PART f

STEREOCHEMISTRY OF THE RAPID EQUlLlBRATION ‘
0F PROTONRTED BICYCLO {3.1 .0] HEXENONES

PART H

CONTROL OF THESTEREOCHEMESTRY or _  I
2,5-CYCLOHEXADEENONE PHOTDISOMERIZATEON ~
BY 3mm: FACTORS.

PART“!

THE SYNTHESIS AND PHOTGREARRA‘NGEMENT
0F 1,6,8,8,9,10- HEXAMETHYLTRECYCLO
[2 .2 .4 ‘0 2.57] DECA- 3,9-DtENE-7-0NE AND
{TS DERIVATIVES

Thesis for the Degree 54‘ Pk. E}.
MICHiGAN STATE fiNWERSlTY
THOMAS RODGERS
1970

 
   

J'fléslg
LIB R A R °_’
Michigaa': ~te

University

 
   

§

'1.“

 

This is to certify that the

thesis entitled

PART I - STEREOCHEMISTRY OF THE RAPID EQUILIBRATION OF
PROTONATED BICYCLO[3.1.0]HEXENONES

PART II - CONTROL OF THE STEREOCHEMISTRY OF 2,5-CYCLOHEXA-
DIENONE PHOTOISOMERIZATION BY STERIC FACTORS

PART III - THE SYNTHESIS AND PHOTOREARRANGEMENT 0F
1,6,8,8,9,1O-HEXAMETHYLTRICYCLO[2.2.4.02:5]DECA—
3,9-DIENE-7-ONE AND ITS DERIVATIVES

Presented by Thomas Rodgers

has been accepted towards fulfillment
of the requirements for

Ph.D. degree in Jhemim

 

LAWQ ~14, MK

Ihknpnmuun

 

T

 

 

Date__lley.emb_er_fi._1219

0-7639

 

*wfln Y..- -
53.11.303.51; .

 

 

 

CCNTRCL CF

N
y.
-

IN.-
C

Io!

m D

1I618I

In the
A!“
~«d~cataliz<
pentame thw] ‘h

1'3'4'5:5~pe

VEStigated .

ABSTRACT

PART I

STEREOCHEMISTRY OF THE RAPID EQUILIBRATION OF PROTONATED
BICYCLO[3.1.0]HEXENONES

PART II

CONTROL OF THE STEREOCHEMISTRY OF 2,5-CYCLOHEXADIENONE
PHOTOISOMERIZATION BY STERIC FACTORS

PART III
THE SYNTHESIS AND PHOTOREARRANGEMENT OF

1,6,8,8,9,10-HEXAMETHYLTRICYCLO[2.2.4.02I5]DECA-3,9-
DIENE-7-ONE AND ITS DERIVATIVES

BY

Thomas Rodgers

In the first part of this thesis, the mechanism of the
acid-catalized rearrangement of ggggf6-propy1-1,3,4,5,6—
pentamethylbicyclo[3.1.0]hexenone (11) and géng-propyl-
1,3,4,5,6-pentamethylbicyclo[3.1.0lhexenone (lg) was in-

vestigated.

17

 

Compounds lz'and léu prepared by the photolysis of fig
or EQ and separated by preparative vpc, were deuterated in

1

Thomas Rodgers

tflne: 4—position by base catalyzed exchange. Treatment of
O / O
22 23

' V'W

liabeled leggj with concentrated sulfuric acid at 40 for
2 Inin gave only Q2 whereas labeled lgjgl) under the same

<:onditions gave only g2, The mechanism proposed for the

O O
H SO
\\ ‘JL_JL_) ‘\
*

(1*.5D) *(1.5D)

 

 

£22, ea
0
H3504
\
* * *
(1.5D) (1.5D)
e1, 33

observed rearrangement is a series of suprafacial 'non-

pivoting' 1,4—sigmatr0pic shifts.
2

Thomas Rodgers
The second part of this thesis is concerned with a
study of the stereoselectivity in the photolysis of 2,5-
cyclohexadienones. Various 2,5-cyclohexadienones fig

(R1 = H or Me, R2 = H or Me, R3 = H or Me) were synthesized

92

and photolyzed to determine the ratio of photoproducts.

Photolysis of 22'(R1 = R2 = H and R3 = H or Me) gave a 1:1

ratio of photoproducts (gg'and 22). On the other hand,

 

0 s
wlll" R3
R2 R1
232, as

photolysis of gg'(Ri= R2 = Me and R3 = H) gave a 10:1 ratio
of fig‘to 2;. Photolysis of gg‘(R1 = Me, R2 = H, and R3 = H
or Me) also showed a stereoselectivity in the ratio of photo-
products, for in these photolyses the ratio of gg'to gg'was
between 2:1 (R3 = H) and 4:1 (R3 = Me). In this last men—

tioned photolysis, only photoproducts with R2 = Me and

3

 

He -
‘ r
a A-
T..€ .t'_
. .
a!“ A " A G
O

 

 

 

4.

H
A

f

Thomas Rodgers
R1 = H were formed, with no evidence for the formation of
the other set of isomeric photoproducts (R2 = H and R1 =
Me).
The reason proposed for the stereoselectivity in these
photolyses is that the ratio of photoproducts is determined
by the relative amount of each conformer of 2£.(l§§)' which

is controlled by the size of R1, R2, and R3. As the size

 

of R1, R2, and R3 become larger, preference for the con-
former on the right increases. Steric factors therefore
favor the formation of that photoproduct (22) with the
larger substituent at C-6 endg.

The formation of only one set of photoproducts in the
photolysis of unsymmetrical 22'(R1 = Me, R2 = H, and R3 = H
or Me) was postulated to be due to the difference in the
stability of the transient intermediates formed during the

photorearrangement (143 is favored over 142).

0' Pr 0‘ Pr
Me Me
0/
12% 122:

 

“3"“ "5 L$~€

t"- b V. b. _
.~ I

:fi“ “ “F. A:

-mu;—.‘~.A. V‘

()

 

Thomas Rodgers
The synthesis and photochemistry of the B,y-unsaturated

ketone 176 and its derivatives was the subject of the third

176

part of this thesis. Compound 176 was synthesized by the
addition of cyclobutadiene to 2,3,4,5,6,6-hexamethyl-2,4-
cyclohexadienone. In order to prove the endo orientation

of 176, the compound was transformed into 177 by reduction

 

 

O.
1. LiAlH4
\
3. hv /
5. CIOS, H2504
176 1 7

to its alcohol derivative, which was acetylated and photo-
lyzed to an acetoxyhomocubane structure. This acetoxyhomo-
cubane was reduced to the alcohol and oxidized to generate
111,

Photolysis of the B,y-unsaturated ketone lzg'gave the
expected products. Direct irradiation of 112,1n methanol or
pentane gave compound lgfi'by a 1,3-acyl migration pathway

and compound 197 from the decarbonylation of 196. Acetone-

5

 

sensitized 1

tion pathwav

 

Thomas Rodgers
sensitized irradiation of 176 gave 202 by a 1,2-acyl migra-

tion pathway.

 

 

196 197
176 L9 0
“‘“’acetone
corex Q
202

m

u

 

"an

Us 9"? '
STERLo'erH‘.

f‘“ VFW“, r '
Lea 15.9.;

a-

\u

'(l

PART I
STEREOCHEMISTRY OF THE RAPID EQUILIBRATION OF PROTONATED
BICYCLO[3.1.0]HEXENONES
PART II
CONTROL OF THE STEREOCHEMISTRY OF 2,5-CYCLOHEXADIENONE
PHOTOISOMERIZATION BY STERIC FACTORS
PART III
THE SYNTHESIS AND PHOTOREARRANGEMENT OF
l,6,8,8,9,10-HEXAMETHYLTRICYCLO[2.2.4.02o5]DECA-3,9-
DIENE-7-ONE AND ITS DERIVATIVES
BY
Roberf

Thomaisodgers

A THESIS

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

DOCTOR OF PHILOSOPHY
Department of Chemistry

1970

747154)

To My Wife,
Kay
and

My Daughter,

Barbara

ii

 

_ . a g l
e . . a to
C _ a. C C 2 Cu
H. u. a . e .2 1
.c r. 2“ Mu um o
.«u .
8 ”L 1. C. flu W
A . .
.7. S v. A. a “a
E C. J
..
C. «C. .3. A“
I a f .

 

C4
s

te...ber , 1

av.
.C
~,\u

Apprecia

2.

L

A

Ring and

‘
50“

Providing f '

‘7‘
‘A.

triene

to:-
V»

ACKNOWLEDGMENT

The author wishes to express his appreciation to
Professor Harold Hart for his enthusiasm, encouragement,
and guidance throughout the course of this study.

Appreciation is extended to Michigan State University
for a Graduate Teaching Assistantship from September, 1966
to June, 1967, from September, 1967 to June, 1968, and from
September, 1968 to December, 1968.

Appreciation is also extended to The Minnesota
Mining and Manufacturing CO., to the National Science
Foundation, and to the National Institutes of Health for
providing financial support.

Appreciation is also extended to Badische Anilin- und
Soda-Fabrik AG for their generous donation of cycloocta-

tetraene.

iii

 

9‘..- A

o.\ ‘ xv

PES'JLI‘S Ago I;

D

h.

(3

E.

A.
B.
C.

 

Von” _\.

UV. »-\

The S“:

 

Prfifle“. -
hexencxo
“Ethyl:

n ‘n‘.'
Tie a».‘

IAL

TABLE OF CONTENTS

PART I

STEREOCHEMISTRY OF THE RAPID EQUILIBRATION OF
PROTONATED BICYCLO[3.1.0]HEXENONES

Page
INTRODUCTION . . . . . . . . . . . . . . . . . . . . . 1

RESULTS AND DISCUSSION 0 O O O O O O O O O O C O O O O 15

A. The Synthesis and Characterization of endo-6-
Propy1-1, 3, 4, 5, 6-pentamethylbicyclo[3.1.0]-
hexenone (17) and exo-6- -Propyl-1, 3, 4, 5, 6-penta—.
methylbicyclo[3.1.0]hexenone (18) . . . . . 15

B. The Acid Catalyzed Rearrangement of 4-Methyl— —d
endo-6—propyl-1, 3, 5, 6- -tetramethylbicyclo[3. 1. O?-
hexenone (22) . . . . . . . . . . . . . . 28

C. The Acid Catalyzed Rearrangement of 4—Methyl—d3-
exo-6-propyl-1,3,5,6-tetramethylbicyclo[3.1.0]-
hexenone (2;) . . . . . . . . . . . . . . . . . 29

 

D. Mechanism of the Acid Catalyzed Rearrangement of
ER and 3"; O O O O O O O O O O O O O O O O O O O 30

E. Other Examples of 1,4-Sigmatropic Rearrangements 35

EXPERIMENTAL . . . . . . . . . . . . . . . . . . . . . 39
A. General Procedures . . . . . . . . . . . . . . . 39
B. General Photolysis Procedures . . . . . . . . . 39

C. Preparation of 6-Allyl-2, 3, 4, 5, 6-pentamethy1-
2 ,4-cyclohexadienone (20) and 4-Allyl-2, 3, 4, 5, 6-
pentamethyl-Z, 5-cyclohexadienone (21) . . . . . 40

D. Reduction of 6-Allyl-2,3,4,5,6—pentamethyl-2,4-
cyclohexadienone: Synthesis of 6-Propyl-2,3,4,
5,6-pentamethyl-2,4-cyclohexadienone (22). . . . 43

E. Reduction of a Mixture of 6-Allyl-2,3,4,5,6-
pentamethyl-Z,4-cyclohexadienone and 4-A11y1-
2,3,4,5,6—pentamethyl-2,5-cyclohexadienone . . . 45

F. Photolysis of 6-Propyl-2,3,4,5,6-pentamethyl-
2,4-cyclohexadienone (22) . . . . . . . . . . . 47

iv

A?

.4

 

CC

CF CC)

13‘
11'

H)

 

0)“)

‘)' " ’17

J: U! r-‘J
(I)

CT
(I)
y
W H

'r‘.

.\
.’U
’1

:J‘

var-3
um

\

U' U‘)
§ “ l

‘3
.T
I (D
a

J

t
J (
I

(D
25’
J
)
I
I

 

”715A? H
.RVU C:

If)

V-‘

TABLE OF CONTENTS (Continued)

Page

The Photolysis of a Mixture of 6-Propyl-2,3,4,
5,6-pentamethyl-2,4-cyclohexadienone and
4-Propyl-2,3,4,5,6-pentamethyl-2,5-cyclo-
hexadienone . . . . . . . . . . . . . . . . 49

The Synthesis of 4-Methyl- -d 3-endo-6-propyl-
1, 3, 5, 6-tetramethylbicyclo[3.1.0]hexenone (30) 50

The Acid Catalyzed Rearrangement of 4-Methyled3-
endo-6-propyl-1,3,5,6—tetramethylbicyclo[3.1.0]-
hexenone (22) . . . . . . . . . . . . . . . 50

The Synthesis of 4-Methyl-d3-exo-6-propyl-
1,3,5,6-tetramethylbicyclo[3.1.0]hexenone (2;) 51
The Acid Catalyzed Rearrangement of 4-Methylvd3—
exo-6-propyl-1,3,5,6-tetramethylbicyclo[3.1.0]-
hexenone . . . . . . . . . . . . . . . . . . 53

PART II

CONTROL OF THE STEREOCHEMISTRY OF 2,5-CYCLOHEXADIENONE

PHOTOISOMERIZATION BY STERIC FACTORS

INTRODUCTION . . . . . . . . . . . . . . . . . . . 57
RESULTS AND DISCUSSION . . . . . . . . . . . . . . 66
A. The Synthesis of 4-Propyl-2,3,4,5,6-pentamethyl-

2,5-cyclohexadienone (22) . . . . . . . . . 68
B. Photolysis of 4-Propyl~2,3,4,5,6-pentamethyl-
2,5-cyclohexadienone . . . . . . . . . . . . 69
C. The Synthesis of 4-Propyl-2,4,6-trimethyl-2,5-
cyclohexadienone (22) . . . . . . . . . . . 70
D. The Photolysis of 4ePropyl-2,4,6-trimethyl-2,5-
cyclohexadienone (22) . . . . . . . . . . . 74
E. The Synthesis of Isodurenol (117) . . . . . 80
F. Synthesis of 4-Propyl-2,3,4,6vtetramethyl-2,5-
cyclohexadienone (104) . . . . . . . . . . . 81
G. The Photolysis of 4+Propyl-2,3,4,6-tetramethyl-
2,5-cyclohexadienone (104) . . . . . . . . . 86
H. Synthesis of 4-Isobut l~2,4,6-trimethyl-2,5-
cyclohexadienone (122 . . . . . . . . . . . 89
I. The Photolysis of 4~Isobutyl-2, 4, 6- -trimethyl-
2 ,5-cyclohexadienone (122).. . . . . . . . 94

TABLE CF C321

h

NU)
U! ‘1
U
(I

U'
s.‘

mm

f1

uh 'U_

I I

#2.“

e a
'C'

t4
lot: tura
0'0.

EIGERIMEWAL
A. Gene:
3. Genera -

(n

 

fl .
u. Syntne:
2 p
ID-C‘!.;
DI Er‘thes
ID-:.v.c
o

 

 

 

‘5‘ A.\'~

C) Cl“r:e
trlfietn

H' RECLCt.

CYClche
2.4,5-.

The p}-
zls‘c‘JFC

SVh‘L

.A Lhes

CresCl

Synthes
Cyc thE

SYnthe:
Cyclch;

REQuCtl
CYCIP‘A

13'4.€
PhOt;

1t

TABLE

OF CONTENTS (Continued)

Synthesis of 4-Isobutyl-2,3,4,6-tetramethyl-
2,5—cyclohexadienone (123) . . . . . . . . .

The Photochemical Isomerization of 4-Isobutyl-
2,3,4,6-tetramethyl-2,5-cyclohexadienone (123)

Mechanistic Considerations in the PhotolySis of

2,5-Cyclohexadienones . . . . . . . . . .

EXPERIMENTAL O O O O O O O O C O O O O O 0 O O O O

A.
B.
C.

General Alkylation Procedure I . . . . . . .
General Alkylation Procedure II . . . . . .

Synthesis of 4-A11yl-2,3,4,5,6-pentamethyl-
2,5-cyclohexadienone (21) . . . . . . . . .
Synthesis of 4-Propyl-2,3,4,5,6-pentamethyl-
2,5-cyclohexadienone (23) . . . . . . . . .
Photolysis of 4-Propy1~2,3,4,5,6-pentamethyl-
2,5-cyclohexadienone (23 . . . . . . . . .

Synthesis of 6-Allyl-2,4,6-trimethy1-2,4-
cyclohexadienone (24) . . . . . . . . . . .

Rearrangement of 6~Allyl-2,4,6-trimethyl-2,4-
cyclohexadienone: Synthesis of 4-Allyl-2,4,6-
trimethyl-Z,5-cyclohexadienone (21) . . . .

Reduction of 4-Ally1-2,4,6-trimethyl-2,5-
cyclohexadienone: Preparation of 4-Prop 1-
2,4,6-trimethyl-2,5-cyclohexadienone (25’ .

The Photolysis of 4-Prop l-2,4,6-trimethyl-
2,5-cyclohexadienone (95' . . . . . . . . .

Synthesis of 2,4,6-trisvChloromethyl—meta-
cresol (118) . . . .1. . . . . . . . . . . .

Synthesis of Isodurenol (117) . . . . . . .

Synthesis of 6-Allyl-2,4,5,6-tetramethyl-2,4-
cyclohexadienone (119) and 6-Allyl-2,3,4,6-
tetramethyl-Z,4-cyclohexadienone (120) . . .

Synthesis of 4-Allyl-2,3,4,6-tetramethyl-2,5-
cyclohexadienone (121) . . . . . . . . . . .

Reduction of 4-Allyl—2,3,4,6-tetramethyl-2,5-

cyclohexadienone (121): Synthesis of 4-Propyl-

2,3,4,6-tetramethyl-2,5-cyclohexadienone (104)

Photolysis of 4-Propy1-2,3,4,6-tetramethy1-2,5-

cyclohexadienone (104) . . . . . . . . . . .

vi

Page

97

101

103

106
106
117

117

117

118

118

119 '

120

120

123

123

126

128

129

129

 

 

P. Synthe
cyc13:
Q. Reduc:
cycl:h'
trime:
R. h Ph.

 

T 13"":Vflf'f
’ .3HLV:
‘ ‘
IS”. .....
\.«

TABLE OF CONTENTS (Continued)

Page
Synthesis of 4-Methallyl-2,4,6-trimethyl-2,5-
cyclohexadienone (124) . . . . . . . . . . . . 132

Reduction of 4eMethallyl-2,4,6-trimethyl-2,5-
cyclohexadienone: Synthesis of 4-Isobutyl-2,4,6-
trimethyl-Z,5-cyclohexadienone (122) . . . . . 132

The Photolysis of 4-Isobutyl-2,4,6~trimethyl-
2,5-cyclohexadienone (122) . . . . . . . . . . 134

Preparation of 4eMethallyl-2,3,4,6-tetramethyl-
2,5-cyclohexadienone (132) . . . . . . . . . . 138

Hydrogenation of 4-Metha11yl-2,3,4,6-tetra-
methyl-2,5-Cyclohexadienone: Synthesis of 4-
Isobutyl-Z,3,4,6-tetramethyl-2,5-cyclo-

hexadienone (123) . . . . . . . . . . . . . . 138

Photolysis of 4-Isobutyl -2,3,4,6-tetramethyl-
2,5-cyclohexadienone (123) . . . . . . . . . . 140
(PART'III

THE SYNTHESIS AND PHOTOREARRANGEMENT OF
1,6,8,8,9,10-HEXAMETHYLTRICYCLO[Z.2.4.02I5]DECA-
3,9-DIENE-7-ONE AND ITS DERIVATIVES

INTRODUCTION . . . . . . . . . . . . . . . . . . . . 144

RESULTS AND DISCUSSION . . . . . . . . . . . . . . . 151

A.

B.

E.

The Synthesis of Cyclobutadieneiron Tri-
carbonyl (181) . . . . . . . . . . . . . . . . 151

The Synthesis and Characterization of 1,6,8,8,
9,10-Hexamethyltricyclo[2.2.4.02:5]deca-3,9-
diene-7-one (176) . . . . . . . . . . . . . . 153

Direct Irradiation of 1,6,8,8,9,10-Hexamethyl-

.tricyclo[2.2.4.02 5]deca-3,9-diene—7-one (176) 164

Sensitized Irradiation of 1,6,8,8,9,10-HeXae.
meth ltricyclo[2.2.4.02v5]deca-3,9-diene-7-one
(176) . . . . . . . . . . . . . . . . . . . . 168

Consequences of this Study . . . . . . . . . . 169

EDERIMMAL . O O C O C O O O O O O O C O O O C O O 1 71

A.
B.

Synthesis of Cis-3,4-Dichlorocyclobutene (179) 171
The Synthesis of Iron Nonacarbonyl . . . ... . 172

Vii

 

 

 

_ e. . .4 a; p. C.
.\\. v. r. p; h. ... A» .5 J .C A. E rlaL . ‘ r-,L .
C S. a S C S T. .2 2 .2 ... 2. l T...
C C .. C t . 2 C a. I...
e l e C e .l 2 4. K 1 E C E C 7
F L“ I .2 .... .2 I n: - T. I a. “y. C “:8
C pl T T C T t S 2 q. t T C .n C 1
n
O I O C O O O I
3 C D .L F G n: T.

 

LI 0G RAPLJY

‘u
1“

TABLE OF CONTENTS (Continued)

Page
The Preparation of Cyclobutadieneiron
Tricarbonyl (181) . . . . . . . . . . . . . 173
The Synthesis of 2,3,4,5,6,6-Hexamethyl-2,4-
cyclohexadienone (2) . . . . . . . . . . . . 173

The Synthesis of 1,6,8,8,9,10-Hexamethyl—
tricyclo[2.2.4.02I5]deca-3,9-diene-7-one (176) 174

Synthesis of 3-Methyl-d3-2,4,5,6,6-pentamethyl-

2,4-Cyclohexadienone (184) . . . . . . . . . 176
Synthesis of 9-Methyl-dawl,6,8,8,10-pentamethyl-
tricyclo[2.2.4.02r5]deca—3,9—diene-7-one . . 177
The Reduction of 1,6,8,8,9,10-Hexamethyltri-
cyclo[2.2.4.02:5]deca-3,9-diene-7-one (176). 177

Acetylation of 1,6,8,8,9,10-Hexamethyltri~
cyclo[2.2.4.02o5]deca-3,9-diene-7-ol (186 and
187) . . . . . . . . . . . . . . . . (“I'. . 178

The Photolysis of 7-Acetoxy—1,6,8,8,9,10-hexa—
methyltricyclo[2.2.4.02:5]deCa-3,9-diene (190

and 191) . . . . . . . . . . . . . . . . . . 179
Reduction of 9-Acetoxy—1,2,3,4,10,10-hexamethyl-
pentacyclo[4.4.0.0.1I 0.31407I3]decane (192) 179
Reduction of 9-Acetox -1,2,3,4,10,10-hexamethyl-
pentacyclo[4.4.0.0.1o 0.3I4O7I8]decane (193) 180
The Synthesis of 1,2,3,4,10,10-Hexamethylpenta—
cyclo[4.4.0.0.1:20.3r407v8]decan-9-one (177) 181

Direct Irradiation of 1,6,8,8,9,10-Hexamethyl-
tricyclo[2.2.4.02r5]deca-3,9-diene-7-one (176) 181

Sensitized Irradiation of 1,6,8,8,9,10-Hexamethyl—
tricyclo[2.2.4.02'5]deca-3,9-diene-7-one (176) 186

The Absence of Elemental Analyses for the New
Compounds Synthesized in this Study . . . . . 186

BIBLIOGRAPHY . . . . . . . . . . . . . . . . . . . . 189

viii

T.--“

.L-‘XDLL

 

(I)

(I)

(I)

(I)

U)

TABLE

II.

III.

IV.

V.

Nmr Spectra
Nmr Spectra
Nmr Spectra
Nmr Spectra

Nmr Spectra

LIST OF TABLES

ix

Page

25
72
84
91
99

 

10

11,

12,

13.

14_

 

Poss;,
caza-j

LIST OF FIGURES

FIGURE Page

1. Possible intermediate ions in the acid-
catalyzed rearrangement of (2) . . . . . . . 9

2. Intermediate ions of a ‘prvoting' 1,4-sigma-
tropic rearrangement of (2) . . . . . . . . . 12

3. Intermediate ions in a 'nOn-pivoting' 1,4-
rearrangement (2D . . . . . . . . . . . . . . 13

4. Scheme for the ‘pivot' 1,4-rearrangement of (22)31

5. Scheme for the 'non-pivoting' 1,4—sigma-
tropic rearrangement of (22) . . . . . . . . 32

6. The molecular orbitals inVOlved in the 1,4-
sigmatropic rearrangement . . . . . . . . . . 34

7. Nmr spectrum (CCl4) of 6-all l-2,3,4,5,6-penta-
methyl-2,4-cyclohexadienone 22) . . . . . . 42

8. Nmr spectrum (CC14) of 4-all 142,3,4,5,6-penta—
methyl-2,5-cyclohexadienone 22) . . . . . . 44

9. Nmr spectrum (CC14) of 6-propy1—2,3,4,5,6-
pentamethyl-2,4-cyclohexadienone (22) . . . . 46

10. Nmr spectrum (CC14) of 4-propyl-2,3,4,5,6—penta-
methyl—2,5-cyclohexadienone (22) . . . . . . 46

11. Nmr spectrum (CC14) of endo-64prop 1-1,3,4,5,6-
pentamethylbicyclo[3.1.0]hexenone 21) . . . 48

12. Nmr spectrum (CC14) of exo-6-propy141,3,4,5,6-
pentamethylbicyclo[3.1.0]hexenone (22) . . . 48

13. Nmr spectrum (C014) of 4-methyl-d3-endo-6-
propyl-l,3,5,6-tetramethylbicyclo[3.1.0]-
hexenone (22) . . . . . . . . . . . . . . . . 52

14. Nmr spectrum (CC14) of 22,. . . . . . . . . . 52

16.

17.

18.

19.

20.

21.

22.

23.

24.

26.

27

28.

29

30.

D

41
(D

(f '1
’1

 

LIST OF FIGURES (Continued)

FIGURES

15.

16.

17.

18.

19.

20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

30.

Nmr spectrum (CCl4) of 4-methyl-d3-exo-6-
propyl-l,3,5,6-tetramethylbicyclo[3.1.0]—

hexenone (22) . . . . . . . . . ._. . . . . .
Nmr spectrum (CC14) of 22'. . . . . . . . . .

Nmr spectrum (CCl4) of 4-allyl-2,4,6-trimethyl-

2,5-cyclohexadienone (21) . . . . . . . . . .

Nmr spectrum (CC14) of endo-6—propyl-1,3,6-tri-

methylbicyclo[3.1.0]hexenone (100

rwv

Nmr spectrum (CC14) of exo-6-prop l-1,3,6-tri-

methylbicyclo[3.1.0]hexenone (101 . . . . .

Nmr spectrum (CCl4) of 4-allyl-2,3,4,6—tetra-
methyl-2,5-Cyclohexadienone (121) . . . . . .

Nmr spectrum (CC14) of endo-6-propyl-1,3,4,6-
tetramethylbicyclo[3.1.0]hexenone (107) . . .

Nmr spectrum (CC14) of exo-6—propyl—1,3,4,6—
tetramethylbicyclo[3.1.0]hexenone (108) . . .

Nmr spectrum (CC14) of 4-methall 1—2,4,6—tri-
methyl-2,5-Cyclohexadienone (124) . .

WV

 

Nmr spectrum (CC14) of endo-6-isobut 1-1,3,6-
trimethylbicyclo[3.1.0]hexenone (125) . .
Nmr spectrum (CCl ) of exo-6—isobutyl-1,3,6-
trimethylbicyclo[3.1.0]hexenone (126) . . . .

Nmr spectrum (CC14) of 4-methally142,3,4,6-
tetramethy1-2,5-Cyclohexadienone (132) . . .

Nmr Spectrum (CC14) of endo-6-isobutyl-1,3,4,6-

tetramethylbicyclo[3.1.0]hexenone (128) . . .

Nmr spectrum (CC14) of exo-6-isobuty141,3,4,6-

tetramethylbicyclo[3.1.0]hexenone (129) . . .

Nmr Spectrum (CC14) of 1,6,8,8,9,104héxa-
methyltricyclo[2.2.4.02v5]deca-3,9-diene-7-
one (176) . . . . . . . . . . . . . . . . . .

