YuC'f'

 

-#
l

   
 
 

”‘j’: LIBRARY
Michigan State
University

W T— v -T.

 

 

This is to certify that the

thgsis entitled f S d
art : S nt e i an eac 1onso stitute
aphthalehe 9, 3- Epoxy— ¥,4 -Endoperoxi3es

Part 11: Synthesis and Properties of Benzo[l,2-
c:3,4-c':5,6-c"]Trithiophene, a Tristhiahexaradialene

presented by

Michio Sasaoka

has been accepted towards fulfillment
of the requirements for

Ph.D . Organic Chemistry
degree 1n

 

 

 

‘Wan J

Major professor

 

 

 

 

Date £27” 7) /777dy

0-7639

 

OVERDUE FINES:
25¢ per day per item

RETURNING LIBRARY HATERIAL§z
Place in book return to ream
charge from circulation rcco1

 

 

 

PART I

SYNTHESIS AND REACTIONS OF SUBSTITUTED
NAPHTHALENE 2,3-EPOXY-l,4-ENDOPEROXIDES

PART II
SYNTHESIS AND PROPERTIES OF

BENZO[l,2-c:3,4-c':5,6-c"]TRITHIOPHENE,
A TRISTHIAHEXARADIALENE

By

Michio Sasaoka

A DISSERTATION

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

DOCTOR OF PHILOSOPHY

Department of Chemistry
1978

kl

k\

ABSTRACT

PART I

SYNTHESIS AND REACTIONS OF SUBSTITUTED
NAPHTHALENE 2,3-EPOXY—l,4-ENDOPEROXIDES

PART II
SYNTHESIS AND PROPERTIES OF

BENZO[l,2-c:3,4-c':5,6-C"]TRITHIOPHENE,
A TRISTHIAHEXARADIALENE

By

Michio Sasaoka

In Part1 of this thesis the l,4-endoperoxides of l,2,3,4-
tetramethyl- and octamethylnaphthalene were studied. l,2,3,4-Tetra-
methyl and octamethylnaphthalene El and gg react with singlet oxygen
to give their l,4-end0peroxides, £3 and £3 respectively. Oxidation
of the carbon-carbon double bond in $3 and £3 with mfchloroperbenzoic
acid gave the stable, crystalline l,2,3,4-tetramethyl~2,3-epoxy-
l,4-epidioxy-l,2,3,4-tetrahydronaphthalene £3 and l,2,3,4,5,6,7,8-
octamethyl-2,3-epoxy-l,4-epidioxy-l,2,3,4-tetrahydronaphthalene £3-
Epoxidation occurred predominantly syn to the peroxide bridge.

The sygfepoxyperoxides underwent acid-catalyzed solvolytic
rearrangement to the stable peroxyacetals, 4-methoxy-l,4,5,8—tetra-
methyl-6,7-benzo-2,3-dioxabicyclo[3.2.l]oct-6-en-8-ol §la and 4-
methoxy-l,4,5,8-tetramethyl-6,7-(3',4',5',6'-tetramethylbenzo)-2,3-
dioxabicyclo[3.2.l]oct-6-en-8-ol 3% respectively, but the anti:

epoxyendoperoxide $38 was recovered under similar conditions. The

Michio Sasaoka

rearrangement involves a l,2-aryl migration. Catalytic hydrogenoly-
sis of 3% gave cis-l-acetyl-l,2,3-trimethylindan-2,3-diol,éé, which
was obtained independently from the acid-catalyzed methanolysis of
the sygfepoxide of l,2,3,4-tetramethylnaphthalene-l,4-endoxide 32.
Deuterium labeling studies support the pr0posed mechanism for these
rearrangements.

Thermolysis of the epoxyendoperoxide £33 occurs with 0-0 bond
cleavage, as established by trapping the intermediate diradical with
good hydrogen donors (diglyme, benzhydrol) to give l,2,3,4-tetra-
methyl-2,3-epoxy-l,2,3,4-tetrahydronaphthalene-l,4-diol gig, synthe-
sized independently by hydrogenolysis of géé' In the absence of
trapping agent, thermolysis of £33 occurs with loss of a methyl group
to give 2,3,4-trimethyl~2,3-epoxy-4-hydroxy-l,2,3,4-tetrahydro-
naphthalenone 3%. Thermolysis and photolysis of £2 gave hexamethyl-
benzo[b]furan.

Reactions of $3 with bromine or trifluoroacetic acid were also
studied.

In Part II of this thesis, synthesis of benzo[l.2-c:3,4-c':5,6-
c"]trithi0phene 22, the first example of a heteroaromatic molecule
in which the carbon-carbon "double bonds" of the central six-membered
ring are all exocyclic, was described.

Oxidation of l,3,4,6,7,9-hexahydrobenzo[l,2-c:3,4-c':5,6-c"]-
trithiophene lifi’ prepared from hexakis(bromomethyl)benzene and sodium
sulfide (80-90% yield), with 000 or ggchloranil gave QQ- Unlike
benzo[c]thiophene, $2 is unreactive toward dienophiles. It forms
crystalline l:l charge transfer complexes with TCNE, DDO, TCNO and

chloranil. Protonation of fig in FSOBH (-20°C) gives a stable dication.

Michio Sasaoka

fig undergoes electrophilic bromination (Br2

ture). Metallation of fig followed by reaction with DMF or CO2 gave

, CCl4, room tempera—

the monoaldehyde and monocarboxylic acid of fig, respectively.

The overall aromaticity of systems such as $3 was further
illustrated by synthesis of benzo[l,2-c:3,4-c']dithiophene 3; via
the 000 oxidation of l,3,4,6-tetrahydrobenzo[l,2-c:3,4-c']dithio-

phene.

ACKNOWLEDGEMENTS

I wish to express my sincere appreciation to Professor
Harold Hart for his enthusiasm, encouragement and guidance
throughout the course of this study.

Appreciation is extended to Michigan State University,
National Science Foundation, and National Institutes of Health
for financial support in the form of teaching and research
assistantships.

I also thank The Rotary Foundation of Rotary International
for a graduate fellowship which enabled me to study in this
country.

Finally, I thank my wife Kiyomi, my family and Kiyomi's
family for their patience, support and constant encouragement.

I also thank Kiyomi for her typing of the entire manuscript.

ii

TABLE OF CONTENTS

PART I

SYNTHESIS AND REACTIONS OF SUBSTITUTED
NAPHTHALENE 2,3-EPOXY-I,4-ENDOPEROXIDES

Page
INTRODUCTION ......................... 2
RESULTS AND DISCUSSION .................... 8
l. The Epoxyendoperoxide of l,2,3,4-Tetramethyl-
naphthalene ..................... 8
2. Acid-catalyzed Solvolysis of 255 .......... lO
WM
3. l,2,3,4-Tetramethylnaphthalene-l,4-endoxide
Epoxide ....................... l4
Deuterium Labeling Studies ............. l6
5. Octamethylnaphthalene Epoxyendoperoxide ....... l7
6. Thermolysis of Epoxyendoperoxides .......... 20
7. Reaction of El with HBr ............... 24
8. Reaction of £3 with Bromine ............. 24
9. Reaction of £3 with TFA ............... 28
EXPERIMENTAL ......................... 31
l.- General Procedures ................. 3l
2. Epoxidation of l,2,3,4-Tetramethyl-l,4-epidioxy-l,4-
dihydronaphthalene (£3) ............... 3l
3. Epoxidation of l,2,3,4,5,6,7,8-Octamethyl-l,4-epidioxy-
l,4-dihydronaphthalene (3:) ............. 33
4. Catalytic Hydrogenolysis of 25 and 25 ........ 33
. ’Vb 'VL

TABLE OF CONTENTS (continued)

Page
5. 4-Methoxy-l,4,5,8-tetramethyl-6,7-benzo-2,3-dioxa-
bicyclo[3.2.l]oct-6-en-8-ol (31a) .......... 34
W»
6. 4-Ethoxy-l,4,5,8-tetramethyl—6,7-benzo-2,3-dioxa-
bicyclo[3.2.l]oct-6-en—8-ol ............. 35
7. l-Acetyl-l,2,3-trimethylindan-2,3-diol ($5) ..... 36
8. l-Methylene-2,3-dimethylindan-2,3-diol (£2) ..... 36
' 9. l,2,3,4-Tetramethyl-l,4-endoxy-l,4-dihydronaphthalene
(38) ........................ 37
’VD
lO. l,2,3,4-Tetramethyl-l,4:2,3-diepoxy-l,2,3,4-tetra-
hydronaphthelene (39) ................ 37
'\/\1
ll. Reaction of 39 with TFA in CH Cl -MeOH ....... 38
m 2 2
12. l-Methylene-3-acetyl-2,3-dimethylindan-2-ol ..... 38
T3. Deuterium Labeling Experiments ........... 39
14. 4-Methoxy-l,4,5,8-tetramethyl-6,7-(3',4',5',6'-tetra-
methylbenzo)-2,3-dioxabicyclo[3.2.l]oct-6-en-8-ol
(44) ........................ 4l
’Vh
lS. 2,3,4-Trimethyl-2,3-epoxy-4-hydroxy-l,2,3,4-tetra-
hydronaphthalenone (53) ............... 4l
l6. Trapping of Diradical 52 .............. 42
T7. 2,3,4-Trimethyl-4-hydroxy-l,4-dihydronaphthalenone
(55) ........................ 43
’Vb
l8. Epoxidation of 55 .................. 43
"UL
l9. Hexamethylbenzo[b]furan (57); Thermolysis of 26 . . . 44
’VL
20. Photolysis of 26 .................. 44
Mb
21. Reaction of g; with HBr ............... 45
22. Reaction of 2% with Brz ............... 45
23. Reaction of 62 (or 63) with Br ........... 46
'UL ’VM 2
24. Reaction of 2% with TFA .......... . ..... 47

iv

TABLE OF CONTENTS (continued)

PART II

SYNTHESIS AND PROPERTIES OF
BENZO[l,2-c:3,4-c':5,6-c"]TRITHIOPHENE,
A TRISTHIAHEXARADIALENE

INTRODUCTION .........................

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

1.

no)

Synthesis of Benzo[1,2-c:3,4-c':5,6-c"]trithiophene, a
Tristhiahexaradialene (88) ..............

Physical and Chemical Properties of 88 .......

Protonation of 86 ..................
’U’b

Bromination of 86 ..................
’Vb

Metalation of 88 with Butyllithium .........

Application of the Mild Dehydrogenation Method to
the Synthesis of 88 .................

EXPERIMENTAL .........................

I.
2.

10.

General Procedures .................

1,3.4,6,7,9-Hexahydrobenzo[l,2-c:3,4-c':5,5-c"]-
trithiophene (888) .................

l.3,4,6,7,9-Hexahydrobenzo[l,2-cz3,4—c':5,6-c"]-
trithiophene-2,2,5,5,8,8-hexaoxide (LLé) ......

Benzo[1,2-c:3,4—c':5,6—c"]trithiophene (88) .....

3,3 -Bithiophene (LL8) ...............

Deuteration of 118 .................
’VW

C.T. complex of 86 .................
’Vb

Protonation of 86 ..................
’Vh

Deuteration of 86 ..................
’Vb

Bromination of 86 ..................
”Vb

Page
50
61

61
67
72
75
77

80
82
82

82

82
83

86
86
86

TABLE OF CONTENTS (continued)

11.

12.

13.
14.

15.
I6.

Page
l-Benzo[l,2-c:3,4-c':5,6-c"]trithiophenecarbox-
aldehyde ($88) ................. 87
1-Benzo[l,2-c:3,4-c':5,6—c"]trithiophenecarboxylic
acid (133) ................... 88

W
Reaction of 133 with thionyl chloride ...... 88
3, 4, 6- Tetrahydrobenzo[l, 2- -c: 3, 4-c' ]dithiophene
(iWZ) ...................... 89
Benzo[l,2-c:3,4-c' ]dithiophene (88) ....... 89
Reaction of 88 with sulfur monochloride ..... 90

vi

LIST OF FIGURES
FIGURE Page

1. uv spectra of 88 and triphenylene ............ 68

vii

PART I

SYNTHESIS AND REACTIONS OF SUBSTITUTED
NAPHTHALENE 2,3-EPOXY-1,4-ENDOPEROXIDES

INTRODUCTION

Endoperoxides of polycyclic aromatic compounds have been the
subject of many investigations owing to their synthetic usefulness1
and their biological importance.2 Anthracenes and higher polyarenes
are well known to give endoperoxides with singlet oxygen,3 but only a
few naphthalene endoperoxides have been similarly prepared.

The irradiation of 1,4-dimethoxynaphthalene $8 with a photosensi-

4 Direct photooxidation of 1,4-

tizer gave its 1,4-endoperoxide 68'
dimethoxy-5,8-diphenylnaphthalene,LQ at -50°C resulted in the formation
of endoperoxide 88.5 This peroxide decomposed at room temperature

and much faster at 70°C into 48 and singlet oxygen. The decomposition
was instantaneous at 110°C and was accompanied by a blue glimmer. In

the latter case naphthalene'LQ acted as a photosensitizer of the photo-

oxidation. Photosensitization by‘LQ was demonstrated by its acceleration

 

 

a one B 0"
as on
0M0 a-02 0M.
1 .; :ll 2 l:llzll
b;n:Ph biIZPh

3

of the photooxidation of tetraphenylfulvene to the endoperoxide 8.

I... P11 111 P11 PI

The Rose Bengal sensitized photooxidation of 1,4-dimethylnaphtha-

lene 8 afforded an unstable peroxide 8 which dissociated at room

6

temperature to 8 and singlet oxygen. Irradiation of 8 in ether at

-50°C precipitated an isomeric bis-epoxide 8.

 

Electron-donating groups on the l- and 4-positions are necessary
for the reaction of naphthalenes with singlet oxygen. Naphthalene
itself does not react with singlet oxygen. However, the endoperoxide
of naphthalene was prepared by a different method.7 The reaction of
l,6-imino[lO]annulene 8 with singlet oxygen afforded the l,4-endo-
peroxide 8 which was treated with nitrosyl chloride to give the endo-

peroxide of naphthalene 88 by eliminating N20 from the resulting

ea «2 <26

n""
7 s

 

 

on ;._,

= 303nm

N-nitrosoazidine 8. The peroxide 88 underwent thermal cleavage to
naphthalene and oxygen with a half-life of 303 min at 20°C. No evidence
for the formation of the isomeric 1,223,4-diepoxide was obtained.

Only a few reactions of these naphthalene endoperoxides are

 

 

, R 0"! w R on.
I cacao
A
2 ———-> ——>
002'.
l, a un:' g1

11 12

5
known. When 8 was heated, a small amount of an unsaturated aldehyde

88 was obtained, as well as l and oxygen probably via the intermediate

—0

dioxetane l.8 Acid hydrolysis of 88 gave 5,8-dipheny1-l,4-naphtho-

S

01

quinone 14.

