 ’ .0 ' -
nona'B-ene’ V. o o o o o o o o o o o 37

Deuterium Exchange of 3 7-d ca b -
methoxypentacyclo[3.3.1:02'£.Og'g.03’7]-
nonane, VI . . . . . . . . . . . . . 37

Irradiation of Maleic Anhydride and
Norcarane. . . . . . . . . . . . . . 38

Irradiation of 1,1,2,2—tetracarbo-
methoxycyclopropane and cyclohexene. 38

BIBLIOGRAPHY O O O O 0 O O O O O I O O O O 41

iv

Table 1 :

LIST OF TABLES

Mass spectrum of VI .

LIST OF FIGURES

Figure Page
1a and 1b

Infrard Spectrum of VI . . . . . . . 7,8
2 Nuclear Magnetic Resonance Spectrum

Of VI. 0 O O O O O O O O O O O O O O 10
3a and 3b

Infrared Spectrum of V . . . . . . . 21,22

4 Nuclear Magnetic Resonance Spectrum
or v 0 O O O O O O O O O O O O O O O 23

5 Nuclear Magnetic Resonance Spectrum
or v11&. 0 O O 0 O O O 0 O O O O O O 26

6a and 6b
Infrared Spectrum of X . . . . . . . 29,30

7 Nuclear Magnetic Resonance Spectrum
Of x O O O O O O O O O O O O O O I O 31

vi

It has been well known.for a considerable period
of time thatcx,B-unsaturated carbonyl compounds undergo
dimerization when irradiated (1). The products are
cyclobutane derivatives arising from the olefinic port-
ions of the molecules. Sometimes a mixture of products
distinguishable as head-to-head or head-to-tail dimers

result (2). Further examples of such dimerization are

0 O
r3 0 u
_’ Q + 3
.; a

o
seen in the work of Griffen (3) with maleic and fumaric

 

 

 

 

 

 

 

 

acids and Hammond (4) with the coumarin system.
In the presence of olefins certain 01, til-unsaturated
carbonyl systems undergo a similar reaction known as

cycloaddition. (5,6,7)

0 O

C) + 0 > ‘35
C5 + 5k 2 O

0

i + I OCH; “Log/r43

.11.
O CLLLfi3

 

 

 

 

 

 

 

 

\/

 

Intramolecular formation of cyclobutanes occurs

even when a highly strained system results (9,10).

 

\r
O

 

COZH

\/

 

L
fl” cozu

coLH

Since cyclopropane systems exhibit many properties
similar to those of olefins (8), an investigation of
photochemical cycloaddition reactions of three membered
rings seemed worthwhile.

1+1 ——>

 

 

 

 

 

 

Recent reports of intramolecular photochemical

rearrangement of tricyclo[3.2.1.0.2’4]octene derivatives

(I) to tetracyclo[3.3.0.02’8.O4’6]octane analogs (II)

involve such a transformation (11,12), as must the

reaction of 2-butyne with dimethyl 3,6-dihydrophthalate
to give (IV) via an intermediate containing a cyclo-
propane ring and a double bond (III) (13,14).

04,... 1 ”3,,
wh/ —» Air

 

 

 

 

 

 

COZC’G
m
c0101} C _.
I
III
C LvLCHj
, C L LCH3
CH3 \CH3

RESULTS AND DISCUSSION

In order to enhance the desired cycloaddition, an
intramolecular model was chosen for this study. In part-

40 03 ’ 71110118."

icular, 8,9-dicarbomethoxytetracyclo[4.3.0.02'
8-ene (V) was used, since models indicated that the
olefinqt-orbitals and the cyclopropane 2,3—bond over-

lapped well.

 

Synthesis of V was reported by Schrauzer and
Glockner (15) and proceeded in 32% yield by reaction of
dimethyl acetylenedicarboxylate with.bicyclo[2.2.1]hepta-
2,5-diene in the presence of a complex nickel catalyst.
The 1:1 adduct was obtained in up to 50% yield by heating
an equimolar mixture of diene and ester (16). In reflux-
ing toluene the yields range from 57 to 65%.

Interactions of non-conjugated chromophores have

been seen in many systems. The non-conjugated interaction

4

usually falls between the isolated carbonyl (A) and the
conjugated carbonyl (E). The location of the absorption

“31% On 0“

A
A
max 275 280 297 310 330

6 14.1 107 107 288

bands in the non-conjugated ketones (B-D) is related to
the stereochemistry of the lone pair andst-orbitals of
oxygen in relation to the u-orbitals of the carbonecarbon
double bond (19). Clearly, u—orbitals will tend to
interact best, when appropriately oriented with respect
to each other. Thus, the shift in absorption from 297mL1
to 310mu.in compounds 0 and D indicates a better over-
lap of the orbitals in the bicyclo[2.2.1]system.