Nmr speCtrum (CC14) of 1,2,3,4,10,10-hexa-

methylpentacyclo[4.4.0.0.1:30.3:407o3]decan-
9"One (177) o o o o o o o o o o o o o o o O 0

xi

Page

55

55

121

124

125

127

131

131

133

136

137

139

141

141

175

182

fl?
\Ji
'wr
...

\

‘
&

131‘ OF Fl
31.

La

7

C
1.

k»

:~
I .
V...

32.

V
.xrr

33.

 

LIST OF FIGURES (Continued)

FIGURE Page

31. Nmr spectrum (CCl4) of 1,3,3,4,5,6-hexamethyl-
tricyclo[2.6.0.01:4]deca-5,8-diene-2—one (196) 184

32. Nmr spectrum (CC14) of 2,3,3,4,5,6-hexamethy1-
tricyclo[1.6.0.01.7]nona-5,8-diene (197) . . 185

33. Nmr spectrum (CC14) of 1,2,7,8,8,10-hexamethyl-

tetracyclo[5.3.0.0.3I602v1°]deca—4-ene-9-One
(202) O C C O O O O O O I C O O O O C C O I O 187

xii

PART I

STEREOCHEMISTRY OF THE RAPID EQUILIBRATION OF PROTONATED

BICYCLO[3.1.0]HEXENONES

 

B‘.‘ t
‘
.75 e ESCE

t!
(f
31'
(D
'1
(1

'7

-
“’2‘“ no.
-..-‘."_ w‘
.-
‘ a DIRVVV
7‘ VAL!
. ‘

h—

(D

{H

'.I

 

 

 

INTRODUCTION

By the use of orbital symmetry, Woodward and Hoffmann1
have established a set of rules for predicting the outcome
of Concerted reactions, of which sigmatropic rearrange-
ments constitute one Class. These are defined as the migra-
tion of a o-bonded atom or group 4R) from one end to the
other of an allylic (n = 1) or polyenylic (n = 2,3, etc.)
chain of a neutral molecule and from one end to the other

of a polyenylic (n - 1,2, etc.) carbonium ion or carbanion.

R
\\C-(C= )n -—-—-9- (C=C)n-C”R neutral
\C-(C=C)n-C ————> C-(C=C)n-C/ cationic
R _ _ R
\C-(C=C )n-C ———-—> C-(C=C )n-C/ anionic

According to these rules,1t2 orbital symmetry constraints
for a neutral moiety are determined by the highest occupied
molecular orbital (HOMO) of the hydrocarbon radical corre-
sponding to the polyenylic chain for thermal rearrangements
and the lowest vacant molecular orbital (LVMO) for photo-
chemical rearrangements. On the other hand, orbital sym-
metry constraints for the cationic or anionic moiety are

determined by the HOMO of the polyene corresponding to the

(?

maze and

 

anionic mole

H

 

 

3
carbonium ion or carbanion chain over which R migrates
for thermal rearrangements (for n = 1, the polyene is buta-
diene) and the LVMO for photochemical rearrangements. Thus,
for a thermal 1,3-sigmatropic rearrangement of a neutral
molecule, the model v-electron system would be the allyl
radical with its HOMO being the Ta level. Similarly, a
thermal 1,4-sigmatropic rearrangement of a cationic molecule
would be determined by the second bonding MO, Y3, of buta-
diene and the thermal 1,4-sigmatropic rearrangement of an

anionic molecule would be determined by Y, of butadiene.

 

Y2, allyl radical Yz, butadiene Y3, butadiene

The nature of the orbital which binds the migrating
group R to the unsaturated system is also important. If
it is a symmetric orbital (for example, p) a! concerted
thermal 1,3-migration of a neutral moiety or a 1,4-migration
of a cationic species would require that R be bonded
simultaneously to C-1 on one side of the plane of the un-
saturated system and to C-3 or C-4, respectively, on the
other. This is an antarafacial process, difficult to achieve
when the carbon chain is short. However, in a concerted

thermal 1,4—migration of an anionic moiety, the transition

state (3837326’.

 

 

 

 

 

4

state geometry would be suprafacial (on one side of the
plane) if the R-binding orbital were of the below type

(case I), and since the connecting atom of R would be

 

Case I: R uses symmetric orbital or one lobe of anti-

symmetric orbital.

1,3: antarafacial.

or

   

1,4 (cationic): antarafacial with retention.

or

   

1,4 (anionic): suprafacial with retention.

or

   

 

 

case II: R

1,3: 5
1,4 (ca
1:4 (an

 

 

 

Case II: R uses both lobes of antisymmetric orbital.

1,3: suprafacial with inversion

 

1,4 (cationic): suprafacial with inversion

 

1,4 (anionic): antarafacial with inversion.

 

bound to only ggg_gggg, the configuration of R would be
retained during the course of the rearrangement. On the
other hand, if R uses 22£g_1obes of an antisymmetric
orbital (case II). the geometry would be reversed: 1,3-

migration and the 1,4(cationic)-migration would be

 

suprafaCial

 

when t:
‘ 1 4- 'r-vvv-«o-
5 I Sljlyc-

Int until 19

 

4
One POSS
Spelling Cf 3
V
as

6

suprafacial and the 1,4(anioniC)-migration would be antara-
facial. Moreover, since the migrating group R would bond
through opposite faces of its connecting atom, rearrange-
ment would invert the configuration of R.

When the original prediction was made, no examples of
a 1,4-sigmatropic shift existed in the literature. It was
not until 1967 that a possible example appeared. Swatton
and Hart3 discovered that treatment of 1,3,4,5,6,6-hexa-
methylbicyclo[3.1.0]hexenone 2 with sulfuric acid gave

2,3,4,5,6,6-hexamethyl—2,4—cyCthexadienone 2;

O
1 ;
N 2

One possible mechanism for this rearrangement is ring

 

opening of 2'to give 2, which is orbital symmetry forbidden
as a concerted process in the ground state although the
energy barrier may be low. Ion g'could undergo a methyl
shift to give 2, which would then deprotonate to yield 2;
Independent experiments had shown that this mechanism was.
unlikely, for the methyl shift from ion g'occurs in the

other direction, to give the cross-conjugated 2,

' _ fight-r]

l-.
1,1,.“ , ..

 

(H

 

O OH OH

mi: o _,

,1, £2, 4
M Me
0 0H
+
{L
2: é.
0
£5.

Another possible mechanism for the Observed rearrange-
ment would involve the well-known cyclopropylcarbinyl cation
rearrangement. Ion i can rearrange to 1 which then Opens

to Q; deprotonation again gives 2, Since the opening of 7

OH

2w

OPEHing bei:
To test this
Treatment :1

gave Predom:

-W ——-> ———>

£1

2‘]
201

(N

to Q'is symmetry forbidden, it seemed possible that the con-
verSion of g'to Z'could be rapid and reversible, with ring-
opening being the rate-determining step in the rearrangement.
To test this possibility, they3 prepared labelled ketone g”
Treatment of compound Q'with sulfuric acid for a short time

gave predominantly compound 2 with deuterium equally distri-

. o 0
~ —. 0/
+
m 18
1 min
on .
3 1.59 1.50 D *
1 05D
2 a 12.

buted between the methyl groups on C-4 and C-5, in addition
to a small amount of £2, The mechanism, which they proposed
for the rearrangement (see Figure 1), was a series of 1,4-
sigmatropic shifts. Quenching of a 50:50 mixture of ions Ll
and 1;} would give the observed product g'(i.e., a 50:50

of §,and Qf).

l l
“‘mu‘s

 

 

 

ken

 

 

Figure 1. Possible intermediate ions in the acid-catalyzed
rearrangement of g”

Two additional alternate mechanisms, which could not be
eliminated, would involve either successive 1,2-shifts or an
open carbonium ion. The mechanism of successive 1,2-shifts
involves the migration of one leg of the cyclopropane ring
to generate a cyclobutyl ion lg, followed by the 1,2-migra-
tion of the other leg as shown below to generate ion l2, By
a series of these successive 1,2-shifts, the cyclopropane

ring would move around the five-membered ring to give ll'.

5.3g

 

10

. g;

é” OH

Ti

6.— severalé‘9
—9 1,2-

shifts

 

In the open carbonium ion mechanism, cleavage of the
cyclopropane ring in ll'gives the tertiary cation lglwhich
subsequently readds to the cyclopentadiene ring to give 12,
By a series of these cyclopropane ring cleavages and re-"
additions, the three-membered ring moves around the five-

membered ring to give 11}.

11
OH OH + k
11 * 14 * 12

m
m m

1T

OH + OH
fi 6— Zpifiiiéi 23.1““ 9%
‘6 closings fi
* *

1|

m

Since the substituents at C-6 were identical, no dis-
tinction could be made between these mechanisms. However,
if the substituents at 0-6 were different, a distinction
between several of the alternatives would be possible, as
described below.

The 1,4-sigmatropic shift can occur in either of two
ways, which might be described as :pivoting', or 'non-
pivoting', each with different consequences for substituents
at C-6. In the'pivoting' 1,4-sigmatropic shift, the three-
membered ring moves around the five-membered ring by rota-

tion abOut one leg of the three-membered ring. This rotation

12

 

R1
0
-
CD3
g"
R R2 0
/
Q“ 12

Figure 2. Intermediate ions of a 'pivoting' 1,4-sigmatropic
rearrangement of gfl.

causes R1 and R2 to exchange positions for each movement
(if R1 was originally eggg, it becomes 2x2 to the five-mem-
bered ring and vice-versa for R3). Put another way, each
migration occurs with retention of configuration at C-6,
resulting in an interchange in the gngpygxg relationship

of R1 and R2. One would expect that quenching of a 1:1
mixture of ll'and 12'(R1 é R2) would give a 1:1 mixture

0f ’8‘," and gm ..

"' gin-I EJ'LK!

The c
suits. RC
loss of 51
the react:
isomers ::
C-4 and C-

In c:
1,4-sigma:
bered ring

that there

C-6 {"5
\lL R1

““9 durin

 

13

The open carbonium ion mechanism predicts similar re- .
sults. Rotation in the open carbonium ion causes complete
loss of stereochemistry at C-6. As a result, quenching of
the reaction mixture at equilibrium would give both stereo-
isomers of §f with deuterium equally distributed between
C-4 and C-5.

In contrast, as shown in Figure 3, the 'non-pivoting'
1,4-sigmatropic shift involves movement of the three-mem-
bered ring around the five-membered ring in such a manner
that there is no change in the orientation of the groups at
C-6 (if R1 was gndg, it remains egg9_to the five-membered

ring during the entire rearrangement; similarly, R2 remains

 

Rz '1
R2 R1 0 R3 R1 OH . HO ’
0 \fi <— C!
H -—>
2"" ' 2.9,

Figure 3. Intermediate ions in a 'non-pivoting' 1,4-rear-
‘ rangement Qf.

 

 

238331-3311 ST;

9.
readlt .

3

C )

f

F“

I);
U”
(D

)

"n
w“po‘~l.lds
Ce:

14
‘§§g). Each migration can be regarded as an intramolecular
8N2 displacement with inversion at C-6, thus maintaining the
exp-endo relationship of R1 and R2 at each step. Quenching
of a 1:1 mixture of ll'and lg (either R1 # R2 or R1 = R2)
would give a 50:50 mixture cf §f and §fm (i.e., g). The
mechanism involving successive 1,2-shiftspredicts the same
result.

All theoretical calculations predict that either the
'non—pivoting' 1,4-sigmatropic shift or the series of 1,2-
shifts is operative in the observed rearrangement. Wood-
ward-Hoffmannznfles predict that the 'pivoting' 1,4-sigma-
tropic shift in this system can only occur antarafacially
(case I is operative). This is impossible because a trans-
fused bicyclic system with three— and five-membered rings
would be generated. However, the 1,2-shifts and the 'non-
pivoting' 1,4-sigmatropic shift both occur suprafacially to
generate the gigffused bicyclo[3.1.0]system (for a full dis—
cussion see section D of Part I).

To distinguish between these alternative mechanisms,
compounds lz'and lg were synthesized and certain deuterated
derivatives of them were treated with acid. A description

Of these results constitutes Part I of this thesis.

17 18

«an—

RESULES AND DISCUSSION

A. The Synthesis and Characterization of endo-6-Propyl-
_TT§:JTSTG-pentamethylbicycloffi.1.0]hexenoneTlZl'and
gngS-Propyl-l,5j4,5,6-pentamethylbicyclo[3.1.0lhex-
enone (SET

 

 

 

 

The synthetic goal was a bicyclic[3.1.0]hexenone with
two different groups at C-6. Such compounds can be made by
the photolysis of appropriately substituted 2,4-cyclohexa-
idienones4 or 2,5-cyclohexadienones5. The choice of groups
on C-6 was governed by two factors: ability to distinguish
between the groups and assign their stereochemistry readily
by nmr spectroscopy, and ease of synthesis. Furthermore,
to insure similar chemistry, it was desirable that the
overall structure be closely related to the previously
studied hexamethyl compounds. Accordingly, it was decided
that one group at C-6 should be methyl and the other could
be either benzyl, allyl, or a reduced allyl group (propyl,
for example). i

It was anticipated that the required dienone precur-
Sors of the bicyclic ketones could be obtained through C-
alkylation of a substituted phenol. It has been found6 that
‘the highest yield of C-alkylation (as compared with o-alkyla-

tion) occurs when a heterogeneous medium is used with allyl

15

. 7 -. I
"-1— "' ‘1'

1,.— ‘VI‘ _.

‘Ew‘
cw"

or benzlv'l
alkylatix
which diff
group was
nmr signal
different
uents. T1".
The alkyla

the tier ive

 

 

 

16
or benzyl halides. Pentamethylphenol was selected for
alkylation to obtain, ultimately, bicyclo[3.1.01hexenones
which differ from l'only in one group at C-6. The benzyl
group was originally chosen because the phenyl and methylene
nmr signals of the benzyl group would occur at appreciably
different chemical shifts from those of the methyl substit-
uents. This initial attempt had to be abandoned,however.
The alkylation and photoisomerization steps went well, but
the derived photoproducts were difficult to separate and
could not be obtained pure.

The allyl group was not selected, for it might undergo
further reactions during the photolysis or during the acid-
catalyzed rearrangement of the derived bicyclo[3.1.01hex—
enones. Therefore, the propyl group was finally chosen as
the second substituent at C-6. Presumably it could be
readily obtained by selective hydrogenation of the corre—
sponding allyl dienone, and would not introduce any extran-
eous reactions during photoisomerization or acid-catalyzed
rearrangement. Its only drawback was that the nmr spectrum
might not be as distinctive as was desired, but this turned
out not to be a problem.

Pentamethylphenol lg’was alkylated by a modification of
the procedure of Curtin7 and Dybvig. A solution of lg,in
'toluene was added to a stirred dispersion of NaH in toluene
‘xnder nitrogen. This was followed by allyl bromide. The
Scflution was stirred overnight and the C-alkylated products

.‘Were purified by column chromatography. The column was

i
l 7 gurgii'il!

eluted Wit

pentane thy

 

changed to
20. After
w

the eluant
2}; On t‘:

eluant was

{Qandgl

 

17
eluted with hexane until the O-alkylated product, allyl
pentamethylphenyl ether, had been removed. The eluant was
changed to carbon tetrachloride to obtain a 67% yield of
22; After compound gg'had been removed from the column,
the eluant was changed to benzene to give a 15% yield of
2;, On the other hand, if, after eluting with hexane, the
eluant was immediately changed to benzene, a 4:1 mixture of

22 and 21 was obtained.

1 .NaH/ ¢CH3 +

 

 

\
19 22, El.
0 o
5% Rh/C X +
2.9. + 3.1. H2 7
1 atm.
22 251

 

u. A

 

[\3
N
+
523
f7:

 

ov_

22’
The t
by thEir S
2’4‘CYClch
iEtc
ma):
emu” 32
example

18

 

1. hv
MeOH O O
_*g537R \
-- a 0/ +
CH2C12
450 W
Pyrex 17 a
5% Rh/C hv
1 atm. 450W

Pyrex

The two isomers £30 and 21’ could be easily distinguished
by their spectral properties. The ultraviolet spectrum of

2 ,4-cyclohexadienones, for example g and 23? have maxima

o . o l

24 238

~

EtOH

hmax

= 330 (e = 4500) xfiggfi = 318 (e = 3800)

around 320 nm (6 ~ 4000) whereas 2,5-cyclohexadienones, for

eXample Q, 23, and 23, have maxima around 245 nm (s 2: 12,000).

 

  
     

The producf
C

at 333 nm
ture 2; ha:

Other
The nmr Spe
lets at T E
0'4, and c.
the allYlic
fine to the

from T 4.91

 

Cf one‘Pro
t0 the met}
assignments
and On the
methyleme E

lets becaus

19

The product assigned structure gg'had a maximum in methanol

0 o
.0
9,“ at” 221°
EtOH _ EtOH _ EtOH _
“max ‘ 246 7\max - 238 xmax — 246
(e = 14,600) (6 = 11,000) (6 = 14,500)

at 333 nm (e = 3870), whereas the compound assigned struc-
ture gl’had a maximum in methanol at 242 nm (e = 11,000).
Other spectral properties confirmed these assignments.
The nmr spectrum (CC14) of 22 showed a pair of broad sing—
lets at T 8.21 and 8.25 due to the allylic methyls on C-2,
C-4, and C-5, a broad three-proton singlet at T 8.10 due to
the allylic methyl on C-3, a three-proton singlet at T 9.00
due to the methyl group on C-6, a three—proton multiplet

from T 4.91 to 5.58 due to the vinyl protons, and a pair
6.0 Hz) at T 7.70 and 7.46 due

<>f one-proton doublets (J
'tc>the methylene protons of the allyl group on C-6. The

Eissignments are based on well established chemical shifts

and on the similarity of the spectrum to that of g. The

Inethylene protons of the allylic group appear as two doub-

‘163ts because they are adjacent to an asymmetric center.12

 

8.21br

8.10br

The r
0.5 Hz). at
all‘z’lic: rte
f9.08 due

°°° H2) at

 

group On c
5.79 due t
On the sirr.
3.9. and a.
slightly 1:
may be due
large: all

0

20

9.00(s) _ m. 0
8.2113113) 4.918?123i( \8.89(s)

 
   

8.10br 8) ‘,/’/) 7’46(d) 7.97(q
8.25br(S) .,r’§{14br(s)
49, ‘ 2. '

The nmr spectrum of 21 showed six-proton quartets (J
0.5 Hz) at T 8.38 and T 8.33 due to the a-allylic and 5-
allylic methyls reSpectively, a three proton singlet at
T 9.08 due to methyl group on C-4, a two-proton double (J
5.5 Hz) at T 7.80 due to methylene protons of the allyl
group on C-4, and a three-proton multiplet from T 5.10 to
5.79 due to the vinyl protons. The assignments are based
on the similarity of the spectrum to that of g” In both
29’ and 2.1,, the methyl group on the quaternary carbon is at
Slightly higher field than in the hexamethyl analog. This
Inay be due to conformational changes brought about by the

llarger allyl substituent, or by additional shielding by

the extra double bond.

0 o
8.38(q)(J=O.5Hz) 8.17(q)(J=0.8H2)

  

8.33(q) 8.05(Q)
9.08(s) 7'80(d) 8.79(s) ‘
5.10-5.79(m)
g; .Q

 

The 1
whereas th
These band.

are consiS'

 

 

21

The ir spectrum of 22'showed bands at 1630 and 1645 cm-1
whereas that of 22'contained bands at 1625 and 1655 cm-1.
These bands are characteristic of cyclohexadienones13 and
are consistent with the proposed structures.

Hydrogenation of 22'in cyclohexane over 5% rhodium on
charcoal at one atmosphere gave a quantitative yield of 22,
The spectroscopic properties of 22 were very similar to those
of 22, The ir spectrum contained bands at 1630 and 1640 cm'l.
Its uv spectrum in methanol showed a maximum at 335 nm (e =
4750). The nmr spectrum (CC14) showed a pair of broad sing-
lets at T 8.12 and 8.23 due to methyl groups on C-2, C-4,
and C-5, a singlet at T 7.95 due to the methyl group on Cr3,

a singlet at T 8.95 due to the methyl group on C-6, and a
seven-proton multiplet from T 9.05 to 9.40 due to the propyl

group on C-6.

9.05-9.40(m)
0
8.12br s)
8.95(s)
7.95(s)
423br(s)
22

AN

Also the 4:1 mixture of 22'and 22,0btained as described
above by benzene elution from the column of the crude alkyla-
tion product, was reduced in cyclohexane over 5% rhodium on

charcoal at one atmosphere to give a quantitative yield of

 

a 4:1 le'
As will he
separatist

in the 53‘?

2}: in net?
sPectrum
four ally;
methyl gr:

to 9-59 d1;

 

22
a 4:1 mixture of 22 and 22 as determined by vpc analysis.
As will be seen later, it was advantageous to postpone the

separation of this mixture until after the photolysis step,

in the synthesis of 11 and 12.
Compound 22 was purified by preparative vpc and identi-

fied by its spectral prOperties. These were, in fact, simi-

lar to those of compound 22. The ir spectrum of 22 had

cm-l. The ultraviolet Spectrum of

bands at 1620 and 1650
23 in methanol had a maximum at 250 nm (e = 8700) . The nmr
Spectrum (CCl4) showed a broad singlet at T 8.36 due to all

four allylic methyl groups, a singlet at T 9.20 due to the

methyl group on C-4, and a seven-proton multiplet from T 9.39

to 9.59 due to the propyl group on C-4.

8.36br(s)

  

9.20 s.) } 9.39-9.590!!!)

Compounds 11 and 12 were synthesized in two ways.
Irradiation of a1% solution of 22 in methylene chloride
through a pyrex filter using a Hanovia-L 450-W lamp gave,
after three hours, the two photoproducts in the ratio of
1:1.

In the other method, compounds 22 and 22' were not

separated because each could be photolyzed to give the

k_

 

desired 53
lyzed uni
photoproi
Irradiati
quartz ve
low press:

lete con;-

'(1

17:18. T":
W W

discussed
removed 1;:

dissolved

 

 

 

1% SOIUti:
lamp gave,

the twO L.

'U

PIDducts w
The t

v.
"‘e um: 51
at higher
:5 ~
2,8», '1‘:
USQd tC

u! the ’T

23
desired photoproducts, ’11, and 12. Each component is photo-
lyzed under conditions in which the other component and the
photoproducts are not affected by the light source used.
Irradiation of a 5% methanol solution of 22 and 22 in a
quartz vessel at 25372 in a Rayonet apparatus equipped with
low pressure mercury arc lamps gave, after three hours, com-
plete consumption of 22, unreacted 22, and a 9:1 ratio of

17 :22. The significance of this unusual ratio will be

 

discussed in Part II of this thesis. The methanol was
removed under reduced pressure and the reaction mixture was
dissolved in methylene chloride. Irradiation of the resulting
1% solution through a pyrex filter using a Hanovia-L 450-W
lamp gave, after five hours, complete consumption of 22 and
the two photoproducts in a ratio of 7:5. The two photo-
Products were separated by preparative vpc.

The two isomers 1.2 and 1.8.. could be distinguished by

their spectral properties. As noted by Miller and Margulies“,

the nmr signal for the egg-methyl group on C-6 of 22 occurs
at higher field than the signal for the Exp-methyl on C-6
of 22. Thus the chemical shift of the C-6 methyl group can
be used to decide whether the group is w (and shielded
by the 1r system of cyclopentenone ring) or 3:2.

The first photoproduct eluted by vpc, which was assigned

s‘lzructure 134 had an exo-methyl signal at “r 8.96 but no endo-

methyl signal near 1: 9.10. On the other hand, the other

photoproduct, which was assigned structure 12, had an endo-7

methyl signal at T 9.13 but no exo-methyl signal near 1' 8.95.

 

   

24

fifi ‘1’

AN

{Prue other assignments are based on the similarity of the
lunar spectra of iz'and lg’to the spectrum of 2; a complete
lgisting of the spectra, with assignments, is found in Table
I -

The ultraviolet spectra of 12 and 12 were similar to
tihe uv spectra of bicyclo[3.1.01hexenones 1 and 22, as shown

Iaelow.

o

0” f

l, 22.1 5

Kazan = 320 (e = 605) xfizgfi = 303 (e = 200)

= 274 (e = 3240) = 265 (e = 3290)

= 235 (e = 6270) = 220 (e = 5900)
film _ -1 film _ -1
Vmax — 1640 and 1690 cm Vmax - 1630 and 1695 cm

__:'.

 

25

a
Table I. Nmr Spectra

 

 

 

 

 

Compound Chemical Shift (J)b Assignments
9.08 s C-6 endo-methyl
8.90 s C-6 eggfmethyl
8.90 s C-1 methyl
8.78 s C-5 methyl
8.45 .q, 0.9g C-3 methyl
8.12 q, 0.9 C-4 methyl
9.05-9.40 m; } C-6 propyl
8.80-8.95 gm
8.96 s C-6 methyl
8.91~ s C-1 methyl
8.78 s C-5 methyl
8.45 q, 0.9§ C-3 methyl
8.12 q, 0.9 C-4 methyl

11'

9.00-9.25 2mg} C-6 propylC

8.55-8.70 m

9.13 s; C-6 methyl

8.85 s C-1 methyl

8.75 s), C-5 methyl

8.45 q, 1.23 C-3 methyl
C-4 methyl

8.10 q, 1.2

 

Continued

/
f

 

./

Im

26

Table 1. Continued.

 

 

 

Compound Chemical Shift (J)b Assignments
9.08-9.38 gm; } C-6 propyl
8.72-8.88 m
8.94 s ' C-6 methyl
8.88 s d C-1 methyl
8.78 s C-5 methyl
8.42 s d C-3 methyl
8.09 q, 1.0) C-4 methyl
8.95-9.25 m; C-6 propylC
8.55-8.70 m
9.11 s C-6 methyl
8.90 s d C-1 methyl
8.79 s C-5 methyl
8.46 s d C-3 methyl
,8.15 q, 1.0) C-4 methyl

 

s2.

 

aAll spectra are in CCl4.

bShifts are reported in T values and J's in Hertz. Multi—

plicity of peaks is shown in brackets: s, singlet; q,
quartet; m, multiplet.

cThe C-6 methyl and C-6 propyl groups have absorptions
overlapping one another. As a result, only the total area
of ten protons for both groups can be measured.

dIntegrates for 1.5 protons.

 

 

PT'

27

0 0
-
\
1.2, ' 18
MeOH _ ' _ MeOH _ " _
xmax — 332 (e — 105) Kmax — 332 (e — 116)
= 268 (e = 1025) = 270 (e = 1990)
= 236 (e = 3220) = 236 (e = 5840)
film _ ' -1 film _ ' —1
Vmax — 1640 and 1685 cm Vmax — 1640 and 1685 cm

The infrared spectra of both 11 and 12 showed bands at

1640 and 1685 cm-1. These bands are consistent with the

bicyclo[3.1.01hexenone structures.

In the acid catalyzed rearrangement of bicyclo[3.1.0]-
hexenone, it is necessary to ascertain that equilibrium has
occurred in order to eliminate some of the proposed mechan-
isms. The various mechanisms predict different results at
equilibrium if the starting material, the bicyclo[3.1.0]-
hexenone, is labelled at C-4 or C-5. Compounds 11 and 12’
were labelled at C-4 by base catalyzed exchange between
CH,OD and the C-4 methyl groups of lz'and 12;

4eMethy1-d3figggng-propyl-l,3,5,6-tetramethylbicyclo-
[3.1.0]hexenone 22 was prepared by refluxing unlabelled 11'
in CH50D, to which some sodium had been added, for two hours
under nitrogen. The nmr spectrum of 22 showed a 95% reduc-
tion in the signal at T 8.12; also the quartet at T 8.45

sharpened to a singlet. The remainder of the spectrum was

mesae
mthy l'da

hexenone

tion in ‘
sharpene:
the save

Vpc anal'

Tre
4.00 f3.r
ratio of
Of reten

labelled

tian 1i“.

28

the same as that of the unlabelled compound (11). No 4-
methYl-dsfg§gf6-propyl-1,3,5,6-tetramethylbicyclo[3.1.0]—
hexenone 2l'was present, by vpc analysis.

Likewise,‘4-methyl-d3:22276-propy1-1,3,5,6-tetramethyl-
bicyclo[3.1.0]hexenone 2l’was prepared by refluxing unlabel-
led lg'in CHSOD, containing dissolved sodium, for two hours
under nitrogen. The nmr spectrum of 22 showed a 95% reduc-
tion in the signal at T 8.10; also the quartet at T 8.45
sharpened to a singlet. The remainder of the Spectrum was
the same as that of unlabelled 12; No 22 was present, by

vpc analysis.