The intermediate hydroperoxide 13 was isolated.
’\/\a ’Vb

II+IVH20

2b____, 1151120

PPWI IJAIH
3 10 4~4ht

 

15

Reaction of 88 with triphenylphosphine gave the partially deoxygen-
ated productll8.7 Reduction of 88 with LiAlH4 gave the unsaturated
cis diol 16.

Ml:

Naphthalene endoperoxides have an isolated 2,3-double bond which

6
could be used for further functionalization of the molecule, and it
was the purpose of this work to explore that possibility. Indeed, a
similar double bond in oxepin endoperoxide,Lz was found to undergo
epoxidation and bromine addition.9 More recently, the reduction of

such strained double bonds by diimide without reducing the peroxide

 

’1)

I O

P’ __°;_,_ /
17
peracid

/ larz

C) C)

O 0/ Br 0/

B

18 ' 19 °

10

bond has been described. For example, 8 was reduced to the cyclic

peroxide 88.

 

7
We were able to epoxidize the 1,4-endoperoxides of 1.2.3.4-

tetramethylnaphthaleneII

and octamethylnaphthalene12 to give extremely
stable epoxyperoxides 88 and 88. These peroxides were handled without

difficulty and melted without explosion, whereas peroxides‘L8 and L8

were reported to explode when heated. The explosion of crude,LZ was
9

also reported.

   

213R1=H1R2=CH3 23; R1: H, R2=CH3 25: R1=H,R2=CH3
22: R1"‘22‘0‘3 24; R1= R2:043. 26: R1? Rz'CH3

It is the purpose of this part of the thesis to describe the

synthesis and chemistry of these compounds.

Results and Discussion

1. The epoxyendoperoxide of l,2,3,4-Tetramethylnaphthalene

 

Oxidation of endoperoxide 88]] with mfchloroperbenzoic acid
(mePBA) gave two isomeric epoxyendoperoxides in a 9:1 ratio.‘3
These isomers were separated after recrystallization from ether.
The major isomer formed colorless rods and the minor isomer formed
colorless cloudy plates which could be separated mechanically
(tweezers). Catalytic hydrogenolysis of each isomer gave an epoxy-
diol. The NMR spectra of the epoxyendoperoxides and diols were

consistent with the symmetry of their structures.14

The peak assign-
ments for the various methyl groups in 88 and 81 were made through

deuterium labeling (vide infra).

 

The major isomer was assigned the syfl_configuration and the
minor isomer the ggti_configuration on the basis of the chemical shifts
of the methyl groups on the epoxide rings, and on dipole moments.15
The upfield chemical shift of the epoxide methyls in the gyg_isomer
(61.18) compared with those of the ascaridole epoxides 8813 can be
attributed to the shielding effect of the aromatic ring. A similar
upfield shift was observed in the dibenzobarrelene epoxide 88.16

The downfield shift of the benzylic methyls in 888 (61.75; gyn_to
the epoxide oxygen) relative to those in,%z§ (6l.37; anti to the

 

epoxide oxygen) add support to the assignment. Finally, the dipole

moments of the major (p=4.2t0.l D) and minor (u=l.7fO.l D) isomers

confirm the assignment.

1.43 1.55 1.21

285 28a 29

10
2. Acid-catalyzed Solvolysis of figs

 

Nhen 888 was treated with trifluoroacetic acid (TFA) in methylene

chloride (0°C, 20 min) in the presence of methanol or ethanol a

single product, assigned structure 88, was obtained in quantitative

 

 

 

a; R=CH3
25s 30 31 b; R-Czl-ls

yield. Although the mass spectrum of,%l did not show an M+ peak (the
highest m/e peak corresponded to M+-HZO), elemental analysis and a
positive test with H1 in acetone suggested that the peroxide group
was still present.

The 1H NMR spectrum of 888 showed singlets for three of the
C-methyl groups (61.10, 1.38, 1.45), but the fourth C-methyl group
was a doublet at 60.75 (J=l.4 Hz). This methyl group was coupled
(verified by decoupling) with the hydroxyl proton which (at 180 MHz)
appeared as a quartet at 65.78 (J=l.4 Hz). This result is consistent
with protonation and ring opening at the epoxide oxygen and not the
peroxide oxygen, for if the latter had occurred one would expect to
find a HOO§CH3 moiety, with the hydroperoxy proton at lower field
than 65.78, and not coupled to the geminal methyl group. The chemical
shift of the hydroperoxy proton of 88, for example, is 67.9 and

coupling with the geminal methyl protons was not observed.‘7 The

11

  

7-9 H00

32

lower than usual chemical shift of the hydroxyl proton in QAR’and
its coupling with the geminal methyl protons can be explained by
its strong hydrogen bonding with an oxygen of either the peroxide
or methoxy group (VD-H 3450 cm-l). The 1H NMR spectrum of 8AR was
similar to that of RAR (except for ethoxy signals in place of the
methoxy singlet at 63.40).

The 13C NMR spectra of 882 and 888 were also consistent with

the assigned structures. The 13C peak assignments for 8A2) shown
on the structure. are based on deuterium labeling and comparison with
888. The unique quaternary carbon signal at 6109.2 ppm supports the

peroxyacetal structure. The chemical shift of the acetal carbon of

8%, for example, is 107.6 ppm from TMS.‘8

 

To prove the structure,,888 was subjected to catalytic hydro-
genolysis. Reaction with hydrogen (Pd/C) in ethanol was complete in
30 min at room temperature, and gave the acetylindandiol'88 as the

sole product (the presumed intermediate hemiacetal 83 was not isolated).

12

it" “ b
Pd/C HO O

 

 

   

The alcohol and carbonyl functionality in 88 were clear from the

infrared spectrum (v0 3440, v 1700 cm-l), and the 1H NMR spectrum
showed the acetyl methyl (62.03) and hydroxyl protons (62.90 and

4.88) as well as singlets at 61.07(3H) and l.47(6H) for the remaining
methyl groups. Europium shift slopes are shown on the structure.

The 13C NMR spectrum of‘88 showed the presence of four different
methyl groups. The structure of 88 was confirmed by independent

synthesis (vide infra).

 

Thermolysis of1888 (refluxing xylene) gave an unsaturated diol

l3

assigned structure 33 in 22% yield. The infrared spectrum of ag
showed bands for the hydroxyl and terminal methylene groups. The
1H NMR spectrum showed two methyl singlets (61.35, 1.42), two methylene
protons (65.27, 5.52) and a broad singlet for the two hydroxyl protons
(62.53).

The formation of peroxyacetal 31 from 233 can be rationalized
by protonation of the epoxide oxygen and ring—Opening with aryl par-
ticipation to give 30. Stereospecific reaction of 30 with the nucleo-
phile results in 3%.]9 Consistent with this mechanism, the 3331_
epoxyendoperoxide 253 was recovered unchanged on treatment with TFA
under conditions identical to those which caused rapid conversion
0f.%§§ to al.20’2] It is perhaps surprising that the peroxyacetal
31 survived the acidic conditions of its preparation. Indeed, treat-
ment of éié with TFA in ethanol did not bring about alkoxy group

exchange nor any other reaction. There is some precedent for the

acid stability of g). The benzylic peroxyacetal 31 is also a stable

cyclic peroxide, produced by the acid-catalyzed methanolysis of

phenanthrene ozonide.22

 

l4

3. l,2,3,4-Tetramethylnaphthalene-l,4eendoxide Epoxide

 

For comparison with 25s, the naphthalene-l.4-endoxide epoxide
32 was prepared and solvolyzed in acid. Epoxidation of naphthalene-
l,4-endoxide 3&23 with m:CPBA gave the epoxyendoxide,3% as a single
isomer. The syn_stereochemistry is assigned on the basis of the
chemical shifts and eurOpium shift slopes of the methyl signals, which
are nearly identical with those of 253, and quite different from those
of 25a.

When 32 was treated with TFA and methanol under the same conditions
as for 25s, it was converted in analogous fashion to 35. If the
methanol was omitted, the product was instead the unsaturated keto-

alcohol 40, whose structure was assigned from spectral data and

 

 

1.17(1o)
W-CPBA
“ “so" 5’" 3‘

91. 8(1.3)

38

 

15

mechanistic considerations. The formation of $5 or 40 from 33 can

be rationalized by protonation at the epoxide oxygen, ring-opening

with aryl participation to give the bridged ion‘sl which, in the

absence of nucleophile loses a proton to give,gg, or in the presence

of methanol is captured to give the acetal 32. Unlike the peroxyacetal
3A, which is stable and easily isolated from the acidic medium, 42

is much more strained and solvolyzes to 35 under the reaction conditions

(Scheme 11, p. 16).

.. 4.;  

* =C03

l/
l.LiA|H4 as 2*
2 .A HE
' 11

21d

Scheme I

 

 

16
4. Deuterium Labeling Studies

Deuterium labeling was used to verify structural and NMR assign-
ments and to be sure that our mechanistic scheme for the acid-catalyzed
solvolytic rearrangements of'gés and 3g was correct. The preparation
of the labeled compounds is outlined in Scheme I (for details, see
the Experimenta1 Section). The expected and observed labeling

consequences are outlined in Scheme II.

Scheme 11

 

17
Labeled tetramethylnaphthalene 21% was converted, via singlet
oxygen and mePBA to gégag which, according to Scheme II should give

35 labeled at the C and C2 methyls. In fact, the peaks at 61.07

1
and 1.47 were diminished in intensity. Since the peak at 61.47

represented two methyl groups with accidental degeneracy, shift reagent

was used, and established that it was the peak with the higher slope

that was diminished in intensity. These results are consistent

with the label being at C1 (61.47, slope 1.6) and C2 (61.07).
Tetramethylfuran labeled at the 02 and C5 methyls was converted

to 39d which, with TFA and methanol, gave 35 with label in the C
W m 3

methyl (61.47, slope 1.0) and acetyl methyl (62.03). Thus the label

results are entirely consistent with the mechanisms put forth in Scheme II.

The 1H NMR assignments in 35 were made as follows. The peak

at 62.03 is clearly the acetyl methyl. The methyl at C1 has an environ-

ment almost identical with that of the C methyl of,%g (61.47 and 1.48

3

respectively). The C3 methyl is similar to that of the 03

36 (61.47 and 1.42). The central methyl in all these compounds comes

methyl in

at somewhat higher field than the benzylic methyls (61.07 inlgé, 1.35
in‘gg and 1.28 in‘40; it appears. at highest field in,g§, where it is

not adjacent to an exocyclic methylene double bond).
5. Octamethylnaphthalene Epoxyendoperoxide
Octamethylnaphthalene-1,4-endoperoxide12 was oxidized with my

CPBA to give the epoxyendoperoxide 26. Only a single isomer, assigned
’Vb

the syn_geometry, was isolated. The chemical shifts of the epoxide

18
methyl groups in 26 (compare with gas) and of the benzylic methyls in

4% (compare with 27g) favor this assignment. Consistent with this

 

stereochemistry, 26 underwent quantitative, rapid solvolytic rearrange:
ment to the peroxyacetal 44 on treatment with TFA in methanol/methy-

lene chloride. The product liberated iodine instantaneously from

TEA
9...
MeOH -C1~12Cl2

 

   

HI in acetone, showed a parent peak at m/e 320 in its mass spectrum,

and had a 1

H NMR spectrum comparable to that of 31% except for the
aromatic methyl substituents.

It is worth noting that in each epoxidation described here (i.e.,
‘23, 24 and 38) the predominant or exclu51ve product was the gyn_1somer
(corresponding to gxg_if the oxygen bridge is replaced by a carbon

taridge). This result may be rationalized as an effect of the oxygen

t>ridge. It is known that the direction of epoxidation can be controlled

19
by coordination of the oxidizing agent with an oxygen atom already
present in the substrate.24 In the present case, coordination with
the peroxide bridge in 23 or 4, or the endoxide bridge in 38 could
account for the observed syn_orientation.
Alternatively, a steric factor may be involved, particularly

with 38. It is well known that most reagents attack bicyclic [2.2.1]
systems from the gxg_face, thus avoiding the crowdedness of the egg9_
face.25 For example, epoxidation of benzonorbornadiene 45 gave only

. 26
the gngox1de 46.

The [2.2.2] alkene dihydrobenzobarrelene 4;,

27
on the other hand, gave equal amounts of 48 and 32. However, the
steric crowdedness of the endo face of,23 and 24 may be more signifi-

cant than usual for [2.2.2] bicyclic systems because of the peroxide

‘ 0

H3”

445 ~ 46

 

bridge. The 0-0 bond, which is shorter than a C-C single bond, tends

to expand the 0-C-C bond angle and contract the CZ-C -C8a bond angle.

1
Consequently either the steric factor or coordination with the oxygen

bridge (or both) may rationalize the syn_epoxidation of 23, £4 and 38.

20

 

38 23(or 24)

6. Thermolysis of EpoxyendOperoxides

The thermal rearrangement of ascaridol 50 to the diepoxide 51
has been known for many years;28 the mechanism involves rate-determin-
ing homolytic 0-0 bond cleavage followed by rapid addition of the

29 For anthracene endoperoxides, two types

oxygens to the double bond.
of thermal reactions are known. 0-0 bond cleavage may give a diepo-
xide, which can be trapped with a dien0phi1e,30 or 0-0 bond cleavage
may occur, with the extrusion of singlet oxygen and reformation of

31

the anthracene. Naphthalene endoperoxides, on the other hand, only

21

extrude oxygen, the rearrangement to diepoxides so far being unknown.32

00 ——A-—»

‘1 1 AA.

000 +'°=
R

 

 

It was of interest to investigate the thermolysis of naphthalene
epoxyendoperoxides such as 25 or £6, for oxygen extrusion could provide
a useful benzoxepin synthesis. In fact, however, decomposition of

£33 in refluxing gfdichlorobenzene gave instead of an oxepin a ketone

33 The compound showed infrared peaks for the

hydroxyl, conjugated carbonyl and epoxide functions, and had an 1H

assigned structure 53.
Mb

NMR spectrum with three methyl singlets and a broad singlet for the
hydroxyl proton. The structure and the stereochemical relationship
between the epoxide ring and the hydroxyl group (cis) was confirmed
by independent synthesis.

2,3-Dimethyl-1,4-naphthoquinone (54) and methyllithium afforded
the ketoalcohol 53 which was oxidized with m¢CPBA to give a product
identical with 5% obtained from 333' The epoxidation of 33 should be
controlled by and occur cis to the hydroxyl function.24

NMR comparisons further support the stereochemical assignment.