The crystalline diester v, m.p. 65-66°, exhibits
an absorption maximum at 244mu(log e 3.98) in the ultra-
violet, indicating transannular interaction between the
x-orbitals of the double bond and the bonding orbitals of
the three membered ring. The significance of this inter-
action may be judged by comparison with dimethyl dimethyl-
ax 230mu(log€ 3.83)
(12) and 2,3-dicarbomethoxybicyclo[2.2.1]heptadiene,

maleate, kmax 212.5mu(log e 3.95); I,)\m

xmax 237mu (log 6 3.60) (10), inasmuch as the latter

two compounds undergo efficient transannular photo-
chemical cycloaddition.

Early attempts to isomerize V led to extensive
dimer formation, largely because the conditions were too
severe and the concentration of substrate too high. In
dilute pentane solution (100mg. V in 300 ml. solvent) a
five hour irradiation led to a 35% conversion to VI.
This substance, m.p. 53°, was purified by preparative
v.p.c., and identified by a variety of spectroscopic

measurements.

\/

l

V VI

COLCHj

The infrared spectrum of the photoproduct, VI,

1 which can be

(Figure 1) has bands at 3050 and 1025 cm.-
assigned to cyclopropyl hydrogens (26). The absorption
at 1735, 1235, and 1225 cm."1 is assigned to the carbo-
methoxy group and the lack of absorption in the 1700-
1500 cm."1 region indicates the absence of a double bond

(26).

oom—

ooom

A :23 >UZmDOwE

H> no 85.90QO vegans”
3” 0.33%

comm ooom

 

L I\_ll I llAth l1. I‘ll

INFW , ,. . fl _ . .. .. 4 ~ _ i _ .. _ V _ _ #1. « i a . .. x , x x . \\.\n\~

ON

CV

00

om

.0—

00m

ooop

H> .Ho 8593QO pandas”
fl” warm

:5. 02303:

oom_ oovp

ooo_

oow_

000m

 

The mass spectrum of VI (20) confirms its monomeric
nature (parent ion at m/e = 234). Additional prominent

ions in the mass spectrum of VI are listed in Table 1.

 

m/e Assignment
234 / u"
203 / (M- can}?
202 /
187 /

+
174 / (M- 0020113- H)
143 /
116 /
115(base) /
91 / (C7H7)+
.59 / (0330c o)+

Table 1: Mass spectrum of VI

The strongest evidence for structure VI was found
in the n.m.r. spectrum(Figure 2) (20), which displayed
resonance signals at1': 8.11, 7.40, 7.17 and 6.39 (area
ratio 1:2:1:3 respectively). The appearance of the
cyclopropyl hydrogen signal at rather low field is

probably due to a combination of carbon bond angle

lO

 

 

 

O.m

 

esesoefia.mo.mxmo.e.mo.a.m.nuaxoseesopeeeae-a.m
H> mo 8450.0QO ooGSQOmom caposwmz smmaosz "m mssmflh
L. A
- mm. 03%.. m.fl\u

 

 

W

 

 

 

 

 

 

11

constriction and electron withdrawal by the carbomethoxy

substituents. Ekamples of these effects can be seen in
4,6]_

hexane (cyclopropyl hydrogen signals at1-= 8.43 and

other systems, such as 1-methyltricyclol3.1.0.0

8.14) (21), quadricyclene (cyclopropyl hydrogen signal
at T: 8.59) (22), and 1,1,2,2-tetracarbomethoxycyclo-
propane (cyclopropyl hydrogen resonance at T: 7.82) (23).

Since the low field cyclopropyl hydrogen resonance
of VI conceivably reflects an increase in the acidity of
these hydrogen atoms (24), base-catalyzed exchange
reactions of VI were investigated. The photoproduct VI
was, however, recovered unchanged from a solution of
sodium methoxide and methanol-O-d after 72 hours.

To test the possibility of thermal conversion to
VI, oxygen-free samples of V were heated at 200°, 306°,
368°, and 412°. Although compound V was unchanged after
19 hours at 200°, considerable darkening of the samples
was observed at 3060 after heating periods of 15 minutes,
1, 3, and 15 hours. An insoluble carbonaceous material
was filtered off and the infrared spectrum of the soluble
pyrolysate was taken. The spectrum was superimposable
with that of authentic V, and v.p.c. chromatography on
FFAP showed only one component, the starting material.
Pyrolysis of v at 368° for 1 and 3 hours gave similar

12

results. Complete decomposition occurred in 15 minutes
at 412°, and v.p.c. analysis showed that none of the
products corresponded in retention time to either V or
VI. In this higher temperature pyrolysis the carbon-
aceous material was again observed. No attempt was
made to identify these pyrolysis products (about
fifteen in number).

