PART I
THE SYNTHESIS OF HTGHLY SUBSTITUTED
PERI-MET HYLAT ED ANTHRACENES

PART II
MISCELLANEOUS

Dissertation for the Degree of Ph. 9..
MICHTGAN STATE UNIVERSTTY
JACK BRIE-CHEN HANG
1975

w.

.5:

     
   

1!!!" '. .--, .v 5t.»-

.4 I i x .1 4.
Mill-5"“Ire

. 5"“‘3mt State

diversity

      

This is to certify that the

thesis entitled
Part I:

Part II:

The Synthesis of Highly Substituted
Peri-methylated Anthrecenes
Miscellaneous

presented by

Jack B. Jiang

has been accepted towards fulfillment
of the requirements for
Ph .1). degree in Organic Chemistry

 

 

 

Wee LJ
Date July 28, 1975

‘7

Major professor

 

 

0-7639

 

       
   

BINSING BY
"(MS & SflNS'
800K BINDERY INC.

LIBRARY BINDERS

"It'll? misma- I

.-

v
~—.
.

       

. :1 .r” ' ABSTRACT

PART I
THE SYNTHESIS OF

HIGHLY SUBSTITUTED PERI-METHYLATED ANTHRACENES

PART II

MISCELLANEOUS
By

Jack Bau-Chien Jiang

The purpose of the first part of this thesis was to
synthesize highly methylated anthracenes, with decamethyl-
anthracene (3) as the ultimate target compound. This syn-
thesis presented the problem of constructing an aromatic
system and, at the same time, keeping to a minimum the ef-

fects of peri interactions and annelation.

 

2 Jack Bau-Chien Jiang
The first route began with hexamethylnaphthalene (2),
which was converted to its acetyl derivative (2). Genera-
tion of the enolate anion of‘g followed by treatment with
ethyl bromoacetate produced the keto ester lg) which in
turn was hydrolyzed to ll, Ring closure of ll_was achieved
but only in low yield; the further transformation of ll'to

3 was therefore abandoned.

1)L1HS

$3 7W

OEt

w< ....... ”lg”;

The second route was an imitation of the synthesis of

 

g, which was prepared from dimethylbenzyne and hexamethyl-

dienone 1;.

¢. +

Hence, preparation of the 2,3-naphthyne turned out to be

 

3 Jack Bau-Chien Jiang
the primary target in this synthesis. Compounds gl, gg,
and §§_were synthesized as possible precursors of the desired
2,3-naphthyne; unfortunately none of them served this pur-

pose successfully. Preparation of £2, the analogue of the
0

Br :

_ Nak [- flan _
”I

well—known naphthyne precursor 3—amino-2-naphthoic acid, was
also tried. The instability and lack of a high-yield syn-

thesis of gfl_were obstacles to the preparation of compound lg.

 

A bridged ketone gg, which would be expected to yield
a peri-substituted anthracene on photolysis, was the key
synthetic intermediate of the third route. The epoxide 11,
obtained from the Diels-Alder adduct of i with dimethylben-
zyne, did not afford the hoped-for gem-dimethyl analogue of

§§_by a rearrangement reaction, and photolysis of 59 did not

4 Jack Ban—Chien Jiang

eliminate 2-butyne but gave rearranged products.

 

The fourth approach involved the extrusion of oxygens
from the endoxide 29. The common procedures which work suc-
cessfully with systems like tetrahydro-l,D-epoxynaphthalene,
failed with gg, Iodine-catalyzed dehydration in acidic media
only expelled one oxygen atom, to give El, A special reagent,
bromotriphenylphosphonium bromide, proved effective in ex-

truding both oxygens, but gave 58 an isomer of 3,

 

3 H
mm
Ce
:L':
:1:
+
V

 

The last synthetic route was the only satisfactory me-

thod which resulted in anthracenes with substituents at peri

5 Jack Bau-Chien Jiang
positions. This method consisted of (1) chemical reduction
of the appropriate benzyne adducts of acridizinium perchlo-

rates followed by (2) thermolysis of the reduced adducts.

 

 

 

One of several anthracenes made by this method was penta-
methylanthracene (1;). A preliminary study of this compound
revealed several consequences of strong peri interactions

among the C -, 08-, and C -methyls (R1, R3, Ru).

Miscellaneous result: that constitute the second part
of this thesis include (a) Wagner-Meerwein rearrangement in
the dibenzobicyclooctadiene system, (b) non-conventional bro-
mination of octamethylnaphthalene g, and (c) Birch reduction
of‘g.

The epoxide §g and its tetramethyl analogue 11_rear-

ranged, in chloroform solution at room temperature, to §§_

and 1s respectively. Compound 11 rearranged much faster than

6 Jack Bau-Chien Jiang
fig. Further rearrangement of 1§_to 12 in refluxing chloro-
form was observed, whereas that of g; to §fl_was achieved only

by adding traces of acid.

 

 

 

The reaction was initiated by traces of acid contained in

7 Jack Bau-Chien Jiang
the solvent, because when the solvent was changed from
chloroform to pyridine, the above rearrangement was not
observed. Double bond participation was supported by the
related transformation of 85 to §g_on heating with acid.

090+» o 050

85 89

Compound 18 was severely overcrowded. It was not pos-
sible to construct a space-filling model of 18, and its nmr
spectrum did not show the symmetry expected for free rota-
tion of the acetyl group. Compound 8;, however, gave a
beautiful symmetrical nmr spectrum at room temperature due
to the free rotation of the acetyl group. But as the tempera-
ture was lowered to 10°, rotation became hindered and the
nmr spectrum of 8; became consistent with that of 1g. At
much lower temperatures there were signs that rotation of
the quaternary methyl substituent was hindered.

O
Bromination of g'in CS at -78 in the dark gave 29

2
which was readily hydrolyzed and dehydrated to give 21.

8 Jack Bau-Chien Jiang

Br Br 0
Br ,CS
0 O 2 2
———-——9 ————>
.. —78°
g 90 91

Birch reduction of g_with lithium and ethanol in THF and
liquid ammonia produced l,4-dihydrooctamethylnaphthalene (21)
in high yield. The cis—geometry at C1 and CA in 2; was demon—
strated by epoxidation, which afforded an epoxide with a

symmetrical nmr spectrum.

3 .91

 

PART I

THE SYNTHESIS OF
HIGHLY SUBSTITUTED PERI-METHYLATED ANTHRACENES

PART II

MISCELLANEOUS

By

Jack Bau-Chien Jiang

A DISSERTATION

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

DOCTOR OF PHILOSOPHY

Department of Chemistry

1975

The heavens declare the glory of God;

and the firmament sheweth his handywork.
Day unto day uttereth speech,

and night unto night sheweth knowledge.

Psalm 19: 1,2

11

ACKNOWLEDGMENTS

I wish to express my deepest gratitude to Professor
Harold Hart for his patient guidance, timely advice, and
whole-hearted support throughout this research program.

Many thanks are also due to Dr. William H. Reusch for
his careful reading of the manuscript, and to Dr. Bobbie
Barnett for the X-ray analysis.

I would also like to thank all my good friends in our
Chemistry Department who have generously provided me with
suggestions and valuable help in many ways.

Sincere appreciation is extended to Michigan State
University for a Graduate Teaching Assistantship from Sep-
tember 1971 through June 1972, and from September through
December 197H; to the National Institute of Health for
financial assistance from June through December 1972; and to
the National Science Foundation for financial support from
January 1973 through July 1975.

Lastly, I am greatly indebted to my beloved wife, Lily
Yang-Bai, and to my parents for their constant encouragement,

understanding and assistance during these years.

111

TABLE OF CONTENTS

PART I

THE SYNTHESIS OF
HIGHLY SUBSTITUTED PERI-METHYLATED ANTHRACENES

Page
INTRODUCTION .......................................... 2
RESULTS AND DISCUSSION .......... ..... ................. 11
A. Route I: via l,A—Anthraquinone .................. 11
B Route II: via Hexamethyl-2,3-naphthyne .......... 15
C. Route III: via Photolysis ....................... 26

D Route IV: via l,“,5,8-Tetramethylnaphthalene-l,A-
endoxide ........................................ 33

E. Route V: via Acridizinium Compounds ............. A0

EXPERIMENTAL ........ .......... ................... ..... “7
General Procedures .................................. A7
2-Acetyl-1,A,5,6,7,8-hexamethylnaphthalene (2) ...... A7
Ethyl 8-(1,A,5,6,7,8-hexamethyl-2-naphthy1)propionate

lg .. ...... ........................................ “8
Saponification and Cyclization of 19 ................ A9
1,A,5,6,7,8-Hexamethyl-2-nitronaphthalene (2Q) ...... 50
1,”,5,6,7,8-Hexamethyl-2-bromonaphthalene (21) ...... 51
l,”,5,6,7,8-Hexamethyl-2,3-dibromonaphthalene (22) .. 52
Reaction of gl_with Sodium Amide in Ammonia .. ....... 52
1,A,5,6,7,8-Hexamethyl-2-naphthoic acid (2;) ........ 53
1,A,5,6,7,8-Hexamethyl-2-aminonaphthalene (25) ...... 5“

iv

TABLE OF CONTENTS (Continued)

Page
Alternative Preparation of 23 from 33 via 35 ........ 55
l A 5, 6 ,7 8-Hexamethyl- 2- bromo-3- -benzoy1naphthalene
(3g). . ......... ...... .. ..... . 56
Reaction of 3§_with Sodium in Liquid Ammonia ........ 57
3,6-Dimethyl-A-nitroisatin (31) .... ..... ............ 57
3,6-Dimethyl—A-nitroanthranilic acid (3g) ........... 57

1, 3, 3, A 7, 8-Hexamethyl- 5, 6- (3, 6-dimethyl- A-nitrobenzo)-
bicyclo[2. 2. 2]octa-5, 7-dien-2-one (3A) ............... 58

l,A,78 -Tetramethyl- 2, 3- 5, 6- di(3, 6- dimethylbenzo)bi-
cy01012. 2. 2]octa- -2 ,5 7- triene (59) ................... 59

1,2,5,6-Tetramethyl-3,A-7,8-d1(3,5-dimethylbenzo)-
cyclooctatetraene (fig) .............. ..... ............ 60

1,2,5,8-Tetramethyl-3,A-6,7-d1(3,6-d1methy1benzo)tri-
cyclo[3.2.1.02’8]octa-3,6-diene (A3) ................. 61
1A 5, 8-Tetramethyl- l, A- -dihydronaphthalene- -l ,A-endoxide

(A) . .. 62

l ,A ,5 8- -Tetramethyl- 2-methoxynaphthalene (51), and 1-
Methoxymethy1-A 5, 8-trimethylnaphthalene (5 ) ... ..... 62

l,A,5,8-Tetramethylnaphthalene (3g) .................. 63

l ,3 3, A ,5 6-Hexamethy1- 7, 8- (l, A-epoxy-l, 2 3, A- tetra-
hydronaphtho)bicyclo[2. 2. 2]-5-ene-2-one (53) ......... 6A

l,A,5,8,9,lO-Hexamethyl-Aa,9a-dihydroanthracene-l,A-
9,10-diendOX1de (ill) 00.000.00.00 ooooooooo 000000000000 65

Phot01y818 Of -5_l4 0.00.0...00.....0000......0.....0.00. 66

1, A ,5 8 9, 10-Hexamethy1- -2 ,3 Aa ,9a-tetrahydroanthracene-
1.:14-9, 10-diendox1de (5—6) ....0......0 ..... 0.....000... 66

1, A 5, 8 9- -Pentamethyl- 10-methylene- 2, 3, Aa-trihydroan-
thracene-Lu-endOXIde (51) cocoon.00000000000000.0000. 67

1 ,A 5, 8 ,9 Pentamethyl-10-methy1ene-9-monohydroanth-
racene (5__8_) ..... coo oooooooooo oooooooooooooooooooooooo 68

TABLE OF CONTENTS (Continued)
Page

The Preparation of 2-(2-Methy1 [1,3] dioxolan-2-y1)-
pyridine {-6—5.) 0.000.000.000.000...0.0.00.00000 ...... .0 69

l,u-Dimethy1-2-brom0methylbenzene (fig) 0 o o o o o o c o o o o o o o 69

1-(2,5-Dimethylbenzyl)2-(2-methy1[1,3]dioxolan-2-yl)-
pyridinium Br0m1de(_6_7_) 0000.00.000.000000000000000... 70

7,10,ll-Trimethylacridizinium Perchlorate (68) ....... 71
l,A,5,8,9-Pentamethylanthracene (11) ................. 72
1,A,9-Trimethy1anthracene (12) ....................... 73

1,A,5,8,9-Pentamethylanthracene-9,10-end0peroxide (15) 73

PART II
MISCELLANEOUS

A. Wagner-Meerwein Rearrangement in the Dibenzobicyclo-
OCtadiene syStem 0.0.0.0.......OOOOOOOOOOO0.0IOOOOO 76

B. Non-Conventional Electrophilic Aromatic Substitutions
Of octamethylnaphthalene 00.0.0.9000000000000000...

C. The Birch Reduction of Octamethylnaphthalene ...... 97
EXPERIMENTAL 0....0...................................00. 101

1,A,7,8-Tetramethyl-2,3-5,6-di(3,6-dimethylbenzo)bicyclo-
[2 .2 .2] 00138-2,5,7-triene-7-GDOX1de (11) o o o o o o o o o o o o o o o 101

The Rearrangement of 11 ............................... 102

l,A,7,8-Tetramethy1-2,3-5,6-dibenzobicyclo[2.2.2]octa-

2,5,7-tr1ene (fl) 0000000000coco0.000000000000000000000 103
1,A,7,8-Tetramethyl-2,3-5,6-dibenzobicyclo[2.2.2]octa-
2,5,7-triene-7,8-epOX1de (Q) onoooooooocoooooooooooooo 10“
l-Acety1-2,3-6 7-dibenzo-1,A,5-trimethy1cyclohepta—
2,",6-triene (£1) ooooooococoo-oooooooooooooooooooooooo 10”
Thermal Reaction of 82 in a Basic Medium ........ ...... 105

vi

TABLE OF CONTENTS (Continued)

Page
Reduction Of8—3by LAH ...... ......OOCCCOUOCOOOCOOOO 105
1,5,8-Tr1methy1-h-methy1ene- 2 ,3- 6 ,7-d1benzo-8-hydroxy1-
biCyClO 3.2.1 00138-2, 6-d1ene (8h ) 0.0000000000000000 106
1-Acety1- -2 ,3- 6 ,7-dibenzo- l U ,5- -tr1methylcyclopenta—
2,“, 6-tr1ene-U-epoxide (86 6s ........................ 107
1-Acety1-2, 3- 6,7 7-d1(3, 6-d1methy1benzo)- l, u ,S-trimethyl-
cyclopenta-2 ,H, 6-tr1ene-4-epox1de (80) ............. 107
Acid Rearrangement of 8; ........................... 108
The Bromination of Octamethylnaphthalene ........... 109

2.6-Dihydro-naphtho[1,8,8a-c,d]pyran (21) .......... 110
Bisbromomethy1-1,4,5,6,7,8-hexamethy1naphtha1ene (2g) 111
1,3,5,6,7,8-Hexamethy1-2,3-dihydroxymethylnaphthalene

_9__ 0000000000000000000000000...0.0000000000000000. 111

Treatment of 2E with Acid .......................... 112
Dihydro-iso-1,4,5,6,7,8-hexamethy1naphthofuran (26). 113
1,h-Dihydrooctamethylnaphthalene (21) .............. 113

1,H-D1hydro-2,3-epoxy-octamethylnaphthalene (28) ... 114
REFERENCES ...... ................. .................... 116

v11

PART I

THE SYNTHESIS OF

HIGHLY SUBSTITUTED PERI-METHYLATED ANTHRACENES

INTRODUCTION

A three-dimensional X-ray analysis of octachloronaph-
thalene was reported by Gafner and Herbstein in 1963.1 They
found that the molecule is severely deformed due to multiple
displacements of the substituents and the nuclear carbon
atoms. The a-chlorine atoms are displaced out of plane by
0.5H-0.79 X and the B-chlorine atoms are displaced by about
0.37-0.47 Z. The out-of—plane displacement of each carbon
atom is about one-third as much and in the same direction as
its halogen substituent. Furthermore, adjacent a and B chlo-
rines are displaced in the same sense, so that the molecule
takes on a propeller-like conformation (1). These results
are contrary to the earlier findings of Donaldson and Robertson2
regarding the molecular structure of octamethylnaphthalene.

The latter workers suggested, on the basis of a two-dimensional
X-ray analysis, that adjoining methyl groups are displaced

in opposite directions so that they are alternately above

and below the general plane of the molecule (g).

01 01 Me “6

   

Gafner and Herbstein discussed steric strain and its
relief in octachloro- and octamethylnaphthalene. If the
substituents alternate above and below the molecular plane
(i.e. g), causing similar displacements of the ring carbons,
the substituent interaction energy will be minimized but the
skeletal strain will be considerable. In a propeller-like
conformation (i.e. 1) there is more substituent interaction
energy but less skeletal strain. The above authors believe
that the former conformation will be energetically preferable
only when the bonding is tetrahedral, as in cyclohexane; in
overcrowded aromatic systems of the type discussed here, a
propeller-like conformation is considered more likely.

In addition to the X-ray analysis, the peculiar molecular
spectra of these fully substituted naphthalenes also reflect
the fact that the molecule is distorted.3 For example, 1,“,5,8-
tetramethylnaphthalene has an ultraviolet absorption maximum
at 296 nm whereas the all B-substituted 2,3,6,7 isomer has
its absorption maximum at 270 nm. This bathochromic shift
was believed to be the result of peri interactions between
adjoining a-methyl groups. An even longer red shift which
supported this view was observed in the uv spectrum of octa-
methylnaphthalene (308 nm). As for nmr spectra,” a significant
deshielding (about 12 Hz) which appears to result from peri
effects has been observed. For instance, the chemical shift
(relative to water as external reference) of the methyl groups
in 2,7-dimethylnaphthalene and its 1,8 isomer were reported

to be 9u.8 and 78.7 Hz respectively.5 There has been no

clear-cut explanation as to how the peri interaction affects
infrared spectra.

Compared to the original multi-step synthesis6 of octa—
methylnaphthalene (Scheme 1), a simpler, higher-yield synthesis
introduced by Hart and Oku in 1967 (Scheme 2) serves as a

more attractive starting point for further investigation of

highly methylated aromatic systems.