22

Treatment of 52 with CH ONa/CH3OD exchanged protons on the B-enone

3
methyl group, and epoxidation gave labeled 53 lacking the methyl
signal at 61.67; furthermore, exchange of the 0H proton of 53 in 020
sharpened the singlet at 61.38. Consequently the methyl signals in
5% are confidently assigned as shown on the structure. The chemical
shift of the C4 methyl (61.38) compares favorably with the benzylic
methyls of 27: (61.37) and not with those of 273 (61.75), consistent
with all cis stereochemistry. Also, the C2 methyl shift in 53 (61.60)
is similar to that recently reported for 56.34

Although the precise mechanism for the formation of 53 is not
clear, the diradical 5% seemed to be one plausible intermediate.

Indeed, when the thermolysis was carried out in the presence of a

 

 

 

 

d iglymo or
P11201011 m-CPBA

 

0fl3li

 

 

56
good hydrogen donor (diglyme solvent) the diol €Z§ was obtained in

quantitative yield. Also, if the thermolysis was carried out in 97
dichlorobenzene to which benzhydrol was added, the products were ZZR
and benzophenone. Thus 23% seems to decompose thermally by 0-0, not
C-O bond cleavage. Trapping of 52 is one of the few examples of
diradical trapping by external chemical means.35
Thermolysis of 26 was not as clean as 25. Analysis of the re-
action mixture by g.1.c. showed at least seven thermolysis products,
and only one of them could be separated and isolated pure. It was
identified as hexamethylbenzofuranlgz (15%). The 1H NMR (180 MHz)

spectrum of 57 showed six singlets, at 62.24, 2.27, 2.30, 2.33, 2.39

h .
a . ' A °' V + unidentlfiod
\\\ products
26. 57
and 2.52. The mass spectrum showed an M+ peak at m/e 202. The uv

spectrum was very similar to that of benzo[b]furan. Photolysis of

26 also gave 51 with other unidentified products. The precise

24
mechanism for the formation of 5% is not clear.

7. Reaction of 27 with HBr

 

Reaction of EZX,With HBr was attempted because the dibromide 38
could serve as a good precursor to tetramethylbenzoxepin. However,
when gas was treated with dry HBr in ether, a naphthalene derivative

considered to be either 59 or 59a was obtained. The 1H NMR spectrum

GEM-’f

had peaks at 62.45(s,6H) and 4.92(s,4H) besides the multiplet in the
aromatic proton region. The mass spectrum showed M+ peaks at m/e
344, 342 and 340. One possible mechanism for the formation of'gg
or 59a is shown in Scheme III.

8. Reaction of 23 with Bromine

 

When 23_was treated with bromine in carbon tetrachloride, the

tribromide 60 was obtained in 11% yield. The 1

H NMR spectrum of 69
showed four singlets at 62.68(3H), 4.72(2H), 4.88(2H),and 4.97(2H)

and multiplets for four aromatic protons. The mass spectrum proved

25

Scheme 111

"m “ m

 

26

ML' 0

   

O

61

 

Br

23

that the product had three bromine atoms. The structure 61 can also
account for these spectral data. However the melting point of’él
(193-194°C) which was reported in the literature36 was different from
observed melting point of the product (211-213°C).

In order to trap some intermediate in the formation of 69, 23
was treated with a half equivalent mole of bromine. From this reaction
at least five products were observed and three of them were separated
by chromatography on silica gel with chloroform as an eluent. They

were identified as dibromide 62 or 63, naphthoquinone 54,37 and
'VL ’Vb 'Vb

 

27
another naphthalene derivative 64. It is obvious that the ethoxyl
group of 63 came from ethanol which was present in the chloroform used
as an eluent.

The 1H NMR spectrum of the dibromide showed four singlets at 62.57
(3H), 2.67(3H), 4.67(2H), and 4.92(2H) and multiplets for the four aroma-
tic protons. The mass spectrum showed the presence of two bromine atoms.
A unique choice between the two alternative structures was not made, but
its intermediacy in the formation of 60 was proved by treating the di-
bromide with bromine to give 60 in a good yield. A possible mechanism
for the formation of 60 and 62 or 63 from 23 is summarized in Scheme IV.
Other products 63 and 54 were probably formed by the reaction of’gg with

protons which were produced during the reaction (see section 9).

Scheme IV

 

+ Bl’ 3' Br

01'

28
9. Reaction of 23 with TFA
“A;

 

A few reactions of alkyl substituted anthracene endoperoxides
with acids are reported. For example, when 9,10-diphenylanthracene
endoperoxide 65 was treated with aqueous sulfuric acid two products
66 and 61 were isolated. Reaction with an anhydrous acid, however,

afforded 68.38

 

However, no reactions of alkyl substituted naphthalene endo-
Peroxides with acids are known. Therefore 23 was treated with TFA.
The reaction products were separated by silica gel chromatography with
Ch Toroform as an eluent. The five products isolated were 62. 119’ 9%.

45,3 and tetramethylnaphthalene a in 9.5, 4.5, n, 1 and 1% yield

res pectively. Structural assignments of these products were made

29

gm Ogu +64+54+21

1 36113

 

23 69

on the basis of their spectral data. Hydroxymethylnaphthalene ZR is

39 The 1H NMR spectrum of 62 showed three singlets

a known compound.
at 62.05(3H), 2.38(3H), 2.58(3H) and multiplets for four aromatic
protons. The mass spectrum showed an M+ peak at m/e 282. The tri-
fluoroacetate functionality was clear from the infrared spectrum
(v -o 1780, vc_ 1230, 1170, 1130 cm-I). The 1H NMR spectrum of 64
showed ethoxyl [61.23(t,3H), 3.62(q,2H)] and allylic methyl protons
(62.38, 2.47, 2.58) as well as a singlet at 64.88(2H) and multiplets
for four aromatic protons. The infrared spectrum showed the ether
'functionality (v _ 1090 cm-l). The mass spectrum showed an M+ peak
ilt m/e 228. These results suggest that 64 and 54 were produced in the
reaction of 3,3 with bromine by the protonation of 213.

The mechanisms for the formation of 63, 33 and 21 are not clear.

A possible mechamism for the formation of 7,0 is shown in Scheme V.

30

Scheme V

10

EXPERIMENTAL

1. General Procedures

1H NMR spectra were measured at 60 MHz on a Varian T-60 or at
180 MHz on a Bruker NH-180 spectrometer against tetramethylsilane

as an internal standard. 13C NMR spectra were determined on a Varian
CFT-20 spectrometer. UV spectra were determined on Unicam SP-800
spectrometer. IR spectra were determined on a Perkin-Elmer Model

167 spectrometer. Mass spectra were obtained with a Hitachi Perkin-
Elmer RMU-6 spectrometer. High resolution mass spectra were obtained
with a Varian CH5 spectrometer. Elemental analyses were performed by

Spang Microanalytical Laboratories, Eagle Harbor, nichigan. Melting

points are uncorrected.

2. Epoxidation of 1,2,3,4-Tetramethyl-l,4-epidioxy-1,4-dihydro-

 

naphthalene (32)

A methylene chloride solution (50 mL) of 85% mfchloroperbenzoic

acid (m-CPBA, 4.4 g) was added dropwise at 0°C to a solution of 2,3”

(l3..97 g, 18.4 mmol) in 20 mL of CH C1 The mixture was stirred at

2 2.
(2"(3 for 18 h during which time mfchlorobenzoic acid precipitated
1WT)!“ solution. The solid was removed by suction filtration and the

ff 7 trate was washed with aqueous sodium sulfite. It was then washed

31

32
with aqueous Na2C03, water and dried (M9504). The solvent was
removed under vacuum at room temperature to give colorless solids
which showed NMR peaks due to the §yn_(gég, 90%) and Eflii.(éém1 10%)
isomers of 1,2,3,4-tetramethyl-2,3-epoxy-l,4-epidioxy-l,2,3,4-tetra-
hydronaphthalene. The solids were recrystallized from ether in a
crystallizing dish covered by Parafilm. Colorless cloudy plates (23%,
3.42 g, 80%) and colorless rods (25%, 0.28 g, 7%) obtained after slow
evaporation of the ether were separated mechanically.

For 255: mp 154-156°C. IR (nujol) 1125(m), 1100(5), 1070(m),
865 cm"(s); ‘H NMR(CDCl3) 61.18(s,6H), l.67(s,6H), 7.23(5,4H); 13c
NMR(CDC13) 11.7, 14.2, 66.3, 79.7, 122.0, 128.3, 140.0 ppm from TMS;
for europium shift data, see structure; mass spectrum, m/e (rel. in-
tensity) 232(1), 216(1), 200(6), 173(100).

Anal, Calcd for C14H1603: C, 72.39; H, 6.94. Found: C, 72.48;
H, 6.94.

For 22:: mp 174-177°C. IR (nujol) 1120(5), 1075(m), 1065(5),
1050(m), 870(m), 845 cm"(m); ‘H NMR (coc13) 61.48(s,6H), 1.73(s,6H),
7.0—7.4(m,4H); 13C NMR (CDC13) 11.1, 15.4, 56.5, 82.7, 120.8, 128.3,
136.0 ppm from TMS; for europium shift data, see structure; mass
spectrum, m/e (rel. intensity) 232(0.5), 216(2), 200(22), 173(100).

Anal. Calcd for C14H1603: C, 72.39; H, 6.94. Found: C, 72.34;

*1. 6.91.

33
3. Epoxidation of 1,2,3,4,5,6,7,8-0ctamethyl-1,4-epidioxy-l,4-
dihydronaphthalene (24)

 

A methylene chloride solution (15 mL) of 85% myCPBA (897 mg)

was added dropwise at 0°C to a solution of 24 in 10 mL of CH Cl

2 2’
The mixture was stirred at 0°C for 24 h. Hork-up in the manner de-
scribed for the preparation of 25 gave a colorless solid which was
nearly pure. Recrystallization from ether-CHCl3 gave pure 1,2,3,4,5,
6,7,8-octamethyl-2,3-epoxy-1,4—epidioxy-1,2,3,4-tetrahydronaphtha1ene
£6, mp ZOO-201°C. IR (nujol) 1120(5), 1100(5), 1080 cm-1(5); 1H
NMR (CDC13) 61.25(s,6H), 1.83(5,6H), 2.22(5,6H), 2.37(s,6H); uv
(cyclohexane) Amax 275 nm (e=420), 225(6840); mass spectrum, m/e
(rel. intensity) 288(6), 256(4), 202(100).

Anal, Calcd for C H 0 ' C, 74.97; H, 8.39. Found: C, 74.89;

18 24 3'
H, 8.32.

4. Catalytic Hydrogenolysis of 25 and 26

 

Medium pressure catalytic hydrogenolysis (20-30 psi of H2) of
gags, 253 and 28 in ethanol with palladium on charcoal (10%, Matheson,
(Soleman and Bell) at room temperature gave the corresponding diols

1&3223’4653 and gg_with 65%, 62% and 82% yields respectively.

For 215: mp 162-163.5°C. IR (nujol) 3450(5), 3400(5), 3320(5),

1 100(5), 1090 cm-](s); ‘H NMR (cums) 51.370551), 1.57(s,6H), 2.27
(£353w.2H), 7.0-7.3(m,2H), 7.3-7.6(m,2H); mass spectrum, m/e (rel. inten-
Sity) 234(4), 200(3), 173(100).

Anal. Calcd for C14H1803: C, 71.77; H, 7.74. Found: C, 71.73;

34

H, 7.76.
For 273: mp 161-162.5°C. IR (nujol) 3200(5), 1100(5), 1065 cm-
(5); ‘H NMR (00013) 61.57(s,6H), 1.75(s,6H), 2.60(s,2H), 7.1-7.6(m,4H).

1

Anal. Calcd for C14H1803: C, 71.77; H,7.74. Found: C, 71.68;

H, 7.75.

For 4%: mp 198-200°C. IR (nujol) 3590(5), 3450(5), 1325(5),
1085(5), 1055 cm"(s); 1H NMR (00013) a1.4s(s,5n), 1.50(s,6H), 2.18
(s,8H,CH3 and 0H), 2.42(s,6H); mass spectrum, m/e(rel. intensity)
290(10), 273(3), 257(6), 230(24), 229(100).

H 0 ° C, 74.44, H, 9.03. Found: C, 74.08;

Anal. Calcd for C 26 3.

18
H. 8.80.

5. 4-Methoxy-l,4,5,8-tetramethyl—6,7-benzo-2,3-dioxabicyclo[3.2.11:
oct-6-en-8-ol (31g2_

 

To a solution of 255 (900 mg) in a mixed solvent of CH2C12 (15
mL) and MeOH (3 mL), 6 mL of trifluoroacetic acid (TFA) was added
dropwise over 5 min at 0°C with stirring. After additional stirring
'for 15 min at 0°C, the reaction mixture was poured into aqueous Na2C03
slowly with vigorous stirring. The organic layer was separated,
vvashed and dried (MgS04). Evaporation of the solvent under vacuum
left a colorless solid which showed NMR peaks due only to 3,13. No
'fi1rther purification than rinsing with petroleum ether was necessary,
mp 98-100°C. IR (nujol) 3450(5), 1035 cm'](s); 1H NMR (CDC13) 60.75
(d .3H,J=1.4 112,08 methyl), 1.10(s,3H,C4 methyl), 1.38(s,3H,C5methyl).
7 — 45(5,3H,C] methyl), 3.40(s,3H, methoxyl), 5.78(bs,1H, hydroxyl),

35
6.7-7.3(m,4H,arom). With a Bruker HH-180 spectrometer, the broad
singlet at 65.78 was resolved to a quartet with J=1.4 Hz. Irradi-
ation of the doublet at 60.75 changed this quartet to a singlet;
13C NMR (CDC13) 10.5, 12.1, 16.8, 19.9, 49.5, 52.0, 82.8, 89.7,
109.2, 122.2, 123.0, 127.9, 128.7, 141.1, 146.8 ppm from TMS; uv

(MeOH) A 272 nm (e=400), 265(420), 258(340), 252(200); mass spect-

max

rum, m/e (rel. intensity) 246(1.5,M+-H20), 133(100).
Anal. Calcd for C15H2004: C, 68.16; H, 7.63. Found: C, 68.27;
11, 7.40.