The photoproduct, v1, is stable at 300°, being
recovered unchanged after heating for 1 and 3 hours.
At higher temperatures the decomposition of VI did not
include reconversion to V.

Attempts have been made to prepare other deriv-

2’4.03'7]nona-8-ene

atives of the tetracyclo[4.3.0.0
system (VII a-e). Fbr example, V was hydrolyzed with

refluxing aqueous sodium hydroxide until the oily organic

 

v R.= R! = 002cm3
VII a) R.= R' = cogs

b) R = R' = CH20H
c) R.= R3 = CHQOTs
d) R.= R' = CH20Ac
e) R.= R' = CH3

IX a) R.= R' = C6H5
b) R.= C6H5, R' = H

13

layer had dissolved and a crystalline diacid (VIIa),
having oxygen-hydrogen stretching at 3000 cm."1 and

'1 in the infrared,

carbonyl stretching at 1695 cm.
was obtained. Treatment of a slurry of VIIa with
diazomethane in ether gave a low yield of the starting
ester, V.

Reduction of V with excess lithium aluminum
hydride gave only partial conversion to impure VIIb.
This compound exhibited strong absorption in the infrared
at 3350 cm.”1 (broad, ascribed to intramolecular
hydrogen-bonded hydroxyl groups), 3052, 2935, and 2860
cm.-1. There was also weak absorption (ca. 20%) at
1720 cm.'1, indicating some remaining carbonyl function.
A semi-crystalline diacetate derivative of the diol was
made by treatment of VIIb with acetic anhydride under
reflux. The diacetate, VIId, had a melting point below
room tmeperature and could only be recrystallized from
ethyl acetate with difficulty. Infrared bands at 3050,
2915, 2851, 1740, 1370, and 1260 cm.-1 indicated an
ester structure.

As the first step in conversion of V to the deoxy
compound, VIIe, via lithium aluminum hydride reduction
of the ditosylate, VIIc, the diol was treated with
p-toluenesulfonyl chloride. The product exhibited no

hydroxyl absorption in the infrared, but also showed no

14

absorption characteristic of the p-toluenesulfonate
group, 1.9. 1230-1150 and 1440-1350 cm."1 (26 ). A
possible explanation for this puzzling result involves
initial formation of a mono-tosylate followed by Sn2
displacement of the tosyl group by the free hydroxyl.

Such a route would lead to a dihydrofuran.

‘ 1
fl—» 5

0

- of
cm\ /,cui’ 5

-. .0
H
Attempts to prepare derivatives of the tetracyclo-
[4.3.0.02'4.03'7]nona-8—ene system included the reaction
of bicyclo[2.2.1]hepta-2,5-diene with various substituted
acetylenes. Among the acetylenes used were diphenyl-
acetylene, phenylacetylene, acetylenedicarboxylic acid
and 2-butyne-1,4-diol. In each case the starting materials
were recovered unreacted. The work of Schrauzer and
Glockner (15) on the nickel cyanide-triphenylphosphine
catalyzed reactions indicates that the reaction of
diphenylacetylene with.bicyclo[2.2.1]hepta-2,5-diene
yields a tricyclic substance rather than the expected

tetracyclo derivative.

15

 

1 1
fl” + ICH —9
's

Hydrogenation of V at atmospheric pressure using a
10% palladium on charcoal catalyst yielded a dihydro-
derivative, x. Inspection of models shows that the

L

043cc, x bozo),

frontside of the molecule would be less hindered and so,
although not rigorously proven, the stereochemistry of
X is probably as indicated.

Having observed the photoreaction of an olefin
and a cyclopropane ring in an intramolecular process,
the investigation of several analogous intermolecular
systems was undertaken. In order to minimize possible
dimerization reactions, a dilute solution of the cyclo-
propane in the olefin or of the olefin in cyclopropane

was studied.

16

After irradiation for 60 hours, 1,1,2,2-tetra-
carbomethoxycyclopropane in cyclohexene failed to give
any products corresponding to a 1:1 adduct of the
two reactants. Neither cyclopentane product (A and B)

expected from the reaction could be detected.
R K R
R R K
+ <1 +-> ”2+ 00

R R

R R
R

A B

R = C 01013

0 0 0
(rec—e and. d1?
0 o

A mixture of maleic anhydride and norcarane
proved to be equally unreactive when irradiated. In
contrast, Schenk has reported a 95% yield of the
corresponding cycloaddition.between maleic anhydride
and cyclohexene (25).