COOH
A1013 COOH CuCO3
O» ——> ——»o
H
2
0 0

HF
1)cho O
9—00 ”W ..
2)-H20
3)‘H2

Scheme 1

 

 

 

   

NaH-DMSO
“0°, 4.5 hr
87%
V .
1)HCHO
HCl
/
\
2)LAH
71%
2.
Scheme 2

A reexamination of the three-dimensional X-ray structure
of g_indicated that it does have a propeller-like geometry
analogous to that of 1,8 It was expected that twisting of
the molecule should alter the n-overlap in g and consequently
affect its reactivity. An extensive study of its chemistry
revealed the unusual reactivity of g_toward electrophiles
and dienophiles.9 For example, g_is fully protonated at an
a—position in trifluoroacetic acid at room temperature, thus

relieving in part the severe peri interactions.

 

W

TFA ‘ll"

r.t.

 

3

Facile electrOphilic attack by bromine on g_was observed,

further details of which are included in the second part of
this thesis.

Examples of cycloaddition reactions of 2_with dienOphiles

are shown in Scheme 3.

 

 

 

COOEt
COOEt

EtOOC COOEt

Scheme 3

While these reactions proceed in high yield, naphthalene
itself reacts with maleic anhydride to form an adduct in very
10
low yield. There is no report of endoperoxide formation

from naphthalene, and no adduct has been obtained from naph-

ll

thalene and benzyne. With dibromocarbene, 2 gave a homoan-

nular bis adduct and a benzomethylene cycloheptatriene.9

There were related reactions of naphthalene and certain car-
11

benes, but the yields were hardly noticeable.

g’ + :CBr

 

Although a number of anthracene derivatives with peri

11 only one crystal struc-

substituents have been synthesized,
ture of an anthracene derivative with substituents at the l,

8, and 9 peri positions has been determined.12 Dellaca and

coworkers found that overcrowding of the peri-substituents in
1,8-dichloro-9-methylanthracene-(§) is relieved by a combina-
tion of in-plane and out—of-plane deformations. The chlorine
atoms are 0.33 K from the mean molecular plane whereas C9 and
the methyl carbon attached to it are on the opposite side of

0
this plane, 0.19 and 0.80 A respectively from it.

 

.3;

In light of above results, it was of interest to syn-

thesize and study the chemistry of decamethylanthracene (3),
which would have four peri interactions instead of two as
in the octamethylnaphthalene system. This was the ultimate

goal of this research.

 

B.

It was reported in the paper on the original synthesis

of‘g (Scheme 1),6 that the steric influence of the methyl
group at position 4 caused the failure of l,2,3,h,6-penta-
methylnaphthalene to undergo chloromethylation at position 5.
And contrary to the usual a-electrophilic substitution in

naphthalenes, chloromethylation took place at the B-position

9
of 1,2,3,h,5,6-hexamethylnaphthalene.

CH201

-*—~>

2

M

Although the final stage of this synthesis involved a success—
ful chloromethylation at the hindered G-position, the yield
was extremely low (2%).

In view of the above facts, the best strategy for syn-
thesis of a fully methylated anthracene would involve prior
introduction of all the peri methyls; and the primary target
compound would become either §,or 1. Subsequent methylation
at the B-positions would complete the synthetic process and
give the final product 5, The results of different synthetic
approaches to these highly substituted anthracenes are the

subject of the first part of this thesis.

10

 

.5. I

RESULTS AND DISCUSSION
A. Route I: vial,4-Anthraquinone

The first synthesis of anthracene was achieved in 1866

by Limpricht, who obtained it by heating benzyl chloride

with water.

@314:

   

Since then, a large number of anthracene syntheses have been
developed (for example, Scheme A). Most of these processes
are based on a single idea; that is, to construct the middle
ring (ring B) from synthons having two well—established

benzene rings (ring A and ring C).
’/\‘
G E) -—->

11

12

-'-”—> <'—— 0

Scheme A

However, this route seemed to be energetically unfavor-
able for our target system because of strong interactions be-
tween methyls at pre-peri positions. Moreover, the annelation
effect of anthracene would produce less aromaticity in the
middle ring, so that the driving force for the formation of
the anthracene skeleton would probably not be strong enough
to conquer the severe peri interactions. A better approach,
therefore, might be to construct ring C after rings A and B
have been set up with suitably located substituents.

Since a high-yield synthesis of 2_is available (Scheme 2),
it seemed reasonable to try to synthesize the target compound 1
by building up the third ring on g, A primary synthetic in-
termediate might be the quinone 8, with a benzoquinone moiety
fused to the unsubstituted sites of 2. Methylation at both
carbonyl groups followed by aromatization should give the all

peri-methylated anthracene 1.

l3

 

However reaction of §_with either succinic anhydride or
maleic anhydride in the presence of aluminum chloride did not
give the desired §.or its dihydro-derivative. Modest success

was achieved in building the third ring by the sequence shown

below.
MeCOCl 1)BuLi
—' A1c13 2)32
COOEt
9613 OEt lo
2- 22% -_ KOH
100%

....... 2+;

*

..... .. ”Q

1h

Friedel-Crafts acylation of §_with acetyl chloride,
using aluminum chloride as catalyst, gave a 96% yield of 9,
which had a carbonyl absorption at 1690 cm-1. On treatment
with strong base (butyllithium in hexamethyldisilazane), the
enolate anion of 2_was generated. Nucleophilic substitution
on ethyl a-bromoacetate produced the keto ester l9: but in
only 22% yield. Saponificationl3 of 12_with KOH and cycliza-
tion of the resulting 11_with polyphosphoric acid afforded
lg (3%). The infrared spectrum of 1g showed no carbonyl ab-
sorption but a medium O-H stretching band at 3N50 cm-l.
The nmr spectrum gave three singlets in the aromatic methyl
region and one singlet in the aromatic proton region with
relative intensities 3:3:3:l, consistent with the symmetric
structure of 1g, Although this product is only one step away
from the desired quinone intermediate, the overall yield
(0.63%) was low enough to discourage us from continuing this
approach.

1A

In 1971 , Havsigk synthesized l,2,3,4-tetramethyl—

naphthalene in good yield from prehnitene and 1,H-dichloro-

butane.

Cl

+ —-» o. —» to
35% '2H2

01 92%

15
Attempts to prepare anthracene precursor based on a similar
approach were tried under various conditions, but none of

these ended up as expected.

AlBr3

Br

mo”
Br

 

B. Route II: via Hexamethyl-2,3—naphthyne

The second approach to our target compound was based
on the synthesis of 2, which was formed by elimination of
dimethylketene from the Diels-Alder adduct of 2,5-dimethyl-
benzyne and dienone 1; (Scheme 2). Thus, if a naphthyne
could be formed at 02 and C3 of _5_, it should undergo Diels—
Alder addition with 13 or with dimethylfuran, and the adduct

should be fairly easily aromatized.

16

 

For the purpose of obtaining our desired 2,3-naphthyne
intermediate, a brief review of benzyne preparations seems
to be essential. Many syntheses of benzyne involve removal
of two adjacent substituents from an aromatic nucleus.15
Among those well known benzyne precursors, only a few have
been applied to the 2,3-naphthyne system. For instance,16
naphthyne has been generated from a halonaphthalene (15) on
treatment with strong base (e.g. NaNH2); from aminonaphtho—
triazole (1;) on oxidation by Pb(OAc)u followed by elimination

of nitrogen; or from the commonly used 3-amino-2-naphthoic

acid (16) on diazotization and thermal decomposition.

17

XNaNH2 00 ”I

_11
NPb(0Ac)u - .
005% ”N“ 00!
OCOOH OCOOH 0|
ONHZ iso-
amylnitrite
16

Consequently, the correSponding synthetic intermediates,
11, 18, and 12_were considered as promising candidates for

our naphthyne precursors.

x N\\N ZOONHZ COOH

 

However, difficulties were experienced in introducing
functional groups at the unsubstituted B-positions of 2.
For example, being sensitive to strong acid, §_was not readily

l7
nitrated. An unpublished result by Hart and Oku illustrated

18

that an attempt to prepare nitroprehnitene resulted in
producing nitrodurene instead; and carboxylation of durene
with a Lewis acid (e.g. A1013) in carbontetrachloride gave
2,3,“,S-tetramethylbenzoic acid after hydrolysis.

AcOH
COOH
)E o o
AlCl3 H 20

This type of methyl migration in acidic media was also

 

0
observed by us, upon heating §_with AlCl3 at A0 for 4 hr.

Nevertheless, 2 could be converted to 19 under the mild

conditions of Gordon's method;L8 which was previously employed

 

to nitrate benzene and anisole. However, a 2% yield of 19

inhibited any further exploration in this route.

l9

 

IU'I

 

A surprising achievement was obtained in the low-
temperature bromination of Q, The reaction was carried out
in darkness at dry-ice temperature (-78‘0 and gave a 5&2 yield
of 21 within 30 min. The mass spectrum (M+ = 290) of 1 showed
only one bromine (p + 2 = 98% of p), and the nmr spectrum re-

vealed only one aromatic proton, at 6 7.15.

Br CS

2’ 2
-730

30 min, Shz

 
 
 

    

IU'I
CD
"3

2Br
3o-uoz

2

 

20.

Dibromohexamethylnaphthalene (22) was also prepared under
the same condition from either §.or 21, with 30-A01 and l5%
yields respectively. The nmr spectrum of gg_showed no peaks
in the aromatic region, and the mass spectrum gave p + A,

p + 2, and p peaks with relative intensities 1 : 2 : 1.
These data indicated the presence of two bromine atoms on the
aromatic ring.

Unfortunately, attempts to make naphthyne from 11 failed.
Treatment of g1 with sodamide in liquid amonia and 2,5-di-

methylfuran resulted in recovery of g1,

{>1
Br
0 a
O ’ No Reaction
NaNH2

NH

 

3
21

The 100% recovery implied that naphthyne was not formed;
otherwise an amino derivative should have been observed.
This may be due to the electron releasing inductive effect
of the six methyls, which makes the proton too weakly acidic

to be attacked even by as strong a base as NH -.

(E.- NH H2
”NH __,[ ”'1 *N

“/an

21
An alternative means of generating benzyne, designed by
Wittig and by Huisgen, is the reaction of o-dihaloaromatic
compounds with magnesium.19 But treatment of 22 with magnesium
in ether gave only negative results.
Following common procedures, the Grignard reagent of 21

was prepared and treated with carbon dioxide to produce a

50% yield of -_2_;.

Mg MgBr COOH
21 ———+ 00 ———> CO
'"' ether 1)CO2

N00 2)HCl

Since the Grignard reagent could be formed successfully,
it might provide an entry to the synthesis of gfl_inasmuch as
a synthetic method for converting aryl halides into aryl
amines via Grignard reagents had been devised in 1969.20
It was reported that the Grignard reagent derived from o-
bromotoluene reacted with solution of p-toluenesulfonyl azide21
in THF to give a dark red solution was presumed to contain
the species 11. This, on reduction by Raney nickel-aluminum

alloy in aqueous NaOH, led to an overall 82% yield of amine 18.

22

Br

0
+
meggN'l‘kN u QMe Ni-Al NH2
26 Q a 0 a

NaOH(aq.)

 

 

27 28

This method was applied to the present problem. On
treatment with p-tosyl azide at 0° in THF for 1 hr followed
by reduction with Raney nickel-aluminum alloy in NaOH aqueous
solution, the hexamethylnaphthalene Grignard reagent (22)
yielded “0% of 23, which was isolated as the ammonium chloride
salt. A parent peak at m/e 227 in the mass spectrum (70eV)
of 2fl_was also the base peak. The infrared spectrum showed
N-H stretching absorption at 3u00 cm-1 and N-H bending at
1600 cm-1; the nmr spectrum showed six methyls between 62.20-

2.55, two amino protons at 63.40 and one aromatic proton at

 

 

56.30.
M gBr 1)p-Tosylazid:
3‘ THF, 00 ‘
El'_—_—’ ,
2)Ni-A1
NaOH(aq.)
22_ l HCl-ether

H5 “

23
Owing to the inconvenience of the amination of 2, introduction
of the amino group at an earlier stage in the synthesis would
be a better alternative for the preparation of 22,

In 1925, 5-nitroisatin was prepared by the nitration of
isatin with sulfuric acid and fuming nitric acid?2 By fol—
lowing the same procedure, 21 was prepared in a 90% yield
from commercially available 3,6-dimethylisatin. The subsequent
ring opening with 3% hydrogen peroxide in 10% sodium hydroxide
resulted in 22 (99%).

NO
0 HNo 2 0 H 0 NO 00H
0 O ——->2 2
H SO N NaOH H
13 O 2 u H 99% 2
19. 9°” 31 32

The nmr spectrum (CD CN) of 21 showed a singlet for the

3
C3—methy1 at 62.65, which was 10 Hz lower than the corresponding
methyl signal in 22, due to the adjacent nitro group. Compound
22 gave a correct elemental analysis, and its nmr spectrum
consisted of two three-proton singlets, one at 62.10 and the

other at 62.flh. Compared with the nmr (CD CN) of 3,6-di-

3
methylanthranilic acid, these data are consistent with the

structure which has the nitro group at CA’

2H

 

52.3u 52.uu
0 COOH 20.00011
0 NH2 H2
52.00 52.10
31 32

Diazotization of 22 followed by cyclo-addition with 22'
afforded a 33% yield of adduct 25, Reduction of 2fl_with LAH
in absolute ether at reflux temperature produced 22_(99%).
The infrared spectrum of 22_showed that both the carbonyl and

-l
nitro groups were reduced (3500 cm ).

 

0 OH
NO NH2
33% LAH
2g.+1l2 ‘—————fi> 42>
A Ether
99%
314 _3_5_

The transformation of 22 to gfl_was achieved both by

pyrolysis of 22 in benzene at 210-2300under nitrogen (22%)

23 o
and by the reaction with dimslsodium at 40 (10%).

OH

20 <.> 210-23o°
9 at:

or NaH-Dmso,uo°

 

32

25
The instability of 23 (it decomposed into a dark oil at
room temperature) and the lack of a high-yield synthetic
pathway made this approach unattractive for 2,3-naphthyne
synthesis.
Another possible route to the desired 2,3-naphthyne was
based on a reaction of o-halo-aryl ketones that has been shown15

to yield benzyne, probably via o-halogenophenyl anions

(Scheme 5).

 

L 05;? 0'

Scheme 5

Since 21, the bromohexamethylnaphthalene, had been prepared
in adequate yield, benzoylation of g; would provide a conven-
ient route to ;§, the potential precursor of 2,3-naphthyne.

In fact the benzoylation of g; was easily accomplished by a

 

 

FriAEdel-Crafts type of reaction at —5m5° in methylene chloride

26
with aluminum chloride as the catalyst (38% yield). The mass
spectrum (70eV) of 2§_showed one bromine (p + 2 = p) in the
molecule; the infrared spectrum showed carbonyl absorption
at 1680 cm_1. Unfortunately, reaction of 22 with sodium amide
in liquid ammonia in the presence of excess dimethylfuran as

the diene, gave only a 100% recovery of the starting material.

NaNH
2

 

)’ No Reaction

‘\\
/ NH

C. Route III: via Photolysis
2H
Givens and Oettle irradiated benzobicyclo[2.2.fl octa-

dienone (21) in acetone. Among other products, naphthalene

was isolated.

 

 

27
This approach suggested another possible route to the
anthracene system. If we could make the corresponding di-
benzobicyclo[2.2.2]octadienone (22) with three methyls
aligned at all the peri positions, the target compound 2

should be easily obtained through photolysis.

 

Cycloaddition of 2_to 22_gave flg_in 5.1% yield. Its
nmr spectrum had a sharp singlet for the four aromatic protons
at 66.40 and other peaks as expected from the formula.
Addition occurred exclusively on the ring with four methyl
substituents. None of adduct (32) was obtained. The two-
carbon alkene bridge in 39 could not be converted to a
carbonyl-containing bridge, although epoxidation of £9 and
subsequent rearrangement of the epoxdie did provide us with
interesting results which are described in the second part

of this thesis.

28

 

O
O

5.1%

 

 

Pyrolysis (600°) or photolysis (Vycor) of flg_showed no
elimination of 2-butyne but only the rearrangement product
32 (50-60%). Its nmr Spectrum had three sharp singlets, one
with four hydrogens at 66.59 and two with four methyls each,
at 62.08 and 1.91.

—-’O O

42

In addition to £2_(40%), irradiation of flg_in acetone

 

 

 

using a Vycor filter gave fl2_(50%) which was from a di-w—methane

29
rearrangement. This result is consistent with the irradia-
tion of dibenzobarrelene in acetone,25 which produced the di-
benzo analogue of semibullvalene, except that dibenzosemibul-
1valene was formed exclusively (85%) and no product correspond—

ing to 32 was reported.

hv

 

 

In order to avoid the problem presented by the two vinyl
methyls in 39, compound 36 was synthesized (by two different
routes). The endoxide 53, which was prepared from 22 and 2,5-
dimethylfuran, was treated with lighium naphthalenide.26 A
100% yield of ߤ_was obtained. This is the most efficient
synthesis of compound lt_6_ yet reported.ll’27 Compound _’-_l_6_ melts

at 131-2° (111:. 132-3").

3:0: —" .(0 L1-”
g

\32 ‘ Hal/1

.3 MeOH

45

 

30

Another route to £§_involved the reduction of £3 fol-
lowed by acid-catalyzed dehydration of the resulting fl2_(see
page 3h). The yield of 26 by the latter route was no more
than 60%.

The reaction of ߤ_and 22_was then carried out under
the usual conditions. No expected adduct was obtained, and
ߤ_was recovered unchanged. Higher temperatures and longer
reaction times did not improve the inertness of 36 toward

benzyne.

 

ClCH2CH2Cl
o o + O
N2+C1- or ’ No Reaction
diglyme
u5 39 reflux,l6 hr

The above unsatisfactory result: prompted us to modify
2 29. 30
the preparation of 28. Endoxide 31_ was found to

undergo rearrangement in methanol with a trace of acid to
give a-naphthol. A 1965 paper reported the related transforma-

tion 38 9 £9.31

 

 

c? a:
C)
:2

V

 

O
6
O
O

MeOH

31
The presence of two methyl groups in ߤ_apparently affected
the orientation of the methoxyl group. A mechanism for this

reaction was proposed and is shown in Scheme 6.

of: w -

O
H
OH 1
* “—— O

OMe

‘::I"[::I’ OMe llllmlliil

QH
(0H
H+
Scheme 6

In light of the above results, endoxide flfl_was expected
to give the intermediate 29_which might undergo cycloaddition

with 22 to furnish 28.

(3‘1.

50

 

On treatment with acid in methanol, flfl_did rearrange,
to 22_and 22, Unfortunately, the desired product 22_was
formed in only 2.3% yield, whereas 22_was the major product

(6u.h%).

32
The nmr spectrum of 22 revealed two singlets for two distinct
aromatic methyls at 62.80 and 2.55, and one singlet for the
remaining two aromatic methyls at 62.75. Compound 22_only
gave one singlet for three aromatic methyl groups at 62.75,

and another singlet for two methylene protons at 6N.60.