6. 4-Ethoxy-1,4,5,8-tetramethy1-6,7-benzo-2,3-dioxabicyclo[3.2.l]-

 

oct-6-en-8-ol (316)

 

To a solution of 255 (250 mg) in a mixed solvent of CH2C12 (10
mL) and EtOH (2 mL), 3 mL of TFA was added dropwise over 5 min at
0°C with stirring. Stirring was continued for 15 min at 0°C. Work-
up in the manner described for‘gla gave a colorless oil which was
nearly pure 316 (300 mg). The colorless oil was chromatographed on
silica gel with CHCl3 as an eluent to give pure‘glb. IR (neat) 3460
(s), 1045 cm-1(s); ‘H NMR (coc13) 60.75(d,3H,J=1.4 Hz), 1.10(s,3H),
1.28 and 1.38(t and 5 respectively,6H), 1.45(s,6H), 3.2-4.1(m,2H),
€5.07(bs,1H), 6.9-7.4(m,4H); 13C NMR (00013) 10.5(q), 12.1(q),15.3
(<1), 17.4(q), 19.9(q), 52.0(5), 58.0(t), 82.9(5), 89.7(5), 109.0(5),
722.2(d), 123.0(d), 127.9(d), 128.7(d), 141.3(5), 146.9(5) ppm from
TMS; mass spectrum, m/e (rel. intensity) 260(0.5,M+-H20), 190(13),
7 72 (75), 84(100). Addition of a small quantity of 3,13, or m to a

so 1 ution of hydroiodic acid in acetone liberated iodine.

36
7. 1-Acety1-1,2,3-trimethylindan-2,3-diol (35)

Medium pressure catalytic hydrogenolysis (24 psi of H2) of 31%
(200 mg) in ethanol with palladium on charcoal (10%, Matheson Coleman
and Bell) at room temperature for 30 min gave 35 (110 mg, 62%), mp
105-107°C (petroleum ether). IR (nujo1) 3440(5), 3290(5), 1700 cm-1
(s); 1H NMR (00013) 61.07(5,3H), l.47(s,6H), 2.03(s,3H), 2.90(s,1H),
4.88(5,1H), 7.0-7.4(m,4H); for europium shift data, see the structure;
130 NMR (00013) 18.8, 20.3, 20.9, 29.4, 61.0, 80.4, 86.4, 123.7,
124.4, 128.2, 129.0, 144.2, 144.9 ppm from TMS; mass spectrum, m/e
(rel. intensity) 216(2,M+-H20), 200(6), 191(3), 181(2), 174(51),
173(100).

Anal. Calcd for C14H1803: C, 71.77; H, 7.74. Found: C, 71.80;

H, 7.71.
8. l-Methylene-2,3-dimethylindan-2,3-diol (3&1

Peroxyacetal 31% (400 mg) was heated in gfxylene at reflux for
10 h. Evaporation of the solvent under vacuum left a light brown
residue which was purified by g.1.c. (10% SE30, 165°C) to give a
colorless solid. The solid was recrystallized from petroleum ether
‘to give pure 36 (65 mg, 22%), mp 96-97°C. IR (nujol) 3430(5) 3320
(s), 1790(m), 1660(m), 900 cm-](s); 1H NMR (CDC13) 61.35(s,3H), 1.42
(5,3H), 2.53(b5,2H), 5.27(s,lH), 5.52(s,1H), 7.0-7.6(m,4H); mass
s'»l:>ectrum, m/e (rel. intensity) 190(21), 175(54), 172(59), 156(36),
7 29(100).

Anal. Calcd for 012H1402: C, 75.76; H, 7.42. Found: C, 75.88;

37
H, 7.44.

9. l,2,3,4-Tetramethy1-1,4-endoxy-l,4-dihydronaphtha1ene (38)

 

A mixture of 3,4-dimethy1-2,S-hexanedione23 (12.4 g), pftoluene-
sulfonic acid (catalytic amount) and benzene (80 mL) was heated under
reflux for 6 h during which time the water produced was removed
azeotropically. The volatile solvent was removed under vacuum, and
the brown oily residue was distilled under reduced pressure as de-
scribed in the literature23 to give pure 2,3,4,5-tetramethy1furan
(9.5 g, 77%).

A mixture of tetramethylfuran (3.0 g), benzenediazonium-2-carbo-
xylate hydrochloride (5.3 g), propylene oxide (36 mL) and 1,2-dichloro-
ethane (100 mL) was heated gradually until gas evolution commenced.

The mixture was then heated under reflux for 2 h. The volatile
solvents were removed under vacuum, and the residue was dissolved

in ether and washed with dilute aqueous NaOH, water, and dried (MgS04).
The brown oil which remained after removal of the ether was chromato-
graphed on silica gel with CHCl3 as an eluent to give pure 38 23(2.8 g,
57%).

10. 1,2,3,4-Tetramethyl-l,4:2,3—diepoxy-l,2,3,4-tetrahydronaphthalene

£22).

A solution of 85% myCPBA (487 mg) in 13 mL of CH Cl2 was added

2

dropwise to a solution of 38’ (400 mg) in 2 mL of CH2C12 with stirring

ial‘t: 0°C. Stirring was continued for 5 h during which time

38
mrchlorobenzoic acid precipitated from solution. The solid was removed
by suction filtration and the filtrate was washed with aqueous Na2503.
Na2C03, water and dried (MgS04). The solvent was removed under vacuum
at room temperature to give a colorless oily solid (420 mg) which
showed NMR peaks due only to 33. It was used for further reactions
without purification. IR (nujol) 1160(5), 1130(5), 1085(5), 1070 cm"

(s); ‘H NMR (00013) 61.17(s,6H,C methyls), 1.68(s,6H,c1 4 methyls),

2,3

7.03(5,4H); 130 NMR (c001 ) 9.1, 11.8, 68.5, 84.8, 119.7, 126.6, 149.5

3
ppm from TMS. For europium shift data, see structure.

11. Reaction ofl33 with TFA in CHzClz-MeOH

 

TFA (1.4 mL) was added dropwise to a solution of 33 (200 mg) in
a mixed solvent of methylene chloride (30 mL) and methanol (1.4 mL)
with stirring at 0°C. The mixture was stirred at 0°C for an additional
1.5 h. After work-up as in the preparation of 33, the solvent was
removed under vacuum to give a yellow oil which was chromatographed

on silica gel with CHCl3 as an eluent to give pure 33 (80 mg, 37%).

12. l-Methylene-B-acetyl-2,3-dimethylindan-2-ol (33)

 

TFA (1.4 mL) was added dropwise to a solution of 33 (200 mg) in
30 mL of methylene chloride with stirring at 0°C. After additional
Stirring for 1.5 h and work-up as for 33, the solvent was evaporated
tartder vacuum to give a yellow oil whose NMR spectrum showed peaks
<EILJe to 33 and some small impurities peaks. The yellow oil was

<2 hromatographed on silica gel with CHCl as eluent to give pure 3,9,

3

39

as colorless needles (30%), mp 87-90°C. IR (nujol) 3400(5), 1790(m),
1700(5), 1650(m), 900 cm-1(s); 1H NMR (CDC13) 61.28(s,3H), 1.48(S,3H),
1.82(s,3H), 3.25(s,1H), 5.22(s,1H), 5.45(s,1H), 7.0-7.6(m,4H); mass
spectrum, m/e (rel. intensity) 216(5), 200(3), 198(4), 184(4),183
(4), 173(90), 156(100).

Aggl, Ca1cd for C14H1602: C, 77.75; H, 7.46. Found: C, 77.68;
H, 7.35.

When 33 (420 mg) was allowed to stand in a flask for 10 days,
it decomposed to a brown oil. Chromatography of this oil on silica

gel with CHCl as an eluent gave 33 (178 mg, 42%).

3

13. Deuterium Labeling Experiments

 

A. 1,4-Dimethyl-2,3-dimethyl-d3-naphthalene,333 was prepared
as follows. A mixture of 2,4,5,6,6-pentamethyl-3-methyl-d3-cyclohexa-
2,4-dienone4O (6.01 g), benzenediazonium-2-carboxylate hydrochloride
(6.5 g), propylene oxide (10 mL) and 1,2-dichloroethane (85 mL) was
heated gradually, until gas evolution commenced. After the solution
became clear, the reaction mixture was heated under reflux for 1 h.
The volatile solvents were removed under vacuum. The oily residue
was dissolved in ether and washed with dilute aqueous NaOH, water,
and dried (MgS04). The light brown residue which remained after
the solvent evaporated was recrystallized from methanol to give 1,3,'
4,7,7-pentamethyl-2-methy1-d3-5,6-benzobicyclo[2.2.2]octa-2,5-dien-
8-one (5.23 g). This ketone (2.5 g) was reduced with lithum aluminum
hydride in ether at room temperature in the usual manner to give the

corresponding alcohol (2,3 g). This alcohol, without purification,

40
was heated in g:xy1ene under reflux for 51 h. Evaporation of the
solvent to dryness gave a colorless solid which was recrystallized
from petroleum ether to give pure 333 which showed two methyl singlets
at 62.37 and 2.57 with relative intensities of 0.6:1.0.

Starting with 333 and using the same methods as described above,
333, 333, 33333 and‘333 were prepared. For 33333, the 1H NMR peak
at 61.18 was reduced in area. For 33333, the singlet at 61.48 was
reduced in area. For both‘33333 and'33333, the singlet at 61.57 was
reduced in area. For 33333, the doublet at 60.75 and the singlet
at 61.38 were reduced in their intensities. The 13C NMR peaks at
10.5, 19.9, 52.0 and 82.8 ppm showed diminished intensities. For
333, the singlets at 61.07 and 1.47 had reduced intensities.

B. 3333 was prepared as follows. A mixture of 3,4-dimethyl-
hexan-2,5-dione(14.l g), sodium carbonate (4.5 g), ethanol-d (31 mL),
and 020 (60 mL) was stirred at room temperature for 18 h. The mixture
was poured into 250 mL of methylene chloride and washed with aqueous.

NaCl. Any residual base was removed with solid C0 After drying

2.
(M9504), the solvent was evaporated under vacuum to give the deute-
rated diketone as an oil (12.35 g). The diketone was converted to
deuterated tetramethylfuran by catalysis with prtoluenesulfonic

acid as described for the nondeuterated compound, since the reported
ZnCl2 method23 resulted in deuterium loss. The 2,5-dideuteromethy1-
3,4-dimethy1furan was converted to‘333 and 333 by the same method
used for the unlabeled material. The 1H NMR spectrum of‘333 showed
that the peak at 61.68 (C1,4 methyls) was reduced in area by about
50%.

Starting with 39d, 35d was prepared as with the unlabeled material.
’VVD'VVb

41
The 1H NMR spectrum of 333 showed reduced intensity of the signals
at 61.47 and 2.03. Of the methyls which are accidentally degenerate
at 61.47 , the one which moves least with eurOpium shift reagent was

decreased in intensity.

14. 4-Methoxy-1,4,5,8-tetramethyl-6,7-(3',4',5',6'-tetramethy1-
benzo)-2,3-dioxabicyclo[3.2.1]oct-6-en-8-ol (33)

 

The reaction of 33 (150 mg) with TFA (2 mL) in the same manner
as described for the preparation of 33 afforded 33 quantitatively
as a colorless oil. The product liberated iodine from H1 in acetone,

showed a parent peak in its mass spectrum, and had a 1

H NMR spectrum
comparable to that of 333 except for the aromatic methyl substituents.
The elemental analysis was not obtained. IR (neat) 3470(5), 1120
cm'1(s); 1H NMR (CDC13) 60.77(d,3H,J=l.4 Hz), 1.12(s,3H), 1.57(s,6H),
2.17(s,6H), 2.27(s,3H), 2.31(5,3H), 3.35(s,3H), 5.53(bs,1H); mass

spectrum, m/e (rel. intensity) 320(<l), 203(100).

15. 2,3,4-Trimethyl-2,3-epoxy-4-hydroxy-1,2,3,4-tetrahydronaphtha-
lenone (33)

 

Compound 333 (100 mg) was heated in gfdichlorobenzene (20 mL)
under reflux for 2 h. The solvent was removed under vacuum to give
a brown oily residue whose NMR spectrum showed peaks due to the
starting material and to 33. Separation and purification by chromato-
graphy on silica gel with CHCl3 as the eluent gave two fractions.

The first fraction was starting material (27 mg). From the second

42
fraction, 53 was obtained (18 mg), mp ll3-114°C. IR (nujol) 3565(5),
3460(5), 1680(5), 1100(5), 1080 cm-1(s); 1H NMR (CDC13) 61.38(s,3H),
1.60(s,3H), 1.67(s,3H), 2.47(bs,1h), 7.0-7.8(m,4H); mass spectrum,
m/e (rel. intensity) 218(1), 175(100).
5221: Calcd for C H O ’ C, 71.54; H, 6.47. Found: C, 71.42;

13 14 3'
H, 6.44.

16. Trapping of Diradical 52

 

A. 233 (355 mg) in 8 mL of diglyme was heated at reflux under
nitrogen for 2 h. The 1H NMR spectrum of the diglyme solution after
heating showed peaks due only to 643' Evaporation of the solvents
left an oily brown residue to which 3 mL of petroleum ether was added.
0n standing overnight in a refrigerator, colorless crystals deposited
which were recrystallized from petroleum ether to give pure 27g (220
mg, 61%).

B. A mixture of 25g (100 mg) and benzhydrol (80 mg) was heated
in 10 mL of gfdichlorobenzene at reflux for 7 h. The brown oil which
was obtained after the removal of the solvent under vacuum was chroma-
tographed on silica gel with CHCl3 as the eluent. The first fraction
gave a small amount of starting material. From the second fraction,
benzophenone (17 mg) was obtained, identified by its IR spectrum.

From the third fraction, gas (30 mg) was obtained.

43
17. 2,3,4-Trimethyl-4-hydroxy—1,4-dihydronaphtha1enone (55)

 

To a solution of 2,3-dimethy1-l,4-naphthoquinone37 (558 mg)
in dry ether (100 mL) was added dropwise at -78°C under nitrogen
2.0 mL of commercial 1.77 M methyllithium (ether). The mixture
was stirred for 15 min, then warmed to room temperature. After
additional stirring for l h, water (20 mL) was added with vigorous
stirring. The separated aqueous layer was extracted with ether,
and the combined ether layers were washed with dilute hydrochloric
acid, saturated sodium chloride solution, and dried (M9504).
Removal of the ether under vacuum gave a pale yellow solid which
was rinsed with petroleum ether to give colorless 55 (400 mg, 67%),
mp 163-164°C. IR (nujol) 3420(5), 1635 cm"(s); 1H NMR (coc13)
61.55(s,3H,C

methyl), 1.87(s,lH,0H), l.92(bs,3H,C methyl), 2.13

4 2

(bs,3H,C methyl), 7.2-8.0(m,4H,arom); mass spectrum, m/e (rel.

3
intensity) 202(32), 187(100).

Anal, Calcd for C13H1402: C, 77.20; H, 6.98. Found: C, 76.79;
H, 6.36. '

Treatment of 55 with excess sodium methoxide in methanol-d for

1 h at room temperature40 gave labeled 53 lacking the signal at

52.13.