B. The Acid Catalyzed Rearrangement of 4<Methyl-d3-endo-

 

6—propyl-1,3,5,6-tetramethylbicyclo[3.1.0]hexenone 22;

 

Treatment of 22 with concentrated sulfuric acid at
4.00 for 2 minutes gave an eighty per cent yield of an 8:1
ratio of 22 and labelled 22, by vpc analysis. By comparison
of retention times with authentic unlabelled samples, no

labelled lg'could be detected by vpc analysis. The detec-

 

o
\ $129.1...)
40 '
* 2 min
30 22..

tion limit was 0.001%.

(‘J

fer

29

Labelled 22’was not collected for nmr analysis, be-
cause the amount was very small and no useful mechanistic
information would be obtained from its spectra. Compound
22 was collected by preparative vpc. Its nmr spectrum was
identical with that of lzzexcept that the methyl signals
for C—4 and C-5 only integrated for 1.5 H each (see Table I).
This demonstrated that equilibration had occurred without
epimerization at C-6. A discussion of this result will be

reserved for Section D.

C. The Acid Catalyzed Rearrangement of 4-Methyl-d3-exo-6-
propyl-1,3,5,6-tetramethylbicyclo[3.1.0]hexenone 22;

 

 

Reaction of 22 with concentrated sulfuric acid at 4.00

for two minutes gave a seventy-five per cent yield of a 9:1

 

ratio of 22,and labelled 22, by vpc analysis. A vpc
analysis of the crude product showed that no labelled 12’

was present, up to the limit of the detector (0.001%).

30
Only compound 22’was collected for nmr analysis. The
nmr spectrum of 22'was identical with that of compound 22
except that the methyl signals for C-4 and C-5 integrated
for only 1.5 H each (see Table I), which demonstrated that

complete equilibration had occurred.

D. Mechanism of Acid Catalyzed Rearrangement of 22 and 22,

 

To summarize the experimental findings briefly, it was
observed that when either 22 or 22 was separately treated
with concentrated sulfuric acid at 0° for two minutes some
rearrangement of each (about 10%) to the conjugated dienone
22'occurred. But the recovered bicyclic ketones maintained
their orientation at C-6, and they had the label equally
distributed between the C-4 and C-5 methyl groups (222,,
the product from the acid treatment of 22'could be repre-
sented by structure 22). This demonstrated that C-6 had
in fact migrated around the cyclopentenone ring, to equi-
librium. An identical result was obtained with compound 22”

These results rule out the 'pivot' 1,4-sigmatropic
rearrangement mechanism, for it predicts in the experiment
with 22'the recovery of 22'labelled only at C-4, as well
as the formation of 22) labelled only at C-5 (see Figure 4).
This was not observed.

The open carbonium ion mechanism also can be eliminated
as a possibility, because it preducts loss of stereochemistry

at C-6. As a result quenching of the reaction mixture from

 

rte tree
50:50 It.

.k‘
distrls.

Figure

fiVe -me‘
entatio)

the Ste;

31
the treatment of either 22'or 22'with acid should give a
50:50 mixture of the two isomers, with deuterium equally

distributed between C-4 and C-5, an unobserved result.

Q- S’ 3" :2 AT
i

30 34

HO._
~95

Figure 4. Scheme for the ‘pivot' 1,4-rearrangement of 22;

 

5??

On the other hand, the 'non-pivoting' 1,4-sigmatropic
shift requires that the three-membered ring move around the
five-membered ring in such a manner that the exo-endo ori-
entation at C-6 is retained throughout the rearrangement,

the stereochemistry at C-6 being inverted after each movement.

 

003

Figure

32
This mechanism predicts that, at equilibrium, quenching
should give a 1:1 mixture of 22'and 22}, whose nmr Spectrum
would be equivalent to the nmr spectrum of 22, the observed
result. A similar result can be predicted for the rearrange-
ment of 22; These experiments definitely Show that, if the
reaction involves a 1,4-sigmatropic shift, it must be of

the 'non-pivoting' type.

0 0H , OH
\ H+ ‘
—>_: ——->
-H+ *-
CD3 *
30 34

O OH HO
_H+
' so:-
F’\ “’
36

19'

rw

Figure 5. Scheme for the 'non-pivoting' 1,4-sigmatropic
rearrangement of 22;

the In:

able.

33

A series of 1,2-Shifts cannot be eliminated aS-a pos-
sible mechanism for the observed rearrangement because the
'non-pivoting' 1,4-sigmatropic shift and the 1,2—shifts
predict the same result: the exo-endo orientation at C-6
is retained during the rearrangement. No experiment has
yet been devised that is able to distinguish between these
two mechanisms.

The preference for the 'non-pivoting' 1,4-sigmatropic
rearrangement over the 'pivoting' 1,4-sigmatropic rearrange-
ment is understandable if one inspects the molecular orbitals
involved in the two possible mechanisms, The only differ-
ence between these two mechanisms is whether the front or
back lobe of the C-6 carbon is involved in the rearrangement.
The molecular orbital for both rearrangements is 22 of
butadiene. In the 'pivoting' mechanism (224 224 22),
the C-6 orbital bonding to C-5 is the same orbital rebOnding
to the C-2 orbital after migration. As shown in Figure 6,
the migration must occur antarafacially, which would give
the geometrically impossible trans-fused three- and five—
membered ring structure 22, By contrast, in the 'non-
pivoting' mechanism (22+ ’42 4 22,): the back lobe of the
C-6 orbital is attacked by the C-2 orbital in a suprafacial
manner to give the gigffused three- and five-membered ring
22; By these considerations, the exclusive preference for
the 'non-pivoting' 1,4-sigmatropic rearrangement is reason—

able.

29 A? 6 96" 9 9"
HQ?" 4’??? 60m
6 '0 0 6 6 6
21 s: 2.2.

Q

35

B. Other Examples of 1,4-Sigmatropic Rearrangements.

 

While this work was in progress, a few reactions were

reported in the literature which involved rearrangements

in which a single 1,4-sigmatropic Shift occurred. Zimmer-
man16 and his workers treated 2-bromo-6fgggfphenyl-6figggg-
pfbromophenylbicyclo[3.1.0]hexan-3-one 22 with one equiva-
lent of potassium Efbutoxide to give a 76% yield of 65322-
' phenyl-6fggggfipfbrom0phenylbicyclo[3.1.0]hexenone 22; None
of the other isomer gz'was detected. Other tests of the

stereochemistry at C-6 include the rearrangements of 22}5

and g2f7.
Br
0 Br
°./ “—4" /
_t_-BuOH
Br
44 :2.

 

36

48 49

 

w
H
5
5
. {firm-4 "2.9-..-

For the rearrangement of 22 or 22” Zimmerman postulated
that the base abstracts a proton to form the bromo enolate
22, followed by loss of bromide ion to give zwitterion 22,
which, in turn, then undergoes a 1,4-sigmatropic shift as
Shown below (only the rearrangement of 22 is shown on page 37
but the mechanism for gg'is the same). A similar mechanism
can be written for the rearrangement of 22;

Two examples of multiple 1,4-sigmatropic Shifts have
recently been described. Winstein and Childs,18 who in-
.vestigated the low temperature nmr spectrum of 22, observed
that as the temperature was raised the signals for the five
methyl groups on the cyclopentenyl ring coalesced to a
singlet at -48° whereas the signals for the two methyl

groups on C-6 remained distinct up to -9°. Koptyug19

and h;

the 1:

raw“:
9; JUL/t
O

00 whe
cn C-6
of Sn:

eXplai

38
and his workers found a similar result when they studied
the low temperature nmr spectrum of 22, The five methyl

groups on the cyclopentenyl ring coalesced to a singlet at

53

0° whereas the signals for the hydrogen and methyl groups
on C-6 remained distinct up to 25°. In both cases a series
of suprafacial 1,44sigmatropic shifts was postulated to

explain the observed nmr spectra.

 

 

Sac} pi
trsmete
J.E.O.i
Wise n
given

With a
Spectz
Perkin
Chrome
with E
Analys

Ann AI

the So;
a sePtu

the fees

EXPERIMENTAL
A. General Procedures

Infrared spectra were taken as neat smears between
NaCl plates or in CCl4 solution with a Unicam SP-200 spec-
trometer. Nmr spectra were obtained on a Varian A-60 or
J.E.O.L. C-60H spectrometer in CC14 solution (unless other-
wise noted) with tetramethylsilane as an internal reference,
given the value T = 10.00. Ultraviolet spectra were obtained
with a Unicam SP-800 or a Beckmann DB spectrometer. Mass
spectra were determined by Mrs. R. Guile with a Hitachi-
Perkin-Elmer RMU-6 mass spectrometer. Varian Aerograph gas
chromatographs were used. Melting points were determined
with a Gallenkamp Melting Point Apparatus and are uncorrected.
Analyses were performed by Spang Microanalytical Laboratories,

Ann Arbor, Michigan.
B. General Photolysis Procedure

For all irradiations carried out in a Rayonet reactor,
the solutions were placed in quartz test tubes, sealed with
a septum cap, and deoxygenated by bubbling purified nitrogen
through them. The test tubes were placed in the center of

the reactor and irradiated with the appropriate lamps. The

39

tempe

C38 1'

40
temperature was maintained between 25-300 by a fan inside
the reactor.

All other irradiations were carried out with a 450
watt Hanovia Type L mercury lamp. The solutions to be
irradiated were placed in quartz test tubes, sealed with a
septum cap and deoxygenated by bubbling purified nitrogen
through them. The test tubes were secured to the outside
of a quarts well into which the mercury lamp, surrounded by
the appropriate filter, was placed. The temperature was
maintained between 18 to 25° by a water bath.

The reactions were monitored by withdrawing aliquots
of the solutions and examining these by vapor phase chroma-
tography (vpc).

For each irradiation a dark reaction was carried out
in which part of the initial photolysis solution was stored
in a quartz test tube for one week. After this time the
solvent was evaporated and spectra were taken. No dark re-
actions were observed nor is there evidence for thermal
rearrangement of any of the starting materials or products

on the vpc.

C. Preparation of 6-Allyl-2,3,4,5,6-pentamethyl-2,4-cyclo-
hexadienone (£2) and 4-Allyl-2,3,4,5,6-pentamethyl-2,5-
cyclohexadienone (2;):

 

 

In a 500-ml three-necked round-bottomed flask, fitted
with a glass stopper, a 250 ml dropping funnel, and a three—

way stopcock attached to a nitrogen line, were placed 170 ml

hi.)

chra
Sil.

tru:

Ethel
the (
pm:

of 6C

1645

41
of toluene and 4.2 g (0.100 mole) of a 56% mineral e11 dis-
persion of sodium hydride. To this magnetically stirred
solution, which was flushed continuously with nitrogen, 16.4
g (0.100 mole) of pentamethylphenol lg'in 110 m1 of toluene
was added by the use of the dropping funnel. After addition
was complete (30 minutes), the solution was stirred for
five hours; subsequently, 25 ml (0.29 mole) of allyl bromide
was added, and the solution was stirred at room temperature
overnight.

The resulting mixture was washed with water, and the
organic layer was dried over anhydrous magnesium sulfate.
The drying agent was filtered, and the toluene was removed
under reduced pressure to give a yellow oil. This oil was
chromatographed on a 4 x 80 cm column of 60-100 mesh fluori-
sil. The column was eluted with hexane until the ir spec-
trum of the eluted material showed no band at 1180 cm-1,
which was due to the presence of allyl pentamethylphenyl
ether. Eluant was then changed to CCl4 to elute gglfrom
the column (67%). The analytical sample was obtained by vpc
purification and trapping (2' x 1/4" SE-30, 150°, He flow
of 60 ml/min).

Anal. Calcd. for C14H200: C, 82.30; H, 9.87.

Found: C, 82.22; H, 9.92.

The infrared spectrum of EQ'Showed bands at 1630 and
1645 cm-1. The uv spectrum of gg'in methanol had a maximum
at 333 nm (e = 3870). The nmr spectrum (CCl4) showed a pair

of broad singlest at 1 8.21 and 8.25 (9H), a broad singlet

42

.«mmv maocoflomxmn

. m
Ioaumolfi.NIH%LuoEmucmm|®.m.v.m.NIH>HHMI® mo AvHUUV Eduuommm H82 b on: am

Pt~+hki~b>h ~P>xP+—FLLP~’>>rrP>?P —LID(> bl L+"h—L LLi—D b! a} bLDD — 57> bfb

o... +0.0 0.” . OK 0.0 A L. v .59..- 9m . 0.? .

._.... . . . . _ .
3.». _ Vigifiefitfl .gdiihbigékgégi 39 . wixyf‘ft’eeoiiggg 31;}???
.m

2

4‘ l.

K _ _
m

qun

.

 

 

 

~_~.-_

_. -—— —_.._._

 

h m"

_
m
M
.
_
_
_
.
_

 

 

 

43
at r 8.10 (3H), a singlet at T 9.00 (3H), a multiplet from
T 4.91 to 5.58 (3H), a pair of doublets (J = 6 Hz) at T 7.70
(1H) and 7.46 (1H).

Elution with benzene gave 4-allyl-2,3,4,5,6-pentamethyl-
2,5-cyclohexadienone 21 (15%). The analytical sample was
collected by vpc (same conditions as above).

Anal. Calcd. for C14H200: C, 82.30; H, 9.87.

Found: C, 82.42; H, 9.92.

The infrared spectrum of 21 showed bands at 1625 and
1655 cm-1. The ultraviolet speCtrum of 21 in methanol had
a maximum at 242 nm (e = 11,000). The nmr spectrum (CC14)
consisted of quartets (J = 0.5 Hz) at r 8.38 (6H) and 8.33
(6H), a singlet at T 9.08 (3H), a doublet (J = 5.5 Hz) at

r 7.80 (23), and a multiplet from T 5.10 te 5.79 (33).

D. Reduction of 6-Allyl-2,3,4,5,6-pentamethyl-2,4-cyclo-
hexadienone: Synthesis of 6-Propyl—2,3,4,5,6epenta-

 

methyl-2,4-cyclohexadienone (22).

 

In a sloping manifold assembly at one atmosphere, 0.75
g of gg'was hydrogenated over 50 mg of 5% rhodium on char-
coal in 30 ml of cyclohexane. Reaction was run until the
uptake of hydrogen had ceased (3 hours). After the catalyst
'was filtered, the solvent was removed under reduced pressure
to give a quantitative yield of 22, The analytical sample
‘was obtained by vpc trapping (5"2 1/4" SE-30, 180°, He flow
of 120 ml/min).

Anal. Calcd. for C14H220: C, 81.50; H, 10.75.

Found: C, 81.63; H, 10.77.

44

.mmmv 0coomwomxm£

 

 

 

——‘_—_.'—.~_.-

 

 

 

 

 

 

IoHomolm.N|Hm£umEmucomlm.m.v.m.m|H>HHMIv mo “vauoy Esuuowmm HEZ .m musmflm
(. . »-_ .._.. .L p .d.» (H .(+(” r()(-_. (r . .L .)r—f. +-Ll.l r._ .. .,. . . .m
@p Qo 9w QR o3 .L;2%_ ed ow . on
.2... . _. , it.
1...; .IX../ .1: (31).) v 5 3 73.1.1 1
v . m 1 .fi .
. . u 1 \t 2:: t \l- - ..

45
The infrared spectrum of 22 showed bands at 1630 and
1640 cm-1. The ultraviolet spectrum of 22'in methanol showed
a maximum at 335 nm (e = 4750). The nmr Spectrum (CClg)
showed a pair of broad singlets at T 8.12 and 8.23 (9H),
a singlet at T 7.95 (3H), a singlet at T 8.95 (3H), and a

multiplet from T 9.05 to 9.40 (7H).

E. Reduction of a mixture of 6-Allyl-2,3,4,5,6-pentamethyl-

 

2,4-cyclohexadienone and 4-Allyl-2,3,4,5,6-pentamethyl-

2,4-cyclohexadienone.

 

In a sloping manifold assembly at one atmosphere 0.75 g
of a mixture of 22’and 22'was hydrogenated over 50 mg of 5%
rhodium on charcoal in 30 ml of cyclohexane. This mixture
of 2Q'and 22 was obtained by benzene elution immediately
after hexane elution in the column chromatography of the
crude product from the reaction of sodium pentamethylphen—
oxide and allyl bromide. Reaction was run until the uptake
of hydrogen had ceased (3 hours). After the catalyst was
filtered, the solvent was removed under reduced pressure to
give a quantitative yield of 22'and 22;

The vpc trace of the reduCtion mixture (5' x 1/8" SE—30,
195°, 75 ml/min of He) showed two products in the ratio of
4:1. The major component was identical in retention time
and spectroscopic properties with 22, The minor component,
which was 22, had an infrared spectrum with bands at 1620
and 1650 cm*1. Its ultraviolet spectrum in methanol had a

maximum at 250 nm (e = 8700). The nmr Spectrum (CC14) showed

46

. 3th 020:0
twomxmsoHomolm.NIH>£uoE
Imucmmlm.m.¢.m.NIH>QOHm .

um mo AvHUUV Esuuowmm HEZ .oH ousmfim

 

'nD’tp'blllrLIl‘lrp r b rh hLL b —.t-b‘rr{.r
- @— 99 ed
‘1’} "10
.J. \)..\. .\ ll) )1 s . .s
I. w. ’- z' «A a s
.. N~ .. .‘ 1.... \\ I.
L l . .\
- ,e . l .. x
l. _ .\
on ~
1 e
m:
a
a
r
T
t
\\ m
\|.I.l-

 

.«va maoom
Iflomxmsoaomolv.mlamnumfi
Imucomlm.m.¢.meulammoum

am no mvHUUV Esuuommm H82

.m mnsmfim

 

 

 

 

 

 

 

 

 

 

 

—
u

_. ._¥.

47
a broad singlet at T 8.36 (12H), a singlet at 1 9.20 (3H),
and a multiplet from T 9.39 to 9.59 (7H). Trapping by Vpc
(same conditions as above) gave the analytical sample.
Anal. Calc. for C14H220: C, 81.50; H, 10.75.

Found: C, 81.41; H, 10.71.

F. Photolysis of 6-Propyl-2,3,4,5,6-pentamethyl-2,4-cyclo-

 

hexadienone (22).

 

A solution of 0.47 g of 22 in 50 ml of methylene chlor-
ide was irradiated using a Hanovia-L 450W lamp with a pyrex
filter for three hours. A vpc trace (5' x 1/4" SE-30, 190°,
80 ml/min of He) of the reaction mixture showed a 1:1 ratio
of two photoproducts. The two compounds were separated by
gas chromatography (10' x 1/4" 20% DEGS, 150°, 100 ml/min
of He).

The first component, which had a retention time of
12.8 minutes, was endg-G-propyl-l,3,4,5,6-pentamethylbi-
cyclo[3.1.0]hexenone (£1). The infrared spectrum of 21
showed bands at 1640 and 1685 cm‘l. The nmr spectrum (CC14)
showed quartets (J = 0.9 Hz) at T 8.12 (3H) and 8.45 (3H).
singlets at T 8.91 (3H), 8.78 (3H), and 8.96 (3H), and a set
of two multiplets from T 8.80 to 8.90 and T 9.05 to 9.40
(7H). The ultraviolet spectrum of lz'in methanol showed
maxima at 332sh (e = 105), 268 (e -'1025), and 236 nm (e =
3220). Collecting by vpc (same conditions as for separating
the photOproducts\ gave the analytical sample.

Anal. Calcd. for C14H220: C, 81.50; H, 10.75.

Found: C, 81.?5: H, 10.68.

 

48

0..

 

.Amfiv «no
uemxmnflo.H.m_oaomoflnamsuws
unueomae.n.e.m.Hnflmnoneue
mem.mo Avaoov Eduuommm HEz

 

ital . .9.
4. “mg.-. _
.

m M
. a 5...." .... 3,... . n.
. . . . 1.x.
_:_L.f:vef
.cfi— .....nn.
- C

 

 

 

 

 

 

 

 

 

‘
2 ' u'- .
. u
.. .
a
H.
a .—
.. our
5 .lo . . ....
_. . .|,. ..u.. m...
c ..I 1..
In . . 3— I... .n o . . .. .... u . . 7
u... .... .2... _ ....JIT.
...... “.... ...... Ni 47:...
...: fr... -n ...h . .. __ .1 _ .....1
.u ..m a... . . .. ..... u..
o w..... .. , ...- . . ... e a...
.... ..n . o — _ . ___.
o. a q t o- ...
.. .. .. .. . ._. .m. .-
. . . . ..... .u
._ . .... .. ... ....... ..
-.e.-:... green,.:.__
1.. _ ._ ...... .. ..-.17
it? .. .._. :r if... ...
.. ..xr;,.tl 1.;wig.e.sl
... w. . _. .. ........._.._.M_ ... 3...
. .H . . . ......d.,h.... w. nut“.
. _ ... 1. .. .. ...t: . ..-
. ... .M... of w .. H e
.. . ...md. .. ”Quit. .
n .... ._ . ,..,
' .c 0.. '0 .0 . o .
l. m....... .u . ..—.

.NH wusmflm

.mNmV mco
Icwxosno.H.mHoHU>UwQH%£umE
unuemmue.n.e.m.Huasmonmue
lovaw mo AvHUUV Eduuowmm HE?

 

_2&%
. .. .
.11.

 

0..

_L

..."-
“P: _. _
‘ .

 

 

 

 

 

o.
m.~__.HJ__14AAA«41——J«Aul—W—-

.HH muzmflm

49

The other component, which had a retention time of
15.0 minutes, was ggg—6-propyl-1,3,4,5,6-pentamethylbicyclo-
[3.1.0]hexenone 12, Its infrared spectrum showed bands at
1640 and 1685 cmfi1. The nmr spectrum (CC14) showed quartets
(J = 1.2 Hz) at T 8.10 (3H) and 8.45 (3H), singlets at T
8.75 (3H), 8.85 (3H), and 9.13 (3H), and a set of two multi-
plets from T 8.55 to 8.70 and 1 9.00 to 9.25 (7H). Its uv
spectrum in methanol showed maxima at 332 Sb (6 = 116), 270
(e = 1890), and 236 nm (e = 5840). Trapping by vpc (same
conditions as above) gave an analytical sample.

Anal. Calcd. for C14H220: C, 81.50; H, 10.75.

Found: C, 81.52; H, 10.84.

G. The Photolysis of a Mixture of 6-Pr0pyl-2,3,4,5,6-

 

pentamethyl-Z,4-cyclohexadienone and 4-Propyl-2,3,

 

4,5,6-pentamethyl-2,5-cyclohexadienone.

 

A solution of 3.05 g of the mixture was dissolved in
70 m1 of methanol in a quartz test tube and irradiated at
25373 in a Rayonet low pressure mercury arc lamp. The pho-
tolysis was followed by gas chromatography (5‘ x 1/4" SE-30,
180°, 120 ml/min of He). After 22 had essentially been con-
sumed by the photolysis, the methanol was removed under
reduced pressure. The crude product was dissolved in 200 ml
of methylene chloride and irradiated using a Hanovia-L 450W
lamp with a pyrex filter for five hours. Again the photo-
lysis was followed by gas chromatography (same conditions as

above). After the photolysis was complete, the solvent was

50
removed under reduced pressure to give 3.01 g of crude
product. A vpc trace of the reaction mixture showed again

the same two photoproducts 11 and 12,

H. The Synthesis of 4-Methyl-d3-endo-6—propyl-1,3,5,6-
tetramethylbicyclot3.1.0]hexenone (22).

 

 

A solution of 383 mg of 12.1“ 15 ml of deuteromethanol
was treated with0.2 g of sodium metal under an atmosphere
of nitrogen. The solution was heated under reflux for one
hour, and the deuteromethanol was removed under reduced
pressure. The experiment was repeated with an additional
15 ml of deuteromethanol, and after one hour, the solvent
was evaporated under reduced pressure. The residue was
dissolved in methylene chloride and washed with cold (0°)
water three times. The organic layer was dried over anhy-
drous magnesium sulfate and evaporated to give 363 mg (95%)
of a pale yellow oil, which consisted of a single isomer
in 95% purity by vpc analysis (10' x 1/4" DEGS, 160°, 75
ml/min of He). The vpc analysis detected none of 22, The
nmr Spectrum of the oil showed greater than 95% of the label
on the methyl in the 4-position and consisted of methyl
singlets at T 8.45, 8.82, 8.92, and 8.98 and a set of two

multiplets from T 8.81 to 8.90 and T 9.05 to 9.41 (7H).

1. The Acid Catalyzed Rearrangement of 4-Methyl-ds-endo-6—
propyl-1,3,5,6-tetramethylbicyclo[3.1.0]hexenone (23).

 

 

One hundred mg of 22 was cooled to 4.0° and added to

5 ml of similarly cooled 98% sulfuric acid. The mixture was

51
shaken for 2.0 minutes, then quenched by pouring over
crushed ice. The mixture was diluted to 50 ml with water
and extracted with methylene chloride. The organic layer
was washed with water and dried over anhydrous magnesium
sulfate. Evaporation of the dried methylene chloride solu—
tion gave 80 mg of a pale yellow oil, which was examined
by nmr and vpc. The remainder of the material was lost due
to mechanical loss in workup.

The pale yellow oil consisted of two components in the
ratio of 8:1 by vpc analysis (10' x 1/4" DEGS, 160°, 70
ml/min of He). The major component was deuterated lz'and
the minor component was deuterated 22” by comparison of
their retention times with those of authentic unlabelled
samples. No 22 was detectable under these vpc conditions
(limit of detector is 0.001%).

The major component 22'was separated by preparative
vpc (same conditions as above). The minor component was
not isolated. The nmr spectrum of 22'consisted of a quartet
(J = 1.0 Hz) at T 8.09 (1.5H), singlets at T 8.42 (3H), 8.78
(1.5H), 8.88 (3H), and 8.94 (3H), and a set of two multi-

plets from T 8.72 to 8.88 and T 9.08 to 9.38 (7H).

J. The Synthesis of 4-Methyl-d3-exo-6-propyl-1,3,5,6-
tetramethylbicyclo[3.1.0]hexenone (22).

 

 

A solution of 260 mg of lg'in 15 ml of deuteromethanol
'was treated with 0.2 g of sodium metal under an atmosphere

of nitrogen. The solution was heated under reflux for one

52

2mm merHooV ssnuommn nsz .vH enemas

2&2 . o a

 

 

 

 

 

 

...W_ze_ eagle .fngJ,iee_inm
_ _t. fivmfffi i m.fi_ f
. . _. _~..n . .. .
fl. 331/. . \fixfifix
g
_ n
_ ..
. _ _

 

 

.«mmv meoemxmn_e.fi.nl
-oaomonnflmnnmsmnuouue.n.m.fi
IammoumlmloucmlmOIthumE

um mo Avaoov Eduuowmm H82 .mH ousmflm

 

. .
2&0— . o . o
.lezmgneenm_q4zfl;_asl_n

 

c

 

 

 

 

 

 

 

 

 

 

53
hour, and the deuteromethanol was removed under reduced
pressure. An additional 15 ml of deuteromethanolxyas added
and the solution was refluxed another hour. Most of the
deuteromethanol was removed under reduced pressure, and
the residue was dissolved in methylene chloride and washed
with cold (0°) water three times. The organic layer was
dried over anhydrous magnesium sulfate, and on evaporation
gave 250 mg (96%) of a very pale yellow oil, which con-
sisted of a single isomer in 95% purity by vpc analysis
(10' x 1/4" DEGS, 160°, 75 ml/min of He). The vpc analysis
detected none of QR (detection limit is 0.001%). ‘The nmr
spectrum of the oil showed greater than 95% of the label
on the methyl in the 4-position and consisted of methyl
singlets at T 8.45, 8.77, 8.89, and 9.13 and a set of two

multiplets from T 8.55-8.70 and r 8.95 to 9.25 (73).

K. The Acid Catalyzed Rearrangement of 4-Methyl-d3-exo—

 

6-pr0pyl-1,3,5,6—tetramethylbicyclo[3.1.0]hexenone.

 

One hundred mg of §l was cooled to 4.00 and added to
5 ml of similarly cooled 98% sulfuric acid. The mixture
was shaken for 2.0 minutes, then quenched by pouring over
crushed ice. The mixture was diluted to 50 ml with water
and extracted with methylene chloride. The organic layer
was washed with water and dried over anhydrous magnesium
sulfate. Evaporation of the dried methylene chloride solu-
tion gave 75 mg of a light yellow oil, which was examined

by nmr and vpc.

‘3'".