OMe
+
H OMe
010—» 00 +00
MeOH
an 51 52

The formation of 22_probab1y can be rationalized as in

Scheme 7.

OMe

O
Q
I
O
6

Scheme 7

33

D. Route IV: via 1,“,5,8-Tetramethylnaphthalene-l,H—endoxide

It has been reported that 21 is a good dienOphile, and

it has been used successfully in an anthracene synthesis

 

 

(Scheme 8).32 :1:
.10 . 00‘!
._ ”0

Scheme 8

In contrast, our investigation of the similar reaction
between flfl_and 2,h-hexadiene resulted in no reaction. Of
course 2,4-hexadiene is not nearly as successful a diene as
2,3-dimethylbutadiene in Diels—Alder reactions. 0n the other
hand, l,fl-dihydggnaphthalene-l,h-endoxide fll_(prepared by

Fieser's method ) underwent a beautiful reaction with 22,

affording 50% of 22 which might be a potential anthracene

+ x N

over night

precursor.

 

3h.

 

In 22, the coupling constant between the proton at the endoxide
bridgehead and the proton at C7 was 5 Hz. This indicated that
the dihedral angle was about #00, which is consistent with a

structure in which the hydrogens at C and 08 are exo. The

7
low chemical shift ( 61.7-l.8) of the vinyl methyls, presum-
ably due to deshielding by the benzene ring, as well as the
larger europium shift numbers for the exo hydrogens compared
with the vinyl methyls supported the configuration shown.
However, the failure of the addition of 22 to 22 made

us give up any further investigation on the chemistry of 22

and its related systems.

0
+ H —"'——> No Reaction
£3 .13.

Another synthetic example, using the same endoxide system,
30
was given by Wittig and Pohmen in 1956. This inspired us

to investigate compound 22, which was prepared from 22 and

35
2,5-dimethylfuran in diglyme at reflux temperature for 72 hr

(60% yield). The nmr spectrum of compound 2fl_was comprised

of three sharp singlets ( 62.32, 1.73, and 1.60) for the six

 

reflux ,72 hr .6

60%

methyl groups at the peri positions, one singlet for two vinyl
protons ( 66.M5), and another singlet for two aromatic protons
( 66.85). Protons at Gila and C9a appeared as a singlet at
62.h0. The configuration of 22_was determined by an X-ray
analysis of its reduced derivative, 22, which was obtained
from a reaction of 22 with diimide in 70.9% yield. This re-
duction can likewise be achieved under hydrogen atmosphere
at room temperature with Pd/C as the catalyst.

Extrusion of the endoxide oxygens in 22 unfortunately

could not be accomplished by refluxing the substrate in methanol

36
33
with a catalytic amount of HCl. Since iodine is often used
to dehydrate alcohols, a few crystals of iodine were added
to the mixture to facilitate extrusion of the oxygens. The
result was that only the oxygen in the middle ring was removed.
Furthermore, even under such mild acidic conditions, the product

21_was found to have an exocyclic double bond rather than an

aromatic central ring.

 

The infrared spectrum of 21 showed C-O-C absorption at

-1 -1
1180 cm and a strong terminal methylene absorption at 960 cm ;

the nmr spectrum had two vinyl protons with different chemical
shifts ( 65.50 and “.95), and one vinyl methyl at 61.69.

It is obvious that extrusion of two endoxide oxygens is
more difficult than that of one, so that a search for a more
powerful reagent became essential to complete the investiga-
tion of this approach.

Recently DeWit and Wynberg published their results on
the application of an effective deoxygenation reagent, tri—

phenylbromophosphonium bromide, in the system shown?“

37

¢3PBr+B:- / ‘a
S Q

 

As a model compound, the saturated derivative of 21,
1,2,3,N-tetrahydronaphthalene-l,H-endoxide, was chosen for
a mechanistic study of this deoxygenation reaction.314 The
prOposed mechanism involved an initial weak complex formation
between the oxygen and the phosphorus, which facilitates 3N1
cleavage of the oxygen-carbon bond. Then reaction with
bromide ion followed by abstraction of triphynylphosphine
oxide leads to the dibromide. The final stage is a dehalogena-
tion reaction, which was reported to occur on a silica gel

column.
Br

 

3+
w-wicn
a 81*:
w
W

 

M

O Silica gel .. + Q, PO
3
-2HBr

Br

38

It is known that iodide ion is both a good nucleophile
and leaving group. Hence, an iodide derivative of the phos-
phonium salt should be even more effective for the purpose of
deoxygenation. The iodide reagent was made by treatment of
triphenylphosphine with an equimolar amount of iodine in
DMF at low temperature (-5m00). The reaction of 22 and tri-
phenyliodophosphonium iodide was carried out at 70°for 2 hr
and l35-140°for ”.5 hr. After work-up, 20% of 22_was obtained.
Its uv spectrum had an intense maximum at 252 nm (51.3 X 164)
and a shoulder at 289 nm (8 X 162); the nmr spectrum showed
one singlet for two vinyl protons at 65.62, one quartet for
the C9-proton at 611.110 (J = 6 Hz), and a doublet for the C9-
methyl at 61.30 (J = 6 Hz). The chemical shifts of the aromatic
methyls are shown in the formula. Compound 22 is a tautomer

of 2 in which the strain due to four peri interactions is

avoided by not having an aromatic central ring.

 

 

An attempt to remove the oxygens photochemically was
also tried by irradiating 22 in anhydrous ether with a Corex
filter for 8 hr. An ether-insoluble crystalline solid was

obtained in 5-6% yield. The mass spectrum gave m/e 592 as

39
the parent peak and the ir spectrum revealed C-O-C absorption
at 1170 and 1160 cm- . A correct analysis was obtained for
CAOHA8OM’ which indicated the formation of the dimer of 22
(i.e. 22). Two more compounds were isolated from the ethereal
solution by chromatography on alumina with 50% ether in petro-
leum ether as the eluent. The first fraction was 22 (17.5%)

which was formed presumably by hydrogen abstraction from

ether. The second fraction was unreacted 22_(10%).

hv
2 > 0 +fi+fl
Corex

 

 

The above result suggested one way in which the strain
energy in an all peri-methylated anthracene system might be
dissipated under the conditions used to form the system.
Because of the annellation35 effect in the anthracene series,
the peri-methylated anthracene would prefer a structure with
minimum peri interactions to one with a fully conjugated
aromatic skeleton. A new synthetic method which would furnish
non-acidic media at the key step, the step where peri interac-

tions and aromaticity are introduced, would be more suitable

for decamethylanthracene synthesis.

N0

E. Route V: via Acridizinium Compounds

Acridizinium salts (22) were first reported by Bradsher
and Beavers in 1951!.36 The salts were prepared by cyclization
of the quaternary salts (22) formed when picolinic aldehyde
reacted with an appropriate benzyl halide.

O . Kim: @1‘6“,

22 60

 

Oxime derivatives of picolinic aldehyde were found to be more

desirable37 than picolinic aldehyde itself for the purpose of

a high-yield synthesis of 22. But the most effective synthesis

was achieved38

by using the dioxolan ketal derivatives (22)
which were prepared by refluxing a mixture of the picolinic

aldehyde, ethylene glycol and p-toluenesulfonic acid in benzene

 

for 6” hr.
HO OH
<:) > (:) O
O TSA
p—
H O
.6;
An extensive study of the applicationuof the acridizinium
39 0
salts in synthesis was made by Fields. , Several substi-

tuted anthracenes (63a - 63g) were prepared by a simple and

A1
convenient synthesis consisting of catalytic or chemical re-
duction and thermolysis of the appropriate benzyne adducts

39
of ua-azoniaanthracene perchlorates (62a - 62g), (Scheme 9).

 

 

R1 Ru
R.
ooo + @. >
R3 + x‘
Q
b. R1=Me, RZNRN=H

C. R1=R3=RH=H, R2=Me

 

i

d. R1=N02, R2~Ru=H r

e. R1=ACO, R2=RH=H,
R3=t-Bu

r. R1=R2=Ac0, R =Ph,

3

 

Ru=H
g. R1~R3=H, Ru=Ph
Considering the previous unfavorable results, this route
appeared to be worth trying.
Before investing much effort on the synthesis of the
synthetic intermediate 22_which would lead to the product 2
according to Field's method, a simpler system was investigated

first.

ON
:-
ION

A2
The dioxolan ketal of 2-acetyl pyridine (22) was pre-
pared by following the literature procedure.38 Bromomethyl-
p-xylene (22) was synthesized in uoz yield by bromomethylation
of p-xylene under the conditions shown. A by-product in this
reaction was 2,5-dibromomethy1-p—xylene, the structure being

proved by reduction to durene.

HCHO,HBr
O ' +
glacial AcOH

95°, 2 hr 3,.

 

LAH, THF
reflux I

O\,
0‘

 

over night

Quaternization of 22 by reaction with 22 in the presence
of tetramethylene sulfone was carried out at 64°‘in a sealed

flask for six days. Compound 21_was obtained in 76.5% yield.

62. 25

00
O/\> ©51. 8
614° 6 days

52. 20

 

6h.0

£2 67
Compound 21_was soluble in chloroform, and its nmr (CDCIB)
absorptions are shown in the formula. Cyclization of 21_was
achieved by heating in HBr (48%) at 120 °fOr 12 hr. The
resulting oil was dissolved in methanol, and 22 was obtained

(100%) upon addition of 35% perchloric acid. Since 21_was

“3

insoluble in most organic solvents and undetectable in the

 

—— 120° 35%

12 hr Br H010 _
u 010,l

68

r MeOH
.7 WE. @OfiO ..., @ONO
4.

mass spectrometer, it was characterized by a correct elemental
analysis and further conversions.

Diels-Alder reaction of 22 with dimethylbenzyne (or
benzyne) produced the ether-insoluble adduct 22 ( or 12),
which was then reduced using sodium borohydride in methanol
containing sodium methoxide. The resulting oily material was
not purified, and the subsequent thermolysis step was satis-
factorily accomplished at reflux temperature in acetic an-
hydride with added sodium acetate. 1,“,5,8,9-Pentamethyl-
anthracene (12) was obtained in 27% yield (1,“,9-trimethyl-
anthracene 12_was obtained in comparable yield, mp. 80-810,

11
lit. 81° ).

 

 

 

I NaOMe/MeOH ’
r.t. @
§§_ > a>
2) Ac20 o R
NaOAc,lh0
l
22 R = Me hr 71 R = Me
19 R = H 72 R = H

22

Pentamethylanthracene (12) is a yellow crystalline
solid and its solutions in organic solvents fluoresce.
In addition to an intensive maximum at 267 nm (86.2 X 10“),
its uv spectrum had bands at the borderline near the uv and
visible regions of the spectrum: 207 nm (3.6 X 103), 386
(2.2 X 103), and 368 (3.5 X 103). The absorption bands of
anthracene itself occur at shorter wavelengths:1 max 375
(9 X 103) and 256 (18 X 10“). The nmr spectrum of 12 showed
two singlets for aromatic protons (one for the Clo-hydrogen

at 58.25, the other for the C C C -, and C7-hydrogens

2" 3" 6
9-methyl at 63.02;

and two singlets with two aromatic methyls each, at 62.80 and

at 67.02); one sharp singlet for the C

2.70.

It was of great interest to investigate the orientation
of the protonation in the middle ring of 12 (it was reported“1
that anthracene itself was protonated at meso-positions in
sulfuric acid). Instead of generating the tertiary carbonium
ion 12_(9-methy1anthracene was protonated on C10 in HF + BF3u2),
12 was fully protonated on the methyl-bearing carbon (09) at

room temperature in trifluoroacetic acid. The nmr data are

shown in the formula 12. The relief of strong peri interactions

61.50(d,J=6Hz)
62.67 H65.60(q,J=6Hz)

    

45

must provide the driving force for protonation. This is
similar to exclusive a-protonation of octamethylnaphthaleneg
(see Introduction).

Irradiation of an undegassed solution of 12 in cyclohexane
with a Pyrex filter for 3 hr, gave 100% conversion to 12.
This endOperoxide 12_was first obtained simply by allowing
a cyclohexane solution of 12 to stand at room temperature
for about a week, exposed to the flourescent lamps of the
laboratory.

Compound 12 gave one singlet at 66.75 for the four aro-
matic protons and one singlet at 56.25 for Clo-proton. The

nmr data for the methyl groups are shown in the formula.

 

 

The above preliminary results revealed the consequences

of strong peri interactions among the C -, 08-, and C -methyls

9
in compound 12; and the ultimate target compound, decamethyl-
anthracene (2), becomes more attractive to us.

However, attempts to synthesize compound 22 (see page
41) were made using 12 as the starting halide. No quaterniza-
tion product was isolated; instead a salt of 22.was obtained.

This probably is due to elimination from 12 brought about by

the base 22.

46

O
o
o

 

Compound 22_cou1d be alternatively used as the substrate

for introducing the Clo-methyl group. We leave this problem

for further research.

 

.OFD ? —> 6.1:

010”
68

EXPERIMENTAL

General Procedures

 

All nmr spectra were measured in the organic solvents
noted, with tetramethylsilane as an internal standard. The
60 MHz spectra were recorded on a Varian T-60 or A56/60
spectrometer. Ultraviolet spectra were recorded on a Unicam
SP—800 spectrophotometer in the solvent noted. Infrared Spec-
tra were recorded on a Unicam SP-200 spectrophotometer, and
major peaks are reported in units of cm-l. Mass Spectra were
obtained using a Hitachi-Perkin Elmer RMU-6 at 70eV, Operated
by Mrs. Ralph Guile. Elemental analyses were performed by
Spang Microanalytical Laboratories, Ann Arbor, Michigan and

Clark Microanalytical Laboratories, Urbana, Illinois.

2_Acety1—1,4,5,6,7,8-hexamethy1naphtha1ene (2)

 

7

Hexamethylnaphthalene (2, 500 mg, 2 mmol) in CH C12(10 ml)

2
was added slowly to a suspension of AlCl3 (550 mg, 4 mmol)
and acetyl chloride (320 mg, 4 mmol) in methylene chloride
(5 ml). The mixture was stirred for 2 hr at 0 - 5° . The

greenish brown solution was then poured into 10% HCl solution.

The aqueous layer was washed successively with dilute HCl,
47

48

5% NaOH solution (X 2) and water, and was dried (MgSO ).
Chromatography on silica gel with CHC13/hexane (2/1.5§ gave
a 96% yield of the title compound, mp 78-9°. Ir (001“)
2990 (s), 1689(3); mass spectrum (70eV) 254 (M), 100), 239
(9A), 211 (3a), 196 (2A), 181 (27), 165 (25); nmr (001“)
67.21 (1H, s, aromatic proton), 2.70 (3H, s, aromatic methyl
at Cl), 2.62 (3H, s, aromatic methyl), 2.59 (6H, 3, two aromatic
methyls), 2.56 (3H, s, aromatic methyls), 2.36 (6H, 3, one
aromatic methyl and one acetyl methyl).

2222. Calcd for C H O: C, 82.99; H, 8.72. Found:

18 22
C, 85.02; H, 8.71.

Ethyl B-(1,4,5,6,7,8-hexamethyl-2—naphthyl)propionate (22)

 

A three-necked flask which was flushed with nitrogen sev-
eral times, was immersed in an ice-water bath. All chemicals
were injected through a septum in the following order: 0.5 m1
(1 mmol) of 2.1 M butyllithium in ether, 0.15 ml (1 mmol) of
hexamethyldisilazane, 4 m1 of dry THF, and 300 mg (1 mmol)
of 2 in 2 m1 of dry THF. The mixture was stirred for 15 min
at 00 and 0.16 ml (1.5 mmol) of ethyl B-bromoacetate was intro-
duced by injection all at once. The flask was allowed to warm
to room temperature and the contents were stirred overnight.

A white solid (LiBr) precipitated. The reaction mixture was
then diluted with water, extracted with ether and dried (MgSO”).
Chromatography on silica gel with benzene gave 78 mg (22%) of

22 as an oil (the 4th fraction; the previous three fractions

29

were unidentified). Ir (neat) 3000 (s), 1735 (s), 1690 (s),
1310 (s); nmr (CClu) 67.25 (1H, s, aromatic proton), 4.15

(2H, q, J = 7 Hz, methylene on the ethoxy group of the ester),
3.15 (2H, t, J = 6 Hz, d-methylene), 2.70 (3H, s, aromatic
methyl at C ), 2.65 (5H, multiplet, one aromatic methyl and
B-methylene), 2.50 (6H, 8, two aromatic methyls), 2.36 (6H,

5, two aromatic methyls), 1.30 (3H, t, J = 7 Hz, methyl on the
ethoxy group).

28
Saponification and Cyclization of 22

 

Ester 22 (78 mg) in 5% alcoholic KOH (1.5 ml) was stir-
red at room temperature. After 4 hr, 10% HCl (1 ml) was added.
A brown oil separated after the acidic solution was diluted
with water. The organic layer was dried (CaC12). Evaporation
of the solvent gave a brownish oil which solidified on standing
in the refrigerator overnight.

2222. Calcd for C20H24O3: C, 76.89; H, 7.74. Found:

C, 76.70; H, 7.78.

This solid was then added to l g of polyphosphoric acid
which was previously heated to 100°. The mixture was stirred
vigorously while the temperature was raised to 150°. After
15 min, 20 m1 of water was added, and the mixture was extracted
with ether. The ethereal solution was washed with water three

times and then with 5% NaOH solution. The organic layer was

separated and dried over MgSOu. The solvent was removed under

vacuo. Thick-layer chromatography on silica gel with CHC13/

50

benzene (1/1) gave a brown oil (1st fraction) which solidified
(lg) after removing the solvent and chromatography on silica
gel with hexane as the eluent. The yield (for the two steps)
of 22 was 3%. Ir (KBr) 3450 (m), 2950 (s); nmr (CClu) 67.6

(2H, s, aromatic protons), 2.52 (6H, 3, two methyls at C9 and

), 2.40 (6H, s, two methyls at C and 08), 2.35 (6H, s,

010
two methyls at C6 and C7).

5

18
1,4,5,6,7,8-Hexamethyl-2-nitronaphthalene (2Q)

 

(A) Preparation of silver nitrate impregnated silicic

 

£322: Chromatographic silicic acid (1.8 g) was slurried with
absolute methanol (5 ml) and added to a solution of silver
nitrate (0.47 g) in 50% aqueous methanol (3 ml). The resulting
slurry was then evaporated to dryness. The solid was dried
overnight in an oven at 150°.