18. Epoxidation of 56

 

To a solution of 65 (202 mg) in methylene chloride (10 mL) was
added 300 mg of 85% mePBA in 10 mL of the same solvent. The mixture

was stirred for 7 h at room temperature, washed with aqueous sodium

44
sulfite (2x), aqueous sodium carbonate (4x), saturated sodium chloride
(2x) and dried (M9504). Removal of the solvent under vacuum gave a
colorless residue which, after rinsing with petroleum ether (30-60°C)

gave 180 mg (83%) of pure 5% (IR, NMR, mp and mmp identical with the

 

thermolysis product 0f.§%§)- Starting with labeled 55 (vide supra)
the resulting labeled 5% lacked the signal at 61.67 in its NMR spectrum.

19. Hexamethylbenzo[b]furan (57); Thermolysis of 26
W ’Vb

 

Epoxyendoperoxide $6 (288 mg, 1 mmol) in a sealed tube was heated
at 210°C for 2.5 h. After heating, the oily material in the tube was

chromatographed on silica gel with CHCl as an eluent. The first

3
fraction was a mixture of 37 and octamethylnaphthalene 2%. The second
fraction was nearly pure 51. After g.1.c. (20% SE30, 200°C), pure

57 was obtained as a colorless solid (30 mg, 15%). IR (CHC13) 1630

(m), 1100 cm-1(s); 1H NMR (CDC13, at 180 MHz) 62.24(s,3H), 2.27(s,3H),
2.30(s,3H), 2.33(s,3H), 2.39(s,3H), 2.52(s,3H); uv(ethanol) Amax 292
nm(e=l400), 288(1320), 280(1670), 256(14320), 224(13130); mass spectrum,
m/e (rel. intensity) 202(94), 187(100). 177(16); high resolution

mass spectrum, calculated for Cl4Hl80 m/e 202.13568, observed m/e
202.13576.

Thermolysis of 26 in refluxing benzene for 340 h gave similar

results.

20. Photolysis of 26
'VD

 

A solution of 26 (360 mg) in 650 mL of CH3CN was irradiated

45
for 4 h (450 H Hanovia, Corex, room temperature). Benzofuran 57
was the only product which was isolated and identified after separa-

tion and purification by g.1.c. (20% SE30, 200°C), (6 mg, 2.4%).

21. Reaction of 27 with HBr
“A;

 

Dry hydrogen bromide gas was passed through a solution of 27
(40 mg) in 5 mL of dry ether at 0°C until the solution was saturated.
The solution was stirred for 30 min at 0°C, then poured into aqueous
Na2C03 solution slowly with stirring. After the ether layer was
washed with water and dried the solvent was evaporated under reduced
pressure to give colorless solids. Recrystallization of the solids
from ether gave colorless needless of 52 (or 59a) in about 30% yield,
mp 229-230°C. IR (nujol) 1210 cm-](s); 1H NMR (CDC13) 62.45(s,6H),
4.92(s,4H), 7.3-7.6(m,2H), 7.7-8.l(m,2H); mass spectrum, m/e (rel.
intensity) 344(2.5), 342(5), 340(2.5), 263(25), 261(25), 182(100).

The elemental analysis was not obtained.

22. Reaction of 23 with Brz
'Vb

 

A. To a solution of $3 (200 mg) in 20 mL of CCl , Br in CCl

4 2 4
was added drapwise at 0°C until the color of Br2 was not decolorized.
The solution was washed with aqueous Nazszo3 solution, water, and
dried (MgS04). Evaporation of solvent left a brown residue to which
a small amount of CHCl3 was added. Colorless solids precipitated
and the solids were separated and rinsed with ether to give 60 (42

mg, 11%), mp 211-213°c. IR (nujol) 1210(5), 1190(m), 765(5).

46
760 cm-](s); 1H NMR (C0013) 62.68(5,3H), 4.72(s,2H), 4.88(s,2H),
4.97(s,2H), 7.3-7.6(m,2H), 7.8-8.1(m,2H); mass spectrum, m/e (rel.
intensity) 424(2), 422(5), 420(5), 418(2), 343(18), 341(35), 339
(19), 262(21), 260(20), 181(100), 165(71).

8. To a solution of 22 (751 mg) in 20 mL of 0014, 3.28 mL of
bromine solution (0.53 M) in carbon tetrachloride was added drap-
wise at 0°C with stirring. After stirring for 30 min the solvent
was evaporated under vacuum to leave a pale yellow oily residue.

The residue was chromatographed on silica gel with chloroform as the

eluent. The first fraction gave a colorless solid which was recrys-

tallized from ether to give 82 mg of 62 (or 63). The second fraction
gave an unidentified product (15 mg). From the third and the fourth
fractions, small amounts of 54 and 64 respectively were obtained.

For 62 (or 22): mp 161-165°C. IR (nujol) 1210(5), 1190(m),

760 cm"(s); ‘H NMR (coc13) 62.57(s,3H), 2.67(s,3H), 4.67(5,2H),
4.92(s,2H), 7.1-7.7(m,2H), 7.7-8.3(m,2H); mass spectrum, m/e (rel.
intensity) 344(6), 342(12), 340(6), 263(67), 261(68), 182(100).

For 64, see p. 47.

23. Reaction of 62 (or 63) with Br
rm. rim 2

To a solution of 62 (or 63) (17 mg) in-l mL of carbon tetra-
chloride, 0.5 mL of Br2 solution (0.098 M) in CCl4 was added. The

mixture was stirred for 6 h at room temperature and the solvent was

removed under vacuum to give a colorless solid. The solid was rinsed

with ether twice to give pure 60 (15 mg).

47
24. Reaction of 23 with TFA

 

TFA (3 mL) was added dropwise to a solution of 2% (1.0 g) in
20 mL of carbon tetrachloride at 0°C and the mixture was stirred
for 1.5 h at the same temperature. The mixture was poured into
aqueous sodium carbonate solution with vigorous stirring. The
organic layer was washed with water (3x) and dried (Na2504). The
residue obtained by evaporation of the solvent under vacuum was
chromatographed on silica gel with chloroform as the eluent. The
first fraction gave l-trifluoroacetyl-2,3,4-trimethylnaphthalene
(62) which was purified by g.1.c. (20% SE30, 5 feet x 0.25 inch,
227°C), (125 mg, 9.5%). The second fraction gave l-ethoxymethyl-
2,3,4-trimethy1naphthalene (63) which was similarly purified by
g.1.c. (116 mg, 11%). The third fraction gave a mixture of 34 and
21 which was separated to 3 mg of 54 and 3 mg of 21 by g.1.c. as
above. The fourth fraction gave l-hydroxymethyl-2,3,4-trimethyl-
naphthalene (70) which was recrystallized from cyclohexane to give
colorless needles of 72 (42 mg, 4.5%).

For 63: IR (CHC13) 1780(5), 1230(5), 1170(5), 1130 cm-](s);
‘H NMR (coc13) 62.05(s,3H), 2.38(s,3H), 2.58(5,3H), 7.2-7.7(m,3H),
7.8-8.1(m,1H); uv (ethanol) Amax 297 nm (sh,e=300), 288(400), 275
(400), 234(1400); mass spectrum, m/e (rel. intensity) 282(73), 185
(100).

For pg: IR (cuc13) 1090 cm“(s); 1H NMR (coc13) 61.23(t,3H),
2.38(s,3H), 2.47(s,3H), 2.58(s,3H), 3.62(q,2H), 4.88(s,2H), 7.1-
7.5(m,2H), 7.7-8.2(m,2H); mass spectrum, m/e (rel. intensity) 228
(37), 213(11), 184(55) 183(84), 182(100), 169(59).

48
39 .
For 70: mp 176-177°C (lit. l77-177.5°C). IR (nuJol) 3240(5),
’V‘J
1000cm"(s); ‘H NMR (coc13) 61.50(s,lH), 2.37(s,3H), 2.47(s,3H),
2.58(5,3H), 5.07(s,2H), 7.1-7.5(m,2H), 7.7-8.2(m,2H); mass spectrum,

We (rel. intensity) 200(100), 183(45), 182(60), 171(70), 157(60).

PART II

SYNTHESIS AND PROPERTIES OF
BENZO[1,2-c:3,4—c':5,6-c"]TRITHIOPHENE,
A TRISTHIAHEXARADIALENE

INTRODUCTION

Radialenes are a class of compounds possessing n ring atoms

and n double bonds, all of which are exocyclic.4] They have been
42

the subject of extensive investigation with both theoretical and

synthetic interest. Their feature of main interest is the extent

of conjugation available into the ring. In the series 71 - 73,

43 44 ,
[3] - and [4] radialene (71 and 72) have been known for some years.
[5]Radialene 73 is still unknown. Only hexaalkyl derivatives of

A 0'

71 72 74

   

[6]radialene 74 were known45 until the parent hexaradialene was
reported just recently.46
The first synthesized member of the series was [3]radialene,
an isomer of benzene. The initial success of Blomquist in 1959 was
followed by Griffin's synthesis, which improved the yield and made
further characterization of the compound possible.43 Radialene a
Showed marked sensitivity to oxygen which is in agreement with

theoretical predictions of high free valency anticipated at the

50

51
terminal carbons of the exocyclic double bonds.42 The synthesis of

71 is summarized in Scheme 1.

Scheme I
.0" '0"
If 3'. \+ 4.
'7 " —
A A K
«on u /
71
A A
' 1-
K0“ ’
":1"!
I \+
'2—
' on
7 +
“.3 1m
17(— '-
\ A: o
_\_‘,'; 2
/

52

The hexamethyl derivative of a was prepared by j_n_ situ

addition of a vinyl carbene to an allene.47 Although the parent

hydrocarbon a was air sensitive and polymerized rapidly, hexamethyl

[3]radialene ,7}; turned out to be a surprisingly stable crystalline

 

compound.
Ie\ =c/U I0\ 6 c c/Ile
../ \Bf './c- - - \"e

 

75

[4]Radialene which, like cyclobutadiene, possesses four sp -
44

hybridized carbon atoms in the ring, was prepared by similar methods.

Hydrocarbon 712 was found to be remarkably stable to dilute mineral

acids and strong bases. However, like [3]radialene, it was air

sensitive. The radialene gave a mono Diels-Alder adduct with tetra-

Cyanoethylene (TCNE), but a second cycloaddition did not occur even

With a large excess of TCNE.

 

 

 

X=Br,|

54
An octaphenyl derivative of‘zg was reported as a stable crystal-
line photodimer of tetraphenylbutatriene at almost same time as the
parent radialene was reported.48 However it was found later that

the structure of the photodimer is actually 77, not 76.49

n p
”I n

 

I’ll IV '1
Pkc===x j. I n

\
"I "' PI PI:
9% Ph
Ph Ph

77 76

In 1960 the first [6]radialene was reported by Hopff',45 as its
hexamethyl and hexaethyl derivatives. These hexaalkyl hexaradialenes,

prepared as shown below, were stable crystalline compounds.

X

012 or BIZ Mg
X i
" x
Br Br
an I Illiil II:
----—- Br ----a-
Br Br

   

55
The parent hexaradialene Z4 was synthesized by three different
groups just recently.46 Hexaradialene was found to be an air
sensitive colorless solid, reasonably stable in solution if kept

below -40°C. The various syntheses of,z4 are summarized in Scheme II.

Scheme II

Q 15;
A 11le
5‘12 A-soz 7‘ \

of A CH2R2
Rl=R2=Br
50 0f R1=H,R2:Br
2

 

56
In the case of the [4]- and [6]radialenes, one can replace two
hydrogen atoms by one heteroatom and form radialene derivatives in
which every exocyclic double bond is in a heteroaromatic ring (49,

4%). The tetraphenyl derivative of 44 (X=S) was reported as a stable

x=o,s,lll etc.

 

crystalline compound by Garratt.SO Since the parent compound is

still unknown, it is not known whether the stability of 44 is due

Ph Ph Ph Ph

0 l
'2 $5
I » s

o 0

ph Ph Ph Ph

82

to its heteroaromatic rings or its phenyl substituents.

 

Hexaradialene can be viewed as a benzene which is converted
in a formal sense from having six overlapping p orbitals associated
with endocyclic "double bonds" to having those orbitals associated
with exocyclic carbon-carbon "double bonds". In the same sense 44

is an exocyclic benzene and g4 andigg are partially exocyclic

57

X=0,S,NI etc.

benzenes,5] if we exclude nonclassical forms which use d orbitals
of X or dipolar forms of these compounds. For example it is possible
that structures of the type 444 may contribute to the overall

structure of 86 (81, X=S), because in the structure 86a the endocyclic
’Vb’Vb ’VVb

         

/ S , Eggs:
\
S S
85 861

aromaticity of the six-membered ring is restored. Actually a few
examples of such nonclassical condensed thiophenes are known.52
Thieno[3,4-c]thiophene‘44 has a nonclassical structure in which
both sulfur atoms have partial tetracovalent character. Without
sulfur d-orbital participation the molecule must either have the

character of a thiocarbonyl ylid 87a or a diradical 87b.
’VVL Wb

58

SE3 <—>s I S <—> ECG ”etc. \.
/
87 87a

87b

Tetraphenylthieno[3,4-c]thiophene‘44 was the first example of an

53

isolable nonclassical condensed thiophene. It was found to be

 

n H. n Ph I’ll
,/’ ,,’
s s S \‘\
In. in. n: n. P
as 89 so

54
nonpolar and its benzene solution gave no ESR signal. The structure

55

was confirmed by an X-ray analysis. Other similar examples, such as

5
43 6 and 4457 have been reported.
Orthoquinoidal heterocycles of the type 44 show extreme chemical
reactivity, particularly in cycloaddition reactions which restore

58 The most

the endocyclic aromaticity of the six-membered ring.
recent view of their electronic structure suggests an aromatic hetero-

. . . . . . . . 5
cyclic ring With a relatively noninteracting butadiene m01ety 443.

59

:54.

84:

It seemed likely that, if the number of exocyclic double bonds
were increased (as in 44 or 44), the additional heteroaromatic rings
would diminish reactivity and restore overall aromaticity to the
structure.

Two examples of compounds of the type‘44 have been reported.
Benzo[l,2-c:3,4ec']dithiophene 44 (44, X=S) was obtained from 4,4'-
diformyl-3,3'-bithienyl by reaction with hydrazine.60 7H-Pyrrolo-
[3,4-e]isoindole 44 (44, X=NH) was prepared from 44 by 1,2-e1imination
of methanesulfinic acid followed by spontaneous tautomerization of

the product.61

This synthesis was reported after the present work
was well in progress. Both compounds were remarkably stable compared
to their homologues of the type 44. Compounds of the type gfl'were
unknown until the present work. A synthesis of 44 has recently been
proposed but not yet realized.62 It is the purpose of this part of

the thesis to describe the synthesis and properties of benzo[l,2-c:

3,4-c':5,6-c"]trithiophene‘44, the first example of such a molecule.