54

The light yellow oil consisted of two components in the
ratio of 9:1 by vpc analysis (10' x 1/4" DEGS, 160°, 75
ml/min of He). The major component was deuterated 18 and
the minor component was deuterated 22, by comparison of
their retention times with those 0f authentic unlabelled
samples. No §§ was detectable under these vpc conditions
(limit of detector 0.001%).
8 The major component §§ was purified by preparative vpc
(same conditions as above). The minor component was not
isolated. The nmr Spectrum of gg'consisted of a quartet
(J = 1.0 Hz) at T 8.15 (1.5H), Singlets at T 8.46 (3H),
8.79 (1.5H), 8.90 (3H), and 9.11 (3H), and a set of two

multiplets from T 8.55 to 8.70 and 1 8.95 to 9.25 (7H).

d.

55

.mm mo Avaoov Esnuommm HEZ .mfi mnsmfim

2&2 . a. , m
..:.:4+_.:],_:_1:44

c
s f

 

 

 

 

 

 

 

 

 

 

.Aflmv mcocmxmnao.fi.muoaomu
uflnsmsumemuumuso.m.m.fi
IHwmoumlmloxwlmvlamnuwfilv
mo AvHUUv Esuuoomm H82 .mH wuswfim

_2§3_ .. a o

4,_‘..fiqdlq1fi—~__fi4~___1

 

 

 

 

 

 

 

PART II

CONTROL OF THE STEREOCHEMISTRY OF 2,5-CYCLOHEXADIENONE

PHOTOISOMERIZATION BY STERIC FACTORS

56

INTRODUCTION

The photolysis of 2,5-cyclohexadienones g2 gives
bicyclo[3.1.0]hexenones £1,35 one type of photoproduct.
{the commonly accepted mechanism5 as shown below is the

excitation of g3 to give, in several steps, the dipolar ion

0 0-
¢%a¢
v
R1 R2 R1 R2

2‘1, 55

m

1,4 sigmatropic

 

 

rearrangement
O .-
R
.f @A 1
R2 1.... R2 J
21. 56

QE, which undergoes a 1,4-sigmatropic rearrangement to give

§Zx JDipolar ion §§ may only be a transient intermediate

57

58
along the potential energy curve in the rearrangement of ii
to QZ; The number of photoproducts (£1) which can be ob-'
tained from the photolysis of ég’depends on the substituents
on £2, For R1 # R2, photolysis of £2 should give statisti-
cally a 1:1 ratio of the two possible isomeric photoproducts
(one isomer will have R1 gggg to the cyclopentenone ring and
R2 239 and the other isomer will have reverse orientation).
It is also possible to generate four photoproducts if the
starting 2,5-cyclohexadienone is unsymmetrical as in §§.

(R1 # R2 and R3 # H), since the intermediate dipolar ion 23

can under go a 1,4-sigmatropic shift on either side of the

ring to give 92 and/or g}, Again on a statistical basis the
O

61

A~

59
the four photoproducts (there are two possible isomers of
both gg'and fig) could be formed in equal amounts, but steric
and/or electronic effects may alter the statistically ex-
pected ratios.
Many of the 2,5-cyclohexadienones that have been photo-
lyzed in which R1 # R2 and R3 # H contain fused rings as in

gzso and §§?1. Due to this ring fusion, only one of the

OAc

possible bicyclic[3.1.0]hexenones can be generated by photo-

lysis; for example, the photolysis of g2 only gives 82, An

 

60
examination of the possible photolysis intermediates demon-
strates the reason for this. As shown below, compound g2,
can only form dipolar ion g2 and not Qg’because the latter
'would generate a sterically strained six-membered ring trans-

fused to the bicyclic[3.1.0]system. Ion :52 can theoretically

64

62

.O@

6

m

undergo a 1,4-sigmatropic rearrangement to either side of
the :five-membered ring. Work in Part I of this thesis and
Other workerslfi‘19 have shown this rearrangement must pro-

ceed suprafacially with inversion. As a result, ion Q can

61

only rearrange to give 62 and not Ez'because the latter.
would contain a sterically strained'trangffused ring system.
Therefore, the ring fusion dictates that sterically only
one photoproduct is possible and no stereoselectivity in
the photochemistry of 2,5—cyclohexadienones can be detected.

On the other hand, there have been a few examples in
tjhe literature in which stereoselectivity in the photo-
cihemistry of 2,5-cyclohexadienones has been detected, but
nxo systematic study of the question has been made. Jeger
and Schaffner22 found that photolysis of 68 resulted in a
3:1 ratio of the two possible photoproducts gg'and 19'
Patel and Schuster23, who irradiated 11 at 25378, obtained
cmily bicyclic ketone 72; In contrast, Zimmerman24 and
errnewald obtained a 1.37:1.00 ratio of ii to 22 and a

1.()0:1.13 ratio of ZZ.t° 78 when 12 and 22, respectively,

Miller“:25 irradiated 22 and 82 to give

were photolyzed .

 

62

0 O C 13
hv
2537K a
¢H
CCl3
71 72 (100%)

 

252, '19. (58%) 12 (42%)

   

ZZ.(47%) Z§'(53¢)

63

 

 

2537e
12’ . u(56%) 28 (44%)
O
O hexane 9 ‘\/H
25372
H
\\
:22, Q}, (100%)
O\ o
h a,
he“; ; ‘fi. .V’\
‘\
82 83 (56%) (44%)

:32; r W

g2, (72%) 21, (28%)

64

a 1 .3:1 ratio iof 21 to 28 and 853 to 83‘, respectively; but
11325 found only 81’ when ”82' was'photolyzed. Curran and
Schuster26 found'a 2.5:1'ratio of 82 .to 81 when [83‘ was
photolyzed in methanol, but a ratio'of 0.79:1 was found in
cyclohexane. From these examples, it is observed that the
ratio of the two possible photoproducts is influenced by
the difference in size‘of the two substituents at C-4 and
the size of the substituents at C-3 and C-5 of the 2,5—
cyclohexadienone. In most cases, there is a preference for
the bulkier of the two groups at C-4 of the 2,5-cyclohexa-
dienone to take up an epic; position in the derived photo-
Product. This preference indicates that steric factors may
be important in determining the ratio of photoproducts from
the photolysis of 2,5—cyclohexadienones.

Since we had accidentally discovered, as mentioned in
Part I of this thesis, a high degree of stereoselectivity

in the photolysis of 2,11, we decided to make a systematic

O 0
R1 R3
R3
23 £2.

stUdy of steric effects in the photolysis of 2,5-cyclohexa-

dienones. In Part II of this thesis, the synthesis and

65

photolysis of 8,8 with different groups (R1 = H or Me, R2 _

H or Me, R3 = Pr or iso-Bu) is described.

RESULTS AND DISCUSSION

For this study a general procedure was established for
the synthesis and photolysis of the various 2,5-cyclohexa-
dienones. It involved C-alkylation of the sodium salt of

the appropriate phenol 82, rearrangement of the resulting

 

 

OH
O /
1. Na ¢
H/ CH3 \ R3
2.CH2=C-CH¢ *7?
I
R2 R1 R3 R2 R1

R3 = H or Me

X = Br or 1
£2. c 90

R1 ='H or Me
R9=HorMe

 

 

O O
BF3-Et20 TX 5% Rh/C
c1~1.,,c12 7 H2
or -
heat R3 R1 1 atm. R2 R1
R3 \\ Rs
2.1, 92
O
22f‘—Ex—-)> +
25372
MeOH ' R3
R2 R1
252, 94

66

67

2,4-cyclohexadienone 22 to the 2,5-cyclohexadienone 21,
selective reduction of the substituted allyl group to give
22” and photolysis of 2% to generate the photoproducts 2%
and 21.

This procedure, which is the same as that for the syn-
thesis of 22 in Part I of this thesis except for the acid:
catalyzed rearrangement step (22,—e 21), was chosen because

each step was known to proceed in high yield. Of the two

known methods for the rearrangement of 2,4-cyclohexadienones
to 2,5-cyclohexadienones, acid-catalyzed11 or therma127,
the acid-catalyzed rearrangement was selected because it
gives higher yields. A weak acid (boron trifluoride etherate)
was used because a stronger acid might cause cleavage of the
allyl group as a cation and regenerate the phenol. Photo-
lyses of the 2,5-cyclohexadienones were run using a 25378
lamp, because Kropp28 had obtained the highest yields of
bicyclo[3.1.0]hexenones under these conditions.

The first 2,5-cyclohexadienone selected for photolysis
was,compound gg, which had been previously photolyzed as

described in Part I. This was done in order to ascertain

68
that the stereoselectivity found in the photolysis of the

mixture of ggland gg'in Part I of the thesis was due entirely

23
to the photolysis of £2 and not effected in some unknown
manner by the presence of 32;

A. The Synthesis of 4-Propyl-2,3,4,5,6-pentamethy1-2,5-

cyclohexadienone (fig).

Compound g2 was synthesized by the rearrangement of fig

to £1, followed by selective reduction as shown below.

 

 

o / o o
BF3 'EtzO‘ 5% Rh/C ‘
CH2C12 7 Q I
H2
1 atm
\
22. 21. 29-

Cyclohexadienone 22, whose synthesis was described in

Part I, was dissolved in methylene chloride and treated with

69

boron trifluoride etherate overnight at room temperature.
Quenching over ice and work-up gave an 88% yield of 21,
which was identical in Spectral properties with previously
isolated material (Part I).

Reduction of 21 over 5% rhodium on charcoal at one
atmosphere gave a quantitative yield of 22, Compound 23
was identical in all ways with previously prepared 2%

(Part I).

B. Photolysis of 4-Propyl-2,3,4,5,6-pentamethyl-2,5-cyclo-

hexadienone.

 

A 5% solution of 22 in methanol was photolyzed at
25372 in a Rayonet reaCtor for three hours to give a 10:1
ratio of compounds 11 and 18 reSpectively, by vpc analysis.

The two photoproducts were identical in retention time and

o o o
hv
+
25372} ‘
MeOH
951 17 18

Spectral properties with authentic samples that had been pre-

viously prepared (Part I).

Since it was possible to repeat our earlier result with

the photolysis of 22” we decided to study the photolysis of

70
the trimethyl (compound 22, R1 = R2 - R3 = H) and tetra—
.methyl (compound 22” R1 5 Me; R2 = R5 = H) analogs of gg,
‘We anticipated a 1&1 ratio for the photoproducts of the tri-
:methyl analog and a ratio for the photoproducts of the tetra-
methyl analog between 1:1 and 10:1 by analogy with eXamples

in the Uterature (see Introduction).

C. The Synthesis of 4-Propyl-2,4,6-trimethyl-2,5—cyclo—

hexadienone (22).

As shown below, compound gfi was synthesized by alkyla—
tion of the sodium salt of mesitol 2g, rearrangement of the
resulting conjugated dienone 22 to the isomeric 2,5-cyclo-

hexadienone 97, and selective reduction.

 

OH
0 o
1.NaH/¢CH3 \ BF3-Et20
2.CH2=CH-CH;§r ---—1>
CH2C12
00

96 24 97

O
5% Rh/C
21?
H2
1 atm
22.

 

71

By a modification of the procedure of Curtinaigg_§l.,
6-allyl-2,4,6-trimethyl-2,4—cyclohexadienone gglwas synthe—
sized in a 70% yield by the reaction of the sodium salt of
2g with allyl bromide. Compound 2'11, had a uv spectrum
identical with that described in the literature. In meth-
anol, it displayed a maximum at 318 nm (e = 3800). The
other spectral properties were consistent with the proposed

structure. ,The ir spectrum of 22 showed bands at 1640 and

 

1650 cm—1. The nmr spectrum (see Table II) was assigned on
29

  

the basis of analogy with the spectra of g§?9, gg' , and.22,
0 .1 0 - 4.91—
8.188 8 8888 85 8'25br(-°’)09.00s 5.58m

' \‘8.93s ,-‘

‘\.

  
  

3.40m 4.22m 7-45d

8.103

8'183 8.00br(q

8.185 8. 8d
98 99 20

m rw MI

Compound gg'was dissolved in methylene chloride and
treated with boron trifluoride etherate at 0° for 2.0
minutes. Quenching over ice gave a quantitative yield of
21, Longer reaction times gave increasing yields of gg.

Compound gz'had a uv spectrum in methanol with a'maxi-
mum at 245 nm.(é = 8700); lit.3° xgzgn a 245 nm (e = 14,000).

1

The ir spectrum of 21'showed bands at 1630 and 1665 cm- I

which were consistent with the proposed structure13. The

72

 

 

 

Table II. Nmr spectraa
Compound Chemical Shift (J)b Assignments
8.92 s) C-6 methyl
.8.21 d,1.3) C-2 methyl
C-4 methyl

101

i

 

8.12 d,1.3;
7.86 d,4.0
7.76 d,4.0)
4.40a5.42(m)
4.20-4°33(m)
3.40-3.54 m

8.85 s;
8.16 S
7.70 d,5.5)
4.86-5.33(m)

3.52(s)

8.33-9.27(m)
8.83 s;
8.17 s

3.59(s)

9.05-9.33Em)
8.73-8.78 m
8.85 s)
8.70%5
8.33 d,1.0)

8.08-8.23 m;
2.95-3.15 m

8.92-9.25 m)
8.50-8.63(m
9.08§s;
8.73 s
8.35 d,1.0)

8.12-8.23 m;
2.98-3.15 m

Methylene Protons

Eggfcyclic vinyl
protons

C-5 hydrogen

C-3 hydrogen

C-4 Methyl

C-2 and C-6 methyls

Methylene protons

gggrcyclic vinyl
protons

C-3 and C-5
hydrogens

C-4 propylC
C-4 methyl
C-2 and C-6
methyls
C-3 and C-5
hydrogens

C-6 propyl

C-6 methyl
C-1 methyl
C-3 methyl
C-5 hydrogen
C-4 hydrogen

C-6 propyld

C-6 methyl
C-1 methyl
C43 methyl
C-5 hydrogen
C-4 hydrogen

Continued

73

Table II. Continued

 

aAll spectra are in CC14.

bShifts are reported in T values and J's in Hertz. Multi-

plicity of peaks is shown in brackets: s, singlet; d,
doublet; m, multiplet.

cThe C-4 methyl and C-4 propyl groups have overlapping
absorptions. As a result, only the total area of ten
protons for both groups can be measured.

dThe methyl and propyl groups at C-6 have absorptions over-
lapping one another. Therefore, only the total area of
ten protons for both groups can be measured.

74
nmr spectral assignments (see Table II) follow from similar-

ities to the spectra of 25? and 22}"

3’30d 8.77s 0
3.93d

  
  

3.605

8.173
8 ’83s } 8 .45—9 .1010

“Zi. 12,

Reduction of gz'over 5% rhodium on charcoal at one
atmosphere gave a quantitative yield of 25, Compound 25'
was identified by its Spectroscopic properties and by the
fact that analysis showed that only one double bond had
been reduced. Compounds gé'and gz'had similar spectral
properties, which showed that neither of the cyclic double
bonds had been reduced. The uv spectrum of gé'in methanol
had a maximum at 244 nm (e = 8400). Its ir spectrum showed
bands at l635 and 1660 cm-1. The nmr.spectrum (see Table II)

'was assigned on the basis of the similarities in spectra of

22.. 22,. and 22,-

D. The Photolysis of 4-Propy1-2,4,6-trimethyl-2,5—cyclo—

hexadienone (25).

A 5% solution of gg'in methanol was photolyzed at 25378
in a Rayonet reactor for 1.5 hours until Vpc analysis showed

that gg'had been essentially consumed and the two

75

photoproducts, assigned structures lgg’and 12;, had been
formed in the ratio of 1:1.27. It was difficult to deter-
mine when 35 had been completely consumed because £0}; and 25’
had the same retention time on all vpc columns tried. As"
a result, the photolysis was continued until the area of

the one peak remained constant on further photolysis. On
account of this difficulty, the accuracy of the ratio 1:1.27
is questionable; and the only conclusion about the ratio,
which can be made with certainty, is that the ratio of the

two photoproducts is approximately 1:1.

0 O O
hv ‘. + ..
25373 a
MeOH

22

100 101

Rt ="9.0 min 7.0 min 9.0'min

The two photoproducts were separated by preparative vpc
and were identified as lgg'and lgl’by their spectral proper-
ties. .Analysis showed that lgp'ahd lglvwere isomeric with
g5, The uv spectra of lgg’and 121 were typical of bicyclo-
[3.1.0]hexenones. The uv spectrum of 122,in methanol had
maxima at 325 sh (e = 750), 275 (e = 2000), and 235 nm (e =

4800) while the uv spectrum of 101 in methanol had maxima

76
at 320 sh (c = 350), 276 (e = 2300), and 236 nm (e = 4700).
The ir spectra of lgg'and.121’had bands at 1640 and 1685 cm-1,
consistent with the'proposed structures.

Compounds 120 and 121 were distinguished by their nmr
spectra. The photoproduct with a retention time of 7.0
minutes was assigned structure 122 because it had its C-6
methyl at lower field (T 8.85) than the C-6 methyl (T 9.08)
of the other photoproduct 121, The multiplets for the C-6
propyl of the two isomers showed the same effect. The
multiplets of the C-6 propyl occurred at higher field for
lQQ’than for 121, A similar distinction had been found in
the nmr spectra of 12 and l8 in Part I of this thesis. The

reason for the difference is the same: the v-bond of the

cyclopentenone ring shields the endo groups. The other
nmr assignments of 100 and 101 (see Table II) were based

on spectral similarities to 81/ 10228, and 123?8.

‘P _

77

8.563

     

81 102

Since the predictions about the ratio of the photo-
products of gé’were correct, attention was next turned to
photolysis of the tetramethyl analog 104. Another inter-

esting problem was that there are four possible bicyclic

 

104

compounds 122, 126, 121, and lgg’which can possibly be ob—
tained from the photolysis of 122; In lgg'and 126, the C-3
methyl group of lgg'resides on the cyclopropyl ring of the
photoproducts whereas in lgz’and lggithe C-3 methyl group
resides on the double bond of the cyclopentenone ring.

Product prediction is uncertain, because there is precedent

in the literature for the formation of products of the type

78

107 and 108 only, or 105 and 106 only, or all four, with
107 and 108 predominating. Kropp found10 that the photolysis

   

105 106 10 108

of gfi'gave only photoproduct lgg'with the B-substituted

double bond. On the other hand, Miller found the photo-
product with the less substituted double bond §l.in the photo-
lysis of 82” Yet Schaffner and Jeger found that lgg’and 112’
photolyze in bdth possible manners to give the photoproducts
with the less substituted double bond and the more substi-
tuted double bond. The photoproduct with the more substi-
tuted double bond predominated. They also found that the

photolysis of 115 generated the photoproduct 116 with the

 

2,952, (100%)

 

 

79

 

80

more substituted double bond. They postulated that both
inductive and steric effects were at work in these examples.
The photolysis of lggxmight aid in determining what struc-
tural effects of the starting 2,5-cyclohexadienone dictate
whether steric or inductive or both are operative during

the photochemical rearrangement.

E. The Synthesis of Isodurenol (117).

 

Isodurenol 111 was needed for the synthesis of 122;
Though the compound is well-known, there is no really good
synthetic method for it in the literature. By use of the
chloromethylation procedure of Mathai and Sethma31, iso—
durenol was synthesized in two steps from meta-cresol as

shown:

OH OH OH

CIC 2 cnzcl
(CH20)3 LiAlH4

dioxané f§§"'>
HCl

CHZCl
118 117

Paraformaldehyde was dissolved in dioxane and saturated
'with hydrogen chloride. Subsequently, meta-cresol and hydro-
gen chloride were added to give a 72% yield of 11§,(mp 100-
1020; lit.32 104—106). The nmr spectrum (CC14) of 118 showed

the expected two-proton singlets at T 5.47, 5.43, and 5.28

81
due to the benzyl protons, one—proton singlets at T 2.85
and 5.20 due to the aromatic and phenolic hydrogens, re-
spectively, and a three-proton singlet at T 7.60 due to the
aromatic methyl group.

Compound 118 was reduced with lithium aluminum hydride
in tetrahydrofuran. Work up gave a yellow oil, which was
steam distilled to give a 90% yield of isodurenol (mp 75-
770; lit.33 79-800). The nmr spectrum (CC14) of llZ’Showed
a nine—proton singlet at 1 7.96 and a three-proton singlet
at T 7.93 due to the aromatic methyl groups and one—proton
singlets at T 3.52 and 5.68 due to the aromatic and phenolic

hydrogens, respectively.

F. Synthesis of 4-Propyl-2,3,4,6-tetramethyl-2,5-cyclohexa-

 

dienone 104.

 

As shown below, compound 104 was synthesized by a
method similar to that used for 95” The sodium salt of 117
was alkylated with allyl bromide to give a 62% yield of 119

and 120.

OH

1 .NaH/0CHi \
2 .CH2 =CHCTIIZB

 

H
H
q
H
H
(O
H
N
O

i
E
E

82

 

o O
‘\
BF3 'Etzo \
5 Rh
119 + 120 CH2C127 2 /§
0° <3
H2
1 atm
\
121 12g,

Structures 112 and 122'were assigned on the basis of
spectral data and analysis. Elemental analysis of the
mixture was in agreement with the proposed structures. The
uv spectrum of the mixture had a maximum at 325 nm (e =
3350), which is expected for 2,4-cyclohexadienones. The ir
spectrum showed bands at 1635 and 1650 cm-1, which are in-
dicative of a cyclohexadienone13. The nmr spectrum (CCl4)
of the mixture was consistent with a 1:1 ratio of 112’and
122; the multiplet from T 3.39 to 3.51, which was assigned
to the C-3 proton of 112, was equivalent in area to the
multiplet from T 4.12 to 4.28, which was assigned to the C-5
proton of 122, These assignments are based on a comparison
with the spectra of 22’and 22, Due to the similarities in
structure of 112'and 122, a complete assignment of all the
peaks in the nmr spectrum could not be made (see the experi-

mental section for the complete spectrum).

83

O 0

8.185 8.883 N
\ 8‘185 \8.93s
3.40m
4.22m
e/JgiIBS 8.00br(q
98
' AN 8.08d 99

w

No attempt was made to separate the mixture. It was
rearranged directly to compound 121, The mixture was dis-
solved in methylene chloride and treated with boron tri-
fluoride etherate at 00 for 7.0 minutes. Quenching with
ice gave a quantitative yield of 121, Longer reactibn times
regenerated increasing amounts of 111,

Compound 121,was identified by the fact that it was
isomeric with the starting material and by its spectral
properties. The uv spectrum of 121sin methanol displayed a
maximum at 244 nm (e = 8500), which is indicative of a 2,5-
cyclohexadienone. Its ir Spectrum Showed bands at 1620 and
1660 cm-1. The nmr spectrum (see Table III) was also con—
sistent with the proposed structure.

Reduction of 121’over 5% rhodium on charcoal at one
atmosphere gave a quantitative yield of 122, Compound 122’
was also identified by the fact that analysis showed that
only One double bond had been reduced and by its spectro-
scopic properties. Compounds 12g’and 121’had similar spec-
tral properties, which showed that neither of the cyclic

double bonds had been reduced. The uv spectrum of 104 in

84

Table III. Nmr Spectraa

 

 

 

Compound Chemical Shift (J)b Assignments
‘ 8.80(s) C-4 methyl
8.09-8.18(m) C-2,C-3, and C-6
methyls

7.52-7.74(m)
4.65-5.33(m)

 

3.45—3.55(m)
0 8.89—9.28(m)
8.86(s)
6 2 8.16br(s)
5 3.58(q,1.0)
104

8.80-8.91(m)
9.03-9.39(m)
8.87 (S)

8.78(s)

8.45(Q.0.7)
8.20-8.33(m)
8.06(Q.0.7)

 

8.92-9.28(m) '
8.53-8.68(m)
9.15(s)
8.77(s)
8.43(Q.0.7)
8.23-8.38(m)
8.11(Q.0.7)

 

Methylene protons

Eggfcyclic vinyl
protons

C-5 proton

C-4 propyl

C-4 methyl

C-2,C-3, and C-6
methyls

C-5 proton

C-6 propylC

6 methyl
1 methyl
-3 methyl
5 proton
4 methyl

propylc

C 6

C 6 methyl
C 1 methyl
C-3 methyl
C 5 proton
C 4 methyl

 

Continued

85

Table III. Continued

aAll spectra are in CCl4.

bShifts are reported in T values and J's in Hertz. Multi-

plicity of peaks is shown in brackets: br(s), broad sing-
let; s, Singlet; q, quartet; m, multiplet.

CThe C-6 methyl and propyl groups have overlapping absorp-
tions. Therefore, only the total area of ten protons for
both groups can be measured.

86
methanol had a maximum at 245 nm (e = 8400). Its ir spec-
trum Showed bands at 1630 and 1665 cm-1. The nmr spectrum
(see Table III) was assigned on the basis of the similari—

ties in the spectra of 104 and 121.

G. The Photolysis of 4-Propyl-2,3,4,6-tetramethyl—2,5-

cyclohexadienone (104).

 

A 5% solution of 122 in methanol was photolyzed at
25378 in a Rayonet reactor for 2.5 hours until VpC analysis
showed that 122 had been essentially consumed. The two
photoproducts, assigned structures 121 and 122, were formed

in the ratio of 2.12:1.

0 0
0
hv l”/ +
25378 9 \ ‘
MeOH
104 107 192%

The photoproducts were separated by preparative vpc
and identified as 12Z’and 122'by their elemental analysis,
which showed that they were isomers of 122, and by their
spectral properties. The uv spectrum of 121 in methanol
showed maxima at 320 Sh (e = 650), 278 (e = 2360), and 232
:nm (e - 4440) whereas the uv Spectrum.of 122’in methanol

'had maxima at 320 Sh (e = 455), 278 (e = 2360), and 232 nm

87
(e = 4740), similar to the other bicyclo[3.1.0]hexenones
synthesized in this study. The ir spectra of 12Z'and 122’
had bands at 1640 and 1685 cm-1, which are also indicative
of bicyclo[3.1.0]hexenones.

The two photoproducts, 12Z’and 122, were distinguished
by the difference in the position of the C—6 methyl peak in
their nmr Spectra. The major photoproduct was assigned
structure 12Z’because it had its C-6 methyl at lower field
(T 8.87) than the C—6 methyl (1 9.15) of the other photo-
product, assigned structure 122, The other nmr assignments
of 121 and 122 (see Table III) were based on Similarities in
the Spectra of 121 and 122 to the Spectra of 122, 121, and

103.

      

7.94d 8.105

3.15m 8.23m 3.15m 8.23m

There was no evidence for the presence of the other
possible photoproducts, 122 and 122, in the photolysis mix-
ture, for the nmr spectrum of the crude mixture diSplayed
no vinyl protons aroung T 3.00. It is probable that less

than 0.5% of 122 and 106 was formed, for, when the photolysis

88

was run to 50% completion, the vpc analysis of the mixture
showed the presence of only 121, 122, and starting material.
It was only at 90% completion of the photolysis that other
compounds, which were all less than 1% of the total mixture
were detected by vpc analysis. The formation of only 121’
and 122'in the photolysis of 12g’could reflect the greater
stability of the ionic intermediates for the formation of

107 and 108 relative to those for the formation of 105 and

106.

A definite steric effect had been observed in the
photolysis of the three propyl 2,5-cyclohexadienones 22,

22, and 104. It was anticipated that a group bulkier than

 

o O O
22 104 22,

endo-Pr

——————-: 10.0 2.1 1.0

exo-Pr

ratio in photoproducts

propyl at C~4 (for instance, isobutyl) would cause the endo

to exo ratios of the corresponding photoproducts to increase.

89

The isobutyl group was selected because its precursor is
the methallyl group, and C-alkylation of phenols with meth-
allyl chloride was expected to proceed in high yields.
Furthermore, allylic rearrangement during the BF3-catalyzed
isomerization of the 2,4- to 2,5-cyclohexadienone would not
alter the structure of the product.

Compounds 122’and 122’were synthesized and photolyzed
to test the generality of our predictions about steric

effects in the photochemistry of 2,5-cyclohexadienones.

O O
122 123

Also, photolysis of 123 would provide another test of the
observation that rearrangement occurs in the direction

which produces a B-alkylated enone as the photoproduct.

H. Synthesis of 4-Isobutyl-2,4,6-trimethyl—2,5-cyclohexa-
dienone (122).

 

 

As shown below, compound 122 was synthesized by al-
'kylation of the sodium salt of mesitol 22, followed by
selective reduction. The previous alkylation procedure,

‘which worked well with allyl bromide, was unsuccessful with

90
methallyl chloride. Even after five days at room tempera—
ture only a small amount of 2,4-cyclohexadienone was ob-
tained, in addition to considerable unreacted mesitol.
However, if the alkylation mixture was refluxed overnight,
the desired 2,5-cyclohexadienone was obtained directly in
excellent yield. It is possible that the first cyclohexa-
dienone formed was the 2,4-Cyclohexadienone, which then

rearranged thermally to the 2,5-cyclohexadienone.