(B) Nitration of Hexamethylnaphthalene: To the freshly-

 

dried silver nitrate-silicic acid sample was added a solution
of CClu and hexane (1/1, 10 ml). After 15 min, hexamethylnaph-
thalene (5) (0.5 g, 2 mmol) in the same CClu/hexane solvent

(6 ml) was added. The mixture was agitated and allowed to
stand for 64 hr at 24°, then was transferred to a small silica
gel column with ether as the eluent. A brown oil was obtained.
Chromatography on silica gel with hexane/chloroform (2/1)

gave the yellow crystalline 29_(2%, the second fraction; the
first fraction was unidentified), mp 100—10. Mass Spectrum
(70eV) 257 (M’), 225 (15), 206 (67), 181 (H9), 165 (100),

51
152 (48), 141 (37), 128 (37), 115 (50); nmr(CC2‘) 67.45
(1H, s, aromatic proton), 2.73 (3H, s, methyl at 01), 2.68
(3H, s, aromatic methyl), 2.58 and 2.40 (6H each, s, four

aromatic methyls).

1,4,5,6,7,8—Hexamethyl-2-bromonaphthalene (£2)

 

A solution of 212 mg (1 mmol) of 5 in 5 m1 of CS2 was
cooled to -78° in a flask which was wrapped with aluminum
foil. Bromine (0.06 ml, 1 mmol) in 5 ml of CS2 was added
very slowly. After the mixture was stirred at -780 in the
dark for 30 min, 100 mg of sodium bisulfite was added (exother-
mic). Water was added, and the organic layer was washed with
NaHCO3 solution twice and then with water twice. The organic.
layer was dried (MgSO“) and chromatographed on alumina with
pentane as the eluent. Evaporation of the solvent gave a
colorless oil which, after recrystalization from petroleum
ether (bp 30-60°), afforded 22 (155 mg, 54%), mp 68-9°.

Ir (001“) 3080 (w), 3050 (m), 2950 (s), 1580 (m), 1500 (w),
1470 (s), 1450 (s), 1400(m), 1320 (w), 1260 (m), 1210 (w),
883 (s); mass spectrum (70eV) 292 (98), 290 (M+, 100), 227
(32), 275 (33), 211 (24), 196 (23), 181 (23), 155 (27);

nmr (CClu) 67.15 (1H, s, aromatic proton at C3), 2.75 (3H,
s, aromatic methyl at Cl), 2.65, 2.52, and 2.45 (3H each, 3,
three aromatic methyls), 2.30 (6H, 3, two aromatic methyls).

2232. Calcd for C H 98r: C, 66.21; H, 6.55; Br, 27.24.

16 1
Found: C, 66.27; H, 6.39; Br, 27.21.

52
1,4,5,6,7,8-Hexamethyl-2,3-dibromonaphthalene (22)

 

(A) From hexamethylnaphthalene: A solution of 212 mg

 

(1 mmol) of hexamethylnaphthalene in 5 ml of Gszzwas cooled

to -78° in a dark flask. Bromine (0.12 ml, 2 mmol) in cs 2
(5 ml) was added slowly. After 30 min, the mixture was warmed
to room temperature, and an aqueous solution of sodium bisul-
fite was then added. The organic layer was separated and

washed with NaHCO3 and with water several times. After being

dried over MgSO the solution was evaporated to give an oil

4’
which was eluted through an alumina column by 10% ether in
pentane. The first fraction gave 22 (30-40%), mp 174-50.

Mass spectrum (70eV) 372 (50), 370 (100), 368 (M+), 357 (12),
355 (19), 353 (ll), 292 (30), 290 (32); nmr (CD013)<52.70 (5H,
3, two aromatic methyls at C1 and Cu),'2.45 (6H, 3, two methyls
at C5 and 08), 2.30 (6H, 3, two methyls at C6 and 07).

(B) From 2-bromo-hexamethylnaphthalene: The same proce-
dure was followed except that only 1 mmol of bromine was used.
The yield of 22 was 15%.

2322, Calcd for Cl6H18Br2: C, 52.17; H, 4.89. Found:

C, 52.29; H, 4.83.

Reaction of 22 with Sodium Amide in Ammonia

 

To the solution of 22 and 2,5-dimethylfuran in liquid
ammonia was added sodium metal in small pieces. The mixture
was then stirred at dry-ice temperature for six hr and at room

temperature for another six hr. After work-up, only compound

22 was recovered (100%).

53
1,4,5,6,7,8-Hexamethy1-2-naphthoic acid (22)

 

In a dry three-necked flask was placed 85 mg (3.5 matom)
of magnesium turnings and 10 ml of dry ether. A solution of
290 mg (1 mmol) of 22 and 218 mg of ethyl bromide in 10 m1
of dry ether was placed in a funnel. Stirring was commenced
and about t of the halide solution in the funnel was added;
the solution was then heated to 40°. The rest of the halide
solution was added during the course of 1 hr to the vigorously
refluxing mixture. After completion of the addition, reflux
was maintained by heating for another hour. The reaction mix-
ture was then cooled, and a large excess of solid carbon di—
oxide was added slowly in small pieces with rapid stirring.
The resulting gummy addition product was decomposed by adding
with stirring a solution of 1 m1 of concentrated HCl in 2 m1
of water dropwise. After most of the solid had dissolved, the
mixture was transferred to a separatory funnel with the addi-
tion of ordinary ether. The ethereal layer was washed three
times with cold water and dried (MgSO“). Evaporation of the
solvent gave a gummy yellow solid which was recrystalized as
a pale yellow solid in ethanol (125 mg, 50%), mp 192-3’.

Ir (0014) 3000 (s), 2950 (s), 2900 (s), 1680 (s), 1595 (w),
1450 (m), 1430 (m), 1390 (m), 1350 (m), 1270 (s), 1190 (w),
1130 (m), 923 (5); mass spectrum (70eV) 256 (M+), 220 (22),
212 (100), 197 (59), 165 (27), 142 (40); nmr (00013) 67.70
(1 H, s, aromatic proton at C3), 2.90 (3H, 8, methyl at Cl),
2.72, 2.60, and 2.55 (3H each, s, three aromatic methyls),

2.40 (6H, 3, two aromatic methyls).

54
Anal. Calcd for c H 0 - c, 79.65; H, 7.86. Found:

17 20 2'
Ca 79-503 H: 7-91-

1,4,5,6,7,8-Hexamethyl-2-aminonaphtha1ene (23)

 

21
(A) Prgparation of p-tosyl azide: p-Toluenesulfonyl

 

azide was prepared by adding with swirling 700 mg of sodium
azide in 2 ml of water to 1.7 g of freshly—distilled p-toluene-
sulfonyl chloride“3 in 10 m1 of 95% ethanol. After separation,
the oily sulfonyl azide was washed three times with water and
dried (Nazsou).

(B) Naphthylamine: Grignard reagent was prepared as in
the synthesis of 22_from 2.5 g of 22, 0.5 g of ethyl bromide
and 0.9 g of magnesium turnings in 20 ml of dry THF under
nitrogen. The solution turned into light brown indicating
that the reaction had occured. After the mixture was stirred
for 3 hr at 40°, it was cooled in an ice bath and 1.4 g of dry
p-tosyl azide in 5 ml of dry THF was added dropwise with stir-
ring under nitrogen. The resulting greenish solution was stir-
red for four additional hours at 350. In another flask, was
placed 20 m1 of 50% NaOH and a few pieces of ice. To this
basic solution, a small portion of a total of 7 g of Raney
nickel-aluminum alloy was added. After hydrogen started
evolving, the azide solution was added dropwise. The rest of
the alloy was added after the addition of azide, in five por—
tions at 15 min intervals. This mixture was then extracted

with ether. The ethereal solution was washed with water and

 

55

dried over NaOH. Evaporation of the solvent gave a brownish
oil which was diluted with dry ether. An equal volume of dry
ether was saturated with dry HCl. Then this acidic ether was
dropped into the brownish-red ethereal solution slowly. A
pale yellow solid precipitated (0.9g, 40%), which was the am—
monium chloride salt of compound 23, Ir (neat) 4450 (w),
3400 (m), 3000 (s), 1680 (m), 1640 (m), 1600 (s); mass spec-
trum (70eV) 227 (M+, 100), 212 (75); nmr (CS2)66.3 (1H, s,
aromatic proton at 03), 3.40 (2H, 5, amino protons), 2.55
(3H, 5, methyl at C ), 2.50 and 2.35 (3H each, 8, two aromatic
methyls), 2.25 (6H,ls, two aromatic methyls), 2.20 (3H, s,

aromatic methyl).

Alternative preparation of 2£_from 55 via 55

 

To an ice-cold suspension of lithium aluminum hydride
(150 mg) in absolute ether (10 ml) was added with stirring
over 30 min an ether solution (10 ml) of 55 (327 mg). After
the mixture was refluxed for 2 hr, it was poured into ice-water
and extracted with ether. The ethereal solution was dried
(NaOH) and evaporated to give a reddish oil (55) (99%). The
infrared spectrum of this product 55_indicated that both the
carbonyl and the nitro group were reduced: 3500 (s), 1630 (m),
1603 (m). This product was added into the dimslsodium solu—
tion which was prepared from sodium hydride (50% dispersion
in mineral oil, 300 mg) in DMSO (10 ml).23 After 16 hr at
400, the dark brownish-red solution was poured on an ice—water

mixture. After work-up, a reddish oil was obtained in 10%

56
yield; it gave identical spectral data as 25. The same result

was obtained (22%) from the pyrolysis of 55 with benzene as
the solvent at 210-230(3under nitrogen. The pyrolysis was
carried out by dropping a benzene solution of 55 from the top
of a Pyrex hot tube (25 inches) packed with glass beads and

collecting the product with a flask connected with the tube.

1,4,5,6,7,8—Hexamethyl-2-bromo-3-benzoy1naphthalene (55)

 

A solution of 22 (290 mg) in dichloromethane (3 ml) was
added slowly to a suspension of aluminum chloride (0.3 g) and
benzoyl chloride (0.3 g) in dichloromethane (3 ml) at —5rt5).
After one hour at this temperature, the excess aluminum chlo-
ride was decomposed by adding the reaction mixture to 10% HCl
solution. The aqueous layer was extracted with chloroform
and the combined organic layers were then washed successively
with dilute HCl solution, Na2CO3 solution and water, and dried
(MgSO”). Elution through alumina with 20% ether in petroleum
ether gave starting material (lst fraction, 50 mg) and 5§_
(2nd fraction, 150 mg), mp 189-1900 from CHCl and MeOH.

3
Ir (KBr) 3000 (s), 1680 (s), 1600 (m), 1460 (s), 1400 (m),

1260 (s), 1205 (m); mass spectrum (70eV) 396 (100), 394 (M+,
100), 314 (71), 300 (76), 299 (62); nmr (CDC13) 67.8-7.2 (5H,
m, aromatic protons on benzoyl group), 2.6 (3H, s, aromatic
methyl), 2.47 (6H, 5, two aromatic methyls), 2.38, 2.34, and
2.30 (3H each, 3, three aromatic methyls).

5222. Calcd for C H OBr: C, 70.11; H, 5.88. Found:

23 23
c, 69.90; H, 5.75.

57
Reaction of 55 with Sodium in Liquid Ammonia

 

Sodium (150 mg) was dissolved in liquid ammonia. The blue
color first formed and then disappeared after 10 min. Liquid
ammonia was added while adding sodium to keep the volume to

50 ml. After the blue color completely disappeared, a solu-
tion of 55 and 2,5-dimethy1furan (excess) in THF was added
dropwise. The reaction mixture was warmed to room temperature
for 7 hr, and then was quenched by adding ammonium chloride
(300 mg). After ordinary work-up procedures, only a 100% re—

covery of 55 (400 mg) was obtained.

3,6-Dimethyl-4-nitroisatin (52)

 

A solution of 8.75 g (50 mol) of 3,6-dimethy1isatin 55
in 20 ml of concentrated sulfuric acid was cooled to 0°, and
to this solution was added slowly 3.1 g of fuming nitric
acid. After 30 min, the mixture was poured over 200 g of
cracked ice. Compound 52 was obtained as a dark yellow solid
(9.9 g, 90%), mp 249-250°. Nmr (DMSO-d6)68.12 (1H, s, aroma-
tic proton), 2.65 (3H, 8, methyl at C2), 2.24 (3H, methyl at
C6). Compound 52 was used in the next step, to prepare 52)

without further purification.

5,6-Dimethyl—4-nitroanthranilic acid (52)
To the solution of 9.9 g of 52 in 150 ml of 10% NaOH was

added 150 ml of 3% H202 (exothermic). After the H2O2 was

 

58

added, the mixture was heated to 400 for 6 hr. A light yel-
low solid was sparingly precipitated when the mixture was
acidified with concentrated HCl (foaming). The solid was
filtered and washed with water, giving a 99% Yield of 52,
mp 155 (sublimed) from 80% acetic acid. Mass spectrum (70eV)
210 (M+), 175 (100), 164 (48), 118 (70); nmr (CD3CN) 67.65
(1H, s, aromatic proton), 5.30 (2H, s, amino protons), 2.44
(3H, s, aromatic methyl at Cl)’ 2.1 (3H, s, aromatic methyl
at C6).

2&22. Calcd for C H 0 N : c, 51.42; H, 4.80. Found:

9 10 4 2
C, 51.49; H, 4.82.

1,3,3,4,7,8-Hexamethyl-5,6-(3,6-dimethy1-4-nitrobenzo)bicyclo-

 

[2 . 2 . 2]octa-5 , 7-dien-2-one (55)

 

(A) Diazotization of 3,6-dimethyl-4-nitroanthranilic
acid: To a solution of 52 (2.3 g) and HCl (10 ml) in abso-
lute ethanol (30 ml) was added isoamyl nitrite (3 ml) at 4°

The mixture was then stirred at 00 for 2 hr. Absolute ether

 

(40 ml) was added and the resulting mixture was stirred at
00 for another hour. A pale yellow solid (3,6-dimethy1-4-
nitrobenzenediazonium—2-carboxylate hydrochloride, 55) was
obtained (54%), mp 980 (exploded).

(B) Preparation of the title compound: The mixture of
55 (1.285 g), 25 (0.89 g) and propylene oxide (2 m1, added
last) in 15 m1 of ClCH20H201 was heated at 86’ for 2 hr.
Ether was then added (20 ml). The ethereal solution was washed

three times with dilute sodium hydroxide solution and three

59

times with water, and was dried (Mgsq‘). The solvent was
then evaporated under vacuo. The resulting brownish red
oil crystallized upon dilution with a small portion of absolute
methanol to give 0.54 g (33%) of 53, mp 169-1713 . Mass
spectrum (70eV) 327 (M+), 257 (72), 240 (58), 225 (84), 210
(100); nmr (CD013) 67.24 (1H, s, aromatic proton), 2.65 (3H,
s, aromatic methyl adjacent to nitro group), 2.50 (3H, s, aro-
matic methyl), 1.99 (6H, 5, two vinyl methyls), 1.97-1.85
(6H, multiple, two methyls at bridgeheads), 1.12 and 0.85
(3H each, 8, two methyls at C3).
Anal. Calcd for C H O N: C, 73.36; H, 7.70. Found:
-__— 2O 25 3

C: 73ouu; H, 7.63.

1,4,7,8-Tetramethy1-2,3-5,6-di(3,6-dimethylbenzo)bicyclo-

 

[2.2.2]octa-2,5,7-triene (fig)

 

A mixture of 3,6-dimethylbenzenediazonium-2-carboxylate
hydrochloride (55)(1.06 g, 5 mmol), 5 (0.53 g, 2.5 mmol)
and propylene oxide (2 ml) in ClCH2CH2Cl (15 ml) was refluxed
at 80 - 963 for 1.5 hr. Evaporation of the solvent gave a
dark brown oil which was redissolved in ether and washed with
dilute NaOH solution, with water, and dried (MgSO”). Evapora—
tion of the solvent gave an oil which was diluted with a small
portion of ether and cooled in ice. The solid which formed
was filtered and recrystallized from chloroform and ether

proved to be the dimer of dimethylbenzyne. The filtrate was

subjected to column chromatography on silica gel with cyclohexane

 

60
as the eluent. The first fraction was recovered 5 (400 mg).

The second fraction was an unidentified material with low
yield. The third fraction was a white solid which gave,
after recrystallization from chloroform and petroleum ether,
0

40 mg (5.1%) of 55, mp 160 (sublimed). Mass spectrum (70eV)

+
316 (M ), 301 (100), 286 (82), 271 (61); uv (CH CN) 1 ( e)

3 3 3 max
232 nm (14.2 X 10 ), 288 (shoulder, 4.2 X 10 ), 215 (shoulder,
3

11.5 X 10 ); nmr (CDC13) 66.4 (4H, s, aromatic protons),
2.55 (12H, 5, four aromatic methyls), 2.45 (6H, 3, two methyls
at bridgeheads), 1.75 (6H, 5, two vinyl methyls).

Anal. Calcd for 024H28: C, 91.08; H, 8.92. Found:
C, 90.33; H, 8.93. This analytical data was bad due to the

contamination of 55 by 5.

1,2,5,6-Tetramethyl-3,4-7,8-di(3,6-dimethy1benzo)cycloocta-

 

tetraene(52)

 

(A) via Byrolysis: A solution of 50 mg of 55 in 5 ml of

 

benzene was flashed under nitrogen through a hot quartz tube
at 600°. A dark brown oil was obtained in the receiver.
Column chromatography on silica gel with cyclohexane as the
eluent gave a pale orange colored product 52 (SO-60%), mp 174-6o
from absolute methanol. Mass spectrum (70eV) 316(M+), 301 °
(100), 286 (75), 271 (73), 256 (35); nmr (CD013) 65.59 (“H,
s, aromatic protons), 2.08 (12H, 5, four aromatic methyls),
1.91 (12H, 5, four vinyl methyls).

5522. Calcd for C H : C, 91.08; H, 8.92. Found:

24 28
C, 90.89; H, 8.71.

61
(B) via Photolysis: A solution of 50 mg of 55 in 25 ml

 

of ether was purged with nitrogen in a 2 cm X 15 cm Quartz
test tube, which was then sealed and irradiated using a 456
watt Hanovia lamp with a Vycor filter for 45 min. The major
product was separated by preparative vpc (5' X k" 20% SE-30

on DMCS at 2163 and identified as 52. Two more products which
had longer retention time and were formed in low yields were

unseparable by vpc and were not identified.

2,2,5,8-Tetramet5yl-3,4-6,7-di(3,6-dimethylbenzo)tricyclo-
2 8
[3.2.1.0 ’ Jocta-3,6-diene (55)

 

A solution of 50 mg of 55 was dissolved in 20 ml of
acetone. The acetone solution was degassed with nitrogen and
irradiated using a 450 watt Hanovia lamp with a Vycor filter
for 1 hr. Two major products were separated by vpc (5' X t"
10% SE-30 on DMCS at 206°). The chromatogram revealed a
100% conversion of the starting material to compounds 52
(40%) and 55 (50%) at retention times of 20 min and 30 min
respectively. Compound 55 was recrystalized from methanol.
Mass spectrum (70eV) 316 (M+), 301 (100), 286 (59), 271 (45);
nmr (CD013) 66.45 (4H, s, aromatic protons), 2.45 and 2.20
(6H each, 3, four aromatic methyls), l.95(3H, s, methyl at C3),

1.52 (6H, 3, two methyls at C and 02), 1.19 (3H, 3, methyl

1
at C8).