The lack of reactivity of the two systems may be
attributed to an entropy effect, whereby the olefin
and cyclopropane never achieve the proper orientation to

allow the necessary interaction to occur. Alternatively,

17

the energy initially absorbed by the chromophore may
be lost by collision with solvent molecules prior to
reaction.

After our structure proof in this system was
completed a communication reporting a similar photo-
chemical isomerization appeared in Chemical

Communications (27).

L
as —9 41a

C01R C012
C022 (1012
V R .2: CH3 VI R = CH3
VIIa R = H XI R.= H
;_ _

L@ L
., 5; __) my _

 

 

 

 

 

 

 

 

/ CCLR
C023 C0,)? ,
COLR
_ .1
ant
R C 02 \
L g 2‘
\COIR

XII
C0 R

t

18

The n.m.r. spectrum of the photoacid sodium
salt (X1) is reported to have absorptions at—rz 6.73,
7.07, and 7.65 which agree well with the spectrum of
the photoester, VI (resonance peaks at'rz 7.17, 7.40,
and 8.11. Multiplicity and peak area ratios are the
same in both compounds.)

These authors reported a mixture of liquid
products upon irradiation of the ester, V, while this
work indicates a clean isomerization to one photo-
product, VI. That the photoproduct has structure VI
instead of the alternately proposed structure XII is
best evidenced by the thermal stability of the photo-
isomer and some preliminary x-ray data cited in the
communication (27). Structure XII has two four-
membered rings which would be expected to be thermally

unstable.

EXPERIMENTAL

Melting points were determined on a Koefler hot
stage. Infrared spectra were recorded on a Perkin-
Eflmer Model 237B, Double-beam grating spectrophotometer,
and the ultraviolet spectra were taken on a Beckman DB
spectrophotometer. The nuclear magnetic resonance spectra
were determined in carbon tetrachloride solution using
a varian Associates, A-60, spectrometer and a varian
Associates, HR-100, high resolution spectrometer. Vapor
phase chromatography analyses were made using an
Aerograph, A-90-P, gas chromatograph with a 0.25 in.,
6-ft. column of FFAP on Chromasorb G. Micro-analyses
were performed by Spang Microanalytical Laboratory,

Ann Arbor, Michigan, and by Micro-Tech Laboratories,
Skokie, Illinois.

19

20

8 -Dicarbomethox tetrac clol 4.3.0.02’4.03"fi nona—8-ene, V"
.A solution of bicyclo[2.2.1]hepta-2,5—diene

(9.2 g., 0.1 m.) and dimethyl acetylenedicarboxylate
(5.1 g., 0.036 m.) in 12 ml. toluene was refluxed for 20
hours. The toluene was removed from the amber solution
under reduced pressure and the dark residue was vacuum
distilled at 1 mm. and yielded a clear liquid, b.p. 118-
120°, which solidified on standing, m.p. 59-60°. The
yield of this 1:1 adduct was 4.8 g., 0.02 m., 57%.
Crystallization from pentane gave a white crystalline
material, m.p. 65-660, )‘max 244 ml , loge 3.98; vmax(CCl¢,_)
3060, 2990, 2945. 2860, 1720, 1615, 1440, 1340, 1225, 1160,
1090 cm."1 (Fig. 3);? = 6.31 (6H—sing1et), 7.05 (2HHbroad
multiplet), 7.82 (1Hebroad multiplet), 8.20 (1Hébroad
doublet), 8.41 (4HAbroad multiplet) (Fig. 4).

Analysis: Calc. for 013H1¢04: 0, 66.64%; H, 6.02%

Found: C, 66g3113 H, 5.93%.

2’4.06’8.03’7]nonane,v1

3,I-Dicarbomethoxypentacyclo[3.3.1.0
Photoisomerization of V was accomplished by
irradiating a solution of 100 mg. V and 300 ml. pentane
with a 200 watt medium pressure immersion lamp (Hanovia)
equipped with a vycor filter. The solution was irradiated

for five hours at room temperature, evaporated on a

Rinco flash evaporator, and the resulting oil was

> mo suspoomm consumaH
on ohsmam

 

A _.<<U. >UZwDOmum

00.2 ooom comm ooom 00mm ooov

om

34

21

up

[All

om

 

.9

ON

9..