 

OH 0 o
1.NaH/¢CH,\ 5 Rh c
2.CH2=ccfi2c1 )

6)
H2
1 atm
92 2.2... 3,2},

The sodium salt of gg'was refluxed overnight with
methallyl chloride to give an 84% yield of 4-methallyl-
2,4,6-trimethyl-2,5-cyclohexadienone 122; Structure 124’
was assigned on the basis of spectral properties and analysis.
Elemental analysis was in agreement with the proposed struc-
ture. The ir and uv spectra were similar to the spectra of
other 2,5-cyclohexadienones. The uv spectrum of 124,in
methanol displayed a maximum at 247 nm (e = 14,200). The ir
Spectrum showed bands at 1640 and 1665 cm-1. The nmr spec-

trum (see Table IV) was assigned on the basis of the simi-

larities in the spectra of 124 and 21,

 

91

Table IV. Nmr Spectraa

 

 

 

Compound Chemical Shift (J)b Assignment
8.85(s) C-4 methyl
8.42(d,0.9) exo—cyclic methyl
8.14(s) C-2 and C-6 methyls
7.72(s) methylene protons

 

5.24-5.44(m)

3.47 (s)

9.22(d,6.5)C
8.88-9.35(m)
8.37-8.53(m)
8.85(s)
8.13(s)
3.46(s)

C

9.27(d,6.5)
9.24(d,6.5) d
8.78-9.08(m)

8.89(s)d
8.72(s)
8.32(d,1.0)
8.13-8.25(m)
3.13-3.21(m)

9.08(s)e e
9.05(d,7.0) e
8.95-9.20(m)
8.42-8.83(m)
8.74(s)
8.32(d,1.0)
8.10-8.18(m)
3.00-3.15(m)

exo-cyclic vinyl
protons
C-3 and C-5 protons

gem dimethyls

other isobutyl
protons

C-4 methyl

C-2 and C-6 methyls

C-3 and C-5 protons

one gem dimethyl

other gem dimethyl

other isobutyl
protons

C-6 methyl

C—1 methyl

C—3 methyl

C-5 proton

C-4 proton

C-6 methyl

gem dimethyls
other isobutyl

protons

C-1 methyl

C-3 methyl

C-5 proton

C-4 proton

 

Continued

 

92

Table IV. Continued

 

aAll spectra are in CCl4.

bShifts are reported in T values and J's in Hertz. Multi-

plicity of peaks is shown in brackets: s, singlet; d,
doublet; m, multiplet.

CThe total area of all peaks labelled by this letter repre-
sents nine protons.

dThe total area of all peaks labelled by this letter repre-
sents six protons.

eThe total area of all peaks labelled by this letter repre-
sent twelve protons.

93

8.168

3.523

8.85s

Reduction of 122 over 5% rhodium on charcoal in cyclo-
hexane and triethylamine at one atmosphere gave a mixture
of 122,3nd 96” The mixture was column chromatographed and
elution with benzene gave a 70% yield of 122, If the tri-
ethylamine was not present, the reduction gave only 96 in '
a quantitative yield.

Compound 122 was identified by its spectroscopic
properties and analysis. Elemental analysis showed that
only one double bond had been reduced. Compounds 122 and
lgg'had similar spectral properties, which again demon-
strated that neither of the cyclic double bonds had been
reduced. The uv spectrum of lgg’in methanol showed a maxi-
mum at 244 nm (6 = 14,450). Its ir spectrum showed bands 8
at 1635 and 1662 cm-1. The nmr spectrum (see Table IV) was
assigned on the basis of similarities in the spectra of

122 and 124.

94

I. The Photolysis of 4—Isobutyl-2,4,6-trimethyl-2,5-cyclo-

hexadienone (122).

 

 

A 5% solution of lgg’in methanol was photolyzed at
25373 in a Rayonet reactor for 3.0 hours until vpc analysis
showed that essentially all the starting material lggihad
been consumed. The two photoproducts, assigned structures
122 and 126” were formed in the ratio of 1.13:1.00. It was
again difficult to determine when the starting material,
122” had been completely consumed because one of the photo—
products (122) had the same retention time as 122,0“ all
vpc columns tried. As with the photolysis of 95, the pho-
tolysis was continued until the area of the one peak re-
mained constant on further photolysis. Again the accuracy
of the ratio is questionable; and the only thing which can
be stated with certainty is that the ratio of the two photo-
products is approximately 1:1.

The two photoproducts were separated by preparative

vpc and identified by their Spectral prOperties. The uv

spectrum of 125 in methanol showed maxima at 324 sh (e = 240),

274 (e = 1360), and 240 (e = 4250) and the uv spectrum of

\g/ 324:?ng ‘/ + Of

122 125 126

’M W

1“

95
lfifiIhad maxima at 320 sh (e - 150), 274 (e = 2400), and
238 (e = 5800), which are similar to the spectra of other
bicyclo[3.1.0]hexenones synthesized in this study. The ir

1 and the ir

spectrum of 12§,had bands at 1640 and 1690 cm-
spectrum of lgg’had bands at 1630 and 1690 cm-1, consistent
with the bicyclo[3.1.0]hexenone structure.

Compounds 125 and 126 were distinguished as were the
other photoproducts in this study, by their nmr spectra.
The photoproduct with a retention time of 8.8 minutes was
assigned structure 125 because it had its C-6 methyl at
lower field (T 8.89) than the C-6 methyl (T 9.08) of the
other photoproduct, assigned structure 126, The other nmr
assignments of lgg’and 122 (see Table IV) were based on
similarities betWeen the spectra of lgéiand lgg'and those
of lQQ'and 121, The methyls of the isobutyl group of 122'

appear as tWo doublets because they are non-equivalent.

Silverstein34 and his workers observed a similar result in

9.083

   

'the nmr spectrum of 127 where the two methyl groups appeared

as two doublets at T 9.10 and 9.08 (J = 7 Hz). They postulated

96

that this was due to the single asymmetric center at the

 

hydroxyl-bearing carbon, which resulted in nonequivalence of
the protons of the adjacent methylene groups and also the
methyl groups. This explanation does not seem plausible

for compound 125” because it would predict two doublets for
the isobutyl methyls in lgfi'as well, an unobserved result.

A more probable explanation, which was supported by a study
of molecular models, is that rotation of the isobutyl group
is restricted in 125, so that the methyls are on the average
in different environments and nonequivalent. When the iso-
butyl group is 252, as in 126, there is no such restricted
rotation and only one methyl doublet was observed.

It had been predicted that the ratio of lgé’and‘ggg
from the photolysis of lzg’would be slightly larger than the
ratio of lgg’to lgg'from'the photolysis of 95, To the extent
that the measured ratios were accurate, the prediction was
verified.

Compound 12§,was next synthesized and photolyzed to
determine which‘of the four possible photoproducts, 1287131,

WV

would be produced, and in what ratio.

O

123
O O o O
128 129 130 131

Synthesis of 4-Isobutyl-2,3,4,6-tetramethyl-2,5-cyclo—

 

hexadienone (123).

 

As shown below, compound 123 was synthesized by the

same method used to prepare 122. The sodium salt of 117

was refluxed overnight with methallyl chloride to give a

71.5% yield of 4-methallyl-2,3,4,6-tetramethyl-2,5—cyclo-

hexadienone.132.

98

 

 

OH O
O
1.NaH/¢>CH3 5 \
I
CH3
heat H2
1 atm

117 132 123

Structure 132 was assigned on the basis of Spectral
properties and analysis. The elemental analysis was in
agreement with the proposed structure. The uv and ir spectra
were similar to those of other 2,5—cyclohexadienones. The
uv spectrum in methanol displayed a maximumaflz248nm(e = 8200),
and the ir spectrum showed bands at 1635 and 1665 cm_1.

Peaks in the nmr Spectrum (see Table V) were assigned on the

 

1 1

W

basis of the similarities between the spectra of 132 and 121.
Reduction of 132 over 5% rhodium on charcoal at one
atmosphere gave a 94% yield of 123. The elemental analysis

showed that only one double bond had been reduced. Compounds

99

 

 

 

Table V. Nmr Spectraa
Compound Chemical Shift (J)b Assignment
’0‘ y:§5 s) C-4 methyl
.53 d,0.9) exo-cyclic methyl
3 8.18 q,1.5 C-2 and C-3 methyls
’ 8.14 d,0.9 C-6 methyl,

5 7'67 S) meth lene protons
7.55 s) Y
5.38-5.62(m) vinyl protons
3.59(q,1.2) C-5 proton

1.32.

O c .
9.25(d,6.5) c gem dimethyls
8.25-9.25(m) other isobutyl

6 protons
8.87(S) C-4 methyl

5 8.17br(s) C-2,C—3, and C-6

methyls
3.46(q,1.1) C-5 proton
123

 

 

9.33 d,6.5)C
9.22 d,6.5)cc
8.95-9.38(m)

8.92ésg
8.75 s
8.46br
8.38br
8.08br

S
S
S

 

 

9.17 s)d
9.08 d,7.0
8.50-9.20(m

d
))d

8.77(s)
8.47br
8.32br
8.07br

S
S
S

 

 

one gem-dimethyl

other gem-dimethyl

other isobutyl
protons

C-6 methyl

C-1 methyl

C-3‘proton

C-5 proton

C-4 methyl

C-6 methyl
gem-dimethyls
other isobutyl
protons

C-1 methyl

C-3 methyl

C-5 proton

C-4 methyl

 

Continued

100

Table V. Continued.

 

aAll spectra are in CC14.

b

C

d

Shifts are reported in T values and J's in Hertz. Multipli-
city of peaks is shown in brackets: br(s), broad singlet;

3,.

The
for

The
for

singlet; d, doublet; q, quartet; m, multiplet.

total area of all peaks labelled by this letter accounts
nine protons.

total area of all peaks labelled by this letter accounts
twelve protons.

101
123 and 132 had similar spectral properties, which demon-
strated that neither of the cyclic double bonds had been
reduced. The uv spectrum of 123 in methanol showed a maxi—
mum at 248 nm (e = 8150). Its ir spectrum showed bands at
1635 and 1660 cm-1. The nmr spectrum (see Table V) was
assigned on the basis of the similarities between the

spectra of 123 and 132.

K. The Photochemical Isomerization of 4-Isobutyl—2,3,4,6—

 

tetramethyl-2,5-cyclohexadienone (123).

 

A 5% solution of 123 in methanol was photolyzed at
25373 in a Rayonet reactor until vpc analysis showed that
95% of 123 had been consumed. The two photoproducts, as-
signed structures 128 and 129,were formed in the ratio of
3.35:1.00.

The photoproducts were separated by preparative vpc

and identified as 128 and 129 by their spectral properties.

 

o o o
13)) \ / + /
\
25372 7
MeOH
12.3, 1.2.9, 1.2.2,

The uv Spectrum of 128 had maxima at 325 sh (e = 850), 274

(e = 2510) and 235 nm (e - 4250) and the uv spectrum of 129

102

showed maxima at 322 sh (e = 790), 273 (e = 2390), and 235 nm
(e - 3400), both indicative of the bicyclo[3.1.0]hexenone
structure. The ir spectra of 128 and 129 had bands at 1640
and 1685 cm_1, similar to other bicyclo[3.1.0]hexenones.

Photoproducts assigned structures 128 and 129 were
again distinguished by the differences in the nmr spectra
for the absorptions of the C-6 methyl groups. The major
photoproduct, assigned structure 128” had its C—6 methyl at
lower field (1 8.92) than the C-6 methyl (1 9.17) of the
other photOproduct (122). The other nmr assignments of 128’
and 129 (see Table V) are based on a comparison with the"’
nmr spectra of lQZIand 128” The methyls of the isobutyl
group in 128 appear as two doublets because they are non-
equivalent, the same effect which was observed in the nmr

spectrum of 125 (see Section I).

9.155

8 .45q

  

8.20—
8.06q 8.33m

 

107 108

I'M rwv

There was no evidence for the presence of 130 or 131

 

in the photolysis mixture, for vpc analysis of the reaction

Inixture showed only the two photoproducts (128 and 129) and

103
starting material. The nmr spectrum of the crude mixture
displayed no vinyl protons around T 3.00, which would be

expected for 130 and 131.

L. Mechanistic Considerations in the Photolysis of 2,5—

Cyclohexadienones.

 

The presently accepted mechanismaffor the photoisomeri—
zation of 2,5-cyclohexadienones to bicyclo[3.1.0]hexenones
is shown below. The starting material is excited to the
diradical 133” which bonds across C—3 and C-5 and electron

demotes to give 55. Patel and Schuster23 have demonstrated

0 o’ o

6 2 hv 1.c-3 and c-5
-——————) ' \
5 3 bonding 7

2 .e demotion V

 

 

R1 R2 R1 '2
R1 R2
ii 1,952, :32

1,4 Shift
\V
0

R1
R2

57

104
the reality of intermediate Qé’by trapping it with methanol,

in the photolysis of Z};

O O
OCH3
hv
Er)
MeOH v
CC13 CCl3

71 134

A 1,4-sigmatropic rearrangement of 55 gives 21, the
isolated photoproduct. The Woodward-Hoffmann rules predict
that the 1,4—sigmatr0pic rearrangement proceeds suprafacially
with inversion36 at C-6 in this system. These predictions
have been verified; for Jeger and Schaffner22 have found
that inversion occured at C-10 in the photolysis of 22/ and

several group15-19v36 have discovered that 1,4-sigmatropic

OAc

 

rearrangements occur in a suprafacial manner. This means

that the stereochemistry (that is, the exo or endo position

105
of R1 or R2) of 5Z,must be the same as that of 55. As a
result the stereochemistry of the photoproducts is definitely
determined at the time that intermediate 55Iis formed.
Space filling molecular models Show that the 2,5-
cyclohexadienone probably exists as an equilibrium mixture
of two rapidly interconverting rather flat boat conforma—

tions (135). Miller37v38 pointed this out from a study of

 

 

R
o 2
0 fl L
R2 R1
135

mm

molecular models of 55’and 136. In this conformation, the
2,5-cyclohexadienone is already in the proper orientation
for formation of a C3-C5 bond leading to the bicyclo[3.1.0]-

system 55, As a result, the ratio of the two photoproducts

80 136

106
is most probably determined by the relative amounts of each
conformer of 155, This ratio can be affected by the size
of substituents at C—3, C-4, and C-5.

Steric factors control the population of each conformer
of 155’by affecting the rates of ring flipping from one con-'
former to the other. In flipping, the 2355 group at C-4
goes through intermediate stage 15Z'in which it passes by

and is in close proximity to the C-3 and C-5 substituents.

0 R1
\ R2
W .

 

 

Large groups at C—3 and C-5 or a large substituent at C-4
increase the barrier to flipping in one of the two direc-
tions. Steric repulsion in moving a bulky group at C-4 from
the §g§2_position past the groups at C-3 and C-5 to the 352.
position is large; the rate of ring flipping in this direc-
tion is therefore retarded. Also if the bulkier group at
C-4 were in the §x2_position, it would experience a large
Steric repulsion between itself and the C-3 and C-5 sub-
Stituents so that this conformer is not favored and the ring
flips to the preferred conformer with the bulkier group in
the 2352 position. As a result, the conformer with the
'bulkier group §g§2_is densely populated and the other con-

former is sparsely populated. On the other hand, if the

107
difference in Size between the two groups at C-4 is small,
or the substituents at C-3 and C-5 are small, each conformer
of 555’will be equally populated. That is, steric repulsion
between the g§9_C-4 group and the C-3 and C-5 substituents
for each conformer will be nearly equal, the steric retarda-

tion of ring flipping from one conformer to another will be

’1

small, and the rates in either direction will be nearly'
equal.

From these considerations, stereoselectivity in the
photolysis of 2,5-cyclohexadienones becomes quite understand—
able. The difference in bulk between the methyl and propyl
groups is not great, as indicated by the Similar AG'S for
the equilibrium shown below when R = methyl and when R =
propyl (AGMe - —l.7 XE-AGPr = —2.1 kcal/mole).39 In the

photolysis of 55, which contains small groups at C-3 and

R

C-5 (hydrogen), one would predict that the approximately
equal amounts of each conformer of 55’would be mirrored in
a 1:1 ratio of the two photoproducts, the observed result.
In the photolysis of 122, however, with a methyl group at
C-3, one would predict a predominance of the conformer with

the propyl group in the endo position, because the C-3

108

O O 0
25378
MGOH \ \
h v
25 100 (45%) 101 (55%)
h v I
\ + \
104 107 (68%) 108 (32%)
O O O

23 i2, (91%) is. (9%)

 

109

O O O
hv K ;\ W
fl +
122. 2.22 (56%) 12,9 (44%)
O O
hV a ‘ + \
1.2-2 12.2.). (77%) .129 «23%)

methyl group should experience less steric repulsion with
the 3x2 C-4 methyl group than with the gxg_C-4 propyl group.
As a result, less ring flipping toward and sparser popula-
tion of the exo-propyl conformer of 122'should result. This
Should be mirrored in a predominance of the Egggspropyl
photoproduct, which is the observed result.

By this same reasoning, the gxgrpropyl conformer of 25’

should experience even more steric repulsion and be even

110
less populated than that of 104. Indeed, there is evidence

in the nmr spectrum that compound 55 exists almost entirely
in the Eggg_propyl conformation. The high field absorption
for the propyl hydrogens of 55 (T 9.39—9.59) can only be
explained if one assumes that the propyl hydrogens are 3559
to and shielded by the w electrons of the cyclohexadienone.
From these considerations one would predict that the EEQQ‘
propyl product Should be highly favored in the photolysis
of 55, which is the observed result.

One would also expect that the larger the difference
in the Size of groups at C-4, the greater these effects
should be. This was found when the gfpropyl group was re-
placed by the isobutyl group. Compare the ratio of photo-
products from 152 and 155 with those from 95 and 153
respectively.

From this study it seems clear that the determining
factor in the ratio of photoproducts is the amount of each
conformer of the 2,5-cyclohexadienone present in the ground
state. It is not unusual that the structure of the starting
material Should determine the ratio of photoproducts obtained,
for Dauben4O postulated that compound 155’gave only ringv
opened 1551and 1 2 gave only ringvclosed 125 because of the
differenCe in the stereochemistry of 155 and 159, If the
triene 155 is to be formed from 152, the C-9 hydrogen must
pass by the methyl group to which it was originally opposed.
The repulsion between them should inhibit ring cleavage and

thereby permit the cyclization to a cyclobutene to occur.

 

‘5.

111

hv )

HO HO /
138 139
hv a II
I
HO \ H0
140 141

W
W

Having the §g£i_relationship at the 9 and 10 positions,
compound 155 on the contrary is constructed with no steric
impediment to triene formation.

On the other hand, the formation of only 155'and 152
from the photolysis of 125 and only 121,and 525’from the
photolysis of 192.i3 most reasonably explained by inductive
effects. The photolysis of 155 generates a zwitterion
which can undergo a 1,4-sigmatropic rearrangement in two
different ways to give either the transient intermediate
122,0r 125, Collapse of 125 will give the observed photo-
products, lQZ’and 155, The formation of only 135 is reason-

able because it contains the more stable carbonium ion

 

112

 

O

hV 1,4 shift .\
a "_'-72 .
142 F

104
Pr 0 ' Pr
Me
.0 Me
1,4-shift

(tertiary carbonium ion of 125’y§_a secondary carbonium ion
of 152) and would be preferred. The same reason can be in—
voked for the formation of only 155 and 155 from the photo-
lysis of £25.

A mixture of steric and inductive effects can explain
the formation of the photoproducts from the photolysis of
the other unsymmetrical 2,5—cyclohexadienones mentioned
in Section D. The formation of only 155 from the photolysis
of 55 can also be explained by the difference instability

of the transient intermediates, 144 and 145. Intermediate

0 - O 0‘ 0
11:1,
1
VS 0/
O

 

 

114
144 is favored over 145 because it is the more stable car-

bonium ion. AS a result, 103 and not 146 is formed in the

 

photolysis. However, formation of only 5l’in the photolysis

of 55 can only be explained by steric factors. Inductive

   

 

\
O H
0" 1,4—shift
——-) ——>
H
\
g5
\
O-
0 \
H

lil———) 7‘

149

115

 

81

effects predict the formation of only 125 and not 51, which
is an unobserved result. The difference in the amount of
steric repulsion between 121 and 155 accounts for the forma—
tion of only 51, The large allyl group experiences more
steric repulsion between itself and the C-3 methyl and C—4
Efbutyl group in 121 than between itself and the C-3 :-
butyl group in 155, Therefore intermediate 135 is favored
and only 55 is formed. A mixture of these two effects can
also account for the results of Jeger and Schaffner as

mentioned in Section D.

EXPERIMENTAL
A. General Alkylation Procedure I.

In a 500-ml three-necked round—bottomed flask fitted
with a 250-ml dropping funnel and a three-way stopcock
connected to a nitrogen line, were placed 150 ml of toluene
and 4.2 g (0.100 mole) of a 56% mineral oil diSpersion of
sodium hydride. To this magnetically stirred solution,
which was flushed continually with nitrogen, 0.100 mole
of phenol in 100 ml of toluene was added drOpwise. After
addition was complete (30 minutes), the solution was stirred
for three hours; subsequently, 0.30 mole of allyl bromide
was added, and the solution was stirred at room temperature
overnight.

The solution was washed with 200 ml of water. The
organic layer was separated and dried over anhydrous mag-
nesium sulfate. The drying agent was filtered and the
toluene was removed under reduced pressure to give a yellow
oil. The oil was chromatographed on a 5 x 80 cm column of
60-100 mesh fluorisil. The column was eluted with hexane
until the ir spectrum of the eluted material showed no ab—

1 due to the O—alkylated product,

sorption around 1200 cm-
allyl polymethylphenyl ether. Eluant was then changed to
CCl4 to elute the desired product from the column.

116

 

117

B. General Alkylation Procedure II.

 

The procedure is the same as procedure I except that
methallyl chloride is used instead of allyl bromide, the
solution is refluxed overnight instead of stirred overnight
at room temperature, and the desired product is eluted from

the column with benzene instead of CCl4.

C. Synthesis of 4-Allyl—2,3,4,5,6—pentamethyl-2,5-cyclo-

 

hexadienone (21).

 

Compound 25 (7.82 g, 0.038 mole) was dissolved in 200 ml
of methylene chloride in a 500-ml Erlenmeyer flask, and 60 ml
of 47% boron trifluoride etherate was added. After the solu-
tion was magnetically stirred at room temperature for eight
hours, the solution was quenched by pouring it over ice to
hydrolyze the boron trifluoride etherate. The organic layer
was separated, washed twice with water, and dried over an-
hydrous magnesium sulfate. The drying agent was filtered
and the solvent was removed under reduced pressure to give
6.79 g (88%) of 25, Compound 25 had the same Spectral
properties and retention time as 55 isolated from the alkyla-

tion of pentamethylphenol l2 (see Part I of this thesis).

D. Synthesis of 4-Propyl-2,3,4,5,6-pentamethyl-2,5—cyclo—

hexadienone (25).

 

In a sloping manifold assembly at one atomosphere, 0.75
g of Zl’in 30 ml of cyclohexane was hydrogenated over 60 mg

of five percent rhodium on charcoal. Reaction was run until

 

118
the uptake of hydrogen had ceased (3 hours). The catalyst
was filtered and the solvent was removed under reduced
pressure to give a quantitative yield of 25, Compound 25
had the same spectral properties as previously synthesized

25 (see Part I of this thesis).

E. Photolysis of 4-Propyl-2,3,4,5,6—pentamethyl-2,5—cyclo-

 

hexadienone (55).

 

In a quartz test tube, 0.75 g of 25 was dissolved in
12 ml of methanol and photolyzed at 25372 in a Rayonet
reactor. The photolysis was followed by gas chromatography
(5' X 1/4" SE—30, 1800, 90 ml/min of He). After the photoly-
sis was complete (3 hours), the solvent was removed under
reduced pressure to give 0.77 g of crude product. The vpc
trace of the crude product showed a ratio of ggdg—G-propyl-
1,3,4,5,6-pentamethylbicyclo[3.1.0]hexenone (Rt = 7.3 min)
to‘gxg—G—propyl-l,3,4,5,6—pentamethylbicyclo[3.1.0]hexenone
(R = 7.8 min) of abour 10:1 by comparison with authentic

t
samples of 11 and 15,

F. Synthesis of 6-Allyl—2,4,6-trimethyl-2,4—cyclohexadi—

enone (24).

 

By the use of procedure I, compound 25 was synthesized
in a 70% yield. The infrared spectrum of 52’showed bands
at 1640 and 1650 cm—1. The nmr spectrum (CC14) showed one—
proton multiplets from T 3.40 to 3.54 and 4.20 to 4.33, a

three-proton multiplet from T 4.40 to 5.42, one- proton

 

119
doublets (J = 4.0 Hz) at T 7.76 and 7.86, three-proton doub-
lets (J = 1.3 Hz) at T 8.12 and 8.21, and a three—proton
Singlet at T 8.92. The uv Spectrum in methanol displayed

a maximum at 318 nm (e = 3800).

G. Rearrangement of 6-Allyl—2,4,6-trimethyl—2,4—cyclohexa—

 

dienone: Synthesis of 4—Allyl-2,4,6-trimethyl-2,5-

 

cyclohexadienone (51).

 

Compound 51 (3.00 g, 0.017 mole) was dissolved in 250
ml of methylene chloride in a 500—ml Erlenmeyer flask. The
magnetically stirred solution was cooled in an ice bath and
Six drops of 47% boron trifluoride etherate were added to
the solution. After two minutes, the reaction was quenched
by pouring the solution over ice. The organic layer was
separated, washed twice with water, and dried over anhydrous
CaClZ. The drying agent was filtered, and the solvent was

removed under reduced pressure to give a quantitative yield

The ir spectrum of 51 showed bands at 1630 and 1665 cm_1.
The nmr Spectrum (CCl4) showed a three-proton singlet at
T 8.85, a six-proton singlet at T 8.16, a two-proton doublet
(J = 5.5 Hz) at T 7.70, a three-proton multiplet from T 4.86
to 5.33, and a two—proton Singlet at T 3.52. The uv spectrum

in methanol showed a maximum at 245 nm (e = 8700).

120

H. Reduction of 4—Allyl-2,4,6-trimethyl—2,5—cyclohexadi-
enone: Preparation of 4-Propyl-2,4,6-trimethyl-2,5—

 

cyclohexadienone (g5).

 

In a sloping manifold assembly at one atmosphere, 3.00 g
of 21.1“ 100 ml of cyclohexane was hydrogenated over 200 mg
of 5% rhodium on charcoal. Reaction was run until the up-
take of hydrogen had ceased (5 hours). The catalyst was
filtered, and the solvent was removed under reduced pressure
to give a quantitative yield of 95. The analytical sample
was obtained by vpc purification and trapping (5'x 1/4" SE—30,
180°, He flow of 100 ml/min).

Anal. Calcd. for C12H180: C, 80.85; H, 10.18.

Found: C, 80.88; H, 10.12.

The ir spectrum of 55 showed bands at 1635 and 1660 cm—1.
The nmr spectrum (CC14) showed a six-proton singlet at T
8.17, a three-proton singlet at T 8.83, a seven—proton multi-
plet from T 8.33 to 9.27, and a two-proton Singlet at T 3.59.

The uv spectrum of 95’in methanol displayed a maximum at

244 nm (e = 8400).

I. The Photolysis of 4-Propyl—2,4,6-trimethyl—2,5—cyclo-

 

hexadienone (55).

 

In a quartz test tube, 500 mg of 55 was dissolved in
10 ml of methanol and photolyzed at 25372 in a Rayonet re—
actor. The photolysis was followed by gas chromatography
(S'x 1/4" Hallcomid M-18-OL, 1100, 100 ml/min of He). After

the photolysis was complete (1.5 hours), the solvent was

121

 

 

.Abmvmcocmwwwxm£ofio>01m.mlaxzquHHuI®.v.Nuahaamsv mo Avaoov Esuuommm HEZ .bH muomam
_. . . r - . L > (If) . . _ - . T- Lu; .r-rlvo-r:.~ If)? [v-_.-.itr t _ r _ ... . n L t . » H . r) (If ..r. Ir :_---L.ltc.rlr.L|Ir.lr)ro.bil_l
a. co Qm ox f: ----uo;3 awfafl:-wd. - - as 0%

 

III}! Ils.

ii) 14> l_.7.3.3.323;3):?!)/. IN)

)
5.. writ.
d

 

122

removed under reduced pressure to give 501 mg of a light
yellow oil. The vpc trace of the crude product showed
two compounds in the ratio of 44:56. They were 2555-6—
propyl-1,3,6—trimethylbicyclo[3.1.0]hexenone £59 and gng
6—pr0pyl-1,3,6-trimethylbicyclo[3.1.0]hexenone 15%, re—
spectively. These two compounds were separated by gas
chromatography (10' x 1/4“ DEGS, 140°, 120 ml/min of He).