 

 

62

1,4,5,8-Tetramethyl-l,4-dihydronaphthalene-l,4-endoxide (55)

To 100 ml of C1CH20H2C1 was added 23.2 g (0.14 mol) of
55, 34 m1 of propylene oxide and 59.7 ml (0.56 mol) of 2,5-
dimethylfuran. The mixture was refluxed at 83° for 2.5 hr.
Evaporation of the solvent gave a red brown oil which was then
redissolved in ether. The ethereal solution was washed with
dilute NaOH and with water. The product 55 (15.5 g, 55%
based on 55) was distilled at 93° (0.6 torr). Ir (neat) 3000
(s), 1500 (s), 1460 (s), 1400 (s), 1305 (s), 1150 (3); mass
spectrum (70eV) 200 (M+), 174 (21), 157 (100), 142 (52);
nmr (CClu) 66.65 (2H, s, aromatic protons), 6.50 (2H, 3, two
vinyl protons), 2.25 (6H, 3, two aromatic methyls), 1.87 (6H,
3, two methyls at C1 and C4)°

5522. Calcd for C H O: C, 83.96; H, 8.05. Found:

14 16
C, 83.77; H, 8.17.

1,4,5,8-Tetramethyl-2-methoxynaphthalene (52), and l-Methoxy-

 

methyl-4,5,8-trimethylnaphtha1ene(52)

 

 

A solution of 200 mg (1 mmol) of 55 in 5 m1 of absolute
MeOH with 1 drop of concentrated HCl was heated at 60-65° for
2 hr. After work-up, 130 mg of crude product was obtained.
Column chromatography with 1% ether in petroleum ether (30—60°)
on alumina gave two products. The first fraction was 52
(5 mg, 2.3%), mp 54-56°. Ir (KB:) 2950 (s), 1600 (s), 1110

(3); mass spectrum (70eV) 214 (M , 100), 199 (32), 171 (44),

156 (32); nmr (CDC13) 66.80 (3H, s, aromatic protons), 3.80

53

(3H, s, methoxy methyl), 2.80 (3H, s, methyl at Cl)’ 2.75

(6H, 3, two aromatic methyls), 2.55 (3H, s, aromatic methyl).
5222, Calcd for 015H18O: C, 84.07; H, 8.47. Found:

C, 84.09; H, 8.44.

The 2nd fraction was 52 (140 mg, 64.4%), mp 48-50°.
Ir(KBr) 3000 (s), 1870 (w), 1600 (s), 1105 (s); mass spectrum
(70eV) 214 (M+), 182 (65), 167 (100), 152 (30); nmr (001”)
66.8-7.0 (4H, multiplet, aromatic protons), 4.60 (2H, s,
methylene protons at Cl), 3.20 (3H, s, methoxy methyl), 2.75

(9H, s, three aromatic methyls).

1,4,5,8—Tetramethylnaphthalene (55)

 

(A) From 55 and Lithium Naphthalenide: The stock solu-

 

tion of lithium naphthalene radical anion was prepared in
1,2-dimethoxyethane (DME). Small strips of lithium wire

(100 mg, 14 mmol) were added to a solution of naphthalene

(1.8 g, 14 mmol) in 30 ml of dry solvent under an argon atmos-
phere. The blue-green color of the radical anion appeared
within 2 hr. The solution was stirred (glass-covered stir-
ring bar) overnight. To this solution, 55 (100 mg, 0.5 mmol)
in DME (10 ml) was injected from a syringe. The mixture was
stirred at ambient temperature for 2 hr. The dark green
solution was decomposed with an iodine crystal, and the
resulting colorless cloudy suspension was evaporated in vacuo.
The residue was suspended in ether and washed with 1 N sodium

thiosulfate solution and water. The solution was then dried

64
O 27
(MgSO“) and evaporated to give 100% of 55, mp 131-2 (lit.
132-20). Mass spectrum (70eV) 184 (M+, 100), 169 (96); nmr
(CD013) 66.90 (4H, s, four aromatic protons), 2.79 (12H, 3,

four aromatic methyls).

(B) From 55 via 55: To a solution of 200 mg of 55

 

in 10 ml of methanol was added 100 mg of hydrazine hydrate
and a trace of cupric sulfate. Hydrogen peroxide (30%,
500 mg) was then added dropwise at room temperature. The
reaction was worked up after 1 hr to give 202 mg of 55.
Nmr (CDC13) 66.60 (2H, s, two aromatic protons), 2.30 (6H,
3, two aromatic methyls), 1.40-1.80 (4H, multiplet, four
protons at C2 and C3).

55 was then redissolved in 10 ml of absolute methanol,
and to this solution 2 drops of concentrated HCl was added.
The solution was refluxed for 12 hr. After work-up, the
crude product was chromatographed on alumina with cyclohexane

as the eluent to give 110 mg (60%) of 55.

1,3,3,4,5,6-Hexamethyl-7,8-(1,4-epoxy-1,2,3,4-tetrahydro-

naphtho)bicyclo[2.2.2]-5-ene-2-one (55)

 

A mixture of 1.44 g of 1,4-dihydronaphtha1ene 1,4-endo-

28 and 1.78 g of 25 in 20 m1 of chlorobenzene was re-

oxide
fluxed for 6.5 hr. After evaporation of the solvent, a little
ethanol was added and the resulting liquid was kept in the
refrigerator. A colorless crystalline solid was obtained

which was recrystallized from 50% aqueous MeOH at room

65

temperature to give a 50% yield of 55, mp 123-7°. Ir °°°l4°

3003 (s), 1710 (s), 1470 (m), 1400 (m), 1030 (m); mass spec-
+ .
trum (70eV) 322 (M ), 234 (48), 134 (85), 118 (100); nmr

(CClu) 67.1 (4H, s, aromatic protons), 5.10 (2H, d, J = 5 Hz,
endoxide bridgehead protons), 2.30—1.60 (8H, multiplet, pro-

tons at C and two vinyl methyls), 1.32 (6H, s, methyls

7’ C8

at C1 and C4)’ 0.90 (3H, s, geminal methyl), 0.85 (3H, s,

geminal methyl).
Anal. Calcd for C
c, 82.14; H, 8.18.

22H2602: C, 81.95; H, 8.13. Found:

2,4,5,8,9,10-Hexamethyl-4a,9a-dihydroanthracene-1,4-9,10-

 

diendoxide (55)

 

A solution of 2 g (10 mmol) of 55_and 6 ml (50 mmol) of
2,5-dimethylfuran in 10 m1 of diglyme was refluxed at 150-
l60"for 72 hr. After work-up, a crystalline solid was
obtained (1.8 g, 60%), mp 100-5° from petroleum ether. Ir
(CClu ) 3050 (s), 1400 (s), 1220 (w), 1200 (w), 1150 (m);
mass spectrum (70eV) 296 (M+ ), 200 (74), 173 (87), 157 (100),
142 (68), 128 (61), 115 (59), 106 (56), 96 (98); nmr (00013)
66.85 (2H, 3, two aromatic protons), 6.45 (2H, 3, vinyl protons),
2.40 (2H, s, protons at 04a and 09a), 2.32 (6H, s, aromatic
methyls), 1.73 (6H, s, methyls at C

and C 1.60 (6H, s,

methyls at C1 and C 4°°

Anal. Calcd for C20H24O2: C, 81.04; H, 8.16. Found:

C, 81.18; H, 8.04.

 

66
Photolysis of 55

 

A solution of 296 mg of 55 in 10 ml of anhydrous ether
was degassed with nitrogen and irradiated with a Corex filter
for 8 hr (450 watt lamp). A fine crystalline product (55)
which deposited in the form of plates on the test tube was
filtered (30-35 mg, 5-6%), mp higher than 350°. Mass spectrum
(709V) 592 (M+), 175 (18), 174 (100), 173 (12), 159 (7); 1P
(KBr) 3000 (s), 2960 (s), 1500 (m), 1470 (s), 1400 (s), 1260
(s), 1240 (s), 1170 (m), 1160 (m), 1120 (s), 920 (m), 880 (s),
860 (s), 820 (s), 770 (3). Compound 55 was insoluble in

most organic solvents (CHCl DMSO, acetone, etc.).

3,

5222. Calcd for C40H48O4° C, 81.04; H, 8.16. Found:

C, 81.10; H, 8.27.

The filtrate was evaporated to dryness and chromatographed
on alumina with ether and petroleum ether (1/1). Two fractions
were collected. The second fraction proved to be unreacted
55 (10%). The first fraction was crystallized after distil-
lation under vacuo and gave the same spectral data and mp

as compound 55 (17.5%).

1,4,5,8,9,10-Hexamethy1-2,3,4a,9a-tetrahydroanthracene-l,4—9:

 

10-diendoxide (55)

 

To a solution of 55 (700 mg), hydrazine hydrate (1.5 g)
and a trace of cupric sulfate in 7 m1 of MeOH was added 5.04 g

of 30% H202 in 7 ml of MeOH while the temperature was kept

 

O 67

at 0 - 5 . After being stirred at room temperature for

30 min, the reaction mixture was diluted with water and
extracted with ether. The ethereal solution was dried (MgSO“)
and evaporated (under vacuo). The remaining oil was kept

in the refrigerator overnight, and crystalline 55 was obtained
in 70.9% yield, mp 88-89° . Ir (CCl ) 3020 (s), 1510 (m),
1480 (m), 1400 (s), 1360 (w), 1280 (3), 1260 (m), 1185 (w),
1120 (m); mass spectrum (70eV) 298 (M°), 154 (69), 152 (90),
139 (68), 137 (100); nmr (CClu) 66.63 (2H, s, aromatic protons),
2.30-2.31 (10H, multiplet, two aromatic methyls and four
protons at C2 and 03), 2.08 (2H, s, protons at C4a and 09a),
1.85 (6H, s, two methyls at C and C ), 1.40 (6H, 5, two

9 10

methyls at c and C4)“

1
Anal. Calcd for C20H2602: C, 80.49; H, 8.78. Found:
C, 80.51; H, 8.74.
The same result was obtained by using Pd-C as the cata-

lyst under hydrogen atmosphere at room temperature. The con-

figuration of 55 was determined by X-ray analysis.

 

 

2,4,5,8,9-Pentamethyl-10-methy1ene-2,3,4a-trihydroanthracene-

1,4-endoxide (51)

 

A solution of 500 mg of 55 with five drops of HCl (or
HBr) and a trace of iodine in 10 m1 of MeOH (absolute) was
refluxed at 750 for 126 hr. The mixture was diluted with
ether and washed with sodium bicarbonate and with water sev-
eral times. The organic solution was dried (MgSO“) and eva—

porated (under vacuo). Column chromatography (Florisil, 5%

68
ether in hexane) of the residue afforded 150 mg (30%) of

O
51 (the lst fraction), mp.110-2 from petroleum ether (30 -
o
60 ). Ir (001”) 3000 (s), 2950 (s), 2900 (s), 1500 (m),
+
1180 (s), 1130 (s), 960 (s); mass spectrum (70eV) 280 (M ),

279 (39), 175 (76), 175 (100), 173 (39); nmr (CClu) 66.70
(2H, 3, two aromatic protons), 5.50 (1H, broad, vinyl proton),
4.95 (1H, broad, vinyl proton), 2.20 (6H, s, methyl at C9),

1.60 (6H, 8, two methyls at C and C4)'

1

Anal. Calcd for C2OH24O° C, 85.66; H, 8.63. Found:

C, 85.31; H, 8.35.

1,4,5,8,9-Pentamethyl-lO-methylene-9-monohydroanthracene (55)

 

To a solution of 524 mg (2 mmol) of triphenylphosphine
in 3 ml of DMF was added 508 mg (2 mmol) of iodine in 1 ml
of DMF slowly under nitrogen at -5 ~0° . The temperature was
raised to 70° and 298 mg (1 mmol) of 55 was added in one
portion. After 2 hr at this temperature, 108 mg of NaOMe was
added. The reaction was continued for 4.5 more hours at 1354.
140°. The reddish-brown solution was extracted with pentane,
and the pentane solution was washed with NaHCO3 solution and
then with water. The solution was dried (MgSO”) and eluted
through a short column on Florisil using pentane as the eluent.
Evaporation of the solvent gave a fine crystalline solid (50mg,
20%) which was recrystallized from ethanol. Uv (pentane)
°max(°) 289 (sh.) (0.82 X 103), 252 (1.3 X 10°); mass spectrum
(706V) 262 (M ), 247 (100); nmr (CDC13) 67.0 (4H, s, aromatic

protons), 5.62 (2H, 5, vinyl protons), 4.40 (1H, q, J = 6 Hz,

 

69

C9-proton), 2.55 and 2.40 (6H each, s, four aromatic methyls),

1.30 (3H, d, J = 6 Hz, C9—methyl).

38
The Preparation of 2-(2-Methyl[l,3]dioxolan-2-y1)pyridine (25)

A mixture of 7 g (57 mmol) of 2-acetylpyridine, 11 ml
(200 mmol) of ethylene glycol and 5 g of p-toluenesulfonic
acid in 150 ml of benzene was refluxed for 64 hr in an ap—
paratus provided with a modified Dean-Stark water separator.
The reaction mixture was then poured into concentrated Na2CO3
solution, and the benzene layer was separated. The aqueous
layer was extracted four times with benzene, then the combined
organic layers were washed once with water and dried over

MgSOu. Benzene was removed under vacuo, and the residue was

0
distilled at 70 (0.5 torr). The yield of 55 was m70%.

2,4-Dimethyl-2-bromomethylbenzene ( 6)

 

Dry HBr was bubbled through a solution of paraformaldehyde
(12 g, 4 mol) in glacial acetic acid (150 m1) until the solu—
tion turned clear. p-Xylene (10.6 g, 0.1 mol) was added
dropwise to the above solution at 95°. The mixture was then
stirred at this temperature for 2 hr, and the reaction was
stopped by pouring the mixture onto ice-water. The water
solution was extracted with ether several times. The com-

bined ethereal solution was washed with NaHCO solution twice,

3
with water once, and was dried (MgSO“). Two products were

obtained from column chromatography (Florisil, hexane). The

 

70
lst fraction gave 55 (7.7 g, 40%), bp 57°(0-l torr). Mass
4.
spectrum (70eV) 200 (7), 198 (M ), 120 (17), 119 (100), 117
(8), 105 (14), 91 (17), 77 (9); nmr (00013) 56.94 (2H, multiplet,

aromatic protons at C and C6), 6.90 (1H, s, aromatic proton

5
at C3), 4.38 (2H, s, methylene protons), 2.30 (3H, 5, methyl
at Cl), 2.25 (3H, 5, methyl at C4)'

The 2nd fraction gave 2,5-bis-bromomethyl-p-xylene
which was recrystallized from methanol, mp 145-6°(this compound
has lachrymatory vapor). Mass spectrum (70eV) 294 (2), 292
(4), 290 (M+), 213 (36), 211 (37), 132 (100); nmr (00013)
67.0 (2H, s, aromatic protons), 4.37 (4H, s, methylene protons),
2.30 (6H, s, two aromatic methyls).

This bis-bromomethyl compound (1 g) was dissolved in
THF (20 ml), which was added slowly into a mixture of LAH
(500 mg) and THF (10 m1) at room temperature. Then the resulting
mixture was refluxed over night. After work-up, the oily re-
sidue was chromatographed on alumina with hexane as the eluent.

The first fraction was durene (50%); the second fraction was

the unreacted bis-bromomethyl compound (50%).

1-(2,5-Dimethylbenzy1)-2-(2-methy1[1,3]dioxolan-2-yl)pyridinium

 

Bromide (51).

 

The quaternization of 3.30 g of 55 by reaction with 3.96 g
of 55 in the presence of 4 m1 of dry tetramethylene sulfone
was carried out at 64° in a sealed flask. After 6 days, the
reaction mixture was cooled to room temperature. The resulting

viscous yellow oil was diluted with ethyl acetate. A white

 

71
solid formed. This solid was filtered and washed repeatedly
with ethyl acetate to give 5.50 g (76.5%) of 51, No parent
peak was shown in the mass spectrum. The base peak was at
m/e 198 which is the molecular weight of 55. Nmr (CDC13)
68.90-8.0 (4H, multiplet, aromatic protons on pyridinium
ring), 7.0 (1H, s, proton on p-xylene ring), 6.42 (1H, s,
proton adjacent to methylene on p-xylene ring), 6.30 (1H, s,
proton on p-xylene), 4.0 (4H, multiplet, four protons on
dioxolan), 2.25 (3H, s, aromatic methyl), 2.20 (3H, s, aro-
matic methyl), 1.80 (3H, s, acetyl methyl on pyridinium

ring).

7,10,ll-Trimethylacridizinium Perchlorate (55)

 

A solution of 4.2 g (11 mmol) of 51 in 20 m1 of HBr (48%)
was stirred at 1200 for 12 hr. HBr was removed under vacuo
(aspirator). The remaining yellow syrup was then dissolved
in methanol and cooled in ice. A yellow solid precipitated
upon the addition of 35% of perchloric acid. This yellow
solid was filtered and washed with methanol. Recrystalliza-
tion from acetonitrile and ether gave 3.2 g (100%) of yellow
crystalline 55, mp higher than 265° (turned dark).

5222. Calcd for C H NClO : C, 59.72; H, 5.01. Found:

l6 l6 4
C: 59.69; H, 5.00.