6O

 

0110
can

 

RESULTS AND DISCUSSION

1. Synthesis of Benzo[1,2-c:3,4-c':5,6-c"Jtrithiophene, a Tristhia-

 

hexaradialene (44)

 

Tristhiahexaradialene 44 is a thiophene analogue of triphenylene.
The syntheses of thiophene analogues of polycyclic arenes reported
in the literature fall into two categories. One involves initial
formation of a thiophene derivative followed by cyclization to a
polycyclic system. The synthesis of benzodithiophene 44 is an example
of this type.

The synthesis of naphthotetrathiophene 44, a thiophene analogue
of dibenzo[g,p]chrysene also falls into this category.63 The diol
‘44, when heated in boiling acetic acid with a catalytic amount of
iodine, underwent pinacol rearrangement to give ketone 44, which
was reduced to the alcohol 44. With 44 a similar rearrangement was
repeated to give 44. The starting diol 44 was prepared by a pinacol
reduction of cyclOpentadithiophenone 444 which is a thiophene analogue
of fluorenone. The synthesis of 444,64 which is summarized in Scheme
III, also belongs to the same category. Thiophene analogues of
fluorenes were prepared similarly.65

However, the application of this straightforward coupling method
to the synthesis of benzotrithiophene 44, as shown in Scheme IV, may

not be fruitful, because treatment of 444 with n-butyllithium,

61

62

 

 

 

63

Scheme III

U + U :*:;i. , "U U

 

 

    

o 0
Br I
954“— Oi ¢ “" W
“2'2 \ DMF S S
100
Scheme IV
s S
8
Br I Q
sue->2: Jami:
r r
8
1m \ s as s

64
then with CuClZ, afforded the thiophene tetramer444.66 Therefore
some complexities can be expected in the reaction of 444 with 3,4-
dibromothiophene under similar conditions.
It is well known that cis-stilbene in the presence of an oxi-
dizing agent undergoes a photochemically induced cyclization to yield

phenanthrene.67 This photochemical cyclization reaction was used to

prepare the thiophene analogue of phenanthrene.68 A similar photo-

Qif/w’fl -—hv—->[cis] A. a] \

'2 s

s /
nv
103 s “5

chemical reaction can be envisioned to prepare 44 from‘444, but the

 

result of irradiation ofl444, in which 444 was obtained instead of

l06,69 suggests that an analogous isomerization would occur.

104 106

 

65
The other strategy for making thi0phene analogues of polycyclic
arenes involves the formation of thiophene rings in the final step.
Benzo[c]thiophenei444 was originally prepared from sulfide 444 by

catalytic dehydrogenation at high temperature.70

@L@%m

107 108 109

Later, Cava reported

that the dehydration of sulfoxide 444 yielded 444 in a better yield.7]
Since benzotrithiophene 44 was expected to be more aromatic than
benzo[c]thiophene 444, dehydrogenation or dehydration of a proper
precursor seemed to be a reasonable route to‘44. Actually an isomeric

benzotrithiophene 444 was prepared from the reaction of tris-sulfide

S
(1 l7 Hut 110
a --—-’
S 0 17 Khur s
s 0
110 111

111. "

 

 

112

66

Ill with chloranil.72
’VVU

The tris-sulfide444 was obtained from the
condensation of three molecules of thiolactone 444 under very high
pressure.

Treatment of hexakis(bromomethyl)benzene44473 with Na25-9H20
in a mixture of ethanol, THF and water at reflux gave (80-90%) the
tris-sulfide,444 as a pale yellow solid which, owing to its low
solubility, was not purified. The NMR spectrum ofl444 in DMSO-d6
showed only a singlet at 64.17. Oxidation of 444 with peracetic

r I-zs |OV

113 114

 

acid at room temperature for l4 h gave the corresponding tris-sulfone
.444 whose NMR spectrum in DMSO-d6 showed a singlet at 64.55. The
mass spectrum of the sulfone showed a molecular ion peak at m/e 348
and fragment peaks at m/e 284 (M+-SOZ), 220 (M+-2502), 156 (M+-3SOZ)
and 64 (502+). The mass spectrum suggests that 444 may serve, under
appropriate conditions, as a precursor of hexaradialene 44. This
prediction was recently verified by Boekelheide and his coworkers.46

Dehydrogenation of tris-sulfide,444 with either DDQ or gfchlora-
nil in refluxing chlorobenzene afforded benzotrithiophene 8

colorless needles. The NMR spectrum of 44 showed a singlet at 67.52

in CDCl3. somewhat downfield from that of the C-2 proton in thiophene

 

an
SDI

The

67

DD
114 0 lb 8

or o-chlorauil

 

 

(67.l9).74 This downfield shift is possibly due to deshielding by
the adjacent thiophene rings. A similar downfield shift of thiophene
ring protons was observed for naphthotetrathiophene 44 (68.77, type
H-l; 68.42, type H-3).63 The 13C NMR spectrum of 44 consisted

of two peaks, at 6ll7.4 and l32.4 corresponding to C-l—and C-3a-type
carbons respectively. These shifts may be compared with those of
thiophene (C-2,l24.9;C-3,l26.7)75 and triphenylene (C-l,l23.7;C-2,
l27.6;C-4a,l30.2).76 A large proton chemical shift difference (0.9
ppm) between 44 and 44, and the upfield shift of C-l-type carbons of

44 compared to those of thiophene will be discussed later.

2. Physical and Chemical Properties of 44

 

Benzotrithiophene 44 is a stable crystalline compound. Its
mass spectrum showed a molecular ion peak at m/e 245.964l6 (calcu-

lated for C12H653 245.963l8) as a base peak. The IR spectrum showed

an absorption at 3100 cm'1 (heteroaromatic C-H). The ultraviolet

spectrum of 44 is similar to that of triphenylene (Figure l).77

The longest wavelength absorption (320 nm) occurs at appreciably

log E

 

 

 

-
p-
.-
D
p
_
b
..
.-
.-
—
b
y-
b
b

250 300

Anm
Figure l. uv spectra of 44 (—) and triphenylene (---).

69
lower energy than that of the tris-benzo[b]isomer,444 (286 nm).72
A similar but even larger difference is observed for benzo[c]thiophene
(343 nm)70 and benzo[thhiophene (298 nm).78

The 1H NMR spectrum of 44 was compared with that of 44 (2192.
§gg§g). The large downfield shift of the C-3 proton in 44 from 44

is possibly due to the steric effect of neighboring thiophene rings.

The buttressing effect between C-l and C-lZ protons would cause the

7.52

   

compression of C-3 and C-4 protons which would shift these protons
downfield.

In order to examine the degree of this steric effect, the chemical
shifts of 44 were compared with those ofl444. The difference between
the C-3 protons in 44 and 444 (0.51-0.64 ppm) can be attributed to

the buttressing effect of the C-l and C-l2 protons in 44.

 

70
The 13C NMR spectrum of 44$howed an upfield shift of the C-l-

type carbons (vide supra). The shift can be explained by the B-

 

effect of C-9a-type carbons. For the comparison, 3,3'-bithiophene
.444 and its dideuterated derivative,4444 were prepared and their
130 NMR spectra were measured. The position of deuteration of‘444
with deuterated fluorosulfonic acid was confirmed by studying the
1H NMR spectra of‘444 andl4444. The 1H NMR spectrum of 444 (DMSO-

d6) at 180 MHz showed three sets of doublets of doublets centered

 

 

8
mg
a
n
S
"7 mu
at 67.52 (Jab=4.9 Hz, Jac=l.3 Hz), 7.59 (Jab=4.9 Hz, ch=2.9 Hz) and

7.77 (Jac=l.3 Hz, J =2.9 Hz) due to Ha’ H and Hc respectively.

bc b
The assignment of protons was made based on a comparison of the
coupling constants with those of thiophene (J2,3=4.7 Hz, J2’5=2.9 Hz,
J2,4=l.0 Hz).79 The NMR spectrum of 4444 showed that the peaks
centered at 67.77 were reduced in area, thus proving that deuterium
exchange occurred exclusively at the C-2 position.80

Comparison of the 13C NMR spectrum of‘444 andl4444 made it

possible to assign the high field peak at 6ll9.7 to C-2-type carbons,

71
because it was the only peak which was reduced in intensity in 4444
compared to‘444. The remaining peaks were at 6126.0, 126.3 and
137.2. This proved that the connection of two thiOphene rings at the
0-3 positions shifts the C-2-type carbons upfield probably because
of the B-effect of the C-3 carbon in the other thiophene ring.

Thus the downfield shift of protons and the upfield shift of
C-l-type carbons of 44 compared to those of thiophene were caused
by the connection of thiophene rings. However no evidence that
these shifts were caused by conjugation around the central six
membered ring was obtained.

In contrast to 44, benzotrithiophene 44 did not readily add
dien0philes (j;g;, dimethyl acetylenedicarboxylate, benzyne, singlet
oxygen). This is probably because the aromaticity of the product
from 44 is greater than that of the product from 44. Cycloaddition
of 44 with a dienophile gives a benzodithiophene derivative 444
while 44 with a dien0phi1e gives a benzene derivative. However, the
unreactivity of 44 toward dien0philes may also indicate that the
overall aromaticity is enhanced as the number of exocyclic double

bonds is increased froml44 to‘v4.

-------- -----*

 

72
Compound 44 does form intensely colored, air- and moisture-
stable crystalline l:l charge-transfer (CT) complexes with TCNE, DDQ,
TCNQ (all dark blue, CT band near 610 nm), and chloranil (dark red,

CT band at 527 nm). The CT bands occurred at about 50 nm longer

81

wavelength than for those of triphenylene. Oxidation of‘44 with

SbCl5 in methylene chloride gave a rather stable blue radical cation.82

3. Protonation of 86
’VL

 

It is known that electrophilic substitution of thiophene occurs
preferentially at the 2-position.83 The NMR observation of protonated
thiophenes showed that protonation occurred exclusively at the 2-

position.84

In benzotrithiophene, two types of protonation are possible:
(a) at the sulfur atom; (b) at the C-l-type carbon. For each type of
protonation, mono-, di-, and trications are possible. Every cation
that can be formed from 44 by protonation at either carbon or sulfur
but not both is summarized in Scheme V.

In FSO3H-SOZC1F at -20°C to -78°C compound 44 gave a single species.
The ‘H NMR spectrum (-20°C) had peaks at 65.90(d,4H,J=1.5 Hz), 9.34
(s,2H) and ll.45(t,2H,J=l.5 Hz). The doublet at 65.90 became a sharp
singlet by irradiation of the triplet at 611.45. For a possible
structure of this species, all structures but‘444 and'4444 were ruled
out on symnetry grounds. The doublet, singlet, and triplet were
assigned to Ha’ Hc, and Hb respectively (p. 74). The chemical shifts
of Ha and Hb agree well with those obtained on protonation of thio-

phene with HF-BF at C-2 (65.40(H2) and ll.27(H5)).

3

Scheme V

 

 

123 124

74

H ‘+
F8231! :
119 5

 

1203 Ha

With F5030, exchange was rapid at Ha (65.90 peak nearly absent
in 5 min, 611.45 peak a singlet; quench gave 44442) but extensive at
all positions after 8 h (quench gave mainly 44445 and 86-d6).
Although the monoprotonated speciesl444 could not be detected, rapid
exchange of two protons and slow exchange of the remaining protons
can best rationalized by assuming a rapid equilibrium between 444
or 4444 andl444 but slow deprotonation ofl444 in the strongly acidic
medium.

A unique choice between the diprotonated species was not
possible. Neither the "meta" dication 120b nor the trication 121

WVD ’VVb
was observed. It seems plausible to assume that monocationt444 can
only be protonated to give the "para" dicationl444, in which the
largest charge separation can be attained, and thatl444 is not further

protonated to give trication‘444 because two positive charges would"

be located "ortho" each other. However the "ortho" dication 4444

75

cannot be ruled out completely. One might argue that,444 is proton-
ated to give14444 instead of 444 to avoid steric repulsion between
the two adjacent sets of methylene protons, and the same reason could
explain the absence of 1 1.

WM

4. Bromination of 86
Mb

 

Treatment of 44 with one equivalent of Br in CCl4 at room tem-

2
perature gave a monobromo derivative contaminated with a small amount
of a dibromide. The 1H NMR spectrum of the monobromide 444 showed

a doublet at 68.37 due to Ha’ and a multiplet at 67.47 due to the
remaining protons.

Similarly di- and tribromo derivatives were obtained, depending

Brz Bf
86 ’

-HBr Ha

 

125

on the Br2/44 mole ratio. Isomer separation was difficult, but NMR
analysis showed that three dibromo and two tribromo compounds were
formed, these being the isomers possible without having two sterically
interfering bromines in the same ''wedge" of the structure. These
possibilities are summarized in Scheme VI.

The isomer ratios were statistical. For example, treatment

76

     

Scheme VI
5 .
Br 8
125
two routes
Br 5 S
T
S
B T
125 127 128
two
routes /two 4
routes
3! 5
Br
Bf S

 

129 130

77
of 44 with three equivalents of Br2 afforded a mixture of tribromo
derivatives. The 1H NMR spectrum of the mixture at 60 MHz showed
two singlets with equal areas, but at 180 MHz there were two singlets
at 68.57 and 8.58, and two sets of doublets at 68.62 and 8.65, with
equal areas for each signal. The only way to explain the splitting
pattern and the integration of the spectrum is that both doublets
and one of the singlets are due to 444 and the other singlet is due
to the isomer‘444 with a three to one ratio of'444 tol444. This
three to one ratio agree well with the statistical ratio for the

formation of 129 and 130, as shown in Scheme VI.
mm

5. Metalation of 44 with Butyllithium

 

Metalation of‘44 was also possible. For example, treatment

with excess butyllithium in ether at —25°C followed by addition of

n-Buli li+ DMF 0110

85———> ___.

 

 

131 132

 

78

dimethylformamide gave the yellow aldehyde,444. The product was
proved to be the monoaldehyde by its 1H NMR and maSs spectra. The
NMR spectrum consisted of two singlets at 67.97 (lH,C-3 proton) and
10.l3(lH, aldehyde), one doublet at 68.72(1H,J=2.5 Hz,C-9 proton),
and a multiplet at 67.58(3H,C-4,-6,-7 protons). The IR spectrum
showed a conjugated carbonyl absorption at 1645 cm-1.

It was of interest to investigate the metalation of 44, because
the carboxylic acid‘444 which can be obtained from 444 with C02 could

serve as a useful starting material to prepare the interesting compound

444 in which the remaining "wedges" in 44 are closed.

Scheme VII

133 ---->

 

79

Treatment of 44 with butyllithium followed by passing C0 gas

2
though the reaction mixture afforded the monocarboxylic acid,444.
Because of the low solubility ofl444, its NMR spectrum could not be
measured. However the mass spectrum showed a molecular ion peak at
m/e 290 and the IR spectrum showed chracteristic absorptions for
a carboxylic acid (3600-2400 and 1670 cm-l).