Egggf6-propyl—1,3,6-trimethylbicyclo[3.1.0]hexenone

MeOH

100, R - 7.0 min, was a colorless oil with a xmax = 325 sh

t
(€’= 750), 275 (e = 2000),and 2351un(€ = 4800). Its ir
Spectrum showed bands at 1640 and 1685 cm_1. The nmr spec—
trum (CCl4) showed three-proton singlets at T 8.85 and
8.70, a three~proton doublet (J = 1.0 Hz) at T 8.33, one
proton-multiplets from T 8.08 to 8.23 and T 2.95 to 3.15,
and a seven-proton set of two multiplets from T 8.73 to
8.78 and T 9.05 to 9.33. The analytical sample was collected
by Vpc (same conditions as for separating the photoproducts).
Anal. Calcd. for C12H180: C, 80.85; H, 10.18.
Found: C, 80.48; H, 10.27.
Exg—6-propyl-1,3,6—trimethylbicyclo[3.1.0]hexenone

101, R = 9.0 min, was a colorless oil with Aޤ3H = 320 sh

-~ 't

(e = 350), 276 (e = 2300), and 236 nm (e = 4700). Its ir
spectrum showed bands at 1639 and 1685 cm-1. The nmr spec-
trum (CCl4) showed three-proton singlets at T 8.73 and 9.08,
a three-proton doublet (J = 1.0 Hz) at T 8.35, one-proton
multiplets from T 8.12 to 8.23 and T 2.98 to 3.15, and a

seven proton set of two multiplets from T 8.50 to 8.63 and

123
T 8.92 to 9.25. Trapping by vpc (same conditions as above)
gave the analytical sample.
Anal. Calcd. for C12H180: C, 80.85; H, 10.18.

Found: C, 80.63; H, 10.14.

J. Synthesis of 2,4,6-tris-Chloromethyl-meta—cresol (118).

 

In a three-necked 500-ml round-bottomed flask fitted
with an inlet tube for hydrogen chloride gas, 13.5 g
(0.150 mole) of paraformaldehyde was dissolved in 100 ml
of dioxane. Hydrogen chloride was bubbled through the
solution at room temperature until the solution was saturated
with the gas. Then 10.8 g (0.100 mole) of mfcresol was
added, and hydrogen chloride was bubbled into the solution
for another thirty minutes. The solution stood overnight,
and the white solid was filtered the following morning.
The white solid was recrystallized from 63-750 petroleum
ether to give 18.1 g (72%) of 1,13; (mp 100-1020, lit.32 104-
106). The nmr spectrum (CCl4) showed two-proton singlets
at T 5.47, 5.43, and 5.28, one-proton singlets at T 2.85

and 5.20, and a three—proton singlet at T 7.60.

K. Synthesis of Isodurenol (117).

 

In a two-liter three—necked round-bottomed flask fitted
with a dropping funnel and a reflux condenser, was placed
400 ml of dry tetrahydrofuran and 8.0 g (0.21 mole) of
lithium aluminum hydride. Compound $55 (17.0 g, 0.067 mole)

in 500 ml of dry tetrahydrofuran was dropped into the

 

124

Amdflv mcocmxmz
IHo.H.maoHo>OHflawzumEHuulo.m.Hiawmoumlouopcw mo Avaoov Esuuommm HEZ .wH whomem

 

.lf!‘l|li|¥)|pl P p 1— r) . .l'III IrDL - .— p n n b b ’L it P b L r > .— r + h r ~ L r P .> F bl) P L b > I ) 4 L h + )r r P It.) I? L) r L b P > s h u

I

c. co Qm QR Q5 45:53 Qm Qe Wu
. ...4 .7 r ...4 ._ ... ...... rt...‘ 3 3......7: _ r ... ...;...............:...r ..x of... res.........:.!:....... 6...... ._ ....
.....m 3.4... ...r... Rev/4&4; ...“. ......- .....d....:».~ 2»P_.4._1.._.....4\ .r......f1:11.“...4Jns...\/....45....1:... $1.1: ._ 44... 4:. .... . I. ....\«~

efililmflfkf WW 4 ..i \HTA
e... ... i i... .
I _

 

 

".-..-- _.__..- -Ho-

—_ ...-r 1...

125

 

AHOHV mcocoxmn
Awaouv ESHpommm HEZ .mH musmflm

 

IHo.H.mLOHosuenHmeumsHuuue.m.Hueseoemuesoxm mo

1" DOI‘AF'SI . lr’V" . 5 b p . L b L.) n L ) h) r ‘ r + p L — F s F L — P b n r w . b )l y ? r)— + F) ? F . “P > ) — r
k

2 e... ...n. ox o... ...... 2.... o m o...

(1.8. < 0 17'4“.“
a . 5.45:3. (1 7’

n —- ‘ b .- .
...........u..b......\.-.u._.........».eu.r.l..o1._;r2..$.31. .. 1.... r. p . .. .... . .. . ..
a. 4M—.K.,.J.n.-m......9.1.1.! .....I..~c.|<......\~.J.‘|d.. Lat. .. .. ....o 4“ . 14.-:1. an... -. ... ...-... 1.. v - lit-.— . ... / ... .
o .

 

 

 

 

 

 

--.-——.-—_- .. 4-.. .
~

 

126
magnetically stirred suspension at such a rate as to cause
a gentle reflux. After addition was complete, the mixture
was refluxed for four hours. The mixture was allowed to
stand overnight. Water was cautiously added to decompose
the excess lithium aluminum hydride. Next, 6g hydrochloric
acid was added to dissolve the precipitated salts. The
organic layer was separated, and the solvent was removed
under reduced pressure to give an oil. Steam distillation
of the oil gave 9.0 g (90%) Of,£LZ as a white solid (mp
75-770; lit.33 79-800). The nmr Spectrum (CCl4) showed a
nine—proton singlet at T 7.96, a three—proton singlet at

T 7.93, and one-proton singlets at T 3.52 and 5.68.

L. Synthesis of 6-Allyl-2,4,5,6—tetramethyl-2,4—cyclo—
hexadienone (119) and 6-Allyl-2,3,4,6-tetramethyl—2,4-

cyclohexadienone (120).

 

By the use of procedure I, compounds 119 and 155 were

sunthesized in 62% yield. The mixture of 119 and 120 was

a yellow oil with AMeOH
max

 

= 325 nm (e = 3350). Its ir spec-
showed bands at 1635 and 1050 cm-1. The nmr spectrum (CCl4)
of 155 and 1 5 showed three—proton singlets at T 8.97 and

8.88, a nine proton braod singlet at T 8.20, a nine—proton

multiplet from T 7.96 to 8.11, a four-proton multiplet from
T 7.38 to 7.96, a six—proton multiplet from T 4.41 to 5.46,
and one-proton multiplets from T 4.12 to 4.28 and t 3.39 to
3.51. Collecting by vpc (2' x 1/4" 53-30, 1300, 120 ml/min

of He) gave the analytical sample.

127

..HNHV meocmee

ImxmLoHoholm.muamnumemuumun®.v.m.NIH>HHmIv mo Avauuv Esmuommm H82

.om musmflm

 

 

 

 

 

5- - _
H

 

128
Anal. Calcd. For C13H180: C, 82.06; H, 9.53.

Found: C, 82.01; H, 9.56.

M. Synthesis of 4-Allyl-2,3,4,6—tetramethyl-2,5—cyclo—

hexadineoneL§gL).

 

In a 500-ml Erlenmeyer flask, 3.00 g of a 1:1 mixture
of 155 and 155 was dissolved in 250 ml of methylene chloride.
To this magnetically stirred solution, cooled in an ice
bath, were added six drops of 47% boron trifluoride etherate.
After seven minutes, the reaction was quenched by pouring
the solution over ice. The organic layer was separated,
washed with water, and dried over anhydrous calcium chloride.
The calcium chloride was filtered, and the solvent was re-
moved under reduced pressure to give a quantitative yield
of Ag}, Trapping by Vpc (5' X 1/4" SE-30, 150°, 120 ml/min
of He) gave the analytical sample.

Anal. Calcd. for C13H130: C, 82.06; H, 9.53.

Found: C, 82.09; H, 9.48.

The ir spectrum of £25 showed bands at 1620 and 1660
cm—1. The nmr spectrum (CCl4) showed a three-proton sing-
let at T 8.80, a nine-proton quartet (J = 0.8 Hz) at T 8.13,
a two-proton multiplet from T 7.52 to 7.74, a three-proton
Inultiplet from T 4.65 to 5.33, and a one-proton quartet (J =
1.1 Hz) at T 3.50. The uv Spectrum of 12}.in methanol dis—

Lplayed a maximum at 244 nm (e — 8500).

129

N. Reduction of 4-Allyl-2,3,4,6-tetramethyl-2,5-cyclo-

 

hexadienone (121): Synthesis of 4-Propyl—2,3,4,6—

 

tetramethyl-2,5-cyclohexadienone (104).

 

In a sloping manifold assembly at one atomosphere, 3.00
g of £55 in 100 ml of cyclohexane was hydrogenated over 200
mg of 5% rhodium on charcoal. The reaction was run until
the uptake of hydrogen had ceased (5 hours). After the
catalyst was filtered, the solvent was removed under reduced
pressure to give a quantitative yield of 152, Trapping by
vpc (5' X 1/4" SE—30, 1500, 120 ml/min of He) gave the
analytical sample.

Anal. Calcd. for C13H200: C, 81.20; H, 10.48.

Found: C, 81.19; H, 10.49.

The ir spectrum of 525 showed bands at 1630 and 1665
cm-1. The nmr spectrum (CCl4) Showed a three-proton singlet
at T 8.86, a seven-proton multiplet from T 8.89 to 9.28, a
nine-proton broad singlet at T 8.16, and a one—proton quartet

(J = 1.0 Hz) at T 3.58. Its uv Spectrum in methanol showed

a maximum at 245 nm (e = 8400).

O. Photolysis of 4—Propyl—2,3,4,6—tetramethyl-2,5—cyclo-

 

hexadienone (104).

 

In a quartz test tube, 500 mg of 155 was dissolved in
10 ml of methanol and photolyzed at 25372 in a Rayonet re-
actor. The photolysis was followed by gas chromatography
(5' x 1/4" Hallcomid M-18-OL, 1400, 100 ml/min of He). After

two and a half hours, the starting material was essentially

130

gone, and removal of solvent under reduced pressure gave
498 mg of crude product. The vpc trace showed the crude
product to consist of two photoproducts, present in the
ratio of 68:32. The photoproducts were EEQQ-G—propyl-l,
3,4,6—tetramethylbicyclo[3.1.0]hexenone 151 and 2&2‘6‘
propyl-1,3,4,6—tetramethylbicyclo[3.1.0]hexenone 155,
respectively. The two compounds were separated by vpc
(10' x 1/4" FFAP, 1650, 130 ml/min of He).

Egggf6-propyl-1,3,4,6-tetramethylbicyclo[3.1.0]hex—

MeOH

= 10.1 min, was a colorless oil with a 1m :

enone 107, R

t
320 sh (C = 650), 278 (e = 2360), and 232 nm (e = 4440).
Its ir spectrum showed bands at 1640 and 1685 cm-1. The
nmr spectrum (CC14) showed three—proton quartets (J =
0.7 Hz) at T 8.06 and 8.45, three—proton singlets at T 8.78
and 8.87, a one-proton multiplet from T 8.20 to 8.33, and a
seven—proton set of two multiplets from 8.80 to 8.91 and
T 9.03 to 9.39. Collection by vpc (same conditions as for
separating the photoproducts) gave the analytical sample.
Anal. Calcd. for C13H200: C, 81.20; H, 10.48.
Found: C, 81.03; H, 10.50.
Eng6-propyl—1,3,4,6—tetramethylbicyclo[3.1.0]hexenone
155, Rt = 15.0 min,was a colorless oil with AfiggH = 320 sh

(€‘= 455), 278 (e = 2360), and 232 nm (e = 4740). The ir

1 The

Spectrum of 108 Showed bands at 1640 and 1685 cm-
Ilmr spectrum (CC14) showed three-proton quartets (J = 0.7
112) at T 8.11 and 8.43, three—proton singlets at T 8.77 and

9.15, a one-proton multiplet from T 8.23 to 8.38, and a

 

131

.dmmmv mcocmxmz
IHo.H.m.oHomoHnHSeemsmehmh
Imevem.anammonmlmnoxm

mo AvHOUV Esuuommm HEZ .NN mnsmflm

 

4 .2 ., v. _
. ...... . .... _ . .4
. .4 _. . .. 2.
_ .. . . ..
‘ ._ n :V.\ _..
.. i M ..N .m
_ ‘ ._ ..
H“ ._ .H .44 ..
.. . ._ .
_ .. ..

 

 

.AMNNV msocmxmz
IHO.H.maoaomoflnamnuwfimuump
Imevem.HlH>moumI®Iopcm

mo AVHUUV Enuuommm HEZ

 

.Hm musmflm

132

seven-proton set of two multiplets from T 8.53 to 8.68 and
T 8.92 to 9.28. Trapping by vpc (same conditions as above)
gave the analytical sample. -

Anal. Calcd. for C13H200: C, 81.20; H, 10.48.

Found: C, 81.26; H, 10.57.

P. Synthesis of 4—Methallyl—2,4,6-trimethyl-2,5-cyclohexa-

 

dienone (124).

 

By the use of procedure II, compound 153 was synthe-
sized in 84% yield. The ir spectrum of Agg’Showed bands
at 1640 and 1665 cm_1. The nmr spectrum (CCl4) showed a
six-proton singlet at T 8.14, two—proton broad singlets at
T 3.47 and 7.72, a two-proton multiplet from T 5.24 to 5.44,
a three-proton doublet (J = 0.9 Hz) at T 8.42, and a three-
proton singlet at T 8.85. The uv spectrum of lgg'in meth-
anol showed a maximum at 247 nm (e = 14,200). 'The analyti—
cal sample was obtained by vpc purification and trapping
(5' x 1/4" 53-30, 1500, 120 ml/min of He).

Anal. Calcd. for C13H180: C, 82.06; H, 9.54.

Found: C, 82.25; H, 9.50.

Q. Reduction of 4-Methallyl-2,4,6-trimethyl—2,5—cyclohexa—
dienone: Synthesis of 4-Isobutyl-2,4,6-trimethyl-2,5—

 

 

cyclohexadienone 122.

 

In a sloping manifold assembly at one atmosphere, 3.00
g of 124 in 100 ml of cyclohexane containing one ml of tri-

ethylamine was hydrogenated over 5% rhodium on charcoal.

133

IflomxmonO%OI

 

 

.AVNHV mcocm

a.Nlahsumefiyu|®.v.mlahaamzumfiuv mo Avaoov Esuuommm HEZ .mm musmflm

LP L ? > L _ I) L b L 5 . _ IL o—m D L D .— L L P) b L P L 5 FL 5 DLL
I

as .S.EP.) Qt os

31)....) J. R... ..

......-

I“

 

 

 

134

The reaction was continued until the uptake of hydrogen
had ceased (12 hours). The catalyst was filtered, and the
solvent was removed under reduced pressure to give 3.01 g
of a light yellow oil.

The oil was chromatographed on a 2 x 20 column of 80-
200 mesh activated alumina. Elution with benzene gave 122’
(70%). The analytical sample was collected by vpc (5' x
1/4" 53—30, 1500, 120 ml/min of He).

Anal. Calcd. for C13H200: C, 81.20: H, 10.48.

Found: C, 81.14; H, 10.51.

The ir spectrum of 122 showed bands at 1635 and 1662
cm-1. The nmr spectrum (CCl4) showed a six—proton singlet
at T 8.13, a three-proton singlet at T 8.85, a two-proton
broad singlet at T 3.46, a three-proton set of two multiplets
from T 8.37 to 8.53 and T 8.88 to 9.35, and a six-proton

doublet (J = 6.5 Hz) at T 9.22. The uv Spectrum of 122 in

methanol showed a maximum at 244 nm (e = 14,450).

R. The Photolysis of 4-Isobutyl-2,4,6—trimethyl-2,5—cyclo-

 

hexadienone (122).

 

In a quartz test tube, 0.500 g of 122 was dissolved in
10 ml of methanol and photolyzed at 25378 in a Rayonet re—
actor. The photolysis was followed by vpc (5' x 1/4"
Hallcomid-M-lB-OL, 1250, 100 ml/min of He). After the
photolysis was complete (3 hours), the solvent was removed
under reduced pressure to give 0.510 g of crude product.

The vpc trace of the crude product showed the presence of

‘21: A

135

two compounds in the ratio of 53:47. These were enggf6-iso—
butyl—1,3,6-trimethylbicyclo[3.1.0]hexenone 125 and egng—
isobutyl-1,3,6-trimethylbicyclo[3.1.0]hexenone 126” respec-
tively. The two compounds were separated by preparative
Vpc (5' x 3/8" SE-30, 180°, 100 ml/min of He).

The ir spectrum of 125” Rt = 8.8 min, showed bands at
1640 and 1690 cm_1. The nmr Spectrum (CCl4) showed three—
proton singlets at T 8.72 and 8.89, a three—proton doublet
(J = 1.0 Hz) at T 8.32, one—proton multiplets from T 3.13
to 3.21 and 1 8.13 to 8.25, a three—proton multiplet from
T 8.78 to 9.08, and three—proton doublets (J = 6.5 Hz) at
T 9.24 and T 9.27. The uv spectrum of 125 in methanol
showed maxima at 324 sh (e = 240), 274 (é'= 1360), and 240
nm (e = 4250). The analytical sample was collected by vpc
(same conditions as for separating the photoproducts).

Anal. Calcd. for C13H200: C, 81.20; H, 10.48.

Found: C, 81.31; H, 10.59.

The ir spectrum of 1223 Rt = 10.6 min, showed bands
at 1630 and 1690 cm-1. The nmr Spectrum (CCl4) Showed
three-proton singlets at T 9.08 and 8.74, one-proton multi—
plets from T 3.00 to 3.15 and T 8.10 to 8.18, a three-proton
doublet (J = 1.0 Hz) at T 8.32, a three—proton set of two
Inultiplets from T 8.42 to 8.83 and T 8.95 to 9.20, and a
six-proton doublet (J = 7.0 Hz) at 1 9.05. Its uv Spec-
trum in methanol showed maxima at 320 sh (e = 150), 274

(E = 2400Land 2&3nm (e = 5800). The analytical sample was

collected by vpc (same conditions as above).

136

warp

. mcocmxmn
o co mo Avaoommmwwuommm HEZ .vm musmflm
uHo.H.monosoenHSLumEHHuue.m.H-H»psnomeuel e

_

 

 

 

. . ind.
_ I’lllvlltalr‘l.‘ .) .n)
_ ...:I } {Elk} .Il'l .r \
-. )iltrr- .L .r.rLLi{)UJW: em on . .:a
. r. e. o. .. .. / e<2e.;r:. asexuaywe ...
a . ,. I. .. . ... :.34. e. .a m
(an r; $1.... \r/_!<a$¥1. 7;...hixmfinnfi2. .13”. 2.7; L1... 5.7.9510. ((5 I \
e a ...:z. x f)..-
«as E

 

 

 

“...-...

137

 

.AwNHV mcocmxmn
n20.H.m2oaomoenasnumeHuuae.m.Hnasusnomeueuoxm mo Aeaoov asupoomm Hez

 

 

.mm musmflm

 

 

 

ncllblulb!“ .IL'I .tlnhntr405‘rrltlrlll'srblfIIr’iflt. kl. inn) b > r L > Htllbplrll... Incala0pl.. r up L) hIIF L .( > > r u p b )(ulr. .)_..\
:2 av Q» os 03 ,b_2ee cm 9v cm.
... ...»...5... 2.2.3.6 31:435....) .212....:>.p..$.{ ... 2:332??? .§§.Z>J$§X.. 27332.54/4» 5.2.. ...:xxéé
xx. 5;; 3 3 23 2 3
n. 3 x;2 2
22 2
2 2 . r...\.
2
2 2
2
. .2
2 2 _ 2.
.2 2 2:
. _,2 .
M2 2.
27% g
2 2. . .
2.2 _ r. 2
22
2 ..
21
L
. 22 2
2 . 2
2 f
2 \.

138

Anal. CalCd. for C13H200: C, 81.20; H, 10.48.

Found: C, 81.09; H, 10.51.

8. Preparation of 4-Methallyl—2,3,4,6-tetramethyl—2,5—

cyclohexadienone (122).

 

 

By the use of procedure II, compound 1&2 was synthe—
sized in 71.5% yield. The ir spectrum of 1&2 showed bands
at 1635 and 1665 cm-1. The nmr spectrum (CCl4) showed a
three—proton singlet at T 8.85, three-proton doublets (J =
0.9 Hz) at T 8.53 and 8.14, a six-proton quartet (J = 1.5
Hz) at T 8.18, one-proton singlets at T 7.55 and 7.67, a
two-proton multiplet from T 5.38 to 5.62, and a one-proton
quartet (J = 1.2 Hz) at T 3.59. The uv spectrum of 122 in
methanol showed a maximum at 248 nm (e = 8200). Collecting
by vpc (5' x 1/4" SE—30, 150°, 100 ml/min of He) gave the
analytical sample.

Anal. Calcd. for C14H200: C, 82.30; H, 9.87.

Found: C, 82.31; H, 9.89.

T. Hydrogenation of 4-Methallyl—2,3)4,6-tetramethyl—2,5—

 

cyclohexadienone: Synthesis of 4-Isobutyl-2,3,4,6-

 

tetramethyl-2,5-cyclohexadienone (123).

 

In a sloping manifold assembly at one atmosphere, 1.00
g of 122.1“ 30 ml of cyclohexane was hydrogenated over 50 mg
of 5% rhodium on charcoal. The reaction was continued until
the uptake of hydrogen had ceased (3 hours). After the
catalyst was filtered, the solvent was removed under reduced

pressure to give 0.95 g of 123 (94%). Collecting by vpc

139

 

.hmmfiv mcocmaomxmz
l0HU>UIm~NIH%£memHumU'®sw~msmlflhHngumEIv m0 AVHUUV EDHgommm HEZ on“ mHsmHh

 

. .. .n . 1-3.. Int}. .. .1» , v v|.va .Ii'lxur IIIIIIIIIII'tlbI ..vl
on as an
4 (n o. A . «05;.‘3‘1I‘w-‘W’c’ui 3%
J J. \
-
.. 2
.. .B. |
‘7}.-' \|III‘

 

 

140
(5'x 1/4" SE-30, 150°, 100 ml/min of He) gave the analytical
sample.
Anal. Calcd. for C14H220: C, 81.50; H, 10.75.
Found: C, 81.58; H, 10.63.
The ir spectrum of 123 showed bands at 1635 and 1660
-1

cm . The nmr spectrum (CCl4) showed a nine-proton broad

singlet at r 8.17, a three—proton singlet at T 8.87, a one—

proton quartet (J ‘ 1.1 Hz) at T 3.46, a three-proton multi—
plet from T 8.25 to 9.25, and a six-proton doublet (J =
6.5 Hz) at 1 9.25. The uv Spectrum of 123 in methanol

showed a maximum at 248 nm (E = 8150).

U. Photolysis of 4—Isobutyl—2,3,4,6—tetramethyl-2,5—cyclo-

 

hexadienone (123).

 

In a quartz test tube, 0.500 g of lZ§.waS dissolved in
10 ml of methanol and photolyzed at 2537A in a Rayonet re—
actor. The photolysis was followed by gas chromatography
(10' x 1/4" FFAP, 1700, 80 ml/min of He). The photolysis
was stOpped when the starting material had been 95% con—
sumed (3 hours). The solvent was removed under reduced
pressure to give 0.510 g of crude product. A vpc trace of
the crude product showed lZ§,and 122,t0 be present in the

ratio of 77:23. The two COmpounds were separated under

the same conditions.

Endo-6-isobutyl-1,3,4,6-tetramethylbicyclo[3.1.0]—

. . . e _
hexenone 128: R = 35 min, was a colorless Oll Wlth AfiagH -

t
325 sh (e = 850), 274 (c = 2510), and 235 nm (E = 4250).

141

33%}!: <3 . a2j25 ¥32.
. . 33.. . 3

with

.dWva maocoxw:
n30.H.m3oaosoflnamsumsmuump
no.v.m.Hnamu:QOmHnonoxm

mo 3¢Hoov Esnuommm HEZ .wm musmflm

 

 

 

L:....... ..-r._. ...,..L.- 3.:+1:r
LLu oo Qm
.3 .. ... .
... 7..” ...‘LRA‘
2.<&. « .

 

 

 

 

 

.AwNHv mcocwxmz

-30.H.m3oHo»oHnHmnumamHumu

:©.¢.m.fl-fisusnomfluouowcm
mo Avaoov eouaommm HEZ .bm mmsmflm

 

7\/ J. 3:
2.332 \./ a. 3) ..
.. ... .7. ..3_ r ...\1‘
.. .... 33...... .33. , ... 5.x
: ... 3 . ..
_ ... 3. . . . .. .
.. ...3: .x3 3 . ..3.
. ..p: ... 3 ,—
... 3.33 3 .3 3.
. 3:33 3 . r
33.3 3» .
33.. 3 .
a,, 3 .
33. ... 3.
3 . P
._ .
r.
3
_

142

Its ir spectrum showed bands at 1640 and 1685 cm-1. The
nmr spectrum (CC14) showed three-proton singlets at 1 8.92
and 8.75, three-proton broad singlets at T 8.46 and 8.08,
three-proton doublets (J = 6.5 Hz) at T 9.22 and 9.33, a
one-proton broad singlet at T 8.38, and a three-proton
multiplet from T 8.95 to 9.38. The analytical sample was
collected by vpc (same conditions as above).

Anal. Calcd. for Cl4H220: C, 81.50; H, 10.75

Found: C, 81.69; H,'10.77.
Egng—isobutyl-1,3,4,6—tetramethylbicyclo[3.1.0]-

hexenone 129, R = 46.0 min, was a colorless oil with a

t
AmeOH = 322 sh (e = 790), 273 (e = 2390), and 235 nm (e =

max
3400). The ir spectrum of lggpshowed bands at 1640 and
1685 cm—1. The nmr spectrum (CC14) showed three-proton
broad singlets at T 8.07 and 8.47, a one-proton broad sing-
let at T 8.32, three-proton singlets at T 8.77 and 9.17, a
three-proton multiplet from T 8.50 to 9.20,and a six-proton
doublet (J = 7.0 Hz) at T 9.08. The analytical sample was
collected by Vpc (same conditions as above).

Anal. Calcd. for C14H220: c, 81.50; H, 10.75.

Found: C, 81.46; H, 10.78.

PART III

THE SYNTHESIS AND PHOTOREARRANGEMENT OF
1,6,8,8,9,10-HEXAMETHYLTRICYCLO[2.2.4.02I5]DECA-

3,9-DIENE-7-ONE AND ITS DERIVATIVES

143

W7.

INTRODUCTION

The photochemistry of fi,y—unsaturated ketones had re-
ceived much attention in the literature since 1963. Three
major rearrangement pathways have been observed: 1,3—acyl
migration, 1,2-acyl migration, and decarbonylation.

The most common of these is 1,3-acyl migration, first
observed by Buchi and Burgess41 during the irradiation of
1,4,4—trimethylbicyclo[3.2.0]hept-6-ene-2-one, 122, Other
examples include the rearrangement of lézfz and'lééf3. The
mechanism had generally been described as a NorriSh Type I
cleavage of the ketone to give an allyl-acyl diradical

which then recombined to give either the B,y-unsaturated

O

.0 _._.___) .0

(L____.