 

72

1,4,5,8,9—Pentamethylanthracene (Z2)

 

A solution of 2.4 g (7.4 mmol) of recrystallized 55
in 100 ml of refluxing acetonitrile was cooled to room temper-
ature. One third of 1.7 g (8 mmol) of freshly—prepared 3,6—
dimethylbenzenediazonium-2-carboxylate hydrochloride (55) was
added. The resulting dark solution was stirred at 45° and
then the rest of 55 was added portionwise. After the mixture
was refluxed for 1 hr, it was cooled to room temperature and
diluted with 50 ml of ether—petroleum ether (2/1) mixture.
The solid which deposited as the mixture stood in the refri-
gerator was filtered and washed with ether. This solid was
added to a mixture of 700 mg of NaOMe and 400 mg of NaBHu in
30 ml of methanol at room temperature. After 30 min standing
at room temperature, the clear part of the mixture was added
to water. The resulting white opaque solution was acidified

with concentrated HCl, then neutralized with NaHCO and

3,
finally extracted with ether several times. The combined
ethereal solution was dried (MgSO“) and evaporated to dryness.
The oily residue was dissolved in 25 m1 of acetic anhydride,

and 400 mg of NaOAc (anhydrous) was added. The mixture was

then refluxed at 145° for 1 hr. Water was added to the reac-
tion mixture, which was then filtered and redissolved in ether.
The ethereal solution was washed with NaHCO3 solution and

water, and was dried (MgSO“). Evaporation of the solvent af—
forded 0.5 g (27%) of 22 which was recrystallized from methanol,
mp 158-159°. UV (CH3CN) °max (e) 224 (1.6 X 10°), 267 (6.2

x 10°), 368 (3.5 x 103), 386 (4.2 x 103), 407 (3.6 x 103);

 

73
ir (KBr) 3000 (s), 1830 (w), 1620 (m), 1470 (s), 1400 (w),
1350 (w), 1050 (w), 880 (s), 840 (5); mass spectrum (70eV)
248 (M+), 233 (67); nmr (CD013) 68.25 (1H, s, aromatic pro—

ton at C10), 7.02 (4H, s, aromatic protons at C , C , C

2 3 6
C7), 3.02 (3H, 3, methyl at C9), 2.80 (6H, 3, two methyls at

, and

Cl and C8 ), 2.70 (6H, 8, two methyls at C4

Anal. Calcd for C19H20: C, 91.88; H, 8.12. Found:

11d C .

1,4,9-Trimethy1anthracene (l2)

 

The same procedure of the preparation of Z2_was followed,
using 55 and benzenediazonium-2-carboxylate hydrochloride as
starting materials. Compound 12_was obtained in comparable
yield to that of 12 and recrystallized from methanol, mp 80-

O O
81 (lit. 11 81 ). Nmr (CD013) 68.20 (1H, s, C —proton),

10
5, c6, c7, and 08),
6.97 (2H, s, protons at C2 and C3), 3.15 (3H, s, C

8.15-7.20 (4H, multiplet, protons at C

9-methyl),

2.85 (3H, s, C -methyl), 2.63 (3H, s, C -methy1).

1 4

1,4,5,8,9-Pentamethylanthracene-9,lO-endOperoxide (15)

 

Compound 12 (50 mg) was dissolved in cyclohexane (16 ml).
The solution was then irradiated (Hanovia 450 watt lamp) with-
out being degassed, with a Pyrex filter for 3 hr. After eva-
poration of the solvent, 15 was obtained in 95% yield (100%

conversion), mp 210-212°from methanol. Ir (KBr) 2950 (s),

74

1503 (s), 1460-1480 (broad), 1390 (s), 1090 (s), 840 (s),
800 (s), 750 (s); mass spectrum (70eV) 280 (M+), 265 (14),
264 (14), 250 (20), 249 (50), 248 (100), 247 (22), 233 (50);
nmr (CDC13) 6.75 (4H, s, aromatic protons), 6.25 (1H, s,
proton at C10), 2.48 (6H, 5, two methyls at C1 and C8), 2.38
(6H, 8, two methyls at C4 and C5), 2.29 (3H, s, methyl at C9).

5222, Calcd for 019H2002: C, 81.39; H, 7.19. Found:

C, 81.64; H, 7.34 (corrected for 2.52% of ash).

 

PART II

MISCELLANEOUS

75

76
A. Wagner-Meerwein Rearrangement in the Dibenzobicycloocta-

diene System

 

During attempts to synthesize highly methylated anthra-
cenes, dibenzobicyclooctadiene (55) was synthesized (see
Part 1, Route III). It was hoped that after epoxidation,
11 would rearrange to ketone (112) which in turn would photo-

chemically eliminate the bridge2° or thermally eliminate an

oxirene to give the highly methylated anthracene.

 

A facile epoxidation of 55 with metachloroperbenzoic
acid (MCPBA, washed with pH 7 buffer solution) in methylene
chloride gave a 65% yield of epoxide 11_within seconds.
Compound 11 was recrystallized from methylene chloride-hexane.

After standing at room temperature for six days, a solu—
tion of ZZ_in chloroform was found to contain no Zl_but a
new compound 15 which gave a correct analysis for C24H280

(isomer of 11). The uv spectrum revealed a maximum at

77
Amax°3° nm (2.28 X 10°) with shoulders at 308 (103). 283
(3.4 X 103) and 265 (7.1 X 103); the infrared spectrum indi—
cated the presence of a hydroxyl group (3500 cm-l). The nmr
spectrum gave two singlets for two vinyl protons at 55.45 and

5.20. Other nmr data with europium shift numbers in parenthesis

are shown on the proposed structure.

115(10)110(36)

CHCl

 

r.t., 6 days

 

or reflux
12 hr

This rearrangement process was also achieved by reflux-
ing a solution of Zl_in chloroform for 12 hr or by injection
of 11 on a gas chromatograph (15% SE—30, 2303. It was fur-

ther observed, by following the rearrangement at room temper-

 

ature with nmr, that after 24 hr a completely new set of peaks
appeared different from those of 11_or 15. However, this
intermediate could not be isolated; all attempts to isolate
and purify it gave the final product 15.

To study the above rearrangement more easily, a simpler
system which would give the same reaction sequence but exhibit
less complicated nmr spectra was desired. Therefore compound
52_was synthesized from benzyne and l,2,3,4-tetramethylnaph-

thalene which was prepared by following Havsigk's procedure.1°

 

Using the same conditions as for the epoxidation of 55,
compound 52 was transformed to 52 in 72.4% yield. The nmr
data of 52 and 52 are shown in the formulas. At reflux
temperature for 4.5 hr, a chloroform solution of 52 afforded
a nicely crystalline compound 55 (100%) which had its °maxat

266 nm (9.3 X 103) and 228 (1.54 X 10°). The infrared spec-

trum showed strong absorption at 1703 cm-1

, indicating the
presence of a carbonyl group; the nmr Spectrum consisted of
one multiplet between 67.0-7.4 for the eight aromatic protons,
one singlet at 62.20 for the two vinyl methyls, and two equal

singlets at 61.93 and 1.80 for the Cl- and acetyl methyl

 

groups respectively.

MCPBA
CH2C12
r.t.

72.4%

 

 

79

 

..-... Q Qbo

    

 

CHCl3
reflux,12 hr 55
CHCl3
reflux 100%
4.5hr reflux,4hr
88.3% 0
°: excess MCPBA 00 .
CH2012 , r. t.
2.. 99%
LAH/m F 98% 86
OH
85

The presence of carbonyl and olefin functional groups
was supported by the following experiments:

Compound 55 was reduced by LAH in THF at room tempera—
ture to give the alcohol 55. Its infrared spectrum showed
no carbonyl absorption but a strong band at 3480 cm—1. The

nmr spectrum of 55 showed one quartet at 65.0 (J = 6 Hz) for

the carbinyl proton and one doublet at 50.85 (J = 6 Hz) for

80

the methyl group adjacent to the carbinol group. When com-
pound §§_was treated with an excess of MCPBA in methylene
chloride at room temperature, it yielded the epoxide §§_(99%).
The structure of gg was confirmed by its nmr spectrum (bands
due to the vinyl methyls disappeared and a new singlet was
shown at 61.70 for the two methyls attaching to the epoxide
ring), the C-O-C absorption at 1180 cm-1, and the parent
peak (m/e 292) in its mass spectrum. A tetramethyl analogue
80, was also obtained by the reaction of flg_and 85% MCPBA at
room temperature for 15 hr. The products were separated by
vpc. Compound 12 was obtained in 70% yield, and 89 in 30%

yield (for its nmr spectrum see page8h).

 

O
OH
8
iP. 5% MCPBA ; O +
r.t., 15 hr < :)
O
12 80

No further rearrangement of §§_was observed on prolonged
reflux of its chloroform solution for 12 more hours. Isomer
§fl_was obtained only by refluxing a solution of §g_(or 8;)
in chloroform with a trace of p-toluenesulfonic acid for
12 hr (N hr for 8;). The yield of g3 was 88.3% accompanied
with a 10.3% recovery of 8;, Compound 85 gave a correct

analysis, and its nmr spectrum had multiple peaks between

81
67.50-6.95 for the eight aromatic protbns, two equal singlets
at 65.55 and 5.15 for the two vinyl protons, and three equal
singlets at 61.70, 1.58 and 1.0 for the 01’ C5 and C8 methyls
respectively.
On the basis of this information, a proposed mechanistic

path rationalizing these skeletal rearrangements is shown in

Scheme 10.

W

 

H“

 

 

 

Scheme 10

82
The first three steps (88_+_8;) are consistent with the
44
results of previous work by Cristol in 1960, who found that

epoxide 81 was converted to aldehyde 8§_on heating.

 

Since then, the same research group has published a series
of papers concerned with the rearrangement of the above
system. The existence of the intermediate 818 has been con-
firmed by trapping with halide or acetate anions.”5 According
to our results, the phenyl migration step can be supported
by the fact that 11 rearranged to 18 much faster than 8g_to
8;, because the migratory aptitude of the migrating phenyl
group was enhanced by methyl substituents. Moreover, a
thermal reaction of 8g_in pyridine was conducted at 65-75° to
give a total recovery of 8g, the absence of rearrangement
product 8; implies that the rearrangement was induced by
traces of acid present in the solvent.

The next steps t°.§£ are first a double-bond participation
to regenerate the carbonium ion 888 and then loss of one

proton to form the exocyclic olefin.

 

83
An energy diagram can be drawn to help visualize the
above rearrangement. The activation energy between 828 and
88 must be higher than that between 828_and 8;, The reason
why 888 did not rearrange to 88 in the first place can be
explained if the conversion of 888 to 8; is kinetically
controlled whereas that of 888 to 88 is thermodynamically

controlled.

 

The double bond participation was supported by the reac-

tion shown below.

 

84
On treatment with p-toluenesulfonic acid, a solution of 88
in benzene turned pink. After 3 hr at reflux temperature,
a mixture of two isomeric alkenes was obtained (828 and 8g8).
They were separated by preparative vpc (5% SE-30 on Chromo-
sorb w at 160°) with 53.33% and 10.66% yields for the first
and second fractions respectively. The nmr spectrum of the
first isomer had one multiplet at 66.90-7.60 for the aromatic
protons, two equal singlets at 65.60 and 5.10 for the vinyl
protons, another multiplet at 62.10 for the 08-proton, two
singlets at 61.70 and 1.56 for C - and C -methyls respectively,

1 5

and one doublet (J = 6 Hz) at 60.76 for C -methy1; the nmr

8
spectrum of the 2nd isomer showed one multiplet at 66.80-

7.10 for the aromatic protons, two equal singlets at 52.0

and 1.86 for the methyls at C and C5 respectively, another

1
multiplet at 61.30-1.70 for the proton at CB, and one doublet

(J = 6 Hz) at 60.82 for the methyl at C No configurational

8.
assignment has been made.
The nmr spectrum of 8§_is consistent with its symmetrical

structure, but that of 18 is not. By using a space-filling

 

model, it can be seen that interference between methyl sub-
stituents has extremely distorted the symmetry of the molecule.

The nmr spectrum of 88_shows a similar perturbation.

2.h5

2.20 2.40

   

Figure 1. Models of 1-acetyl-2,3-6,7—dibenzo-l,u,5-
trimethylcyclohepta-Z,“,6-triene (8;): A. methyl group
axial; acetyl cannot be added to the model in the equa-
torial position; B. asymmetric conformation with the acetyl
group axial; this is the predominent conformation at low
temperature; C. two symmetric conformations with the acetyl
group axial; each represents a possible transition state
for the interconversion of enantiomeric asymmetric conformations,

85a

 

 

86
#6
According to a 1975 paper, Cl-substituted cyclohepta-
trienes have two isomeric structures; that is, the substi-
tuent may be in either an equatorial or an axial position.
This gives us another possible explanation for the unsymmet—
rical nmr spectrum of 18, which is the possibility of confor-

mational interconversion.

 

X
H
/
E X H
Equatorial Axial

However, this possibility can be ruled out by the fol-
lowing arguments: (1) 01113“7 and his coworkers reported
numerical data for nonbonded interaction energies in their
study of the energy barriers to conformational inversion of

the system shown below.

11 Kcal/mole

21.7 Kcal/mole
Me200H

 

2
7 Kcal/mole 21.2 Kcal/mole

 

87
In light of the above result, a higher activation energy was
expected for the conformational inversion of our system.
(2) A study of the low temperature nmr spectra of 8;_showed
that at 10°, the sharp singlet due to the two vinyl methyls
began to Split and eventually two sharp singlets (with a
difference of 8 Hz) were observed at -20°. If this split
was caused by conformational interconversion (that is, at
low temperature, the flipping of the seven-membered ring
slowed down so that both conformers were observed), two sets
of peaks for the Cl- and acetyl methyls should also have ap-
peared in the spectrum. However, although the first singlet
(vinyl methyls) split, the other two singlets (Cl' and acetyl
methyls) still remained sharp. This result also revealed that
it was not only a plane of symmetry but also the free rota-
tion of the acetyl group, which caused a symmetric nmr spec-
trum of 8; at room temperature. As the temperature was low-
ered, the rotation became hindered, and the acetyl group was
squeezed out of the plane of symmetry by its crowded sur-
roundings to give an 8 Hz difference between the two vinyl
methyls. The acetyl group of 18 which has an even more
crowded structure, stOpped its rotation at room temperature
to give a comparable difference (8 Hz) between the vinyl
methyls.

It was also observed that at much lower temperatures
(~70m950), the peak due to the quaternary methyl of 8§_broad-
ened, indicating that its rotation was probably also hindered.

Now, we know that the flipping of the seven-membered ring

did not occur and only one of the two conformers of 8; (or 18)

 

.J I. 111‘
vii-ll I!!! III"

88
was observed. Since an axial acetyl group is required by
the mechanism of its formation (Scheme 10, page 81), the
total conversion of 18 (or 8;) to lg (or 88) by double bond

participation can prove that the acetyl group is axial.

 

B. Non-Conventional Electrgphilic Aromatic Substitutions of

Octamethylnaphthalene

A recent review article!48 contained an extensive discus-
sion of non-conventional electrophilic aromatic substitution
and related reactions. This type of reaction can be described
as an electrophilic reaction of aromatic compounds in which
side-chain substitution is involved. For example, hexamethyl-
benzene and molecular chlorine react in acetic acid, in the
absence of light and catalyst, to give mainly chloromethyl-
pentamethylbenzene.’49 The scope of these reactions includes
halogenation,u9'53 nitration,5u"55 and isotope exchange.56

CH2C1

C12
0 O
AcOH

 

\r

 

89
Three mechanisms have been proposed (Scheme 11), which
involve a common slow step; that is, the electrophile may
initially attack any activated position of the aromatic

system.

 
  

Scheme 11

In mechanism I,50 the electrophile migrates to the methyl
group attached to the attacked position. Another mechanism
(II) calls for the electrophile to migrate to an ortho double
bond.50 The last mechanism (III) suggests that proton loss
from the carbonium ion generated in the slow step can lead

to 8,51 Rearrangement may then occur at an ion-pair stage

A8
with high retention of halogen, as observed. Mechanism I

seems to be eliminated by a study of the chlorination of

9O
isodurene. The carbonium ions possibly involved are 8_and 8,
in which the positions attacked are both activated by two
ortho-methyl groups and one para-methyl group. Nevertheless,

substitution occurs almost exclusively on the C -methyl group

57

5
to give 3,“,5-trimethylchloromethylbenzene.

t

l3. 9

 

Some support for ionic intermediates (mechanism III) was
reported by Illuminati and coworkers.58 They found that in
the chlorination of hexaethyl-benzene, the amount of side-
chain substitution (ca. 15%) was significantly larger than
the 5% found in the case of hexamethylbenzene. Another result
favoring ionic intermediates was reported by Cerfontain on
the study of side-chain sulfonation of meso-methylated an-

59
thracenes. The mechanism proposed is shown in Scheme 12.

 

‘1’ “(a

Scheme 12

91
An electron transfer mechanism similar to mechanism III
was proposed not long ago by Kochi60(Scheme 13), who observed
a well-resolved esr spectrum of hexamethylbenzene cation—
radical during the chlorination reaction which was conducted
by mixing acetic acid solutions of chlorine and hexamethyl-
benzene directly in the cavity of an electron spin resonance

spectrometer.

___\ _
+
C12\—- + C12
1 'le
C

. l2 . + Cl-
012 + Cl' etc.
Cl

CH2‘

Scheme 13

One interesting system that has never been explored in

this context is octamethylnaphthalene (g).

 

0
Compound g was treated with bromine in CS2 at -78 in the

dark. The reaction was worked up after 30 min and the viscous
residue was diluted with a small portion of ether. The re-
sulting crystalline solid (57%) analyzed correctly for

C H Br2 and gave a reasonable nmr spectrum for the bisbromo-

18 22
methyl compound (28).

924

 

514.85
Br Br
52.35
A .0
O
-78 , 30 min 62.75
57% 52.50

|l\)
\0
O

Other possible structures for this product whose sym-

metry would also satisfy the observed nmr spectrum are:

Br

..

Br

 

Attempts to purify 29 by column chromatography on alumina
with 15% chloroform in carbon tetrachloride as the eluent gave
a 22% yield of an ether, assigned structure of 8;. This result
was also achieved (50% yield) by hydrolysis of 28 in NaOH and
THF at 140—700 for H0 min. In addition to its correct analysis
for C18H22O’ compound ngdisplayed a consistent nmr spectrum
as shown in the formula with eurOpium shift numbers in paren—

thesis. The infrared spectrum showed strong absorption at

93
—1
1193 cm indicating a C—O—C linkage.

  

 

 

O
Alumina 5'0(2'57)
Column 2.22(0.28)
__22% 2
or 7—
NaOH(aq.) 2-60(0-15)
THF 2.30(0.20)
50%
90 91

As a consequence, all those structures which cannot form a
cyclic ether linkage can be ruled out; only 28 and 28g_remain
as possible structures for the dibromo compound. The following
series of reactions confirmed that the 1,8-bisbromomethy1
isomer (28) was the product obtained.

Hexamethylnaphthalene (8) was bisbromomethylated with
paraformaldehyde and hydr0gen bromide (in glacial acetic
acid to produce a 66% yield of 22, which had an nmr spectrum
similar to but distinctly different from that of 28 (see

 

 

 

 

structure).
2.50 2.62
.90
r OAc
.—~—->
HCHO AgOA°
HBr,AcOH AcOH 23 NaOH
66% 2.50 2.60 EtQH
2 100°
(_____1

 

9H
It was not possible to convert 22_to the corresponding
alcohol 2fl_under the common solvolytic conditions. Thus
a two-step process involving the acetate as an intermediate
was employed.61 The diester 22_was obtained by treatment of
22_with silver acetate in glacial acetic acid at 100-110‘) for
7 hr. Its nmr spectrum had one singlet at 55.H0 for the
four methylene protons, three singlets at 62.60, 2.50 and
2.37 with two methyls each for the six aromatic methyls,
and one singlet atIS2.10 for the two acetyl methyls.