22

00

ow

vo—

 

com

000—

> .8 25.30on @2555

pm 0.533

a $2”: >UZmDOm¢m

oom—

cov—

000—

I

oow_

 

 

Inl\

omouwumqosmhamo.:amo.o.m.:QOHohomspophxogpoeopmmoflpum.w

> mo asapommm ooqmcomom oflpocwmzflnmoaosz ”: ossmflm

FT

 

O.QH O.m

 

 

 

 

i _

 

 

23

 

 

 

24

dissolved in benzene and separated by vapor phase
chromatography. Two peaks appeared in the chromatogram.
The first peak (65%) was collected and found to be
identical to V'in every respect (m.p., i.r., and
n.m.r.). The second peak (35%) solidified upon cooling
in the collection tube and melted at 534°. The infrared
spectrum showed absorption at 3055, 2950, 2860, 1740,
1435, 1360, 1105, and 1070 cm.-1 (Fig. 1). The mass
spectrum of VI showed a parent peak at m/e 234 (Table 1)
and the n.m.r. spectrum had resonance at T = 8.11, 7.40,
7.17, and 6.39 (area ratios 1:2:1:3 respectively)
(Fig. 2).

Analysis: Calc. for 013H1404: C, 66.64%; H, 6.02%

sound: 0, 66.48%; H, 6.33%.

When photolysis was carried out in a more
concentrated solution ( e.g., 200-900 mg.) or for times
exceeding five hours, several secondary reactions
occurred which led to the formation of dimers. Their
structures were not determined but evidence of their
dimeric character came from the retention times on
the gas chromatograph. After irradiation for nineteen
hours a one gram solution of V in pentane showed several

peaks at retention times of 1.1-1.3 hours.

25

TetracycloI4.3.0.02’4.03’71nona-8-ene-8,2-dioic acidI Ella

To a 0.1 g. (0.00043 m.) 8,9-dicarbomethoxy-

294.03’7]nona-8-ene was added 5 8-

tetracyclol4.3.0.0
sodium hydroxide and 50 ml. water. The mixture was
refluxed for two days, during which time the oily
organic material had gone into solution. Upon cooling
and acidification with dilute hydrochloric acid, a
white flocculent precipitate formed. Extraction with
ether gave the diacid, VIIa, in 80% yield, m.p. 217-9°.
The n.m.r. showed resonance at 'r= 6.77, 7.73, 8.02,

and 8.33 with area ratios of 2:1:1:4 respectively.

(The spectrum was run in hexadeuteroacetone with
tetramethylsilane as an internal standard. A quintuplet
at 1': 7.95 is correct for acid hydrogens partially

exchanged for deuteriums in the hexadeuteroacetone.)

(Figure 5).

Method II

An alternate method of synthesis of the diacid,
VIIa, is from bicyclo[2.2.tlhepta-2,5-diene and
acetylenedicarboxylic acid. A solution of 2.73 g.
(0.025 m.) acetylenedicarboxylic acid and 2.78 g.
(0.31 m.) bicyclo[2.2.1]hepta-2,5-diene in 20 ml.

toluene and 5 ml. 95% ethanol was heated under reflux

26

Uwom ooap1mawnosoumumsocnw«no.

mHH> uo abhpoommr

o.®

 

 

 

 

 

:.mo.o.m.:~oaozomspoe
wmgmmomom caposmmz smoaosz ”m ossmwm

cg

\1/1111

 

 

27

for two days. After cooling to room temperature, the
reaction mixture was evaporated to remove the solvents.
No 1:1 adduct could be isolated from the remaining
yellow oil. Thin-layer chromatography showed no less
than nine spots, one of which corresponded to the
diacid derived from the diester, V, upon hydrolysis.

No further attempt was made to use this route

synthetically.

8,Q-Dicarbomethoxytetgacyclo [1}.3.0.02’4.03’.7]nonaneI X
A solution of 0.234 g. (0.001 m.) 8,9-dicarbo-

2'4.03’71nona-8-ene in 100 ml.

methoxytetracyclo[4.3.0.0
ethyl acetate was stirred in a 125 ml. round bottom

flask fitted with a condenser and a three-way stop-

cock, having one outlet leading to a vacuum system

and one to a banbon of hydrogen gas. Palladium on
charcoal (0.1 g., 10%) was added and the system evacuated
until the ethyl acetate started to boil. Hydrogen was
allowed to enter and the evacuation-filling process
repeated three times. The reaction mixture was then
stirred for five hours. The catalyst was removed by
filtration and the resulting clear solution was
evaporated. Crystallization was effected from ethanol,

yielding a product melting at 57°. The infrared of
this material showed bands at 3055, 2950, 2870, 1745,

28

1435, 1350, 1315, 1235, 1190, and 1170 c111."1 (Fig. 6).
The n.m.r. showed absorbance atT'z 6.38 (singlet),
6.89 (broad triplet), 7.75, 8.17, 8.49, and 8.74

(all broad multiplets) with area intensity 6:2:1:1:2:3
respectively (Fig. 7).