150 151

WI rvw

152 153

144

v
[.l

33.- F
4“

145

R-154 S-15

WV I'M

photoproduct or starting ketone. The photolysis of lég'by
Schaffner and co—workers43 placed doubt on the correctness
of this mechanism. Their results suggest that the reaction
proceeds by a concerted photochemically allowed 02 + v2
process.44

1,2-Acyl migration, the second reaction mode, has been
observed in both sensitized and direct irradiation of B,y-
unsaturated ketones. When Williams and Ziffer45 irradiated
léé’and lézidirectly in tfbutanol, they obtained only one

of the two possible cyclopropane ketones from each photolysis.

hV

J,

155 1 6

M W

They postulated that the reaction proceeded by a concerted
photochemically allowed 02 + W2 process.44 Similarly,

Dauben46 and Schaffner and Domb47 obtained cyclopropyl

146

OH OH
{lllllllliilllllllliyhV
O O '
1 7 158

ketones when they irradiated 159 and 1 1, respectively, in

hv
\
%L ”17
¢ acetophenone

NW

 

 

 

147
the presence of sensitizers. Since all four possible
stereoisomers were obtained from lgl/ Schaffner and Domb
postulated that the rearrangement occurred through a di-
radical mechanism. Irradiation of the fi,y-unsaturated

ketone generates the biradical intermediate lggfl which

a .
starting ketone—__) NA —) \"9 fiproduct
.0 o O o

162 152

rwv

cleaves at side a to generate a new diradical 1§§, Closure
of 1§§ gives the observed product. Thus present data seem

to favor a non-concerted biradical process for the 1,2-acyl
migration.

The third general mode of reaction in the photolysis of
B,y-unsaturated ketones is decarbonylation. Only a few ex-
amples exist in the literature. Starr and Eastman48 found
that photolysis of lfig gave a quantitative yield of the
decarbonylated produCts, 1§§ and lgg, and Erman49 found

that photolysis of 167 and 169 gave the cyclopropyl compounds

30—» ‘M

164 Nov
55% 45%

 

148

O.°——. /

167 168

6.0;. /

169 1 0

NW
NW

168 and 170, reSpectively. The E,y-double bond facilitates
ketone decarbonylation because of its ability to stabilize

a biradical intermediate such as 171.

171

rm

Another photochemical reaction of 5,y-unsaturated
ketones, observed in bicyclic[2.2.2]octadienones,has been
loss of the carbonyl-containing bridge and generation of
an aromatic compound. For example, Hart and Murray50 have

found that direct irradiation of 172 and 174 generated a

 

149
substituted benzene in addition to dimethylketene. They
proposed that the reaction proceeded either by direct
elimination of dimethylketene or initially by a 1,3-acyl
shift to give a cyclobutanone, which then eliminated di—

methylketene to generate 173 and 175.

W

 

O
COzMe ”‘
COzMe h V
+ ...
172 173
0
CD
/ " \
hv ' ... ...
¢
¢
17 1 5

If cyclobutadiene would add to 2,3,4,5,6,6-hexamethyl-
2,4-cyclohexadienone in a [4+2] Diels-Alder fashion to
generate 11g, another interesting compound would be available
to study in the general photochemical rearrangement of bi-
cyclic[2.2.2]5,y-unsaturated ketone systems. It not only
might undergo the usual photochemical rearrangements of 6,7-

unsaturated ketones but also there is the possibility for

150

" ‘K‘Nl.

formation of tetramethyl cyclooctatetraene by direct ir—

radiation and for a sensitized [2+2] photochemical cyclo—

Compound 176 and its deriva-

 

addition to give compound 177.

tives were synthesized and photolyzed to determine which of

the possible rearrangements would occur. This constitutes

Part.III of the thesis. Though the study is incomplete, the

results obtained thus far are of sufficient interest to

include here.

RESULTS AND DISCUSS ION

A. The Synthesis of Cyclobutadieneiron Tricarbonyl (181).

It was envisioned that 11g might be produced by the
[4+2] cycloaddition of cyclobutadiene to hexamethyl—2,4—
cyclohexadienone. Accordingly, the cyclobutadieneiron
tricarbonyl complex l§l was prepared, according to litera—
ture procedures.

The synthetic route to compound 1§1,utilizes chlorina-
tion of cyclooctatetraene at -20°, Diels-Alder addition of
dimethyl acetylenedicarboxylate to the dichloro product,

thermolysis of the adduct obtained, and addition of iron

c1
”A Cl
1 .c1z/cc14 \ COZMe
2 .CH302ccEc-CCZCH3 /

COzMe
17

W

C1

C1 //
/

Cl

 

 

H
q
(9
H
m
0

§
§

 

152

 

nonacarbonyl to the ging,4—dichlorocyclobutene obtained
from the thermolysis.

Compound lzg’was synthesized by the procedure of F
Nenitzescu et al.51 A solution of cyclooctatetraene in l ..
carbon tetrachloride at -300 was chlorinated by the addi- {I
tion of one equivalent of chlorine in carbon tetrachloride.

The product was treated with one equivalent of dimethyl
acetylenedicarboxylate, and the resulting oil was thermo—
lyzed at 185° and a pressure of 10 mm. The clear distil-
late showed two peaks on a vpc trace in the ratio of 4:1.
The two components were separated by use of a spinning band
column.

The major component, gig-3,4-dichlorocyclobutene, was
isolated in a 60% yield. Its ir spectrum was the same as
that reported by Nenitzescu gt_al.51, and contained bands at
1220 and 1280 cm-1. Its nmr spectrum (CC14) consisted of
two-proton doublets (J = 1.0 Hz) at T 4.93 and 3.78 due to
the methine and vinyl protons, respectively.

The minor component, trans-1,4-dichlorobutadiene, had
the same ir spectrum as reported in the literature51, with
a strong absorption at 1560 cm-1. Its nmr Spectrum (CCl4)

consisted of a multiplet from T 3.00 to 4.16.

153

The iron tricarbonyl complex of cyclobutadiene (1§1)
was synthesized according to the procedure of Paquette and
Wise.52 Iron nonacarbonyl, obtained by the photolysis of
iron pentacarbonyl,53 was added to a mechanically stirred
benzene solution of lzg’maintained at 600 under nitrogen
until the vpc trace Of the reaction mixture showed no re-
maining 112, The product was a dark orange oil, which was
not characterized or further purified, but was used immedi- V."

ately in the synthesis of 176.

 

B. The Synthesis and Characterization of 1,6,8,8,9,10-
Hexamethyltricyclo[2.2.4.02:5]deca-3,9-diene-7-one (1161,

 

 

Compound 176 was synthesized by freeing the cyclobuta-
diene from its complex in the presence of 2,3,4,5,6-hexa-
methyl-2,4-cyclohexadienone 2, using the procedure of

Paquette and Wise.52 Compound 181 was dissolved in an

 

H
00
H

(N

acetone solution of g and treated with excess ceric ammonium
nitrate, using a short reaction time. Workup and purifica-

tion by chromatography on alumina gave an off-yellow

154
semi-crystalline solid, assigned structure 176. The yield

was 76%.

It was not at all clear in advance that cyclobuta-
diene would behave as a dienophile toward the conjugated

cyclohexadienone. In principle, four types of addition are

possible:

H + [2+2]

 

\/

CE]

 

\/

H + [:11

 

/ + [4+2] >
\

< + -‘——-+41 oi

These lead to two types of products. In the present case,
the product seems to be of the latter type ([4+2] or [4+4]).
Structure 176 was assigned to the adduct on the basis

of the following Spectroscopic properties. The mass spectrum

 

155
of the adduct showed a parent peak at m/e = 230, which demon—
strated that it was a 1:1 adduct of cyclobutadiene and 2.

1

Strong absorption at 1700 cm- in the ir spectrum of the

adduct eliminated structure 182, which has a conjugated

 

 

O
O
.-
' ‘1
O /
Q Ca ~
1&2». 183 184

carbonyl group and should absorb at a lower frequency in

the infrared. The nmr spectrum (CC14) of the adduct showed
three-proton singlets at T 9.13, 9.08, 8.95, and 8.92, which
are due to four aliphatic methyl groups, three-proton
quartets (J = 1.0 Hz) at T 8.30 and 8.42 due to two adjacent
allylic methyl groups, a two-proton pair of doublets (J =
4.0 Hz) at T 7.28 and 7.54'due to two methine protons, and

a two-proton pair of doublets (J = 3.0 Hz) at T 4.25 and
4.00 due to the vinyl protons. The nmr spectrum does not
readily distinguish 11§,from 183, But these two adducts

can be distinguished by examining the nmr Spectrum of the
adduct prepared from labelled dienone 184, If the [2+2]
type structure lgg'were correct, it should show the loss of

one aliphatic methyl when the adduct is prepared from 184.

On the other hand, if the [4+2] type structure 116 were'

156
correct, it should show the loss of one allylic methyl and
the other allylic methyl should sharpen to a singlet, when
the adduct is prepared from 184,

Compound 184, 3-methyl-d342,4,5,6,6-pentamethyl-2,4-
cyclohexadienone was prepared by stirring unlabelled g'in
CHSOD with sodium methoxide as the base required for ex-
change. Integration of the nmr peaks of the labelled pro-
duct indicated that the C-3 methyl group was 95% exchanged.

The nmr spectrum of the adduct from 184 showed that the

quartet at T 8.42 was absent whereas that at T 8.30 had

 

become a singlet. These results are consistent only with
structure 11§,for the adduct. It is therefore possible to
assign the quartet at T 8.42 to the C-9 methyl and the
quartet at T 8.30 to the C-10 methyl of 116, No attempt
was made to characterize the other aliphatic methyl peaks
in the nmr spectrum.

Having demonstrated the gross structure of the adduct,

it was still necessary to determine whether addition occurred

0

185

in the endo (126) or exo (185) sense. The endo orientation

Of the cyclobutene ring was proved by the number of reduction

157
products of 116” the nmr spectrum of the reduction product,
and a photochemical transformation.
Reduction of lzg’with lithium aluminum hydride in ether
gave a 95% yield of two isomeric alcohols, 186’and 181,
The ir spectrum of the resulting mixture showed an abSOrp-
tion at 3450 cm.1 but none at 1700 cm_1, which demonstrated

that 176 had been completely reduced.

HO H H OH

/

1 6 1 7

w W

A Vpc trace of the mixture showed the presence of only
two compounds, in the ratio 7:5. The major component is
tentatively assigned structure l§§.°n the basis of mechan-
istic considerations. Since hydrogen is smaller than methyl,
the lithium aluminum hydride should attack preferentially
on the less hindered side to give lgg'as the major product,
though this argument is somewhat tenuous, since the methyl-
bearing carbons are s2? hybridized, whereas the hydrogen-
bearing carbons are s2? hybridized, thus placing the hydro-
gens closer to the carbonyl bridge. The formation of two
isomeric alcohols in almost a 1:1 ratio is consistent with

an gago orientation of the cyclobutene ring of 176, for in

 

158
lzg'there should be an almost equal probability that lithium
aluminum hydride can attack either side of the ketone func-
tion. On the other hand, in 185, the cyclobutene ring
shields attack from one side of the ketone function so that
only one alcohol should be formed, an unobserved result.

In addition, the nmr spectrum of the mixture of alcohols is
in complete accord with the endo orientation of the cyclo-
butene ring. All the vinyl protons of the mixture appear

at T 3.90, the same value as found in the ketone. If struc-
ture 185’ware correct, the vinyl protons should appear at
two different positions, separated by about 0.5 T. For ex-
ample, Winstein54 g£_al. found that H

b
of 189 occurred at T 7.05.

of 188 occurred at

T 7.6 whereas Hb

 

188 189

To seek a more definitive proof of the §g§2.orienta-
tion of cyclobutene ring, a [2+2] photocyclization of the
acetates of lggland lgz'was attempted. The acetates were
photolyzed rather than the alcohols, for the acetates should

be more easily handled and separated by preparative vpc.

 

 

Er

159
If the assignment of ggdg geometry is correct, photocycli—
zation should generate products whose nmr spectrum would
contain no vinyl protons or allylic methyl groups.
The mixture of alcohols 186 and lgz’was acetylated
with acetic anhydride and pyridine to give an 83% yield of

the corresponding acetates 190 and 191. The ir spectrum

OAc

 

191

190

§
§

of the mixture showed strong absorption at 1725 cm-1, which
is consistent with the formation of an acetate. The mass
spectrum of the mixture had a parent peak at m/e = 274, con-
sistent with the proposed structures. The vpc trace and

the nmr spectrum of the mixture suggested that it was a

7:5 ratio of 122.3nd 121, The 7-proton, appearing at T 5.70,
integrated for one prOton and is assigned to structure 121’
whereas the 7-proton appearing at T 5.60 integrated for 1.4
protons and is assigned to structure 122, These assignments
are based on the fact that the 7-proton of lgA'Should be
shielded by the 9-10 double bond and thus appear at higher

field than the 7-proton of 190. The acetoxy methyl protons

 

160

show the same effect: the acetoxy methyl group of lgg’ap-
pears at T 8.06 whereas that of 121 appears at T 8.00; No
further assignments of the nmr spectrum were made (see the
experimental section for the complete spectrum). The mix—
ture was not separated but the crude product was immedi-
ately photolyzed.

A 1% acetone solution of the 7:5 mixture of lgg’and

191 was photolyzed with a 450 watt Hanovia Type L mercury

 

H
9 OAC
/
190 + 191 _hé_) '7‘
M w ace one
Corex
635‘ 193

arc lamp through Corex for 24 hours. The nmr spectrum of
the crude product showed the absence of vinyl protons and
allylic methyl groups, which is consistent with the cyclized
structures 122 and 123; The resulting orange oil was
chromatographed on 604200 mesh activated alumina to remove
any polymer from the oil. Elution with benzene gave a 63%
yield of a 7:5 mixture of lgg’and 123, These acetates
were separated by preparative vpc.

The major photoproduct was considered to be 122, the

photolysis product of 190. The mass spectrum of 192 had

 

161
a parent peak at m/e = 274, which showed that 122 was an
isomer of acetates 122 and 121, It had an ir spectrum
with a strong absorption at 1725 cm-1. Its nmr spectrum
(CC14) showed singlets at T 9.23 (3H), 9.12 (3H), 9.05
(6H), and 8.90 (6H) due to the aliphatic methyl groups, at
T 8.02 (3H) due to the acetoxy methyl group, and at 1 5.47
(1H) due to the 9-proton, and a multiplet from T 7.08 to
7.48 (4H) due to the methine protons. No further assign-
ments of the nmr spectrum were made.

Compound 123” the minor photoproduct, had spectral
properties very similar to those of 122, The mass spectrum
of lgg’had a parent peak at m/e = 274, which demonstrated
that it also was isomeric with lgg'and 121; The ir Spec-
trum of lgg’had strong absorption at 1725 cm-1. Its nmr
spectrum (CC14) showed singlets at T 9.33 (3H), 9.20 (3H),
9.05 (6H), and 8.94 (6H) due to the aliphatic methyl
groups, at T 8.05 (3H) due to the acetoxy methyl group,
and at T 5.47 (1H) due to the 9-proton, and a multiplet
from T 7.08 to 7.48 (4H) due to the methine protons. No
further assignments of the nmr spectrum were made._ The nmr
spectra of lgg’and 123 are not sufficiently different to
allow a definitive structural assignment. The structures
given here are simply based on the fact that the same
ratio was obtained as the original ratio of photolyzed
acetates.

The structure proof of compound 176 seems reasonably

cxmnplete. It could only have an endo cyclobutene ring

162
because the photolysis of lgg’and lgl’produced the expected
ring-closed photoproducts.

In order to be certain that compounds 122 and lgg'dif-
fered only by the orientation of the acetoxy group at C-9,
each compound was reduced separately to the corresponding
alcohol. Both alcohols gave the same ketone, on oxidation.

Compound lgg’was reduced with lithium aluminum hydride
in ether to give a 90% yield of alcohol 124, Its Spectro—
scopic properties were consistent with the proposed struc-
ture. The mass spectrum had a parent peak at m/e = 232.

The ir spectrum had an absorption at 3500 cm-1, due to the
alcohol function. The nmr spectrum (CC14) of 122'showed
signlets at T 9.33 (3H), 9.18 (3H), 9.07 (3H), and 8.92 (9H)
due to the aliphatic methyl groups and at T 7.00 due to the
9-proton, quartets (J I 3 Hz) at T 7.48 (2H) and 7.30 (2H),
and a broad singlet at T 8.75 (1H) (disappeared on shaking
'with D20) which is due to the OH proton.

Reduction of lgg’with lithium aluminum hydride in ether
gave an 85% yield of alcohol 122, Its spectroscopic proper-

ties are consistent with the proposed structure. The mass

 

 

 

 

163

OH

 

 

H
(O
01

§

spectrum had a parent peak at m/e = 232. The ir spectrum is

 

had an absorption at 3500 cm-1, consistent with the alcohol
structure. The nmr spectrum (CCl4) of 125'showed singlets
at T 9.33 (3H), 9.22 (3H), 9.10 (3H), 9.03 (6H), and 8.97
(3H) which are due to the aliphatic methyl groups and at
T 7.03 (1H) due to the 9-proton, a multiplet from T 7.17 to
7.50 (4H) due to the methine protons, and a broad singlet
at T 8.68 (1H) (disappeared on shaking with D20) which is
due to the OH proton.

Each alcohol, 124,and 122, was separately oxidized
with Jones reagent55 to give a 90% yield of the same ketone,

177. Its spectroscopic properties are consistent with the

Jones

 

\
122.0r 122' Reagent 7

 

164
proposed structure. The mass spectrum of lZZ,had a parent
peak at m/e = 230. The ir spectrum had a strong absorp-
tion at 1690 cm-1, normal for a cyclohexanone. The nmr
spectrum (CC14) consisted of singlets at T 9.18 (3H), 9.12
(an), 8.97 (6H), 8.93 (3H), and 8.87 (3H), all due to the
aliphatic methyl groups, and a multiplet from T 6.66 to
7.26 (4H) due to the methine protons.

The formation of a single ketone at this stage con-
firms the essential correctness of the assigned structures,
the only possible errors being in the assignment of geometric
isomers.

Next, attention was turned to photolysis of the parent
compound 116; In particular, it was of interest to determine

whether the photolysis of 176 would generate 177.

C. Direct Irradiation of 1,6,8,8,9,10-Hexamethyltricyclo—
[2.2.4.02I5]deca—3,9-diene-7-one (176).

 

Irradiation of a 2% pentane or methanol solution of
lzg’with a 450 watt Hanovia Type L mercury arc lamp through
Corex gave two products, whose relative yields depended on
the irradiation time. The products were separated by
preparative vpc and were assigned structures lgg'and 121”
The ratio of lgg/ng'was initially high, ranging from 3:1
in pentane to 5:1 in methanol. Prolonged irradiation of

176 gave an essentially quantitative yield of 197.

an”? .

 

 

165

 

o
H
/ b _h_v__, $3
Corex
MeOH 1' H
or
pentane V
\\
176 196 197 rm

twv
rwv ’VW

Structures 126 and 191 were assigned on the basis of
spectroscopic and mechanistic criteria. The compound as—
signed structure 196 was a colorless oil with a carbonyl
band at 1760 cm"1 (cyclobutanone55) and a parent peak in
its mass spectrum at m/e = 230. The uv Spectrum of 126 in
methanol showed maxima at 318 (e = 397) and 235 nm (e'=
1860). Its uv spectrum was very similar to that of 116,
which demonstrates that it must be also a fi,y-unsaturated
ketone with interaction between the ketone function and the
cyclohexene double bond. Its nmr Spectrum (CCl4) showed
four aliphatic methyl singlets at T 9.10, 8.98, 8.88, and
8.77, a broad Six-proton Singlet at T 8.33 due to allylic
methyls, a broad two-proton singlet at T 6.84 due to the
methine protons, and a two-proton singlet at T 3.82 due to
the vinyl protons. The presence of the cyclobutanone ring
required that Alfiihad undergone a 1,3-acyl shift. Only struc-

ture 196 is consistent with this mechanistic interpretation.

166
However, the nmr spectrum in carbon tetrachloride is not
entirely consistent with this structure, for one might not
have expected that the allylic methyl groups and the vinyl
and methine protons would be singlets. The nmr spectrum was
taken in another solvent (C6D6) to determine whether the
Spectrum would change. In this solvent, the nmr Spectrum
consisted of four aliphatic methyl singlets at T 9.15, 9.13,
8.93, and 8.85, two allylic methyl quartets (J = 0.9 Hz) at
T 8.32 and 8.54, a two methine quartet (J = 0.9 Hz) at T
7.05, and a vinyl proton quartet (J = 0.9 Hz) at T 3.90.
These data are entirely consistent with structure 122,

Compound lgg'was assigned with the cyclobutene'and
cyclobutanone rings anti to one another on the basis of
mechanistic and steric criteria. The presently accepted
mechanism for 1,3-acyl migration is that they occur in a
concerted suprafacial manner. This requires that the two
rings be anti to one another, for it has already been Shown
that the ketone function and cyclobutene ring are 332i to
one another in 116, Also, compound 126 in the anti orienta-
tion has less steric repulsion than its §y2_isomer.

Compound lgz’was a light yellow oil with a parent peak
at m/e = 202 and no carbonyl absorption in its ir spectrum:
these data indicate that carbon monoxide was lost from 126,
By analogy with the photochemical behavior of 128?7 and
since photolysis of 126 generates 121, the struCtural as-
signment seems correct. The nmr Spectrum (CCl4 and C6D6)

eliminates the alternative structure 201. In carbon

 

 

167

 

 

1 8 19 200 ,
'vvv rvvv ’vvv T—
tetrachloride, the nmr spectrum of 197 showed aliphatic E

methyl singlets at T 9.23 (3H), 8.97 (3H), and 8.99 (6H)

xb

two allylic methyls as a singlet at T 8.38, two methine
protons as a pair of doublets (J = 4 Hz) at T 7.04 and 7.22,
and a two—vinyl proton singlet at T 3.82. The appearance

of singlets for the allylic methyls and vinyl protons was
somewhat disconcerting, though the pair of doublets for the
methine protons favors (but does not absolutely require)
structure 121, To better resolve the spectrum, it was deter-
mined in CéDé. Here there were four aliphatic methyls at

T 9.13, 9.02, 8.98, and 8.95, two allylic methyl quartets

 

168

(J = 0.5 Hz) at T 8.37 and 8.43, a two—methine proton multi-
plet from T 6.86 to 7.25, and a two vinyl proton pair of
doublets (J = 1.5 Hz) at T 3.72 and 3.74. These data are
only consistent with structure 191,

Compound lgz’was assigned with the cyclopropane and
cyclobutene rings in an anti_orientation on the basis of
steric effects. The anti orientation has less steric re-

pulsion than the syn orientation.

D. Sensitized Irradiation of 1,6,8,8,9,10-Hexamethyl-
tricyclo[2.2.4.02r5]deca-3,9-diene-7-one (176).

 

 

Irradiation of a 1% acetone solution of 176 with a 450
watt Hanovia Type L mercury arc lamp through Corex for Six
hours gave a 92% yield of 202. Structure 202 was assigned

on the basis of spectroscopic and mechanistic criteria.

176 hv \

acetone7 0
450W
Corex

202

 

The ir spectrum of the photoproduct had a strong absorption
at 1715 cm.1 (cyclopentanone56). The mass Spectrum of Egg
had a parent peak at m/e = 230, which showed that it was
isomeric with the starting material. The nmr spectrum

(CCl4) showed no allylic methyls but only aliphatic methyl

MHZQL‘» .‘
l

169
singlets at T 9.17 (3H), 9.08 (3H), 8.97 (6H), 8.92 (3H),
and 8.87 (3H), a two methine proton quartet J = 1.0 Hz) at
T 7.06, and a two vinyl proton multiplet from T 3.69 to
3.98. From analogy with the photochemical behavior of 128’

and the lack of allylic methyls in the nmr spectrum, struc-

hv *; \
x
12§, aceEOne7 0 -C02Me
pyrex K

‘0 2M8

 

 

203

ture 222 seems correct for this compound. In this irradia-
tion of 116, there was no evidence for the presence of 111”
by vpc analysis.

These photolyses of 116 constitute yet another example
of the general photochemical rearrangement paths of B,y-un-
saturated ketones. On direct irradiation, they undergo 1,3-
acyl shifts to generate photoproducts which decarbonylate
if the resulting diradical is well stabilized whereas on
sensitized irradiation, they undergo 1,2-acyl shifts to form

cyclopropane compounds.
E. Consequences of This Study.

The synthesis of 176 and the photochemical transforma-
tions of its derivatives offer a general scheme for the

iiynthesis of substituted homocubanes. By the selection of

170
appropriate substituents, it is possible that the two carbon
bridge of the homocubane could be eliminated, thus afford-
ing a synthetic entry to various substituted cubanes. Per-
haps even more interesting is the demonstration that cyclo-
butadiene acts as a dienophile and adds 1,4 to conjugated
cyclohexadienones. It will be interesting to determine
whether this also holds for less substituted dienones, and

dienones in larger rings.

 

“a

EXPERIMENTAL

A. Synthesis of cis-3,4-Dichlorocyclobutene (179).

 

A solution of 21.0 g of cycloectatetraene in 50 ml of
CCl4 was placed in a 500—ml three-necked round-bottomed
flask fitted with a mechanical stirrer, a dropping funnel
to which was attached a condenser cooled with dry ice and
trichloroethylene, and a drying tube. The solution was
stirred and cooled to -400 by means of a bath of dry ice
and trichloroethylene. A solution of 16 g (0.226 mole) of
dry chlorine in 50 ml of CCl4 (prepared between -200 and
-30°) was added dropwise over a period of twenty-five
minutes to the cyclooctatetraene solution maintained be-
tween -200 and —40°. After addition was complete, the solu-
tion was stirred while permitting it to warm to room tempera-
ture. Solid CaC03 was added and the solution was shaken
vigorously and filtered.

To this filtrate, 28 g (0.198 mole) of dimethyl acety-
lenedicarboxylate was added, and the solution was refluxed
for 3 hours. The solvent was removed under reduced pressure
to give a bright orange oil, which was not purified but

thermolyzed immediately.

171

 

172

Distillation of the oil at 10 mm gave a clear liquid
with a boiling range of 70—800. Analysis of the liquid by
gas chromatography (5' x 3/8" SE-30, 120°, 100 ml/min of
He) showed that there were two components present, in the
ratio of 4:1. The two components were separated by means
of a spinning band column.

The major and higher boiling component, gi§f3,4—di-
chlorocyclobutene (119), was isolated in a 60% yield. Its

-1 and was the

ir spectrum showed bands at 1220 and 1280 cm
same as that of the previously prepared 11251“ Its nmr
spectrum (CCl4) consisted of two-proton doublets (J = 1.0
Hz) at T 4.93 and 3.78.

The minor and lower boiling component, trans-1,4—di-
chlorobutadiene, showed strong absorption in its ir Spectrum
at 1560 cm_1. The spectrum was identical with that previ-

ously described.51 The nmr Spectrum (CC14) consisted of a

multiplet from T 3.00 to 4.16.
B. The Synthesis of Iron Nonacarbonyl 5?

Fifty ml of iron pentacarbonyl was dissolved in 300 ml
of glacial acetic acid and irradiated through quartz with
a 450 watt Hanovia Type L mercury arc lamp. Nitrogen was
continuously bubbled through the magnetically stirred solu-
tion. Periodically, the reaction was stopped and crystals
Of iron nonacarbonyl were filtered and washed with 95%
ethanol and anhydrous ether. This was continued until no

Crystals were formed on further photolysis. The iron

 

 

173

nonacarbonyl was dried in vacuo to give 51.40 g (78%) of

 

these orange crystals.

C. The Preparation of Cyclobutadieneiron Tricarbonyl (181)52.

 

To a mechanically stirred solution of 4.15 g of cis-
3,4-dichlorocyclobutene (179) in 100 ml of dry benzene,
'maintained in a nitrogen atmOSphere, was added over a twenty—

minute period, 27 g of Fe2(CO)9. During the course of the

 

reaction, the temperature of the mixture was maintained at
approximately 60°. The reaction was complete after an addi-
tional hour of stirring at 60°, as confirmed by vpc analysis
(5' x 3/8" SE-30, 120°, 90 ml/min of He) which showed the
absence of 112.1“ the reaction mixture. The mixture was
filtered through alumina, the alumina was washed with pen-
tane, and the benzene and pentane solutions were evaporated
in vacuo to give 2.92 g (45%) of cyclobutadieneiron tricar-
bonyl as a dark orange oil. Compound 181 was not purified
but immediately used for the reaction with 2,3,4,5,6,6-hexa-

methyl-2,4-cyclohexadienone (2).

D. The Synthesis of 2,3,4,5,6,6-Hexamethyl-2,4—cyclohexa—

dienone (£24.

 

In a two-liter three-necked round-bottomed flask were
placed 100 g of hexamethylbenzene, 200 ml of methylene chlor-
ide, 400 ml of acetic acid, and 300 ml of sulfuric acid.

This slurry was mechanically stirred at 0° while a solution

containing 30 ml of 90% hydrogen peroxide, 45 ml of acetic

174
acid, and 30 ml of sulfuric acid was added over a forty-
five minute period. The solution was stirred for eight
hours, then poured over 1500 g of ice. The organic layer
was washed three times each with water, aqueous sodium bi—
carbonate, 10% sodium hydroxide, and water again. The
organic layer was dried over anhydrous magnesium sulfate,
and the drying agent was filtered. Evaporation of the sol-
vent under reduced pressure gave a dark oil. Vacuum dis-
tillation at 83-85° (1.0 mm) resulted in 81 g (73.5%) of
2,3,4,5,6,6-hexamethyl-2,4-cyclohexadienone as a yellow oil,
which was identical in all aspects with previously prepared

2.