This diester was subsequently hydrolyzed in aqueous
sodium hydroxide and ethanol at 100‘> for 2 hr to give the
diol 22 in 82% yield (for the two steps).

An attempt to dehydrate 22 with HCl in methanol only
afforded the dimethoxy compound 22. Its nmr spectral data

are illustrated in the formula.

2.96 2.60

   

 

 
 
  

 

HCl 2.30 “'66
OH (2 drops; 0 OMe3 . L43
(:3 (:3 OH MeOH OMe
reflux,12hr
115%
93 22
2.30 2.60
2.20
p-TSA
; O
CH2C12
reflux,1hr

 

95
The dehydration of 22_was accomplished by reflux for
1 hr in methylene chloride in the presence of p-toluene-
sulfonic acid.
In a manner similar to Kochi's mechanism, an electron
transfer pathway for the formation of 28 from octamethylna-

phthalene is proposed in Scheme 14.

 

B Br

 

Scheme 14

Possible reasons why bromination occurred at the Cl- and C8-
methyls are (l) the loss of a peri-methyl proton from the
cation radical to give intermediate 8 may minimize the strong
peri interaction; (2) the loss of the second proton from

C8-methyl and C -bromomethyl groups, which is expected to be

1
greater than that between C4— and CS-methyls; and (3) the

96
radical of intermediate 8 could be stablized by the adjoining

bromine in the manner of neighboring-group participation.

Another reasonable pathway is shown in Scheme 15.

 

Scheme 15

Since in compound 2_the most reactive position toward elec-

trophiles are a-positions,9 bromine may initially attack one

 

of the peri-carbons. The second bromine-attack may have oc-

curred at C5 instead of C8 because steric hindrance by the

bulky bromomethyl group blocked the C position.

8
Clearly much remains to be learned about the mechanisms

of these reactions, but the results described here show that

bromination of octamethylnaphthalene gives mainly a simple

dibromo product, 28.

97
C. The Birch Reduction of Octamethylnaphthalene

 

When aromatic rings are reduced by sodium (or potassium
or lithium) in liquid ammonia, 1,4-addition of hydrogen takes
place, and nonconJugated cyclohexadienes are produced. This
reaction is known as the Birch reduction.62 The rule of addi—
tion. of hydrogen atoms to a benzene ring based on the experi-
mental evidence is as follows: The hydrogen atoms are added
in positions para to each other, avoiding carbon atoms car-
rying electron-repelling groups, and being attracted to car-
boxyl groups.63

¢~© s t—U

R = MeO, NMe Alkyl

2’

For the naphthalene system without substituents, hydrogen

atoms have been found to add at the a-positions.6u

M

A mechanism was proposed for this type of reaction as

shown in Scheme 16.65

98

H _ H H
Li EtOH
o ——» ———>
EtOH
11 . NH ' H '
q 3 H

Li
H H H
EtOH
(_.___._
H H H '

Scheme 16

The lithium transfers an electron to the ring, becoming
oxidized to Li+ and creating an ion-radical. The ion-radical
accepts a proton from ethanol to give a radical, which is
then reduced to a carbanion by another lithium atom. The
relative stabilities of alkylbenzene anion radicals in tetra-
hydrofuran-l,2-dimethoxyethane mixtures have been measured

by Lawler and Tabit.66

o >o>o>o>$>0 >0:

This implies that the Birch reduction of any aromatic system

should be retarded by alkyl substituents. Since compound 2

 

99

carries methyl groups in all possible locations, it seemed'
interesting to investigate the orientation of the proton
additions and the effect of substituents on the Birch reduction

of 2.

 

Birch Reduction
> :2

3

The reaction of 2_was carried out in a solution of
liquid ammonia, THF and absolute ethanol, with the slow ad—
dition of lithium metal at dry ice temperature. The reac-
tion was continued until the blue color vanished. Evapora—
tion of the ammonia and work-up in the usual way afforded a
39-86.7% yield of product (yield varies with different ratio
of 2/Li, and with temperature). The product was isolated by
recrystallization from petroleum ether (30-600), and was iden-
tified as the 1,4-dihydro derivative 21, Its nmr spectral

data are shown in the formula.

2.15 l.20(d,J=7Hz)

 

 

2.
Li, NH
o. 3 >
. EtOH, THF
2
O
#15:. 22:0 222::
1/24 _50° 86.7% Kb 2

 

100
The stereochemistry of 21 was cis, as shown by the

symmetrical nmr spectrum of its epoxide derivative 28.

.‘ 85% MCPBA
CH2C12

40%

 

 

The epoxidation was carried out at room temperature over
night using 85% MCPBA in methylene chloride to give a 40%
yield of 28.

EXPERIMENTAL

1,4,7,8-Tetramethyl-2,3-5,6-di(3,6-dimethylbenzo)bicyclo

 

[2.2.2]octa-2,5,7-triene-7-epoxide (11)

 

A methylene chloride solution of meta-chloroperbenzoic
acid (50 mg, washed with pH 7 buffer solution and dried under
vacuo) was added dropwise to a solution of 88_(64 mg) in
methylene chloride (5 ml) at room temperature. After 1 min,
the reaction mixture was poured into an aqueous solution of
sodium sulfite. The aqueous layer was separated and extracted
with ether, and the combined organic layer was then washed
with KHCO3 solution and water, and was dried over MgSOu.
The solvent was removed under vacuo at room temperature. A
white solid remained which was recrystallized from methylene
chloride/hexane. The yield of 11_was 42.5 mg (65%). Compound
11 was unstable in organic solutions and decomposed when the
melting point was measured(125°). Mass spectrum (70eV)

332 (M+), 317 (43), 289 (100); nmr (00013) 66.60 (2H, 3, two
aromatic protons on the benzene ring which is syn to the
epoxide), 6.52 (2H, s, two aromatic protons on the benzene
ring which is anti to the epoxide), 2.52 (6H, 5, two aromatic
methyls on the benzene ring which is syn to the epoxide),

2.50 (6H, s, two aromatic methyls on the benzene ring which

is anti to the epoxide), 2.35 (6H, s, two methyls at C
101

1 and C4)’

102

1.30 (6H, 5, two methyls at C and C8).

7

The Rearrangement of 11

 

Compound 11 was dissolved in chloroform in an nmr tube,
and allowed to stand at room temperature over night. The
reaction was followed by changes in the nmr spectrum. A 100%
conversion of 11 to 18 was observed. Nmr (CDC13) 66.70 and
6.65 (2H each, s, aromatic protons), 2.60 and 2.55 (3H each,
s, two aromatic methyls at carbons ortho to C2 and C7), 2.25
and 2.09 (3H each, 3, two aromatic methyls at carbons ortho
to C3 and C6), 2.15 and 1.99 (3H each, s, vinyl methyls at C4
and C5), 1.90 (3H, 3, methyl at Cl), 1.87 (3H, s, acetyl methyl).
Then the solution in the nmr tube was evaporated to dryness
at room temperature under vacuo. The White solid obtained
was quickly mixed with dry KBr, and the infrared spectrum of
18 was taken: 3000 (s), 1690 (s), 1470 (s), 1390 (m), 1360 (m),
830 (s), 800 (m).

Further rearrangement was observed on redissolving 18
in chloroform and allowing it to stand for six more days.

After evaporation of the chloroform, 18 was obtained in 100%
yield. This rearrangement was also accomplished in 100%
yield either by refluxing 11 in chloroform for 12 hr or by
injecting 11 on the gas chromatograph (5' X k" 15% SE-3O on
30/60 Chromosorb W at 230 ). Recrystallization of 12 from
chloroform and absolute methanol gave a 90% yield of pure
Q, mp 205-7°. UV (MeOH) x (e) 308 (sh.) (1 x 103), 283

max
(sh.) (3.4 x 103), 265 (sh.) (7.1 x 103), 235 (2.28 x 10");

 

103

ir (KBr) 3500 (s), 3000 (s), 1620 (m), 1470 (s), 1390 (s),
1340 (s), 920 (m), 830 (3); mass spectrum (70eV) 332 (M+),
317 (43), 289 (100), 27“ (“9), 259 (63), 244 (30), 235 (30);
nmr (CDC13) 56.65 (2H, s, aromatic protons), 6.52 (2H, s,
aromatic protons), 5.45 and 5.20 (1H each, 5, vinyl protons),
2.60 and 2.22 (3H each, s, two aromatic methyls), 2.20 (6H,
3, two aromatic methyls), 2.10 (1H, broad, hydroxyl proton),
2.0 (3H, 3, methyl at cl), 1.60 (3H, 8, methyl at C5), 1.15
(3H, 8, methyl at C8).

522}: Calcd for C24H28O: C, 86.70; H, 8.49. Found:
C, 86.59: H, 8.46.

1,4,7,8-Tetramethyl-2,3-5,6-dibenzobicyclo[2.2.2]octa-2,5,7—

 

triene (81)

 

A mixture of 1,2,3,4-tetramethylnaphthalenel“(l g, 5.4
mmol), benzenediazonium-2-carboxylate hydrochloride (1.2 g,
5.6 mmol), pr0pylene oxide (15 ml) and 1,2-dichloroethane
(50 ml) was gradually heated until gas evolution occurred.
After the solution became clear, the reaction was continued
at reflux temperature for 2 hr. The volatile solvents were
removed under vacuo, and the oily residue was dissolved in
ether. The ethereal solution was washed with cold 2% NaOH,

with water, and dried over MgSO The brown residue which

“0
remained after the solvent was evaporated was chromatographed
on silica gel with cyclohexane as the eluent. The first frac—
tion proved to be the unreacted starting material; the second

0
fraction, which was recrystallized from methanol, mp 178-180 ,

104
was the title compound with a yield of 32% (448 mg). Mass
spectrum (70eV) 260 (M+), 245 (100), 230 (42), 215 (32),
206 (18); nmr (CDC13) 66.74-7.15 (8H, m, aromatic protons),

2.05 (6H, 8, two methyls at C and 04)’ 1.65 (6H, 8, two vinyl

l
methyls).

Anal. Calcd for c H c, 92.26; H, 7.74. Found:

20 20‘

C, 92.29; H, 7.74.

1,4,7,8-Tetramethy1-2,3—5,6-dibenzobicyclo[2.2.2]octa-2,5,7-

 

triene-7,8-epoxide (82)

 

The same procedure for the epoxidation of fl8_was fol-
lowed to transform 125 mg (0.48 mmol) of 81 into 96 mg
0 O
(72.4%) of 83, mp 154-155 (decomposed at 120 ). Uv (CH3CN)

A (.) 233 (1.64 i 103), 266 (1.28 x 103), 274 (1.46 x 103);

max
mass spectrum (70eV) 276 (M+), 233 (100), 218 (26), 203 (22),

202 (25), 191 (23); nmr (CDC13) 67.0 (8H, m, aromatic protons),

2.93 (6H, s, two methyls at C and C4)’ 1.21 (6H, s, two

1

methyls at C7 and C8).

Anal. Calcd for C20H2OO: C, 86.92; H, 7.29. Found:

C, 86.98; H, 7.34.

l-Acetyl-2,3-6,7-dibenzo-l,4,5-trimethylcyclohepta-2,4,6-

 

triene (82)

 

A solution of §8_(98 mg) in chloroform (5 ml) was

refluxed for 4.5 hr. Evaporation of the solvent gave

105

98 mg (100%) of 88, mp 173-5o from methanol. Uv (CH CN)

1 max (c) 266 (9.3 x 103), 228 (1.54 x 10”); ir (KBr) 3050 (m),
1703 (s), 1480 (m), 233 (100), 218 (25), 202 (29); nmr (00013)
57.0-7.4 (8H, m, aromatic protons), 2.20 (6H, 3, two vinyl
methyls), 1.93 (3H, 3, methyl at Cl), 1.80 (3H, 8, methyl at
carbonyl).

Anal. Calcd for C2OH O: C, 86.92; H, 7.29. Found:

20
C, 86.85; H, 7.34.

Thermal Reaction of 82 in a Basic Medium

 

Compound 32_(20 mg) was dissolved in pyridine (3 ml).
The flask was previously rinsed with concentrated ammonium
hydroxide and dried in an oven. The reaction mixture was
heated at 65-750 for 5 hr and then evaporated to dryness
under vacuo. The nmr spectrum showed only recovered starting

material, 22.

Reduction of 82_by LAH

 

A solution of 82 (99 mg) in THF (5 ml) was added slowly
at room temperature into a suspension of LAH (70 mg) in THF
(5 ml). After the mixture was stirred at room temperature
for 1 hr, the excess LAH was decomposed by pouring the mix-
ture over ice. The resulting suspension was then extracted
with ether, and the ethereal solution was washed repeatedly

with water and dried over MgSOu. After chromatography on

106
alumina with 50% ether in hexane as the eluent, 95 mg of
82 (98%) was obtained from the second fraction (the lst frac-
tion was an unidentified solid). The colorless liquid 82
gave no carbonyl absorption in its ir spectrum (neat): 3480
(s), 3010(s), 1485 (s), 1400 (m), 1270 (m), 1050 (s). Mass
spectrum (70eV) 278 (M+), 260 (34), 233 (100), 218 (28), 206
(34), 202 (28); nmr (CDC13) 66.80-7.30 (8H, m, aromatic pro-
tons), 5.0 (lH, q, J = 6 Hz), 3.26 (1H, s, hydroxyl proton),
2.24 (6H, 3, two vinyl methyls), 1.85 (3H, 3, methyl at Cl)’
0.85 (3H, d, J = 6 Hz).

1,5,8-Trimethy1-4-methylene-2,3-6,7-dibenzo-8-hydroxylbicyclo-

 

[3.2.1]octa-2,6-diene (82)

 

A solution of 82 (68 mg) in chloroform (5 ml) with a
trace amount of p-toluenesulfonic acid was refluxed for 12 hr.
The chloroform solution was then washed with NaHCO3 solution
and dried over MgSOu. The residue which was obtained after
evaporation of the solvent was chromatographed on silica gel
with 20% ether in hexane as the eluent. The first fraction
was 82 (7 mg, 10.3%), and the second fraction was 82 (60 mg,
88.3%), mp 170-3o from methanol. Mass spectrum (70eV) 276
(M+), 233 (100); nmr (CDC13) 67.50-6.95 (8H, m, aromatic pro-
tons), 5.55 and 5.15 (1H each, 8, vinyl protons), 1.70 (3H,
s, methyl at Cl), 1.58 (3H, 8, methyl at CS), 1.0 (3H, 5,
methyl at C8).

2821. Calcd for C2OH2OO: C, 86.92; H, 7.29. Found:

C, 86.93; H, 7.33.

107
The same result was obtained by refluxing the chloroform
solution of 82 in the presence of a small amount of p-toluene-

sulfonic acid for only four hours.

l-Acetyl-2,3-6,7-dibenzo-l,4,5-trimethylcyclopenta-2,4,6-

 

triene-4-epoxide (88)

 

A solution of 8§_(276 mg, 1 mmol) in methylene chloride
(5 ml) was treated with MCPBA (excess, washed with pH 7 buf-
fer solution and dried under vacuo) over night at room temp-
erature. After work-up, 88_was obtained in 99% yield, mp 140-
1430 from ether and petroleum ether (30—600). Ir (KBr)
3000 (m), 1710 (s), 1480 (w), 1390 (m), 1180 (m), 1100 (m),
770 (3); mass spectrum (70eV) 292 (M+), 206 (42), 146 (52),
43 (100); nmr (CD013) 67.50-7.0 (8H, m, aromatic protons),
2.15 (3H, s, methyl at Cl), 1.97 (3H, s, acetyl methyl), 1.70
(6H, s, two methyls at C4 and C5).

1-Acetyl-2,3-6,7-di(3,6-dimethylbenzo)-l,4,5-trimethylcyclo-

 

penta-2,4,6-triene-4-epoxide (88)

 

To a solution of 88 (64 mg) in methylene chloride (5 ml)
was added 85% MCPBA (50 mg, 25% excess) at room temperature.
After 15 hr, the reaction was worked up as usual. A mixture
of two products was obtained. They were separated by pre—
parative VpC (5' X k" 15% SE—30 on 30/60 Chromosorb W at 2250).
The first fraction was 12 in 70% vpc yield, and the next
fraction was 88 in 30% vpc yield. Ir (KBr) 3510 (w), 3000 (s),

108
1620 (m), 1460 (s), 1390 (s), 1340 (s), 1150 (m), 1120 (m),
920 (s), 830 (3); mass spectrum (70eV) 348 (M+), 330 (44),
287 (64), 272 (28), 257 (20), 174 (100), 173 (88), 159 (38);
nmr (CDC13) 66.77 and 6.74 (2H each, s, aromatic protons),
2.45 (6H, two 8, aromatic methyls at carbons ortho to C and

2
C7), 2.40 (6H, two 3, aromatic methyls at carbons ortho to

 

C3 and C6), 2.20 (6H, two s overlaped, methyls at C1 and
carbonyl), 1.70 (3H, 3, methyl at C4)’ 1.57 (3H, s, methyl
at .

05)

Acid Rearrangement of 82

 

p-Toluenesulfonic acid (20 mg, 0.12 mmol) was added to
a solution of 82 (68 mg, 0.24 mmol) in benzene (100 ml).
The solution turned pink when the acid started to dissolve.
After 1.5 hr at reflux temperature, 20 mg of anhydrous
NaZSOu was added. The reaction mixture was refluxed for
another 1.5 hr. The solution was passed through a short
silica gel column with benzene as the eluent. Evaporation
of the solvent followed by chromatography on alumina with
hexane as the eluent gave a mixture of two isomers which
were separated by vpc (5' X k" 5% SE-30 on Chromosorb W at

160°

). The lst fraction, with retention time of 20 min,
was completely separated from the 2nd fraction with reten-
tion time of 25 min. These two isomers gave exactly the
same mass spectrum (70eV) 260 (M+), 245 (34), 215 (24),

206 (100), but different nmr spectra. Nmr (lst isomer, CDC13)

109
66.90-7.60 (8H, m, aromatic protons), 5.60 and 5.10 (1H each,
8, vinyl protons), 2.10 (1H, m, proton at C8), 1.70 (3H, 3,
methyl at Cl), 1.56 (3H, 3, methyl at C5), 0.76 (3H, d, J =
6 Hz, methyl at C8); nmr (2nd isomer, CDC13) 66.80-7.10 (8H,
m, aromatic protons), 4.94 and 4.60 (1H each, two d, J = 2 Hz,
vinyl protons), 2.0 (3H, s, methyl at Cl), 1.86 (3H, 3, methyl
at 05), 1.30-1.70 (1H, m, proton at 08), 0.82 (3H, d, J =
6 Hz, methyl at C8). The total yield of the two isomers was
64%. The relative yield from vpc was 5/1 (lst/2nd).