2’4.03’?lnona-8-ene, VIIQ

8,Q-Dihydroxymethyltetracyclo[4.3.0.0

A slurry of 0.4 g. lithium aluminum hydride and
0.24 g. (0.001 m.) 8,9-dicarbomethoxytetracyclo-
[4.3.0.02'4.03’7] nona-8-ene in 70 ml. ether was heated
under reflux for 1.5 hours. The excess hydride was
destroyed by careful addition of ethyl acetate. A
solution of saturated aqueous ammonium chloride (25 ml.)
was then added. The reaction mixture was filtered and
the filtrate extracted with ether. Some additional
product was also obtained by washing the collected
precipitates with portions of ether. Evaporation of
the solvent from the combined extracts yielded an oil,
the infrared spectrum of which still showed a weak
carbonyl absorption at 1720 cm.-1.

Upon treatment with excess acetic anhydride in
pyridine the infrared absorption band at 3350 cm...1
present in the diol disappeared. The product from
this reaction, presumably the diacetate, exhibited a

typical ester odor and showed acetate absorption at

N mo asapoomm 60.3.”qu

mm mhsmfim

 

om

um

29

co

ow

con.

ooom

 

:38 05:02:

oonm

ooom

comm

ooow

 

'7“.L I\_ll ll"

 

"u ‘1‘ 1" .‘l‘l.ln.

x .8 €9.30on umnmflHnH
pm THREE

:28 02303:

89 8: 8o— 82 88

 

cm

00—

51

masses: .mo .J .mo .0 . m I: oaozomapofinkogpmaophmoflwum .w
\P x Mo 38on mogqomomfowposmmz amoaosz " N. 8de

Tm 9m To

)IIP

 

 

 

32

1740 and 1260 cm.-1 in the infrared. The diester was a
non-crystalline material which formed small crystals
from a cold solution of ethyl acetate; however, these

melted below room temperature.

Method II
An alternate preparation of 8,9-dihydroxymethyl-

2’4.03'7)nona-8-ene was attempted

tetracyclol4.3.0.0
utilizing a reaction similar to that employed in the
preparation of the tetracyclic ester, V. A solution of
8.6 g. 2-butyne-1,4-diol and 9.2 g. bicyclo[2.2.1]hepta-
2,5-diene in 15 ml. dimethylsulfoxide was heated under
reflux for two days. vacuum distillation at 4 mm.
yielded the bicycloheptadiene in a forerun, dimethyl-
sulfoxide as the major distillate, and residues which
were essentially unreacted 2-butyne-1,4-diol. Further

preparations were not attempted.

Attempted synthegis 0f 8,9-dimethyltetracyclo-

2,4

[4.3.0.0 .03’71nona-8-ene, VIIQ

A sample of the impure 8,9-dihydroxymethyl-

2’4,03’?lnona-8-ene (0.1 8o: 0°00043 m.)

tetracyclo[4.3.0.0
was dissolved in 5 ml. pyridine and 2 g. of p-toluene-

sulfonyl chloride in 5 m1. pyridine was added. The

33

mixture was heated under reflux for two hours. After
cooling to room temperature, 25 ml. of water was added
and the aqueous solution was extracted twice with ether.
The ether was washed with a 10% sodium bicarbonate sol-
ution and twice with water. The organic layer was
dried over sodium sulfate and evaporated, yielding a
colorless oil. Except for carbon-hydrogen absorption,
the infrared spectrum of this oil showed no absorptions
ascribable to the expected tosylate, VIIc. The
p-toluenesulfonate grouping is known tohave absorption
in the infrared at 1230-1150 and 1440-1350 cm.-1 (26).
Absorption was lacking in these regions.

The oil was dissolved in ether and 0.2 g. lithium
aluminum hydride was added. The reaction mixture was
allowed to remain at room temperature for several days,
at which time the excess hydride was destroyed by adding
ethyl acetate. EXcept for a slight increase of 0-H
absorption in the infrared, presumably due to reduction
of any residual carbonyl present after the first reduction,
the infrared spectrum was essentially the same as that

prior to reduction.

34

‘Attempted synthesis of 8-phenyl;
14.3.0.02’4.03’7]nona-8éene,IXQ

Phenylacetylene (30.6 g., 0.3 m.) and bicyclo-
[2.2.1]hepta-2,5-diene (27.6 g., 0.3 m.) were refluxed
for twelve hours. The initially light yellow solution
turned a deeper yellow and finally a rich amber red.
Upon distillation of the reaction mixture, only

unreacted starting materials were recovered.