(V

E. The Synthesis of 1,6,8,8,9,10-Hexamethyltricyclo-
[2.2.4.02c5]deca-3,9-diene—7-one (176).

 

 

To a mechanically stirred solution of 2.50 g of 2
and 2.92 g of 181’in 500 ml of acetone at room temperature
was added 30 g (0.055 mole) of ceric ammonium nitrate over
a two-minute period. Vigorous stirring was continued another
ninety seconds after addition was complete. The reaction
mixture was poured quickly into 500 ml of a saturated salt
solution. The solution was extracted with ether three times
and the combined organic extracts were dried over anhydrous
calcium chloride. Evaporation of the solvent under reduced
pressure gave a black-tarry residue, which was chromatographed
on alumina. Elution with hexane gave 2.45 g (76%) of 116’

as a light yellow semi-crystalline solid.

 

 

175

 

.Aobav maoablwsmflplm.mnmomp
Ian.mo.v.m.NaoHoaoflHuamnummemnloa.m.m.w.©.H mo Avauov Ednuommm HEZ .mm madmam

l . l . _ r
. . -. . , l na.oz. «tn. , . . tl.‘bunltlbl|>t|.’(lf t.ll’|"rlllu[l r P )L h > >

..T

o T.

w” co 9. T, :cm. - 93 -xsléa .3

Ill—lit » I) F p (T — b)» Irtlrlr!o ,u

C)
CH

. c. . o .. . . . .m
3 ...“.a. ... . n - . . o . . . ... \....o.. -. ,. .
.- .- a . _ . , . . . . .c ‘ . . . . . s I) ... . . . o. .. .... . o‘.

.. " ... .. ..... “a. . .. . .. p. . g ,.. . a . . . r . ... (...1 .. ..r. ..
.. ... . ... .. . . ...... .31.; v .. .- . ..s ._ . :\ T\ f. \
.. . . .. . D $\

-
-

u-
‘

.
' r

-
B

 

_
:
.,_

ml

—--._....._...__
.~‘ ?:_-.- w "

m
_
. . —
"i m
__ __. :
r. m__ ..N _ .l
... l... L
. 1. h.
u ._ i:
e a
2 g
F. a
L ._
S.” _
n “a.

176
The ir spectrum of 116 showed a band at 1700 cm-1.

Its uv spectrum in cyclohexane had maxima at 300 nm (e =
384) and 223 nm (e - 1630). The nmr spectrum (CCl4).con—
tained three—proton singlets at T 9.13, 9.08, 8.95, and
8.92, three—proton quartets (J = 1.0 Hz) at T 8.30 and 8.42,
a two-proton pair of doublets (J = 4.0 Hz) at T 7.29 and
7.54, and a two-proton pair of doublets (J = 3.0 Hz) at

T 4.25 and 4.00. The mass spectrum of lZ§.had a parent

peak at m/e = 230.

F. Synthesis of 3-Methyl-d3-2,4,5,6,6—pentamethyl-2,4-

cyclohexadienone (184)4.

 

 

Five grams of 2,3,4,5,6,6-hexamethyl-2,4-cyclohexa—
dienone g’was dissolved in 20 ml of methanol-d1 and 0.5 g
of sodium methoxide was added. The solution was magneti-
cally stirred for three hours. Periodically samples were
withdrawn and checked by nmr. The disappearance of the
peak at T 7.92 was observed. By integration it was deter-
mined that 95% of the protons on the methyl group at C-3
were exchanged. After three hours, the methanol-d1 was re—
moved under reduced pressure. The residue was dissolved in
methylene chloride and washed twice with ice water. The
organic layer was separated and dried over anhydrous mag-
nesium sulfate. Evaporation of the solvent gave a quantita-

tive yield of 184.

177

G. Synthesis of 9eMethyl-d3-1,6,8,8,10-pentamethyl-
tricyclo[2.2.4.02o5]deca-3,9-diene-7-one.

 

 

To a mechanically stirred solution of 0.62 g of 182’
and 0.75 g of cyclobutadieneiron tricarbonyl in 250 ml of
acetone at room temperature was added 10 g of ceric ammon-
ium nitrate over a one-minute period. Vigorous stirring
was continued another ninety seconds, then the reaction
mixture was poured into 250 ml of a saturated salt solution.
The solution was extracted three times with ether and the
combined organic extracts were dried over anhydrous calcium
chloride. Evaporation of the solvent under reduced pres-
sure gave a black-tarry residue, which was chromatographed
on alumina. Elution with hexane gave 0.52 g (75%) of the
adduct as a pale-yellow semi-crystalline solid.

The nmr spectrum (CCl4) of the adduct consisted of
three-proton singlets at T 9.13, 9.08, 8.95, 8.92, and 8.30,
a two-proton pair of doublets (J = 4.0 Hz) at T 7.28 and
7.54, and a two-proton pair of doublets (J = 3.0 Hz) at

T 4.25 and 4.00.

H. The Reduction of 1,6,8,8,9,10-Hexamethyltricyclo-
[2.2.4.02:5]deca-3,9-diene-7-one (176).
#fi.

 

 

In a 100-ml three-necked round-bottomed flask fitted
with a reflux condenser, one gram of lithium aluminum
hydride was added to thirty ml of anhydrous ether. One
gram of lzg’dissolved in 20 ml of anhydrous ether was added

to the magnetically-stirred solution of lithium aluminum

3,

Mgom-‘z‘.
(
I

178

hydride at such a rate as to cause gentle reflux. The re-
sulting solution was stirred at room temperature for thirty
minutes and quenched by carefully adding four ml of water.
The solution was filtered, and the ether was evaporated to
give 0.98 g of a colorless oil which had an adsorption in
its ir spectrum at 3450 but not at 1700 cm-1. Vpc analysis
(10' X 1/4" FFAP, 200°, 90 ml/min of He) of the colorless
oil showed the presence of two compounds in the ratio of

7:5. This oil was used in the next step without purifica-

I. - I”
l.” t
.L
1

tion or further characterization.

I. Acetylation of 1,6,8,8,9,10-Hexamethyltricyclo-
[2.2.4.02I5]deca-3,9-diene-7-ol (186 and 187).

 

 

The oil obtained in part H was dissolved in five ml of
pyridine and 1.0 ml of acetic anhydride. The reaction mix-
ture was permitted to remain overnight, then was diluted
with water and extracted with ether. The organic layer was
separated and dried over anhydrous magnesium sulfate.
Evaporation of the solvent under reduced pressure gave 0.96 g
(83%) of a bright yellow oil. Vpc analysis (10' x 3/8"
FFAP, 190°, 200 ml/min of He) of the oil showed the presence
of two compounds in the ratio of 7:5.

The ir spectrum of the oil showed a strong band at 1725
cm-1. The nmr spectrum (CCl4) contained singlets at T 9.40
(4H), 9.27 (3H), 9.07 (4H), 8.97 (17.6H), 8.35 (14H), 8.06
(4H), 8.00 (3H), 5.70 (1H), and 5.60 (1.4H) and multiplets

from T 7.18 to 7.70 (5H) and 4.05 to 4.33 (4.8H). The mass

 

179
Spectrum of the oil had a parent peak at m/e = 274. No

attempt was made to separate the oil into its two components;

the mixture was photolyzed as described in the next section.

J. The Photolysis of 7-Acetoxy-1,6,8,8,9,10-hexamethyl-
tricyclo[2.2.4.02:5]deca—3,9-diene (190 and 191).

 

 

The acetates obtained in step I were dissolved in 50 ml
of acetone in a quartz test tube. The solution was deoxy-
genated and irradiated through a Corex filter with a 450 NA
watt Hanovia Type L mercury arc lamp for 24 hours. The E

solution was evaporated in vacuo to give 0.98 g of an orange

 

oil, which was chromatographed on 60-200 mesh activated
alumina. Elution with benzene gave 0.60 g of a 7:5 mixture
of acetates 192 and 193, respectively, which were separated
by preparative vpc (10' X 3/8" FFAP, 200°, 200 ml/min of He).

Compound 192 was a colorless oil with a viiim = 1725

cm-1. Its nmr spectrum (CC14) showed singlets at T 9.23
(3H), 9.12 (3H), 9.05 (6H), 8.90 (6H), 8.02 (3H), and 5.47
(1H) and a multiplet from T 7.08 to 7.48 (4H), the mass
Spectrum of lgz'had a parent peak at m/e = 274.
Compound‘lgg'had a light yellow oil with a viiim= 1725
cm-1. Its nmr spectrum (CC14) showed singlets at T 9.33
(3H), 9.20 (3H), 9.05 (6H), 8.94 (6H), 8.05 (3H), and 5.47
(1H) and a multiplet from T 7.08 to 7.48 (4H). The mass
spectrum of lgg'had a parent peak at m/e = 274.
K. Reduction of 9-Acetoxy-1,2,3,4,10,10-hexamethylpenta-
cyclo[4.4.0.0.1I30.3r407:3]decane (1221.

In a 50-ml three-necked pear-shaped flask, 0.50 g of
lithium aluminum hydride was added to fifteen ml of anhydrous

ether. A solution of 200 mg of 122,1n ten ml of anhydrous

 

180
ether was added to the mechanically stirred solution of
lithium aluminum hydride at such a rate as to cause gentle
reflux. The resulting solution was stirred at room tempera-
ture for thirty minutes and quenched by carefully adding
two ml of water. The solution was filtered, and the ether

was evaporated under reduced pressure to give 165 mg of 194.

film

-1 . .

The ir spectrum of lgg’had a v
nmr Spectrum (CCl4) showed singlets at T 9.33 (3H), 9.18
(3H), 9.07 (3H), 8.92 (9H), and 7.00 (1H), quartets (J =
3 Hz) at T 7.48 (2H) and 7.30 (2H), and a broad singlet at
T 8.75 (1H), which disappeared on shaking with D20. The

mass spectrum of 194 had a parent peak at m/e = 232.

L. Reduction of 9-Acetoxy—1,2,3,4,10,10-hexamethylpenta-
cyclo[4.4.0.0.1I20.3I407I3]decane (193).

 

 

In a 50—ml three—necked pear—shaped flask, 0.50 g of
lithium aluminum hydride was added to 15 ml of anhydrous
ether. A solution of 250 mg of 193 in 10 ml of anhydrous
ether was added to the mechanically stirred solution of
lithium aluminum hydride at such a rate as to cause gentle
reflux. The resulting solution was stirred at room tempera-
ture for 30 minutes and quenched by carefully adding 2 ml
of water. The solution was filtered, and the ether was
evaporated under reduced pressure to give 180 mg of 195,

The ir spectrum of 195 had a viiim = 3500 cm-1. The
nmr Spectrum (CC14) showed singlets at T 9.33 (3H), 9.22
(3H), 9.10 (3H), 9.03 (6H), 8.97 (3H), and 7.03 (1H), a

multiplet from T 7.17 to 7.50 (4H), and a broad singlet at

.L's..(_e- [a

181
T 8.63 (1H), which disappeared on shaking with D20. The

mass spectrum of 195 had a parent peak at m/e = 232.

M. The Synthesis of 1,2,3,4,10,10-Hexamethylpentacyclo-
[4.4.0.0.1r20.3a407:8] decan-9-one (177).

 

 

A solution of 150 mg of either 194 or 195 in five ml
of acetone was added to a test tube at 0°. One ml of
Jones reagent55 was added and the mixture was permitted to

stand for 25 minutes. Water was added and the solution was

 

extracted with ether. The organic layer was dried over ]
anhydrous magnesium sulfate, filtered, and evaporated in

ygggg to give 140 mg of a light yellow oil. The vpc trace

(10' x 1/4" FFAP, 200°, 120 ml/min of He) showed the pres-

ence of only one compound, 111” which was purified by

preparative vpc under the same conditions.

 

film

Compound_lzz_was a colorless oil with a vmax = 1690

cm-1. Its nmr spectrum (CC14) showed singlets at T 9.18
(3H), 9.12 (3H), 8.97 (6H), 8.93 (3H), and 8.87 (3H) and a
multiplet from T 6.66 to 7.26 (4H). The mass Spectrum of

177 had a parent peak at m/e = 230.

N. Direct Irradiation of 1,6,8,8,9,10-Hexamethyltricyclo-
[2.2.4.02'5]deca-3,9-diene-7-one (176).

 

 

In a quartz test tube, 0.20 g of 176 was dissolved in
10 ml of either pentane or methanol and photolyzed in the
presence of a 450 watt Hanovia Type L mercury arc lamp

through a Corex filter. Photolysis was followed by vpc

182

 

.Ahbfiv ozonmlcmomflmm.50v.m.ON.H.o.o.v.vH
IoHomomucmmahsummewnloH.oH.v.m.m.H mo Avaoov Ednuummm HEZ .om musmflm

— u — . t . H F _
kirlr'lll’ltLllbluilbi . P p r . r -.- 1ft--. .Ins-)->-!!larl.|'(;[tl;.b!uftllblcrb ILIQLIIKII {[1.leourl|rnlvlluflf.

I.” O.Ns .‘ua: Q...“ “v.0 AF” fur-cm ai'.m
. . . ... . 1. . .
.... .. w .... . _ F. I...” ..r...- ”.0 ...“ w. ‘v - s1... ‘mw._..~..a,—...._ v... ......cJK’thLM). ..JV. .. . ......I .94 s— ,2. .

1"—}&- v,. on n o .. .. .u .. ...u o . .v .- “.ovit. fia... ...r.. .o o .‘ ‘a.. o o .......

u a . . o . . . .. g . . . a a .

...\ . _. 1.71.: J. . . . . . 4.. ......r ....
, . .. . . .fl...¢.

 

 

 

 

 

 

183
(10' x 1/4" OV-25, 170°, 100 ml/min of He). After 3 hr, the

reaction was stopped and the solvent was evaporated in vacuo

 

to give 0.182 g of a dark orange oil. The vpc trace of the
oil showed the presence of two photoproducts. Under these

same conditions, the two photoproducts 196 and 197 were sep-

arated and collected by preparative gas chromatography. Ir-
radiation for another three hours gave an essentially quanti-
tative yield of 121;

= 23.8 minutes, was a colorless oil

t
. max _ -1 MeOH _ :
With a Vfilm — 1760 cm and °max — 318 (e 397) and 235

Compound 196, R

nm (e = 1860). Its nmr spectrum (CCl4) showed three-proton
singlets at T 9.10, 8.98, 8.88, and 8.77, a six-proton broad
singlet at T 8.33, a two-proton broad singlet at T 6.84, and
a two-proton singlet at T 3.82. Its nmr spectrum (C6D6)
showed three-proton singlets at T 9.15, 9.13, 8.93, and 8.85,
three-proton quartets (J = 0.8 Hz) at T 8.32 and 8.54, a
two-proton quartet (J = 1.0 Hz) at T 7.05, and a two-proton

quartet (J = 0.9 Hz) at T 3.90. The mass spectrum of 196

w

had a parent peak at m/e = 230.

= . MeOH _
Compound 197, Rt 6.8 minutes, had a xmax - 223 nm

(e = 1900). The nmr spectrum (CCl4) of 191’showed singlets
at T 9.23 (3H), 8.99 (6H), 8.97 (3H), 8.38 (6H), and 3.83
(2H) and a two-proton pair of doublets (J - 4.0 Hz) at T

7.04 and 7.22. Its nmr spectrum (C6D6) showed three-proton
singlets at T 9.13, 9.02, 8.98, and 8.95, three-proton

quartets (J - 0.5 Hz) at T 8.37 and 8.43, a two-proton multi-
plet from T 6.86 to 7.25, and a two-proton pair of doublets
(J = 1.5 Hz) at T 3.72 and 3.74. The mass spectrum of 197

had a parent peak at m/e = 202.

 

184

 

.Awmac mco-m:mcmflw-w.m

Imompmw.ao.o.m.Naoaomofluuamnummemnlw.m.v.m.m.H mo ATHOOV Esuuommm H82 .Hm musmflm

 

. '
.r‘r r . h b b b LOT.) 5 P) x— b b 5 Pi... I P t L I} a L) )lrh L D L p Ollie}. ,' .6 . I. Q.HII ptl .! v-oil' u ‘slllvull'nIID-IL‘. n‘ilL‘li'le}‘i.-lill’_ -
o C a o \. c, .... ...... .. 2:...“ 0.... o s C ....
I... . .JJ-.:~: qr ~ — P
\ ... . . . .... a p _ ... ... _ om . .. m . ... o. . . . r .
.. r 2... ...L. .. .\.....A 1... ... Ur. ..JK....\.... . ~. 4.. .....o. ....) t. .. . T... .. ...s... ...r... ,.. . 1...; ... ~.~.. 5%.. ... .. . _. .. ..v ... ......,...m ~ ...:ns: ...... ..- ..k m. ...oirr...n..r.. ..4... .. ._._ .
. 1 .Sb) 2T:.T4aa . :3 M fi.+.... ...: . ..x. _. .25.f..§ :wa:::; .,
. .... .. L m _ . a
_ .. . _ . . .v
w “ i ._ H.“ _
m _ .. l t _ t _.
___ u it ‘ . r r.
u“ _ x. _ . t
_ . p a .. ._
T T... l m. __ 1.. __
:_ .... L _ . .
:. .u _ n . _ _
. . .. .s . n
h. g T. .w . .
:._l¢ .L
e.;_. .
.— “a ..— _" x
. J o t .,
N. in ,u m. _
..n .n .. ~_ 0
. _. . .
_ _. _
r _. L .h _
: n. J. z. r. m
.2 : 1 _. .
in l i __ i
..u a“ y. .1 u
m. . r. n
,. _. m.
w . .i
M. .. .
. . __ ._

 

 

.AbmHv mswaplw. mlmco:

lap Ho. 0. m. Huoaowoauuamnqumxmnlo m. w m m N mo AvHOUV Esnuommm HEZ .Nm musmflm

.L'|I r b ~ - F b p _ p h h b F) L , b u — h» L > r h L L n L _ P p p > b . . by) _ . m L . L L w > Flt » h n P > > L b b b!) b . b L b
9m 98 as fibrfiu 98 as or
. \(J p \.p§s1¥(\....,ol

xx???...........<<...?...}s... .11.. .... ...J; \. 33:58:34.... . 4 ..
..x ,

( b

r. a). .. .... R/N
.1 . / 3 ...... 1....sc}sé< .....1..........€... 2.5.. ...; \

 

185

- .. H. _ . .
Mn... ..-

186

O. Sensitized Irradiation of 1,6,8,8,9,10-Hexamethyl-
tricyclo[2.2.4.02o5]deca-3,9-diene-7-one (176).

 

 

In a quartz test tube, 100 mg of 116 was dissolved in
10 ml of acetone and photolyzed through a Corex filter with
a 450 watt Hanovia Type L mercury arc lamp for six hours.
The photolysis was followed by vpc (10' x 1/4" OV-25, 180°,
100 ml/min of He). Evaporation of the solvent under re-
duced pressure gave 97 mg of a light yellow oil, which was
92% compound 222 by vpc analysis. Compound 222 was collected
under the same vpc conditions.

film

Compound 202 had a v MGOH
'VV” max

= 1715 cm’1 and a °max = 290
Sh (e = 1050), 230 (e = 11,800), and 217 nm (€ = 9850). Its
nmr spectrum (CCl4) contained singlets at T 9.17 (3H), 9.08
(3H), 8.97 (6H), 8.92 (3H), 8.87 (3H), a quartet (J = 1.0

Hz, 2H) at T 7.06 and a multiplet from T 3.69 to 3.98 (2H).

The mass Spectrum of 202 had a parent peak at m/e - 230.

P. The Absence of Elemental Analyses for the New Compounds
Synthesized in this Study.

 

Several attempts at elemental analysis of compounds

176, 196, 197, and 202 gave results not in close agreement

 

with the theoretical values. All compounds gave low carbon
values. These compounds tended to pick up oxygen rather
easily. Compound 121 was especially susceptible to oxygen,
for the mass Spectrum after standing in the room for two

days showed that it had picked up a mole of oxygen. As a

 

 

187

 

 

.Amouc maoumumcmnsumomsmos.«08.8.o.o.m.m_

. Hsmam
IoHomomHumuH>Lumfimxm£Ioa.m.m.b.m.H mo Avaoov Esuuommm HEZ mm m .

d! p . »L > L'r‘l.)_'|

 

h n _ F L L by r r P b M r l? L ’ ~ 5’! ‘ Ll) b o.“
.r..-. _ .+LL...rr_..L.drT.-.tr-L-.Q.-r--. cs mkas OH on
..L. {PA . o_-.v I r - 1,015: .2?!“ ‘Fné‘2§()l§am¢— \A... ‘.-.v‘fC-.VVVO<J~.)\.' .JP
1......“ 181...... _ .M .... ......t?...§§. .... ......12..... 12.....1}...2......I..t.. .-.... .. ..2
. _ . ~. ~ {Jug/.2 A .e ...... . . .. ._
M _ _ i . _ a . .. *
... . I I r i.
Y . — . v
z i s. . 1
in .;“_
M. “L“.
..H U“ .0 , _
f _ .h .1
. _.
.l ... .1 .. l
l .. . m
l. r
n .

—_._.,‘,.,-" \

--.-.--.--

188
result, the molecular weight of the particular compound was
based solely on its mass spectrum without its complementary
elemental analysis.
Due to difficulty with the above mentioned compounds,
no attempts were made to obtain elemental analyses of the
homocubane compounds but their molecular weights were ob-

tained solely from the mass spectrum.

'—'.T:'u. a

a"! ‘1'“:

'55

 

 

 

BIBLIOGRAPHY '

10.

11.

12.

13.

BI BLIOGRAPHY

Woodward, R. B., and R. Hoffmann, J. Amer. Chem. Soc.,
81” 2511 (1965). F‘

 

 

. :1.

Woodward, R. B., and R. Hoffmann, "The Conservation of
Orbital Symmetry," Academic Press Inc., New York, N.Y.,
1970, pp. 114—115.

Swatton, D., and H. Hart, J. Amer. Chem. Soc., 82” 5075
(1967).

 

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

 

Kropp. Paul J., in "Organic Photochemistry," Vol. I,
O. L. Chapman, Ed., Marcel Dekker, Inc., New York, N.Y.,
1967. PP. 13-20.

 

Curtin, D. Y., R. J. Crawford, and M. Wilhelm, J. Amer.
Chem. Soc., 82, 1391 (1958).

Curtin, D. Y. and D. H. Dybvig, J. Amer. Chem. Soc.,
82” 225 (1962).

Kalberer, F. and H. Schmidt, Helv. Chim. Acta, 32, 779
(1957).

 

Caine, D. and J. B. Dawson, J. Org. Chem., 22/ 3108
(1964).

Kropp: P. J., Tetrahedron, 21, 2183 (1965).

 

Swatton, D. and H. Hart, J. Amer. Chem. Soc., 82, 1874
(1967).

Silverstein, R. and G. Bassler, "Spectrometric Identi-
fication of Organic Compounds," 2nd ed., John Wiley and
Sons, Inc., New York, N.Y., 1967, p. 129.

waring, A. J., in "Advances in Alicyclic Chemistry,"

Vol. I, H. Hart and G. Karabatsos, Eds., Academic
Press, New York, N.Y., 1966, pp. 184-188.

189

14.

15.

16.

17.

18.

19.

20.

21.

22.

23.

24.

28.
29.

30.

31.

190

Miller, B. and H. Margulies, J. Amer. Chem. Soc., 89,
1678 (1967).

 

 

Eastman, R. H., J. Amer. Chem. Soc., 26, 4115 (1954).

Zimmerman, H., D. S. Crumrine, Dietriéh Dopp, and Paul
S. Huyffer, J. Amer. Chem. Soc., 91, 434 (1969).

Brennan, T. and R. H. Hill, J. Amer. Chem. Soc., 92,
5614 (1968).

 

Childs, R. F. and S. Winstein, J. Amer. Chem. Soc.,
92, 7146 (1968).

 

Koptyug, V. A., L. I. Kuzubova, I. S. Isaev, and V. I.
Mamtyuk, Chem. Commun., 389 (1969).

 

Barton, D. H. R., P. deMayo, and M. Shafig, J. Chem.
Soc., 140 (1958); D. Arigoni, H. Bosshard, H. Bruderer,
G. Buchi, O. Jeger, and L. J. Krebaum, Helv. Chim.

Acta, 40, 1732 (1957).

 

 

Weinberg, K., E. C. Utzinger, D. Arigoni, and O. Jeger,
Helv. Chim. Acta., 43, 236 (1960).

 

Schaffner, K. and O. Jeger, Helv. Chim. Acta, 49, 1049
(1966).

Patel, D. J. and D. I. Schuster, J. Amer. Chem. Soc.,
92, 5137 (1968).

Zimmerman, H. E. and J. O. Grunewald, J. Amer. Chem.
Soc., 89, 5163 (1967).

Miller, Bernard, J. Amer. Chem. Soc., 89, 1690 (1967).

 

Curran, W. V. and D. I. Schuster, Chem. Commun., 699
(1968).

Rhoads, S. J., "Molecular Rearrangements,“ Vol. I,
Paul DeMayo, Ed., John Wiley and Sons, New York, N.Y.,
1963, pp. 660-684.

Kropp, P. J., J. Amer. Chem. Soc., 86, 4053 (1964).

 

Collins, P. M., and H. Hart, J. Chem. Soc., 895 (1967).

Fahrni, P. and H. Schmid, Helv. Chem. Acta, 42, 1102
(1959).

Mathai, K. P. and S. Sethma, J. Indian Chem. Soc.,
42, 347 (1963).

 

 

 

32.

33.

34.

35.

36.
37.
38.

39.

40.

41.

42.

43.

44.
45.

46.

47.

48.

49.

50.

51.

191
Reese, J., Chem. Ber., 22, 979 (1953).
"Handbook of Chemistry and Physics," Forty-second
edition, The Chemical Rubber Publishing Company,
Cleveland, p. 1058.

Silverstein, R. M., J. O. Rodin, D. L. Wood, and L. E.
Brown, Tetrahedron, 22, 1929 (1966).

 

Zimmerman, H. E. and J. S. Swenton, J. Amer. Chem. Soc.,
g2, 906 (1962).

 

See Part I of this thesis.

 

Miller, B., J. Amer. Chem. Soc., 22, 1685 (1967).

Miller, B., J. Amer. Chem. Soc., 22, 751 (1969).

 

Eliel, E. L., N. L. Allinger, S. J. Angyal, and G. A.
Morrison, "Conformational Analysis," Interscience
Publishers, New York, N.Y., 1966, p. 44.

Dauben, W. G., Chem. Weekblad, 62, 381 (1964).

 

Buchi, G. and E. M. Burgess, J. Amer. Chem. Soc., 22,
4333 (1960).

Schuster, D. I., M. Axelrod, and J. Auerbach, Tet.

Letters, 1911 (1963).

Baggiolini, E., K. Schaffner, and O. Jeger, Chem.

Commun., 1103 (1969).

Reference 2, p. 97.
Williams, J. R. and H. Ziffer, Chem. Commun., 194 (1967)

Dauben, W. G. et al., J. Amer. Chem. Soc., 22, 1786
(1970).

Domb, S. and K. Schraffner, Helv., 22, 677 (1970).

Starr, J. E. and R. H. Eastman, J. Org. Chem., 22,
393 (1966).

Erman, W; F., J. Amer. Chem. Soc., 22, 3828 (1967).

Hart, H. and R. K. Murray,Jr., Tetrahedron Letters,
4995 (1968).

Nenitzescu, C. D. et al., Chem. Ber., 22, 378 (1964).

 

 

 

52.

53.

54.

55.

56.

57.

192

Paquette, L. and L. D. Wise, J. Amer. Chem. Soc., 22,
6665 (1967).

Braye, E. H. and W. Hubel, Inorg. Syn., 2, 178 (1966).

Winstein, S., et al., J. Amer. Chem. Soc., 21, 5247
(1965).

Bowers, A., T. G. Halsall, and E. R. H. Jones, g,
Chem. Soc., 2548 (1953).

See K. Nakanishi, "Infrared Sbsorption Spectroscopy,"
Holden-Day, Inc., San Francisco, 1962, p. 42.

Hart, H., R. K. Murray, and G. D. Appleyard, Tetra—
hedron Letters, 4785 (1969)

 

H
H
I..I
“
“I
N
H
H
II
I
E N
"
"ll

!

IIHINHIH

31293 03168 8108