Anal. Calcd for C2OH C, 92.26; H, 7.74. Found:

, 20‘
c, 91.91; H, 7.44.

The Bromination of Octamethylnaphthalene

 

A mixture of carbon disulfide (5 m1) and octamethylnaph-
thalene (2) (480 mg, 2 mmol) was kept in a flask wrapped
with aluminum foil at -78°. To this mixture was added slowly
a solution of bromine (0.12 ml, 2 mmol) in CS2 (3 ml). After
30 min at this temperature, the reaction was stopped by ad-
ding sodium bisulfite solution. The organic layer was sepa-
rated, diluted with chloroform, and washed with sodium bicar—
bonate solution. Solvent was removed under vacuo after the
solution was dried (MgSO“). The viscous residue was then
diluted with a small portion of ether. A crystalline solid
(28) was obtained in 57% yield (450 mg), mp 165-1700.
Mass spectrum (70eV) 398 (10), 396 (M+), 319 (14), 317 (14),
238 (100), 223 (45), 207 (30), 193 (26); nmr (00013) 54.85

110
(4H, s, methylene protons), 2.75 (6H, s, aromatic methyls
at C3 and C6), 2.50 (6H, s, aromatic methyls at C4 and CS),
2.35 (6H, s, aromatic methyls at C2 and C7).
Anal. Calcd for C18H22Br2: C, 54.59; H, 5.60. Found:

C, 54.47; H, 5.63.

2,6-Dihydro-naphtho[1,8,8a-c,d]pyran (21)

 

Column chromatography 450 mg of 28 on alumina with 15%
chloroform in carbon tetrachloride as the eluent gave three
fractions. The first two fractions were insignificant, and
attempts to identify them were unsuccessful. 'The.third frac—
tion, after recrystallization from ether, gave 63 mg (22%)
of 21, mp 194-80. Ir (001“) 3050 (s), 2950 (s), 2850 (m),

1470 (m), 1400 (m), 1143 (s), 1060 (m), 980 (m); mass spectrum
(706V) 254 (M+), 239 (80), 225 (55), 211 (43), 195 (19):
179 (20), 165 (21); nmr (CDC13) 65.0 (4H, s, methylene pro-

tons), 2.60 (6H, s, aromatic methyls at C and C6), 2.30

3

(6H, s, aromatic methyls at C and C5), 2.22 (6H, s, aromatic

u
methyls at c2 and c7).

fl£§l° Calcd for C18H22O: C, 84.99; H, 8.72. Found:
C, 85.02; H, 8.69.

Compound 28 (250 mg) was dissolved in THF (10 m1). This
solution was then added into an aqueous solution of NaOH,

0
and the resulting mixture was heated at 40-70 for 40 min.

After work-up, a 50% yield of 21_was obtained.

lll

Bisbromomethyl-l,4,5,6,7,8-hexamethylnaphtha1ene (22)

 

Dry HBr was bubbled through a suspension of paraform-
aldehyde (300 mg) in glacial acetic acid (5 m1) until the
solution turned clear. To this solution was added at 35°
a glacial acetic acid (5 ml) solution of.2 (212 mg). The
resulting solution was stirred at this temperature for 2 hr,
and at 40-450 for another hour. A white solid was precipi-
tated. The mixture was then poured onto ice, and the solid
was filtered and washed with water. The solid was redis-
solved in ether and the ethereal solution was washed with
aqueous NaHCO3 and water, and was dried (MgSO“). Evaporation
of the solvent and recrystallization from ether and chloroform
gave a 66% yield of 22, mp 182-4o. Mass spectrum (70eV)

398 (ll), 396 (M+), 319 ("3), 317 (43), 238 (100), 223 (43),
208 (30), 193 (30); nmr (CDC13) 64.90 (4H, s, methylene pro-

tons), 2.62 (6H, s, methyls at C and C4)’ 2.50 (6H, s,

l
methyls at C5 and C8), 2.30 (6H, s, methyls at C6 and C7).

Anal. Calcd for ClBH2 Br2. C, 54.59; H, 5.60. Found:

C, 54.58; H, 5.67.

58
1,4,5,6,7,8-Hexamethyl-2,3-dihydroxymethylnaphthalene (28)

 

Silver acetate was precipitated by adding excess aqueous
KOAc to AgNO3 (1 g) in water. The solid was filtered and
washed three times with glacial acetic acid, and was then

diluted with glacial acetic acid and dried with acetic

. 112
anhydride. The dibromo compound 22 (800 mg) was then added
to this solution, and the mixture was held at lOO-llOofor
7 hr. The white solid (AgBr) was filtered after the reaction
was stopped. The filtrate was concentrated under reduced pre-
ssure and purified with elution on alumina with 50% ether

in hexane as the eluent. The diester 22, nmr in CDCl 55.40

3
(4H, s, methylene protons), 2.60 (6H, s, methyls at C1 and C4)’

2.50 (6H, s, methyls at C and C8), 2.37 (6H, s, methyls at

5
C6 and C7), 2.10 (6H, 3, two acetyl methyls) obtained was
redissolved in 10% of aqueous NaOH (30 ml) and absolute
ethanol (30 ml). The solution was stirred at 1000 for 2 hr,
and gave 450 mg (82%) of diol 28) after work-up, mp 178-1800
from chloroform and petroleum ether (30-600). Ir (KBr) 3350
(s), 2950 (s), 1580 (w), a series of bands between 1490-
1100, 1000 (3); mass spectrum (70eV) 272 (M+), 254 (100),
239 (“3), 225 (45), 211 (31), 195 (20), 179 (22), 155 (23);
nmr (CDC13) 64.97 (4H, s, methylene protons), 2.60 (6H, s,

and C8),

methyls at C1 and C4)’ 2.50 (6H, s, methyls at C5

2.37 (6H, s, methyls at C6 and C7).

Treatment of 22 with Acid

 

The diol 22 (100 mg) was dissolved in absolute methanol
(10 m1), and 2 drops of concentrated HCl was added. The
mixture was refluxed for 12 hr, and was worked up and dried
(MgSO“). Chromatography on alumina with 50% ether in petro-

leum ether (30-600) as the eluent gave 50 mg (45%) of 22,

113
mp 96-970, Mass spectrum (70eV) 300 (M+, 100), 268 (86),
253 (94), 238 (“0), 225 (25), 223 (25), 207 (27); nmr (CD013)
64.66 (4H, s, methylene protons), 3.43 (6H, s, methoxy methyls),
2.60 (6H, s, methyls at C6 and C7).
8821. Calcd for C20H28O2: C, 79.95; H, 9.39. Found:
C, 79.87; H, 9.39.

Dihydro-iso-1,4,5,6,7,8-hexamethylnaphthofuran (28)

 

To a 1,2-dich1oroethane solution of diol 22_(100 mg),
was added a saturated solution of p-toluenesulfonic acid
in 1,2-dichloroethane. The mixture was heated at 40-500
for 1 hr and then worked up. Chromatography on alumina with
30% ether in petroleum ether (30-60) as the eluent gave a
crystalline solid of 22_(30%), mp 1800(decomposed). Nmr
(CDC13) 65.20 (4H, s, methylene protons), 2.60 (6H, s,

methyls at C1 and C4)’ 2.30 (6H, s, methyls at C and C8),

5

2.20 (6H, s, methyls at C and C7).

6

67
l,4-Dihydrooctamethylnaphthalene (97)

 

A three-necked flask equipped with a dry-ice condenser,
a mechanical stirrer, and an inlet tube, was charged with
240 mg of octamethylnaphthalene (2). The stirrer was started,
and to the rapidly stirred flask contents was added 50 m1 of
ammonia as rapidly as possible. A portion (ca. 5 ml) of THF
was then added to dissolve the insoluble organic material.

The liquid ammonia was passed through a sodium hydroxide tube

114
before being condensed into the flask. When all the octa-
methylnaphthalene had gone into solution, 3 ml of absolute
ethanol was added. To this mixture was then added 168 mg
of lithium metal in small pieces and at such a rate as to
prevent the ammonia from refluxing too violently. After
the addition of the lithium had been completed (ca. 45 min),
the solution was stirred for a while until the blue color
disappeared. Evaporation of the ammonia and decomposition
of the residue with cold water were followed by extraction
with ether. Evaporation of ether under vacuo gave 210 mg
(86.7%) of the product 21. Recrystallization was successful
with proper amount of petroleum ether (30-600), mp 70-730.
Mass spectrum (70eV) 242 (M+), 227 (80), 212 (100), 198 (23),
162 (25), 147 (56); nmr (00013)53.0 (2H, q, J . 7 Hz, protons
at C and C4)’ 2.20 (6H, s, methyls at C

l
s, methyls at C

6 and C7), 2.15 (6H,

and C8), 1.75 (6H, s, methyls at C and C3),

5
1.20 (6H, d, J = 7 Hz, methyls at C1 and C4)’

2

Anal. Calcd for 018H26: C, 89.19; H, 10.81. Found:
C, 89.63; H, 10.87.

1,4-Dihydro-2,3-epoxy-octamethy1naphthalene (28)

 

The epoxidation of 21_was carried out by adding MCPBA
(85%, 150 mg) to a solution of 21 (100 mg) in 5 ml of methylene
chloride. The reaction mixture was stirred at room tempera-
ture for 15 hr. The organic solution was washed with Na CO

2 3
solution and dried (MgSO“). Chromatography on silica gel with

115

10% ether in cyclohexane as the eluent gave 50 mg (40%)
of 28, mp 128-9O from petroleum ether (30-600). Ir(KBr)
2950 (s), 1480 (s), 1400 (s), 1120 (s), 1080 (s), 1020 (m),
890 (8); mass spectrum (70eV) 258 (M+), 241 (70), 229 (73),
225 (100), 215 (71), 201 (70), 185 (62); nmr (CClu) 63.22
(2H, q, J = 7 Hz, protons at C1 and C4)’ 2.10 (12H, 3, four
aromatic methyls), 1.35 (6H, s, methyls at C2 and C3), 1.15
(6H, d, J = 7 Hz, methyls at C1 and C“).

2821. Calcd for C18H26O: C, 83.66; H, 10.14. Found:
c, 83.78; H, 10.20.

10.

ll.

12.

13.

14.
15.

16.

REFERENCES

G. Gafner and F. H. Herbstein, Nature, 200, 130 (1963).

N. M. Donaldson and J. M. Robertson, J. Chem. Soc.,
17 (1953). N

For a review, see V. Balasubramaniyan, Chem. Rev.,

66, 567 (1966).

B. A. Nagasampagi, R. C. Pandey, V. S. Pansare, J. R. Prahlad,
and S. Dev, Tet. Lett., 411 (1964).

C. MacLean and E. L. Mackor, Mol. Phy., 2, 223 (1960).

B. J. Abadir, J. W. Cook, and D. T. Gibson, J. Chem. Soc.
8 (1953).

A. Oku, T. Kakihana, and H. Hart, J. Am. Chem. Soc., 82,
4554 (1967).

H. Hart and 1. Paul, unpublished result.
H. Hart and A. Oku, J. Org. Chem., 21, 4269 (1972).

M. C. Kloetzel, R. P. Dayton and H. L. Herzog, J. Am.
Chem. Soc., 72, 273, 1991 (1950).

For a review, see E. Clar, "Polycyclic Hydrocarbons",
p. 288, Academic Press (1964).

R. J. Dellaca, B. R. Penfold and W. T. Robinson, Acta.
Cryst., B25, 1589 (1969).

C. F. Allen and M. J. Kalm, Org. Syn., Coll. Vol. 2,
608 (1963).

D. Havsigk, Synthesis, 307 (1971).

T. L. Gilchrist and C. W. Rees, "Carbenes, Nitrenes, and
Arynes", Appleton-Century-Crofts, N. Y. (1969).

For a review of naphthynes, see R. W. Hoffmann, "Dehydro-
benzene and Cycloalkynes", Academic Press (1967); R. Huisgen
and L. Zirngib1,Angew. Chem., 69, 389 (1957); R. Huisgen
and J. Sauer, Angew. Chem., 69,390 (1957); L. Friedman

and F. M. Logullo, J. Am. Chem. Soc., 85, 1549 (1963);

116

117

J. Org. Chem., 22, 3089 (1969); W. Amrein and K. Schaffner,
Helv. Chim. Acta., 28, 380 (1975).

17. H. Hart and A. Oku, unpublished result.
18. J. E. Gordon, J. Org. Chem., 22, 2722 (1970).

19. G. Wittig and E. R. Wilson, Chem. Ber., 28, 451 (1965):
J. D. Cook and B. J. Wakefield, Chem. Comm., 297 (196%);
3.

R. J. Martens and H. J. denHertog, Rec. Trav. Chim.,
621 (1964).

20. P. A. S. Smith, C. D. Rowe, and L. B. Bruner, J. Org. Chem.,
35. 3430 (1969).

21. W. vonE. Doering and C. H. DePuy, J. Am. Chem. Soc., _2,
5955 (1953).

22. H. O. Calvery, C. R. Noller, and R. Adams, J. Am. Chem.
Soc., 21, 3058 (1925).

23. E. J. Corey and M. Chaykovsky, J. Am. Chem. Soc., 81,
1345 (1965).

24. R. S. Givens and W. F. Oettle, Chem. Comm., 1164 (1969).
25. E. Ciganek, J. Am. Chem. Soc., 88, 2882 (1966).

26. S. B. Polovsky and R. W. Franck, J. Org. Chem., 22,
3010 (1974).

27. W. Mosby, J. Am. Chem. Soc., 12, 2564 (1952).

28. L. F. Fieser, Can. J. Chem., 22, 1599 (1965).

29. E. Wolthuis, B. Bosseubroek, G. DeWall, E. Geels, and
A. Leegwater, J. Org. Chem., 28, 148 (1963).

30. G. Wittig and L. Pohmer, Chem. Ber., 82, 1334 (1956).

31. Migggtizon and N. Trong Anh, Bull. Soc. Chim. Fr., 3208

32. E. Wolthuis, J. Org. Chem., 28, 2215 (1961).
33. H. Hibbert, J. Am. Chem. Soc., 21, 1748 (1915).
34. J. DeWit and H. Wynberg, Rec. Tran. Chim., 22, 281 (1973).

35._ E. Clar, "The Aromatic Sextet", John Wiley and Sons Inc.
1972).

36. C. K. Bradsher and L. E. Beavers, Chem. and Ind., 1394

37.

38.

39.
40.

41.

42.

43.
44.

25.

46.

27.

48.
49.
50.

51.

52.

53.

54.

118
(1954); J. Am. Chem. Soc., 11, 4812 (1955); 19, 2459 (1956).

C. K. Bradsher, T. W. G. Solomons and F. R. Vaughan,
J. Org. Chem., 22, 757 (1960).

C. K. Bradsher and J. C. Parham, J. Org. Chem., 28,
83 (1963).

D. L. Fields, J. Org. Chem., 28, 3002 (1971).

D. L. Fields, T. H. Regan, and R. E. Graves, J. Org. Chem.,
39. 2995 (1971).

W. 13 Aalbersberg, G. J. Hothink E. L. Mackor, and
w. P. WeiJland, J. Chem. Soc., 3049, 3055 (1959).

J. P. Colpa, C. MacLean and E. L. Mackor, Tetrahedron,
SUPpl- l9. 65 (1963).

S. W. Pelletier, Chem. Ind., 1034 (1953).

S. J. Cristol and R. K. Bly, J. Am. Chem. Soc., 82,
5155 (1960).

S. J. Cristol, R. P. Arganbright, and D. D. Tanner,

J. Org. Chem., 28, 1374 (1963); S. J. Cristol, F. P. Parungo,
D. E. Plorde, afia K. Schwarzenbach, J. Am. Chem. Soc., 81,
2870, 2879 (1965); S. J. Cristol, R. J. Bopp, and

A. E. Johnson, J. Org. Chem., 34, 3574 (1969); S. J. Cristol
and R. J. Bopp, J. Org. Chem.,822, 1336 (1974).

W. Tochtermann, K. Gieger and G. Rissmann, Liebigs Ann.
Chem-. 323 (1975).

M. Nogradi, W. D. Ollis, and I. O. Sutherland, Chem. Comm.,
158 (1970).

S. R. Hartshorn, Chem. Soc. Rev., 2, 167 (1974).
E. Baciocchi and G. Illuminati, Tet. Lett., 637 (1962).

E. Baciocchi, A. Ciana, G. Illuminati, and C. Pasini,
J. Am. Chem. Soc., 81, 3953 (1965).

E. Baciocchi and G. Illuminati, Ricerca Scient, 22, 462
(1964). CA. 63. 14655a (1965).

L. J. Andrews and R. M. Keefer, J. Am. Chem. Soc., 88,
2158 (1962).

E. Baciocchi, M. Casula, G. Illuminati, and L. Mandolini,
Tet. Lett., 1275 (1969).

K. Nakamura, Bull. Chem. Soc. Japan, 22, 133 (1971);

119
H. Suzuki, ibid., 22, 879 (1970).

55. E. Hunziker, J. R. Penton, and H. Zoolinger, Helv. Chim.
Acta., 22, 2043 (1971).

56. C. Eaborn and G. J. Wright, J. Chem. Soc. (B), 2263 (1971).

57. G. Illuminati, L. Mandolini, and A. Patara, Tet. Lett.,
4161 (1972).

58. G. Illuminati, L. Mandolini, E. M. Arnett, and R. Srnoyer,
J. Chem. Soc. (B), 2206 (1971).

59. A. Kfieberg-Telder and H. Cerfontain, Tet. Lett., 3535
(197 ).

60. J. K. Kochi, Tet. Lett., 2315 (1972).

61. w. w. Hartman and E. J. Rahrs, Org..Syn., Coll. Vol.III,
550, 652 (1955).

62. For reviews, see A. J. Birch, Quart. Rev., 4, 69-93 (1950);
A. J. Birch and H. Smith, Quart. Rev., 12, 17-33 (1958).

63. A. J. Birch, J. Chem. Soc., 430 (1944).
62. A. J. Birch and D. Nasipuri, Tetrahedron, 8, 148 (1959).

65. A. P. Krapcho and A. A. Bothner-By, J. Am. Chem. Soc.,
8_1. 3658 (1959); 8a. 751 (1960).

66. R. G. Lawler and C. T. Tabit, J. Am. Chem. Soc., 21,
5671 (1959).

67. C. D. Gutsche and H. H. Peter, Org. Syn., Coll. Vol.IV,
887 (1963).