‘Attempted synthesis of 8,9-diphenyltetracyclo-
14.3.0.02'4
A solution of 2 g. (0.02 m.) bicyclo[2.2.1]hepta-

.03’7lnona-8-ene,IXa

2,5—diene and 0.32 g. (0.002 m.) diphenylacetylene (tolan)
in 5 ml. toluene was refluxed for 19 hours. Upon flash
evaporation of the volatile components, a creamy yellow
powder was obtained. Recrystallization from pentane
gave quantitative recovery of the diphenylacetylene.

When the reaction was repeated in the absence of
toluene at the temperature of refluxing bicyclo[2.2.1]-
hepta-2,5-diene (approximately 150°) no adduct was

observed.

35

Pygglysis of 8,2-dicarbomethoxytetracyclo-
[4.3.0.02’4.03’7]nona-8-ene

In the pyrolysis eXperiments the glass vials were
prepared in the following manner.

Pyrex tubing (8 mm. 0.D.- 4 mm. I.D.) was cut into
20 cm. lengths and placed in a bath of nitric acid for
5 hours. The tubes were rinsed thoroughly with distilled
water and covered with ammonium hydroxide for 30 minutes.
After rinsing well with distilled water, they were placed
in a drying oven at 120° for 7 hours. The tubes were
then sealed and stored in a desiccator.

Samples of compound V (ca. 100 mg.) were degassed,
sealed under nitrogen, and heated in an aluminum block
furnace at 200°, 302°, 368°, and 412° for various
lengths of time. The temperature was controlled to
t 2°C. After 19 hours at 2000 a sample of V was completely
unchanged. The clear, colorless liquid solidified upon
cooling to room temperature, and its infrared spectrum
was identical with that of the unheated V.

Upon heating to 3020 for 15 minutes, 1, 3, and
15 hours, each of the samples became quite dark and
discolored. At the longer reaction times dark, chunky
flakes of carbonaceous material formed. After cooling,

the vial was broken and the organic material dissolved

36

in.benzene. vapor phase chromatographic analysis

of these benzene solutions exhibited one peak at

the retention time of the starting material in each case.
Infrared spectra of the samples in 0014 were identical
to those of V.

At 368°, samples heated for 15 minutes and 1 hour
again exhibited decompostion. The soluble organic
material again proved to be the starting ester V.

Pyrolysis of V at 4120 also led to extensive
decomposition, but v.p.c. analysis of the soluble
components showed a complex mixture of products, none
of which corresponded to the starting diester, V, or
to the photoproduct, VI. No attempt was made to
identify these substances.

 

Pyrolysis of 3,z-dicarbomethoxypentacyclo-
[j.3.1.02’4.0°’°.03’7]nonane, V;

Photoisomer, VI, remained essentially unchanged
when heated to 300°. At higher temperatures the
decomposition did not involve conversion to the

starting diester, V.

37

Deuterium Exchange of 8,g-dicarbomethoxytetracyclo-

 

A mixture of 145 mg. (0.00062 m.) V, 56 mg. sodium
methoxide and 30 ml. methanol-O-d (80%1d) was heated
under reflux for 20 hours. The reaction mixture was
cooled, added to 30 ml. heavy water (99% d), and the
aqueous layer extracted with pentane. After drying
over sodium sulfate, the pentane solvent was evaporated,
yielding a white solid. The infrared and n.m.r.
spectra of this material was identical to that of
the starting material, V.

Deuterium EXchange of 3,z-dicarbomethoxypentacyclo-
I3.3.1.02'4.06’8.03’7]nonane,,V1

 

After 72 hours of warming on a steam bath, a
mixture of 40 mg. VI and 48 mg. sodium methoxide in
30 ml. methanol-O-d (80% d) was cooled, poured into
deuterium oxide (99% d), and extracted with ether. The
solid material recovered proved to be identical to the
photoproduct, VI.

After 3 weeks of sitting at room temperature a
20 mg. sample of VI in a solution of deuteromethanol/
sodium methoxide was also recovered unchanged. The
infrared and n.m.r. spectra were taken to make comparisons
for possible incorporation of deuterium. The results

were negative.

38

Irradiation of Maleic Anhydride and Norcarane
Irradiation of a solution of 35 mg. maleic

anhydride in 1 ml. norcarane (bicyclo[4.1.0]heptane)
for two hours in a vycor test tube using a 250 watt.
medium pressure mercury lamp led to no observable
products. Analysis was by vapor-phase chromatog-
raphy.