The preparation of the acid chloride of 444 was not successful.
The reaction of‘444 with thionyl chloride afforded a yellow, unidenti-
fied product. The NMR spectrum could not be measured owing to low
solubility. The IR spectrum showed a carbonyl absorption (1720 cm")
and a heteroaromatic C-H absorption (3100 cm'1). The absorption of
the carboxylic acid group disappeared completely. The mass spectrum
(m/e 386) and the IR spectrum (1720 cm") were not suitable for either
the acid chloride or the conjugated cyclic ketone 444.

The reaction of 86 with sulfur monochloride and aluminum chloride

to give the similar compound 444 was also unsuccessful.

5 Cl
86 2 2

 

AICI3

 

80

6. Application of the Mild Dehydrogenation Method to the Synthesis

of 44

 

The same mild dehydrogenation used to synthesize‘44 was success-
ful for the preparation of benzodithiophene 44. Treatment of,444‘
with 000 afforded‘44 in 35% yield. The precursorl444 was prepared
in two steps (46%) from prehnitene by an improved modification of the

85

literature procedures. The overall yield (three steps) of the

previous synthesis of 44 was 4.2% from 3,4-dibromothiophene, which

 

137

 

itself must be synthesized from thiophene (two steps, 42%).60

This simple mild dehydrogenation was also successful for the

synthesis of naphthodithiophene,444 from444.86

 

 

81
In conclusion, benzo[l,2-c:3,4—c':5,6-c"]trithiophene, a tris-
thiahexaradialene, is a new type of heteroaromatic compound. It
is clear that, although exocyclic benzenes of type g3 are highly
reactive toward cycloaddition, those of type Q3 and‘gl are much

more like aromatic compounds in their reactivity.

EXPERIMENTAL

1. General Procedures (see Part I of Thesis).

 

2. l,3,4,6,7,9-Hexahydrobenzo[l,2—c:3,4-c':5,6—c"Jtrithiophene

(M)

 

2S-9H20 in 40 mL of water was diluted

with 800 mL of ethanol and 400 mL of THF. Then hexakis(bromomethyl)

A solution of 15 g of Na

benzene,LL% (6.4 g) was added to the solution and the mixture was
heated at reflux for 19 h with vigorous stirring. The mixture was
cooled to room temperature and the resulting light yellow precipitate
was separated by suction filtration. The solids were washed well
with water, then with ethanol, and dried under vacuum. The crude

ng (2.4 g) thus obtained was subjected to further reaction without
purification. However the purification of,LL4 was possible by subli-
mation under reduced pressure, mp 250-256°C dec. IR (nujol) 1195(m),

l

875(m), 725 cm'](s); H NMR (DMSO-d6) 64.l7(s); high resolution mass

spectrum, calculated for C m/e 252.01013, observed m/e 262.01065.

12H1253

3. l,3,4,6,7,9-Hexahydrobenzo£l,2-c:3,4-c':5,6-c"]trithiophene

 

2,2,5,S,8,8-hexaoxide (JJ§)

 

A mixture of llé (2.0 g) and 50 mL of peracetic acid was stirred

82

83
for 14 h at room temperature. The colorless precipitate was
separated by centrifugation. The solid was repeatedly washed with
water and separated (centrifuge) until the peracetic acid was removed,
then dried under vacuum to give crude L15 (1.5 9), mp > 320°C.
IR (nujol) 1315(5), 1145(5), 1125(5), 1105 cm-1(s); 1H NMR (DMSO-
d6) 64.55(s); mass spectrum, m/e (rel. intensity) 348(M+, trivial),

4.
284(M -SOZ,25), 220(M+-250 .42), 156(M+-3502,85), 64(SOZ+,100); high ,

2
resolution mass spectrum, calculated for C12H12S306 m/e 347.97961,

observed m/e 347.98022.

4. Benzo[1,2-c:3,4-c':5,6-c"]trithiophene (@Q)

 

To a solution of gfchloranil (1.6 g) in 250 mL of chlorobenzene,

0.5 g of,114 was added and the mixture was heated at reflux for 4 h
with vigorous stirring. After the solution was cooled to room temper-
ature, the brown precipitate was removed and the solvent was evaporated
from the filtrate under vacuum. The brown residue was dissolved in
benzene and chromatographed on basic alumina with benzene as an eluent.
Colorless needles of 06 (225 mg, 46%) were obtained from the first
fraction, mp 236-238°C. IR (nujol) 3100(m), 870(m), 855(5), 850(5),
780 cm"(s); ‘H NMR (coc13) 67.52(s); 13c NMR (coc13) 117.4, 132.4

ppm from TMS; uv (EtOH) A 320 nm (e=6210), 306(8730), 294(6420),

max
277(sh,10050), 265(Sh,19680), 256(59250), 248(50840); mass Spectrum
m/e (rel. intensity) 246(100), 201(13), 170(12), 123(13); high resolu-
tion mass spectrum, calculated for c12H6S3 m/e 245.96318, observed

m/e 245.96416. 000 (1.35 9) could be used in place of 9:ch10rani1 to

give $6 in a similar yield.

84
5. 3,3'-Bithiophene (115)

 

To a solution of 3-bromothiophene (5.0 g) in 50 mL dry ether,
17 mL of 2.2 M n—BuLi in hexane was added at -78°C under nitrogen
atmosphere. The mixture was stirred for 15 min. At the same tem-
perature dry Cu012 (5.2 g) was added to the solution, and the mixture
was stirred for an additional 2 h. After the mixture was warmed
to room temperature, it was stirred over night. Then the mixture
was cooled to 0°C and 50 mL of 4N HCl was added with vigorous stir-
ring. The ether layer was separated and washed with 4N HCl and

water. After it was dried over MgSO4 the solvent was removed to

give light brown solids. The solids were purified on a short

alumina column with CHCl3 as the eluent to 9iV¢.LL§ (1.6 9). mp

1

87
128-130°C (lit. 132°C). H NMR (DMSO-dfi) 67.52(dd,2H,Jab=4.9 Hz,

Jac=1.3 Hz), 7.59(dd,2H,Ja =4.9 Hz,J c=2.9 Hz), 7.77(dd,2H,Jac=

b b

1.3 Hz,J c=2.9 Hz); 13c NMR (coc13) 119.7, 126.0, 126.3, 137.2 ppm

b
from TMS,

6. Deuteration of,LL§

 

In a 50-mL round-bottomed flask was placed 400 mg 0ftLL§2 and

about 20 mL of $0261F was condensed in the flask at -78°C under a

N2 atmosphere. Then approximately 1 mL of DSO3F was added and the

mixture was stirred vigorously to give an orange solution of carbonium

ion. The stirring was continued for 5 h at -78°C under a N2 atmos-

phere. The solution was poured into aqueous NaZCO with vigorous

3
stirring. The mixture was extracted with ether and the ether layer

__W!

85
was washed with water and dried (MgSO4). Evaporation of the ether
gave‘LL8882 as a colorless solid (328 mg). IR (nujol) 2310 cm-](m,

heteroaromatic C-D).

7. C.T. complex of 88

 

To a solution of 88 (100 mg) in 20 mL of benzene, 52 mg of TCNE
in 10 mL of benzene was added with stirring. Evaporation of the sol-
vent left a dark blue solid. The solid was recrystallized from CHCl3
to give dark blue rods (120 mg), mp 183-190°C (dec.). When the
solid was heated to about 160°C it changed color from blue to
white and started melting at 183°C with the color changing to blue
again. The C.T. complex of 88 with TCNE thus obtained showed IR and
mass spectra corresponding to a mixture of 88 and TCNE. 1H NMR (CDC13)
67.53(s); C.T. band (CHC13) Amax 608 nm (5:90).

Similarly C.T. complexes of 88 with 000, TCNQ and chloranil
were prepared. The complex with 000 melted at 255-260°C with decom-
position. When it was heated to about 200°C, the color was changed
from blue-black to brown. The complex with TCNQ melted at 239-240°C.
It showed a Cgll.band at.612 nm (€=190). The complex with chloranil
melted at l95-200°C with decomposition. When it was heated to about

115°C, the color was changed from red to brown. The complex showed

a C.T.band at527 nm (60).

86

8. Protonation of 88

 

About 30 mg of 88 and a small amount of tetramethylammonium
tetrafluoroborate were placed in a 10 mm NMR tube into which about
2.5 mL SOZCTF was condensed at -78°C. To the mixture, 1 mL of FSO3H
was added and the contents were mixed using a "super mixer" (Matheson
Scientific, cat. No. 601-0005) to give a dark red solution. The
1H NMR spectrum of the carbonium ion thus prepared was obtained at
180 MHz using (CH3)4NBF4 (63.13) as an internal standard. For a

description of the spectrum, see text.

9. Deuteration of 86
’VD

 

Deuterated 88 was obtained by the same method which was used
for deuteration oflLL8. The degree of deuteration was checked by
mass spectrometry. When the carbonium ion solution was quenched
5 min after addition of acid, 88%82 was obtained. Benzotrithiophene
recovered after 8 h was mainly penta- and hexadeuterated. IR (nujol)

2300 cm’](m, heteroaromatic C-D).

10. Bromination of 88

 

To a solution of 88 (50 mg) in 8 mL of CCl4 three equivalents
of Br2 in CCl4 were added dropwise at room temperature with stirring.
The mixture was stirred for 30 min and the solvent was removed under
vacuum to give colorless solids (82 mg). The solids were recrystal-

lized from CHCl3 to give a mixture of tribromide 888 andIL88 in a

87
3 to 1 ratio. IR (nujol) 3110 cm-1(w, heteroaromatic C-H); 1H NMR
(CDC13) 58.57(s,1H), 8.58(s,1H), 8.62(d,J=2.9 Hz,1H), 8.65(d,J=2.9 Hz,
1H); mass spectrum, m/e (rel. intensity) 486(43), 484(100), 482(94),
480(32), 405(18), 403(27), 401(13), 324(21), 322(18), 243(12); high

resolution mass spectrum, calculated for C12H3Br353 m/e 479.69467,

observed m/e 479.69275.
Similarly monobromide and a mixture of dibromides were obtained

by changing the amount of Br However the contamination by other

2.
bromides could not be avoided. The separation of isomers was not

achieved.

For 888: mp 110°C (dec.); IR (nujol) 3100(m), 870(m), 855(m),

1

780(5), 735 cm"(m); H NMR (coc1 67.47(m,4H), 8.37(d,J=3.0 Hz);

3)

high resolution mass spectrum, calculated for chHSBrSB m/e 323.87369,

observed m/e 323.87302.

1

For dibromides : H NMR (coc13) 67.45(m,2H), 8.40(m,2H); high

resolution mass spectrum, calculated for C12H4Br253 m/e 401.78421,

observed m/e 401.78348.

ll. 1-Benzo[l,2-c:3,4-c':5,6-c"]trithiophenecarboxaldehyde (L88)

 

To a solution of 88 (100 mg) in 50 mL of dry ether, about 6.6
mmol of n-BuLi in hexane (3 mL) was added at -25°C under a N2 atmos-
phere with stirring. The mixture was stirred for l h at the same
temperature, then DMF (1.0 mL) was added. Stirring was continued
for an additional 2 h at -25°C, after which the mixture was warmed
to room temperature. About 10 mL of water was added with vigorous

stirring. The ether layer was separated and the aqueous layer was

88
extracted with ether. The combined ether layers were washed with
dilute hydrochloric acid and water, then dried (M9504). Evaporation
of the ether gave a yellow solid which was purified by preparative
thin layer chromatography (silica gel) with CHCl3 as a developing
solvent to give pure 888 (69 mg), mp 170-173°C. IR (nujol) 3100(m),

2720(5), 1645 cm’](s); ‘H NMR (coc1 67.58(m,3H), 7.97(s,lH), 8.72

3)
(d,J=2.5 Hz,lH), 10.13(s,lH); mass spectrum, m/e (rel. intensity)
274(51), 246(100), 201(35).

0. .uhn— ' fiY~
-nEL.EJH J

12. 1-Benzo[l,2-c:3,4-c':5,6-c"]trithiophenecarboxylic acid (L88)

 

To a solution of 88 (100 mg) in 50 mL dry ether, about 3.3 mmol
of n-BuLi in hexane (1.5 mL) was added at -25°C under a N2 atmosphere
with stirring. The mixture was stirred for l h at the same temperature.

Then dry C0 gas was passed through the solution for 5 min. The

2
solution was warmed to room temperature, 10 mL of water was added

and the mixture was stirred vigorously for 2 h. The aqueous layer
was separated and acidified with concentrated hydrochloric acid to
give a yellow precipitate. The precipitate was collected by suction
filtration, washed with water and dried under vacuum to give crude
$33 (113 mg), mp 230-235°c (dec.). IR (nujol) 3600-2400(5), 1670 cm“

(5); mass spectrum, m/e (rel. intensity) 290(3), 246(100), 201(6).

13. Reaction of 133 with thionyl chloride
’VW

 

The crude acid L88 (100 mg) was heated in 3 mL of SOCl2 at reflux

for 6 h with vigorous stirring. Distillation of the excess thionyl

89
chloride under reduced pressure left a yellow solid (82 mg). The
difficult solubility properties of the solid prevented the measure-
ment of its NMR spectrum. Its IR spectrum in nujol showed the
absence of the carboxylic group. IR (nujol) 3100(m, heteroaromatic
C-H), 1720 cm-1(s); mass spectrum, m/e (rel. intensity) 386(21),
384(18), 352(24), 350(45), 342(18), 340(18), 316(100), 308(23), 306
(44), 272(95), 244(44).

l4. l,3,4,6-Tetrahydrobenzo[l,2-c:3,4-c']dithiophene (L31)

 

l,2,3,4-Tetrakis(bromomethyl)benzene was prepared from prehnitene

88

by the literature method. This tetrabromide (13.0 g) was added

to a mixture of Na25-9H20 (20.8 g), EtOH (350 mL), and water (30

mL). The mixture was heated under reflux for 16 h with vigorous
stirring, then cooled to room temperature. About 400 mL of water

was added and the mixture was cooled in an ice bath. The light brown
precipitate which formed was collected by suction filtration, washed
with water several times and with EtOH twice and dried under vacuum

85 l

to give‘ng (4.5 9). mp 110-115°C (lit. mp 115-117°C). H NMR

(CDC13) 64.12(bs,4H), 4.20(bs,4H), 7.00(s,2H).

15. Benzo[1,2-c:3,4-c']dithiophene (31)

 

Treatment of 881 (0.5 g) with 000 (1.3 g) as described in the

60
preparation of 88 afforded 88 (0.17 g, 35%), mp lll-112°C (lit.

mp 112-113°C).