Irradiation of 1,1,2,2—tetracarbomethoxycyclopropang
and cyclohexene

Irradiation of solutions of 45 mg. 1,1,2,2-
tetracarbomethoxycyclopropane in 3 ml. cyclohexene
for from 1 to 60 hours in quartz test tubes using a
250 watt medium pressure mercury lamp led to no
observable products. Analysis was by vapor-phase

chromatography.

SUMMARY

This investigation has disclosed a second example
of intramolecular cycloaddition of a cyclopropane
system to an olefin. The cycloaddition occurs
readily even though the conversion of 8,9-dicarbomethoxy-

2’4.03’7]nona-8-ene (V) to

tetracyclo[4.3.0.0
3,7-dicarbomethoxypentacyclo[3.3.1.02’4.06'8,03’7]-

nonane (VI) results in an increase in strain.

1 1
ii 1'.
Q r m

co1m3 €09,013 C010); C01013
V VI
The structure of the pentacyclic photoproduct
was deduced from mass spectral, infrared, and nuclear
magnetic resonance data, and the relative thermal
stability of VI.
The photoisomer, VI, is stable to heating to

368° but decomposes above that temperature. The

tetracyclic diester, V, is thermally stable at 2000

39

40

but decomposes rapidly at 306°, 368°, and 412°.
The only recoverable material is unreacted V. No
trace of the photoproduct is found.

An intermolecular reaction between an olefin
and a cyclopropane ring did not occur when the
reactants were maleic anhydride/ norcarane or

cyclohexene/ 1,1,2,2-tetracarbomethoxycyclopropane.

10.

11.

12.

13.
14.

15.
16.

17.

J.

R.

S.

BIBLIOGRAPHY

Mustafa, Chem. Rev., 51, 1(1952).
E. Eaton, J. Am. Chem. Soc., 84, 2344(1962).

W. Griffen and L.I. Peterson, J. Am. Chem. Sog.,
84. 3398(1962).

S. Hammond, C. A. Stout, and A. A. Lamola,
J. Am, Chem. Soc., 86, 3103(1964).

E. Eaton, J. Am. Chemé:§og,, 84, 2454(1962).

J. Corey, R. B. Mitra, and H. Uda,
J. Am. Chem. Soc., 86, 485(1964).

De Mayo, S. T. Reid, and R. W. Yip, Can. J. Chem.,
42, 2828(1964).

D. Roberts and M. Caserio, Basic Principles of

Dr anic Chemistr (New York: Benjamin Press,
1954), p. 114.

C. Cookson, E. Crundwell, and J. Hudec,
Chem. Ind. London , 1003 (1958) and D. F. O'Brian
and J. w. Gates, Jr., J. Org. Chem., 30, 2593(1965).

J. Cristol and R. L. Snell, J. Am. Chem. Soc.,
80, 1950(1958).

K. Freeman, D. G. Kuper, and V. N. M. Rao,
Tetrahedrom Letters, 3301 (1965).

Eberbach and G. Vbn Veh,

Prinzbach, W.

Ang. Chemie(Int'1. Ed.), 4, 436(1965).

Askani, Chem. Ber., 98, 3618(1965).

Criegee, C._and.E. News, March 1965.

N. Schrauzer and P. Glockner, Chem. Ber., 97, 2451(1964).

F. Huebner, E. Donaghue, L. Dorfman, A. Stuber,
N. Daniele, and E. Wenkert, Tetrahedron Letters, 1185(1966).

Labhart and G. wagniere, Helv. Chim. Acta,
42, 2219(1959).

41

18.

19.

20.

21.

22.

23.

24.

25.

26.

27c

42

S. Winstein, L. Devries, and R. Orlaski,
J. Am. Chem. Soc., 83, 2020(1961).

R. C. Cookson and R. waryar, J. Chem. Soc., 2302(1956).

Thanks go to Dr. Lois Durham and Mr. Robert Ross of
Stanford University for the mass spectral analysis
and the HR-100 n.m.r. spectrum.

G. L. 01088 and R. L. Larrabee, Tetrahedron Letters,
287 (1965)-

w. G. Dauben and R. L. Cargill, getrahedron, 15, 197(1961).
R. H. Starkey, M.S. Thesis, Michigan State University(1965).
The proton on C-2 of tricyclo[1.1.0.0°'lfi pentane

is acidic enough to be exchanged using methyl

lithium. of. reference 21.

G. 0. Schenk, W. Hartmann, and R. Steinmutz,
Ber., 95, 1642(1962) and 96, 498(1963).

L. J. Bellamy, The Infrared Spectra of Com lex Molecules,
(New York: John Wiley and Sons, Inc., 1958), p. 350.

C. F. Huebner, E. Donoghue, L. Dorfman, E. wenkert,
W. E. Streth, and S. W. Donely, Chem. Commg, 419(1966).