90

16. Reaction of 88with sulfur monochloride

 

To a solution of 88 (246 mg) in 20 mL of carbon disulfide,
440 mg of AlCl3 and 405 mg of S2C12 were added at 0°C. The mixture
was stirred for 3 h at 0°C and for 17 h at room temperature. Then
it was poured onto cracked ice. The reddish brown precipitate formed
was collected by suction filtration and washed with water then with
methanol and dried under vacuum (357 mg), mp > 350°C. The solid
was not soluble in either water or organic solvents, and could not

be identified. The carbon disulfide layer was separated from aqueous

layer and evaporation of the solvent gave sulfur (65 mg).

BIBLIOGRAPHY

10.

11.

12.
13.

14.

BIBLIOGRAPHY

W. Adam, Angew. Chem., Int. Ed. Engl.,'L8, 619 (1974).

 

J.E. Baldwin, H.H. Basson, and H. Krauss, jun., Chem. Commun.,
984 (1968); M.K. Logani, W.A. Austin, and R.E. Davies, Tetra—
hedron Lett., 511 (1978).

 

 

K. Gollnick and 6.0. Schenck, “Oxygen as a Dienophile" in "1,4-
Cycloaddition Reactions", ed. J. Hamer, Academic Press, New York,
1967, p. 299; J. Rigaudy, Pure Appl. Chem., 88, 169 (1968); J.
Rigaudy and D. Sparfel, Bull. Soc. Chim. Fr., 3441 (1972); J.
Rigaudy, M.C. Perlat, and N.K. Cuong, BUll. Soc. Chim. Fr., 2521
(1974); J. Rigaudy, C. Breliere, and P. Scribe, Tetrahedron Lett.,
687 (1978). .

 

 

 

 

J. Rigaudy, C. Deletang, and J.J. Basselier, C. R. Acad. Sci.,
Paris, Ser. C, 888, 344 (1969).

 

 

(1966

J. Rigaudy, C. Deletang, and J.J. Basselier, ibid., 888, 1435

J. Rigaudy, D. Maurette, and N.K. Cuong, ibid., 888, 1553 (1971).

M. Schfifer-Ridder, U. Brocker, and E. Vogel, Angew. Chem., Int.
Ed. Engl., 88, 228 (1976).

J. Rigaudy, C. Deltang, D. Sparfel, and N.K. Cuong, C. R. Acad.
Sci. Paris, Ser. C, 888. 1714 (1968).

 

 

 

C.H. Foster and G.A. Berchtold, J. Org. Chem., 88, 3743 (1975).

 

W. Adam and H.J. Eggelte,Angew. Chem., Int. Ed. Engl., 16, 713
(1977); W. Adam, A.J. Bloodworth, H.J.Eggelte,and M.EJVLoveitt,
ibid., L1, 209 (1978); W. Adam and I. Erden, ibid.,,Lz, 210 (1978).

 

H.H. Wasserman and D.L. Larson, Chem. Commun., 253 (1972).

 

H. Hart and A. 0ku, Chem. Commun., 254 (1972).

 

For examples of other epoxidations of endoperoxides, and for

leading references, see W. Herz, R.C. Ligon, J.A. Turner, and
J.F. Blount, J. Org. Chem., 42, 1885 (1977); J.A. Turner and

w. Herz, ibid., 65’ 1895, 1900, 2006 (1977).

 

The numbers on structures correspond to the chemical shifts

91

15.

16.

17.

18.

19.

20.

21.

22.

23.

24.

25.

26.
27.
28.
29.

92

expressed as 6, and the numbers in parentheses are the relative
slopes of the LIS shifts using Eu(fod)3 shift reagent.

We are indebted to Dr. Rodney Willer for suggesting and arranging
for the dipole moment measurments, and to Professor E.L. Eliel
for obtaining the results.

8. Hagt, J.B.-C. Jiang and M. Sasaoka, J. Org. Chem., 88. 3840
1977 .

 

C.J.M. van den Heuvel, H. Steinberg, and Th.J. de Boer, Rec. Trav.
Chim., 88, 157 (1977).

 

L.F. Johnson and W.C. Jankowski, "Carbon-l3 NMR Spectra", Wiley-
Interscience, Inc., New York, 1972, spectrum 359.

A similar rearrangement was observed with dibenzobarrelene oxide
;16 see also S.J. Cristol and R.K. Bly, J. Am. Chem. Soc., 88.
55 (1960).

 

After a prolonged reaction time, 888 gave several unidentified
products in low yield.

Although a few examples of 1,3-peroxide oxygen shifts have been
reported [A.P. Schaap, P.A. Burns, and K.A. Zaklika, J. Am. Chem.
Soc., 88, 1270 (1977); J. Griffiths, K. Chu, and C. Hawkins, Chem.
Commun., 676 (1976); T. Wilson, Photochem. and Photobiol., 10. 441
(19691I we are not aware of 1,2-shifts such as might possibY? have
occurred with 888.

 

 

J.N. Brown, R.L.R. Towns, M.J. Kovelan, and A.H. Andrist, J. Org.
Chem., 88, 3757 (1976).

An improvement over literature procedures [E. Wolthuis, B.
Bossenbroek, G. Dewall, E. Geels, and A. Leegwater, J. Org. Chem.,
88, 148 (1963)] was developed; see Experimental Section for details.

 

H.O. House, "Modern Synthetic Reaction", W.A. Benjamin, Inc., Menlo

Park, 1972, p. 303; H. Hart, M. Verma, and 1. Wang, J. Org. Chem.,
, 3418 (1973); P.G. Gassman and J.L. Marshall, J. Am. Chem. Soc.,
, 2822 (1966).

 

 

H.C. Brown, "The Nonclassical Ion Problem", Plenum Press, New York,
1977, p. 123.

P.D. Bartlett and W.P. Giddings. J. Am. Chem. Soc., 88, 1240 (1960).

 

K. Kitahonoki and Y. Takano, Tetrahedron Lett., 1597 (1963).

 

E.K. Nelson, J. Am. Chem. Soc., 88, 1404 (1911).

 

J. Boche and 0. Runquist, J. Org. Chem., 88. 4285 (1968).

 

30.

31.

32.

33.

34.
35.
36.

37.
38.
39.

40.

41.
42.

43.

44.

93

J. Rigaudy, J. Baranne-Lafont, A. Defoin, and N.K. Cuong, C.R.
Hebd. Seances Acad. Sci., Ser. C, 280, 527 (1975); Chem. Abstr.,
83, 8829 m (1975).

 

H.H. Hasserman and J.R. Scheffer, J. Am. Chem. Soc., 89, 3073
1967 .

 

The photochemical rearrangements of naphthalene endoperoxides to
diepoxides is known.6

An analogous reaction was observed when 9,10-demethy1anthracene
endoperoxide inwas heated, hydroxyketone jj_being one of the
products; J. Rigaudy, M. Moreau, and N.K. Cuong, C.R. Acad. Sci.,
Ser. C, 274, 1589 (1972).

 

K. Maruyama and Y. Naruta, Chem. Lett., 431 (1978).

 

w. Adam and J. Arce, J. Am. Chem. Soc., 27, 926 (1975).

 

I. Kishida, T. Hiraoka, and J. Ide, Tetrahedron Lett., 1139
1968 .

 

L.I. Smith and I.M. Webster, J. Am. Chem. Soc., 52, 662 (1937).

 

J. Rigaudy and C. Breliere, Bull. Soc. Chim. Fr., 1390 (1972).

 

E. Wolthuis, w. Cady, R. Room, and B. Neidenaar, J. Org. Chem.,
31, 2009 (1966).

 

 

H. Hart, P.M. Collins, and A.J. Waring, J. Am. Chem. Soc., 88,
1005 (1966). ’V”

8.5. Thyagarajan, Intra-Science Chem. Reports, 4, 42 (1970).

 

J.D. Roberts, A. Streitwieser, Jr., and C.M. Reagan, J. Am.
Chem. Soc., 74, 4579 (1952); E. Weltin, F. Gerson, J.N. Murrell.

 

and E. Heilbronner, Helv. Chim. Acta, 44, 1400 (1961); J. Aihara,
Bull. Chem. Soc. Japan, gg, 517 (1975YY”

 

 

A.T. Blomquist and D.T. Longone, J. Am. Chem. Soc., 81, 2012
(1959); E.A. Dorko, ibid., , 55187(1965); P.A. waitkus, 5.3.
Sanders, L.I. Peterson, and .w. Griffin, ibid., 89, 6318 (1967).

 

G.w. Griffin and L.I. Peterson, ibid., Q4, 3398 (1962).

45.

46.

47.

48.

49.

50.
51.
52.
53.
54.

55.

56.
57.
58.

59.

60.
61.

62.

63.

94

H. Hopff4 and A.K. Wik,He1v. Chim. Acta,43, 1473 (1960);
ibid., 19 (1961); H. Hopff and A. Gat1, ibid. ,m8, 509

(1965);M1’bid.,4,4/§, 1289 (1955).
A.J. Barkovich, E.S.Strauss,and K.P.C. Vollhardt, J. Am. Chem.

 

 

 

Soc., , 8321 (1977); P. Schiess and M. Heitzmann, Helv. Chim.
Acta, , 844 (1978); L.G. Harruff, M. Brown, and V. Boekelheide,
J. Am hem. Soc.,,LgQ, 2893, (1978).

 

G. K6brich and H. Heinemann, Angew. Chem. , Int. Ed. Engl., 4,
594 (1965).

 

R.0. Uhler, H. Schechter, and G.V.D. Tiers, J. Am. Chem. Soc.,
54. 3397 (1962).

 

Z.Berkovitch-Ye11in, M. Lahav, and L. Leiserowitz, ibid. ,96,
918 (1974). mm

 

P.J. Garratt and 5.8. Neoh, J. Org. Chem., 40, 970 (1975).

 

H. Hart and M. Sasaoka, J. Am. Chem. Soc., 180, 4326 (1978).
M.P. Cava and M.V. Lakshmikantham, Acc. Chem. Res., 8, 139 (1975).

 

M.P. Cava and G.E.M. Husbands, J. Am. Chem. Soc., 21, 3952 (1969).

 

M.P. Cava, M. Behforouz, G.E.M. Husbands, and M. Srinivasan,
ibid., 95, 2561 (1973).

M..D.§1ick and R. E. Cook, Acta Crystallogr. ., Sect. B,2§, 1336
1972

 

K.T. Potts and D. McKeough, J. Am. Chem. Soc., 25, 2750 (1973).

 

J.M. Hoffman and R.H. Schlessinger, ibid., 91, 3953 (1969).

Benzo[c]furans: R.C. Elderfield,Heterocyc1. Compd.,2. 68 (1951);
R.N. Warrener, J. Am. Chem. Soc., 93, 2346 (1971). Benzo[c]thio-
phenes: B. Iddon, Adv. Heterocyc1.'VChem., 1A4, 331 (1972);
Isoindoles: J. D. White and M. E. Mann, ibid. , 10, 113 (1969).

 

 

 

E. Chacko, J. Bornstein, and D.J. Sardella, J. Am. Chem. Soc.,
22, 8248 (1977), and leading references therein.

 

D.W.H. MacDowell and M.H. Maxwell, J. Org. Chem., 35, 799 (1970).

 

R. Kreher and K.J. Herd, Angew. Chem., Int. Ed. Engl., 17, 68
(1978). ‘9'

 

J. Ciric, S. L. Lawton, G. T. Katotailo, and G. W. Griffin, J. Am.
Chem. Soc. , 100, 2173 (1978).

 

G.H. Heeres and H. Wynberg, Synthetic Comm., 2, 335 (1972).

 

64,

65.

66.

67.

68.

69.

70.

71.

72.

73.

74.

75.

76.

77.

78.
79.
80.

81.

95

P. Jordens, G. Rawson, and H. Wynberg, J. Chem. Soc. C, 273
(1970).

 

A. Kraak, A.K. Wiersema, P. Jordens, and H. Wynberg, Tetrahedron,
88, 3381 (1968).

 

T. Kauffmann, B. Greving, J. K6ning, A. Mitschker, and A. Woltermann,
Angew.Chem., Int. Ed. Engl.,,Lfi, 713 (1975).

 

F.B. Mallory, 0.5. Wood, J.T. Gordon, L.C. Lindquist, and M.L.
Savintz, J. Am. Chem. Soc., 88, 4361 (1962).

 

 

R.M. Kellogg, M.B. Groen, and H. Wynberg, J. Org. Chem., 32, 3093
(1957). “W

H. Wynberg, H. van Oriel, R.M. Kellogg. and J. Buter, J. Am. Chem.
Soc., 88, 3487 (1967).

 

R. Meyer, H. Kleinert, S. Richter, and K. Gewald, J. Prakt. Chem.,
88, 244 (1963).

 

M.P. Cava, N.M. Pollak, 0.A. Mamer, and M.J. Mitchell, J. Org.
Chem., 36, 3932 (1971).

8. Prgetzsch, F. Bieniek, and F. Korte, Tetrahedron Lett., 543
1972 .

 

H.J. Backer, Recl. Trav. Chim. Pays-Bas, 88, 745 (1935).

 

L.M. Jackman and S. Sternhell, "Applications of Nuclear Magnetic
Resonance Spectrosc0py in Organic Chemistry", Pergamon Press,
Oxford, 1969, p. 209.

L.F. Johnson and W.C. Jankowski, "Carbon-13 NMR Spectra", Wiley-
Interscience, New York, N.Y., 1972, spectrum 50.

A.J. Jones and D.M. Grant, Chem. Commun., 1670 (1968), data are
for CS2 solvent.

 

"UV Atlas of Organic Compounds", Vol. 3, Plenum Press, New York,
N.Y., 1967, spectrum E 5/2.

ibid., Vol. 5, 1971, spectrum 5/1.

Reference 74, p. 307.

Wynberg, et al. reported the same results in the deuteration of
Why CH CO 0. R.M. Kellogg, A.P. Schaap, and H. Wynberg, J. Org.
Chem., 8%, 343 (1969).

G. Briegleb, J. Czekalla, and G. Reuss, Z. Phys. Chem., Abt. B,
88, 316 (1961).

82.
83.

84.
85.

86.
87.
88.

96
Unpublished result of Dr. Avraham Teuerstein.

L.A. Paquette, "Principles of Modern Heterocyclic Chemistry",
W.A. Benjamin, Inc., London, 1968, p. 116.

H. Hogeveen, Recl. Trav. Chem. Pays-Bas,l88, 1072 (1966).

 

8. Gigvannini and H. Vuilleumier, Helv. Chim. Acta, 88, 1452
1977 .

 

Unpublished result of Mr. David Makowski.
K. Auwers and T.V. Bredt, Chem. Ber., 88, 1714 (1894).

 

8.T. 8tapler and J. Bornstein, J. Heterocycl. Chem.,,LQ, 983
1973 .